CN108009375A - Control signal characterizing method and PWM modulator model, switching device model and electromagnetical transient emulation method - Google Patents

Abstract

The invention discloses a control signal characterization method of a PWM gate control system, a PWM modulator model, a controlled power electronic switch device model and a high-precision PWM converter electromagnetic transient simulation method based on the method, which can accurately reflect events The time of occurrence, and the accurate change time of the output logic quantity of the PWM modulator can be obtained and output without reducing the memory usage efficiency, and the accurate simulation of the PWM converter can be realized by cooperating with various subsequent interpolation and integration methods. The control signal characterization method disclosed by the invention can simultaneously express the PWM control logic and possible precise logic change moments, optimize the memory usage and numerical calculation efficiency, and improve the reliability of the simulation algorithm program implementation. The invention is highly compatible with EMTP and other electromagnetic transient simulation methods based on fixed step size, and can be conveniently applied in simulation software based on the above electromagnetic transient simulation method.

Description

Control signal characterization method and PWM modulator model, switching device model and electromagnetic transient state simulation method

technical field

The invention relates to a control signal characterization method of a PWM gate control system and a PWM modulator model, a controlled power electronic switch device model and a high-precision PWM converter electromagnetic transient simulation method based on the method, which belong to the power system simulation technology .

Background technique

The power system is increasingly showing the trend of power electronics. The rapid development of technologies such as flexible DC transmission, flexible AC transmission, and distributed new energy generation has made power electronic equipment more widely used in power systems. Power electronic equipment is generally composed of a PWM converter using PWM (Pulse Width Modulation, pulse width modulation) technology. The PWM modulator in the PWM converter, as the interface between the converter control subsystem and the electrical subsystem, presents both continuous change characteristics and discrete event characteristics, and it is difficult to accurately simulate this characteristic. Transient simulation presents great challenges.

The PWM modulator usually adopts natural sampling or regular sampling method and modulates with triangular wave or sawtooth wave as the carrier. When the input modulation wave is greater than or equal to the carrier, it outputs logic "1", otherwise it outputs logic "0". The most widely used electromagnetic transient simulation method for power systems is the EMTP (Electro-Magnetic Transient Program) method proposed by Professor Dommel. Since the EMTP method is based on a fixed integral step size for simulation calculation, the output logic of the PWM modulator in the conventional EMTP method can only be changed at the full step size. However, since the change in the relationship between the modulating wave and the carrier size in practice often occurs at a non-integral step time, the conventional EMTP method will cause the loss of information at the exact moment when the PWM output signal changes, and will delay the result of this change to the next It is reflected in the simulation results of the whole step long time, which will cause errors and even get wrong results.

Aiming at the above problems, a large number of studies often only focus on how to use various interpolation and integration methods to obtain more accurate simulation results and at the same time suppress possible numerical oscillations after knowing the exact moment when the PWM output signal changes. The critical phenomenon that the conventional EMTP method misses the exact moment when the PWM output signal changes is often overlooked. And the few solutions that are available are more or less less accurate or memory efficient. Two existing solutions and their advantages and disadvantages are described below.

Solution 1: Do V Q, McCallum D, Giroux P, et al.A backward-forward interpolation technique for a precise modeling of power electronics in HYPERSIM[C].Proc.Int.Conf.Power System Transients,Rio de Janeiro,Brazil,2001. A modeling method of PWM modulator is proposed. In the PWM modulator model, the calculation function of the exact moment when the output logic quantity changes is added, and a double-type output variable is added to express it. At the same time, the matching IGBT , MOSFET, GTO and other controlled power electronic switching device models correspondingly add a double type input variable.

Advantages: This method enables the PWM modulator model to obtain and output the exact change time of the logic quantity, and cooperate with various subsequent interpolation and integration methods to achieve more accurate simulation.

Disadvantages: A double variable must be added to the PWM modulator and IGBT, MOSFET, GTO and other power electronic switching device models, which increases the memory burden of the simulation software.

Solution 2: Chinese patent ZL201310119795.3 proposes an improved EMTP algorithm suitable for the average model of PWM converters. The modified algorithm omits the complex switch processing sub-module in the traditional EMTP method, and adds 2 sub-modules for predicting segmentation The simulation error caused by the inaccurate parameters of the average model and correction parameters.

Advantages: Due to the use of the average model, the simulation speed of the PWM converter will be greatly improved.

Disadvantages: (1) The average model is used, which loses the change characteristics of the model on the microscopic time scale, and the simulation results are not accurate enough; (2) It cannot reflect the soft switching characteristics of some PWM converters using soft switching technology; (3) Prediction and correction modules increase the computational workload of the simulation.

Contents of the invention

Purpose of the invention: In order to overcome the deficiencies in the prior art, the present invention provides a control signal characterization method of a PWM gate control system and a PWM modulator model, a controlled power electronic switching device model and a high-precision PWM transformer based on the method. The current electromagnetic transient simulation method can accurately reflect the time when the event occurs, and can obtain and output the accurate change time of the output logic quantity of the PWM modulator without reducing the memory usage efficiency. Accurate simulation of PWM converters is achieved.

Technical scheme: in order to achieve the above object, the technical scheme adopted in the present invention is:

A control signal characterization method for a PWM gate control system. The control signal is a digital variable, and its value range and memory usage are set to be exactly the same as double-precision floating-point data. It can not only express logic "1" and logic "0", it can also provide the important information of the accurate change time of the logic quantity, and is compatible with the electromagnetic transient simulation algorithm based on the fixed step size; record the current simulation time as t _k , and the previous fixed step size simulation time as t _{k- 1.} According to the value of the control signal y _k at time t _k , it is represented as the following event extended logic variable:

① When y _k ∈ (1.0,+∞), the event expansion logic variable of the control signal within the time range t ∈ (t _k-1 ,t _k ] is represented as 1;

② When y _k ∈ (0.0,1.0], the event expansion logic variable of the control signal within the time range of t ∈ (t _k-1 ,t _k ] is represented as a rising edge transition from 0→1, and the transition time is t _s =y·t _k +(1-y)·t _k-1 ;

③ When y _k = 0.0, it indicates that there is no change in the event expansion logic variable within the time range of the control signal t∈(t _k-1 ,t _k ], and the event expansion logic variable at the end of the previous step is maintained;

④ When y _k ∈ [-1.0,0.0), the event expansion logic variable of the control signal within the time range t ∈ (t _k-1 , t _k ] is represented as a falling edge transition from 0→1, and the transition The moment is t _s =-y·t _k +(1+y)·t _k-1 ;

⑤ When y _k ∈ (-∞, -1.0), the event expansion logic variable of the control signal in the time range of t ∈ (t _k-1 , t _k ] is represented as 0.

A PWM modulator model based on the above-mentioned control signal characterization method, the PWM modulator model includes an input terminal and an output terminal; the input terminal is used to input a modulated wave signal u; the modulated wave signal u is compared with the carrier signal C to generate PWM control signal y; the output terminal is connected to the model of the controlled power electronic switching device for outputting PWM control signal y; record the current simulation time as t _k , and the input and output at time t _k are uk and _{y k respectively} . The simulation time with a fixed step size is t _k-1 , the input and output at time t _k-1 are u _k-1 and y _k-1 respectively, and the carrier period is T, then the update of PWM control signal y _k at time t _k The process is:

(a) calculate the time offset to _ff =t _k -floor(t _k /T) in the carrier period at the time t k, wherein floor ( ) represents rounding down; _enter step (b);

(b) according to the type of the carrier and t _off , calculate the carrier signal value C _k at t _k moment; enter step (c);

(c) Calculate y _k according to the relationship between C _k and u _k :

(c1) If u _k > C _k and u _k-1 ≥ C _k-1 , it means that the modulated wave is larger than the carrier in the time range of t∈(t _k-1 ,t _k ], set y _k > 1.0, for example, y _k = 2.0;

(c2) If u _k > C _k and u _k-1 < C _k-1 , it indicates that the relationship between the modulating wave and the carrier has changed within the time range of t∈(t _k-1 ,t _k ], let y _k = (C _k-1 -u _k-1 )/(u _k -C _k +C _k-1 -u _k-1 );

(c3) If u _k ＝C _k and u _k-1 >C _k-1 , it means that the modulating wave is larger than the carrier in the time range of t∈(t _k-1 ,t _k ), and the modulating wave and the carrier are exactly equal at time t _k , let y _k =-1.0;

(c4) If u _k ＝C _k and u _k-1 ＝C _k-1 , it is abnormal, let y _k ＝y _k-1 ;

(c5) If u _k ＝C _k and u _k-1 <C _k-1 , it means that the modulating wave is smaller than the carrier in the time range of t∈(t _k-1 ,t _k ), and the modulating wave and the carrier are exactly equal at time t _k , let y _k =1.0;

(c6) If u _k <C _k and u _k-1 >C _k-1 , it indicates that the relationship between the modulation wave and the carrier wave has changed within the time range of t∈(t _k-1 ,t _k ], let y _k = -(u _k-1 -C _k-1 )/(C _k -u _k +u _k-1 -C _k-1 );

(c7) If u _k <C _k and u _k-1 ≤ C _k-1 , it means that the modulated wave is smaller than the carrier in the time range of t∈(t _k-1 ,t _k ], set y _k <-1.0, for example, y _k = -2.0.

The carrier wave can be a triangle wave, a forward sawtooth wave or a backward sawtooth wave, and the carrier period and upper and lower limit values can be set.

A controlled power electronic switching device model based on the above control signal characterization method, the controlled power electronic switching device model is connected to the PWM modulator model, receives the PWM control signal y output by the PWM modulator model, and according to the PWM control signal y Interpolation or integral calculation is performed on the logic value and jump time to obtain its own follow-up switching action command, so as to realize reinitialization and suppress numerical oscillation; the controlled power electronic switching device model, by reading and By processing the PWM control signal output from the PWM modulator model, the precise change moment of the PWM modulator model can be obtained, so that various interpolation and integration methods can be used for subsequent switching action processing. Note that the current simulation time is t _k , and the previous fixed-step simulation time is t _k-1 , and the switching action state of the controlled power electronic switching device model is judged according to the PWM control signal y _k at time t _k :

① If y _k ∈ (1.0,+∞), it means that there is no switching action within the time range of t ∈ (t _k-1 ,t _k ], and it is in the open state;

② If y _k ∈ (0.0,1.0], it means that t ∈ (t _k-1 ,t _k ] changes from off to on within the time range, and the change time is t _s = y _k t _k + (1-y _k )·t _k-1 ;

③ If y _k = 0.0, it means that there is no switching action within the time range of t∈(t _k-1 ,t _k ];

④ If y _k ∈ [-1.0,0.0), it means that t ∈ (t _k-1 ,t _k ] time range changes from on to off, and the change time is t _s = -y _k t _k + (1 +y _k )·t _k-1 ;

⑤ If y _k ∈ (-∞, -1.0), it means that there is no switching action within the time range of t ∈ (t _k-1 , t _k ], and it is in the off state.

The controlled power electronic switching device can be SCR, IGBT, MOSFET, GTO or ideal switch.

A high-precision PWM converter electromagnetic transient simulation method based on the above-mentioned control signal characterization method, comprising the following steps:

(1) Simulation initialization, including control sub-model sorting, electrical equation matrix and state quantity initialization, and initial simulation time setting;

(2) Perform one-step simulation calculation on all control subsystem models to obtain modulated wave signals;

(3) Carry out one-step simulation calculation to all PWM modulator models, obtain the logic value and transition time of PWM control signal;

(4) According to the logic value and transition time of the PWM control signal, interpolation or integral calculation is performed to obtain the subsequent switching action instructions of the controlled power electronic switching device model, so as to realize reinitialization and suppress numerical oscillation, and finally obtain more accurate Simulation results of PWM converter;

(5) End the simulation calculation of the current step length, push the simulation time one step further and judge whether the simulation is over: if yes, end; otherwise, return to step (2).

Beneficial effects: the control signal characterization method of the PWM gate control system provided by the present invention and the PWM modulator model, the controlled power electronic switching device model and the high-precision PWM converter electromagnetic transient simulation method based on the method are different from the existing Compared with technology, it has the following advantages: 1. The control signal characterization method adopted by the present invention can not only express the PWM control logic but also express its precise change time, and realize more accurate PWM converters for subsequent use of various interpolation and integration methods The simulation has established a necessary and effective premise; 2. On the basis of realizing the high-precision modeling and simulation of the PWM converter, the present invention optimizes memory usage and numerical calculation efficiency, and multiplies the value of the event expansion logic type variable by a negative sign to realize Logical non-operation improves the reliability of simulation algorithm program implementation; 3. The present invention has very little changes to the models of existing PWM modulators and controlled power electronic switching devices, and is highly compatible with electromagnetic transients based on fixed step sizes such as EMTP The simulation method can be conveniently applied in the simulation software based on the above electromagnetic transient simulation method.

Description of drawings

Fig. 1 is the implementation flowchart of the electromagnetic transient simulation method of high-precision PWM converter;

Fig. 2 is the variable distribution of the PWM control signal of the present invention and the schematic diagram of expression meaning thereof;

Fig. 3 is the schematic diagram of the PWM modulation model of the present invention;

Fig. 4 is a schematic diagram of a model of a controlled power electronic switching device of the present invention.

Detailed ways

The present invention will be further described below in conjunction with the accompanying drawings.

A control signal characterization method for a PWM gate control system. The control signal is a digital variable, and its value range and memory usage are set to be exactly the same as the double-precision floating-point data. The current simulation time is t _k , and the previous one The fixed-step simulation time is t _k-1 , and according to the value of the control signal y _k at time t _k , it is represented as the following event extended logic variable:

① When y _k ∈ (1.0,+∞), the event expansion logic variable of the control signal within the time range t ∈ (t _k-1 ,t _k ] is represented as 1;

② When y _k ∈ (0.0,1.0], the event expansion logic variable of the control signal within the time range of t ∈ (t _k-1 ,t _k ] is represented as a rising edge transition from 0→1, and the transition time is t _s =y·t _k +(1-y)·t _k-1 ;

③ When y _k = 0.0, it indicates that there is no change in the event expansion logic variable within the time range of the control signal t∈(t _k-1 ,t _k ], and the event expansion logic variable at the end of the previous step is maintained;

④ When y _k ∈ [-1.0,0.0), the event expansion logic variable of the control signal within the time range t ∈ (t _k-1 , t _k ] is represented as a falling edge transition from 0→1, and the transition The moment is t _s =-y·t _k +(1+y)·t _k-1 ;

⑤ When y _k ∈ (-∞, -1.0), the event expansion logic variable of the control signal in the time range of t ∈ (t _k-1 , t _k ] is represented as 0.

By adopting the above control signal characterization method, it is possible not only to express the logic of the PWM gate control system, but also to effectively express the exact moment when the logic changes. The common logical NOT operation on the event extension logic variable can be realized by simply multiplying by the negative sign, which improves the calculation efficiency and does not increase memory consumption. Cooperating with various subsequent interpolation and integration methods, accurate simulation of the PWM converter can be realized.

Fig. 1 is a flow chart of an electromagnetic transient simulation method for a high-precision PWM converter. Outside the dotted line frame is the electromagnetic transient simulation process of a traditional power system, and inside the dotted line frame is the core of the present invention. The electromagnetic transient simulation of a power system containing a PWM converter mainly has the following process: after the initialization, a step of simulation calculation of the control subsystem is performed to obtain the modulated wave signal, and then a step of PWM modulator simulation is performed, and then the current Whether there is a control logic jump in the step size and corresponding processing is carried out, and finally the simulation time is pushed forward one step until the end of the simulation. Let the current simulation time be t _k , and the previous fixed-step simulation time be t _k-1 , the specific steps are as follows:

Step 1, simulation initialization, including control sub-model sorting, electrical equation matrix and state quantity initialization, and initial simulation time setting.

Step 2: Perform one-step simulation calculation on all control subsystem models to obtain the modulated wave signal u.

Step 3: Perform a one-step simulation calculation on all PWM modulator models to obtain the logic value and transition time of the PWM control signal; the PWM control signal here is characterized by the above method, and the event expands the distribution of logic variables and the logic changes expressed The situation is shown in Figure 2.

Described PWM modulator model as shown in Figure 3, comprises input terminal and output terminal; Described input terminal is used for input modulation wave signal u; Produces PWM control signal y after modulation wave signal u compares with carrier signal C; Said output The terminal is connected to the model of the controlled power electronic switching device to output PWM control signal y; record the current simulation time as t _k , the input and output at time t _k are u _k and y _k respectively, and the previous fixed step simulation time is t _k-1 , the input and output at time t _k-1 are u _k-1 and y _k-1 respectively, and the carrier cycle is T, then the update process of PWM control signal y _k at time t _k is:

(a) calculate the time offset to _ff =t _k -floor(t _k /T) in the carrier period at the time t k, wherein floor ( ) represents rounding down; _enter step (b);

(b) according to the type of the carrier and t _off , calculate the carrier signal value C _k at t _k moment; enter step (c);

(c) Calculate y _k according to the relationship between C _k and u _k :

(c1) If u _k ＞C _k and u _k-1 ≥C _k-1 , it means that the modulated wave is larger than the carrier wave within the time range of t∈(t _k-1 ,t _k ], set y _k ＞1.0;

(c2) If u _k > C _k and u _k-1 < C _k-1 , it indicates that the relationship between the modulating wave and the carrier has changed within the time range of t∈(t _k-1 ,t _k ], let y _k = (C _k-1 -u _k-1 )/(u _k -C _k +C _k-1 -u _k-1 );

(c3) If u _k ＝C _k and u _k-1 >C _k-1 , it means that the modulating wave is larger than the carrier in the time range of t∈(t _k-1 ,t _k ), and the modulating wave and the carrier are exactly equal at time t _k , let y _k =-1.0;

(c4) If u _k ＝C _k and u _k-1 ＝C _k-1 , it is abnormal, let y _k ＝y _k-1 ;

(c5) If u _k ＝C _k and u _k-1 <C _k-1 , it means that the modulating wave is smaller than the carrier in the time range of t∈(t _k-1 ,t _k ), and the modulating wave and the carrier are exactly equal at time t _k , let y _k =1.0;

(c6) If u _k <C _k and u _k-1 >C _k-1 , it indicates that the relationship between the modulation wave and the carrier wave has changed within the time range of t∈(t _k-1 ,t _k ], let y _k = -(u _k-1 -C _k-1 )/(C _k -u _k +u _k-1 -C _k-1 );

(c7) If u _k <C _k and u _k-1 ≤ C _k-1 , it means that the modulated wave is smaller than the carrier in the time range of t∈(t _k-1 ,t _k ], set y _k <-1.0, for example, y _k = -2.0.

Step 4. Perform interpolation or integral calculation according to the logic value and transition time of the PWM control signal to obtain the subsequent switching action instructions of the controlled power electronic switching device model, so as to realize reinitialization and suppress numerical oscillation, and finally obtain a more accurate PWM converter simulation results.

The controlled power electronic switching device model is shown in Figure 4, the controlled power electronic switching device model is connected to the PWM modulator model, receives the PWM control signal y output by the PWM modulator model, and according to the logic of the PWM control signal y Interpolation or integral calculation is performed on the value and jump time to obtain its own subsequent switch action instructions to realize reinitialization and suppress numerical oscillation; record the current simulation time as t _k , and the previous fixed step simulation time as t _k-1 , according to t The PWM control signal _{y k} at time k judges the switching action state of the controlled power electronic switching device model:

① If y _k ∈ (1.0,+∞), it means that there is no switching action within the time range of t ∈ (t _k-1 ,t _k ], and it is in the open state;

② If y _k ∈ (0.0,1.0], it means that t ∈ (t _k-1 ,t _k ] changes from off to on within the time range, and the change time is t _s = y _k t _k + (1-y _k )·t _k-1 ;

③ If y _k = 0.0, it means that there is no switching action within the time range of t∈(t _k-1 ,t _k ];

④ If y _k ∈ [-1.0,0.0), it means that t ∈ (t _k-1 ,t _k ] time range changes from on to off, and the change time is t _s = -y _k t _k + (1 +y _k )·t _k-1 ;

⑤ If y _k ∈ (-∞, -1.0), it means that there is no switching action within the time range of t ∈ (t _k-1 , t _k ], and it is in the off state.

Step 5. End the simulation calculation of the current step length, push the simulation time one step further and judge whether the simulation is over: if yes, end; otherwise, return to Step 2.

The above is only a preferred embodiment of the present invention, it should be pointed out that for those of ordinary skill in the art, without departing from the principle of the present invention, some improvements and modifications can also be made, and these improvements and modifications are also possible. It should be regarded as the protection scope of the present invention.

Claims (6)

1. A control signal characterization method of a PWM gate control system, characterized in that: the control signal is a digital variable, and its value range and memory occupation are set to be exactly the same as double-precision floating-point data, and the current simulation time is recorded is t _k , and the previous fixed-step simulation time is t _k-1 , according to the value of the control signal y _k at time t _k , it is represented as the following event extended logic variable: ① When y _k ∈ (1.0,+∞), the event expansion logic variable of the control signal within the time range t ∈ (t _k-1 ,t _k ] is represented as 1; ② When y _k ∈ (0.0,1.0], the event expansion logic variable of the control signal within the time range of t ∈ (t _k-1 ,t _k ] is represented as a rising edge transition from 0→1, and the transition time is t _s =y·t _k +(1-y)·t _k-1 ; ③ When y _k = 0.0, it indicates that there is no change in the event expansion logic variable within the time range of the control signal t∈(t _k-1 ,t _k ], and the event expansion logic variable at the end of the previous step is maintained; ④ When y _k ∈ [-1.0,0.0), the event expansion logic variable of the control signal within the time range t ∈ (t _k-1 , t _k ] is represented as a falling edge transition from 0→1, and the transition The moment is t _s =-y·t _k +(1+y)·t _k-1 ; ⑤ When y _k ∈ (-∞, -1.0), the event expansion logic variable of the control signal in the time range of t ∈ (t _k-1 , t _k ] is represented as 0. 2. A PWM modulator model based on the control signal characterization method described in right 1, characterized in that: the PWM modulator model includes an input terminal and an output terminal; the input terminal is used to input the modulation wave signal u; the modulation wave signal The PWM control signal y is generated after u is compared with the carrier signal C; the output terminal is connected to the model of the controlled power electronic switching device for outputting the PWM control signal y; record the current simulation time as t _k , the input and output at time t _k are u _k and y _k respectively, the previous fixed-step simulation time is t _k-1 , the input and output at time t _k-1 are u _k-1 and y _k-1 respectively, and the carrier period is T, then t _k The update process of the PWM control signal y _k at the moment is: (a) calculate the time _{offset toff} = _tk -floor( _tk /T) in the carrier period at moment tk, wherein floor( ) represents rounding down; enter step (b); (b) according to the type of the carrier and t _off , calculate the carrier signal value C _k at t _k moment; enter step (c); (c) Calculate y _k according to the relationship between C _k and u _k : (c1) If u _k ＞C _k and u _k-1 ≥C _k-1 , it means that the modulated wave is larger than the carrier wave within the time range of t∈(t _k-1 ,t _k ], set y _k ＞1.0; (c2) If u _k > C _k and u _k-1 < C _k-1 , it indicates that the relationship between the modulating wave and the carrier has changed within the time range of t∈(t _k-1 ,t _k ], let y _k = (C _k-1 -u _k-1 )/(u _k -C _k +C _k-1 -u _k-1 ); (c3) If u _k ＝C _k and u _k-1 >C _k-1 , it means that the modulating wave is larger than the carrier in the time range of t∈(t _k-1 ,t _k ), and the modulating wave and the carrier are exactly equal at time t _k , let y _k =-1.0; (c4) If u _k ＝C _k and u _k-1 ＝C _k-1 , it is abnormal, let y _k ＝y _k-1 ; (c5) If u _k ＝C _k and u _k-1 <C _k-1 , it means that the modulating wave is smaller than the carrier in the time range of t∈(t _k-1 ,t _k ), and the modulating wave and the carrier are exactly equal at time t _k , let y _k =1.0; (c6) If u _k <C _k and u _k-1 >C _k-1 , it indicates that the relationship between the modulation wave and the carrier wave has changed within the time range of t∈(t _k-1 ,t _k ], let y _k = -(u _k-1 -C _k-1 )/(C _k -u _k +u _k-1 -C _k-1 ); (c7) If u _k <C _k and u _k-1 ≤ C _k-1 , it means that the modulated wave is smaller than the carrier in the time range of t∈(t _k-1 ,t _k ], and set y _k <-1.0. 3. The PWM modulator model according to claim 2, characterized in that: the carrier wave is a triangle wave, a forward sawtooth wave or a backward sawtooth wave. 4. A controlled power electronic switching device model based on the control signal characterization method described in right 1, characterized in that: the controlled power electronic switching device model is connected with the PWM modulator model, and receives the PWM control output of the PWM modulator model signal y, and perform interpolation or integral calculation according to the logic value and transition time of the PWM control signal y, to obtain its own follow-up switching action command, so as to realize reinitialization and suppress numerical oscillation; record the current simulation time as t _k , and the previous fixed step The long simulation time is t _k-1 , and the switching action state of the controlled power electronic switching device model is judged according to the PWM control signal y _k at time t _k : ① If y _k ∈ (1.0,+∞), it means that there is no switching action within the time range of t ∈ (t _k-1 ,t _k ], and it is in the open state; ② If y _k ∈ (0.0,1.0], it means that t ∈ (t _k-1 ,t _k ] changes from off to on within the time range, and the change time is t _s = y _k t _k + (1-y _k )·t _k-1 ; ③ If y _k = 0.0, it means that there is no switching action within the time range of t∈(t _k-1 ,t _k ]; ④ If y _k ∈ [-1.0,0.0), it means that t ∈ (t _k-1 ,t _k ] time range changes from on to off, and the change time is t _s = -y _k t _k + (1 +y _k )·t _k-1 ; ⑤ If y _k ∈ (-∞, -1.0), it means that there is no switching action within the time range of t ∈ (t _k-1 , t _k ], and it is in the off state. 5. The controlled power electronic switching device model according to claim 4, wherein the controlled power electronic switching device is an SCR, IGBT, MOSFET, GTO or an ideal switch. 6. A high-precision PWM converter electromagnetic transient simulation method based on the control signal characterization method described in right 1, is characterized in that: comprise the steps: (1) Simulation initialization, including control sub-model sorting, electrical equation matrix and state quantity initialization, and initial simulation time setting; (2) Perform one-step simulation calculation on all control subsystem models to obtain modulated wave signals; (3) Carry out one-step simulation calculation to all PWM modulator models, obtain the logic value and transition time of PWM control signal; (4) According to the logic value and transition time of the PWM control signal, interpolation or integral calculation is performed to obtain the subsequent switching action instructions of the controlled power electronic switching device model, so as to realize reinitialization and suppress numerical oscillation, and finally obtain more accurate Simulation results of PWM converter; (5) End the simulation calculation of the current step length, push the simulation time one step further and judge whether the simulation is over: if yes, end; otherwise, return to step (2).