An EAST fast control power supply output current control method and device

Abstract

The application discloses an EAST fast control power supply output current control method and device, and the method comprises the following steps: predicting current values after two periods according to EAST fast control power supply output current sampling values; establishing a multi-target optimization cost function with constraint conditions; seeking optimal solutions of the multi-target optimization cost function on line by using a particle swarm algorithm; comparing the optimal solution obtained by the particle swarm algorithm with an output voltage amplitude of a single branch H-bridge inverter circuit of the EAST fast control power supply to obtain an output voltage coefficient; multiplying the output voltage coefficient with a carrier amplitude to obtain a modulation value; comparing the modulation value with a carrier modulation to obtain an optimal duty cycle to control the single branch H-bridge inverter circuit of the EAST fast control power supply, so that the EAST fast control power supply output follows a given signal current value; and the application has the advantages of small calculation amount and high control precision.

Description

Technical Field

[0001] This invention relates to the field of EAST fast control power supply operation technology, and more specifically to an EAST fast control power supply output current control method and device. Background Technology

[0002] In the context of current prominent energy issues, nuclear fusion power generation has broad development prospects due to its readily available fusion raw materials and pollution-free fusion products. The Experimental Advanced Superconducting Tokamak (EAST) is an important component of nuclear fusion power generation.

[0003] Nuclear fusion power generation relies on the reaction between plasmas, and plasma displacement control is a key issue. An active feedback controller for plasma vertical displacement detects the plasma's vertical displacement and calculates the setpoint signal for the Fast Control Power Supply (FCPS) based on this displacement. The FCPS then excites the active feedback coil according to the setpoint signal, generating a rapidly changing magnetic field to perform closed-loop feedback control of the plasma in the vertical direction, ensuring the stability of the plasma's vertical displacement. The voltage signal given by the FCPS is linearly correlated with the output current of the FCPS; a setpoint voltage signal of ±10V corresponds to a total output current of ±9000A. Since the entire FCPS system consists of six parallel branches, a setpoint voltage signal of ±10V corresponds to a single branch output current of ±1500A. The supply voltage of a single branch of the FCPS is obtained from the three-phase voltage of the mains through uncontrolled rectification. A single branch is formed by three cascaded H-bridge inverter circuits, therefore the output voltage of a single branch is between ±1620V. The plasma control system (PCS) issues corresponding voltage command signals based on the plasma displacement. The EAST fast control power supply rapidly outputs current based on the voltage command signals to establish the required magnetic field to ensure the balance of plasma displacement.

[0004] In EAST fast-control power supply systems, rapid output of voltage and current signals is a crucial performance indicator. However, due to the switching losses of high-power devices, the switching frequency of the power supply is difficult to increase, thus limiting the rapid output voltage and current performance. To achieve rapid voltage and current output, the duty cycle of the switching devices in the circuit undergoes drastic changes. However, in the initial stage of current build-up, overshoot may occur, causing the output current to exceed the rated current value, which is detrimental to the long-term operation of the power devices. Simultaneously, to quickly establish the magnetic field, the output voltage may experience large abrupt changes in adjacent operating cycles, increasing the ripple of the output current and affecting the stable operation of the circuit. Therefore, the rapid output of voltage and current may be accompanied by problems of output current overshoot and excessive ripple, posing a significant challenge to the operation of power devices. In severe cases, temporary overcurrent may even occur, burning out the power devices.

[0005] The EAST fast control power supply features a six-channel parallel structure, all of which share a similar configuration: a single-cascaded H-bridge inverter circuit. For a detailed introduction to the EAST fast control power supply, please refer to Bi Nanxia's master's thesis, "Research and Optimization of the EAST New Fast Control Power Supply Control System," published in April 2018 at Hefei University of Technology. Model predictive control (MMC) is a predictive method that has been applied in cascaded H-bridge inverter circuits, capable of predicting the next stage's output voltage and current values ​​based on the current sampled signal. However, to constrain output current ripple and suppress overshoot, multi-objective optimization terms need to be added, along with constraints in the cost function. Solving cost functions with constraints is challenging. Currently, there are no methods for online solving of constrained MMC equations; the optimal analytical solution must be manually calculated and then incorporated into the control algorithm. When the cost function equations with constraints are complex, this significantly increases the computational load and leads to larger control accuracy errors. Summary of the Invention

[0006] The technical problem to be solved by this invention is that the existing EAST fast control power supply output current control method has a large amount of calculation and large control accuracy error.

[0007] This invention solves the above-mentioned technical problems through the following technical means: an EAST fast control power supply output current control method, the method comprising:

[0008] Step 1: Predict the current value two cycles later based on the EAST fast control power supply output current sampling value;

[0009] Step 2: Establish a multi-objective optimization cost function with constraints;

[0010] Step 3: Use the particle swarm optimization algorithm to find the optimal solution of the multi-objective optimization cost function online;

[0011] Step 4: Compare the optimal solution obtained by the particle swarm optimization algorithm with the output voltage amplitude of the single-branch H-bridge inverter circuit of the EAST fast control power supply to obtain the output voltage coefficient. Multiply the output voltage coefficient with the carrier amplitude to obtain the modulation value. Compare the modulation value with the carrier modulation to obtain the optimal duty cycle to control the single-branch H-bridge inverter circuit of the EAST fast control power supply, so that the output current value of the EAST fast control power supply follows the given signal.

[0012] This invention establishes a multi-objective optimization cost function with constraints, transforming the EAST fast-control power supply output current control problem into a problem of finding the optimal solution of the multi-objective optimization cost function. The particle swarm optimization algorithm is used to find the optimal solution of the cost function online, avoiding the process of manually finding analytical solutions, reducing computational load, and avoiding the selection of weight coefficients for multi-objective optimization terms. The optimal solution obtained by the particle swarm optimization algorithm is compared with the output voltage amplitude of the single-branch H-bridge inverter circuit of the EAST fast-control power supply to obtain the output voltage coefficient, which is then used to derive the duty cycle to control the single-branch H-bridge inverter circuit of the EAST fast-control power supply. This allows the EAST fast-control power supply output to follow the current value of the given signal, achieving precise control of the output current with small control accuracy error.

[0013] Further, step one includes:

[0014] Through formula

[0015]

[0016] Establish the output voltage equation of the H-bridge inverter circuit, where i o Where L is the output current, L is the output load inductance, and R is the internal resistance of the load inductance.

[0017] In a switching cycle T s Within time k, the output voltage equation is discretized. The output voltage equation known at the previous time is discretized and expressed as follows:

[0018]

[0019] The predicted output current at the next cycle time is expressed as:

[0020]

[0021] The predicted output current for the next two cycles is expressed as follows:

[0022]

[0023] Among them, u o (k) represents the output voltage of the H-bridge inverter circuit at time k, i o (k) represents the output current of the H-bridge inverter circuit at time k.

[0024] Furthermore, step two includes:

[0025] The purpose of the entire EAST fast control power supply is to ensure that the output current value follows the given reference signal current i. ref Therefore, the target for the output current in the next two cycles is to match the reference signal current value, hence the establishment of a cost function.

[0026] J1 = [i ref -i o (k+2)] 2 (5)

[0027] To limit the increase in output voltage of the H-bridge inverter circuit, a multi-objective optimization cost function with weight λ is established.

[0028] J2 = [i ref -i o (k+2)] 2 +λ[u o (k)-u o (k-1)] 2 (6)

[0029] Limit the output current and establish a multi-objective optimization cost function with constraints.

[0030]

[0031] Furthermore, step three includes:

[0032] Set the initial parameters for the particle swarm optimization algorithm, use the output voltage of the H-bridge inverter circuit as the individual particles in the population, and use the multi-objective optimization cost function with constraints as the fitness function of the particle swarm optimization algorithm. Execute the particle swarm optimization algorithm within the constraints of the particle optimization parameters. When the maximum number of iterations is reached, the optimal position of the output population is the optimal solution of the multi-objective optimization cost function with constraints.

[0033] Furthermore, the initial parameters for the particle swarm optimization algorithm include: a population size of 200 and a maximum number of iterations of 30.

[0034] Furthermore, the range of limitations for the particle optimization parameters includes:

[0035] Set the particle optimization position parameter limit range as follows:

[0036]

[0037] The speed parameter is set to a limit range of ±1.

[0038] Furthermore, when the particle swarm optimization algorithm is executed, after the maximum number of iterations is reached, the optimal position of the output swarm is the optimal solution of the multi-objective optimization cost function with constraints, including:

[0039] The particle swarm optimization algorithm is executed, iteratively calculating the fitness value of each particle and finding the optimal position of the individual that minimizes the fitness value. While the particles are searching for the minimum individual optimal position, the optimal position of the entire swarm is also constantly updated. By comparing the minimum fitness values ​​of each individual, the optimal position of the swarm that minimizes the overall fitness value is found. After updating the optimal positions of each individual and the optimal position of the swarm, the velocity of the individual is updated within the velocity parameter limit. Within the number of iterations, the algorithm continuously performs the optimization process. When the maximum number of iterations is reached, the output optimal position of the swarm is the optimal solution of the multi-objective optimization cost function with constraints.

[0040] The present invention also provides an EAST fast control power supply output current control device, the device comprising:

[0041] The current prediction module is used to predict the current value two cycles later based on the sampled output current value of the EAST fast control power supply.

[0042] The objective function creation module is used to create a multi-objective optimization cost function with constraints.

[0043] The solver module is used to find the optimal solution of the multi-objective optimization cost function online using the particle swarm optimization algorithm;

[0044] The output current control module is used to compare the optimal solution obtained by the particle swarm optimization algorithm with the output voltage amplitude of the single-branch H-bridge inverter circuit of the EAST fast control power supply to obtain the output voltage coefficient. The output voltage coefficient is multiplied by the carrier amplitude to obtain the modulation value. The modulation value is then compared with the carrier modulation to obtain the optimal duty cycle to control the single-branch H-bridge inverter circuit of the EAST fast control power supply, so that the output current value of the EAST fast control power supply follows the given signal.

[0045] Furthermore, the current prediction module is also used for:

[0046] Through formula

[0047]

[0048] Establish the output voltage equation of the H-bridge inverter circuit, where i o Where L is the output current, L is the output load inductance, and R is the internal resistance of the load inductance.

[0049] In a switching cycle T s Within time k, the output voltage equation is discretized. The output voltage equation known at the previous time is discretized and expressed as follows:

[0050]

[0051] The predicted output current at the next cycle time is expressed as:

[0052]

[0053] The predicted output current for the next two cycles is expressed as follows:

[0054]

[0055] Among them, u o (k) represents the output voltage of the H-bridge inverter circuit at time k, i o (k) represents the output current of the H-bridge inverter circuit at time k.

[0056] Furthermore, the objective function establishment module is also used for:

[0057] The purpose of the entire EAST fast control power supply is to ensure that the output current value follows the given reference signal current i. ref Therefore, the target for the output current in the next two cycles is to match the reference signal current value, hence the establishment of a cost function.

[0058] J1 = [i ref -i o (k+2)] 2 (5)

[0059] To limit the increase in output voltage of the H-bridge inverter circuit, a multi-objective optimization cost function with weight λ is established.

[0060] J2 = [i ref -i o (k+2)] 2 +λ[u o (k)-u o (k-1)] 2 (6)

[0061] Limit the output current and establish a multi-objective optimization cost function with constraints.

[0062]

[0063] Furthermore, the solution module is also used for:

[0064] Set the initial parameters for the particle swarm optimization algorithm, use the output voltage of the H-bridge inverter circuit as the individual particles in the population, and use the multi-objective optimization cost function with constraints as the fitness function of the particle swarm optimization algorithm. Execute the particle swarm optimization algorithm within the constraints of the particle optimization parameters. When the maximum number of iterations is reached, the optimal position of the output population is the optimal solution of the multi-objective optimization cost function with constraints.

[0065] Furthermore, the initial parameters for the particle swarm optimization algorithm include: a population size of 200 and a maximum number of iterations of 30.

[0066] Furthermore, the range of limitations for the particle optimization parameters includes:

[0067] Set the particle optimization position parameter limit range as follows:

[0068]

[0069] The speed parameter is set to a limit range of ±1.

[0070] Furthermore, when the particle swarm optimization algorithm is executed, after the maximum number of iterations is reached, the optimal position of the output swarm is the optimal solution of the multi-objective optimization cost function with constraints, including:

[0071] The particle swarm optimization algorithm is executed, iteratively calculating the fitness value of each particle and finding the optimal position of the individual that minimizes the fitness value. While the particles are searching for the minimum individual optimal position, the optimal position of the entire swarm is also constantly updated. By comparing the minimum fitness values ​​of each individual, the optimal position of the swarm that minimizes the overall fitness value is found. After updating the optimal positions of each individual and the optimal position of the swarm, the velocity of the individual is updated within the velocity parameter limit. Within the number of iterations, the algorithm continuously performs the optimization process. When the maximum number of iterations is reached, the output optimal position of the swarm is the optimal solution of the multi-objective optimization cost function with constraints.

[0072] The advantages of this invention are:

[0073] (1) This invention establishes a multi-objective optimization cost function with constraints, transforming the EAST fast control power supply output current control problem into a problem of finding the optimal solution of the multi-objective optimization cost function. The particle swarm optimization algorithm is used to find the optimal solution of the cost function online, avoiding the process of manually finding analytical solutions, reducing the amount of computation, and avoiding the selection of weight coefficients of multi-objective optimization terms. The optimal solution obtained by the particle swarm optimization algorithm is compared with the output voltage amplitude of the single-branch H-bridge inverter circuit of the EAST fast control power supply to obtain the output voltage coefficient, which is then converted into the duty cycle to control the single-branch H-bridge inverter circuit of the EAST fast control power supply, so that the output of the EAST fast control power supply follows the current value of the given signal, realizing precise control of the output current, thereby reducing the control accuracy error.

[0074] (2) In step one of the present invention, the output current of the EAST fast control power supply is predicted after two cycles. One cycle is to predict the output current in advance so that the output current can quickly follow the reference given signal. The other cycle is to compensate for the loading delay of one cycle in the process of loading the carrier value by the digital control processor. The model prediction is used to predict the output current of the EAST fast control power supply two steps in advance so that the output current can quickly follow the given signal and ensure the balance of the plasma in the vertical displacement.

[0075] (3) The multi-objective optimization cost function with constraints established in step two of this invention mainly consists of multi-objective optimization terms and constraints. The two objectives of the multi-objective optimization terms are to make the output current follow the reference signal quickly after two cycles and to minimize the increase in output voltage. The objective of the constraint terms is to limit the output current value of a single branch to within ±1500A. The multi-objective optimization terms make the output current follow the reference signal quickly, while the output current has a small ripple. The constraint terms make the output current not overshoot, and the output current value will not exceed the range of ±1500A due to the unreasonable selection of the weight coefficient λ. Overall, it avoids the output current overshoot, makes the output current ripple smaller, which is beneficial to the long-term operation of the circuit devices and reduces the impact of the large overshoot current on the circuit.

[0076] (4) In step three of this invention, the particle swarm optimization algorithm is used to find the optimal solution for the multi-objective optimization cost function with constraints established in step two. The process of finding the optimal solution is online, avoiding the process of manually finding analytical solutions and also avoiding the selection of weight coefficient λ. The process of finding the optimal solution is an iterative process, in which the state variable solution u that minimizes the value of the multi-objective optimization cost function with constraints is found within a set number of iterations and a set solution range. o (k).

[0077] (5) The optimal state variable solution u in step four of this invention o (k) The voltage coefficient is obtained by comparing it with ±1620V. The voltage coefficient is then multiplied by the carrier amplitude and modulated with the carrier to obtain the optimal duty cycle. The carrier of the single-branch cascaded H-bridge inverter circuit is modulated by 60° phase shift control, which further increases the switching frequency of the inverter circuit and speeds up the response of the output current. Attached Figure Description

[0078] Figure 1 This is a flowchart of an EAST fast control power supply output current control method disclosed in Embodiment 1 of the present invention;

[0079] Figure 2 This is a flowchart of the particle swarm algorithm in the EAST fast control power supply output current control method disclosed in Embodiment 1 of the present invention;

[0080] Figure 3 This is a schematic diagram comparing the output current waveform of a single branch of the EAST fast control power supply and the given reference voltage waveform in an EAST fast control power supply output current control method disclosed in Embodiment 1 of the present invention. Detailed Implementation

[0081] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0082] Example 1

[0083] The overall block diagram of the EAST fast control power supply in this invention is as follows: Figure 1 As shown, a single-branch cascaded H-bridge inverter circuit is composed of three cascaded inverter H-bridges. An EAST fast-control power supply output current control method is also provided, the method comprising:

[0084] S1: Predict the current value two cycles later based on the EAST fast control power supply output current sampling value; the specific process is as follows:

[0085] Output voltage equation and output current i o The output load inductance L and the load inductance internal resistance R are related. The output voltage equation of the H-bridge inverter circuit can be expressed as:

[0086]

[0087] In a switching cycle T s Within time k, the output voltage equation is discretized. The output voltage equation known at the previous time is discretized and expressed as follows:

[0088]

[0089] The predicted output current at the next cycle time is expressed as:

[0090]

[0091] The predicted output current for the next two cycles is expressed as follows:

[0092]

[0093] Among them, u o (k) represents the output voltage of the H-bridge inverter circuit at time k, i o (k) represents the output current of the H-bridge inverter circuit at time k.

[0094] S2: Establish a multi-objective optimization cost function with constraints; the specific process is as follows:

[0095] The purpose of the entire EAST fast control power supply is to ensure that the output current value follows the given reference signal current i. ref Therefore, the target for the output current in the next two cycles is to match the reference signal current value, hence the establishment of a cost function.

[0096] J1 = [i ref -i o (k+2)] 2 (5)

[0097] To avoid large output current ripple, a multi-objective optimization cost function with weight λ is established to limit the increase in output voltage of the cascaded H-bridge inverter circuit, thereby preventing excessive output current ripple.

[0098] J2 = [i ref -i o (k+2)] 2 +λ[u o (k)-u o (k-1)] 2 (6)

[0099] To prevent overshoot in the output current of the cascaded H-bridge inverter circuit and avoid the current amplitude of a single branch exceeding ±1500A, the output current is limited, and a multi-objective optimization cost function with constraints is established.

[0100]

[0101] The optimal solution for a multi-objective optimization cost function with constraints is the state output variable u that minimizes the cost function value. o (k).

[0102] S3: Use the particle swarm optimization algorithm to find the optimal solution of the multi-objective optimization cost function online; the specific process is as follows:

[0103] The process of finding the optimal solution using the particle swarm optimization algorithm is as follows: Figure 2 As shown, at the beginning of the particle swarm optimization algorithm, the population size is initially selected as 200, and the maximum number of iterations is set to 30. Based on formula (7) and the limitation of ±1620V for the single-branch output voltage, the particle optimization position parameter limit range is set to...

[0104]

[0105] The speed parameter is set to a limit range of ±1.

[0106] After setting the initial parameters of the particle swarm optimization algorithm, the algorithm begins to iteratively calculate the adaptive value of each particle within the constraints of the particle optimization position parameters and finds the optimal position of the individual that minimizes the adaptive value.

[0107] As a particle seeks the smallest individual optimal position, the optimal position of the entire population is also constantly updated. By comparing the minimum fitness values ​​of each individual, the optimal position of the population that minimizes the population's fitness value is found.

[0108] After updating the optimal position for each individual and the optimal position for the group, update the speed of the individuals within the speed parameter limit;

[0109] Within the range of iterations, the algorithm continuously performs the optimization process. When the maximum number of iterations is reached, the optimal position of the output population is the optimal solution of the multi-objective optimization cost function with constraints.

[0110] It should be noted that the particle swarm optimization algorithm provided by this invention belongs to the existing algorithm. The output voltage of the H-bridge inverter circuit is used as the individual particles of the population, and the multi-objective optimization cost function with constraints is used as the fitness function of the particle swarm optimization algorithm. The population size is 200, the maximum number of iterations is 30, and the algorithm execution process and update process are existing conventional technologies, which will not be described in detail here.

[0111] For example: 200 voltage points are randomly selected within a voltage range of ±1620V. These 200 voltage points are considered as individual particles in the population. Substituting the voltages of each of these 200 points into the multi-objective optimization cost function yields 200 multi-objective optimization cost function values. These 200 values ​​represent the fitness value of each individual particle. After each particle completes its multi-objective optimization cost function calculation, its voltage value is updated once according to a certain velocity and direction. The update formula uses the existing velocity and position update formulas for particle swarm optimization algorithms. After the update, a new fitness value (multi-objective optimization cost function value) is obtained. The number of updates is preset to 30. After each particle's voltage is updated 30 times, a minimum fitness value is obtained, which is the individual's optimal fitness value. The voltage corresponding to the minimum fitness value is the individual's optimal position. When 200 particles are updated 30 times collectively, the population collectively produces a minimum fitness value, which is the population's optimal fitness value. The voltage of the particle that obtains the population's optimal fitness value is the population's optimal position. This is also the optimal position that the particle swarm optimization algorithm of this invention needs to find. Finally, this optimal position is output.

[0112] S4: The single-branch output voltage is between ±1620V, and the optimal output variable u is obtained based on the particle swarm optimization algorithm. o(k) The output voltage coefficient is obtained by comparing the output voltage amplitude of a single branch; the output voltage coefficient is multiplied by the carrier amplitude to obtain the modulation value, and the modulation value is then compared with the carrier modulation to obtain the optimal duty cycle to control the cascaded H-bridge inverter circuit of the single branch, so that the EAST fast control power supply can quickly output a current value that follows the given signal.

[0113] like Figure 3 As shown, a simulation analysis of the EAST fast-controllable power supply output current control method based on particle swarm optimization (PSO) is performed. Simulation parameters are: DC supply voltage E = 540V, inductance L = 400mH, inductance internal resistance R = 0.08Ω, triangular carrier frequency = 5kHz, and a 60° carrier phase shift method is selected for phase shift control. The output current prediction equation of the EAST fast-controllable power supply is established. A cost function is established by combining multi-objective optimization and constrained control concepts. The optimal solution of the multi-objective optimization cost function with constraints is sought online using the PSO algorithm. The optimal duty cycle is output through carrier modulation to control the EAST fast-controllable power supply. Figure 3 The figures show the output current waveform and the given reference voltage waveform of the single-branch EAST fast control power supply in this invention. As can be seen from the figures, when the reference voltage waveform is a ±10V 100Hz AC square wave, the EAST fast control power supply can quickly output a ±1500A 100Hz AC square wave current. The ratio between the given reference voltage value and the output current value of the EAST fast control power supply fully meets the requirements for normal operation of the EAST fast control power supply. The simulation waveforms demonstrate that this invention enables the EAST fast control power supply to promptly and accurately follow the given reference voltage signal to output the corresponding current value.

[0114] Through the above technical solution, this invention establishes a prediction model cost function based on the sampling of the output current. The cost function consists of multi-objective optimization terms and constraints. The two objectives of the multi-objective optimization terms are to ensure that the output current quickly follows the reference signal after two cycles and to minimize the increase in output voltage. The constraint objective is to limit the single-branch output current value to within ±1500A. The multi-objective optimization terms ensure that the output current quickly follows the reference signal while having low ripple. The constraint terms ensure that the output current does not overshoot and that the output current value does not exceed the ±1500A range due to an unreasonable selection of the weighting coefficient λ. The particle swarm optimization algorithm finds the optimal solution for the established multi-objective optimization cost function with constraints. This online solution-finding process avoids the need for manual analytical solution-finding and also avoids the need for careful selection of the weighting coefficient λ. After the particle swarm optimization algorithm obtains and outputs the optimal solution of the cost function, it compares it with ±1620V to obtain the voltage coefficient. The voltage coefficient is then multiplied by the carrier amplitude and modulated with a 60° phase-shift controlled carrier to obtain the optimal duty cycle, achieving precise control of the output current.

[0115] Example 2

[0116] Based on Embodiment 1, Embodiment 2 of the present invention also provides an EAST fast control power supply output current control device, the device comprising:

[0117] The current prediction module is used to predict the current value two cycles later based on the sampled output current value of the EAST fast control power supply.

[0118] The objective function creation module is used to create a multi-objective optimization cost function with constraints.

[0119] The solver module is used to find the optimal solution of the multi-objective optimization cost function online using the particle swarm optimization algorithm;

[0120] The output current control module is used to compare the optimal solution obtained by the particle swarm optimization algorithm with the output voltage amplitude of the single-branch H-bridge inverter circuit of the EAST fast control power supply to obtain the output voltage coefficient. The output voltage coefficient is multiplied by the carrier amplitude to obtain the modulation value. The modulation value is then compared with the carrier modulation to obtain the optimal duty cycle to control the single-branch H-bridge inverter circuit of the EAST fast control power supply, so that the output current value of the EAST fast control power supply follows the given signal.

[0121] Specifically, the current prediction module is also used for:

[0122] Through formula

[0123]

[0124] Establish the output voltage equation of the H-bridge inverter circuit, where i o Where L is the output current, L is the output load inductance, and R is the internal resistance of the load inductance.

[0125] In a switching cycle T s Within time k, the output voltage equation is discretized. The output voltage equation known at the previous time is discretized and expressed as follows:

[0126]

[0127] The predicted output current at the next cycle time is expressed as:

[0128]

[0129] The predicted output current for the next two cycles is expressed as follows:

[0130]

[0131] Among them, u o (k) represents the output voltage of the H-bridge inverter circuit at time k, i o (k) represents the output current of the H-bridge inverter circuit at time k.

[0132] More specifically, the objective function establishment module is also used for:

[0133] The purpose of the entire EAST fast control power supply is to ensure that the output current value follows the given reference signal current i. ref Therefore, the target for the output current in the next two cycles is to match the reference signal current value, hence the establishment of a cost function.

[0134] J1 = [i ref -i o (k+2)] 2 (5)

[0135] To limit the increase in output voltage of the H-bridge inverter circuit, a multi-objective optimization cost function with weight λ is established.

[0136] J2 = [i ref -i o (k+2)] 2 +λ[u o (k)-u o (k-1)] 2 (6)

[0137] Limit the output current and establish a multi-objective optimization cost function with constraints.

[0138]

[0139] More specifically, the solution module is also used for:

[0140] Set the initial parameters for the particle swarm optimization algorithm, use the output voltage of the H-bridge inverter circuit as the individual particles in the population, and use the multi-objective optimization cost function with constraints as the fitness function of the particle swarm optimization algorithm. Execute the particle swarm optimization algorithm within the constraints of the particle optimization parameters. When the maximum number of iterations is reached, the optimal position of the output population is the optimal solution of the multi-objective optimization cost function with constraints.

[0141] More specifically, the initial parameters for the particle swarm optimization algorithm include: a population size of 200 and a maximum number of iterations of 30.

[0142] More specifically, the limiting range of the particle optimization parameters includes:

[0143] Set the particle optimization position parameter limit range as follows:

[0144]

[0145] The speed parameter is set to a limit range of ±1.

[0146] More specifically, when the particle swarm optimization algorithm is executed and the maximum number of iterations is reached, the optimal position of the output swarm is the optimal solution of the multi-objective optimization cost function with constraints, including:

[0147] The particle swarm optimization algorithm is executed, iteratively calculating the fitness value of each particle and finding the optimal position of the individual that minimizes the fitness value. While the particles are searching for the minimum individual optimal position, the optimal position of the entire swarm is also constantly updated. By comparing the minimum fitness values ​​of each individual, the optimal position of the swarm that minimizes the overall fitness value is found. After updating the optimal positions of each individual and the optimal position of the swarm, the velocity of the individual is updated within the velocity parameter limit. Within the number of iterations, the algorithm continuously performs the optimization process. When the maximum number of iterations is reached, the output optimal position of the swarm is the optimal solution of the multi-objective optimization cost function with constraints.

[0148] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for controlling the output current of an EAST fast-control power supply, characterized in that, The method includes: Step 1: Predict the current value two cycles later based on the EAST fast control power supply output current sampling value; Through formula （1） Establish the output voltage equation for a single-branch H-bridge inverter circuit, where, For output current, For output load inductance, The load inductance internal resistance; In a switching cycle Inside, Discretize the output voltage equation at each step. The output voltage equation known at the previous step is expressed as follows: （2） The predicted output current at the next cycle time is expressed as: （3） The predicted output current for the next two cycles is expressed as follows: （4） in, For single-branch H-bridge inverter circuit Output voltage at any given time For single-branch H-bridge inverter circuit Output current at any given moment; Step 2: Establish a multi-objective optimization cost function with constraints; The purpose of the entire EAST fast control power supply is to ensure that the output current value follows the given reference signal current. Therefore, the target for the output current in the next two cycles is to match the reference signal current value, hence the establishment of a cost function. （5） The increase in output voltage of a single-branch H-bridge inverter circuit is limited, and a weighted system is established. Multi-objective optimization cost function （6） Limit the output current and establish a multi-objective optimization cost function with constraints. （7）； Step 3: Use the particle swarm optimization algorithm to find the optimal solution of the multi-objective optimization cost function online; Step 4: Compare the optimal solution obtained by the particle swarm optimization algorithm with the output voltage amplitude of the single-branch H-bridge inverter circuit of the EAST fast control power supply to obtain the output voltage coefficient. Multiply the output voltage coefficient with the carrier amplitude to obtain the modulation value. Compare the modulation value with the carrier modulation to obtain the optimal duty cycle to control the single-branch H-bridge inverter circuit of the EAST fast control power supply, so that the output current value of the EAST fast control power supply follows the given signal.

2. The EAST fast-control power supply output current control method according to claim 1, characterized in that, Step three includes: The initial parameters of the particle swarm optimization algorithm are set, the output voltage of the single-branch H-bridge inverter circuit is used as the individual particles of the population, and the multi-objective optimization cost function with constraints is used as the fitness function of the particle swarm optimization algorithm. The particle swarm optimization algorithm is executed within the limit of the particle optimization parameters. When the maximum number of iterations is reached, the best position of the output population is the optimal solution of the multi-objective optimization cost function with constraints.

3. The EAST fast-control power supply output current control method according to claim 2, characterized in that, The initial parameters for the particle swarm optimization algorithm are set as follows: the population size is 200 and the maximum number of iterations is 30.

4. The EAST fast-control power supply output current control method according to claim 2, characterized in that, The particle optimization parameter limitation range includes: Set the particle optimization position parameter limit range as follows: （8） The speed parameter is set to a limit range of ±1.

5. The EAST fast-control power supply output current control method according to claim 2, characterized in that, The particle swarm optimization algorithm, after reaching the maximum number of iterations, outputs the optimal position of the swarm, which is the optimal solution of the multi-objective optimization cost function with constraints, including: The particle swarm optimization algorithm is executed, iteratively calculating the fitness value of each particle and finding the optimal position of the individual that minimizes the fitness value. While the particles are searching for the minimum individual optimal position, the optimal position of the entire swarm is also constantly updated. By comparing the minimum fitness values ​​of each individual, the optimal position of the swarm that minimizes the overall fitness value is found. After updating the optimal positions of each individual and the optimal position of the swarm, the velocity of the individual is updated within the velocity parameter limit. Within the number of iterations, the algorithm continuously performs the optimization process. When the maximum number of iterations is reached, the output optimal position of the swarm is the optimal solution of the multi-objective optimization cost function with constraints.

6. An EAST fast-control power supply output current control device, characterized in that, The device includes: The current prediction module is used to predict the current value two cycles later based on the sampled output current value of the EAST fast control power supply. Through formula （1） Establish the output voltage equation for a single-branch H-bridge inverter circuit, where, For output current, For output load inductance, The load inductance internal resistance; In a switching cycle Inside, Discretize the output voltage equation at each step. The output voltage equation known at the previous step is expressed as follows: （2） The predicted output current at the next cycle time is expressed as: （3） The predicted output current for the next two cycles is expressed as follows: （4） in, For single-branch H-bridge inverter circuit Output voltage at any given time For single-branch H-bridge inverter circuit Output current at any given moment; The objective function creation module is used to create a multi-objective optimization cost function with constraints. The purpose of the entire EAST fast control power supply is to ensure that the output current value follows the given reference signal current. Therefore, the target for the output current in the next two cycles is to match the reference signal current value, hence the establishment of a cost function. （5） The increase in output voltage of a single-branch H-bridge inverter circuit is limited, and a weighted system is established. Multi-objective optimization cost function （6） Limit the output current and establish a multi-objective optimization cost function with constraints. （7）； The solver module is used to find the optimal solution of the multi-objective optimization cost function online using the particle swarm optimization algorithm; The output current control module is used to compare the optimal solution obtained by the particle swarm optimization algorithm with the output voltage amplitude of the single-branch H-bridge inverter circuit of the EAST fast control power supply to obtain the output voltage coefficient. The output voltage coefficient is multiplied by the carrier amplitude to obtain the modulation value. The modulation value is then compared with the carrier modulation to obtain the optimal duty cycle to control the single-branch H-bridge inverter circuit of the EAST fast control power supply, so that the output current value of the EAST fast control power supply follows the given signal.

Images