Circulating current suppression method and system for parallel connection of two inverters in electric energy feedback device

Abstract

The invention discloses a method and system for restraining parallel circulating current of two inverters in an electric energy feedback device, and relates to the technical field of electric energy conversion in the new energy technology, and the method comprises the steps: constructing an inverter parallel circulating current mathematical model; processing the circulating current difference and the circulating current difference change rate of the parallel connection of the two inverters as the input of a neural network; and inputting the output of the neural network into a PI controller, and processing the inverter. According to the method, factors generated by the circulating current during parallel operation of the two inverters are comprehensively considered, the circulating current is suppressed by using the fuzzy neural network PI controller and the virtual impedance, and the method has remarkable advantages in improving the control performance and the feedback efficiency of power plant battery electric energy feedback.

Description

Technical Field

[0001] The present invention relates to the technical field of power conversion in new energy technologies, and particularly to a method and system for suppressing circulating current in parallel connection of two inverters in a power feedback device. Background Art

[0002] With the development of renewable energy, especially the wide application of wind energy and solar energy, the stability and reliability of the power system face new challenges. To ensure the stability of power supply, power plants adopt battery energy storage systems to prevent emergencies and improve the safety of the power system. However, for the health of the battery, charge and discharge experiments must be carried out on the battery at regular intervals. The battery energy storage system usually converts direct current into alternating current through an inverter and feeds it back to the power grid. However, when multiple inverters operate in parallel, circulating current may be generated, affecting the stability and efficiency of the system. Therefore, how to suppress the circulating current between parallel inverters has become a technical problem that urgently needs to be solved in the power system.

[0003] When the capacity of a single inverter is insufficient, we will take multiple inverters in parallel for capacity expansion. When multiple inverters are connected in parallel, due to the possible slight differences in the control methods and output voltages of each inverter, low-frequency circulating current may be generated between the two inverters. These circulating currents do not participate in actual energy transmission, but consume the power of the system and may cause heating, reduced efficiency, and even damage to equipment. When connecting to the grid, current tracking or voltage tracking methods are usually used for adjustment. Due to the parameter inconsistencies between the power grid and the inverter (such as the phase and amplitude of voltage and current, etc.), even under ideal conditions, there may be slight differences in the alternating current output by the two inverters. Factors such as the operating states, control algorithms, temperature, and aging of different inverters will all lead to inconsistencies in output current and voltage. When these inverters are connected through a common DC bus, due to the voltage difference, a closed-loop current may be formed between them, which is the so-called "circulating current".

[0004] The existence of circulating current will not only increase the power loss of the system, but may also cause the inverter to overheat, resulting in system instability or damage. Therefore, how to suppress this circulating current and ensure the coordinated operation between grid-connected inverters has become a research hotspot. For this reason, a method for suppressing circulating current in parallel connection of two inverters in a power plant battery power feedback device based on a fuzzy neural network PI controller is proposed. Summary of the Invention

[0005] In view of the above existing problems, the present invention is proposed.

[0006] Therefore, the problem to be solved by the present invention is: when multiple inverters are connected in parallel, due to possible slight differences in the control methods and output voltages of each inverter, low-frequency circulating currents may be generated between two inverters. How to suppress such circulating currents and ensure the coordinated operation between grid-connected inverters?

[0007] To solve the above technical problems, the present invention provides the following technical solution: a method for suppressing the parallel circulating current of two inverters in an electric energy feedback device, including constructing a mathematical model of the parallel circulating current of the inverters; using the circulating current difference and the change rate of the circulating current difference of the two parallel inverters as the inputs of a neural network for processing; the neural network refers to a fuzzy neural network including four parts, an input layer, a membership function generation layer, a fuzzy inference layer, and a normalization layer. The input layer adopts two-bit input, and the input quantities are the circulating current and the change rate of the circulating current e`, and the output layer also adopts a two-layer neural network structure, and the output quantity is that of the PI controller and ; input the output of the neural network into the PI controller to process the inverter.

[0008] As a preferred scheme of the method for suppressing the parallel circulating current of two inverters in the electric energy feedback device of the present invention, wherein: the mathematical model of the parallel circulating current of the inverters includes mathematical modeling of the parallel connection of the two inverters, and according to the Thevenin equivalent theorem, the two inverters are equivalent to a two-port network formed by connecting an ideal voltage source and an equivalent impedance in series and then in parallel.

[0009] As a preferred scheme of the method for suppressing the parallel circulating current of two inverters in the electric energy feedback device of the present invention, wherein: the mathematical model of the parallel circulating current of the inverters is expressed as: , Analyze to obtain the relationship between the output voltage and the equivalent output impedance and the circulating current between the two parallel inverters; when , , the expression of the circulating current is: , When , , the expression of the circulating current is: , When , , the expression of the circulating current is: , Among them, is the equivalent impedance of inverter , is the current of inverter , is the voltage and phase of inverter , is the equivalent impedance of the inverter ; is the current of the inverter ; is the voltage and phase of the inverter ; is the voltage and phase across the load is the circulating current difference

[0010] As a preferred scheme of the method for suppressing the parallel circulating current of two inverters in the power energy feedback device described in the present invention, wherein: the input layer includes two neuron nodes, and the output to the next layer of neurons through the activation function is expressed as: , wherein is the activation function is the th input quantity is expressed as the circulating current e is expressed as the circulating current change rate e'; the membership function generation layer includes respectively performing fuzzy processing on the circulating current and the circulating current change rate by using 7 membership functions, and using the Gaussian function as the membership function, which is expressed as: , wherein is the central value of the Gaussian membership function is the width of the Gaussian membership function is the th membership function is the th input variable under the th fuzzy rule

[0011] As a preferred scheme of the method for suppressing the parallel circulating current of two inverters in the power energy feedback device described in the present invention, wherein: the fuzzy inference layer includes that each node represents a fuzzy rule between the input and the output, and each corresponding fuzzy rule is applied to match the antecedent of the fuzzy rule and calculate the fitness of the fuzzy rule, and the activation function of each node is expressed as: , , wherein is the fuzzy segmentation number of the th input is the activation function value of the kth output node is the product of the fuzzy segmentation numbers of all input variables; the normalization layer includes normalizing the output of the fuzzy inference layer, and the output is expressed as: , Among them, is 's normalization function; the output layer includes clarifying the data, and the output data is , 's result after tuning, and the activation function is expressed as: , Among them, is the activation function of the output layer, is the connection weight matrix between the fuzzy inference layer and the output layer; the result output is expressed as: , Among them, is the difference of the proportionality coefficient, is the difference of the integral coefficient.

[0012] As a preferred solution of the method for suppressing the parallel circulating current of two inverters in the power energy feedback device of the present invention, among them: the processing of the inverter includes making the impedances of the two inverters consistent by adding a virtual impedance. Let the added virtual impedance be , then it is expressed as: , Among them, is the virtual impedance, is the adaptive proportionality coefficient, is the adaptive integral coefficient, is the circulating current difference, is the Laplace operator; the correction formula obtained according to the fuzzy neural network is expressed as: , Among them, is the proportionality coefficient in the PI controller after correction, is the integral coefficient in the PI controller after correction, and are the initial values obtained by the traditional PI.

[0013] Another object of the present invention is to provide a system for the method of suppressing the parallel circulating current of two inverters in the power energy feedback device, which can solve the problem of suppressing the parallel circulating current of two inverters in the power energy feedback device by constructing an inverter parallel circulating current suppression system.

[0014] To solve the above technical problems, the present invention provides the following technical solutions: A two-inverter parallel circulating current suppression system in an electric energy feedback device, including a model construction module, a neural network processing module, and an inverter processing module; the model construction module is used to construct a mathematical model of the parallel circulating current of the inverters; the neural network processing module is used to process the circulating current difference and the change rate of the circulating current difference of the two parallel inverters as the input of the neural network; the inverter processing module is used to input the output of the neural network into a PI controller to process the inverter.

[0015] A computer device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the method for suppressing the parallel circulating current of two inverters in the electric energy feedback device as described above.

[0016] A computer-readable storage medium stores a computer program thereon. When the computer program is executed by a processor, it implements the steps of the method for suppressing the parallel circulating current of two inverters in the electric energy feedback device as described above.

[0017] The beneficial effects of the present invention are as follows: The method for suppressing the parallel circulating current of two inverters in the electric energy feedback device provided by the present invention comprehensively considers the factors generating the circulating current during the parallel operation of the two inverters, and uses a fuzzy neural network PI controller and virtual impedance to achieve the suppression of the circulating current, which has significant advantages in improving the control performance and feedback efficiency of the battery electric energy feedback in the power plant.

[0018] The present invention constructs a mathematical model of the parallel circulating current of the inverters, takes the circulating current difference and the change rate of the circulating current difference as inputs, and uses the input layer, membership function generation layer, fuzzy inference layer, and normalization layer of the fuzzy neural network for processing, outputs the parameter adjustment value of the PI controller, and then adjusts the inverter control strategy in real time. Combining with the virtual impedance technology, the circulating current between the parallel inverters is effectively suppressed, thereby significantly improving the stability and reliability of the power system, reducing the power loss, enhancing the control performance and efficiency of the electric energy feedback, and at the same time ensuring the flexibility and scalability of the system, providing an efficient, stable and highly adaptable technical solution for the electric energy feedback system of renewable energy power plants. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0020] Figure 1 It is a flowchart of the method for suppressing the parallel circulating current of two inverters in the electric energy feedback device provided by the first embodiment of the present invention.

[0021] Figure 2 This is the overall block diagram of the method for suppressing the parallel circulating current of two inverters in the power energy feedback device provided by the first embodiment of the present invention.

[0022] Figure 3 This is the equivalent circuit diagram of the parallel connection of two inverters in the power plant battery power energy feedback system of the method for suppressing the parallel circulating current of two inverters in the power energy feedback device provided by the first embodiment of the present invention.

[0023] Figure 4 This is the Gaussian membership function graph of the input variables in the fuzzy neural network of the method for suppressing the parallel circulating current of two inverters in the power energy feedback device provided by the first embodiment of the present invention.

[0024] Figure 5 This is the structure diagram of the fuzzy neural network of the method for suppressing the parallel circulating current of two inverters in the power energy feedback device provided by the first embodiment of the present invention.

[0025] Figure 6 This is the structure diagram of the system for suppressing the parallel circulating current of two inverters in the power energy feedback device provided by the second embodiment of the present invention.

[0026] In the figure: 100, model construction module; 200, neural network processing module; 300, inverter processing module. Specific embodiments

[0027] In order to make the above objects, features, and advantages of the present invention more obvious and understandable, the following will describe the specific embodiments of the present invention in detail with reference to the accompanying drawings of the specification.

[0028] In the following description, many specific details are set forth in order to fully understand the present invention. However, the present invention can also be implemented in other ways different from those described herein. Those skilled in the art can make similar extensions without departing from the connotation of the present invention. Therefore, the present invention is not limited by the specific embodiments disclosed below.

[0029] Embodiment 1, referring to Figures 1 to 5 , which is the first embodiment of the present invention. This embodiment provides a method for suppressing the parallel circulating current of two inverters in a power energy feedback device, including: constructing a mathematical model of the parallel circulating current of the inverters; using the circulating current difference and the change rate of the circulating current difference of the parallel connection of the two inverters as the inputs of a neural network for processing; inputting the output of the neural network into a PI controller to process the inverters.

[0030] To overcome the deficiency that circulating current will be generated when two inverters are connected in parallel during the power feedback process of the power plant battery, the present invention aims to provide a method for suppressing the circulating current of two inverters connected in parallel in the power plant battery power feedback device based on a fuzzy neural network PI controller. This method makes the equivalent impedances of the two inverters equal by adding virtual impedance, and uses the fuzzy neural network PI controller to control the inverter adjustment parameters to reduce the circulating current and achieve the purpose of stable operation.

[0031] To achieve the above goal, the technical route adopted by the present invention is: a method for suppressing the circulating current of two inverters connected in parallel in the power plant battery power feedback device based on a fuzzy neural network PI controller, including the following steps: S1. Construct a mathematical model of the circulating current of the parallel inverters.

[0032] Perform mathematical modeling on the parallel connection of two inverters. According to Thevenin's equivalent theorem, the two inverters can be equivalent to a two-port network in which an ideal voltage source and an equivalent impedance are connected in series and then in parallel, as Figure 2 and 3 shown. represents the equivalent impedance of inverter , represents its current, represents its voltage and phase. Similarly, the equivalent impedance, current, voltage and phase of inverter can be obtained. represents the voltage and phase across the load. The mathematical expression of its circulating current is: . From the above formula, the relationship between the output voltage, equivalent output impedance and the circulating current of the parallel connection of the two inverters can be analyzed as follows: When , , the circulating current expression is: . When , , the circulating current expression is: . When , , the circulating current expression is: . Among them, is the equivalent impedance of inverter , is the current of inverter , is the voltage and phase of inverter , is the equivalent impedance of inverter , is the inverter current, is the inverter voltage and phase, is the voltage and phase across the load, is the circulating current difference.

[0033] It can be seen from the above formula that when the impedances of the inverters are the same, the reason for the generation of the circulating current is the difference in the amplitudes and phases of the output voltages of the two inverters. When the output impedances are different and the voltage phase amplitudes are different, the circulating current will also be generated.

[0034] S2. Use the circulating current difference and the change rate of the circulating current difference of the two parallel inverters as the inputs of the neural network for processing.

[0035] The fuzzy neural network includes four parts: the input layer, the membership function generation layer, the fuzzy inference layer, and the normalization layer. Its structure is as Figure 5 shown. The input layer adopts two-bit input, and the input quantities are the circulating current e and the change rate of the circulating current e`. The output layer also adopts a two-layer neural network structure, and the output quantities are Δkp and Δki of the PI controller. The processing process of each layer of the fuzzy neural network is as follows: Input layer: There are two neuron nodes in this layer, and they are output to the neurons in the next layer through the activation function, expressed as: , where, is the activation function, is the th input quantity, is expressed as the circulating current e, is expressed as the change rate of the circulating current e`.

[0036] Membership function generation layer: In this layer, the circulating current and the change rate of the circulating current are respectively fuzzified using 7 membership functions. The fuzzy sets are {NB, NM, NS, ZO, PS, PM, PB}, as Figure 4 shown. The fuzzy language domain is {-3, -2, -1, 0, 1, 2, 3}. The Gaussian function is used as the membership function, and the formula is as follows, where is the central value of the Gaussian membership function, is the width of the Gaussian membership function: , where, is the central value of the Gaussian membership function, is the width of the Gaussian membership function, is the th membership function, is the th input variable at the The fuzzy quantity under a fuzzy rule is obtained by comparing the input value with the corresponding membership function.

[0037] Fuzzy inference layer: Each node in this layer represents a fuzzy rule between the input and the output. Each corresponding fuzzy rule is applied to match the antecedent of the fuzzy rule and calculate the fitness of the fuzzy rule. The activation function of each node can be expressed as the following formula: , , where, is the number of fuzzy partitions of the th input, is the activation function value of the kth output node, which represents the fitness of the kth fuzzy rule, is the product of the number of fuzzy partitions of all input variables, N = 49.

[0038] Normalization layer: Normalize the output of the fuzzy inference layer, and the output expression can be expressed as: , where, is the normalization function; The source of 49 is that both e and e` have 7 linguistic variables {NB, NM, NS, ZO, PS, PM, PB}, so the number of rules in the fuzzy rule base covering the entire solution space consists of 49 fuzzy conditional statements, k = {1, 2, 3, 4,..., 49}.

[0039] Output layer: The output layer is also called the defuzzification layer. This layer can clarify the data, and the output data is , the tuned result of, and its activation function is: , where, is the activation function of the output layer, is the connection weight matrix between the fuzzy inference layer and the output layer.

[0040] The final result is output as follows: , where, is the difference in proportionality coefficient, is the difference in integral coefficient.

[0041] S3. Input the output of the neural network into the PI controller to process the inverter.

[0042] Use the addition of virtual impedance to make the impedances of the two inverters consistent. Let the added virtual impedance be , that is: , Among them, is the virtual impedance, is the adaptive proportionality coefficient, is the adaptive integral coefficient, is the circulating current difference, is the Laplace operator.

[0043] According to the above fuzzy neural network, we can obtain its correction formula: , Among them, is the proportionality coefficient in the PI controller after correction, is the integral coefficient in the PI controller after correction, and are the initial values obtained by the traditional PI.

[0044] Example 2, referring to Figure 6 , is the second embodiment of the present invention. Different from the previous embodiment, it provides a parallel circulating current suppression system for two inverters in the power feedback device, including: a model construction module 100, a neural network processing module 200, and an inverter processing module 300.

[0045] The model construction module 100 is used to construct a mathematical model of the parallel circulating current of the inverters.

[0046] The neural network processing module 200 is used to process the circulating current difference and the change rate of the circulating current difference of the two parallel inverters as the input of the neural network.

[0047] The inverter processing module 300 is used to input the output of the neural network into the PI controller to process the inverter.

[0048] If the above functions are implemented in the form of software function units and sold or used as independent products, they 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 a part of this technical solution, can be embodied in the form of a software product. This 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 the various embodiments of the present invention. And the aforementioned storage medium includes: USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical disks, etc., which can store program codes.

[0049] The logic and / or steps represented in the flowchart or otherwise described herein can, for example, be considered a defined sequence of executable instructions for implementing logical functions, which can be embodied in any computer-readable medium for use by or in connection with an instruction execution system, apparatus, or device, such as a computer-based system, a system including a processor, or other systems that can fetch and execute instructions from the instruction execution system, apparatus, or device. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.

[0050] More specific examples (a non-exhaustive list) of the computer-readable medium include the following: an electrical connection (electronic device) having one or more wirings, a portable computer diskette (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disc read-only memory (CDROM). Additionally, the computer-readable medium can even be paper or other suitable media on which the program can be printed, as the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing as appropriate, and then storing it in a computer memory.

[0051] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above-described embodiments, the multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, any one of the following techniques known in the art or a combination thereof can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGAs), field programmable gate arrays (FPGAs), etc.

[0052] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A method for suppressing circulating current of two inverters in parallel in an electric energy feedback device, characterized in that: include, Construct the inverter parallel circulation mathematical model; The circulating current difference and the changing rate of the circulating current difference of the two inverters in parallel are processed as the input of the neural network; The neural network refers to a fuzzy neural network which includes four parts: input layer, membership function generation layer, fuzzy reasoning layer and normalization layer. The input layer takes binary input, and the input quantity is the circulation. The output layer also adopts a two-layer neural network structure, and the output is the PI controller and ; The output of the neural network is input into the PI controller to process the inverter.

2. The method for suppressing circulating current of two inverters in parallel in an electric energy feedback device according to claim 1, characterized in that: The inverter parallel circulation mathematical model includes mathematical modeling of two inverters in parallel, and according to the Thevenin equivalent theorem, the two inverters are equivalent to a two-port network in which an ideal voltage source and an equivalent impedance are connected in series and then in parallel.

3. The method for suppressing circulating current of two inverters in parallel in an electric energy feedback device according to claim 2, characterized in that: The inverter parallel circulation mathematical model is expressed as: ， The relationship between the output voltage and equivalent output impedance and the circulating current between two inverters in parallel is analyzed; when , When , the circulation expression is: ， when , When , the circulation expression is: ， when , When , the circulation expression is: ， in, For inverter The equivalent impedance of For inverter The current, For inverter The voltage and phase, For inverter The equivalent impedance of For inverter The current, For inverter The voltage and phase, is the voltage and phase across the load, is the circulation difference.

4. The method for suppressing circulating current of two inverters in parallel in an electric energy feedback device according to claim 3, characterized in that: The input layer includes two neuron nodes, which are output to the neurons of the next layer through the activation function, expressed as: ， in, is the activation function, For the Input quantity, Expressed as circulation e, Expressed as circulation change rate e`; The membership function generation layer includes fuzzifying the circulation and the circulation change rate using 7 membership functions respectively, and using Gaussian function as the membership function, which is expressed as: ， in, is the central value of the Gaussian membership function, is the width of the Gaussian membership function, For the membership function, For the The input variable is The fuzzy quantity under a fuzzy rule.

5. The method for suppressing circulating current of two inverters in parallel in an electric energy feedback device according to claim 4, characterized in that: The fuzzy reasoning layer includes each node representing a fuzzy rule between input and output. Each corresponding fuzzy rule is applied to match the antecedent of the fuzzy rule and calculate the fitness of the fuzzy rule. The activation function of each node is expressed as: ， ， in, For the The number of fuzzy segmentations of the input, is the activation function value of the kth output node, is the product of the fuzzy segmentation numbers of all input variables; The normalization layer includes normalizing the output of the fuzzy inference layer, and the output is expressed as: ， in, for Normalization function of ; The output layer includes clarifying the data, and the output data is , After the adjustment, the activation function is expressed as, ， in, is the activation function of the output layer, is the connection weight matrix between the fuzzy reasoning layer and the output layer; The resulting output is expressed as, ， in, is the proportional coefficient difference, is the integral coefficient difference.

6. The method for suppressing circulating current of two inverters in parallel in an electric energy feedback device according to claim 5, characterized in that: The processing of the inverter includes adding a virtual impedance to make the impedances of the two inverters consistent. The added virtual impedance is assumed to be , then it is expressed as, ， in, is the virtual impedance, is the adaptive scale factor, is the adaptive integral coefficient, is the circulation difference, is the Laplace operator; According to the fuzzy neural network, the correction formula is expressed as: ， in, is the proportional coefficient in the PI controller after correction, To correct the integral coefficient in the PI controller, and is the initial value obtained by traditional PI.

7. A system using the method for suppressing circulating current of two inverters in parallel in an electric energy feedback device according to any one of claims 1 to 6, characterized in that: It includes a model building module (100), a neural network processing module (200) and an inverter processing module (300); The model building module (100) is used to build a parallel circulating current mathematical model of the inverter; The neural network processing module (200) is used to process the circulating current difference and the circulating current difference change rate of two inverters connected in parallel as inputs of the neural network; The inverter processing module (300) is used to input the output of the neural network into a PI controller to process the inverter.

8. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method for suppressing circulating current of two inverters in parallel in the electric energy feedback device according to any one of claims 1 to 6 are implemented.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method for suppressing circulating current of two inverters in parallel in the electric energy feedback device according to any one of claims 1 to 6 are implemented.

Images