Boost Current Loop Control Method, System, Medium and Processor for Sodium-Ion Battery Energy Storage System

Abstract

The present invention relates to the technical field of DC conversion control, and discloses a Boost current loop control method, system, medium and processor for a sodium-ion battery energy storage system. It is applied to a Boost circuit. Based on the Boost circuit, the transfer function between the inductor current and the duty cycle is obtained. According to the transfer function, a double-loop control structure of a current inner loop and a voltage outer loop is established, and a PI controller based on a BP neural network is constructed according to the double-loop control structure. The control method includes obtaining the initial parameters set by the current inner loop PI controller; according to the initial parameters, collecting the circuit parameters of the Boost circuit, calculating the output value of the voltage outer loop PI controller, and processing it through a BP neural network to update the parameters of the current inner loop PI controller; according to the updated parameters, optimizing the dynamic characteristics of the current inner loop PI controller. The present invention updates the PI controller parameters in real time based on a BP neural network to optimize the dynamic characteristics of the controller.

Description

Technical Field

[0001] The present invention relates to the technical field of DC conversion control, and particularly relates to a Boost current loop control method, system, medium and processor for a sodium-ion battery energy storage system. Background Art

[0002] The DC boost (Boost) circuit is often used as the front-stage circuit of the DC / AC conversion circuit and is widely used in new energy power generation systems.

[0003] With the development of new energy storage and material technologies, new energy storage sodium-ion energy storage batteries have also been developed and produced. The sodium-ion energy storage battery also uses the DC boost (Boost) circuit as the front-stage circuit of the DC / AC conversion circuit. However, the traditional Boost circuit uses voltage and current double-loop PI control, and the control parameters are all fixed values. It can basically meet the requirements when the input DC voltage is relatively stable. However, due to the wide voltage input characteristics that the sodium-ion energy storage battery may exhibit, it may lead to problems such as unstable output and poor dynamic performance of the Boost circuit. Summary of the Invention

[0004] Aiming at the problem that the Boost circuit in the prior art has a wide voltage input characteristic, resulting in unstable output and poor dynamic performance of the Boost circuit, the present invention provides a Boost current loop control method, system, medium and processor for a sodium-ion battery energy storage system, which can realize real-time updating of the parameters of the PI controller based on the BP neural network, optimize the dynamic characteristics of the controller, and solve the problems of unstable output and poor dynamic performance of the Boost circuit in the prior art. The specific technical solutions are as follows:

[0005] The present invention provides a Boost current loop control method for a sodium-ion battery energy storage system, which is applied to the Boost circuit of the sodium-ion battery energy storage system. Based on the Boost circuit, the transfer function between the inductor current and the duty cycle is obtained, a double-loop control structure of the current inner loop and the voltage outer loop is established according to the transfer function, and a PI controller based on the BP neural network is constructed according to the double-loop control structure; based on the Boost circuit and the PI controller of the BP neural network, the control method includes:

[0006] Obtain the initial parameters set by the PI controller of the current inner loop;

[0007] According to the initial parameters, collect the circuit parameters of the Boost circuit, calculate the output value of the PI controller of the voltage outer loop, and update the parameters of the PI controller of the current inner loop after being processed by the BP neural network;

[0008] According to the updated parameters, optimize the dynamic characteristics of the PI controller of the current inner loop.

[0009] Preferably, the initial parameters of the PI controller include the proportional gain and the integral gain .

[0010] Preferably, according to the initial parameters, the circuit parameters of the Boost circuit are collected, the output value of the voltage outer-loop PI controller is calculated, and after being processed by a BP neural network, updating the parameters of the current inner-loop PI controller includes:

[0011] Set the initial and according to the current inner-loop PI controller, and collect the inductor current and capacitor voltage of the Boost circuit;

[0012] According to the collected capacitor voltage and the preset voltage reference value, calculate the output value of the voltage outer-loop PI controller and set it as the current reference value;

[0013] Input the deviation value between the current reference value and the collected inductor current, as well as the current reference value, into the BP neural network for processing, and obtain the latest and of the current inner-loop PI controller;

[0014] According to the latest and , obtain the output value of the current inner-loop PI controller, and through PWM modulation, obtain the latest inductor current and capacitor voltage of the Boost circuit, and continuously update the and of the current inner-loop PI controller until the number of loop iterations of the BP neural network reaches the upper limit or the error function value is within the set range, and then terminate the iteration.

[0015] Preferably, inputting the deviation value between the current reference value and the collected inductor current, as well as the current reference value, into the BP neural network for processing, and obtaining the latest and of the current inner-loop PI controller includes:

[0016] Input the deviation value between the current reference value and the real-time collected inductor current, as well as the current reference value, into the BP neural network with the corresponding parameters and structure already set;

[0017] Through forward propagation calculation in the BP neural network, calculate the input deviation value in combination with the weight matrix and bias matrix of each layer of the neural network, and process the calculation result through a preset activation function to obtain the final output;

[0018] Obtain the latest and of the current inner-loop PI controller according to the final output.

[0019] Preferably, the forward propagation calculation in the BP neural network combines the input deviation value with the weight matrix and bias matrix of each layer of the neural network for calculation, and processes the calculation result through a preset activation function. The final output includes:

[0020] In the forward propagation calculation of the BP neural network, the input deviation value is transmitted from the input layer to the hidden layer;

[0021] Multiply the input deviation value by the updated weight matrix of the hidden layer, add the preset bias matrix, and perform a non-linear transformation through the preset activation function to obtain the output of the hidden layer;

[0022] Multiply the output of the hidden layer by the updated weight matrix of the output layer, add the preset bias matrix, and then perform a non-linear transformation through the preset activation function to obtain the output of the output layer.

[0023] Preferably, the update process of the updated weight matrix includes:

[0024] Establish an error performance function based on the current reference value and the deviation value of the inductor current collected in real time;

[0025] Use the steepest descent method, the set learning rate, and the inertia coefficient to update the weight matrices of the hidden layer and the output layer along the negative gradient direction of the error performance function surface.

[0026] The embodiment of the present invention also provides a Boost current loop control system for a sodium-ion battery energy storage system, which is applied to the aforementioned Boost current loop control method for a sodium-ion battery energy storage system, and includes:

[0027] A Boost circuit, including a low-voltage DC power supply, a power switch device, a freewheeling diode, a filter capacitor, and a load resistor, is used for voltage boost conversion and controls the output voltage of the circuit by adjusting the duty cycle of the power switch device;

[0028] A PI control module based on a BP neural network is used to obtain the initial parameters set by the current inner loop PI controller; according to the initial parameters, collect the circuit parameters of the Boost circuit, calculate the output value of the voltage outer loop PI controller, and update the parameters of the current inner loop PI controller after being processed by the BP neural network; optimize the dynamic characteristics of the current inner loop PI controller according to the updated parameters.

[0029] Preferably, the PI control module based on the BP neural network includes:

[0030] A collection module for initializing according to the current inner loop PI controller and , collect the inductor current and capacitor voltage of the Boost circuit;

[0031] A voltage outer-loop PI controller, which is used to calculate the output value of the voltage outer-loop PI controller according to the collected capacitor voltage and the preset voltage reference value, and set it as the current reference value; input the deviation value between the current reference value and the collected inductor current, and the current reference value into the BP neural network;

[0032] A BP neural network module, which is used to process the deviation value between the input current reference value and the collected inductor current, and the current reference value to obtain the latest and , and continuously update the and of the current inner-loop PI controller until the number of loop iterations of the BP neural network reaches the upper limit or the error function value is within the set range, and then terminate the iteration;

[0033] A current inner-loop PI controller, which is used to obtain the output value of the current inner-loop PI controller according to the latest and .

[0034] The present invention also provides a computer-readable storage medium, which includes a stored program. When the program runs, it controls the device where the computer-readable storage medium is located to execute the foregoing Boost current loop control method for a sodium-ion battery energy storage system.

[0035] The present invention also provides a processor, which is used to run a program. When the program runs, it executes the foregoing Boost current loop control method for a sodium-ion battery energy storage system.

[0036] Compared with the prior art, the beneficial effects of the present invention are:

[0037] The Boost current loop control method for a sodium-ion battery energy storage system of the present invention realizes flexible regulation of the input voltage and output voltage by constructing a Boost circuit and using PWM to control power switching devices. A state-average model is established for small-signal analysis, and a double-loop control structure of a current inner loop and a voltage outer loop is established, improving the dynamic response speed and stability of the system. By collecting the circuit parameters of the Boost circuit, calculating the output value of the voltage outer loop PI controller, and processing it through a BP neural network, the parameters of the current inner loop PI controller are updated; according to the updated parameters, the dynamic characteristics of the current inner loop PI controller are optimized. A PI controller with a BP neural network is constructed and the parameters are updated in real time to optimize the dynamic characteristics of the controller, solving the problems of unstable output and poor dynamic performance of the Boost circuit in the prior art. At the same time, through the real-time learning and parameter optimization of the BP neural network, combined with PWM modulation, precise current loop and voltage loop control are achieved, reducing control errors. Description of the Drawings

[0038] In order to more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the following will briefly introduce the drawings required for use in the description of the specific embodiments or the prior art. In all the drawings, similar elements or parts are generally denoted by similar reference numerals. In the drawings, the elements or parts are not necessarily drawn to scale.

[0039] Figure 1 It is a flowchart of the Boost current loop control method for a sodium-ion battery energy storage system according to an embodiment of the present invention.

[0040] Figure 2 It is a Boost circuit diagram of the Boost current loop control method for a sodium-ion battery energy storage system according to an embodiment of the present invention.

[0041] Figure 3 It is a Boost circuit topology diagram of the Boost current loop control method for a sodium-ion battery energy storage system according to an embodiment of the present invention.

[0042] Figure 4 It is a schematic diagram of two working modes of the Boost circuit of the Boost current loop control method for a sodium-ion battery energy storage system according to an embodiment of the present invention.

[0043] Figure 5 It is a Boost double-loop control structure diagram of the Boost current loop control method for a sodium-ion battery energy storage system according to an embodiment of the present invention.

[0044] Figure 6 It is a schematic diagram of forward propagation of the Boost current loop control method for a sodium-ion battery energy storage system according to an embodiment of the present invention.

[0045] Figure 7 The control flowchart of the PI controller for the Boost circuit current loop based on the BP neural network for the Boost current loop control method of the sodium-ion battery energy storage system in the embodiment of the present invention.

[0046] Figure 8 The Bode diagram of the current loop for the Boost current loop control method of the sodium-ion battery energy storage system provided in the embodiment of the present invention.

[0047] Figure 9 The Bode diagram of the voltage loop for the Boost current loop control method of the sodium-ion battery energy storage system provided in the embodiment of the present invention.

[0048] Figure 10 The main circuit diagram a for the Boost current loop control method of the sodium-ion battery energy storage system provided in the embodiment of the present invention.

[0049] Figure 11 The control circuit diagram b for the Boost current loop control method of the sodium-ion battery energy storage system provided in the embodiment of the present invention.

[0050] Figure 12 The PI controller and the control effect diagram of the present invention when the input voltage fluctuates by ±600V for the Boost current loop control method of the sodium-ion battery energy storage system provided in the embodiment of the present invention.

[0051] Figure 13 The schematic diagram of the Boost current loop control system for the sodium-ion battery energy storage system provided in the embodiment of the present invention.

[0052] Figure 14 The schematic diagram of the PI control module of the BP neural network for the Boost current loop control system of the sodium-ion battery energy storage system provided in the embodiment of the present invention. Detailed implementation manners

[0053] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all of the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0054] It should be understood that when used in this specification, the terms "include" and "comprise" indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their combinations.

[0055] It should also be understood that the terms used in the specification of the present invention are only for the purpose of describing specific embodiments and are not intended to limit the present invention. As used in the specification of the present invention, unless the context clearly indicates otherwise, the singular forms "a", "an" and "the" are intended to include the plural forms.

[0056] It should be further understood that the term "and / or" used in the specification of the present invention refers to any combination and all possible combinations of one or more of the associated listed items, and includes these combinations.

[0057] Please refer to the following embodiments Figures 1 to 14 。

[0058] The present invention provides a Boost current loop control method for a sodium-ion battery energy storage system, which is applied to the Boost circuit in the sodium-ion battery energy storage system. Based on the Boost circuit, a transfer function between the inductor current and the duty cycle is obtained, and a double-loop control structure of a current inner loop and a voltage outer loop is established according to the transfer function. A PI controller based on a BP neural network is constructed according to the double-loop control structure; based on the Boost circuit and the PI controller of the BP neural network, the control method includes:

[0059] Step S1, obtaining the initial parameters set by the current inner loop PI controller; the initial parameters of the PI controller include the proportional gain and the integral gain ;

[0060] Step S2, according to the initial parameters, collecting the circuit parameters of the Boost circuit, calculating the output value of the voltage outer loop PI controller, and updating the parameters of the current inner loop PI controller through BP neural network processing; specifically including:

[0061] According to the initially set and of the current inner loop PI controller, collect the inductor current and capacitor voltage of the Boost circuit;

[0062] According to the collected capacitor voltage and the preset voltage reference value, calculate the output value of the voltage outer loop PI controller and set it as the current reference value;

[0063] Input the deviation value between the current reference value and the collected inductor current, and the current reference value into the BP neural network for processing to obtain the latest and of the current inner loop PI controller;

[0064] According to the latest and , obtain the output value of the current inner-loop PI controller, and through PWM modulation, obtain the latest inductor current and capacitor voltage of the Boost circuit, and continuously update the and , until the number of loop iterations of the BP neural network reaches the upper limit or the error function value is within the set range, then terminate the iteration.

[0065] Step S3, optimize the dynamic characteristics of the current inner-loop PI controller according to the updated parameters.

[0066] Optimize the dynamic characteristics of the controller by based on the PI parameters updated in real time, and set the voltage-loop control parameters according to the circuit parameters through the current-loop control of the Boost circuit.

[0067] In this embodiment, the BP neural network can continuously update the proportional gain and integral gain of the current inner-loop PI controller according to the real-time circuit parameters such as the collected inductor current and capacitor voltage. In the face of wide voltage input or load changes, the controller can quickly adjust the parameters to adapt to the dynamic changes of the system, improving the adaptability of the system to different working conditions. And by continuously iterating and updating the PI parameters, set the output of the voltage outer-loop PI controller as the current reference value, and process the deviation between the current reference value and the inductor current, which can accurately control the output of the Boost circuit. Continuous optimization makes the output inductor current and capacitor voltage closer to the expected values, reducing the steady-state error and achieving precise control of the output voltage and current.

[0068] Specifically, please refer to Figure 7 , in an embodiment of the present application, input the deviation value between the current reference value and the collected inductor current, as well as the current reference value, into the BP neural network for processing to obtain the latest and including:

[0069] Input the deviation value between the current reference value and the inductor current collected in real time, as well as the current reference value, into the BP neural network with preset corresponding parameters and structure;

[0070] (2) Through forward propagation calculation in the BP neural network, calculate the input deviation value in combination with the weight matrix and bias matrix of each layer of the neural network, and process the calculation result through a preset activation function to obtain the final output; specifically including:

[0071] Through forward propagation calculation in the BP neural network, transfer the input deviation value from the input layer to the hidden layer;

[0072] Multiply the input deviation value by the updated weight matrix of the hidden layer, add the preset bias matrix, and perform a non-linear transformation through the preset activation function to obtain the output of the hidden layer;

[0073] Multiply the output of the hidden layer by the updated weight matrix of the output layer, add the preset bias matrix, and then perform a non-linear transformation through the preset activation function to obtain the output of the output layer;

[0074] (3) Obtain the latest and .

[0075] By inputting the current reference value and its deviation value from the inductor current into the BP neural network together, the network can comprehensively consider the system set target and actual current state information. Through forward propagation calculations within the network, including operations with the weight matrix, bias matrix, and non-linear transformation of the activation function, relevant information can be effectively extracted and processed, and the latest proportional gain and integral gain parameters of the PI controller adapted to the current system state can be accurately output.

[0076] In this embodiment, please refer to Figure 7 , set the parameters of the voltage-loop PI controller according to the circuit parameters, perform real-time sampling on the inductor current and capacitor voltage , calculate the output value of the voltage outer-loop PI controller, calculate the current loop and , output the duty cycle, perform PWM modulation, output the capacitor voltage , determine whether k reaches the upper limit (k is a measurement value representing the number of iteration termination times). When k does not reach the upper limit, k = k + 1, update the parameters of the voltage-loop PI controller in real time and reset them, repeat the process, and when k reaches the upper limit, end the control process.

[0077] It should be noted that for the BP neural network, the input column vector is X of dimension m×1, the weight matrix of the kth hidden layer (in this embodiment, k takes 1, using one hidden layer), that is, the coefficient matrix is of dimension n×n , the bias matrix of dimension n×1 , the activation function is f, and the output layer is the column vector O of dimension 1×n. Then the output of the first hidden layer in forward propagation is:

[0078]

[0079]

[0080] The weight matrix corresponding to the output layer O is , the bias matrix , then there is:

[0081]

[0082] Among them, the first-layer weight matrix , the second-layer weight matrix They are respectively:

[0083]

[0084]

[0085] The first - layer bias matrix and the second - layer bias matrix They are respectively:

[0086]

[0087]

[0088] The activation function f(x) takes:

[0089]

[0090] where h is the output of the hidden layer. The schematic diagram of forward propagation is as shown in the appendix Figure 6 as follows.

[0091] According to the current - moment current reference value of the input, that is, the parameters of the current inner - loop PI controller corresponding to the output layer and :

[0092]

[0093] Specifically, the update process of the updated weight matrix includes:

[0094] Based on the current reference value and the deviation value of the inductor current collected in real - time, establish an error performance function;

[0095] Using the steepest - descent method, the set learning rate, and the inertia coefficient, update the weight matrices of the hidden layer and the output layer along the negative - gradient direction of the surface of the error performance function.

[0096] In this embodiment, set the number of hidden layers of the BP neural network, the number of elements in each hidden layer, the activation function, and the output of the discrete - system PI controller is expressed as , assume that the output reference value at the k - th moment (representing real - time acquisition) is , the actual output is , the error is = , define the error performance function :

[0097]

[0098] Use the steepest - descent method to find the optimal solution of the weight matrix, and update the weight matrix along the negative - gradient direction of the surface of the error function :

[0099]

[0100] Wherein:

[0101]

[0102] Simplified to a sign function:

[0103]

[0104] From the PI controller output expression and the corresponding PI parameters of the output layer and , the output of the PI controller based on the BP neural network is expressed as:

[0105]

[0106] Then:

[0107] ,

[0108] 、 Respectively obtained from 、 to get:

[0109]

[0110] The gradient of the error function is:

[0111]

[0112] The update of the weight matrix elements needs to be based on the set learning rate xite and inertia coefficient alpha:

[0113]

[0114] It should be noted that the data composition of the BP neural network includes the input layer, hidden layer and output layer. The execution process of this algorithm includes two parts: forward propagation and backward propagation. In forward propagation, the upper-layer data is calculated based on the weight matrix, bias matrix and activation function to obtain the lower-layer data, and the lower-layer data is used to obtain the final output in turn. In backward propagation, through the defined error function, the partial derivative of the weight matrix with respect to the error function is calculated layer by layer from the output layer data, and the weight matrix is updated using the steepest descent method. When the number of loop iterations of forward propagation and backward propagation reaches the upper limit, or the error function value is within the set range, the iteration can be terminated and the final result can be output.

[0115] The Boost current loop control method for the sodium-ion battery energy storage system of the present invention realizes flexible adjustment of the input voltage and the output voltage by constructing a Boost circuit and using PWM to control the power switch device, establishes a state-average model, conducts small-signal analysis, and establishes a double-loop control structure of an inner current loop and an outer voltage loop, improving the dynamic response speed and stability of the system. By collecting the circuit parameters of the Boost circuit, calculating the output value of the outer voltage loop PI controller, and processing it through a BP neural network, the parameters of the inner current loop PI controller are updated; according to the updated parameters, the dynamic characteristics of the inner current loop PI controller are optimized. Constructing a PI controller of a BP neural network and updating the parameters in real time to optimize the dynamic characteristics of the controller solve the problems of unstable output and poor dynamic performance of the Boost circuit in the prior art. At the same time, through the real-time learning and parameter optimization of the BP neural network, combined with PWM modulation, precise current loop and voltage loop control are realized, reducing the control error.

[0116] In one embodiment, based on the Boost circuit, the transfer function between the inductor current and the duty cycle is obtained, and a double-loop control structure of an inner current loop and an outer voltage loop is established according to the transfer function, which specifically includes:

[0117] The first step: According to the Boost circuit, use PWM to control the conduction and disconnection of the power switch device, adjust the input voltage and the output voltage, control the output voltage by adjusting the duty cycle, and establish a state-average model of the Boost circuit;

[0118] In one embodiment, the Boost circuit is composed of a low-voltage DC power supply , a power switch device T, a freewheeling diode , a filter capacitor C, and a load resistor R; as Figure 3 shown, the power switch device T (usually generated by PWM) is controlled by a square wave to realize different working modes of the circuit;

[0119] When the power switch device T is turned off, the power supply charges the filter capacitor C, and the output voltage increases;

[0120] When the power switch device T is turned on, the power supply , the inductor L, and the power switch device T form a loop, and the filter capacitor C discharges to the load resistor R, and the output voltage decreases;

[0121] The relationship between the input voltage and the output voltage and the duty cycle D of the square wave satisfies:

[0122]

[0123] By changing the duty cycle D of the square wave to control the output voltage , where L is the inductance value.

[0124] It should be noted that, as Figure 4 shown, the averaged state model of the Boost circuit is:

[0125]

[0126] where is the inductor current; is the input voltage; is the capacitor voltage, that is = ; L is the inductance value; r is the equivalent DC resistance of the inductor; Let the left side of the equal sign be 0 to obtain the steady-state solution as:

[0127]

[0128] Using the small-signal analysis method, each state variable of the averaged state model of the Boost circuit can be written as the sum of the steady-state component and the transient component:

[0129]

[0130] In the formula, is the small-signal variation of the inductor current; is the small-signal variation of the input voltage; is the small-signal variation of the capacitor voltage; is the small-signal variation of the square wave duty cycle.

[0131] In the steady-state process, it is considered that the variations of the inductor current and the capacitor voltage are 0, and there is:

[0132]

[0133] where , , , are the disturbance of the inductor current, the disturbance of the capacitor voltage, the disturbance of the input voltage, and the change of the duty cycle respectively; is the steady-state solution of the inductor current, is the steady-state solution of the capacitor voltage.

[0134] Step 2: Conduct small-signal analysis, deduce the transfer function between the inductor current and the duty cycle, and establish a double-loop control structure of the current inner loop and the voltage outer loop;

[0135] Considering that r is usually small and can be ignored in the process of designing the control law, i.e., r = 0. At the same time, when the disturbance quantity is small, the product of the two disturbance quantities can be ignored. Then, based on the state-average model of the Boost circuit, the small-signal model of the Boost circuit is as follows:

[0136]

[0137] Taking the Laplace transform of the second equation of the Boost circuit small-signal model, we get:

[0138]

[0139] Substituting the result of the Laplace transform into the first equation of the Boost circuit small-signal model, we get:

[0140]

[0141] Among them, s is the symbol of the frequency-domain variable, i.e., s = jw;

[0142] Substituting the Boost circuit state-average model into The transfer function from the inductor current to the duty cycle D is obtained as:

[0143]

[0144] Substituting the first equation of the Boost circuit small-signal model into the second equation to eliminate We get:

[0145]

[0146] Substituting the steady-state solution to eliminate In the equation, the transfer function from the capacitor voltage to the inductor current is obtained as:

[0147]

[0148] Taking the inductor current as the input of the current inner loop and the output of the voltage outer loop, the duty cycle as the output of the current inner loop, and the capacitor voltage as the input of the voltage outer loop, and using a PI controller as the controller for the current inner loop and the voltage outer loop, the double-loop control structure is as shown in the appendix Figure 5 as shown.

[0149] The embodiment of the present invention also provides a Boost current-loop control system for a sodium-ion battery energy storage system, which is applied to the aforementioned Boost current-loop control method for a sodium-ion battery energy storage system, and includes:

[0150] The Boost circuit, including a low-voltage DC power supply, a power switch device, a freewheeling diode, a filter capacitor and a load resistor, is used for voltage boost conversion and controls the output voltage of the circuit by adjusting the duty cycle of the power switch device;

[0151] The PI control module based on the BP neural network is used to obtain the initial parameters set by the current inner-loop PI controller; according to the initial parameters, collect the circuit parameters of the Boost circuit, calculate the output value of the voltage outer-loop PI controller, and after being processed by the BP neural network, update the parameters of the current inner-loop PI controller; according to the updated parameters, optimize the dynamic characteristics of the current inner-loop PI controller.

[0152] Specifically, the PI control module of the BP neural network includes:

[0153] The acquisition module is used to collect the inductor current and capacitor voltage of the Boost circuit according to the initial and set by the current inner-loop PI controller;

[0154] The voltage outer-loop PI controller is used to calculate the output value of the voltage outer-loop PI controller according to the collected capacitor voltage and the preset voltage reference value, and set it as the current reference value; input the deviation value between the current reference value and the collected inductor current, and the current reference value into the BP neural network;

[0155] The BP neural network module is used to process the input deviation value between the current reference value and the collected inductor current, and the current reference value to obtain the latest and of the current inner-loop PI controller, and continuously update the and of the current inner-loop PI controller until the number of loop iterations of the BP neural network reaches the upper limit or the error function value is within the set range, and then terminate the iteration;

[0156] The current inner-loop PI controller is used to obtain the output value of the current inner-loop PI controller according to the latest and ;

[0157] The technical effect of this embodiment for the Boost current-loop control system of the sodium-ion battery energy storage system is the same as that of the embodiment for the Boost current-loop control method of the sodium-ion battery energy storage system, and the functions of the corresponding modules correspond to the method, so it will not be repeated here.

[0158] The present invention also provides a computer-readable storage medium, which includes a stored program. When the program runs, it controls the device where the computer-readable storage medium is located to execute the aforementioned Boost current loop control method for a sodium-ion battery energy storage system.

[0159] The technical effect of this embodiment is the same as that of the Boost current loop control method for a sodium-ion battery energy storage system in the embodiment, and will not be repeated here.

[0160] The present invention can be used in many general or specific computer system environments or configurations. For example: personal computers, server computers, handheld devices or portable devices, tablet devices, multi-processor systems, microprocessor-based systems, set-top boxes, programmable consumer electronic devices, network PCs, minicomputers, mainframe computers, distributed computing environments including any of the above systems or devices, and so on.

[0161] The present invention also provides a processor, which is used to run a program. When the program runs, it executes the aforementioned Boost current loop control method for a sodium-ion battery energy storage system.

[0162] The technical effect of this embodiment is the same as that of the Boost current loop control method for a sodium-ion battery energy storage system, and will not be repeated here.

[0163] In this embodiment, the processor can be a central processing unit (CPU), a controller, a microcontroller, or other data processing chips.

[0164] To verify the beneficial effects of the present invention, scientific demonstration is carried out through experiments. Taking the Figure 3 shown circuit as an example, with the parameters shown in Table 1, a current loop PI controller based on a BP neural network is designed.

[0165] Table 1 Boost circuit parameters

[0166] Parameter Value Inductance L 2.5 mH Capacitance C 500 μF Load resistance R 100 Ω Input voltage Uin 800V Output voltage target value 1000V Switching frequency fs 10 kHz

[0167] First, tune the parameters of the voltage and current loop PI controllers. Set the crossover frequency f of the current loop open-loop transfer function Gio ic to be 1 / 20 of the switching frequency f s , then f ic = 500 Hz. To make the current loop have good dynamic characteristics, according to the automatic control principle, let the phase margin of Gio be 45°. The parameters k pi , k ii are the quantities to be solved, and construct the equations:

[0168]

[0169] The value of k is obtained by solving pi = 5.29678, k ii = 4587.67459; Similarly, specify the crossover frequency f of the open-loop transfer function of the voltage loop uc = 0.1fic, the parameters k pu 、k iu to be determined. Let the open-loop transfer function of the current loop be:

[0170]

[0171] The value of k is obtained by solving pu = 0.13108, k iu = 51.61707.

[0172] Corresponding to the Bode diagram of the current loop and the Bode diagram of the voltage loop, as Figure 8 and Figure 9 shown; Construct the main simulation circuit and the control circuit, as Figure 10 and Figure 11 shown.

[0173] Set the input voltage to generate a wide voltage fluctuation of ±600V. Compare the control method set by the present invention with the traditional constant PI parameter control method, and the output results are as shown in the appendix Figure 12 shown. The abscissa is the time sampling point; the ordinate is the voltage. There are two curves in the figure. The dotted line represents the Boost current loop control method (BPPI) for the sodium-ion battery energy storage system, and the solid line represents the traditional constant PI parameter control method (PI). Facing the wide voltage fluctuation, the fluctuation amplitude of the BPPI control method curve is relatively larger, with an obvious dip at the time sampling point 2 and a large rise and then a rapid decline at about the time sampling point 4.5; the PI control method curve is relatively smooth with small fluctuations. The large fluctuations of the BPPI curve indicate that it can respond to voltage fluctuations faster and adjust the output, quickly returning to a stable state close to the initial one; while the small fluctuations of the PI control method curve indicate a slower adjustment speed. Therefore, the BPPI control method of the present invention has a better control effect in dealing with wide voltage fluctuations, can more effectively regulate the output to adapt to the change of the input voltage, and has practicality.

[0174] Those of ordinary skill in the art will appreciate that the units of each example described in connection with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components of each example have been generally described in terms of function in the above description. Whether these functions are executed in hardware or software depends on the specific application and design constraints of the technical solution. Skilled artisans can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present invention.

[0175] In the embodiments provided by the present invention, it should be understood that the division of units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units can be combined into one unit, one unit can be split into multiple units, or some features can be ignored, etc.

[0176] In addition, the functional units / modules in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

[0177] If the above-mentioned integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present invention. The foregoing storage medium includes: various media such as USB flash drives, read-only memories (ROMs), random access memories (RAMs), mobile hard disks, magnetic disks, or optical discs that can store program codes.

[0178] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements on some or all of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention, and they should all be covered by the scope of the specification of the present invention.

Claims

1. A Boost current loop control method for a sodium-ion battery energy storage system, characterized in that, The Boost circuit applied to the sodium-ion battery energy storage system, based on the Boost circuit, obtains the transfer function between the inductor current and the duty cycle, establishes a double-loop control structure of the current inner loop and the voltage outer loop according to the transfer function, and constructs a PI controller based on the BP neural network according to the double-loop control structure; Based on the Boost circuit and the PI controller of the BP neural network, the control method includes: Obtain the initial parameters set by the current inner-loop PI controller, including the proportional gain and the integral gain ; According to the initial parameters, collect the circuit parameters of the Boost circuit, calculate the output value of the voltage outer loop PI controller, and update the parameters of the current inner loop PI controller after being processed by the BP neural network; specifically including: Set the initial and according to the current inner-loop PI controller, and collect the inductor current and capacitor voltage of the Boost circuit; According to the collected capacitor voltage and the preset voltage reference value, calculate the output value of the voltage outer loop PI controller and set it as the current reference value; Input the deviation value between the current reference value and the sampled inductor current, as well as the current reference value, into a BP neural network for processing to obtain the latest and ; According to the latest and , the output value of the current inner-loop PI controller is obtained, and through PWM modulation, the latest inductor current and capacitor voltage of the Boost circuit are obtained, and the and of the current inner-loop PI controller are continuously updated until the number of loop iterations of the BP neural network reaches the upper limit or the error function value is within the set range, and then the iteration is terminated; According to the updated and parameters, optimize the dynamic characteristics of the current inner-loop PI controller.

2. The Boost current loop control method for a sodium-ion battery energy storage system according to claim 1, wherein The deviation value between the current reference value and the inductor current collected in real time, as well as the current reference value, are input into a BP neural network for processing to obtain the latest and including: Input the deviation value between the current reference value and the real-time collected inductor current, and the current reference value into the BP neural network with the corresponding parameters and structure set; Through forward propagation calculation in the BP neural network, calculate the input deviation value combined with the weight matrix and bias matrix of each layer of the neural network, and process the calculation result through the preset activation function to obtain the final output; Obtain the latest and according to the final output for the current inner-loop PI controller.

3. The Boost current loop control method for a sodium-ion battery energy storage system according to claim 2, wherein The above-mentioned forward propagation calculation in the BP neural network, calculating the input deviation value combined with the weight matrix and bias matrix of each layer of the neural network, and processing the calculation result through the preset activation function to obtain the final output includes: Through forward propagation calculation in the BP neural network, transfer the input deviation value from the input layer to the hidden layer; Multiply the input deviation value by the updated weight matrix of the hidden layer, add the preset bias matrix, and perform nonlinear transformation through the preset activation function to obtain the output of the hidden layer; Multiply the output of the hidden layer by the updated weight matrix of the output layer, add the preset bias matrix, and then perform nonlinear transformation through the preset activation function to obtain the output of the output layer.

4. The Boost current loop control method for a sodium-ion battery energy storage system according to claim 3, characterized in that, The update process of the updated weight matrix includes: Establish an error performance function according to the deviation value between the current reference value and the real-time collected inductor current; Use the steepest descent method, the set learning rate and inertia coefficient to update the weight matrices of the hidden layer and the output layer along the negative gradient direction of the error performance function surface.

5. A Boost current loop control system for a sodium-ion battery energy storage system, characterized in that, The Boost current loop control method applied to any one of claims 1 to 4 for the sodium-ion battery energy storage system includes: The Boost circuit includes a low-voltage DC power supply, a power switch device, a freewheeling diode, a filter capacitor and a load resistor, and is used for voltage step-up conversion, and controls the output voltage of the circuit by adjusting the duty cycle of the power switch device; The PI control module based on the BP neural network is used to obtain the initial parameters set by the current inner loop PI controller; according to the initial parameters, collect the circuit parameters of the Boost circuit, calculate the output value of the voltage outer loop PI controller, and update the parameters of the current inner loop PI controller after being processed by the BP neural network; according to the updated parameters, optimize the dynamic characteristics of the current inner loop PI controller; The PI control module based on the BP neural network includes: The acquisition module is used to set the initial and according to the current inner-loop PI controller, and acquire the inductor current and capacitor voltage of the Boost circuit; A voltage outer-loop PI controller is used to calculate the output value of the voltage outer-loop PI controller based on the collected capacitor voltage and the preset voltage reference value, and set it as the current reference value; input the deviation value between the current reference value and the collected inductor current, and the current reference value into a BP neural network. BP neural network module, which is used to process the deviation value between the input current reference value and the collected inductor current, as well as the current reference value, to obtain the latest and , and continuously update the and of the inner current loop PI controller until the number of loop iterations of the BP neural network reaches the upper limit or the error function value is within the set range, and then terminate the iteration; The inner current loop PI controller is used to obtain the output value of the inner current loop PI controller according to the latest and .

6. A computer-readable storage medium, characterized in that, The computer-readable storage medium includes a stored program, wherein when the program runs, it controls the device where the computer-readable storage medium is located to execute the Boost current loop control method for a sodium-ion battery energy storage system according to any one of claims 1 to 4.

7. A processor, characterized in that, The processor is used to run a program, wherein when the program runs, it executes the Boost current loop control method for a sodium-ion battery energy storage system according to any one of claims 1 to 4.

Images