JP3961350B2 - Training equipment - Google Patents

Description

[0001]
BACKGROUND OF THE INVENTION
The present invention relates to a training apparatus used for disaster prevention training or the like.
[0002]
[Prior art]
In recent years, interest in nuclear disaster prevention has increased, and training for electric power companies as well as disaster prevention drills organized by the national government, including local governments, electric power companies, residents, and others have been implemented. In 2000, the Nuclear Disaster Prevention Measures Law was enacted, and laws on nuclear disaster prevention have been established.
[0003]
In order to carry out nuclear disaster prevention drills more practically, it is necessary to simulate plant events, but the actual plant behavior is long and does not fall within the normal training time. It is common practice to create data that matches the training time, such as by deleting.
[0004]
The training data for disaster drills is edited, plant parameters are displayed, and used for disaster drills. The Japan Atomic Energy Society "2001 Autumn Conference" draft (H40) "Nuclear Disaster Prevention" “Development of Training Operation Parameter Display Program” discloses a technique for displaying data in real time based on analysis data based on MAAP4 code.
[0005]
[Problems to be solved by the invention]
However, in the above-mentioned technology, when omitting part of the event progress in accordance with the training time, the data is deleted while looking at the numerical data. It is not suitable for simultaneous training at a plurality of locations because it does not correspond to the above, and no means for displaying the same data at a remote location is provided.
[0006]
Further, in the analysis code, even a value in a range that cannot be measured in an actual plant is output by calculation. In addition, it is impossible to simulate a situation in which a data value cannot be acquired due to a failure of a device due to an event in which the device is not operating normally in a plant emergency. Therefore, even if disaster drills are conducted using the analysis data, there is a possibility that the actual situation will be different.
[0007]
Therefore, in order to solve the above-described problems, the present invention creates training data, generates data, and improves the efficiency of the display method. Also, considering the reliability of data values, the training is performed more practically. It is an object of the present invention to provide a training device that can perform such a task.
[0008]
[Means for Solving the Problems]
  The present invention achieves the above object, and the invention according to claim 1In a training apparatus comprising a device basic unit and a terminal device connected to the device basic unit via a communication line,
  The device basic part is:A scenario data storage unit for storing scenario data in which data values of a plurality of parameters representing plant behavior obtained by analysis are recorded in time series;
  Stored in the scenario data storage unitAdded information indicating reliability index of data value to the scenario dataCreate training dataA reliability information setting unit;
  By the reliability information setting unitInformation indicating the reliability indexTheCreated by adding to the scenario dataIsTheSaidA training data storage unit for storing training data;
  A display unit for displaying at least a part of the training data together with information indicating the reliability index;
  A display condition setting unit for setting an update condition of the scenario data to be displayed on the display unit;
  Display contents displayed on the display unitA display data distribution unit that performs transmission to the terminal device and reception of an update condition transmitted from the terminal device via the communication circuit,
  The display condition setting unit sets, as the update condition, one of fast-forwarding, thinning-out, and skipping as a request type based on a setting operation from a user, and processing by the processing method set as the request type. Set the time period of the training data to be targeted by the time series number of the training data,
  The display unit is selected from the training data stored in the training data storage unit based on the request type set as the update condition by the display condition setting unit and the period of the training data. Get what to display, display this acquired training data,
  The terminal device receives display content displayed on the display unit transmitted from the display data distribution unit via the communication line, displays the received display content, and from the user in the terminal device Transmitting an update condition set based on the setting operation to the display data distribution unit via the communication line as information on the update condition set by the display condition setting unit,It is characterized by providing.
[0009]
According to the invention of claim 1, by adding the reliability information of the data value to the scenario data created by the analysis code, it becomes closer to the process value obtained from the actual plant, and training closer to practice is performed. Is possible.
[0010]
The invention according to claim 2 is the training apparatus according to claim 1, wherein the reliability index is determined to be unreliable when the data value is outside a predetermined range. It is characterized by. According to the invention of claim 2, in addition to the action and effect of the invention of claim 1, it is possible to eliminate an abnormal value as a data value and perform training closer to practice.
[0014]
  Claims3The display condition setting unit is configured to change an update condition of the scenario data when the scenario data is displayed on the display unit. Claim3According to the invention, since the display speed of the training data can be arbitrarily changed according to the situation of the place, not only can abrupt change during training be flexibly handled, but also, for example, preparation before training. When confirming the display of training data at the stage, it is also possible to efficiently confirm the preliminary preparation stage while displaying the part to be noted at a low speed and displaying the gentle part of the behavior at a high speed.
[0015]
  Claims4The display condition setting unit can be set in a state where a trend graph of the scenario data is displayed on the display unit. Claim4According to this invention, it is possible to easily perform settings such as fast turning and thinning in accordance with changes in plant behavior.
The
[0016]
  Claims5In the invention described in the above, the display condition setting unit selects a maximum value and a minimum value of each period for each parameter from the scenario data stored in the time series in terms of a time interval for each predetermined interval. It can be set as described above. Claim5According to this invention, it is possible to perform data thinning in consideration of both the upper limit value and the lower limit value without changing the overall thinning rate of the data.
[0017]
  Claims6In the invention described in the above, the display condition setting unit divides a time period for each predetermined interval from the scenario data stored in the time series, and one specific parameter of the plurality of parameters is It is characterized in that it is possible to select a timing at which a maximum value and a minimum value are taken in a period, and to select data of all parameters displayed on the display unit at the selected timing.
[0018]
    Claim6According to the invention, when thinning out data, paying attention to a specific parameter that matches the characteristics of the event, the data is extracted and thinned according to the time position indicating the upper limit value and lower limit value of the parameter. Thus, it is possible to thin out the data reflecting the characteristics of the event without impairing the mutual relationship between the parameters at the same time.
[0023]
DETAILED DESCRIPTION OF THE INVENTION
A configuration example of an embodiment of a training apparatus according to the present invention will be described with reference to FIG. The apparatus basic unit 20 includes a scenario data storage unit 1, a reliability information setting unit 2, a training data storage unit 3, a data display unit 4, a display condition setting unit 5, a peak value determination unit 6, a display data storage unit 7, and a display. A data distribution unit 8 and a password registration unit 10 are included.
[0024]
The display data distribution unit 8 is connected to a router 22, and the router 22 is connected to the terminal device 9 via a LAN or a communication line such as a line network or the Internet. Functions of the display data distribution unit 8, the router 22, the terminal device 9, and the like will be described later.
[0025]
The scenario data storage unit 1 stores scenario data. The scenario data is time-series data indicating plant behavior created using, for example, a plant analysis code. The reliability information setting unit 2 reads the scenario data, sets information indicating the reliability index of the data value according to the user's request and sets it as training data, and stores this in the training data storage unit 3.
[0026]
Configuration examples of scenario data and training data are shown in FIGS. 2 and 3, respectively.
The information indicating the reliability index is the reliability information shown in FIG. 3, for example, 0 (zero) when normal, 1 when deviating from the upper and lower limits, and when measurement cannot be performed due to failure of the sensor itself, etc. Is set to the data value in the form of 2.
[0027]
For example, on the dialog screen shown in FIG. 4A, the user selects whether to set the upper / lower limit value or the period. When setting is selected using the upper and lower limit values, as shown in FIG. 4B, the upper and lower limit values are entered in the table for each parameter. When each parameter exceeds the input upper and lower limit values, the reliability information of the data value is set to 1 (deviation from the upper and lower limits), and this is set as “bad”, otherwise 0 (normal) ).
[0028]
When setting by a period, as shown in FIG.4 (c), the start time series number and the end time series number are input into a table | surface for every parameter. In the range between the input start time series number and end time series number, the reliability information of the data value is set to 2 (defect), otherwise it is set to 0 (normal).
In addition, when data conversion etc. are required between scenario data and training data, it converts into the data suitable for a display in the reliability information setting part 2. FIG.
[0029]
In general, in the analysis code, even a value in a range that cannot be measured in an actual plant is output by calculation. On the other hand, it is impossible to simulate a situation in which a data value cannot be acquired due to a failure of a device or the like even in an event where the device is not operating normally in a plant emergency. For this reason, even if disaster prevention training is conducted using the analysis data, it may be different from the actual situation. According to the above embodiment, by adding the reliability information of the data value to the scenario data created by the analysis code, it becomes closer to the process value obtained from the actual plant, and training closer to practice can be performed. It becomes possible.
[0030]
Next, functions of the display condition setting unit 5 and the like of the embodiment of the training apparatus shown in FIG. 1 will be described. The data display unit 4 reads the content of the training data from the training data storage unit 3 in accordance with a command from the display condition setting unit 5 and performs display while periodically updating the data. The data display unit 4 has a screen necessary for grasping the state of several types of plants. FIG. 5A is an example of a screen for selecting the main parameters of the plant and displaying them as a trend graph. FIG. 5B is an example of a screen for displaying the main parameters of the plant as a summary.
For example, the user sets display conditions on the dialog screen shown in FIG. 6, and the data display unit 4 displays data according to the set conditions.
[0031]
First, the time condition of training data is set on the data time condition setting screen shown in FIG. Here, the data head time is the time of the first time-series data of the training data, and the unit time indicates the data interval between the first and second data. In the example shown in FIG. 6A, the data head time is 2001/10/20 00:00:00 and the unit time is 60 seconds. In this case, the second data of the training data is interpreted as 2001/10/200 00:01:00. The data start time is the default, and the internal clock time of this device is preset.
[0032]
Next, a display method is selected for the data time displayed on the screen. Select whether to display the internal clock as it is or to display the data time according to the time set in the data time condition setting of the training data.
[0033]
Next, designation of data reading conditions will be described with reference to FIG. Here, the time range for setting the conditions is set by the registration start time series number and the registration end time series number, and the processing method is designated from among fast-forwarding, thinning-out, skipping reading, and stopping reading.
[0034]
In the case of fast rotation, when the update speed is selected between 1 to 10 times speed and the registration button is pressed (clicked), the set condition is registered. In the selection of the double speed, the selection range is limited so that the maximum double speed is, for example, 5 seconds or more in relation to the unit time set in advance in consideration of the display update load.
[0035]
In the case of thinning, each parameter is read by skipping data in the time series direction by the specified numerical value. For example, when 5 is specified, every five pieces of data are read, such as time series numbers 0, 5, 10,. When skipping is specified, data between the specified time series numbers is skipped. When reading stop is designated, reading is stopped with the data designated as the registration start time series number as the last. It is set as 1 × speed during the time series number that the user has not specified anything.
[0036]
The contents set by designating the data reading conditions can be stored in a predetermined number in the display condition setting unit 5, and the setting contents can be reused from the next time. Data reading conditions can also be set online during data reading. The current time series number is displayed on the dialog screen so that the progress can be grasped when setting online while reading data.
[0037]
Next, the data reading start request will be described with reference to FIG. The data reading start can be selected either immediately after the request is made or by specifying the reading start time as a timer.
The display condition setting unit 5 calculates the update cycle from the display conditions and passes it to the data display unit 4 together with the set display conditions.
[0038]
An example of information passed to the data display unit 4 is shown in FIG. First, the data start time and display time are specified. For example, when the display time is 1, the display time is set to the internal clock, and 2 is set to the data time. The reading start time is passed in the form of the time when the request is executed even if the current reading is designated. There are a start time series number, an end time series number, a request type, a data reading interval, and an update cycle. The request type is, for example, 1 for fast turn, 2 for thinning, -1 for skipping, and 99 for reading stop.
[0039]
In the case of fast rotation, the update cycle is
Update cycle = unit time / double speed
In the case of thinning,
Update cycle = unit time
It becomes.
The data reading interval specifies how many times data is read, and is set to 1 for fast-forwarding and set to thinning for thinning-out.
[0040]
The data display unit 4 periodically reads and displays the data in the training data storage unit 3 according to the conditions at the update cycle time. This detailed procedure will be described below with reference to FIG.
[0041]
In step S1, the display condition input by the user passed from the display condition setting unit 5 is read. In step S2, the data reading start time is determined. If the data reading start time is zero, the process immediately proceeds to step S5 to read training data.
[0042]
If the data reading start time is not zero in step S2, that is, if the data reading start time is designated as a future time, the time is periodically checked in step S3, and the time is after the designated time. In other words, when the internal time becomes larger than the data reading start time, the process proceeds to step S5, and the training data is read from the training data storage unit 3.
[0043]
After reading the training data, display conditions are read in step S6. Here, when the training data ends, the process proceeds from step S7 to the end. If the training data has not ended, a request is made in steps S8 and S9 by comparing the time series number of the data with the start time series number and end time series number of the display condition data passed from the display condition setting unit 5. Determine the type.
[0044]
If the request type is “stop” in step S9, the process immediately ends. If the request type is “skip skip” in step S9, an operation of advancing the pointer to the designated time series number is performed in step S10, and then the process returns to step S5.
[0045]
If the request type is any one of “decimation”, “fast turn”, and “1 × speed” in step S9, screen display is performed in step S11. After that, in step S12, the reading cycle time is waited, then the pointer is advanced in step S13, and then the process returns to step S5.
[0046]
The content of the screen display is based on the content set by the display condition setting unit 5, and the data time is displayed on the screen. When the data time is set to the internal time, the actual time of the internal clock is displayed regardless of whether the data is thinned out or fast-forwarded. When adjusting to the data, both fast and thinning,
Start time + unit time x serial number
It is represented by
[0047]
As a thinning method when thinning data, there is a method of adopting a peak value in addition to thinning at a predetermined interval for each specified number. The peak value determination unit 6 determines the peak value in this case.
[0048]
The peak value specification method includes, for example, a method of picking up the maximum and minimum values within the period for each parameter, and paying attention to one of the parameters, and showing the maximum and minimum values within the range of the parameter. There is a method to read all parameters according to the time series number. Select one of the methods to determine the peak value.
[0049]
Here, the maximum and minimum extraction methods will be described. When the thinning interval S is designated, the peak value determination unit extracts the maximum and minimum values from S data, and the thinning rate becomes S / 2. In order to determine the maximum and minimum peak values without changing the thinning rate, the data extraction is performed by determining the maximum value and the minimum value for S × 2 data. For example, if the set value of thinning is 5, the maximum and minimum extraction is determined in 10 data periods. When the same value overlaps with a plurality of time series numbers, for example, the smaller time series number is selected.
[0050]
An example of the processing procedure of the maximum and minimum determination methods will be further described with reference to the flowchart of FIG. In step S31, a thinning interval S is set, and training data is read in step S32. Next, the count value i is set to 1 in step S33, and the value of n + i is set to MAX and MIN in step S34. Here, n is a time series number.
[0051]
Next, i is increased by 1 in step S35, and the n + i-th training data is read in step S36. In step S37, MAX is compared with the value of the (n + i) th training data. If the former is smaller, MAX is replaced with the (n + i) th training data in step S38, and MAXn is set to n. Next, in step S39, MIX is compared with the value of the (n + i) th training data. If the former is larger, in step S40, MIX is replaced with the value of the (n + i) th training data, and MIXn is set to n.
[0052]
Next, in step S41, i and S × 2 are compared. If the former is smaller, the process returns to step S35, and otherwise, the process proceeds to step 42 to wait for the next activation command. When the maximum and minimum values are calculated for all parameters, the maximum and minimum determination processes are repeated for the parameters.
[0053]
Next, a method of reading all the parameters in accordance with the time series number indicating the maximum and minimum values within the parameter range will be described by paying attention to any one parameter. The peak value determination unit 6 reads S (decimation interval) × 2 pieces of data to be read next for the designated parameter from the training data storage unit 3, and sets time series numbers having upper and lower peaks. Determine and notify the data display unit 4. The data display unit 4 extracts and displays all parameters according to the notified time series number.
[0054]
For example, the procedure shown in FIG. 9 is performed for one designated parameter, and the time series numbers of MAXn and MINn determined thereby are passed to the data display unit 4. The data display unit 4 receives MAXn and MINn from the training data storage unit 3. The data of the position of is read for all parameters.
Data read and displayed by the data display unit 4 can be recorded in the data storage unit 7 and reused.
[0055]
In the display condition setting unit 5, for example, as shown in FIG. 10, the period can be set by specifying the registration start time and the registration end time on the trend graph screen. When the “designate with graph” button is pressed (clicked) on the data reading method designation dialog screen of FIG. 6, the screen (window) of FIG. 10 is displayed in an overlapping manner. When the cursor of the graph is brought to the place where registration is desired and the start number determination or end number determination button is pressed, the period is set to the registration start time series number and registration end time series number of FIG.
[0056]
As described above, according to this embodiment, by arbitrarily setting the display speed of the data in a plurality of arbitrary periods, the portion having a large behavior change is matched to the plant behavior change without missing the data. Since a low-speed portion with little change can be displayed at high speed, disaster prevention training can be performed efficiently and effectively even when the training time is short.
[0057]
In addition, since the display speed of training data can be arbitrarily changed online according to the situation at the site, not only can abrupt changes during training be flexibly handled, but also training can be performed at the pre-training stage. When confirming the display of data for use, it is possible to efficiently confirm the advance preparation stage while advancing the display at a low speed for the part to be noted and at a high speed for the gentle part of the behavior.
[0058]
Furthermore, by arbitrarily setting the data thinning interval in a plurality of arbitrary periods, it is possible to thin out or skip reading according to the behavior of the plant rather than uniformly, so that the unit time of training data is, for example, 1 second. Even in a short case, by thinning out the data, it is possible to smoothly update the screen without increasing the display load.
[0059]
Moreover, by thinning out according to the peak value of a specific parameter that is a point of a plant event, it is possible to perform data thinning without impairing the characteristics of the event. In addition, by changing the data thinning interval online, it is possible not only to respond to sudden changes in the disaster prevention training day, but also to adjust the thinning interval while viewing the display screen. The stage can be checked efficiently.
[0060]
Furthermore, by setting while observing the behavior of the plant in the graph, it is possible to easily perform settings such as fast turning and thinning in accordance with changes in the plant behavior. Also, when thinning out data, by extracting and thinning out the upper limit value and lower limit value at an interval twice the thinning interval, both the upper limit value and the lower limit value are changed without changing the overall thinning rate of the data. Can be thinned out in consideration of the above.
[0061]
Also, when thinning out data, pay attention to a specific parameter that matches the characteristics of the event, and by extracting and thinning out the data according to the time position indicating the upper limit value and lower limit value of the parameter, the same It is possible to thin out data reflecting the characteristics of the event without damaging the mutual relationship of the parameters at the time.
[0062]
In addition, it is easy to understand when changing the display conditions online by specifically displaying the progress of reading the scenario data on an interactive screen for setting display conditions by number, position, and the like.
[0063]
Next, functions of the display data distribution unit 8, the password registration unit 10, the router 22, the communication line 24, the terminal device 9, and the like illustrated in FIG. 1 will be described.
The display data distribution unit 8 distributes the screen displayed on the data display unit 4 via the web, and distributes the data on the terminal device 9 having a browser function connected via a communication line 24 such as a LAN, a public line, or the Internet. The displayed screen can be referred and the display condition can be set.
[0064]
In order to connect the terminal device 9, the terminal device 9 that can be connected is limited in advance by, for example, registering the telephone number of the other party in the router 22 or the like, and the password is received from the password registration unit 10. It is necessary to register in advance.
[0065]
Furthermore, in order to set display conditions from the terminal device 9, it is necessary to register a dedicated password. When trying to open the display condition setting dialogue screen from a menu or the like, the user ID and password must be entered. Prompt. A user is identified by a password, and display condition setting operations are not permitted for users whose passwords do not match. For the password registration, an interactive function for registration is provided so that the deletion and change can be easily performed. In addition, a password is given a time limit so that passwords used in a past fixed number of times, for example, the past 1000 times cannot be registered.
[0066]
Thus, by displaying on the terminal device 9 at a remote location, the same data can be displayed from anywhere in the location where the training is participated using the existing line infrastructure, and the progress of the plant behavior is also possible at a remote location. Therefore, effective training can be performed. In addition, since the display setting can be performed at a remote place, the functions of the apparatus can be used at various places without having to move the basic unit 20 to the place of use for every training. In addition, by providing the user with the use of the training device function, the user can use the function of the present device only with the terminal device 9 without having the device basic unit 20.
[0067]
【The invention's effect】
  Main departureAccording to Ming, by adding the reliability information of the data value to the scenario data created with the analysis code, it becomes closer to the process value obtained from the actual plant, and it becomes possible to perform training closer to practice .
[0068]
  In addition, this departureAccording to Ming, by setting the data display speed arbitrarily, for example, when the training time is short, the part with a large change in behavior is displayed at a low speed and the part with a small change is displayed at a high speed according to the change in plant behavior. However, disaster drills can be efficiently and effectively performed.
[Brief description of the drawings]
FIG. 1 is a block diagram showing a configuration of an embodiment of a training apparatus according to the present invention.
FIG. 2 is a diagram showing a configuration example of scenario data in the embodiment of FIG. 1;
FIG. 3 is a diagram showing a configuration example of training data in the embodiment of FIG. 1;
4A and 4B are diagrams showing an example of a data value reliability information setting screen in the embodiment of FIG. 1, wherein FIG. 4A is a screen for selecting a reliability information setting method, and FIG. A screen when setting by upper and lower limit values, (c) shows a screen when setting by period.
5 is a diagram showing an example of a display screen of the display data section in the embodiment of FIG. 1, wherein (a) is a screen for selecting main parameters of a plant and displaying them as a trend graph, and (b) is a screen. The screen which displays the main parameter of a plant as a summary is shown.
6 is a diagram showing an example of a display condition setting screen in the embodiment of FIG. 1, wherein (a) is a screen for setting a data time condition, and (b) is a screen for designating a data reading method. , (C) shows a screen for setting a data reading start request.
7 is a diagram showing a configuration example of data passed from the display condition setting unit to the data display unit in the embodiment of FIG. 1;
8 is a flowchart showing an example of a display condition setting procedure in the embodiment of FIG.
FIG. 9 is a flowchart showing an example of a maximum value and minimum value extraction method used in the display condition setting procedure in the embodiment of FIG. 1;
10 is a diagram showing an example of a period designation screen displayed when “designate by graph” is selected on the screen of FIG.
[Explanation of symbols]
DESCRIPTION OF SYMBOLS 1 ... Scenario data storage part, 2 ... Reliability information setting part, 3 ... Training data storage part, 4 ... Data display part, 5 ... Display condition setting part, 6 ... Peak value determination part, 7 ... Display data storage part, DESCRIPTION OF SYMBOLS 8 ... Display data delivery part, 9 ... Terminal device, 10 ... Password registration part, 20 ... Basic apparatus part, 22 ... Router, 24 ... Communication line.

Claims (6)

In a training apparatus comprising a device basic unit and a terminal device connected to the device basic unit via a communication line,
The apparatus basic unit is a scenario data storage unit that stores scenario data in which data values of a plurality of parameters representing the behavior of the plant obtained by analysis are recorded in time series, and
A reliability information setting unit for creating training data by adding information indicating a reliability index of a data value to the scenario data stored in the scenario data storage unit ;
Wherein the reliability information set the training data storage unit information indicating a reliability index to store the training data created in addition to the scenario data by unit,
A display unit for displaying at least a part of the training data together with information indicating the reliability index;
A display condition setting unit for setting an update condition of the scenario data to be displayed on the display unit;
A display data distribution unit that performs transmission of display contents displayed on the display unit to the terminal device and reception of update conditions transmitted from the terminal device via the communication circuit;
The display condition setting unit sets, as the update condition, one of fast-forwarding, thinning-out, and skipping as a request type based on a setting operation from a user, and processing by the processing method set as the request type. Set the time period of the training data to be targeted by the time series number of the training data,
The display unit is selected from the training data stored in the training data storage unit based on the request type set as the update condition by the display condition setting unit and the period of the training data. Get what to display, display this acquired training data,
The terminal device receives display content displayed on the display unit transmitted from the display data distribution unit via the communication line, displays the received display content, and from the user in the terminal device A training apparatus , wherein an update condition set based on a setting operation is transmitted to the display data distribution unit via the communication line as information relating to an update condition set by the display condition setting unit .   The training apparatus according to claim 1, wherein the reliability index is determined to be unreliable when the data value is outside a predetermined range. The display condition setting section, said that it is configured to be able to change the conditions for updating the scenario data when the scenario data is displayed on the display unit, the training of claim 1, wherein apparatus. The display condition setting section, in a state in which a trend graph of the scenario data on the display unit is displayed, it is intended that can be set, training device according to claim 1 or 2 wherein. The display condition setting unit can be set so as to select a maximum value and a minimum value of each period for each parameter by dividing a period at a predetermined interval in time from the scenario data stored in the time series The training device according to claim 1 or 2, wherein The display condition setting unit divides a period at predetermined intervals in time from the scenario data stored in the time series, and one specific parameter of the plurality of parameters has a maximum value and a minimum value in each period. select when taking the value at the selected timing, the all displayed on the display unit that is intended to be set to select the data of the parameters, according to claim 1 or 2, wherein Training device.

Images