Time domain simulation method and system for circuit containing switching device, storage medium and equipment

Abstract

The invention relates to the technical field of circuit system time domain simulation, and provides a time domain simulation method and system of a circuit containing a switching device, a storage medium and equipment. The method comprises the following steps: establishing a differential equation set containing a switching device circuit; performing simulation test on the action moment of the positioning switch device; finding out a correct combination of the synchronous action switching devices, and re-initializing the equation set at the action moment of the switching devices; and returning to the time sequence before the action moment of the switching device, and restarting simulation by adopting TR-BDF2. When simulation of a circuit containing a switching device is carried out, a TR-BDF2 format and a Lagrange interpolation method are adopted, and high precision and L stability can be kept in the whole simulation process.

Description

Technical Field

[0001] The present disclosure belongs to the technical field of time domain simulation of circuit systems, and specifically relates to a simulation method for a circuit containing switching devices, and in particular to a time domain simulation method, system, storage medium and device for a circuit containing switching devices that maintains high precision and L-stable. Background Art

[0002] The frequent operation of switching devices places higher demands on transient simulation of circuit systems, especially circuits containing power electronic devices.

[0003] The difficulties are mainly reflected in three aspects:

[0004] 1. It is difficult to accurately solve the action time of the switching device. The fixed-step simulation algorithm can only change the state of the switching device at the integer multiple of the step length. The delay in the action time of the switching device will cause the waveform to have "spikes" and "oscillations" that are inconsistent with the true solution.

[0005] 2. Linear interpolation often reduces the accuracy of the numerical integration format. PSCAD (Power Systems Computer Aided Design, electromagnetic transient simulation software) uses an implicit trapezoidal integration format with second-order accuracy to solve electromagnetic transient equations. When a switch state change is detected, PSCAD uses linear interpolation to locate the switching moment of the switch device. Linear interpolation will reduce the simulation accuracy to first order.

[0006] 3. Eliminate the numerical oscillation caused by the non-state quantity jump as soon as possible after the switch device is actuated. The power electronic switch device is modeled as a piecewise linear resistance model, and the circuit equation often shows "rigidity". d There is a "jump" of non-state variables. For example, when the inductor current x is forced to turn off by the switch device, the resistance of the switch device in the off state (MΩ level) causes some variables in the circuit to change rapidly in a short time. Electromagnetic transient simulation does not use a small enough step size to accurately solve this transient change process, so the instantaneous value of the switch device after switching is not accurate (the error is recorded as ε(t d )).

[0007] In order not to affect subsequent calculations, we hope that ε(t d ) can be damped in a single step, that is, it is necessary to use a numerical integration format with L-stable (L-stable: if a certain time discrete format satisfies that when the single step size is large enough, the local truncation error can be completely damped in a single step, then the format is called L-stable) property to damp ε(t d ). PSCAD is used to find Then, the trapezoidal integration and midpoint interpolation method are used to The error is damped at all times. This method is L-stable as verified by the principle formula, but The transient change process in the interval cannot be solved correctly, which may cause misjudgment of the state of other switching devices in the interval. Where h is the integration step.

[0008] Unlike PSCAD, EMTP-RV (a professional software for developing full-featured and technologically advanced simulation and analysis of electromagnetic transients in power systems) and PS IM (a simulation application package software for the power electronics and motor control fields) use one or two half-step backward Euler damping ε (t d ), but the transient change process within this step or half step cannot be solved correctly, and the backward Euler method will lose the second-order accuracy of the trapezoidal integration format.

[0009] "Damped Second-Order Accuracy Interpolation Method for Power Electronic Switching Circuits" (Yi Zhongfei, Zhang Lei, Cao Shuli, etc., Modern Electronic Technology, 2022, 45-01) explains how to find t d and x(t d ) and then find x n+1 A damped second-order precision interpolation method, but does not specify the switching action time t d At the same time, the document adopts a TR-BDF2-TR integration method, using the result x of TR-BDF2 d+1 , the result of TR-BDF2-TRx d+1+γ , and the Lagrange interpolation point x d+α Fit a new quadratic polynomial and interpolate to t n+1 At this moment, we get x n+1 , from x d To find x n+1 A total of three numerical integrations and two parabolic interpolations are required. It can be seen that the amount of calculation (numerical integration operations) in this document is very large and the calculation efficiency is not high. Summary of the invention

[0010] To solve the above problems, the present disclosure provides a simulation method, system, storage medium and device for a circuit containing a switching device, which adopts the TR-BDF2 format and the Lagrange interpolation method, and can maintain high accuracy and L stability throughout the simulation process.

[0011] A simulation method for a circuit containing a switching device, comprising:

[0012] S1, establish a differential equation system containing a switching device circuit;

[0013] S2, solving the equation group and detecting the positioning switch action; including,

[0014] S2.1, using BDF2 to solve the equations and find the state variable 1;

[0015] S2.2, detect whether there is a switch action. If there is a switch action, go back one integration step h and perform one step of implicit trapezoidal integration on the equation group with a step length of yh to find x tr ;

[0016] S2.3, based on the x tr , use Lagrange interpolation method to solve the state variable 2 for the equation group;

[0017] S2.4, based on the x tr , use the midpoint interpolation method to solve the state variable three for the equation group;

[0018] S2.5, make parabola fitting for the guard functions at the first, second and third state variables, and find the zero-crossing point of the guard function, i.e. the action moment of the switch device;

[0019] S3, finding the correct combination of synchronous action switching devices, and re-initializing the equation group at the action time of the switching devices;

[0020] S4, resynchronize using the TR-BDF2 format to return to the timing before the switching device is actuated;

[0021] S5, using the TR-BDF2 method to complete the restart simulation of the BDF2 format.

[0022] Further,

[0023] S3 includes:

[0024] S3.1, find the action time of all switch devices, take the minimum action time as the initial value, and update the equation group:

[0025] S3.2, under the condition that the state variable cannot change suddenly before and after the switch device is actuated, the zero positive value of the non-state variable is obtained by using the equation group, and it is determined whether there is a synchronous switch device actuation; if there is a synchronous switch device actuation, the moment before the switch device is actuated is used as the initial value, and the equation group is updated again; if not, proceed to the next step;

[0026] S3.3, repeat the previous step until all the switching devices that act synchronously are found, and use the time before the switching devices act as the initial value to completely reinitialize the equation group.

[0027] Further,

[0028] S4 includes:

[0029] S4.1, based on Use TR-BDF2 method to find x t and x b ;

[0030] S4.2, use (t d , x d )、(t t , x t )、(t b , x b ) Perform Lagrange interpolation to t α =t d +αh, find x α =x(t d +αh);

[0031] S4.3, use (t d , x d )、(t t , x t ) do midpoint interpolation and get x l ;

[0032] S4.4, detection interval (t d , t b ) Whether there is any switch device operation;

[0033] If yes, go to S4.5; if no, go to S4.6;

[0034] S4.5, for x l 、x α and x b The guard function at is fitted with a parabola to find the zero crossing point of the guard function, i.e., the action time of the switch device t 2d ; If multiple switching devices are activated, take the smallest t 2d ;

[0035] Use (t l , x l )、(t α , x α )、(t b , x b ) Perform Lagrange interpolation to t 2d At 2d - ); if t2 d >t n+1 , go to S4.6; if t 2d ≤t n+1 , go to S4.7;

[0036] S4.6, use (t l , x l )、(t α , xα )、(t b , x b ) Perform Lagrange interpolation and find x n+1 , so far, the S4 process is completely completed and goes to S5;

[0037] S4.7, from the differential algebraic equations under the new switching device state, find y 2d + , by y 2d + Determine the action of synchronous switching devices;

[0038] Repeat S4.7 until the correct combination of synchronous switching device actions is found and the correct y is obtained. 2d + ;

[0039] If 2d =t n+1 ,but Go to S5; if t 2d <t n+1 , let t d =t 2d ,x(t d + )=x(t 2d + ), y(t d + )=y(t 2d + ), return to S4.1.

[0040] Further,

[0041] The action of the switching device occurring within the first step of the restart is positioned and re-initialized at the action moment, and high accuracy and L stability are always maintained.

[0042] Further,

[0043] If the switch action time t d In the interval (t n , t n+1 ), and x is obtained by S4 process n+1 After that, if the BDF2 format still cannot be started, use the TR-BDF2 method to restart the BDF2 format;

[0044] S5 includes:

[0045] S5.1, based on x n+1 , use the TR-BDF2 method to find x tr and x bdf2 ;

[0046] S5.2, detection interval (tn+1 , t bdf2 ) Whether there is any switch device operation;

[0047] If not, let x n+2 =x bdf2 , the S5 process is completely finished; if there is, go to S5.3;

[0048] S5.3, use (t n+1 , x n+1 )、(t tr , x tr )、(t bdf 2,x bdf2 ) Perform Lagrange interpolation to t α =t n+1 +α h At α =x(t n+1 +αh);

[0049] Use (t n+1 , x n+1 )、(t tr， x tr ) Perform midpoint interpolation to find x l ;

[0050] x l 、x α and x bdf2 The guard function at is fitted with a parabola to find the zero crossing point of the guard function, i.e., the action time of the switch device t d If multiple switching devices are in operation, take the minimum t d ;

[0051] Use (t l , x l )、(t α , x α )、(t bdf2 , x bdf2 ) Perform Lagrange interpolation to t d , find x(t d + ); Set n = n + 1, and go to S4.1 until the new x is obtained after the S4 process is completed. n+1 , then the S5 process ends; at this point, return to the conventional solution steps.

[0052] The present disclosure also provides a time domain simulation system for a circuit containing a switching device, comprising:

[0053] An equation group module, wherein the equation group module is established according to an improved node voltage method and includes state variables, time variables and non-state variables;

[0054] The action time positioning module of the switch device adopts the BDF2 format to solve the differential equation group, and performs Lagrange interpolation operation according to the solution of BDF2 to obtain the action time of the switch device;

[0055] An equation group reinitialization module, used to determine a correct combination of switch devices for synchronous action and reinitialize the equation group;

[0056] The timing resynchronization module uses the TR-BDF2 format and Lagrange interpolation method to solve the state variable value of the next step at the switching device action moment to return to the original timing;

[0057] The restart module of the BDF2 method calculates the value of the state variable at the second full-step time after the switch action moment, so that the BDF2 integration method returns to normal.

[0058] Further,

[0059] The system is used to implement any one of the aforementioned methods.

[0060] Based on the same inventive concept, the present disclosure also provides a computer-readable storage medium.

[0061] One or more programs are stored, and when the one or more programs are executed, any of the aforementioned methods can be implemented.

[0062] In addition, the present disclosure provides a device, including a processor, a communication interface, a memory, and a communication bus; the processor, the communication interface, and the memory communicate with each other via the communication bus;

[0063] The processor is used for the program stored in the computer-readable storage medium. Compared with the prior art, the present disclosure has the following advantages:

[0064] 1. When simulating a circuit containing switching devices, the present invention proposes an interpolation method in the format of BDF2 (two-step backward difference formula, an implicit linear multi-step method used to solve differential equations of the form y=f(t,y)), reinitialization (finding the correct combination of synchronous switching actions, finding the correct t d , Correctly reinitialize equation (1)), resynchronize (solve for x n+1 , return to the original timing), BDF2 format restart (according to x α and x n+1 Find x n+2 ) are 3 different things. The method disclosed in this disclosure can ensure that each process has L stability.

[0065] 2. In the embodiment of the present disclosure, the midpoint interpolation point of the trapezoidal integral also has the property of L stability, so only two numerical integrations, two parabolic interpolations, and one linear interpolation are required. Since the amount of calculation for interpolation is significantly less than that for numerical integration, the method of the present disclosure can significantly improve the calculation efficiency compared with the method in the document "Damped Second-Order Precision Interpolation Method for Power Electronic Switch Circuits".

[0066] 3. For passive switching devices such as diodes, t d Still need to calculate. This disclosure not only gives t d and x d The clear method of finding , and it is explained in principle: as long as x n+1 is obtained via an inherently L-stable format (such as BDF2), then the constant γ can be arbitrarily chosen.

[0067] Other features and advantages of the present disclosure will be described in the following description, and partly become apparent from the description, or be understood by implementing the present disclosure. The purpose and other advantages of the present disclosure can be realized and obtained by the structures pointed out in the description, claims and drawings. BRIEF DESCRIPTION OF THE DRAWINGS

[0068] In order to more clearly illustrate the embodiments of the present disclosure or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present disclosure. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0069] Figure 1 A schematic flow chart of a time domain simulation method for a circuit containing a switching device is shown;

[0070] Figure 2 An equivalent topological diagram of a simulation test circuit according to an embodiment of the present disclosure is shown;

[0071] Figure 3 Shows Figure 2 Comparison diagram of the solution results of the terminal voltage of the switching device;

[0072] Figure 4 A schematic diagram of the structure of an electronic device according to an embodiment of the present disclosure is shown. DETAILED DESCRIPTION

[0073] In order to make the purpose, technical solution and advantages of the embodiments of the present disclosure clearer, the technical solution in the embodiments of the present disclosure will be clearly and completely described below in conjunction with the drawings in the embodiments of the present disclosure. Obviously, the described embodiments are part of the embodiments of the present disclosure, not all of the embodiments. Based on the embodiments in the present disclosure, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present disclosure.

[0074] Attached Figure 1 A schematic flow chart of a time domain simulation method for a circuit containing a switching device according to an embodiment of the present disclosure is shown, wherein the time domain simulation method comprises:

[0075] S1, establish a differential equation system containing a switching device circuit;

[0076] S2, solving the equation group and detecting the action of the positioning switch;

[0077] S3, finding the correct combination of synchronous action switching devices, and re-initializing the equation group at the action time of the switching devices;

[0078] "Synchronous switches" refer to a group of switches whose on and off states are causally related, such as the IGBT and diode on the same arm of an H-bridge. The initial state of a bridge arm is that the IGBT (denoted as s1) is on and the diode (denoted as d1) is off. At a certain moment, s1 receives a shutdown signal, and the simulator will solve a set of algebraic equations in which s1 is turned off and other switches remain in their original states, solve all non-state variables in the circuit, and determine the operation of other switches. If the terminal voltage of d1 is (V f is the threshold voltage of diode conduction), then d1 is turned on. The simulator continues to solve a set of algebraic equations in which s1 is turned off, d1 is turned on, and other switches (s2, d2, s3, d3, etc.) remain in their original states, and finds the values ​​of all non-state variables. On the one hand, it verifies whether d1 is turned on, and on the other hand, it determines the action of other switches (s2, d2, s3, d3, etc.). Until the action of all synchronous switches caused by the turning off of s1 is found, that is, "the correct combination of synchronously acting switch devices".

[0079] S4, resynchronize using the TR-BDF2 format to return to the timing before the switching device is actuated;

[0080] S5, using the TR-BDF2 method to complete the restart simulation of the BDF2 format.

[0081] See attached Figure 2, shows the equivalent topological structure of the test circuit of the time domain simulation method of the circuit containing the switching device of the embodiment of the present disclosure, in which the GTO is used as the switching device to be simulated and tested. The test circuit includes a boost chopper circuit composed of an input power supply E1, an inductor L1, a switching device, a freewheeling diode D1, and a load resistor R1, and an input power supply E2 connected in parallel with the load resistor R1. The positive and negative poles of the input power supply E2 are consistent with the positive and negative poles of the output end of the boost chopper circuit. Specifically, the input power supply E1, the switching device, the input power supply E2, and the load resistor R1 are connected in parallel and grounded; the inductor L1 is connected between the positive pole of the input power supply E1 and the positive pole of the switching device, the positive pole of the freewheeling diode D1 is connected to the positive pole of the switching device and the inductor L1, and the negative pole of the freewheeling diode D1 is connected to the positive pole of the input power supply E2 and the load resistor R1.

[0082] The simulation test parameters of this embodiment are set as follows: E1 = 100V, E2 = 260V, R1 = 100Ω. The switch resistances of GTO and D1 are R on =1mΩ, R off =10MΩ. The initial conditions of the simulation test are: i L =0.91A, D1 is turned on and GTO remains off.

[0083] A fixed-step simulation is used, with a step length of h and the nth time step recorded as t n , the solution of all state variables in the circuit at this moment is denoted as x n . Find the solution x at the n+1th time step n+1 After that, it is determined that the diode D1 is turned off, that is, D1 is turned off during the time interval (t n , t n+1 )Inside.

[0084] Step S1 includes:

[0085] According to the improved node voltage method, we establish Figure 2 The system of differential algebraic equations of the topology shown

[0086]

[0087] In formula (1), x is a state variable (such as inductor current, capacitor voltage, etc.); t is a time variable; y is a non-state variable.

[0088] Step S2 includes:

[0089] S2.1, use BDF2 to solve the differential algebraic equations and find x n+1 ;

[0090] S2.2, judge (t n , t n+1) Whether there is a switch action within the time: If there is no switch action, output x n+1 , set n = n + 1, and return to S2.1; if there is a switch action, return to t n , perform a one-step implicit trapezoidal integration with a step length of γh to find the values ​​of all state variables in the circuit, denoted as x tr .

[0091] Where h represents the integration step size, and γ takes any value in (0, 1).

[0092] S2.3, with (t n , x n )、(t tr， x tr )、(t n+1 , x n+1 ) Perform Lagrange interpolation to t α =t n +ah, find x α =x(t n +α h ),in x α is a point obtained by Lagrange interpolation with high accuracy and L-stable properties. n represents the nth time step. α is a constant.

[0093] S2.4,

[0094] Use (t n , x n )、(t tr， x tr ) do midpoint interpolation and get t l =t n The value of the state variable x at +0.5γh l ,

[0095] x l The proof that it is also L-stable is as follows:

[0096] According to the trapezoidal integral formula, Use (t n , x n )、(t tr , x tr ) to perform linear interpolation to t l (α) = t n At +αh:

[0097]

[0098] x n The errors at n , x lThe error at (α) is ε l (α) then: lim hλ →- ∞ ε l (α)=(1-2α)ε n , when α=0.5, lim hλ→-∞ ε l (α)=0, so It is also L-stable.

[0099] S2.5, for x l 、x α and x n+1 The guard function at t is used to make a parabola fit (the guard function is the criterion function for judging whether the switch device is actuated) to find the zero crossing point of the guard function, that is, the action time of the switch device t d .

[0100] S2.6, with (t l , x l )、(t α , x α )、(t n+1 , x n+1 ) Perform Lagrangian interpolation to the switching device action time t d , we can find is the instantaneous value of each state variable in the circuit before the switch is activated. Note that the state variables cannot change suddenly before and after the switch is activated, so the instantaneous value of the state variable in the circuit after the switch is activated is Later it is directly recorded as x d .

[0101] Since the BDF2 format and Lagrange interpolation are used, the whole process maintains high accuracy. The following proves that the method can obtain is L-stable:

[0102] Due to x n+1 and x l are all L-stable, as long as we prove that x α It is also L-stable. Then any point obtained by Lagrange interpolation from these three points is L-stable. The proof process is as follows:

[0103] According to the Lagrange interpolation formula:

[0104]

[0105] x n-1 、x n 、x α The errors at n-1 , ε n , ε α ,but:

[0106]

[0107] For formula (4), only when hour, x α is L stable, so x l 、x α and x n+1 The three points are obtained by Lagrange interpolation. It is also L-stable.

[0108] At the same time, the solution component with a large step size hλ will tend to decay to 0, which also shows that L stability has a damping characteristic. Therefore, the method disclosed in the present invention can effectively eliminate numerical oscillations.

[0109] Step S3 includes:

[0110] S3.1, if the interval (t n , t n+1 ) has multiple switching devices in action. Find the action time of all switching devices and take the minimum time as t d With the minimum t d As the initial value, reinitialize equation (1).

[0111] S3.2, the state variable x (inductor current, capacitor voltage, etc.) cannot change suddenly before and after the switching device is activated, that is, Substituting into the equation g(x, y, t) = 0, we can find the zero positive value of the non-state variable. Depend on and The value of meets the conditions for the switching device to operate, and determines whether there is a synchronous switching device operation. If yes, use t d + As the initial value, reinitialize equation (1).

[0112] S3.3, repeat the process of S3.2 until all the synchronously acting switching devices are found and t is completely reinitialized. d + Equation (1) at the moment.

[0113] In order to output the simulation results at the whole time point and to facilitate the analysis and comparison of the results, it is necessary to calculate x d Then, a single-step numerical integration format is used to solve the next step, and a high-precision, L-stable interpolation method is used to solve x n+1 (S4 process); and because the BDF2 discrete format requires two steps of historical values ​​to solve the next time step, it is also necessary to use some high-precision and L-stable method to solve x n+2 (S5 process). (Note that, to find x n+1 and xn+2 The process may still have new switching actions. n+1 and x n+2 After that, you can jump out of the interpolation strategy, x n+i (i≥3) can be solved according to the normal process of BDF2.

[0114] Step S4 specifically includes:

[0115] S4.1, from First, according to the correctly reinitialized system of equations ( is the instantaneous value of the state variable after the switch is actuated, also denoted as x d ), use the TR-BDF2 method to find x t and x b ,in (The TR-BDF2 method here refers to d To begin, do an implicit trapezoidal integration with a step length of γh to find x t , then according to x d , x t Make a BDF2 with a step length of (1-γ)h and find x b . )

[0116] S4.2, use (t d , x d )、(t t , x t )、(t b , x b ) Perform Lagrange interpolation to t α =t d +αh, find x α =x(t d +ah), where t t =t d +γh,t b =t d +h is similar to formula (3), which can prove that x α is L-stable.

[0117] S4.3, use (t d , x d )、(t t , x t ) Do midpoint interpolation: t l =t d +0.5yh, Similar to formula (2), it can be proved that x l is L-stable.

[0118] S4.4, detection interval (td , t b ) to see if any switch device is operating. If yes, go to S4.5; if no, go to S4.6.

[0119] S4.5, for x l 、x α and x b The guard function at t is used to make a parabola fit (the guard function is the criterion function for judging whether the switch device is actuated) to find the zero crossing point of the guard function, that is, the action time of the switch device t 2d (If multiple switching devices are in operation, take the smallest t 2d ).

[0120] Use (t l , x l )、(t α , x α )、(t b , x b ) Perform Lagrange interpolation to t 2d At 2d - ). If t 2d >t n+1 , go to S4.6; if t 2d ≤t n+1 , go to S4.7.

[0121] S4.6, use (t l , x l )、(t α , x α )、(t b , x b ) Perform Lagrange interpolation and find x n+1 , go to S5.

[0122] S4.7, from the differential algebraic equations (1) under the new switching device state, find y 2d + , by y 2d + Determine the action of the synchronous switching device.

[0123] Repeat S4.7 until the correct combination of synchronous switching device actions is found and the correct y is obtained. 2d + If t 2d =t n+1 ,but Go to S5; if t 2d <t n+1 , let t d =t 2d ,x(t d +)=x(t 2d + ), y(t d + )=y(t 2d + ), return to S4.1.

[0124] Step S5 specifically includes:

[0125] S5.1, based on x n+1 , use the TR-BDF2 method to find x tr and x bdf2 ;

[0126] S5.2, detection interval (t n+1 , t bdf2 ) whether there is any switch device action, if not, let x n+2 =x bdf2 , the S5 process is completely finished. If yes, go to S5.3;

[0127] S5.3, use (t n+1 , x n+1 )、(t tr， x tr )、(t bdf2 , x bdf2 ) Perform Lagrange interpolation to t α =t n+1 +αh, find x α =x(t n+1 +αh); use (t n+1 , x n+1 )、(t tr , x tr ) Perform midpoint interpolation to find x l ; for x l 、x α and x bdf2 The guard function at is fitted with a parabola to find the zero crossing point of the guard function, i.e., the action time of the switch device t d If multiple switching devices are in operation, take the minimum t d ; Use (t l , x l ),(tα ， x α )、(t bdf2 , x bdf2 ) Perform Lagrange interpolation to t d , find x(t d + ). Set n = n + 1 and go to S4.1. Until a new x is found n+1, S5 process ends. At this point, the interpolation strategy can be jumped out and the conventional solution step can be returned to (set n = n + 1, and return to step S2.2).

[0128] In the method proposed in the embodiment of the present disclosure, the positioning of the switching action time (S2), the re-initialization of the equation group (S3), the resynchronization of the timing (S4) and the restart process of BDF2 (S5) always maintain high precision and L stability, and compared with the method proposed in the document "Damped Second-Order Accuracy Interpolation Method for Power Electronic Switching Circuits", the midpoint interpolation is used to replace the first implicit trapezoidal integration, thereby improving the calculation efficiency.

[0129] Calculation example

[0130] See again the attached Figure 2 In the example shown in the figure, we focus on the self-turn-off process of diode D1.

[0131] The simulation was performed for 10us with a step length of h = 0.63us, and the terminal voltage v of the GTO was calculated using the analytical solution (drawing according to the step length of 0.63us), PSCAD, and the method disclosed in the present invention. gto Solve. The results are compared as follows Figure 3 As shown, we can see that in v gto There is an obvious discrepancy between PSCAD and the analytical solution near the jump point. However, the method proposed in the present disclosure is basically consistent with the analytical solution. This shows that the method of the present disclosure is more accurate in capturing the action moment of the switching device and solving the zero positive value of the variable.

[0132] Figure 3 In v gto Near the jump point of , the solution error of the disclosed method is 3 orders of magnitude smaller than that of PSCAD. In the rest of the parts, the solution error of the disclosed method is 8 orders of magnitude or more smaller than that of PSCAD. In addition, the error of the disclosed method decays rapidly, while the error of PSCAD is almost iso-oscillating.

[0133] Based on the same inventive concept, the present disclosure also provides a computer-readable storage medium storing one or more programs, which, when executed, can implement the aforementioned time-domain simulation method of a circuit containing switching devices.

[0134] like Figure 4 As shown, an embodiment of the present disclosure further provides a device, including a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory communicate with each other via the communication bus.

[0135] The memory is a computer-readable storage medium for storing one or more programs.

[0136] The processor is used to execute the program stored in the computer-readable storage medium.

[0137] The computer-readable storage medium may be included in the device / apparatus described in the above embodiments; or may exist independently without being assembled into the device / apparatus.

[0138] Although the present disclosure has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present disclosure.

Claims

1. A time domain simulation method for a circuit containing a switching device, characterized in that: The method comprises: S1, establish a differential equation system containing a switching device circuit; S2, solving the equation group and detecting the positioning switch action; including, S2.1, using BDF2 to solve the equations and find the state variable 1; S2.2, detect whether there is a switch action. If there is a switch action, go back one integration step h and perform one step of implicit trapezoidal integration on the equation group with a step length of γh to find x tr ; S2.3, based on the x tr , use Lagrange interpolation method to solve the state variable 2 for the equation group; S2.4, based on the x tr , use the midpoint interpolation method to solve the state variable three for the equation group; S2.5, make parabola fitting for the guard functions at the first, second and third state variables to find the zero crossing point of the guard function, i.e. the action moment of the switch device; S3, finding the correct combination of synchronous action switching devices, and re-initializing the equation group at the action time of the switching devices; S4, resynchronize using the TR-BDF2 format to return to the timing before the switching device is actuated; S5, use the TR-BDF2 method to complete the restart simulation of the BDF2 format.

2. The method according to claim 1, characterized in that S3 includes: S3.1, finding the action time of all switch devices, taking the minimum action time as the initial value, and updating the equation group; S3.2, under the condition that the state variable cannot change suddenly before and after the switch device is actuated, the zero positive value of the non-state variable is obtained by using the equation group, and it is determined whether there is a synchronous switch device actuation; if there is a synchronous switch device actuation, the moment before the switch device is actuated is used as the initial value, and the equation group is updated again; if not, proceed to the next step; S3.3, repeat S3.2 until all the switching devices that act synchronously are found, and use the time before the switching devices act as the initial value to completely reinitialize the set of equations.

3. The method according to claim 2, characterized in that S4 includes: S4.1, based on Use TR-BDF2 method to find x t and x b ; S4.2, use (t d ,x d )、(t t ,x t )、(t b ,x b ) Perform Lagrange interpolation to t α =t d +αh, find x α =x(t d +αh); S4.3, use (t d ,x d )、(t t ,x t ) do midpoint interpolation and get x l ; S4.4, detection interval (t d ,t b ) Whether there is any switch device operation; If yes, go to S4.5; if no, go to S4.6; S4.5, for x l 、x α and x b The guard function at is fitted with a parabola to find the zero crossing point of the guard function, i.e., the action time of the switch device t 2d ; If multiple switching devices are activated, take the smallest t 2d ; Use (t l ,x l )、(t α ,x α )、(t b ,x b ) Perform Lagrange interpolation to t 2d At 2d - ); if t 2d >t n+1 , go to S4.6; if t 2d ≤t n+1 , go to S4.7; S4.6, use (t l ,x l )、(t α ,x α )、(t b ,x b ) Perform Lagrange interpolation and find x n+1 , so far, the S4 process is completely completed and goes to S5; S4.7, from the differential algebraic equations under the new switching device state, find y 2d + , by y 2d + Determine the action of synchronous switching devices; Repeat the process of S4.7 until the correct combination of synchronous switching device actions is found and the correct y is obtained. 2d + ; If 2d =t n+1 ,but Go to S5; if t 2d <t n+1 , let t d =t 2d ,x(t d + )=x(t 2d + ), y(t d + )=y(t 2d + ), return to S4.

1.

4. The method according to claim 3, characterized in that The action of the switching device occurring within the first step of the restart is positioned and re-initialized at the action moment, and high accuracy and L stability are always maintained.

5. The method according to claim 1, characterized in that: If the switch action time t d In the interval (t n ,t n+1 ), calculate x by S4 process n+1 After that, the BDF2 format still cannot be started, and the TR-BDF2 method is used to restart the BDF2 format; S5 includes: S5.1, based on x n+1 , use the TR-BDF2 method to find x tr and x bdf2 ; S5.2, detection interval (t n+1 ,t bdf2 ) Whether there is any switch device operation; If not, let x n+2 =x bdf2 , the S5 process is completely finished; if there is, go to S5.3; S5.3, use (t n+1 ,x n+1 )、(t tr ,x tr )、(t bdf2 ,x bdf2 ) Perform Lagrange interpolation to t α =t n+1 +αh, find x α =x(t n+1 +αh); Use (t n+1 ,x n+1 )、(t tr ,x tr ) Perform midpoint interpolation to find x l ; x l 、x α and x bdf2 The guard function at is fitted with a parabola to find the zero crossing point of the guard function, i.e., the action time of the switch device t d If multiple switching devices are in operation, take the minimum t d ; Use (t l ,x l )、(t α ,x α )、(t bdf2 ,x bdf2 ) Perform Lagrange interpolation to t d , find x(t d + ); Set n = n + 1, and go to S4.1 until the new x is obtained after the S4 process is completed. n+1 , then the S5 process ends; at this point, return to the conventional solution steps.

6. A time domain simulation system for a circuit containing a switching device, characterized in that: The system comprises: An equation group module, wherein the equation group module is established according to an improved node voltage method and includes state variables, time variables and non-state variables; The action time positioning module of the switch device adopts the BDF2 format to solve the differential equation group, and performs Lagrange interpolation operation according to the solution of BDF2 to obtain the action time of the switch device; An equation group reinitialization module, used to determine a correct combination of switch devices for synchronous action and reinitialize the equation group; The timing resynchronization module uses the TR-BDF2 format and Lagrange interpolation method to solve the state variable value of the next step at the switching device action moment to return to the original timing; The restart module of the BDF2 method calculates the value of the state variable at the second full-step time after the switch action moment, so that the BDF2 integration method returns to normal.

7. The system according to claim 6, characterized in that The system is used to implement the method according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that: One or more programs are stored, and when the one or more programs are executed, the method described in any one of claims 1 to 5 can be implemented.

9. A device comprising a processor, a communication interface, a memory and a communication bus; the processor, the communication interface and the memory communicate with each other via the communication bus; characterized in that: The processor is used to execute the program stored in the computer-readable storage medium according to claim 8.

Images