JP6524338B2 - Search apparatus, search method, and search program - Google Patents

Description

  The present invention relates to a search apparatus, search method, and search program for searching a network in which a plurality of nodes constituting a node group are connected.

  In recent years, the seriousness of large-scale blackouts in various countries has promoted the introduction of PMU (Phasar Measurement Unit). Also, PMU measures data such as current, voltage, and frequency several hundreds times more frequently than conventional SCADA (Supervisory Control And Data Acquisition). Therefore, for example, if 500 PMU data are continuously stored, an amount of data of several tens to several hundreds [TB] will be accumulated in one year. Analysis of the pattern of such a large amount of data enables stable operation of a normal power system.

  The configuration of the power system is updated by daily system operation and maintenance. That is, the configuration of the network showing the power system changes in time series due to the new installation, the shutdown, and the removal of the facilities. Therefore, even if PMU data of the same equipment are similar, they are not similar because the network configuration, specifically, other equipment of the connection partner, for example, is different. In order to consider this update, the configuration of the network updated daily is archived as a snapshot. Then, the similar case search process searches similar snapshots from the archive. The similar case search process is a process of searching a partial network for a large number of networks managed in time series, that is, time-series snapshot groups for the networks.

  Patent Document 1 discloses a graphic data management method that makes searching in the time axis direction fast and easy. The graphic data management method is a method of managing the existence period of the feature in a database by registering the generation time Ts at which the feature is built and the annihilation time Te being destroyed in the figure data.

  Patent Document 2 stores the process of refinement of a design process as a directed graph consisting of nodes having design information as an attribute, and newly reads a plurality of nodes, and is similar to the design information of each of the plurality of nodes. Searching a node (similar node) in the directed graph having the designed information, searching a path between the similar nodes, and a design process up to refinement which is a path between the similar nodes This is a thinking support method that presents information for filling in between the jumped design items by performing the displaying step.

JP, 2008-176808, A JP, 2004-199418, A

  However, in the above-described prior art, for example, when dealing with a wide area power system such as a power company, the scale of the network also becomes large. Therefore, computational cost increases when creating a snapshot for each ever-changing network configuration. As a result, it takes time for partial network search necessary for detecting a sign of abnormality. On the other hand, when a snapshot is not generated for each network configuration that changes every moment, it can not respond to the network change and the search accuracy is reduced.

  An object of the present invention is to realize high-speed and high-precision search of a network whose configuration changes temporally.

  A search device, a search method, and a search program according to one aspect of the invention disclosed in the present application are a search device, a search method, and a search program for searching a network in which a plurality of nodes constituting a node group are connected. The search device includes a processor that executes a program, and a storage device that stores the program, and the storage device stores snapshot information including a snapshot and a generation time of the snapshot. And one or more pieces of update information including update content indicating addition of a new node in the network, change of the state of the node, or deletion of the node, and update time of the addition, modification or deletion A time-series update information group is stored, and the snapshot is stored in the network. The connection information of the node group, and node information holding at least one of the type or state of each node of the node group, the processor acquiring the search partial network, and the search by the acquisition process When the partial network is acquired, the snapshot and the update information up to the specific update time are used using the snapshot and the update information up to the specific update time in the time-series update information group Search processing for searching for a partial network that matches the search partial network, associating a valid period from the search for the partial network until the search is no longer performed with the partial network, and output processing for outputting a search result by the search processing , And is characterized in that.

  According to the representative embodiment of the present invention, it is possible to realize high-speed and high-precision search of a network whose configuration changes temporally. Problems, configurations, and effects other than those described above will be clarified by the description of the following embodiments.

FIG. 1 is an explanatory view showing a time-varying network, a table and update information in time series. FIG. 2 is an explanatory view showing an example of a system configuration of the search system according to the present embodiment. FIG. 3 is a block diagram showing an example of the hardware configuration of the search device. FIG. 4 is an explanatory diagram of an example of the network NWi at time ti. FIG. 5 is an explanatory view showing an example of a sensor table. FIG. 6 is an explanatory diagram of an example of the time-series DB. FIG. 7 is an explanatory view showing an example of the data structure of the update information Ri. FIG. 8 is a block diagram showing an example of the functional configuration of a search device. FIG. 9 is an explanatory view showing a snapshot and extraction of partial network candidates. FIG. 10 is an explanatory diagram of an addition example 1 of a new node at time t1. FIG. 11 is an explanatory diagram of a second example of addition of a new node at time t1. FIG. 12 is an explanatory diagram of a deletion example 1 of the existing node at time t2. FIG. 13 is an explanatory diagram of a second example of deletion of a new node at time t2. FIG. 14 is an explanatory diagram of a first modification of the existing node at time t3. FIG. 15 is an explanatory diagram of a second modification of the existing node at time t3. FIG. 16 is a flowchart of an example of a process of storing time-series data by the storage unit. FIG. 17 is a flowchart of an example of a storage processing procedure of update information by the storage unit. FIG. 18 is a flow chart showing an example of the search processing procedure by the search device. FIG. 19 is a flowchart of a detailed process procedure example of the partial network search process (step S1803). FIG. 20 is a flowchart of a detailed process procedure example of the addition process (step S1905) illustrated in FIG. FIG. 21 is a flowchart showing a detailed processing procedure example of the deletion process (step S1906) shown in FIG. FIG. 22 is a flowchart of a detailed process procedure example of the change process (step S1907) depicted in FIG. FIG. 23 is a flowchart showing a detailed processing procedure example of the similar waveform search process (step S1804) shown in FIG.

  The search apparatus according to the present embodiment uses a snapshot indicating the network configuration at a certain time of the network that changes from moment to moment, and a search partial network given as a search key using update information indicating an update history after that time. Find out during which period the network was effectively present. Hereinafter, the present embodiment will be described.

<Search example>
FIG. 1 is an explanatory view showing a time-varying network, a table and update information in time series. In order to simplify the description, the network column indicates a part of the network (nodes A to E,...), And the table or the column of update information also indicates a part of the network (nodes A to E,. Indicates information about

  First, the line at time t0 will be described. The network NW0 at time t0 has a partial network candidate SNa configured by nodes A to C. Node A indicates a power plant, and nodes B and C indicate substations. In the network NW0, nodes A to C are in operation.

  The configuration of the network NW0 at time t0 is specified by the connection information table 101 and the node information table 102. The connection information table 101 and the node information table 102 show a snapshot of the network NW0 at time t0. Time t0 is a snapshot generation time. The combination of the snapshot generation time (time t0) and the snapshot (network NW0) is referred to as snapshot information.

  The connection information table 101 is a table showing a connection configuration between nodes of the network NW0. Out indicates a connection source, and In indicates a connection destination. The node information table 102 specifies a node by a node ID, a node type, and a state. The node ID is identification information that uniquely identifies a node. The node type is information indicating the type of the node specified by the node ID, for example, the type of power station, substation, switching station. The state indicates the state (for example, operation, stop) of the node specified by the node ID and the node type.

  Next, the line after time t1 will be described. In the row after time t1, the update information Ri (i is a nonnegative integer) is stored instead of the snapshot.

  The update information R1 at time t1 specifies the nodes A and B to be updated in the network NW1 based on the update content, the node ID, the node type, and the state. The update content specifies what kind of update has occurred in the network NW2. Specifically, for example, there are three types of update contents: addition of a new node, deletion of a node, and state change of a node. At time t1, since the states of the nodes A and B are changed from "active" to "stopped", the updated contents of the nodes A and B are all "changed".

  The update information R2 at time t2 specifies the nodes D and E to be updated in the network NW2 based on the update content, the node ID, the node type, the state, Out, and In. At time t2, nodes D and E are added to the network NW2.

  The update information R3 at time t3 specifies the nodes A and B to be updated in the network NW3 based on the update content, the node ID, the node type, and the state. At time t3, since the states of the nodes A and B are changed from “stop” to “operation”, the update contents of the nodes A and B are “change”.

  The update information R4 at time t4 specifies the nodes A and B to be updated in the network NW4 based on the update content, the node ID, the node type, Out, and In. At time t4, nodes A and B are deleted from the network NW4.

  The search device acquires a search partial network Q as a query. The search partial network Q is a partial network serving as a search key. Here, as an example, the search condition of the search partial network Q is a configuration in which three nodes are connected in series, the node type of the node at one end is "power plant", and the node type of the remaining nodes is " It is a substation, and the state of each node is “operational”.

  The search apparatus uses the connection information table 101 and the node information table 102, which are snapshots at time t0, and the update information R1 to R4 to determine at which time ti the network NWi of the search partial network Q exists. Identify. At this time, the search device generates a snapshot only at time t0 and does not generate a snapshot at time t1 to t4.

  In the network NW0 at time t0, there is a partial network candidate SCa that matches the search partial network Q. There is no partial network candidate that matches the search partial network Q in the network NW1 at time t1. In the network NW2 at time t2, there are partial network candidates SCb that match the search partial network Q. In the network NW3 at time t3, partial network candidates SCa and SCb that match the search partial network Q exist. A partial network candidate SCb matching the search partial network Q exists in the network NW4 at time t4.

  Therefore, the search apparatus sets the search results 110 to the partial network candidate SCa of the period [t0, t1), [t3, t4), the partial network candidate SCb of the period [t2, t4], and the final valid period. Output as partial network. Since the partial network candidate SCa in the periods [t0, t1) and [t3, t4] does not exist at the times t1 and t4, the partial network candidate SCa is a half-open interval not including the times t1 and t4.

  As described above, by using the snapshot at time t0 and the update information R1 to R4 at time t1 to t4, the generation of the snapshot at time t1 to t4 can be suppressed, and the search speed can be increased. it can. Moreover, thereby, the load reduction of production | generation of the snapshot of time t1-t4 in a search device can be aimed at. Further, by specifying the existence periods [t0, t1), [t3, t4), and [t1, t4] of partial networks that match the search partial network Q, it is possible to improve search accuracy. Therefore, by using the search results obtained from the search device, the administrator can achieve speeding up and high reliability of decision making on the network to be searched.

<System Configuration Example of Search System>
FIG. 2 is an explanatory view showing an example of a system configuration of the search system according to the present embodiment. The search system 200 includes a search device 201, a power network system 202 which is an example of a search target, and a network DB (Database) 203. The search device 201 is a computer that executes the search process shown in FIG. The power network system 202 is a network system in which a plurality of nodes 204 indicating facilities such as power stations, substations and switching stations are connected by a transmission line (not shown). Node 204 has one or more facilities 205. The equipment 205 is an observation object observed by one or more sensors 206 (for example, PMU).

  The search device 201 is connected to the sensor 206 via the first communication network 210. The search device 201 acquires time-series observation values from the sensor 206 for the facility 205 via the first communication network 210. The network DB 203 is connected to the node 204 via the second communication network 220. The network DB 203 holds a snapshot, that is, the connection information table 101 and the node information table 102 in the network NWi at a certain time ti. The node stores the change of the internal configuration in the network DB 203 as the update information Ri. The search device 201 accesses the network DB 203 to acquire a snapshot and update information Ri.

<Hardware Configuration Example of Search Device 201>
FIG. 3 is a block diagram showing an example of the hardware configuration of the search device 201. As shown in FIG. The search device 20130 includes a processor 301, a storage device 302, an input device 303, an output device 304, and a communication interface (communication IF 305). The processor 301, the storage device 302, the input device 303, the output device 304, and the communication IF 305 are connected by a bus. The processor 301 controls the search device 20130. The storage device 302 is a work area of the processor 301. Further, the storage device 302 is a non-temporary or temporary storage medium storing various programs and data. The storage device 302 is, for example, a read only memory (ROM), a random access memory (RAM), a hard disk drive (HDD), or a flash memory. The input device 303 inputs data. Examples of the input device 303 include a keyboard, a mouse, a touch panel, a ten key, and a scanner. The output device 304 outputs data. Examples of the output device 304 include a display and a printer. The communication IF 305 is connected to the first communication network 210 to transmit and receive data.

<Various data structure examples>
Hereinafter, various data structure examples will be described. In the description of the table, the value xx of the AA field bb (AA is a field name and bb is a code) may be described as “AAxx” for convenience.

  FIG. 4 is an explanatory diagram of an example of the network NWi at time ti. The network NWi is stored in the network DB 203 and the storage device 302 of the search device 201. Specifically, for example, the network NWi is managed by the connection information table and the node information table shown in FIG. 1 and the subsequent update information group. The network NWi shown in FIG. 4 is an example, and in practice, for example, since it is a power network covering the entire country, it is a network that is more extensive than the number of nodes shown in FIG.

  FIG. 5 is an explanatory view showing an example of the sensor table 500. As shown in FIG. The sensor table 500 is a table for specifying the location of the sensor 206. The sensor table 500 has a sensor field 501 and a node ID field 502. The sensor field 501 is a storage area for storing a sensor ID as a value. The sensor ID 501 is identification information that uniquely identifies the sensor 206. The node ID field 502 is a storage area for storing a node ID as a value. The node ID 502 is identification information that uniquely identifies the node 204.

  FIG. 6 is an explanatory diagram of an example of the time-series DB. The time-series DB 600 has an observation value table 601 for each sensor ID 501. The observation value table 601 has a time stamp field 602 and an observation value field 603. The timestamp field 602 is a storage area for storing a timestamp as a value. The time stamp 602 is a date and time when the sensor 206 acquires an observation value by observing the facility 205. The observation value field 603 is a storage area for storing the observation value as a value.

  FIG. 7 is an explanatory view showing an example of the data structure of the update information Ri. The update information Ri includes an update time field 701, an update content field 702, a node ID field 703, a node type field 704, a state field 705, an Out field 706, and an In field 707. The update time field 701 is a storage area for storing the update time as a value. The update time 701 is the time when the network NWi is updated.

  The update content field 702 is a storage area for storing update content as a value. The update content 702 indicates either addition (addition) of the new node 204 into the network NWi, status change (change) of the node 204, or deletion (deletion) of the existing node 204. The node ID field 703 is a storage area for storing a node ID as a value. The node ID 703 is identification information that uniquely identifies the node 204. The node type field 704 is a storage area for storing a node type as a value. The node type 704 is information indicating a power plant, a substation, or a switching station. A state field 705 is a storage area that stores a state as a value. State 705 indicates that the node 204 is up or down.

  The Out field 706 is a storage area for storing the node ID 703 of the node 204 targeted for the update content 702. When the update content 702 is "addition", the node ID 703 of the new node 204 to be added is stored. If the update content 702 is “deleted”, the node ID 703 of the existing node 204 to be deleted is stored. The In field 707 is a storage area for storing the node ID 703 of the node 204 that is the connection destination of the node 204 that is the target of the update content 702. When the update content 702 is “addition”, the node ID 703 of the node 204 which is the connection destination of the new node 204 to be added is stored. When the update content 702 is “deleted”, the node ID 703 of the node 204 which is the connection destination of the existing node 204 to be deleted is stored.

<Functional Configuration Example of Search Device 201>
FIG. 8 is a block diagram showing a functional configuration example of the search device 201. As shown in FIG. The search device 201 includes an acquisition unit 801, a search unit 802, an output unit 803, a storage unit 804, and a creation unit 805. Specifically, the acquiring unit 801 to the creating unit 805 realize the function by causing the processor 301 to execute a program stored in the storage device 302 illustrated in FIG. 3, for example.

  The acquisition unit 801 acquires a search partial network Q. Specifically, for example, when the user operates the input device 303, the acquisition unit 801 receives the search partial network Q. In addition, the acquisition unit 801 receives the search partial network Q from the outside via the first communication network 210. Further, the search partial network Q has a search condition. The search conditions include at least the connection configuration of the node 204. The connection configuration of the node 204 is connection information between one or more nodes 204 and the nodes 204 that constitute the search partial network Q. Further, the search condition may include node ID, node type (power station, substation, switching station), state of node 204 (operation, stop), or search target period.

  Also, the search condition may include the node ID of the node 204. For example, as shown in FIG. 1, the node ID “X” of the node 204 of the power plant at the left end of the search partial network Q may be designated. When the node IDs of all the nodes 204 constituting the search partial network Q are specified, the search unit 802 searches except the partial network specified by the node ID. In the example of FIG. 1, assuming that the node IDs of the nodes 204 constituting the search partial network Q are X, Y and Z, search conditions excluding combinations of node IDs X, Y and Z, ie, in operation A partial network consisting of a power plant, an operating substation connected to it, and an operating substation connected to it will be searched.

  The search unit 802 has a partial network search unit 820 and a similar waveform search unit 830. The partial network search unit 820 searches for partial networks when the acquisition unit 801 acquires the search partial network Q. The similar waveform search unit 830 corresponds to time-series observation values observed by the sensor 206 corresponding to the node 204 configuring the search partial network Q, and the node 204 configuring the partial network searched by the partial network search unit 820. It is determined whether or not the time-series observation values observed by the sensor 206 are similar. The partial network search unit 820 will be described.

  The partial network search unit 820 executes partial network search processing using a snapshot and update information up to a specific update time in the time-series update information group. The snapshot is the network NWj at a certain time tj. The time-series update information group is a set of update information Ri stored in the storage device 302. That is, the search unit 802 updates the snapshot of the generation time tj and the update information R (j + 1) from the next update time t (j + 1) next to the time tj to a specific update time tk (k is an integer satisfying kkj + 1). The search process is executed using ~ Rk. For example, k is designated as the end time of the search target period which is one of the search conditions. Further, when there is no particular designation for k, the update information Rk is update information of the latest update time.

  Then, as illustrated in FIG. 1, the partial network search unit 820 searches for a partial network that matches the search partial network Q from the snapshot and the update information Rk up to the specific update time tk. In addition, the partial network search unit 820 associates, with the partial network, a valid period from when the partial network is searched until it is not searched. Specifically, for example, the partial network search unit 820 associates the existence period from the update time when the partial network is searched to the update time when the partial network is not searched, as the effective period of the partial network, to the partial network.

  The partial network search unit 820 includes an extraction unit 821, a selection unit 822, a determination unit 825, a generation unit 824, and a setting unit 823. The extraction unit 821 extracts partial network candidates that match the search partial network Q from the snapshot. Specifically, for example, as shown at time t0 in FIG. 1, the extraction unit 821 extracts a partial network candidate SCa that matches the search partial network Q from the network NW0 that is a snapshot. The partial network candidate is data that can be a partial network, and finally can be a partial network by the partial network search unit 820.

  The selection unit 822 selects the oldest update information that has not been selected from the time-series update information group. Specifically, for example, the selection unit 822 selects unselected update information Ri from a set of update information Ri that is a time-series update information group. In the case of FIG. 1, the selecting unit 822 first selects the update information R1 at time t1, and then selects the update information R2 at time t2, the update information R3 at time t3, and the update information R4 at time t4.

  When the update content 702 of the update information Ri selected by the selection unit 822 is “addition”, the determination unit 825 determines whether the search partial network Q includes the configuration of the new node 204 or not. When search partial network Q includes the configuration of new node 204, new partial network candidates may exist in network NWi at time ti.

  The configuration of the new node 204 is configuration information of the new node 204, that is, information including the node ID of the new node 204, the node type, the state, OUT, and IN. When there is one new node 204, OUT and IN are not necessary. Specifically, for example, new nodes D and E are added at time t1 in FIG. The new node D indicates a power plant, and the new node E indicates a substation. The states of the new nodes D and E are both "active". Also, the new nodes D and E are connected to each other. That is, the configuration of the new nodes D and E including the node types and states of the new nodes D and E and OUT and IN is included in the search partial network Q. In this case, the determination unit 825 outputs an execution instruction to the generation unit 824.

  On the other hand, when any one of the node types of the new nodes D and E, the state, and OUT and IN is not included, there is no partial network candidate including the configuration of the new node 204 in the network NWi at time ti.

  When the determination unit 825 determines that the configuration of the new node 204 is included, the generation unit 824 applies the update information group from the oldest update information to the selected update information Ri to the snapshot. The network NWi at the update time ti of the selected update information Ri is generated. At time t2 in FIG. 1, the snapshot is the network NW0 at time t0, the oldest update information Ri is the update information R1 at time t1, and the selected update information Ri is the update information R2 at time t2. It is. Therefore, the generation unit 824 generates the network NW2 at time t2 by applying the update information R1 and R2 to the network NW0.

  In this case, the extraction unit 821 extracts a partial network candidate that includes the new node 204 and matches the search partial network Q from the network NWi generated by the generation unit 824. Specifically, for example, the extraction unit 821 extracts a partial network candidate SCb that includes the new nodes D and E and matches the search partial network Q from the network NW2 generated by the generation unit 824.

  The setting unit 823 sets a start time and an end time that define the effective period of the partial network. Specifically, for example, the setting unit 823 sets the snapshot generation time to the start time of the effective period regarding the partial network candidate extracted by the extraction unit 821. The snapshot generation time is the time when the snapshot was obtained.

  For example, in the above-described example, time t0 at which the network NW0 which is a snapshot is obtained is a snapshot generation time. In addition, the setting unit 823 sets a time t0, which is a snapshot generation time, as the start time of the validity period of the partial network candidate SCa.

  Then, based on the update content 702 of the update information Ri selected by the selection unit 822, the setting unit 823 sets the update time ti of the update information Ri as the end time of the effective period regarding the partial network candidate. Specifically, for example, when the update content 702 of the update information Ri is “deletion”, the setting unit 823 sets the update time ti of the update information Ri as the end time of the effective period regarding the partial network candidate. In this case, since the partial network candidate does not exist at the end time ti, the effective period regarding the partial network candidate is a half open interval. In the example of FIG. 1, the setting unit 823 sets the end time of the partial network candidate SCa from which the nodes A and B are deleted at time t4 to the time t4, and sets the effective period of the partial network candidate SCa to [t3, t4). Do.

  When the update content 702 of the update information Ri is "change" and the search partial network Q and partial network candidate do not match due to the change, the setting unit 823 performs the update as in the case of "deletion". The update time ti of the information Ri is set to the end time of the effective period of the partial network candidate. Also in this case, since the partial network candidate does not exist at the end time ti, the effective period regarding the partial network candidate is the half open interval. In the example of FIG. 1, the end time of the partial network candidate SCa in which the state of the nodes A and B is changed at time t1 is set to time t1, and the effective period of the partial network candidate SCa is [t0, t1).

  When the selectable update information Ri does not exist, the setting unit 823 sets the update time ti of the update information Ri selected last to the end time of the effective period of the partial network candidate. In this case, since the partial network candidate exists at the end time ti, the effective period regarding the partial network candidate is a closed segment. In the example of FIG. 1, the setting unit 823 sets the end time of the partial network candidate SCb remaining at time t4 to time t4, and sets the effective period of the partial network candidate SCb to [t2, t4].

  In addition, when the update content 702 of the update information Ri is "change" and the search partial network Q and partial network candidate match due to the change, the determination unit 825 determines the search partial network as in the case of "addition". It is determined whether Q includes the configuration of the change target node 204. When the search partial network Q includes the configuration of the change target node 204, partial network candidates may exist in the network NWi at time ti.

  Specifically, for example, the states of the nodes A and B are “stopped” at time t2 in FIG. 1, but are changed to “operating” at time t3. The new node A indicates a power plant, and the new node B indicates a substation. The states of the new nodes A and B are both "active". Also, the new nodes A and B are connected to each other. That is, the configuration of the change target node including the node type, the state, and OUT and IN of the new nodes A and B is included in the search partial network Q. In this case, the determination unit 825 outputs an execution instruction to the generation unit 824.

  When the determination unit 825 determines that the configuration of the change target node 204 is included, the generation unit 824 applies the update information group from the oldest update information R1 to the selected update information Ri to the snapshot. Thus, the network NWi at the update time ti of the selected update information Ri is generated. At time t3 in FIG. 1, the snapshot is the network NW0 at time t0, the oldest update information Ri is the update information R1 at time t1, and the selected update information Ri is the update information R3 at time t3. It is. Therefore, the generation unit 824 generates the network NW3 at time t3 by applying the update information R1 to R3 to the network NW0.

  The partial network search unit 820 associates, with the partial network candidate, the effective period of the partial network candidate for which the start time and the end time are set by the setting unit 823. Thereby, the search result 110 shown in FIG. 1 is obtained. The partial network candidates SCa and SCb obtained as the search results 110 become partial networks.

  The similar waveform search unit 830 is similar to a waveform that is a time-series observation value obtained from the sensor 206 that observes the node 204 when the node ID of the node 204 that configures the search partial network Q is specifically identified. The waveform to be generated is searched from the search result 110 of the partial network search unit 820. Specifically, for example, when the node ID (for example, X) of the left end node indicating the power plant of the search partial network Q shown in FIG. 1 is specified in the search condition, the similar waveform search unit 830 The sensor ID 501 corresponding to the node ID of X is specified from the sensor table 500, and the time-series observation value 603 is acquired from the time-series DB 600 using the specified sensor ID 501.

  The similar waveform search unit 830 also identifies the corresponding sensor ID 501 for the nodes A and D of the partial network candidates SCa and SCb that are the search results 110, and uses the identified sensor ID 501 to obtain time series from the time series DB 600 To obtain an observed value 603. Then, the similar waveform search unit 830 determines whether or not the time-series observation value 603 for the nodes A and D is similar to the time-series observation value 603 for the node X.

  Also in the case where the node ID (for example, Y) of the central node indicating the substation of the search partial network Q shown in FIG. 1 is designated by the search condition, the similar waveform search unit 830 similarly similarly outputs the node Y, The time-series observation values 603 of B and E are specified, and it is determined whether or not the time-series observation values 603 of the nodes B and E are similar to the time-series observation value 603 of the node Y.

  Also in the case where the node ID (for example, Z) of the right end node indicating the substation of the search partial network Q shown in FIG. 1 is designated by the search condition, the similar waveform search unit 830 similarly performs the node Y, The time-series observation values 603 of B and E are specified, and it is determined whether or not the time-series observation values 603 of the nodes B and E are similar to the time-series observation value 603 of the node Y.

  The similar waveform search unit 830 calculates the similarity of two waveforms (time-series observed values 603) having different cycles but similar changes using Dynamic Time Warping. The calculated distance is referred to as a DTW distance. The similar waveform search unit 830 determines that the two waveforms are similar if the DTW distance is less than or equal to a predetermined threshold.

  If the similar waveform search unit 830 determines that the two waveforms are similar, the similar waveform search unit 830 determines whether the search partial network Q and the partial network are similar. For example, when only X is specified in the search partial network Q, if the waveform of the node A of the partial network candidate SCa and the waveform of the node X are similar, it is assumed that the search partial network Q and the partial network candidate SCa are similar. to decide.

  For example, when only X and Y are specified in the search partial network Q, if the waveforms of the nodes A and B of the partial network candidate SCa and the waveforms of the nodes X and Y are similar to each other, the search partial network Q and the partial It is judged that it is similar to the network candidate SCa. If any one is not similar, the similar waveform search unit 830 determines that the search partial network Q and the partial network candidate SCa are not similar.

  For example, when X, Y, Z are specified as node IDs in the search partial network Q, the waveforms of nodes A, B, C of partial network candidate SCa and the waveforms of nodes X, Y, Z are similar to each other. Then, it is determined that the search partial network Q and the partial network candidate SCa are similar. If any one is non-similar, the similar waveform search unit 830 determines that the search partial network Q and the partial network candidate SCa are not similar.

  It should be noted that the similar waveform search unit 830 is a time-series observation value 603 of the sensor 206 that monitors the node 204 that configures the partial network within the effective period of the partial network that is the search result 110 of the partial network search unit 820. It may be acquired only from the series DB 600.

  The output unit 803 outputs the search result by the search unit 802. Specifically, for example, the output unit 803 outputs the search result 110 of the partial network search unit 820. When the similar waveform search unit 830 is executed, the output unit 803 also outputs the search result of the similar waveform search unit 830. The output format of the output unit 803 includes display on a display, transmission to an external device, print output from a printer, and projection by a projector. Alternatively, storage in the storage device 302 may be employed.

  The time-series observation value 603 of the sensor 206 monitoring the node 204 constituting the partial network in the validity period of the partial network which is the search result 110 of the partial network search unit 820 by the similar waveform search unit 830 The output unit 803 may associate the search result 110 of the partial network search unit 820 with the acquired time-series observation value 603 and output the result.

  The storage unit 804 acquires the update information Ri and stores the update information Ri in the storage device 302. The storage unit 804 includes a reception unit 841 and a storage unit 842. For example, the receiving unit 841 detects an update of the network NWi, queries an external system (for example, the power network system 202) that manages the network NWi, and receives update information Ri from the external system. The storage unit 842 stores the received update information Ri in the storage device 302.

  Also, the receiving unit 841 receives time-series data that is a time-series observation value 603 from the sensor 206. The storage unit 842 stores the received time series data in the time series DB 600 in association with the node ID of the node 204 monitored by the sensor 206 that has transmitted the time series data.

  The creation unit 805 creates a network NWk at a specific update time tk by applying a chronological update information group to the snapshot, and stores the network NWk as a subsequent snapshot in the storage device 302. For example, assuming that a specific update time tk is time t2 in FIG. 1, the creating unit 805 adds update information R1 of the update time t1 and update information R2 of the update time t2 to the network NW0 of time t0 which is a snapshot. Apply to create a new snapshot. The new snapshot is the network NW0 at time t2.

  That is, when the elapsed time until the specific update time tk and the number of the update information Ri until the specific update time tk become equal to or more than the threshold, the calculation load of network generation in the generation unit 824 becomes high, and the time for network generation is increased. It takes Therefore, when the time up to the specific update time tk and the number of pieces of update information Rk up to the specific update time tk become equal to or more than the threshold, the creating unit 805 creates a new snapshot. In the example of FIG. 1, when the specific update time tk is time t2, the search device 201 uses the network NW0 at time t0 as a snapshot until time t2, and after time t3 as a snapshot The network NW2 will be used. That is, by creating the network NW2 in advance as a snapshot, when a snapshot is required after time t3, the calculation for applying the update information R1 and R2 becomes unnecessary.

<Search processing content>
Details of search processing by the partial network search unit 820 will be specifically described with reference to FIGS. 9 to 15.

  FIG. 9 is an explanatory view showing a snapshot and extraction of partial network candidates. The network NW0 at time t0 indicates a snapshot. In the network NW0, ps # (# is a number) indicates a power plant, su # indicates a substation, and sw # indicates a switching station. Further, as in FIG. 1, the white-painted node 204 indicates the “operation” state, and the black-painted node 204 indicates the “stop” state. The extraction unit 821 extracts partial network candidates SC1 to SC4 from the snapshot. The extracting unit 821 generates a partial network candidate list 900 that manages the extracted partial network candidates SC1 to SC4.

  The partial network candidate list 900 is a table having an ID field 901, a start time field 902, an end time field 903, and a configuration information field 904. An ID field 901 is a storage area for storing the ID of the extracted partial network candidate as a value. The start time field 902 is a storage area for storing, as a value, the start time of the effective period of the extracted partial network candidate. Since the extraction source is a snapshot, all start times 902 become snapshot generation time t0.

  The end time field 903 is a storage area for storing the end time of the effective period of the extracted partial network candidate as a value. At time t0, the end time 903 is undecided. The configuration information field 904 is a storage area for storing configuration information on partial network candidates as a value. The configuration information 904 is information including the node ID of the node 204 included in the partial network candidate, the node type, and connection information (Out, In) between nodes.

  FIG. 10 is an explanatory diagram of an addition example 1 of a new node at time t1. The update information R1 indicates that at time t1, a node ps6 indicating a power plant is added, and is connected to a node su1 indicating a substation and operating. Since the update content 702 is "addition", the determination unit 825 determines whether the search partial network Q includes the configuration of the node ps6 which is the new node 204. Since the search partial network Q includes the node 204 (the node at the left end) indicating the power plant in the operating state, the generation unit 824 applies the update information R1 to the network NW0 that is a snapshot to obtain the network NW1 at time t1. Generate

  The extraction unit 821 extracts a partial network candidate that includes the new node ps6 and matches the search partial network Q from the network NW1 generated by the generation unit 824. Specifically, for example, since the search partial network Q is configured to be connected from the node 204 indicating the power plant at the left end to the node 204 two ahead, the extraction unit 821 starts from the new node ps6, The search range 1000 is set up to the node two nodes ahead two. Then, the extraction unit 821 extracts partial network candidates that match the search partial network Q within the search range 1000. In the case of this example, the extraction unit 821 extracts a partial network candidate SC5 configured by the new node ps6 and the existing nodes su1 and su2.

  FIG. 11 is an explanatory diagram of a second example of addition of a new node at time t1. FIG. 11 shows partial network candidates SC1 to SC5 and a partial network candidate list 900 at time t1. At time t1, a partial network candidate SC5 is added, and an entry of the partial network candidate SC5 is added to the partial network candidate list 900. The start time 902 of the entry is the update time t1.

  FIG. 12 is an explanatory diagram of a deletion example 1 of the existing node at time t2. The update information R2 indicates that at time t2, the node ps2 connected to the node su3 indicating a substation and indicating a power plant in operation has been deleted. Although the network NW2 at the update time t2 is shown in FIG. 12 for convenience of explanation, the network NW2 is not generated because the update content 702 is "deleted". As described above, when the update content 702 is “deletion”, the network NW2 as illustrated in FIG. 12 is not generated, and thus the load of generation of the network by the search device 201 can be reduced.

  FIG. 13 is an explanatory diagram of a second example of deletion of a new node at time t2. FIG. 13 shows partial network candidates SC3 to SC5 and a partial network candidate list 900 at time t2. Since node ps2 is deleted at time t2, partial network candidates SC1 and SC2 including node ps2 are deleted. Partial network search unit 820 registers time t2 in end time field 903 of the entry of partial network candidates SC1 and SC2 in partial network candidate list 900, and writes the entry in search result list 1300. In partial network candidate list 900, the entries of partial network candidates SC1 and SC2 are deleted.

  FIG. 14 is an explanatory diagram of a first modification of the existing node at time t3. The update information R3 indicates that at time t3 at entry 1401, the state of the node su2 connected to the nodes su1, su3 and su4 indicating the substations and indicating the power plant in operation has been changed due to the stop. Further, the update information R3 indicates that at time t3 in the entry 1402, the state is changed due to the operation of the node su6 connected to the node su4 or the like indicating the substation and indicating the substation being stopped. Although the network NW3 at the update time t3 is shown in FIG. 14 for convenience of explanation, the network NW3 is not generated because the update content 702 is "change". As described above, when the update content 702 is “changed”, the network NW3 as illustrated in FIG. 14 is not generated, so that the load on the network NWi by the search device 201 can be reduced.

  Since the state of the node su6 is changed from “stopped” to “active” at the entry 1402, the node su6 is treated equally to the newly added node 204. Specifically, for example, determination unit 825 determines whether or not search partial network Q includes the configuration of node ps 6 which is change target node 204. Since the search partial network Q includes the node 204 (right end node) indicating a substation in operation, the generation unit 824 applies the update information R1 to R3 to the network NW0, which is a snapshot, to obtain the network at time t3. Generate NW3.

  The extraction unit 821 extracts a partial network candidate that includes the new node su6 and matches the search partial network Q from the network NW3 generated by the generation unit 824. Specifically, for example, since the search partial network Q is configured to be connected from the node 204 indicating the substation at the right end to the node 204 two ahead, the extraction unit 821 starts from the new node su6, The search range 1400 is assumed to be the node two nodes ahead two. Then, the extraction unit 821 extracts partial network candidates that match the search partial network Q within the search range 1400. In the case of this example, the extraction unit 821 extracts a partial network candidate SC6 configured by the new node su6 and the existing nodes su4 and ps5.

  FIG. 15 is an explanatory diagram of a second modification of the existing node at time t3. FIG. 15 shows partial network candidates SC3 to SC6 and a partial network candidate list 900 at time t3. At time t3, since the state of the node su2 is changed, the partial network candidates SC4 and SC5 including the node su2 are deleted. Partial network search unit 820 registers time t3 in end time field 903 of the entry of partial network candidates SC4 and SC5 including node su2 in partial network candidate list 900, and writes the entry in search result list 1300. In partial network candidate list 900, entries of partial network candidates SC4 and SC5 are deleted.

  Further, the partial network candidate SC6 is added by the entry 1402, and the entry of the partial network candidate SC6 is added to the partial network candidate list 900. The start time 902 of the entry is the update time t3. At the next update time t4, when there is a state change to the deletion or "stop" for the nodes constituting the partial network candidate SC3 or SC6, or the update information R4 at the update time t4 is the last update information Ri The partial network search unit 820 registers the time t4 in the end time field 903 of the entry of the partial network candidates SC3 and SC6, and writes the entry in the search result list 1300. In partial network candidate list 900, entries of partial network candidates SC3 and SC6 are deleted.

<Example of various processing procedures>
Various processing procedure examples by the search device 201 will be described with reference to FIGS. 16 to 23.

  FIG. 16 is a flowchart of an example of a process of storing time-series data by the storage unit 804. The receiving unit 841 receives time-series data that is the time-series observation value 603 from the sensor 206 (step S1601). The storage unit 842 stores the received time series data in the time series DB 600 in association with the node ID of the node 204 monitored by the sensor 206 that has transmitted the time series data (step S1602). Thereby, time series data of each sensor 206 is accumulated.

  FIG. 17 is a flowchart of an example of a storage processing procedure of update information by the storage unit 804. The receiving unit 841 detects the update of the network NWi and receives the update information Ri from the external system (step S1701). The storage unit 842 stores the received update information Ri in the storage device 302 in association with the node ID included in the update information Ri (step S1702). Thereby, the update information Ri for each update time is accumulated.

  FIG. 18 is a flowchart showing an example of a search process procedure by the search device 201. The acquisition unit 801 acquires a search partial network Q. The acquiring unit 801 also acquires the start time and the end time of the time series data obtained from the sensor 206 that observes the equipment 205 in the node 204 constituting the search partial network Q (step S1801).

  The acquiring unit 801 uses the search partial network Q as a network search query, and generates time-series data between the start time and the end time as a time-series data search query (step S1802).

  The search unit 802 causes the partial network search unit 820 to execute partial network search processing using the search partial network Q, which is a network search query (step S1803). The partial network search process (step S1803) is the process shown in FIG. 1 and FIG. 9 to FIG. Details will be described with reference to FIGS.

  The search unit 802 executes the similar waveform search process by the similar waveform search unit 830 using the time-series data search query (step S1804). Details of the similar waveform search process (step S1804) will be described with reference to FIG. The output unit 803 outputs the search result by the search unit 802 (step S1805). Thus, the search process by the search device 201 ends.

  FIG. 19 is a flowchart of a detailed process procedure example of the partial network search process (step S1803). The partial network search unit 820 reads out from the storage device 302 the network NW0 to be a snapshot as shown at time t0 in FIG. 1 and FIG. 9 (step S1901).

  The partial network search unit 820 causes the extraction unit 821 to extract a partial network candidate that matches the search partial network Q from the snapshot, and registers the snapshot generation time t0 in the partial network candidate list 900 as the start time of the valid period. (Step S1902).

  The partial network search unit 820 causes the selection unit 822 to select update information not selected from among the time-series update information groups stored in the storage device 302 and having the oldest update time (step S1903). The partial network search unit 820 confirms the update content 702 of the selected update information Ri (step S1904).

  If the update content 702 is “addition” (step S1904: addition), the partial network search unit 820 executes an addition process (step S1905), and proceeds to step S1908. The addition process (step S1905) is a process when the new node 204 is added as described in FIGS. 10 and 11. Details of the addition process (step S1905) will be described with reference to FIG.

  If the update content 702 is "delete" (step S1904: delete), the partial network search unit 820 executes the delete process (step S1906), and proceeds to step S1908. The deletion process (step S1906) is a process when the existing node 204 is added as described with reference to FIGS. 12 and 13. Details of the deletion process (step S1906) will be described with reference to FIG.

  If the update content 702 is “change” (step S1904: change), the partial network search unit 820 executes the change process (step S1907), and proceeds to step S1908. The change process (step S1907) is a process when the state of the existing node 204 is changed as described with reference to FIGS. 14 and 15. Details of the change process (step S1907) will be described with reference to FIG.

  The partial network search unit 820 causes the selection unit 822 to determine whether there is any unselected update information Ri (step S1908). If there is unselected update information Ri (step S1908: YES), the process returns to step S1903. On the other hand, when there is no unselected update information Ri (step S1908: No), the partial network search unit 820 causes the setting unit 823 to update the end time of the partial network candidate whose end time is not set The time is set and stored in the search result list 1300 (step S1909). Thereby, the partial network candidate is determined as a partial network, and the partial network search unit 820 ends the partial network search process.

  FIG. 20 is a flowchart of a detailed process procedure example of the addition process (step S1905) illustrated in FIG. The partial network search unit 820 causes the determination unit 825 to determine whether the search partial network Q includes the configuration of the additional node 204 (step S2001). When it does not contain (step S2001: No), it transfers to step S1908. On the other hand, when the search partial network Q includes the configuration of the additional node 204 (step S2001: No), the partial network search unit 820 causes the generation unit 824 to apply the update information group up to the selected update information Ri to the snapshot. Then, the network NWi at the update time ti is generated (step S2002). For example, as shown in FIG. 10, in the update information R1 in which the update content 702 is "addition", the configuration of the additional node ps6, that is, the node type "power plant" and the state "operation" of the additional node ps6 are searched This corresponds to the leftmost node of the partial network Q. In this case, the generation unit 824 generates the network NW1 at the update time t1.

  The partial network search unit 820 causes the extraction unit 821 to set a search range starting from the additional node 204 in the generated network NWi (step S2003). The partial network search unit 820 causes the extraction unit 821 to search a partial network that matches the search partial network Q and that includes the additional node 204 in the set search range (step S2004). For example, as illustrated in FIG. 10, the extraction unit 821 extracts a partial network candidate SC5 including the configuration of the additional node ps6 from the search range 1000.

  If there is no partial network that matches (step S2005: No), the process proceeds to step S1908. On the other hand, if there is a matching partial network (step S2005: Yes), the partial network search unit 820 extracts the partial network and associates the update time ti of the update information Ri with the extracted partial network as the start time of the effective period. Are registered in the partial network candidate list 900 as partial network candidates (step S2006). For example, as shown in FIG. 11, the setting unit 823 associates the partial network candidate SC5 with the update time t1 of the update information R1 as the start time of the effective period, and registers the partial network candidate in the partial network candidate list 900. Thus, the addition process (step S1905) ends, and the process proceeds to step S1908.

  FIG. 21 is a flowchart showing a detailed processing procedure example of the deletion process (step S1906) shown in FIG. The partial network search unit 820 causes the determination unit 825 to determine whether the deletion target node 204 is the node 204 in the partial network candidate (step S2101). When the deletion target node 204 is not the node 204 in the partial network candidate (step S2101: No), the process proceeds to step S1908.

  On the other hand, if the deletion target node 204 is the node 204 in the partial network candidate (step S2101: YES), the partial network search unit 820 causes the setting unit 823 to delete the update time ti of the selected update information Ri. The end time of the effective period regarding the partial network candidate including the node 204 is set (step S2102). For example, as shown in FIG. 12, the setting unit 823 ends the effective period of the partial network candidates SC1 and SC2 including the deletion target node ps2 in the partial network candidate list 900 with the update time t2 of the selected update information R2. It is set to time 903.

  The partial network search unit 820 causes the setting unit 823 to save the entry of the partial network candidate for which the end time 903 is set in the partial network candidate list 900 in the search result list 1300 and delete it from the partial network candidate list 900 (step S2103). For example, as illustrated in FIG. 13, the setting unit 823 stores entries of the partial network candidates SC1 and SC2 in which the update time t2 is set to the end time 903 in the partial network candidate list 900 in the search result list 1300 It is deleted from the candidate list 900. Thus, the deletion process (step S1906) is completed, and the process proceeds to step S1908.

  FIG. 22 is a flowchart of a detailed process procedure example of the change process (step S1907) depicted in FIG. The change process (step S1907) is a process combining the deletion process (step S1906) and the addition process (step S1905). Specifically, for example, the processing of steps S2201 to S2203 corresponds to the deletion processing (step S1906), and the processing of steps S2204 to S2209 corresponds to the addition processing (step S1905).

  The partial network search unit 820 causes the determination unit 825 to determine whether the change target node 204 is the node 204 in the partial network candidate (step S2201). When the change target node 204 is not the node 204 in the partial network candidate (step S2201: No), the process proceeds to step S1908.

  On the other hand, if the change target node 204 is the node 204 in the partial network candidate (step S2201: YES), the partial network search unit 820 causes the setting unit 823 to change the update time ti of the selected update information Ri The end time of the effective period of the partial network candidate including the node 204 is set (step S2202). For example, as shown in FIG. 14, the setting unit 823 sets the update time t3 of the entry 1401 of the selected update information R3 to the effective period regarding partial network candidates SC4 and SC5 including the change target node su2 in the partial network candidate list 900. Set to the end time 903 of.

  The partial network search unit 820 causes the setting unit 823 to save the entry of the partial network candidate for which the end time 903 is set in the partial network candidate list 900 in the search result list 1300 and delete it from the partial network candidate list 900 (step S2203). For example, as shown in FIG. 15, the setting unit 823 stores entries of the partial network candidates SC4 and SC5 for which the update time t3 is set to the end time 903 in the partial network candidate list 900 in the search result list 1300, and the partial network It is deleted from the candidate list 900.

  The partial network search unit 820 causes the determination unit 825 to determine whether the search partial network Q includes the configuration of the change target node (step S2204). When it does not contain (step S22204: No), it transfers to step S1908. On the other hand, when the search partial network Q includes the configuration of the change target node 204 (step S2204: No), the partial network search unit 820 causes the generation unit 824 to create the snapshot, update information group up to the selected update information Ri. Applying, the network NWi of the update time ti is generated (step S2205). For example, as shown in FIG. 14, in the entry 1402 of the update information R3 whose update content 702 is "change", the configuration of the change target node su6, that is, the node type "substation" of the change target node su6 and the state " “Operation” corresponds to the central or leftmost node of the search partial network Q. In this case, the generation unit 824 generates the network NW3 at the update time t3.

  The partial network search unit 820 causes the extraction unit 821 to set a search range starting from the change target node 204 in the generated network NWi (step S2206). The partial network search unit 820 causes the extraction unit 821 to search a partial network that matches the search partial network Q and that includes the change target node 204 in the set search range (step S2207). For example, as illustrated in FIG. 14, the extraction unit 821 extracts a partial network candidate SC6 including the configuration of the change target node su6 from the search range 1400.

  If there is no partial network candidate that matches (step S2208: NO), the process proceeds to step S1908. On the other hand, when there is a matching partial network candidate (step S2208: Yes), the partial network search unit 820 extracts the partial network candidate, and the extracted partial network candidate has the update time ti of the update information Ri in the valid period. It associates with the start time and registers it in the partial network candidate list 900 (step S2209). For example, as shown in FIG. 15, setting unit 823 associates partial network candidate SC6 with update time t3 in entry 1402 of update information R3 as the start time of the effective period, and registers it in partial network candidate list 900. Thus, the change process (step S1907) is completed, and the process proceeds to step S1908.

  FIG. 23 is a flowchart showing a detailed processing procedure example of the similar waveform search process (step S1804) shown in FIG. The similar waveform search unit 830 identifies, from the configuration information, the node ID of the node that configures the partial network that is the search result of the search result list 1300, and searches the sensor table 500 (step S2301).

  The similar waveform search unit 830 searches the time series DB 600 for time series data which is time series observation values observed by the sensor 206 of the searched sensor ID in the effective period of the partial network in the time zone of the time series data search query. (Step S2302). The similar waveform search unit 830 searches the time series DB 600 as time-series data having a threshold value or less from the searched time-series data as a similar waveform based on a certain distance scale, for example, the dynamic time scaling method (step S2303) ).

  The similar waveform search unit 830 associates each search result in the search result list 1300 with the information related to the searched similar waveform, and rearranges them in ascending order of the distance scale (step S2304). The information on the similar waveform includes a time scale observation value 603 which is a similar waveform and a distance scale with time series data which is a time series observation value observed in the effective period of the partial network, for example, a DTW distance. It is information. Thus, the similar waveform search unit 830 ends the similar waveform search process (step S1804).

  As described above, the search device 201 according to the present embodiment is given as a search key using a snapshot indicating the network configuration at a certain time of the network, which changes from moment to moment, and update information indicating an update history after that time. A search partial network Q is searched for which period is effectively present in the network.

  By using the snapshot and the update information Ri, generation of a snapshot at each time can be suppressed, and the search speed can be increased. Also, with this, it is possible to reduce the load of snapshot generation in the search device 201. Further, by specifying the existence period of the partial network that matches the search partial network Q, the search accuracy can be improved. Therefore, by using the search result obtained from the search device 201, the administrator can achieve speeding up and high reliability of decision making on the network to be searched.

  In addition, the generation of the network NWi at a certain update time ti is minimized. Specifically, for example, when deleting a node and when it is excluded from the search partial network Q due to a change in the state of the node, the network NWi is not generated. This can reduce the calculation load at the time of partial network search.

  In addition, when a predetermined time has elapsed from the time of snapshot generation (t0), or when a predetermined number of update information Ri is obtained, the search device 201 uses the update information group up to the latest update information Rk. Create a network NWk to be a snapshot. Then, the search device 201 uses the network NWk instead of the snapshot of the network NW0 for the update information Rm (m> k) and thereafter. As a result, when the network NWm at the update time tm is generated, calculation for applying the update information R0 to Rk becomes unnecessary, and the generation of the network NWm can be speeded up.

  Also, if there are multiple snapshots, the partial network search process (step S1803) for each snapshot may be executed by different threads, different processes, or different servers. As a result, load distribution of partial network search processing (step S1803) can be achieved, and speeding up of partial network search processing can be achieved.

  The present invention is not limited to the embodiments described above, and includes various modifications and equivalent configurations within the scope of the appended claims. For example, the embodiments described above are described in detail to explain the present invention in an easy-to-understand manner, and the present invention is not necessarily limited to those having all the configurations described. Also, part of the configuration of one embodiment may be replaced with the configuration of another embodiment. In addition, the configuration of another embodiment may be added to the configuration of one embodiment. In addition, another configuration may be added to, deleted from, or replaced with part of the configuration of each embodiment.

  In addition, each configuration, function, processing unit, processing means, etc. described above may be realized by hardware, for example, by designing part or all of them with an integrated circuit, etc., and the processor realizes the respective functions. It may be realized by software by interpreting and executing the program to

  Information such as a program, a table, and a file for realizing each function can be stored in a memory, a hard disk, a storage device such as a solid state drive (SSD), or a recording medium such as an IC card, an SD card, or a DVD.

  Further, control lines and information lines indicate what is considered to be necessary for explanation, and not all control lines and information lines necessary for mounting are shown. In practice, it can be considered that almost all configurations are mutually connected.

Claims (15)

A search device for searching a network in which a plurality of nodes constituting a node group are connected,
The search device includes a processor that executes a program, and a storage device that stores the program.
The storage device stores snapshot information including a snapshot and the generation time of the snapshot, and adds a new node in the network, changes the state of the node, or deletes the node. Storing a time-series update information group including one or more update information including the update content indicating and the update time of the addition, change, or deletion;
The snapshot includes connection information of the nodes in the network, and node information holding at least one of the type or the state of each node of the nodes;
The processor is
Acquisition processing for acquiring a search partial network;
When the search partial network is acquired by the acquisition process, the snapshot and the specific update are performed using the snapshot and update information up to a specific update time in the time-series update information group Search processing for searching a partial network that matches the search partial network from update information up to time, and associating, with the partial network, a valid period from the search for the partial network until the search is no longer performed;
An output process for outputting a search result by the search process;
A search device characterized by performing. The search device according to claim 1, wherein
The processor is
A search apparatus which executes storage processing for acquiring the update information and storing it in the storage device. The search apparatus according to claim 1 or 2, wherein
The processor is
Extracting processing for extracting a first partial network candidate that matches the search partial network from the snapshot;
A selection process of selecting the oldest update information not selected from the time-series update information group;
The generation time is set to the start time of the effective period of the first partial network candidate extracted by the extraction process, and based on the update content of the update information selected by the selection process, Performing a setting process of setting an update time to an end time of an effective period of the first partial network candidate,
In the search process, the processor relates the validity period of the first partial network candidate for which the start time and the end time are set by the setting process to the first partial network candidate. apparatus. The search device according to claim 3, wherein
In the setting process, the processor sets an update time of the specific update information to an end time of a validity period of the first partial network candidate, when the unselected update information which has not been selected yet exists. Search device characterized by. The search device according to claim 3, wherein
The processor is
A determination process of determining whether the search partial network includes the configuration of the new node when the update content of the update information selected by the selection process is the addition of the new node;
When it is determined by the determination process that the configuration of the new node is included, the update information group from the oldest update information to the selected update information is applied to the snapshot to select the selected update information Execute generation processing for generating a network at the update time,
In the extraction process, the processor extracts a second partial network candidate that includes the new node and matches the search partial network from the network generated by the generation process.
In the setting process, the processor sets an update time of the selected update information as a start time of an effective period of the second partial network candidate extracted by the extraction process. The search device according to claim 5, wherein
In the setting process, the processor sets the update time of the update information based on the update content of the update information selected after the setting of the start time of the effective period regarding the second partial network candidate, to the second part. Set to set the end time of the effective period for the network candidate,
In the search process, the processor relates the validity period of the second partial network candidate for which the start time and the end time are set by the setting process to the second partial network candidate. apparatus. The search device according to claim 3, wherein
The processor is
If the update content of the update information selected by the selection process is deletion of the node, a determination process is performed to determine whether the deletion target node is included in the first partial network candidate,
In the setting process, when it is determined by the determination process that the processor includes the deletion target node, the processor is configured to update the selected update information with the first partial network including the deletion target node. A search apparatus characterized in that it is set to an end time of an effective period regarding a candidate. The search device according to claim 3, wherein
The processor is
When the update content of the update information selected by the selection process is the change of the state of the node, a determination process is performed to determine whether the change target node of the state is included in the first partial network candidate. Run
In the setting process, when it is determined that the change target node of the state is included in the setting process, the processor includes an update time of the selected update information including the change target node of the state. A search apparatus characterized in that it is set to the end time of the effective period regarding one partial network candidate. The search device according to claim 3, wherein
The processor is
A determination process of determining whether the change target node of the state is included in the search partial network if the update content of the update information selected by the selection process is a change of the state of the node;
When it is determined that the change target node in the state is included in the determination processing, the snapshot is selected by applying an update information group from the oldest update information to the selected update information to the snapshot. Execute generation processing for generating a network at the update time of the updated information;
In the extraction process, the processor extracts a second partial network candidate that includes the change target node in the state and matches the search partial network from the network generated by the generation process.
In the setting process, the processor sets an update time of the selected update information as a start time of an effective period of the second partial network candidate extracted by the extraction process. The search apparatus according to claim 1 or 2, wherein
The storage device stores time-series observation values from a sensor that observes an observation target corresponding to the node,
In the search process, the processor acquires, from the storage device, time-series observation values of nodes configuring the partial network within the valid period.
In the output process, the processor outputs, together with the partial network, time-series observation values of nodes constituting the partial network within the effective period. The search apparatus according to claim 1 or 2, wherein
The storage device stores time-series observation values from a sensor that observes an observation target corresponding to the node,
In the search process, the processor acquires, from the storage device, time-series observation values of the nodes forming the partial network within the effective period, and the time-series observation values of the nodes forming the partial network; Searching a partial network similar to the search partial network based on time-series observation values of nodes constituting the search partial network within the valid period;
In the output processing, the processor outputs a partial network similar to the search partial network. The search device according to claim 1, wherein
A search apparatus characterized by creating a network at the specific update time by applying the time-series update information group to the snapshot, and executing creation processing to be stored in the storage device as a subsequent snapshot. . The search apparatus according to claim 12, wherein
The processor is
A search apparatus characterized in that the creation process is executed when the number of the time-series update information groups becomes equal to or more than a predetermined number. A search method executed by a search apparatus for searching a network in which a plurality of nodes constituting a node group are connected,
The search device includes a processor that executes a program, and a storage device that stores the program.
The storage device stores snapshot information including a snapshot and the generation time of the snapshot, and adds a new node in the network, changes the state of the node, or deletes the node. Storing a time-series update information group including one or more update information including the update content indicating and the update time of the addition, change, or deletion;
The snapshot includes connection information of the nodes in the network, and node information holding at least one of the type or the state of each node of the nodes;
The processor is
Acquisition processing for acquiring a search partial network;
When the search partial network is acquired by the acquisition process, the snapshot and the specific update are performed using the snapshot and update information up to a specific update time in the time-series update information group Search processing for searching a partial network that matches the search partial network from update information up to time, and associating, with the partial network, a valid period from the search for the partial network until the search is no longer performed;
An output process for outputting a search result by the search process;
A search method characterized by performing. A search program for causing a processor to search a network in which a plurality of nodes constituting a node group are connected,
The processor is accessible to the storage device;
The storage device stores snapshot information including a snapshot and the generation time of the snapshot, and adds a new node in the network, changes the state of the node, or deletes the node. Storing a time-series update information group including one or more update information including the update content indicating and the update time of the addition, change, or deletion;
The snapshot includes connection information of the nodes in the network, and node information holding at least one of the type or the state of each node of the nodes;
In the processor,
Acquisition processing for acquiring a search partial network;
When the search partial network is acquired by the acquisition process, the snapshot and the specific update are performed using the snapshot and update information up to a specific update time in the time-series update information group Search processing for searching a partial network that matches the search partial network from update information up to time, and associating, with the partial network, a valid period from the search for the partial network until the search is no longer performed;
An output process for outputting a search result by the search process;
A search program characterized in that it performs.

Images