Power Converter and Network Model Construction Method Based on Parallel Control of MPC and PIC

Abstract

The present invention discloses a power converter and a network model construction method based on parallel control of MPC and PIC, including: a model predictor MPC, which receives an externally input external regulation signal, adjusts the duty cycle of the output pulse switch signal, and generates an output signal; a proportional-integral controller PIC, which receives an externally input analog signal or an internal local reference signal to obtain a PWM signal; a PWM merging unit, which acquires the output signal and the PWM signal and performs a merging and superposition process; a converter and a load, which output a feedback signal; a detection unit, which outputs a switching control signal; a first multiplexer, whose output is connected to the proportional-integral controller PIC, receives the switching control signal, and inputs an analog signal corresponding to the internal local reference signal or the external regulation signal to the proportional-integral controller PIC; and a second multiplexer is connected between the model predictor MPC and the PWM merging unit and performs output signal g MPC output or cut off the output.

Description

Technical Field

[0001] The present invention relates to the technical field of power converters, and in particular to a power converter based on parallel control of MPC and PIC and a method for constructing a network model. Background Art

[0002] The power converter in this article is used in but not limited to the hydropower industry. Taking hydropower as an example, it is a reliable, carbon-free and widely used renewable energy source. Hydropower can provide rapid power supply to the power grid. At present, the digital transformation of hydropower stations can improve energy production efficiency. However, with the establishment of digital and Internet of Things systems, the control systems of power stations have gradually become vulnerable to attacks. Among various hydropower stations, variable-speed hydropower stations equipped with doubly-fed induction motors are widely used due to their high dynamic stability characteristics. The excitation control system of a doubly-fed induction motor consists of a power electronic converter (i.e., a power converter), a controller, and a sensor. During the digitalization process, digital control methods provide the Internet of Things capabilities for the power electronic converters in the system. These networked converters can communicate with each other and can also serve as a control center and an energy Internet center. However, the Internet of Everything has also raised network security issues.

[0003] Currently, network security issues related to the production of hydropower plants are usually studied from the perspectives of systems and communications. In these studies, the possible intentional and unintentional attacks that the converter itself may be subjected to are not considered as possible events. These attacks will cause changes in the set points or reference values of the system, and may cause the converter output (voltage or current) to exceed the acceptable range. In addition, for variable-speed hydro-generator units, the converter plays an important role in the continuous synchronization of the units, and the impact caused by network attacks will be greater. At the same time, the load will also be adversely affected and the power semiconductor may be damaged.

[0004] Therefore, there is an urgent need to propose a safe, reliable, and simple-structured power converter based on parallel control of MPC and PIC and a method for constructing a network model. Summary of the Invention

[0005] In view of the above problems, the purpose of the present invention is to provide a power converter based on parallel control of MPC and PIC and a method for constructing a network model. The technical solution adopted by the present invention is as follows:

[0006] In the first part, the present technology provides a power converter based on parallel control of MPC and PIC, which includes:

[0007] A model predictor MPC, which receives an externally input external regulation signal r * ext [k], and adjusts the duty cycle d of the output pulse switch signal MPC , and according to the duty cycle d MPCGenerate the output signal g MPC ;

[0008] A proportional-integral controller PIC, externally connected to the external control signal r * ext The analog signal r corresponding to [k] * ext (t) or the internal local reference signal r * in (t), and perform proportional-integral processing to obtain the PWM signal g ana ;

[0009] A PWM combining unit, connected to the model predictor MPC and the proportional-integral controller PIC, to obtain the output signal g MPC and the PWM signal g ana , and perform combining and superposition processing;

[0010] A converter and a load, to obtain the signal after combining and superposition processing, and feedback the feedback signal r(t) to the model predictor MPC and the proportional-integral controller PIC;

[0011] A detection unit, connected to the model predictor MPC and the proportional-integral controller PIC, to compare the output signal g MPC with the PWM signal g ana and output a switching control signal;

[0012] A first multiplexer, with its output connected to the proportional-integral controller PIC, receiving the switching control signal, and inputting the internal local reference signal r * in (t) or the external control signal r * ext The analog signal r corresponding to [k] * ext (t); and

[0013] A second multiplexer, connected between the model predictor MPC and the PWM combining unit, and outputting or cutting off the output signal g MPC output or the signal g MPC output.

[0014] The second part, the present technology provides a method for constructing a network model of a power converter using parallel control based on MPC and PIC, which includes the following steps:

[0015] Establish a mathematical model of the parallel framework;

[0016] Obtain the equivalent power plant model in the model predictor MPC according to the mathematical model;

[0017] Obtain the duty cycle of the pulse switching signal output by the model predictor MPC.

[0018] Compared with the prior art, the present invention has the following beneficial effects:

[0019] (1) The present invention ingeniously adopts the parallel operation of the model predictor MPC and the proportional-integral controller PIC. Under normal circumstances, parallel control is performed according to the external regulation signal of the external wide area network; when the system is attacked, it is judged according to the detection unit, and the first multiplexer and the second multiplexer are switched to ensure that the system uses the internal local reference signal for closed-loop control, reliably isolating the model predictor MPC and ensuring its safety and reliability;

[0020] (2) The present invention is provided with a PWM combining unit, which combines the output signal g MPC and the PWM signal g ana . Its functions are as follows: First, under normal circumstances, the signal g is composed of g ana and g MPC . Once the digital MPC is attacked by the network, the PWM combining unit will make g become g ana ; Second, g MPC takes physical isolation measures for the transformation of the signal g after the converter is attacked by the network, and this isolation action does not affect the normal operation of g ana . At the same time, the operation of the analog PIC is also completely independent of the MPC.

[0021] (3) The present invention ingeniously sets a time delay element z -n and a zero-order hold. Its functions are as follows: After the feedback r is transmitted to the MPC through the ADC sampling unit, the output d MPC of the MPC will be delayed by z -n within the control time of n cycles (reflecting the sampling time delay and the control delay). The delayed signal is then added to d ana after passing through the zero-order hold (ZOH). The obtained output d will be used to drive the power converter.

[0022] (4) For the proportional-integral controller PIC of the present invention, when under network attack, the operation of the detection unit should not depend on the digital controller, thus avoiding network attacks.

[0023] In summary, the present invention has the advantages of safety, reliability, simple structure, etc., and has high practical value and popularization value in the technical field of power converters. Description of the Drawings

[0024] To more clearly illustrate the technical solutions of the embodiments of the present invention, the following will briefly introduce the accompanying drawings required for the embodiments. It should be understood that the following drawings only show some embodiments of the present invention, and thus should not be regarded as limiting the scope of protection. For those skilled in the art, without creative efforts, other related drawings can also be obtained based on these drawings.

[0025] Figure 1 It is a schematic structural diagram of the present invention.

[0026] Figure 2 It is a diagram of the parallel control mathematical model of the present invention.

[0027] Figure 3 It is the PWM signal d in the calculation of the equivalent power plant model of the present invention ana (s) Division schematic diagram.

[0028] Figure 4 It is a schematic diagram of the combination of the external regulation signal and the digital feedback signal in the present invention.

[0029] Figure 5 It is a schematic diagram of the principle of the equivalent power plant model of the present invention.

[0030] Figure 6 It is a schematic diagram of the working principle of the present invention. Specific embodiments

[0031] To make the purpose, technical solutions and advantages of the present application clearer, the following further illustrates the present invention with reference to the drawings and embodiments. The embodiments of the present invention include but are not limited to the following embodiments. Based on the embodiments in the present application, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present application.

[0032] In this embodiment, the term "and / or" only describes the associated relationship of the associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist simultaneously, and B exists alone.

[0033] The terms "first" and "second" in the description and claims of this embodiment are used to distinguish different objects, rather than to describe the specific order of the objects. For example, the first target object and the second target object are used to distinguish different target objects, rather than to describe the specific order of the target objects.

[0034] In the embodiments of the present application, words such as "exemplary" or "for example" are used to represent examples, illustrations, or explanations. Any embodiment or design solution described as "exemplary" or "for example" in the embodiments of the present application should not be construed as being more preferred or having more advantages than other embodiments or design solutions. Rather, the use of words such as "exemplary" or "for example" is intended to present relevant concepts in a specific manner.

[0035] In the description of the embodiments of the present application, unless otherwise specified, the meaning of "a plurality of" refers to two or more. For example, a plurality of processing units refers to two or more processing units; a plurality of systems refers to two or more systems.

[0036] As Figures 1 to 6 shown, this embodiment provides a power converter and a network model construction method based on parallel control of MPC and PIC, which uses parallel MPC and PIC to simultaneously implement closed-loop control to enhance the network security of the power electronic converter. Among them, the digital MPC uses the model of the controlled system to predict the system behavior and future optimal actions under multivariable constraints, and the analog PIC is used to meet the basic uninterrupted requirements and zero-order steady-state error requirements.

[0037] In this embodiment, the power converter mainly includes a model predictor MPC, a proportional-integral controller PIC, a PWM merging unit, a converter and a load, a detection unit, a first multiplexer, and a second multiplexer. Among them, the model predictor MPC is externally connected to an external regulation signal r * ext [k], and is internally connected to a feedback signal, and adjusts the duty cycle d * ext of the output pulse switch signal according to the feedback signal and the external regulation signal r MPC . A PWM module is set in the model predictor MPC, which generates an output signal g MPC according to the duty cycle d MPC . It should be noted that in this embodiment, t is a continuous-time variable, and n and k are discrete-time variables.

[0038] In addition, the proportional-integral controller PIC is normally externally connected to the analog signal r * ext [k] corresponding to the external regulation signal r * ext (t), and when a network attack occurs, it switches to an internal local reference signal r * in (t). The proportional-integral controller PIC uses the feedback signal as the adjustment comparison.

[0039] In this embodiment, the PWM combining unit is composed of a multiplexer of CD54HC153 and an exclusive-OR gate of model 74LVC1G86. It obtains the output signal gMPC and the PWM signal gana, and performs a combining and superposition process. In this embodiment, the converter and the load are an equivalent power plant model, and the output feedback signals are respectively given to the model predictor MPC and the proportional-integral controller PIC. In this embodiment, the detection unit does not rely on a digital controller. It compares the outputs of the model predictor MPC and the proportional-integral controller PIC to obtain an accurate and reliable attack signal, and drives the first multiplexer and the second multiplexer to perform reliable switching.

[0040] The following details the network model construction process of this embodiment.

[0041] The first step is to establish a mathematical model of the parallel framework, including the following steps:

[0042] The transfer function G ana (s) of the proportional-integral controller PIC, and the expression of the mathematical model of the parallel framework in the s domain is:

[0043] G ana (s) = K p + K i / s (1)

[0044] d ana (s) = (K p + K i / s)(r* - r) (2)

[0045] Wherein, K p is the proportional coefficient, K i is the integral coefficient, s represents the independent variable in the Laplace domain (s domain), r * represents the external regulation signal r * ext [k] corresponding to the analog signal r * ext (t) and the external regulation signal r * ext [k]; r represents the abbreviation of the feedback signal r(t) and the digital feedback signal r[k]; d ana (s) represents the PWM signal output by the proportional-integral controller PIC in the s domain.

[0046] The second step is to obtain the equivalent power plant model in the model predictor MPC according to the mathematical model, including the following steps:

[0047] In the s domain, the PWM signal d ana (s) output by the proportional-integral controller PIC is divided into the following two parts, and its expression is:

[0048]

[0049] Among them, r * (s) represents the external control signal r in the s-domain * ext The analog signal corresponding to r[k] * ext (t) and the external control signal r * ext [k]; r(s) represents the feedback signal r(t) and the digital feedback signal r[k] in the s-domain ana-1 and d ana-2 respectively represent r * and r output through the same PIC;

[0050] The external control signal r * ext The analog signal corresponding to r[k] * ext (t) and the external control signal r * ext [k] is abbreviated as r * and the feedback signal r(t) and the digital feedback signal r[k] are merged. Here, the variable r(t) representing the analog signal and the digital signal r[k] after sampling the analog signal are represented by the same symbol r, resulting in formula (4), and its expression is:

[0051]

[0052] Among them, d ana-1 (s) represents the form of the output signal of r * after Laplace transform in the s-domain after passing through the proportional-integral controller PIC; d ana-2 represents the form of the output signal of r after Laplace transform in the s-domain after passing through the proportional-integral controller PIC; d equ (s) represents the duty cycle d MPC delayed by n time units (the delay is manifested as multiplying by z -n ) and then merged with d ana-1 The form after Laplace transform in the s-domain of the merged signal; H(s) is the transfer function of the power converter, that is, the mathematical model in the s-domain.

[0053] Substitute r(s)G ana (s) = d ana-2 into formula (1) to obtain formulas (5) and (6); according to formulas (5) and (6), obtain the equivalent power plant model between the feedback signal and the duty cycle of the pulse switch signal in the model predictive controller MPC, and its expression is:

[0054] r(s) = d equ (s)H(s) / [1 + G ana (s)H(s)] (5)

[0055] H equ (s) = H(s) / [1 + G ana (s)H(s)] (6)

[0056] [d MPC (s) + d ana-1 (s)]H equ (s) = r(s) (7)

[0057] Among them, H equ (s) represents the transfer function of the equivalent power plant model, that is, the corresponding function in the s-domain, d MPC (s) represents d MPC The corresponding function in the s-domain.

[0058] In the third step, obtain the duty cycle of the pulse switch signal of the output of the model predictor MPC, including the following steps:

[0059] The equivalent power plant transfer function H equ The expression of the discrete form (i.e., after z-transform) of (s) is:

[0060]

[0061] Among them, z represents the independent variable in the z-transform space, z -n Represents the control delay of n cycles of the time delay element, e represents the natural logarithm; Represents the transfer function of the zero-order hold; T s Is the time constant.

[0062] Substitute formula (8) into formula (5) to obtain the relationship between the feedback signal r(z) and d equ( z) in the z-domain:

[0063] r(z) / d equ (z) = H equ (z) (9)

[0064] Among them, r(z) represents the function corresponding to the feedback signal in the z-domain, d equ (z) is the sum of d ana-1 And d MPC The function in the z-domain.

[0065] Obtain the predicted value of the feedback data after n + 1 cycles Its expression is:

[0066]

[0067] Among them, a and b represent weighting coefficients, and their values range from 0 to 1.

[0068] Determine the minimized predefined cost function W of the model predictor MPC, and its expression is:

[0069]

[0070] Among them, λ represents the control coefficient of the model predictor MPC;

[0071] The minimized predefined cost function W obtains the minimum value through its partial derivative d MPC[k] being zero, and its expression is:

[0072] δW / δdMPC[k] = 0 (12)

[0073] Eliminate d equ contained in formula (10) to obtain formula (13);

[0074] d equ [k] = d MPC [k] + d ana-1 [k] (13)

[0075] Among them, d equ [k] represents the duty cycle d MPC after being delayed by n time units through a zero-order hold and the digital signal formed by the sum with d ana-1 d MPC [k] represents the digital signal output by the model predictor MPC d ana-1 [k] represents the external regulation signal r * the digital signal d output after passing through the PIC ana-1 (s) represents the digital signal output after r * passes through the proportional-integral controller PIC.

[0076] According to the discrete form of G ana (s) in the z-domain, eliminate d ana-1 [k], and its expression is:

[0077] G ana (z) = K p + zK i T s / (z - 1) (14)

[0078] According to formula (3) and formula (14), obtain the digital signal d * output after the external regulation signal r ana-1 [k] passes through the PIC, and it is expressed as:

[0079] (K p +K i T s )r * [k]-K p r * [k - 1]+d ana-1 [k - 1](15)

[0080] Substitute equations (13) to (15) into equation (10) to obtain the prediction value that is only related to the output d MPC [k] of the MPC Its expression is:

[0081]

[0082] where a p is the coefficient of the proportional operator, and a int is the coefficient of the integral operator.

[0083] Substitute equation (16) into equation (12) to obtain the calculated value of the discrete signal of the equivalent power plant output obtained without considering the feedback variable constraint Its expression is:

[0084]

[0085] where f(x) is used to represent the variable related to the calculated value of the equivalent power plant output

[0086] Optimize equation (17) according to the duty cycle interval to obtain the final discrete signal d MPC [k] of the equivalent power plant output, and its expression is:

[0087]

[0088] where d ana [k] represents the sampled value of the square wave signal.

[0089] The control process of this embodiment is described in detail below:

[0090] Under normal circumstances, the control signal required by the power plant is composed of the output signal g MPC and the PWM signal g ana together. Once the model predictor MPC is under a cyber attack, the separate PWM signal g ana is used for control. In addition, the output signal g MPC is physically isolated after the converter is under a cyber attack, and this isolation action does not affect the normal operation of g ana ; meanwhile, the operation of the analog PIC is completely independent of the MPC.​

[0091] As shown in Figure 6 (a), Module 1 is a carrier generation circuit for obtaining the sawtooth wave carrier v of the analog PIC car . The period of v car is determined by the clock signal v clk of the oscillator. The analog PIC consists of an operational amplifier, resistors, and capacitors. Its output is compared with vcar to obtain an analog PWM signal. Such a signal is processed by the latch in Module 2 to avoid mis-triggering and outputs g ana . In this way, the operation of the analog PIC will be completely independent of the DSP. At the same time, v clk is also used to trigger the DSP; then the switching period of the DSP can be synchronized with the switching period of the analog PIC. The synchronized g MPC can be directly generated by the PWM unit of the DSP. Finally, the PWM signals g ana and g MPC are transmitted to the multiplexer (CD54HC153) and the exclusive-OR gate (74LVC1G86) to complete the merging and isolation functions.

[0092] In Figure 6 , the truth table of the multiplexer CD54HC153 is shown in the figure. The multiplexer has two control ports S0 and S1 and two output ports 1Y and 2Y. According to the states of S0 and S1, 1Y and 2Y are selected to be connected to one of the inputs g ana , g MPC or logic 0. The outputs 1Y and 2Y are combined together through the exclusive-OR gate 74LVC1G86 to obtain the combined PWM output signal.

[0093] As shown in Figure 6 (c) to (g), under normal circumstances, S0 = 0, S1 = 0, then gana and gMPC are selected. Since in Equation (17) can be positive or negative, and the combined PWM signal g can be greater than or less than g ana . Therefore, when the combined PWM signal g is greater than g ana , the DSP generates g MPC , and the falling edge of g MPC coincides with the rising edge of g ana (as shown in Figure 6 (c)); while when g is less than g ana , the DSP generates g MPC , and its rising edge coincides with the rising edge of g ana (as shown in Figure 6 (d)). Through the exclusive-OR operation, the duty cycle of g will always be the sum of d ana and d MPC .

[0094] The process of detecting an attack is as follows: During the initial parallel operation, the reference inductor signal is changed under the control of the supervised controller, and the inductor current i L can quickly track At this time, the signal overshoot is zero, the rise and settling times are short, and the steady-state error is zero. Assume that the digital MPC is under a cyber attack and is hijacked. Due to the cyber attack detection delay, i L will still track the hijacked Due to such a delay, the output g of the PWM merging unit is still composed of g ana and g MPC At this time, once a cyber attack is identified, S0 is set to 1, and the hijacked g MPC is physically isolated from the control loop through the multiplexer; therefore, g is only determined by g ana At the same time, the analog PIC makes i L track the local reference. Since only the analog PIC is used, signal overshoot and long-term stable signals can be observed. Finally, once the cyber attack disappears, the detection unit sets S0 to 0, and g MPC is reconnected to the control loop, and i L tracks the reference of the supervised controller again.

[0095] Once the DSP is under a cyber attack, Figure 1 the detection unit shown in Module 3 sets S0 to 1; then the 2Y port is forced to be set to O, while g ana is still transmitted to the 1Y port. In this way, the digital MPC is in a physically isolated state, the XOR gate acts, and the value of g becomes g ana When S0 = 0 and S1 = 1, the 1Y port is forced to be set to 0, while g MPC is still transmitted to the 2Y port (as shown in Figure 6 (f)). Then g actually becomes g MPC after the XOR operation. Finally, when S0 = 1 and S1 = 1, then g is always 0 and the converter is turned off (as shown in Figure 6 (g)).

[0096] The above embodiments are only the preferred embodiments of the present invention and do not limit the protection scope of the present invention. Any changes made by adopting the design principle of the present invention and non-creative labor on this basis shall fall within the protection scope of the present invention.

Claims

1. A power converter based on parallel control of MPC and PIC, characterized in that Including: Model Predictor MPC, with an externally regulated signal r input externally * ext [k], and adjust the duty cycle d of the output pulse switch signal MPC , according to the duty cycle d MPC Generate the output signal g MPC ; Proportional-integral controller PIC, external control signal r input externally * ext Analog signal r corresponding to [k] * ext (t) or internal local reference signal r * in (t), and perform proportional-integral processing to obtain PWM signal g ana ; A PWM merging unit, connected to a model predictor MPC and a proportional-integral controller PIC, obtains an output signal g MPC and a PWM signal g ana , and performs merging and superposition processing; A converter and a load, which obtain the signal after merging, superposing and processing, and feedback the feedback signal r(t) to the model predictor MPC and the proportional-integral controller PIC; The detection unit, connected to the model predictor MPC and the proportional-integral controller PIC, for the output signal g MPC with the PWM signal g ana is compared and a switching control signal is output; The first multiplexer, the output of which is connected to the proportional-integral controller (PIC), receives a switching control signal and inputs an internal local reference signal r * in (t) or an external regulation signal r * ext The analog signal r * ext (t); And A second multiplexer, connected between the model predictor MPC and the PWM merging unit, and outputs the signal g MPC Outputs or cuts off the signal g MPC Output; A time delay unit z is also arranged between the model predictor MPC and the PWM combining unit -n and a zero-order hold unit; the model predictor MPC delays the duty cycle d MPC by z-n in the control time of n cycles and holds it with the zero-order hold unit; The PWM merging unit is an adder.

2. The power converter based on parallel control of MPC and PIC according to claim 1, characterized in that The PWM merging unit is composed of a multiplexer of model CD54HC153 and an exclusive-OR gate of 74LVC1G86.

3. The power converter based on parallel control of MPC and PIC according to claim 1, wherein The converter and the load are an equivalent power plant model.

4. A method for constructing a network model of a power converter based on parallel control of MPC and PIC as described in claim 1 or 2 or 3, characterized in that, Including the following steps: Establishing a mathematical model of the parallel framework; Obtaining the equivalent power plant model in the model predictor MPC according to the mathematical model; Obtaining the duty cycle of the pulse switching signal output by the model predictor MPC.

5. A method for constructing a network model of a power converter using parallel control based on MPC and PIC according to claim 4, characterized in that, Establishing a mathematical model of the parallel framework, including the following steps: The transfer function G ana of the proportional-integral controller (PIC), and the expression of the mathematical model of the parallel framework in the Laplace domain (s-domain) is as follows: G ana (s) = K p + K i / s (1) d ana (s) = (K p + K i / s)(r * - r') (2) Among them, K p is the proportionality coefficient, and K i is the integral coefficient. s represents the independent variable in the Laplace domain (s-domain), and r * represents the external control signal r * ext The analog signal corresponding to [k] is r * ext (t), and it is the abbreviation of the external control signal r * ext [k]; r It represents the abbreviation of the feedback signal r(t) and the digital feedback signal r[k]; d ana (s) represents the square wave signal output by the proportional-integral controller PIC in the s-domain.

6. A method for constructing a network model of a power converter using parallel control based on MPC and PIC according to claim 5, characterized in that, Obtaining the equivalent power plant model in the model predictor MPC according to the mathematical model, including the following steps: In the s-domain, the PWM signal d ana (s) output by the proportional-integral controller (PIC) is divided into the following two parts, and its expression is: where r * (s) represents the external control signal r in the s-domain * ext The analog signal corresponding to r[k] * ext (t) and the external control signal r * ext Abbreviation of r[k]; r(s) represents the abbreviation of the feedback signal r(t) and the digital feedback signal r[k] in the s-domain; d ana-1 and d ana-2 respectively represent the outputs of r * and r through the same PIC; The external regulation signal r * ext The analog signal r corresponding to [k] * * ext (t) and the external regulation signal r * * ext The abbreviation r of [k] * * And the abbreviation r of the feedback signal r(t) and the digital feedback signal r[k] are combined to obtain Equation (4), and its expression is: where d ana-1 (s) represents the form in the s-domain after performing the Laplace transform on the output signal after passing through the proportional-integral controller PIC, d * represents the form in the s-domain after performing the Laplace transform on the output signal after passing r through the proportional-integral controller PIC; d ana-2 represents the form in the s-domain after performing the Laplace transform on the output signal after passing r through the proportional-integral controller PIC; d equ (S) represents the duty cycle d MPC after delaying by n time units and merged with d ana-1 the form in the s-domain after performing the Laplace transform on the signal; H(s) is the transfer function of the power converter, i.e., the mathematical model in the s-domain; Substitute r(s)G ana (s) = d ana-2 into formula (1) to obtain formulas (5) and (6); an equivalent power plant model between the feedback signal and the duty cycle of the pulse switch signal in the model predictor MPC is obtained according to formulas (5) and (6), and its expression is: r(s) = d equ (s)H(s) / [1 + G ana (s)H(s)] (5) H equ (s) = H(s) / [1 + G ana (s)H(s)] (6) [d MPC (s) + d ana-1 (s)]H equ (s) = r(s) (7) Among them, H equ (s) represents the form of the equivalent power plant model in the s-domain, that is, the transfer function of the equivalent power plant, and d MPC (s) represents the duty cycle d MPC after performing the Laplace transform and its form in the s-domain.

7. A method for constructing a network model of a power converter using parallel control based on MPC and PIC as claimed in claim 6, characterized in that, Obtaining the duty cycle of the pulse switching signal output by the model predictor MPC, including the following steps: The equivalent power plant transfer function H equ (S) in discrete form (i.e., after z-transform) is expressed as: where z represents the independent variable in the z-transform space, and z -n represents the n control time delays of the time delay unit, and e represents the natural logarithm; represents the transfer function of the zero-order hold; T s is the time constant; Substituting Equation (8) into Equation (5) gives the relationship between the feedback signal r(z) and d equ (z) in the z-domain: r(z) / d equ (z) = H equ (z) (9) where r(z) represents the form of the feedback signal in the z-domain after z-transform, and d equ (z) is d MPC after being delayed by n time units through a zero-order hold and then added to d ana-1 and represents the form in the z-domain after z-transform of the sum; Obtain the predicted value of the feedback data after n + 1 cycles Its expression is: Wherein, a and b represent weighting coefficients, and their values are between 0 and 1; Determining the minimized predefined cost function W of the model predictor MPC, and its expression is: Wherein, λ represents the control coefficient of the model predictor MPC; Minimize the predefined cost function \(W\) by setting its partial derivative \(\frac{\partial W}{\partial \mathbf{d}} MPC [k]\) to zero, and the expression for the minimum value is: δW / δd MPC [k] = 0 (12) Eliminate d contained in formula (10) equ , to obtain formula (13); d equ [k] = d MPC [k] + d ana-1 [k](13) Among them, d equ [k] represents the duty cycle d MPC After being delayed by n time units through a zero-order hold and added to d ana-1 to form a digital signal, d MPC [k] represents the digital signal output by the model predictor MPC; d ana-1 [k] represents the external regulation signal r * The digital signal output after passing through the proportional-integral controller PIC; d ana-1 (s) represents r * The digital signal output after passing through the PIC; Eliminate d ana in the discrete form of G ana-1 [k] in the z-domain, and its expression is: G ana G(z) = K p + zK i T s / (z - 1) (14) The external regulation signal r is obtained according to formula (3) and formula (14). * The digital signal d output after passing through the proportional integral controller PIC ana-1 [k], which is expressed as: (K p +K i T s )r * [k]-K p r * [k - 1]+d ana-1 [k - 1] (15) Substitute formulas (13) to (15) into formula (10) to obtain the prediction value that is only related to the output d MPC [k] of MPC Its expression is as follows: Among them, a p is the coefficient of the proportional arithmetic unit, and a int is the coefficient of the integral arithmetic unit; Substitute Equation (16) into Equation (12) to obtain the calculated value of the equivalent power plant output discrete signal obtained without considering the feedback variable constraint Its expression is: where f() is used to represent the variable related to the calculated value of the equivalent power plant output ; According to the duty cycle interval, optimize formula (17) to obtain the final discrete signal d MPC [k] of the equivalent power plant output, and its expression is: where d ana [k] represents the sampled value of the square wave signal.

Images