JP4845526B2 - System operation training equipment - Google Patents

Description

  TECHNICAL FIELD The present invention relates to a system operation training apparatus for maintaining and improving techniques, skills, etc., related to system operation during normal times and accidents of operators such as power supply command centers and control stations.

With the increasing scale and complexity of power systems and the increasing demand for supply reliability, daily system operations require advanced technical judgment. In particular, when a grid accident occurs, the social impact is extremely large, and the system operator is required to have a high level of skill to implement the accident recovery procedure quickly and accurately. In order to meet these demands, system operation training equipment has been introduced.
By the way, in the conventional training device, the trainer (training instructor) sets accident information such as accident occurrence time, accident point, accident aspect, etc. from the trainer table, initial system data setting and initial system state creation, Register as a training scenario. During training, trainees (trainees) monitor and control the system via the trainee table and system monitoring panel. The system simulation function is based on the transient stability calculation method according to the trainee operation during an accident, and based on the frequency / tidal flow (consolidation) method after the accident is removed. (See Patent Document 1).

Japanese Patent No. 3121647 (paragraphs 0008 to 11, FIG. 2)

  In the conventional method, the system simulation is usually executed by frequency / tidal flow calculation, stability calculation during an accident, and after switching off to frequency / tidal flow calculation, and stability calculation is performed only during the accident. Therefore, simulation of a control system such as PSS (Power System Stabilizer), AVR (Automatic Voltage Regulator), etc. remains in a short time, and the effect of the control system is not reflected. For this reason, for example, there is a possibility that the generator does not step out in an actual system, but it may be simulated as step out, and there is a problem in simulation accuracy, and it is not possible to faithfully simulate the accident spillover phenomenon, and the realism of training There was a problem that the training effect was not obtained. In addition, since the stability calculation is switched to frequency / tidal flow calculation by eliminating accidents, there are significant differences in frequency, voltage distribution, line power flow, etc. between before and after switching. At that time, the TM display, which is the numerical display of the frequency, voltage distribution, line power flow, etc., is displayed on the trainer and trainee man / machine device, so the previous display contents and TM value change greatly. There was also the problem of giving a great sense of incongruity.

  More recently, integration of power supply command centers and control stations has been attempted, and therefore the jurisdiction of power supply command centers and control stations has been greatly expanded, and the system simulation scale of the training equipment has also increased to 1000 buses. ~ 5000 bus, 200 generator ~ 300 generator. For generators, a PSS device is installed in the generator in order to stabilize the power system, and in order to calculate the stability in the event of a system fault, faithful simulation of the control system and simulation of the generator It is necessary to carry out a detailed model (such as a park model). For this reason, it is indispensable to apply a dynamic stability calculation method that simulates the control system in detail, instead of the conventional stability calculation method.

  This invention was made to solve the above-mentioned problems, faithfully simulates the system accident ripple phenomenon at the time of the occurrence of a system accident, fosters a sense of reality in training and obtains the maximum training effect, and training In response to a request for longer time, a system operation training apparatus capable of maintaining the calculation accuracy of system simulation during training is provided.

The system operation training apparatus according to the present invention includes training management function simulation means for managing system operation training, system simulation function simulation means for simulating the response of the power system based on a control command from the training management function simulation means, A training apparatus comprising:
The system simulation function simulation means is a first system simulation means for performing system simulation calculation including dynamic stability calculation with a real time response, and a system over a longer time than the first system simulation means including frequency calculation and power flow calculation. The second system simulation means that can perform the simulation calculation, from the time when the accident occurs until the time when the accident is removed and the system is stabilized, the system simulation calculation is performed by the first system simulation means until the accident occurs from the start of training Switching between the first and second system simulation means so that the system simulation calculation is performed by the second system simulation means during the period between the time when the system is stabilized after the accident is removed and the time when the next accident occurs or until the end of the training. have a system simulation switching judgment and switching means for performing,
The determination of the system stability is performed when the amount of time change in frequency in the bus voltage of all the generators connected to the simulation target system is equal to or less than a predetermined set value .

As described above, since the system simulation calculation including the dynamic stability calculation is performed by the first system simulation means by the first system simulation means from the time when the accident occurs until the time when the system becomes stable, the present invention Faithfully simulate accident spillovers and foster realism in training. Also, from the start of training to the occurrence of an accident or from the time when the system is stabilized after the accident is removed until the next accident occurs or until the end of the training, the second system simulation means includes a frequency / tidal flow calculation. Since the system simulation calculation over time can be executed, the calculation accuracy of the system simulation during training over a long time can be maintained.
Furthermore, the determination of system stability is made when the amount of time change in frequency at the bus voltage of all the generators connected to the simulation target system is below a predetermined set value. It is important to determine the stability of the grid.

Embodiment 1 FIG.
FIG. 1 is a block diagram showing the entire system operation training apparatus of the present invention. The training apparatus generally has a training management function, a system simulation function, and an automated simulation function. Among these, the training management function is a function that supports various work of the trainer, such as training scenario creation, training execution, and training evaluation. It consists of support functions such as. The system simulation function is a function that faithfully simulates system characteristics statically, dynamically, in terms of performance and accuracy, such as the system range of the trainee's jurisdiction or the system range in which the trainer performs a power station substitution operation. The automated simulation function is a software function implemented in a computer system in a power supply command station or a control station that Traini uses in actual work.

  In the figure, the training management server 1 executes training management function software that performs overall management of the training apparatus. The system simulation device 2 executes system simulation, receives data on various system conditions related to system simulation, accident occurrence conditions, etc. from the training management server 1, dynamic stability calculation, frequency / tidal flow calculation (consolidation) based system simulation, etc. A system simulation server 3 that executes system simulation function software that controls the system and passes the calculation result to the training management server 1, and a symmetry that executes dynamic stability calculation, frequency / tidal flow calculation (connection) based system simulation, and the like SMP-type PCs 4 and 5 having a multi-processor (Symmetric Multiple Processor) configuration are provided. The reason why the two series of SMP PCs 4 and 5 having the same configuration are provided is to increase the reliability of operation, and both PCs are connected by a LAN 11 for calculation.

  Furthermore, the power supply command station used by the trainee, the automation simulation server 6 in which the automation function software at the control station is installed, the trainer table 7 used by the trainer during the training, the trainee table 8 used by the trainee during the training, and the trainee during the training. And a system LAN 10 for connecting the training management server 1, the system simulation server 3, the automation simulation server 6 and the like.

FIG. 2 is a diagram illustrating the functional configuration of the system operation training apparatus of the present invention. The training management function simulation means 12 is software implemented in the training management server 1 of FIG. 1, and the automated simulation function simulation means 13 is software implemented in the automation simulation server 6 of FIG.
The system simulation condition / result data storage area 14 is a shared data storage area for the training management function simulation means 12 and the system simulation function simulation means 15, and the system simulation function simulation means 15 is implemented in the system simulation server 3 of FIG. Software. System simulation condition / result data storage area (PC) 16, dynamic stability calculation base system simulation means 17 as first system simulation means, and frequency / tidal flow calculation base system simulation means 18 as second system simulation means The system simulation switching determination / switching means 19 is data and software mounted on the SMP type PC1 and PC2 in FIG.

  The system simulation condition / result data storage area (PC) 16 stores the same data as the system simulation condition / result data storage area 14, and the dynamic stability calculation-based system simulation means 17 performs “failure calculation + The main protection / retro protection Ry simulation + dynamic stability calculation + generator simulation + load simulation + control device simulation + protection device simulation "is executed. The frequency / tidal current calculation base system simulation 18 is" frequency / tidal current calculation (consolidated) ". + Generator simulation + load simulation + control device simulation + protection device simulation ", and the system simulation switching determination / switching means 19 will switch from dynamic stability calculation to frequency / tidal flow calculation (connection) in detail later. However, after the accident was removed, the frequency change of the bus voltage of all the generators was within the allowable value. Calculated switching from Consolidated to dynamic stability calculation on condition accident and performs switching of both systems simulation means 17 and 18.

The characteristics of the power system vary greatly depending on whether or not an accident occurs in the system. Fig. 3 illustrates the necessity of system simulation based on dynamic stability calculation and frequency / tidal flow calculation from the viewpoint of system phenomena, and shows calculation methods, simulation targets, etc. that need to be implemented in the analysis of each phenomenon. .
First, when a system fault does not occur, as a factor that fluctuates the system frequency, bus voltage, and active / reactive power distribution, there is a change in load active / reactive power. As for the entire system, the daily load curve is predicted every day as the daily load curve, and the output of the generator is predicted accordingly, and the generator output is controlled on the day. The situation where the frequency deviates from the reference value occurs because the supply / demand balance does not match, but this is maintained almost at the reference value by the control of the speed control system of the generator. At the same time, the bus voltage fluctuates, but this is also adjusted within a specified range by adjusting the transformer tap and controlling SC and ShR by the VQC device (voltage / reactive power control device). Therefore, such a normal state can be calculated by applying a method of calculating the entire system by combining frequency calculation and power flow calculation based on a long-term dynamic characteristic analysis method.

  That is, “Phase 1: Start of training-Immediately before the occurrence of accident” and “Phase 3: System stability after accident removal-Immediately before the occurrence of accident” indicate that no system accident has occurred, so fluctuations along the daily load curve of the load Therefore, changes in system frequency, bus voltage, and active / reactive power flow are small, and therefore, system simulation based on frequency / power flow (concatenation) based on long-term dynamic characteristic analysis method can be applied. Moreover, what should be simulated includes a generator, a load, a control device, and a protection device.

  On the other hand, when a grid fault occurs, the electrical output of the generator is greatly reduced during the accident, but the mechanical input does not decrease immediately, so the generator near the accident point is accelerated, far away from the accident point, Alternatively, the generators of other power companies are not far from each other because they are electrically distant from each other. As a result, step-out occurs between the systems, and the generator steps out. May become impossible. Such a system fault spillover phenomenon cannot be calculated with high accuracy unless it is calculated by a dynamic stability calculation method that faithfully simulates a control system such as PSS, AVR, and governor of the generator. In addition, when a system fault occurs, the main protection Ry or the back-end protection RY operates to remove these accidents. Therefore, it is important to perform failure calculation and simulate the logic of the main protection / rear-end protection Ry. .

  In other words, “Phase 2: Accident occurrence until system stabilization after accident removal” and “Phase 4: Next accident occurrence-until system stabilization after accident removal” are due to the occurrence of a system accident. -System configuration changes greatly due to the operation of the back-up protection Ry, so the system may be shaken, and in some cases, system out-of-step or generator out-of-step may occur. System frequency, bus voltage, valid / invalid Since the power distribution also fluctuates greatly, system simulation based on dynamic stability calculation that can consider the operation of various control systems of the generator is indispensable. Moreover, what should be simulated includes failure calculation + main protection / rear protection Ry, generator, load, control device, and protection device.

  From the above, system simulation is usually possible by frequency / tidal flow calculation (concatenation), but it is necessary to perform system simulation based on dynamic stability calculation when an accident occurs. The problem is how to switch between the two. Basically, the precondition of frequency / tidal flow calculation (consolidation) is that the entire system is treated as an equivalent generator, and the system frequency is calculated. It is necessary to determine that the system frequency has been reduced after the accident is removed, and it is appropriate that the change in the frequency of the bus voltage of all the generators is within the set value.

In other words, switching from dynamic stability calculation to frequency / tidal flow calculation (linked) base is based on the condition that the frequency change of the bus voltage of all the generators is within an allowable value after the accident is removed.
FIG. 4 is a diagram for explaining a procedure for determining a stable state and an unstable state of the system by the above method.
In the figure, the determination unit 20 determines whether or not the accident has been removed. If No (no), the system instability flag 24 is set and the determination process ends. If Yes (accident removal), the calculation unit 21 calculates the amount of time change in frequency at the bus voltage of all the generators. Next, the determination unit 22 determines whether or not all the calculated values are equal to or less than a predetermined set value. Specifically, this determination is made, for example, based on whether or not the frequency change Δf per 10 ms, which is the integration calculation step Δt of the dynamic stability calculation, is ± 0.01 Hz or less. If Yes (frequency change is less than or equal to the set value for all generators), the system stability flag 23 is set and the determination process is terminated. If No (no), the system instability flag 24 is set and the determination process is terminated.

Next, a system simulation process based on dynamic stability calculation including determination of system stability and instability will be described with reference to the flowchart of FIG.
In step 25, it is checked whether or not the accident occurrence time set in the training scenario creation has been reached. If an accident has occurred (Yes in step 25), the normal phase and reverse phase are determined based on the symmetric coordinate method corresponding to the accident type. A network calculation (fault calculation) is performed using an equivalent circuit of a phase and a zero phase circuit, and finally a symmetric component is converted into a three-phase component (steps 26, 27, and 28). For example, in the case of 1LG, an equivalent circuit is obtained by connecting a normal phase, a negative phase, and a zero phase circuit in series. Therefore, the calculation of the positive phase, the negative phase, and the zero phase circuit is performed, and the calculation results are combined to obtain a three-phase circuit. The voltage and current of the circuit are calculated, and the input voltage and current of the main protection / rear protection Ry are calculated. This process can be performed in parallel.
If no accident occurs (No in step 25), the network calculation of only the positive phase circuit is performed (step 26-2).

Next, in step 29, the main protection / reserve protection Ry of the simulation target corresponding to the accident selected in the training preparation stage is extracted by the main protection / reserve protection Ry simulation, and the operation simulation of the Ry is calculated by the network calculation result. And Ry settling value. If it is in the operating range, count the timer.
Thereafter, the presence or absence of the Ry operation is determined in Step 30, and if it is present (when the timer count ends: Yes in Step 30), the trip target circuit breaker (CB) of the Ry is disconnected (Step 31). Thereafter, if there is a CB non-operation setting or the like, the accident removal is not completed. Therefore, it is determined in step 25-1 whether or not the system fault is still continued after the CB is shut off. Yes) again, the network calculation (fault calculation) is performed by the equivalent circuit of the normal phase, negative phase, and zero phase circuit based on the symmetric coordinate method corresponding to the accident type, and finally the symmetric component is converted into the three-phase component (Steps 26-1, 27-1, 28-1). This process can be performed in parallel.
When the accident has been removed (No in Step 25-1), the network calculation of only the positive phase circuit is performed because it is normal (Step 26-2).

Next, since the network calculation is completed, the equation of motion of the generator is solved (calculation of the internal phase angle of the generator: step 32), and the solution of the PSS system is detected (the power fluctuation of the generator is detected and suppressed). Therefore, an input signal for AVR control is output: Step 33), excitation system solution (adjustment of generator internal voltage: Step 34), speed adjustment system solution (turbine output adjustment: Step 35), generator equation solution The differential equation is solved (calculation of d-axis and q-axis electric circuits of the generator: step 36). The differential equation of this generator can be processed in parallel.
Next, if it is determined whether a system fault is occurring (step 25-2) and the accident is continuing (Yes), the process again branches to the main protection / repair protection Ry simulation (step 29). If the accident is not continuing (No in step 25-2), it is checked in step 37 whether the system is stable. If not stable (No in step 37), the apparatus simulation cycle is determined in step 38, If it coincides with the simulation cycle (Yes in step 38), the control device simulation (step 39) and the protection device simulation (step 40) are performed. Since the simulation of the control device and the protection device can be executed at an interval longer than the integral step of the dynamic stability calculation, the simulation cycle of the device is determined to avoid unnecessary processing. If it is not a simulation cycle (No in step 38), the process branches to the normal phase network calculation (step 26-2).
If the system is stable (Yes in step 37), the system simulation based on the dynamic stability calculation is terminated.
In FIG. 5, fault calculation (circuit network calculation), generator differential equation solving, and control device / protection device simulation can be performed in parallel.

  As described above, in the system operation training apparatus according to the first embodiment of the present invention, the system simulation based on the dynamic stability calculation is executed during the system oscillation from the time of the system accident occurrence to the time after the accident removal, and the normal operation or the accident removal Since the above two system simulations were switched to execute the system simulation based on frequency / tidal flow calculation (concatenation) after the system became stable later, the system accident propagation phenomenon was highly accurate in real time when an accident occurred, The simulation results can be faithfully simulated, and the continuity of the calculation results can be ensured even when switching to frequency / tidal flow calculation (concatenation). Therefore, the continuity of the trainee's large screen system panel display can be secured, and the system simulation can be performed even during long training. As a result, the high training effect can be obtained.

Embodiment 2. FIG.
For example, in the explanation of FIG. 5 above, the possibility of parallel processing of various operations was mentioned, but a system operation training apparatus in the case where a plurality of computers are specifically employed and the parallel processing is effectively utilized is implemented. This will be described below as Form 2.
In the following description, the case where a personal computer PC is used as a computer will be described. However, the present invention is not limited to this, and the present invention can be applied in the same manner even when an EWS (Engineering Work Station) or the like is used.

FIG. 6 shows an SMP type PC (SMP in FIG. 1) in which the system simulation of the system operation training apparatus is composed of eight computers (PCs) based on the switching between the dynamic stability calculation base and the frequency / tidal flow calculation (connection) base. This is an example in the case where it is realized with 8 types of PCs 4 and 5).
As for processing, the master PC (PC-1) controls the whole and the slave PCs (PC-2 to PC-8) execute the given processing to share the processing and realize parallel processing. Realize processing performance in real time. Note that when the master PC and each slave PC are activated at the start of training, they continue to execute processing until the training is completed. Therefore, for the slave PC, while the process is not executed, it is repeatedly monitored whether the key code is set in the interface table. In addition, the same software is installed in each slave PC to facilitate maintenance.

Basically, according to the processing flow of FIG. 5, the key code is set in the interface table of the slave PC in charge of the process that the master PC (PC-1) wants to execute, and the slave PC always checks it. Then, the process in charge of the own PC is executed without idling, and the key code is cleared after the process. With this method, there is no OS overhead compared to the normal parallel processing method such as threading, and the integral step of the dynamic stability calculation is, for example, within a very short time of 10 ms. Thus, the idle time can be eliminated, and the processing performance can be improved in proportion to the number of computers to be processed in parallel.
FIG. 6 shows key codes that the master PC sets in the interface table as the arithmetic processing proceeds. The key code is expressed by connecting the processing type and the slave PC number with a hyphen. For example, the key code “1-2” means that the processing type 1 is executed by the PC-2. The data position in the interface table referred to by each slave PC may be determined, and only the processing type may be recorded as the key code. The key code may be any key code as long as the master PC can correctly indicate the type of arithmetic processing to be executed by the slave PC between the master PC and the slave PC.

In the processing types 1 to 9, the main part of the processing shown in the flow of FIG. 5 is extracted, and if this processing is processed by the slave PC, all system simulations of the training apparatus can be realized.
The processing type 0 is a training time obtained by integrating the integration step ΔT of the dynamic stability calculation, indicates a relative time from the start of training, and is set by the master PC in the interface table of all PCs. It is used as the time for judging the occurrence of an accident.
Process type 0-1 indicates the relative time from the start of dynamic stability calculation, and is set by the master PC in the interface table of all PCs. Processing type 1 performs parallel processing of network calculations during an accident. Set key codes 1-2, 1-3, and 1-4, and execute parallel processing on PC-2, PC-3, and PC-4. Let it be processed. The processing type 2 sets the key code 2-5 and processes the main protection / rear protection Ry simulation on the PC-5. Processing type 3 has a CB non-operation setting after CB shutoff with Ry operation, and the network calculation when the accident continues, sets key codes 3-2, 3-3, 3-4, and PC -2, PC-3, and PC-4. In the process type 4, the network calculation of only the positive phase circuit after the accident is removed is executed by the PC-2 with the key code 4-2 set. Process type 5 sets key codes 5-2, 5-3, 5-4, and 5-5 to divide the differential equation of the generator into groups of generators for parallel processing, and sets PC-2, PC-3, PC-4, and PC-5 are processed in parallel. Here, an example is shown in which the generator is divided into four groups, but the first 1/4 generator is PC-2, the next 1/4 is PC-3, and the next 1/4 is PC. At -4, all remaining generators are processed by PC-5. Since the network calculation and the solution of the differential equation are performed in series, the PC can be shared.

  Process type 6 is a simulation of a control device and a protection device. Key codes 6-6 and 6-7 are set, and the control device processing is performed by PC-6, and the protection device simulation is performed by PC-7. Let Processing type 7 is a system simulation end process based on dynamic stability calculation, and key codes 7-2, 7-3, 7-4, 7-5, 7-6, 7-7 are set, The PC termination process is executed. In PC-5, the counter of the timer process in the main protection / repair protection Ry simulation is cleared. In the processing type 8, the key code 8-8 is set in accordance with the switching from the dynamic stability calculation base to the frequency / tidal current calculation (consolidated) base, and the system simulation based on the frequency / tidal current calculation (consolidated) is set to PC-8. To run. The processing type 9 sets the key code 9-8 in accordance with the switching from the frequency / tidal flow calculation (consolidated) base to the dynamic stability calculation base, and causes the PC-8 to execute the end processing.

Next, the parallel processing operation will be described. The operation of the master PC will be described with reference to FIG. 7, and the operation of each slave PC will be described with reference to FIGS.
FIG. 7 shows the operation of the key code setting process performed by the master PC, especially when the system simulation calculation is executed by switching the system simulation based on the dynamic stability calculation and the system simulation based on the frequency / tidal flow calculation by parallel processing. It is a flowchart explaining these.

In FIG. 7, in step 41, PC-1, which is a master PC that controls the entire system simulation, first initializes training time and stability calculation time, and interfaces between PC-1 and PC-2 to PC-8. The table is initialized, and then the PC-2 to PC-8 programs are started. Since PC-2 to PC-8 determine the command from PC-1 by reading the key of the interface table, the check is always repeated continuously.
Next, it is determined whether or not a system fault has occurred (step 25). If it has occurred (Yes in step 25), the network calculation key 1 is set in order to execute the network calculation in parallel (step 25). 42) Since the subsequent processing is performed in series, the determination whether the network calculation is completed (step 43) is continued.
If no accident has occurred (No in step 25), the network calculation key 4 for performing the normal phase network calculation is set (step 47), and the end determination of the network calculation (step 43-2) is ended. Continue until you do.

  If the network calculation is completed (Yes in step 43), then the key 2 is set to perform the main protection / repair protection Ry simulation (step 44), and the determination of the Ry simulation end (step 45) is completed. Continue until you do. When the Ry simulation is completed (Yes in step 45), it is determined again whether a system fault has occurred (step 25-1) in order to check whether the CB is shut off by the Ry operation and the accident removal is completed (step 25-1). If not (No in Step 25-1), the process branches to Step 47. If it is occurring (Yes in step 25-1), a backup protection Ry operation can be considered, so the network calculation key 3 is set in order to execute the network calculation after the system configuration change (step 46). Subsequently, the calculation is continued until the end of the determination (step 43-1). If the network calculation is completed (Yes in step 43-1), then the differential equation solving key 5 is set (step 48) in order to execute the differential equation solving of the generator in parallel (step 48). In order to determine whether the equation solution has been completed (step 49), the check is repeated continuously.

  When the solution of the differential equation is completed (Yes in step 49), in order to simulate various control devices and protection devices using the stability calculation result as input data, a determination is made as to whether the simulation is executed (step 38), In the case of the execution cycle (Yes in step 38), the key 6 for executing in parallel is set (step 50), and this process is left to the PC-6 and PC-7. Next, it is determined whether or not it is a cycle for transmitting the system simulation result to the system simulation server 3 (step 51). If it is a transmission cycle (Yes in step 51), CB change, voltage, frequency, active power, Reactive power or the like is transmitted to the system simulation server 3.

Next, it is determined whether or not it is time to switch the system simulation to the frequency / tidal flow calculation (consolidated) base (steps 25-2 and 37), and if not (Yes in step 25-2 or No in step 37), The training time and the stability calculation time are updated (step 53), and the process branches to step 25.
In the case of timing to switch to the frequency / tidal flow calculation (connection) base (Yes in step 37), set the key 7 to end the dynamic stability calculation (step 54), and then execute the frequency / tidal flow calculation (connection). The key 8 to be set is set (step 55). Finally, it is determined whether or not the training time is over (step 56). If the training time is different (No in step 56), the training time is updated and the dynamic stability calculation time is reset (step 57). When branching to 25 and it is determined that the process is terminated (Yes in step 56), the system simulation is terminated.

As described above, the parallel processing of the circuit network calculation (fault calculation) is, for example, in the case of 1LG accident, the normal phase network calculation with PC-2, the three phase circuit calculation, the reverse phase network calculation with PC-3, the PC Perform the zero-phase network calculation at -4. The PC-2 three-phase circuit calculation is a process of converting the voltage and current of the normal phase, reverse phase, and zero-phase network calculation results into three-phase (a-phase, b-phase, and c-phase) components. Moreover, in the differential equation solution of the generator, for example, the group of the generator is divided into four, and the equation of motion → PSS system → excitation system → regulator system → generator equation (dq axis circuit) on each PC. A method of calculating ¼ at a time. Accordingly, the equation of motion 1 group is executed in parallel with PC-2, the equation of motion 2 group in PC-3, the equation of motion 3 group in PC-4, and the equation of motion 4 group in PC-5.
In addition, about the division | segmentation of a generator group, it is not necessary to limit to four groups, and when an improvement in performance is required, you may divide | segment into more groups. For parallel processing on a PC, there are generally a thread method and a process parallelization method. However, there is an OS overhead, and calculation is performed with an integral step of 10 ms or less as in dynamic stability calculation. Is executed, this OS overhead becomes an obstacle, and the effect of parallel processing cannot be exhibited. However, in the present invention, the interface table between the PCs is provided on the shared memory, and each PC always monitors the key code, which is a calculation execution instruction, to execute parallel execution. Thus, the parallel processing can be executed with almost no waiting time, and the performance improvement is not saturated even when the number of PCs is increased.

  FIG. 8 and subsequent FIG. 9 are flowcharts showing the processing of the program to be installed in the slave PCs PC-2 to PC-8 based on the key code set in the master PC by the flow of FIG. . The program of each PC reads the command of the PC-1 which is the master PC from the interface table, determines whether or not it should be executed, and if executed, clears the key of the corresponding interface table after completion, The basic idea is to notify PC-1 that the processing has been completed.

In FIG. 8, first, it is determined whether or not a key related to the user has been set (step 58). If the key is not set, the key check is continued. , Monitor. If the key is set (No in step 58), the subsequent processing proceeds.
First, key 1 is checked (step 59), and if the PC key matches the key, normal phase network calculation and three-phase circuit calculation (step 26) or vice versa depending on the type of accident. The phase network calculation (step 27) or the zero phase network calculation (step 28) is performed, and after each calculation is completed, the key 1 is cleared (steps 60, 61, 62). Next, the key 2 is checked (step 63), and the PC that matches the key is subjected to the main protection / repair protection Ry simulation (step 29). Yes) shuts off the trip target CB of the Ry (step 31) and clears its own key 2 (step 64).

  Next, since there is a possibility that the accident has not been eliminated only by the main protection Ry operation due to the CB non-operation setting or the Ry non-operation setting, the key 3 is checked (step 65), If they match (Yes in step 65), normal phase network calculation, three phase circuit calculation (step 26-1), or negative phase network calculation (step 27-1), or zero phase network calculation (step 27-1). Step 28-1) is performed, and after completion of each, own key 3 is cleared (Steps 66, 67, 68). Next, in order to perform the network calculation after the accident removal, the key 4 is checked (step 69). If the PC key matches the key, the normal phase network calculation and the three-phase circuit calculation are performed. (Step 26-2) is performed, and after completion, the user's own key 4 is cleared (Step 70).

  Next, when the network calculation is completed, the key 5 is checked in order to execute the differential equation of the generator in parallel (step 71), and if the PC key and the key match (step 71). Yes), the motion equation 1 group (step 72), the motion equation 2 group (step 73), the motion equation 3 group (step 74), or the motion equation 4 group (step 75) is processed in parallel. After completion, the user's own key 5 is cleared (steps 76, 77, 78, 79). Next, the key 6 is checked in order to determine whether or not to execute simulation of various control devices and protection devices using the dynamic stability calculation result as input data (step 80). In the case (Yes in Step 80), the control device simulation (Step 39) or the protection device simulation (Step 40) is executed in parallel, and after each processing is completed, the own key 6 is cleared (Steps 81 and 82). . Next, in order to determine whether or not the calculation of the dynamic stability is completed, the key 7 is checked (step 83). If it does not match the own key (No in step 83), the process branches to step 59. If they match (Yes in step 83), the key 7 of the own PC, the integral amount, and the main protection / rear protection Ry simulation timer are cleared (step 84).

  Subsequently, in FIG. 9, the key 8 is checked in order to perform frequency / tidal flow calculation (concatenation) (step 85). If the key 8 matches the key (Yes in step 85), the frequency / tidal flow calculation (concatenation) is checked. A check is made as to whether or not the calculation cycle is reached (step 86). If they do not match (No in step 86), the process branches to step 95 in the subsequent stage. In the case of the calculation cycle (Yes in Step 86), the presence or absence of the occurrence of the separated system based on the CB interruption is checked (Step 87), and when the system separation occurs (Yes in Step 87), the system constant of the separated system is determined. Is calculated (step 88). The system constant of the separated system is an inertia constant of the entire system necessary for frequency calculation, and is obtained by summing up the inertia constant of each generator included in the system.

  After that, the power flow calculation at the start cross section of the frequency calculation is executed (step 89), the generator output, the load power, etc. are calculated, the frequency calculation is performed for each separated system (step 90), and then the device simulation that responds to the frequency is performed. (Step 91) is performed, and the calculation of the required steps is repeated. When the frequency calculation is completed (Yes in Step 92) in the determination of the completion of the required step (Step 92), the power flow calculation (Step 93) of the cross section is performed, and then the device simulation (Step 94) that responds to the power flow calculation result is performed. In order to complete the training or to calculate the dynamic stability again, it is determined whether or not the frequency / tidal flow calculation is to be terminated (step 95). If it is not completed (No in step 95), step 58 (Fig. Branch to 8). In the case of completion (Yes in Step 95), the user's own key 8 is cleared (Step 96), and finally, it is determined whether or not the training time is over (Step 56-1). No) branches to step 58 (FIG. 8). In the case of termination (Yes in step 56-1), the system simulation is terminated.

  As described above, frequency / tidal flow calculation (concatenation) is performed by linking frequency calculation and tidal current calculation. In frequency calculation, integration is performed, but the number of steps is small, and after frequency calculation, Since the tidal current calculation is always performed and the convergence accuracy of the tidal current calculation is high, even if training is performed for a long time of several hours, the calculation accuracy of the system simulation will not deteriorate. Proven.

  The connection between the frequency calculation and the tidal current calculation described above will be described using mathematical expressions. The generator oscillation equation is expressed by the following equation.

  Summing up equation (1) for all generators gives:

P _ACC represents the acceleration force of the entire system, and indicates that the system frequency is increasing when P _ACC > 0, and is decreasing when P _ACC <0.
From the formulas (1) and (2), if Δω _SYS and ω ₀ are eliminated, the following formula is obtained.

  By adding the following equation to the above equation regarding the voltage and reactive power, the equation of tidal current balance during system frequency change is obtained.

The frequency of the system can be obtained by numerical integration of equation (2). In Formula (2), since the tidal current calculation is performed in advance, the electrical output of the generator has been calculated, and the mechanical input of the generator can be calculated by performing the frequency calculation. Next, tidal current calculation is performed using equations (3), (4), and (5). In equation (3), the mechanical input of the generator and the acceleration power P _{ACC of the} system are known numbers, The electrical output of the machine will be calculated. In this way, it can be seen that the frequency calculation and the power flow calculation are coupled via the mechanical input of the generator. In addition, in the frequency calculation, in the integration step where the power flow calculation is not performed, the electrical output of the generator is extrapolated from the frequency calculation result, but the amount of change in the active power of the load is calculated from the frequency characteristics of the load, Approximating that the grid active power loss does not change, the frequency calculation is performed on the assumption that the amount of change in the active power of all loads becomes the change of the electrical output of all generators.

Embodiment 3 FIG.
The third embodiment relates to a load setting method necessary for circuit network calculation, and in particular, a setting method capable of reducing the number of times of failure calculation convergence when a system fault occurs will be described.
In general, there are constant impedance load, constant current load, and constant power load as load types. Of these, the constant impedance load can be treated as an admittance matrix in network calculation. When simulating both loads with only a constant current source, the admittance matrix becomes unstable (the diagonal element is not sufficiently large compared to the non-diagonal element), and the convergence characteristics of the network calculation deteriorate. Accordingly, the constant current load and the constant power load cannot be handled as they are, and the constant impedance load and the constant current source component (adjustment) as shown in FIG. Generally, it is set in the form of 98.

  First, in connection with this load setting, the network calculation process will be described in the case of normal phase network calculation, and on the premise thereof, the characteristics of the load setting method according to the third embodiment of the present invention will be described. To do. The network calculation is expressed by equation (6).

In the network calculation, the generator is an equivalent current source (I _G ), and the load includes a constant impedance load, a constant power load, and a constant power load. The constant impedance load is treated as a part of the Y matrix. The constant current load and the constant power load are simulated by “constant impedance component + constant current source component (I _L )”. For a general node, the node current _IN is zero. Therefore, in Equation (6), the current on the left side is a known number and the voltage on the right side is an unknown, and a solution can be obtained. However, some loads exhibit nonlinear characteristics with respect to the bus voltage such as a constant current load and a constant power load. The formula is shown below.

For a constant current load:
_{(P LI0 + jQ LI0) /} | V LI0 | = (P LI + jQ LI) / | V LI | = constant (7)
_{I LI0 = {P LI0 -jQ LI0} - (| V LI0 | 2 / Z LI)} / V LI0 * ··· (8)
For a constant power load:
P _LP0 + jQ _LP0 = constant (9)
_{I LP0 = {(P LP0 -jQ} LP0) - (| V LP0 | 2 / Z LP)} / V LP0 * ··· (10)
Here, the variables have the following meanings.
P: Active power load.
Q: Reactive power load.
Z: Impedance for the constant impedance of the load.
The subscripts have the following meanings.
I: Constant current load.
P: Constant power load.
0: Means the value in the initial power flow state.
*: Conjugate.

First, the current I _G of the generator can be calculated reactance of d-q-axis circuit of the armature, from behind the voltage or the like. Next, as the voltage of each load node, the bus voltage of the initial power flow calculation result may be used in the initial convergence step. Current I _L of the load, a constant current load by the formula (8), the constant power load can be calculated by Equation (10). In each convergence step, a voltage value that is the previous convergence result is used. The constant impedances of both loads are set in advance according to equations (15) and (17) described later. A constant impedance load (100% impedance load) has a node current of zero. Furthermore, the general load current I _N is also zero. Thus, the left side of the equation (6) is obtained. Next, the Y matrix of the circuit network is created by a normal method including the impedance components of the constant impedance load, constant current load, and constant power load. Thus, the equation (6) can be solved. Then, the constant current source is used to calculate the active / reactive power of the constant current and constant power load from the node voltage and current obtained by the network calculation, and to adjust the difference between it and the active / reactive power of the default load. The size of (for adjustment) is changed, calculation is repeated, and the convergence calculation of the network calculation is performed.

When calculating the correction amount of I _LI using the voltage V _LI calculated for the current I _LI flowing into the current source of the constant current load in the convergence calculation that repeatedly obtains the current and the voltage alternately by the network calculation , the correction amount [Delta] I _LI for the last _{I LI} is calculated as follows. First, the power consumed in the previous _ILI is as follows.

_{P LI + jQ LI = | V} LI | 2 / Z LI * + V LI × I LI * ··· (11)

_Assuming that the load power after adding ΔI _LI as a correction amount to the current source is P _{LI (k)} + jQ _{LI (k)} , the following equation is obtained as in the equation (11).
P _{LI (k)} + jQ _{LI (k)
^{= | V LI | 2 / Z}} LI * + V LI × (I LI + ΔI LI) *
^{_{= | V LI | 2 / Z}} LI * + V LI × I LI * + V LI × ΔI LI *
= P _LI + jQ _LI + V _LI × ΔI _LI ^*

Since this load power P _{LI (k)} + jQ _{LI (k)} also needs to satisfy Expression (7), the following expression is obtained.
P _{LI (k)} + jQ _{LI (k)} = (P _LI0 + jQ _LI0 ) × (| V _LI | / | V _LI0 |)
If P _{LI (k)} + jQ _{LI (k)} is eliminated from these two equations, the following equation is obtained.
(P _LI0 + jQ _LI0 ) × (| V _LI | / | V _LI0 |)
= P _LI + jQ _LI + V _LI × ΔI _LI ^*
Solving this equation for ΔI _LI yields:
ΔI _LI = {(P _LI0 −jQ _LI0 ) × (| V _LI | / | V _LI0 |)
−P _LI + jQ _LI } / V _LI ^* (12)

  The same applies to the case of constant power load.

_{P LP + jQ LP = | V} LP | 2 / Z LP * + V LP × I LP * ··· (13)
ΔI _LP = (P _LP0 −jQ _LP0 −P _LP + jQ _LP ) / V _LP ^* (14)

  Decreasing the values of equations (12) and (14) in each convergence step improves the convergence characteristics. For this purpose, both constant current load and constant power load have a constant impedance in which the power changes with the square of the voltage. It is desirable to make the ratio of minutes as small as possible. However, the purpose of providing a constant impedance component is to set the ratio of the diagonal elements and non-diagonal elements of the Y matrix of the network calculation to a predetermined value or more even when all loads are configured with constant current loads and constant power loads. It is to be. It is known that the solution of simultaneous linear equations becomes unstable when the ratio of diagonal elements and non-diagonal elements of the Y matrix is not large, and the solution of simultaneous linear equations is obtained by providing a constant impedance component. Can be prevented from becoming unstable. Therefore, it is necessary to express a part of the constant current load and the constant power load as a constant impedance load. When the predetermined ratio α of the constant current load is expressed by a constant impedance and the predetermined ratio β of the constant power load is expressed by a constant impedance, the impedance value and the initial value of the current source are as follows.

In the case of a constant current load, the impedance value Z _LI and the current source initial value I _LI0 are calculated by the following equations.
Z _LI = | V _LI0 | ² / ((P _LI0 −jQ _L0 ) × α) (15)
I _LI0 = ((P _LI0 −jQ _LI0 ) / V _LI0 ^* ) × (1−α) (16)
In the case of a constant power load, the impedance value Z _LP and the current source initial value I _LP0 are calculated by the following equations.
Z _LP = | V _LP0 | ² / ((P _LP0 −jQ _LP0 ) × β) (17)
I _LP0 = ((P _LP0 −jQ _LP0 ) / V _LP0 ^* ) × (1−β) (18)
Here, 0 ≦ α <1 and 0 ≦ β <1. Further, when Expression (15) is substituted into Expression (8), Expression (16) is obtained, and when Expression (17) is substituted into Expression (10), Expression (18) is obtained.

  In the network calculation in a state where no accident has occurred, the voltage of the system is a normal magnitude and does not change much. For example, during the occurrence of a three-phase ground fault, the node voltage in the vicinity of the fault point greatly decreases. In a constant current load or constant power load at a node where the voltage during the accident changes greatly from the voltage before the accident occurs, the load amount due to the constant impedance changes greatly from before the accident occurs. In some cases, the number of times of convergence until the load amount for the constant current source that compensates for the change is extremely large may be obtained. Here, the convergence number is the number of repetitions required for the circuit network calculation to converge.

  In order to avoid an excessive increase in the number of times of convergence of the constant current load and constant power load near the accident point during the accident occurrence, in this third embodiment, the load near the accident point during the accident occurrence and the other The load adjusts the ratio of the constant impedance and the constant current source. In other words, the constant current load and the constant power load are adjusted so as to be the constant current source by eliminating the constant impedance as shown in the equations (19) to (21). Here, the vicinity of the accident point is defined as a node including about k pieces of facilities such as a bus, a power transmission line, and a transformer connected to the accident point (k needs to be adjusted according to the system scale).

In the case of a constant current load, since α = 0, the impedance value Z _LI and the initial value I _LI0 of the current source are as follows.
Z _LI = ∞ (19)
I _LI0 = ((P _LI0 −jQ _LI0 ) / V _LI0 ^* ) (20)
In the case of a constant power load, β = 0 and similarly, the impedance value Z _LP and the current source initial value I _LP0 are calculated by the following equations.
Z _LP = ∞ (21)
I _LP0 = ((P _LP0 −jQ _LP0 ) / V _LP0 ^* ) (22)

For loads that are not in the vicinity of the accident point even during an accident, equations (15) to (18) when the accident is not occurring are used.
In the convergence calculation of the network, the correction amount obtained theoretically as in the equations (12) and (14) is often smaller than the required correction amount, and in this case, the number of convergence increases. In addition, it is known that the number of times of convergence can be reduced by correcting the correction amount theoretically obtained by a predetermined multiple (for example, 1.2 times). In order to reduce the number of times of convergence, a method may be employed in which the correction amount is a predetermined multiple of the values calculated by equations (12) and (14).

FIG. 11 is a diagram for explaining the processing flow of circuit network calculation when the load setting method according to the third embodiment of the present invention is adopted, and shows normal phase network calculation in normal times.
First, a Y matrix that is an admittance matrix including the constant current load of the circuit network and the constant impedance of the constant power load is created (step 99), and initial values for the constant current sources of the constant current load and constant power load nodes are obtained. An initial voltage value of the load node for calculation is set (step 100).
Next, it is determined that an accident has occurred (step 25-3). If an accident has occurred (Yes in step 25-3), in step 101, constant current load and constant power load near the accident point are determined. The impedance value and the initial value of the current source are set by the equations (19) to (22), assuming that the impedance is 0% and the constant current source is 100%. At the same time, the portion corresponding to the constant impedance of the constant current load and constant power load is corrected in the admittance matrix. For the constant current load other than the vicinity of the accident point, the constant impedance component is set to 100 × α%, the constant current source component is set to 100 × (1−α)%, and the initial value of the current source is set by the equation (16). For constant power loads other than the vicinity of the accident point, the constant impedance component is set to 100 × β% and the constant current source component is set to 100 × (1−β)%, and the initial value of the current source is set according to the equation (18).
When no accident has occurred (No in Step 25-3), in Step 102, the constant impedance of all constant current loads is set to 100 × α%, and the constant current source is set to 100 × (1-α)%. In (16), the initial value of the current source is set. In all constant power loads, the initial value of the current source is calculated by Expression (18), where the constant impedance component is 100 × β% and the constant current source component is 100 × (1−β)%.

  Thereafter, the node voltage is calculated from “I = Y · V” (step 103), the constant current load and the constant power load are calculated from the calculation result of the node voltage (step 104), and the initially set constant current load is calculated. Then, a mismatch with the power of the constant power load is calculated (step 105), and it is determined whether all of these values are less than or equal to an allowable value (step 106). If it is not less than the allowable value (No in step 106), the values for the constant current source of the constant current load and constant power load are corrected (step 107), the process branches to step 103, and the convergence calculation is continued. If all the values are less than or equal to the allowable value (Yes in step 106), the calculation is terminated because the network calculation has converged.

As described above, the constant current load and constant power load near the accident point are set so that the constant impedance is eliminated and all constant current sources are used, so the node voltage near the accident point changes greatly during the accident. Even so, the number of convergence times of the network calculation does not become extremely large, and the calculation process is performed quickly.
In the third embodiment, the constant impedance of the constant current load and the constant power load near the accident point where the accident occurred is set to 0%. However, even if it is not 0%, the power fluctuation due to the constant impedance is ignored. If the value is as small as possible, the same effect can be obtained. Even if either the constant current load or the constant power load is used and the constant impedance is set to 0% or a small value in the vicinity of the accident point during the accident, the load near the accident point during the accident is also different. The number of times of convergence can be reduced as compared with the case of handling the same load.

Embodiment 4 FIG.
In the main protection / rear protection Ry (relay) simulation based on the dynamic stability calculation, how to set the simulation target range greatly affects the calculation processing time. The simulation target range is generally set to a substantially constant range based on the system configuration. However, considering the diversity of training scenarios, this simulation target range must be set wider, and the processing time tends to increase.
In the fourth embodiment of the present invention, based on a training scenario created by a trainer, a simulation target range of main protection / repair protection Ry is logically determined and registered in advance in consideration of various conditions. In some cases, it was devised to solve the above problems by selecting the simulation target range according to the accident equipment, and the contents thereof will be described below with reference to FIG.

FIG. 12 is a flowchart showing an outline of selecting a simulation target for main protection / repair protection Ry performed after registration of a training scenario in the training preparation stage. First, after registration of the training scenario, the process of extracting the set accident setting data is performed (step 108), and thereafter, the simulation target Ry is selected according to the accident setting contents. In addition, Ry non-operation setting and CB non-operation setting shall be implemented with respect to accident setting equipment.
First, Ry non-operation setting for main protection Ry is checked (step 109), and CB non-operation setting is checked regardless of the presence or absence (steps 110 and 110-1). The CB non-operation setting is classified into 4 patterns.

  For type A (step 111: Ry, no CB inoperative), accident removal is possible only with main protection Ry, so only main protection Ry is selected, and short circuit or ground fault Ry is finally selected according to the type of accident. To do. In the B type (step 112: CB non-operational, Ry non-operational), the main protection Ry alone has the CB non-operation setting and the accident cannot be removed. Therefore, the main protection Ry and the back-up protection Ry related to the CB non-operation setting (The back-up protection Ry that can remove the accident even if there is a non-operation CB by the back-up protection Ry operation) is selected, and the ground fault or the short-circuit Ry is finally selected according to the accident type. For the C type (Step 113: Ry not activated, CB not activated), the main protection Ry cannot be operated, so only the protection Ry that can eliminate the accident is selected, and the ground fault or short circuit Ry corresponding to the accident type is selected. Is finally selected. For the D type (Step 114: Ry inactive and CB inactive), the main protection Ry cannot operate, so only the back-up protection Ry is selected. However, the CB non-operation is added compared to the C type. A wider range of rear-end protection Ry is selected, and a ground fault or short-circuit Ry is finally selected according to the type of accident.

  Next, the data related to the selected simulation target Ry is stored with the accident occurrence time, the accident equipment, the accident type, etc. added (step 115). It is checked in step 116 whether or not there is an accident setting. If there is (Yes in step 116), the process branches to step 108. If not (No in step 116), the process is terminated.

  As described above, the appropriate simulation range of the main protection / rear protection Ry is logically determined in advance according to the presence / absence of the Ry non-operation setting and the presence / absence of the CB non-operation setting. Regardless, the processing time can be kept to a minimum.

Further, in the embodiment of the present invention, the system simulation operation by the first and second system simulation means is executed by an arithmetic device having a plurality of computers,
The arithmetic device has a master computer that performs overall control, a plurality of slave computers that perform arithmetic processing, and an interface table that instructs the arithmetic processing performed by the slave computer, and the slave computer When the master computer sets the data to monitor and instruct the arithmetic processing performed by the slave computer in the interface table, the slave computer executes the instructed arithmetic processing, so the idle time of the parallel processing is eliminated and the parallel processing is performed. The processing performance can be greatly improved in proportion to the number of computers.

In addition, as a method of simulating the constant current load in the dynamic stability calculation executed by the first system simulation means by separating the impedance and the current source, when the initial voltage value in the convergence calculation of the network is applied to the impedance When setting the value of the constant current load impedance so that the predetermined ratio α of the constant current load flows to this impedance, and adopting the method of calculating the current source value by the convergence calculation of the network calculation,
For the constant current load near the accident point during the accident, the ratio of the current flowing through the impedance to the constant current load when the initial voltage value in the convergence calculation of the network was applied to the impedance was made smaller than α. In particular, the dynamic stability calculation process is performed quickly.

In addition, as a method of simulating the constant power load in the dynamic stability calculation executed by the first system simulation means by separating the impedance and the current source, when the initial voltage value in the convergence calculation of the network is applied to the impedance When setting the value of the constant power load impedance so that the predetermined ratio β of the constant power load is consumed by this impedance, and adopting the method of calculating the current source value by the convergence calculation of the network calculation,
In the constant power load near the accident point during the accident occurrence, when the initial voltage value in the convergence calculation of the network is applied to the impedance, the ratio of the power consumed by the impedance to the constant power load is made smaller than β. Therefore, in particular, the dynamic stability calculation process is performed quickly.

Also, in the main protection / rear protection Ry (relay) simulation based on the dynamic stability calculation executed by the first system simulation means,
Regarding the accident contents to be set with the training management function simulation means, the range of main protection / repair protection Ry to be simulated according to the accident equipment, accident type, Ry non-operation setting and CB (breaker) non-operation setting Since it is logically determined in advance, and when the training is executed, when the set accident contents are generated, the simulation target range of the main protection / repair protection Ry determined in advance is selected. The processing time of the main protection / rear protection Ry simulation can always be kept to the minimum necessary regardless of the setting condition of the presence / absence of the Ry non-operation setting or the CB non-operation setting.

It is a block diagram which shows the whole system | strain operation training apparatus in Embodiment 1 of this invention. It is a figure explaining the functional structure of the system | strain operation training apparatus of this invention. It is a figure which shows systematic simulation from a system phenomenon, and shows the calculation method and simulation object which are required by the analysis of each phenomenon in the form of a table | surface. It is a figure explaining the point which judges the stable state of a system | strain, and an unstable state. It is a flowchart explaining the process of the system simulation based on dynamic stability calculation. It is a system | strain operation training apparatus in Embodiment 2 of this invention, Comprising: It is a figure which shows the processing content in the case of performing a system | strain simulation in the parallel process of SMP type | mold PC comprised from eight PCs in the form of a table | surface. It is a flowchart explaining the operation | movement of the setting process of the key code which a master PC bears in parallel processing. It is a flowchart explaining operation | movement of each slave PC in parallel processing. It is a flowchart explaining the operation | movement of the parallel processing following FIG. In the system | strain operation training apparatus of Embodiment 3 of this invention, it is a figure explaining the form of the load set by system | strain simulation. FIG. 12 is a diagram for explaining a processing flow of circuit network calculation when the load setting method according to the third embodiment is adopted. In the system | strain operation training apparatus of Embodiment 4 of this invention, it is a figure explaining the processing flow which selects Ry simulation object range.

Explanation of symbols

1 training management server, 2 system simulation device, 3 system simulation server,
4,5 SMP type PC, 6 automation simulation server, 12 training management function simulation means,
13 automated simulation function simulation means, 15 system simulation function simulation means,
17 Dynamic stability calculation based system simulation means,
18 Frequency / tidal flow calculation based system simulation means,
19 System simulation switching determination / switching means.

Claims (5)

In a training apparatus comprising training management function simulation means for managing system operation training, and system simulation function simulation means for simulating the response of a power system based on a control command from the training management function simulation means,
The system simulation function simulation means has a longer time than the first system simulation means for performing system simulation calculation including dynamic stability calculation in response to real time, and the first system simulation means including frequency calculation and power flow calculation. Second system simulation means capable of system simulation calculation, from the time when the accident occurs until the time when the accident is removed and the system is stabilized, the first system simulation means performs system simulation calculation, and an accident occurs from the start of training. The first and second systems so that the second system simulation means performs the system simulation calculation until the next accident occurs or until the end of training until the system is stabilized after the accident is removed. have a system simulation switching determination processing and switching means for switching the simulation means,
The system operation training apparatus is characterized in that the determination of the system stability is performed when a time change amount of a frequency in the bus voltage of all the generators connected to the simulation target system is equal to or less than a predetermined set value . The system simulation calculation by the first and second system simulation means is executed by an arithmetic unit having a plurality of computers,
The arithmetic device includes a master computer that performs overall control, a plurality of slave computers that perform arithmetic processing, and an interface table that instructs the arithmetic processing performed by the slave computer. 2. The slave computer executes the specified arithmetic processing when the master computer sets data instructing arithmetic processing performed by the slave computer in the interface table, which is monitored by the slave computer. system operation training device. As a method of simulating the constant current load in the dynamic stability calculation executed by the first system simulation means by separating the impedance and the current source, when the initial voltage value in the convergence calculation of the network is applied to the impedance When setting the value of the constant current load impedance so that the predetermined ratio α of the constant current load flows to this impedance, and adopting the method of calculating the current source value by the convergence calculation of the network calculation,
For a constant current load near the accident point during an accident, when the initial voltage value in the convergence calculation of the network is applied to the impedance, the ratio of the current flowing through the impedance to the constant current load should be smaller than α. The system operation training apparatus according to claim 1, wherein As a method of simulating the constant power load in the dynamic stability calculation executed by the first system simulation means by separating the impedance and the current source, the voltage initial value in the convergence calculation of the network is applied to the impedance. When setting the value of the constant power load impedance so that the predetermined ratio β of the constant power load is consumed by this impedance, and adopting the method of calculating the current source value by the convergence calculation of the network calculation,
For the constant power load near the accident point during the accident occurrence, when the initial voltage value in the convergence calculation of the network is applied to the impedance, the ratio of the power consumed by the impedance to the constant power load is made smaller than β. The system operation training apparatus according to claim 1 . In the main protection / rear protection Ry (relay) simulation based on the dynamic stability calculation executed by the first system simulation means,
Range of main protection / repair protection Ry to be simulated according to accident equipment, accident type, Ry non-operation setting, and CB (breaker) non-operation setting regarding the accident contents set by the above training management function simulation means Is logically determined in advance, and at the time of occurrence of the set accident content at the time of training execution, the simulation target range of the corresponding main protection / rear protection Ry determined in advance is selected. The system | strain operation training apparatus in any one of Claim 1 thru | or 4 characterized by the above-mentioned .

Images