Leveling method and system for energy storage converter of multi-level topological structure

Abstract

The invention discloses an energy storage converter leveling method and system of a multi-level topological structure, and the method comprises the steps: obtaining the working circuit information of an energy storage converter to be leveled, and generating a corresponding multi-level topological structure; collecting power supply characteristic information of a power supply system connected with the multi-level topological structure, and analyzing and obtaining theoretical power information of the multi-level topological structure based on the power supply characteristic information; judging a level unbalance phenomenon in the multi-level topological structure based on the theoretical power information, processing the level unbalance phenomenon based on a multi-level power fuzzy balance strategy, and generating a balance leveling instruction; and leveling the multi-level topological structure based on the balance leveling instruction, and evaluating the leveling effect. According to the invention, the problem of abnormal operation power caused by multi-level imbalance in the working process of the energy storage converter is effectively solved, the stability and safety of the energy storage converter are improved, and the working efficiency of the energy storage converter is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of energy storage converters, and particularly to a leveling method and system for an energy storage converter with a multi-level topology structure. Background Art

[0002] With the progress of social science and technology, the new energy industry is also booming. Especially in the fields of wind power generation and solar power generation, as the power generation gradually increases, the importance of energy storage technology becomes increasingly prominent. Among them, the energy storage converter (PCS, Power Conversion System), as a key link connecting the energy storage system and the power generation equipment, its performance directly affects the efficiency and safety of the entire power generation system and the energy storage system. The energy storage converter generally adopts a multi-level inverter circuit structure. However, due to various factors such as device characteristic differences, uneven capacitor charging and discharging, and improper control strategies, the multi-level inverter circuit of the energy storage converter often has problems of voltage and current imbalance and level imbalance during operation, and then abnormal operation power problems occur, which not only affect the stability and safety of the energy storage system, but also affect the working efficiency of the energy storage system. Summary of the Invention

[0003] The purpose of the present invention is to overcome the deficiencies of the prior art. The present invention provides a leveling method and system for an energy storage converter with a multi-level topology structure, which effectively solves the problem of abnormal operation power caused by multi-level imbalance in the working process of the energy storage converter, improves the stability and safety of the energy storage converter, and improves the working efficiency of the energy storage converter.

[0004] The present invention provides a leveling method for an energy storage converter with a multi-level topology structure, and the method includes:

[0005] Obtain the working circuit information of the energy storage converter to be leveled, and generate a corresponding multi-level topology structure;

[0006] Collect the power characteristic information of the power supply system connected to the multi-level topology structure, and analyze and obtain the theoretical power information of the multi-level topology structure based on the power characteristic information;

[0007] Judge the level imbalance phenomenon in the multi-level topology structure based on the theoretical power information, and process the level imbalance phenomenon based on the multi-level power fuzzy equalization strategy to generate an equalization leveling instruction;

[0008] Level the multi-level topology structure based on the equalization leveling instruction, and evaluate the leveling effect.

[0009] Further, the obtaining the working circuit information of the energy storage converter to be leveled, and generating a corresponding multi-level topology structure includes:

[0010] Extract the switching capacitor converter information from the working circuit information of the energy storage converter to be leveled;

[0011] Construct a multi-order grid model based on the switching capacitor converter information, and match the switching capacitor converter information with the nodes of the multi-order grid model to generate an adjacency matrix grid model of the switching capacitor converter;

[0012] Calculate the topological conversion efficiency of the adjacency matrix grid model of the switching capacitor converter, and judge the accuracy of the adjacency matrix grid model of the switching capacitor converter according to the topological conversion efficiency;

[0013] Fill the remaining circuit element information in the working circuit information of the energy storage converter to be leveled into the adjacency matrix grid model of the switching capacitor converter to generate a multi-level topological structure corresponding to the energy storage converter to be leveled.

[0014] Further, the calculating the topological conversion efficiency of the adjacency matrix grid model of the switching capacitor converter includes:

[0015] Calculate the conduction angle of the switching capacitor converter on each node in the adjacency matrix grid model of the switching capacitor converter, and generate a switching control signal based on the conduction angle;

[0016] Calculate the switching loss, conduction loss and drive loss of each switching capacitor converter, and perform normalization processing on the switching loss, conduction loss and drive loss;

[0017] Construct a convolutional neural network model, introduce an error function into the convolutional neural network model, and perform training;

[0018] Based on the trained convolutional neural network model, comprehensively calculate the switching control signal and the normalized switching loss, conduction loss and drive loss to obtain the topological conversion efficiency of the adjacency matrix grid model of the switching capacitor converter.

[0019] Further, the collecting the power supply characteristic information of the power supply system connected to the multi-level topological structure and analyzing and obtaining the theoretical power information of the multi-level topological structure based on the power supply characteristic information includes:

[0020] Obtain the charging rate of the power supply system connected to the multi-level topological structure, and analyze and obtain the theoretical power information of the multi-level topological structure according to the charging rate.

[0021] Further, the obtaining the charging rate of the power supply system connected to the multi-level topological structure includes:

[0022] Generate an equivalent circuit model of the power supply system, and extract the voltage, resistance, and capacitance information of the equivalent circuit model in the initial state;

[0023] Simulate the operation of the equivalent circuit model, and extract the voltage, resistance, and capacitance information of the equivalent circuit model in the operating state;

[0024] Compare the voltage, resistance, and capacitance information of the equivalent circuit model in the initial state and the operating state to generate the dynamic error parameters of the equivalent circuit model;

[0025] Based on the dynamic error parameters, use the Kalman filter model to correct the equivalent circuit model;

[0026] Extract the voltage, resistance, and capacitance information of the corrected equivalent circuit model, and generate the charging rate information of the corrected equivalent circuit model.

[0027] Further, the obtaining the theoretical power information of the multilevel topology structure according to the charging rate analysis includes:

[0028] According to the number of hybrid cascade levels of the multilevel topology structure, combine the charging rate analysis to obtain the theoretical power information of the multilevel topology structure.

[0029] Further, the determining the level imbalance phenomenon in the multilevel topology structure based on the theoretical power information includes:

[0030] Simulate the operation of the multilevel topology structure, obtain the actual power information of the multilevel topology structure in the simulated operation state, and compare the actual power information with the theoretical power information to determine whether there is a level imbalance phenomenon in the multilevel topology structure based on the comparison result.

[0031] Further, the obtaining the actual power information of the multilevel topology structure in the simulated operation state by simulating the operation of the multilevel topology structure includes:

[0032] Simulate the operation of the multilevel topology structure, analyze several switching modes of the multilevel topology structure, and respectively extract the real-time output voltage waveforms of each switching mode;

[0033] Perform carrier PWM modulation on the real-time output voltage waveforms, and perform filtering processing based on the Fourier transform function to generate the real-time drive signal waveforms of each switched capacitor converter in the multilevel topology structure;

[0034] Based on the modulation parameters used in the carrier PWM modulation, calculate the port power ratio of each switched capacitor converter in the multilevel topology structure;

[0035] Integrate the real-time drive signal waveforms and port power ratios of all switched-capacitor converters to generate the actual power information of the corresponding switched-capacitor converters, and further generate the actual power information of the multi-level topology structure.

[0036] Further, the processing of the level imbalance phenomenon based on the multi-level power fuzzy equalization strategy to generate the equalization leveling instruction includes:

[0037] Generate a double-layer switched-inductor equalization circuit topology;

[0038] According to the power characteristic information of the power supply system connected to the multi-level topology structure, define the average power fuzzy language variable, the maximum difference power fuzzy language variable, and the output power fuzzy language variable of the multi-level power fuzzy controller;

[0039] Set the membership functions of the average power fuzzy language variable, the maximum difference power fuzzy language variable, and the output power fuzzy language variable, and convert the output value of the multi-level power fuzzy controller into an equalization current instruction based on the membership functions;

[0040] Combine the equalization current instruction with the maximum difference in inductor current in the power characteristic information and input it into a PID regulator to obtain the maximum duty cycle;

[0041] According to the maximum duty cycle and in combination with the adopted multi-level power fuzzy equalization strategy, obtain the duty cycle of each switch tube in the double-layer switched-inductor equalization circuit topology;

[0042] Perform PWM modulation on the double-layer switched-inductor equalization circuit topology according to the duty cycle of each switch tube to generate an equalization drive signal, and generate an equalization leveling instruction according to the equalization drive signal.

[0043] The present invention also provides a leveling system for an energy storage converter with a multi-level topology structure. The leveling system for the energy storage converter with a multi-level topology structure is used to implement the above-mentioned leveling method for the energy storage converter with a multi-level topology structure. The system includes:

[0044] A multi-level topology structure generation module, which is used to obtain the working circuit information of the energy storage converter to be leveled and generate the corresponding multi-level topology structure;

[0045] A theoretical power information acquisition module, which is used to collect the power characteristic information of the power supply system connected to the multi-level topology structure and analyze and obtain the theoretical power information of the multi-level topology structure based on the power characteristic information;

[0046] An equalization leveling instruction generation module, which is used to judge the level imbalance phenomenon in the multilevel topology structure based on the theoretical power information, process the level imbalance phenomenon based on the multilevel power fuzzy equalization strategy, and generate an equalization leveling instruction;

[0047] A leveling evaluation module, which is used to level the multilevel topology structure based on the equalization leveling instruction and evaluate the leveling effect.

[0048] The present invention provides a leveling method and system for an energy storage converter with a multilevel topology structure. By analyzing the switch capacitor converter information in the energy storage converter to be leveled, an accurate multilevel topology structure is generated. By obtaining the charging rate of the power supply system connected to the multilevel topology structure, comparing the theoretical power information and the actual power information, it is judged whether there is a level imbalance phenomenon. An equalization leveling instruction is generated through the multilevel power fuzzy equalization strategy, effectively solving the problem of abnormal operating power caused by multilevel imbalance in the working process of the energy storage converter, improving the stability and safety of the energy storage converter, and improving the working efficiency of the energy storage converter. Description of the Drawings

[0049] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0050] Figure 1 is the flowchart of the leveling method for the energy storage converter with a multilevel topology structure in the embodiment of the present invention;

[0051] Figure 2 is the flowchart of generating the multilevel topology structure of the energy storage converter in the first embodiment of the present invention;

[0052] Figure 3 is the flowchart of calculating the topology conversion efficiency in the first embodiment of the present invention;

[0053] Figure 4 is the flowchart of obtaining the charging rate of the power supply system in the first embodiment of the present invention;

[0054] Figure 5 is the flowchart of obtaining the actual power information of the multilevel topology structure in the first embodiment of the present invention;

[0055] Figure 6 is the flowchart of generating the equalization leveling instruction in the first embodiment of the present invention;

[0056] Figure 7 It is the architecture diagram of the energy storage converter leveling system with a multi-level topology structure in the second embodiment of the present invention. Specific implementation manners

[0057] 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 only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

[0058] In the present invention, it should be understood that terms such as "including" or "having" are intended to indicate the existence of features, numbers, steps, actions, components, parts, or combinations thereof disclosed in this specification, and do not intend to exclude the possibility of the existence or addition of one or more other features, numbers, steps, actions, components, parts, or combinations thereof.

[0059] In addition, it should be noted that, without conflict, the embodiments in the present invention and the features in the embodiments can be combined with each other. The present invention will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0060] Embodiment 1

[0061] The first embodiment of the present invention provides a method for leveling an energy storage converter with a multi-level topology structure. The method includes: obtaining the working circuit information of the energy storage converter to be leveled, and generating a corresponding multi-level topology structure; collecting the power characteristic information of the power supply system connected to the multi-level topology structure, and analyzing and obtaining the theoretical power information of the multi-level topology structure based on the power characteristic information; judging the level imbalance phenomenon in the multi-level topology structure based on the theoretical power information, processing the level imbalance phenomenon based on the multi-level power fuzzy equalization strategy, and generating an equalization leveling instruction; leveling the multi-level topology structure based on the equalization leveling instruction, and evaluating the leveling effect.

[0062] In an optional implementation manner of this embodiment, as Figure 1 shown, Figure 1 shows the flowchart of the method for leveling an energy storage converter with a multi-level topology structure in the embodiments of the present invention, including the following steps:

[0063] S101. Obtain the working circuit information of the energy storage converter to be leveled, and generate a corresponding multi-level topology structure;

[0064] In an optional implementation manner of this embodiment, as Figure 2 shown, Figure 2The flowchart of generating the multi-level topology structure of the energy storage converter in Embodiment 1 of the present invention is shown, including the following steps:

[0065] S201. Extract the switching capacitor converter information in the working circuit information of the energy storage converter to be leveled;

[0066] In an optional implementation manner of this embodiment, extract the information of all the switching capacitor converters in the inverter circuit serving as the working circuit in the energy storage converter to be leveled.

[0067] Specifically, here it is considered that the energy storage converter is a medium connecting the DC micro-source and the power grid or power system, and its mainstream development trend is multi-level. In the multi-level energy storage converter, multiple switching capacitor converters are usually set, which are the main electrical components reflecting the working conditions of the energy storage converter.

[0068] S202. Construct a multi-order grid model according to the switching capacitor converter information, and match the switching capacitor converter information with the nodes of the multi-order grid model to generate an adjacency matrix grid model of the switching capacitor converter;

[0069] In an optional implementation manner of this embodiment, construct a multi-order network model according to the number of switching capacitor converters set in the inverter circuit of the energy storage converter, and match each switching capacitor converter with each node in the multi-order network model. After the matching is completed, generate an adjacency matrix network model of the switching capacitor converter of the energy storage converter.

[0070] Specifically, in this embodiment, a fourth-order network model is mainly adopted, with a total of 26 nodes, corresponding to the switching capacitor converters in the energy storage converter.

[0071] It should be noted that a bidirectional switch is default set between each adjacent switching capacitor converter.

[0072] S203. Calculate the topology conversion efficiency of the adjacency matrix grid model of the switching capacitor converter, and judge the accuracy of the adjacency matrix grid model of the switching capacitor converter according to the topology conversion efficiency;

[0073] In an optional implementation manner of this embodiment, as Figure 3 shown, Figure 3 The flowchart of calculating the topology conversion efficiency in Embodiment 1 of the present invention is shown, including the following steps:

[0074] S301. Calculate the conduction angles of the switching capacitor converters on each node in the adjacency matrix grid model of the switching capacitor converter, and generate a switching control signal based on the conduction angles;

[0075] In an alternative implementation of this embodiment, the conduction angles of the switched capacitor converters located at each node in the adjacency matrix network model of the switched capacitor converter are calculated based on the nearest level approximation method.

[0076] Specifically, the calculation formula for the conduction angle includes:

[0077]

[0078] In the formula, θ i is the conduction angle of the switched capacitor converter, n is the number of levels of the multilevel circuit structure adopted by the energy storage converter, is the level sequence.

[0079] In an alternative implementation of this embodiment, a switching control signal is generated based on the calculated conduction angle, and the expression of the modulation wave includes:

[0080] e1 = Asin(2πf)

[0081] The expression of the carrier wave includes:

[0082] e2 = Asinθ i

[0083] In the formula, e1 is the modulation wave, e2 is the carrier wave, A is the amplitude of the modulation wave, f is the modulation frequency, and θ i is the conduction angle.

[0084] S302. Calculate the switching loss, conduction loss, and driving loss of each switched capacitor converter, and perform normalization processing on the switching loss, conduction loss, and driving loss;

[0085] In an alternative implementation of this embodiment, the losses during the operation of the switched capacitor converter mainly include three types: switching loss, conduction loss, and driving loss. Among them, the switching loss is mainly the current and voltage loss at the switching moment of the switched capacitor converter, the conduction loss is mainly the loss of the equivalent series resistance (ESR) of the capacitors in the switched capacitor converter and the conduction loss of the switching transistors, and the driving loss is mainly the loss caused by the driving charge of the switching transistors.

[0086] Specifically, the calculation formula for the switching loss includes:

[0087]

[0088] In the formula, Q1 is the switching loss, S i is the i-th switching transistor, I1 is the current of the i-th switching transistor at the switching moment, V1 is the voltage of the i-th switching transistor at the switching moment, t1 is the duration of the switching process, t2 is the duration of the turn-off process, and T is the switching period.

[0089] Furthermore, the calculation formula of the conduction loss includes:

[0090]

[0091] In the formula, Q2 is the conduction loss, C i is the i-th capacitor, S i is the i-th switching transistor, i(t) is the conduction current, R1 is the equivalent series resistance of the capacitor, R2 is the on-resistance of the switching transistor, T is the switching period, and d is the duty cycle corresponding to the switching mode.

[0092] Furthermore, the calculation formula of the drive loss includes:

[0093]

[0094] In the formula, Q3 is the drive loss, S i is the i-th switching transistor, V2 is the drive voltage of the switching transistor, q is the drive charge of the switching transistor, and T is the switching period.

[0095] In an alternative implementation of this embodiment, after the calculation is completed, the switching loss, conduction loss, and drive loss of each switched-capacitor converter obtained by the calculation are normalized.

[0096] S303. Construct a convolutional neural network model, introduce an error function into the convolutional neural network model, and perform training;

[0097] In an alternative implementation of this embodiment, constructing a convolutional neural network model includes a pooling layer, a convolutional layer, an activation layer, etc., and an average absolute percentage error function is introduced into the convolutional neural network model for training.

[0098] S304. Based on the trained convolutional neural network model, perform comprehensive calculations on the switching control signal and the normalized switching loss, conduction loss, and drive loss to obtain the topological conversion efficiency of the adjacency matrix grid model of the switched-capacitor converter.

[0099] In an alternative implementation of this embodiment, based on the trained convolutional neural network model, comprehensive calculations are performed on the switching control signal (e1, e2)) and the normalized switching loss Q1, conduction loss Q2, and drive loss Q3 to obtain the topological conversion efficiency of the adjacency matrix grid model of the switched-capacitor converter.

[0100] In an alternative implementation of this embodiment, the topological conversion efficiency of the obtained adjacency matrix grid model of the switched-capacitor converter is used to judge the accuracy of the adjacency matrix grid model of the switched-capacitor converter.

[0101] Specifically, compare the topology conversion efficiency with a preset conversion efficiency threshold. When the obtained topology conversion efficiency is greater than the preset conversion efficiency threshold, it is considered that the accuracy of the adjacency matrix grid model of the switched-capacitor converter meets the standard.

[0102] S204. Fill the information of the remaining circuit elements in the working circuit information of the energy storage converter to be leveled into the adjacency matrix grid model of the switched-capacitor converter to generate the multilevel topology structure corresponding to the energy storage converter to be leveled.

[0103] In an alternative implementation of this embodiment, fill the information of the remaining circuit elements in the working circuit information of the energy storage converter to be leveled except for the switched-capacitor converter, including resistors, inductors, etc., into the adjacency matrix grid model of the switched-capacitor converter to generate the multilevel topology structure corresponding to the energy storage converter to be leveled.

[0104] Specifically, in this embodiment, the generated multilevel topology structure is an asymmetric cascaded full-bridge topology structure.

[0105] S102. Collect the power characteristic information of the power supply system connected to the multilevel topology structure, and analyze and obtain the theoretical power information of the multilevel topology structure based on the power characteristic information.

[0106] In an alternative implementation of this embodiment, obtain the charging rate of the power supply system connected to the multilevel topology structure, and analyze and obtain the theoretical power information of the multilevel topology structure according to the charging rate.

[0107] Specifically, the charging rate of the power supply system is the SOC (State of Charge, SOC), which also refers to the state of charge of the battery, specifically the ratio of the remaining power or available capacity of the battery to its capacity in the fully charged state, and can specifically reflect the health status and charging efficiency of the battery.

[0108] In an alternative implementation of this embodiment, as Figure 4 shown, Figure 4 shows the flowchart of obtaining the charging rate of the power supply system in the first embodiment of the present invention, including the following steps:

[0109] S401. Generate an equivalent circuit model of the power supply system, and extract the voltage, resistance, and capacitance information of the equivalent circuit model in the initial state.

[0110] In an alternative implementation of this embodiment, any one of the equivalent circuit models including the Rint model, Thevenin model, PNGV model, and RC model can be used as the equivalent circuit model of the power supply system.

[0111] In an alternative implementation of this embodiment, after generating the equivalent circuit model of the power supply system, obtain the voltage, resistance, and capacitance information of the equivalent circuit model in the initial state, that is, in the static state.

[0112] S402. Simulate and run the equivalent circuit model, and extract the voltage, resistance, and capacitance information of the equivalent circuit model in the operating state;

[0113] In an alternative implementation of this embodiment, simulate and run the equivalent circuit model, and extract the voltage, resistance, and capacitance information of the equivalent circuit model in the operating state, that is, in the dynamic state.

[0114] S403. Compare the voltage, resistance, and capacitance information of the equivalent circuit model in the initial state and the operating state to generate the dynamic error parameters of the equivalent circuit model;

[0115] In an alternative implementation of this embodiment, compare the voltage, resistance, and capacitance information of the equivalent circuit model in the initial state and the operating state respectively, and generate a voltage dynamic error parameter V′, a resistance dynamic error parameter R′, and a capacitance dynamic error parameter C′ respectively.

[0116] S404. Based on the dynamic error parameters, use the Kalman filter model to correct the equivalent circuit model;

[0117] In an alternative implementation of this embodiment, use the Kalman filter model to predict the voltage, resistance, and capacitance information of the corresponding operating state at the next moment based on the voltage, resistance, and capacitance information of the equivalent circuit model in the initial state. In a discrete-time system, its system state equation is:

[0118] X(k + 1) = F(k)X(k) + G(k)μ(k) + V(k)

[0119] The measurement equation is:

[0120] Z(k + 1) = H(k + 1)X(k + 1) + W(k + 1)

[0121] Then the further prediction of the system state and covariance is:

[0122]

[0123] P(k + 1|k) = F(k)P(k|k)F ′ (k) + Q(k);

[0124] Wherein, X(k + 1) is the system state at the next moment, F(k) is the first state matrix, X(k) is the system state at the current moment, G(k) is the second state matrix, μ(k) is the control quantity of the system at the next moment, V(k) is the predicted process noise, Z(k + 1) is the measured value at the next moment, H(k + 1) is the measurement matrix, and W(k + 1) is the measurement process noise. is the state prediction result at the next moment, is the optimal state prediction result at the current moment, and P(k + 1|k) is the corresponding covariance, and P(k|k) is the corresponding covariance, and F ′ (k) is the transpose matrix of F(k), and Q(k) is the covariance of the system prediction process.

[0125] In an alternative implementation of this embodiment, after predicting the voltage, resistance, and capacitance information of the equivalent circuit model in the operating state, the predicted value is compared with the true value, and after verifying the dynamic error parameters, the equivalent circuit model is corrected according to the verification result to obtain the corrected equivalent circuit model.

[0126] S405. Extract the voltage, resistance, and capacitance information of the corrected equivalent circuit model, and generate the charging rate information of the corrected equivalent circuit model.

[0127] In an alternative implementation of this embodiment, the voltage, resistance, and capacitance information of the corrected equivalent circuit model is extracted. Based on the initial SOC of the equivalent circuit model being 100%, through HPPC discharge simulation, the accuracy of the corrected equivalent circuit model is verified, and the charging rate information of the corrected equivalent circuit model is generated through the voltage, resistance, and capacitance information of the corrected equivalent circuit model, which is the charging rate of the power supply system connected to the multi-level topology structure.

[0128] In an alternative implementation of this embodiment, the obtaining of the theoretical power information of the multi-level topology structure according to the charging rate analysis includes: according to the hybrid cascade level number of the multi-level topology structure, combining the charging rate analysis to obtain the theoretical power information of the multi-level topology structure.

[0129] Specifically, in this embodiment, the hybrid cascade level number of the multi-level topology structure is the level number of the multi-level circuit structure adopted by the energy storage converter when calculating the conduction angle in S301.

[0130] S103. Judge the level imbalance phenomenon in the multi-level topology structure based on the theoretical power information, and process the level imbalance phenomenon based on the multi-level power fuzzy equalization strategy to generate an equalization leveling instruction;

[0131] In an alternative implementation of this embodiment, the multi-level topology is simulated to obtain the actual power information of the multi-level topology in the simulation running state, and the actual power information is compared with the theoretical power information, and based on the comparison result, it is judged whether there is a level imbalance phenomenon in the multi-level topology.

[0132] In an alternative implementation of this embodiment, as Figure 5 shown, Figure 5 The flowchart of obtaining the actual power information of the multi-level topology in the first embodiment of the present invention is shown, including the following steps:

[0133] S501. Simulate the multi-level topology, analyze several switching modes of the multi-level topology, and extract the real-time output voltage waveforms of each switching mode respectively;

[0134] In an alternative implementation of this embodiment, the multi-level topology is simulated, and the switching tubes and DC sources included in the switched-capacitor converter therein are analyzed for modes to obtain different switching modes.

[0135] Specifically, in the asymmetric cascaded full-bridge topology, the switching tubes on the same arm are in a complementary relationship, that is, their corresponding switching functions are in a complementary relationship. Usually, each two switching tubes include two states of ±1 and a fully truncated 0 state. Therefore, usually, the number of switching modes is the number of levels n of the multi-level circuit structure adopted by the energy storage converter plus 1, that is, the number of switching modes is n + 1.

[0136] In an alternative implementation of this embodiment, after all the switching modes are extracted, according to the actual on-off situation of the switching modes, when each switching mode is extracted, the real-time output voltage waveform corresponding to the multi-level topology is extracted.

[0137] S502. Perform carrier PWM modulation on the real-time output voltage waveform, and perform filtering processing based on the Fourier transform function to generate the real-time drive signal waveforms of each switched-capacitor converter in the multi-level topology;

[0138] In an alternative implementation of this embodiment, carrier PWM modulation is performed on the real-time output voltage waveform of each switching mode, and after filtering processing based on the double Fourier transform function analysis, the real-time drive signal waveforms of each switched-capacitor converter in the multi-level topology are generated.

[0139] S503. Calculate the port power ratio of each switched-capacitor converter in the multi-level topology based on the modulation parameters used in the carrier PWM modulation;

[0140] In an alternative implementation of this embodiment, based on the modulation parameters used in carrier PWM modulation, including frequency modulation ratio, fundamental frequency, switching frequency, etc., and combining with the power characteristic information of the power system connected to the multilevel topology structure, including the total power output thereof, calculate the port power ratio of each switched-capacitor converter in the multilevel topology structure.

[0141] Specifically, the port power ratio (PPDR) is the ratio of the DC micro-source output power of a single switched-capacitor converter in the energy storage converter to the total output power of the energy storage converter, which can reflect the efficiency of the actual output power of each switched-capacitor converter, and further determine whether its output is normal.

[0142] In an alternative implementation of this embodiment, the calculation formula of the port power ratio includes:

[0143]

[0144] In the formula, PPDR is the port power ratio, P out is the total output power of the energy storage converter, f1 is the fundamental frequency, V c is the capacitor voltage of the switched-capacitor converter, i o (t) is the original current of the switched-capacitor converter, and d(t) - d(t - 1) is the duty cycle term.

[0145] S504. Synthesize the real-time drive signal waveforms and port power ratios of all switched-capacitor converters to generate the actual power information of the corresponding switched-capacitor converters, and further generate the actual power information of the multilevel topology structure.

[0146] In an alternative implementation of this embodiment, synthesize the real-time drive signal waveforms and port power ratios of all switched-capacitor converters, and generate the actual power information of the corresponding switched-capacitor converters after superposition, and further generate the actual power information of the multilevel topology structure.

[0147] In an alternative implementation of this embodiment, as Figure 6 shown, Figure 6 shows the flowchart of generating the equalization leveling instruction in the first embodiment of the present invention, including the following steps:

[0148] S601. Generate a double-layer switched-inductor equalization circuit topology;

[0149] In an alternative implementation of this embodiment, a dual-layer switched-inductor equalization circuit topology is generated. Specifically, two batteries are used as one energy storage unit. A single battery is connected in parallel with the first-layer switched-inductor unit, and a single energy storage unit is then connected in parallel with the second-layer switched-inductor unit. That is, equalization is performed between the energy storage units through the second-layer switched-inductor unit, and equalization is performed between the batteries of the energy storage unit through the first-layer switched-inductor unit.

[0150] Here, the dual-layer switched-inductor equalization circuit topology is set. Compared with the single-layer switched-inductor equalization circuit, there are more equalization paths, faster equalization speed, and better scalability.

[0151] S602. Define the average power fuzzy language variable, the maximum difference power fuzzy language variable, and the output power fuzzy language variable of the multilevel power fuzzer according to the power characteristic information of the power system connected to the multilevel topology structure.

[0152] In an alternative implementation of this embodiment, according to the power characteristic information of the power system connected to the multilevel topology structure, that is, the state of charge (SOC) information, including the average SOC value and the SOC difference value ΔSOC, and the preset equalization current I of the dual-layer switched-inductor equalization circuit topology ave are respectively defined as the average power fuzzy language variable a1, the maximum difference power fuzzy language variable a2, and the output power fuzzy language variable a3 of the multilevel power fuzzer. b

[0153] S603. Set the membership functions of the average power fuzzy language variable, the maximum difference power fuzzy language variable, and the output power fuzzy language variable, and convert the output value of the multilevel power fuzzer into an equalization current command based on the membership functions.

[0154] In an alternative implementation of this embodiment, a triangular membership function is used to describe the membership relationship between the average power fuzzy language variable a1, the maximum difference power fuzzy language variable a2, and the output power fuzzy language variable a3, and the output value of the multilevel power fuzzer is converted into an equalization current command according to this membership relationship.

[0155] Specifically, the conversion criterion for the equalization current command is that when both the average SOC value ave and the SOC difference value ΔSOC are small, a small equalization current is used to avoid over-discharging of the power system; when both the average SOC value ave and the SOC difference value ΔSOC are large, a large equalization current is used to accelerate the equalization speed.

[0156] S604. Combine the balanced current command with the maximum difference in inductor current in the power supply characteristic information and input it into a PID regulator to obtain the maximum duty cycle.

[0157] In an alternative implementation of this embodiment, combine the balanced current command with the maximum difference in inductor current ΔI in the power supply characteristic information of the power supply system. LMAX Input it into a PID regulator (Proportional-Integral-Derivative regulator) to output the maximum duty cycle D. MAX .

[0158] S605. Based on the maximum duty cycle and the multi-level power fuzzy equalization strategy adopted, obtain the duty cycle of each switching tube in the double-layer switched-inductor equalization circuit topology.

[0159] In an alternative implementation of this embodiment, adopt a multi-level power fuzzy equalization strategy of mean-difference, and calculate the duty cycle of each switching tube in the double-layer switched-inductor equalization circuit topology through the maximum duty cycle. The calculation formula includes:

[0160]

[0161] In the formula, D i is the duty cycle of the i-th switching tube, SOC i is the SOC of the i-th battery, SOC ave is the average value of the SOC of all batteries, SOC MAX is the SOC of the battery with the maximum power.

[0162] S606. Perform PWM modulation on the double-layer switched-inductor equalization circuit topology according to the duty cycle of each switching tube to generate an equalization drive signal, and generate an equalization leveling command according to the equalization drive signal.

[0163] In an alternative implementation of this embodiment, perform PWM modulation on the double-layer switched-inductor equalization circuit topology according to the duty cycle of each switching tube in the double-layer switched-inductor equalization circuit topology to generate an equalization drive signal, and generate an equalization leveling command according to the equalization drive signal.

[0164] S104. Level the multi-level topology based on the equalization leveling command and evaluate the leveling effect.

[0165] In an alternative implementation of this embodiment, perform active equalization leveling on the multi-level topology based on the equalization leveling command generated in step S103, and evaluate the leveling effect after leveling ends.

[0166] In summary, Embodiment 1 of the present invention provides a leveling method for an energy storage converter with a multi-level topology structure. By analyzing the switch capacitor converter information in the energy storage converter to be leveled, an accurate multi-level topology structure is generated. By obtaining the charging rate of the power supply system connected to the multi-level topology structure, comparing the theoretical power information and the actual power information, it is determined whether there is a level imbalance phenomenon. An equalization leveling instruction is generated through a multi-level power fuzzy equalization strategy, effectively solving the problem of abnormal operating power caused by multi-level imbalance in the working process of the energy storage converter, improving the stability and safety of the energy storage converter, and improving the working efficiency of the energy storage converter.

[0167] Embodiment 2

[0168] Embodiment 2 of the present invention provides a leveling system for an energy storage converter with a multi-level topology structure. The leveling system for the energy storage converter with the multi-level topology structure is used to implement the leveling method for the energy storage converter with the multi-level topology structure described in Embodiment 1. The system includes a multi-level topology structure generation module, a theoretical power information acquisition module, an equalization leveling instruction generation module, and a leveling evaluation module.

[0169] In an alternative implementation of this embodiment, as Figure 7 shown, Figure 7 shows the architecture diagram of the leveling system for the energy storage converter with the multi-level topology structure in Embodiment 2 of the present invention, including the following modules:

[0170] The multi-level topology structure generation module 10 is configured to obtain the working circuit information of the energy storage converter to be leveled and generate a corresponding multi-level topology structure;

[0171] The obtaining of the working circuit information of the energy storage converter to be leveled and generating a corresponding multi-level topology structure includes: extracting the switch capacitor converter information in the working circuit information of the energy storage converter to be leveled; constructing a multi-order grid model according to the switch capacitor converter information, and matching the switch capacitor converter information with the nodes of the multi-order grid model to generate an adjacency matrix grid model of the switch capacitor converter; calculating the topology conversion efficiency of the adjacency matrix grid model of the switch capacitor converter, and judging the accuracy of the adjacency matrix grid model of the switch capacitor converter according to the topology conversion efficiency; filling the remaining circuit element information in the working circuit information of the energy storage converter to be leveled into the adjacency matrix grid model of the switch capacitor converter to generate the multi-level topology structure corresponding to the energy storage converter to be leveled.

[0172] Calculating the topological conversion efficiency of the adjacency matrix grid model of the switched-capacitor converter includes: calculating the conduction angles of the switched-capacitor converters at each node in the adjacency matrix grid model of the switched-capacitor converter, and generating a switching control signal based on the conduction angles; calculating the switching loss, conduction loss, and driving loss of each switched-capacitor converter, and normalizing the switching loss, conduction loss, and driving loss; constructing a convolutional neural network model, introducing an error function into the convolutional neural network model, and training it; comprehensively calculating the switching control signal and the normalized switching loss, conduction loss, and driving loss based on the trained convolutional neural network model to obtain the topological conversion efficiency of the adjacency matrix grid model of the switched-capacitor converter.

[0173] The theoretical power information acquisition module 20 is configured to collect the power characteristic information of the power supply system connected to the multilevel topology structure, and analyze and obtain the theoretical power information of the multilevel topology structure based on the power characteristic information.

[0174] Collecting the power characteristic information of the power supply system connected to the multilevel topology structure and analyzing and obtaining the theoretical power information of the multilevel topology structure based on the power characteristic information includes: obtaining the charging rate of the power supply system connected to the multilevel topology structure, and analyzing and obtaining the theoretical power information of the multilevel topology structure according to the charging rate.

[0175] Obtaining the charging rate of the power supply system connected to the multilevel topology structure includes: generating an equivalent circuit model of the power supply system, and extracting the voltage, resistance, and capacitance information of the equivalent circuit model in the initial state; simulating the operation of the equivalent circuit model, and extracting the voltage, resistance, and capacitance information of the equivalent circuit model in the operating state; comparing the voltage, resistance, and capacitance information of the equivalent circuit model in the initial state and the operating state to generate the dynamic error parameters of the equivalent circuit model; correcting the equivalent circuit model based on the dynamic error parameters by using a Kalman filter model; extracting the voltage, resistance, and capacitance information of the corrected equivalent circuit model, and generating the charging rate information of the corrected equivalent circuit model.

[0176] Analyzing and obtaining the theoretical power information of the multilevel topology structure according to the charging rate includes: analyzing and obtaining the theoretical power information of the multilevel topology structure in combination with the charging rate according to the hybrid cascade level number of the multilevel topology structure.

[0177] An equalization leveling instruction generation module 30, where the equalization leveling instruction generation module 30 is used to judge the level imbalance phenomenon in the multilevel topology structure based on the theoretical power information, process the level imbalance phenomenon based on the multilevel power fuzzy equalization strategy, and generate an equalization leveling instruction;

[0178] The judging the level imbalance phenomenon in the multilevel topology structure based on the theoretical power information includes: simulating the operation of the multilevel topology structure, obtaining the actual power information of the multilevel topology structure in the simulated operation state, and comparing the actual power information with the theoretical power information, and judging whether the level imbalance phenomenon occurs in the multilevel topology structure based on the comparison result.

[0179] The simulating the operation of the multilevel topology structure and obtaining the actual power information of the multilevel topology structure in the simulated operation state includes: simulating the operation of the multilevel topology structure, analyzing several switching modes of the multilevel topology structure, and respectively extracting the real-time output voltage waveforms of each switching mode; performing carrier PWM modulation on the real-time output voltage waveforms, and performing filtering processing based on the Fourier transform function to generate the real-time drive signal waveforms of each switched capacitor converter in the multilevel topology structure; calculating the port power ratio of each switched capacitor converter in the multilevel topology structure based on the modulation parameters used during carrier PWM modulation; synthesizing the real-time drive signal waveforms and port power ratios of all switched capacitor converters to generate the actual power information of the corresponding switched capacitor converters, and further generating the actual power information of the multilevel topology structure.

[0180] The processing the level imbalance phenomenon based on the multilevel power fuzzy equalization strategy and generating an equalization leveling instruction includes: generating a double-layer switched inductor equalization circuit topology; defining the average power fuzzy language variable, the maximum difference power fuzzy language variable, and the output power fuzzy language variable of the multilevel power fuzzy controller according to the power characteristic information of the power supply system connected to the multilevel topology structure; setting the membership functions of the average power fuzzy language variable, the maximum difference power fuzzy language variable, and the output power fuzzy language variable, and converting the output value of the multilevel power fuzzy controller into an equalization current instruction based on the membership functions; inputting the equalization current instruction combined with the maximum difference of inductor current in the power characteristic information into a PID controller to obtain the maximum duty cycle; obtaining the duty cycle of each switch tube in the double-layer switched inductor equalization circuit topology according to the maximum duty cycle combined with the adopted multilevel power fuzzy equalization strategy; performing PWM modulation on the double-layer switched inductor equalization circuit topology according to the duty cycle of each switch tube to generate an equalization drive signal, and generating an equalization leveling instruction according to the equalization drive signal.

[0181] The leveling evaluation module 40 is configured to level the multilevel topology based on the equalizing leveling instruction and evaluate the leveling effect.

[0182] In summary, Embodiment 2 of the present invention provides a leveling system for an energy storage converter with a multilevel topology. The leveling system for the energy storage converter with a multilevel topology is used to implement the leveling method for the energy storage converter with a multilevel topology described in Embodiment 1. By obtaining the charging rate of the power system connected to the multilevel topology, comparing the theoretical power information and the actual power information, and determining whether there is a level imbalance phenomenon, an equalizing leveling instruction is generated through a multilevel power fuzzy equalizing strategy, effectively solving the problem of abnormal operating power caused by multilevel imbalance during the operation of the energy storage converter, improving the stability and safety of the energy storage converter, and improving the working efficiency of the energy storage converter.

[0183] The above has introduced in detail a leveling method and system for an energy storage converter with a multilevel topology provided by the present invention. Those of ordinary skill in the art can understand that all or part of the steps in the various methods of the above embodiments can be completed by instructing relevant hardware through a program, and this program can be stored in a computer-readable storage medium. The storage medium can include: read-only memory (ROM, Read Only Memory), random access memory (RAM, Random Access Memory), magnetic disk or optical disk, etc.

[0184] In addition, the above has introduced the embodiments of the present invention in detail. Specific examples are used in this article to elaborate on the principle and implementation manner of the present invention. The description of the above embodiments is only used to help understand the method and its core idea of the present invention; at the same time, for those of ordinary skill in the art, according to the idea of the present invention, there will be changes in the specific implementation manner and application scope. In summary, the content of this specification should not be construed as a limitation to the present invention.

Claims

1. A leveling method for an energy storage converter with a multi-level topology structure, characterized in that The method includes: Obtaining the working circuit information of the energy storage converter to be leveled and generating a corresponding multilevel topology structure; Collecting the power characteristic information of the power supply system connected to the multilevel topology structure and analyzing and obtaining the theoretical power information of the multilevel topology structure based on the power characteristic information; Judging the level imbalance phenomenon in the multilevel topology structure based on the theoretical power information, processing the level imbalance phenomenon based on the multilevel power fuzzy equalization strategy, and generating an equalization leveling instruction; Leveling the multilevel topology structure based on the equalization leveling instruction and evaluating the leveling effect.

2. The leveling method of the energy storage converter with a multi-level topology structure according to claim 1, characterized in that The obtaining the working circuit information of the energy storage converter to be leveled and generating a corresponding multilevel topology structure includes: Extracting the switched capacitor converter information in the working circuit information of the energy storage converter to be leveled; Constructing a multi-order grid model according to the switched capacitor converter information, matching the switched capacitor converter information with the nodes of the multi-order grid model, and generating an adjacency matrix grid model of the switched capacitor converter; Calculating the topological conversion efficiency of the adjacency matrix grid model of the switched capacitor converter and judging the accuracy of the adjacency matrix grid model of the switched capacitor converter according to the topological conversion efficiency; Filling the remaining circuit element information in the working circuit information of the energy storage converter to be leveled into the adjacency matrix grid model of the switched capacitor converter to generate the multilevel topology structure corresponding to the energy storage converter to be leveled.

3. The leveling method for an energy storage converter with a multilevel topology structure as described in claim 2, characterized in that, The calculating the topological conversion efficiency of the adjacency matrix grid model of the switched capacitor converter includes: Calculating the conduction angle of the switched capacitor converter on each node in the adjacency matrix grid model of the switched capacitor converter and generating a switching control signal based on the conduction angle; Calculating the switching loss, conduction loss and driving loss of each switched capacitor converter and normalizing the switching loss, conduction loss and driving loss; Constructing a convolutional neural network model, introducing an error function into the convolutional neural network model and training it; Based on the trained convolutional neural network model, comprehensively calculating the switching control signal and the normalized switching loss, conduction loss and driving loss to obtain the topological conversion efficiency of the adjacency matrix grid model of the switched capacitor converter.

4. The leveling method of the energy storage converter with a multi-level topology structure according to claim 1, characterized in that The collecting the power characteristic information of the power supply system connected to the multilevel topology structure and analyzing and obtaining the theoretical power information of the multilevel topology structure based on the power characteristic information includes: Obtaining the charging rate of the power supply system connected to the multilevel topology structure and analyzing and obtaining the theoretical power information of the multilevel topology structure according to the charging rate.

5. The leveling method of the energy storage converter with a multi-level topology structure according to claim 4, characterized in that, The obtaining the charging rate of the power supply system connected to the multilevel topology structure includes: Generating an equivalent circuit model of the power supply system and extracting the voltage, resistance and capacitance information of the equivalent circuit model in the initial state; Simulating the operation of the equivalent circuit model and extracting the voltage, resistance and capacitance information of the equivalent circuit model in the operating state; Compare the voltage, resistance, and capacitance information of the equivalent circuit model in the initial state and the operating state to generate the dynamic error parameters of the equivalent circuit model; Based on the dynamic error parameters, use the Kalman filter model to correct the equivalent circuit model; Extract the voltage, resistance, and capacitance information of the corrected equivalent circuit model, and generate the charging rate information of the corrected equivalent circuit model.

6. The leveling method of the energy storage converter with a multi-level topology structure according to claim 4, characterized in that The obtaining the theoretical power information of the multilevel topology structure according to the charging rate analysis includes: According to the number of hybrid cascade levels of the multilevel topology structure, combine the charging rate analysis to obtain the theoretical power information of the multilevel topology structure.

7. The leveling method of the energy storage converter with a multi-level topology structure according to claim 1, characterized in that The determining the level imbalance phenomenon in the multilevel topology structure based on the theoretical power information includes: Simulate the operation of the multilevel topology structure, obtain the actual power information of the multilevel topology structure in the simulated operation state, and compare the actual power information with the theoretical power information. Based on the comparison result, determine whether there is a level imbalance phenomenon in the multilevel topology structure.

8. The leveling method of the energy storage converter with a multi-level topology structure according to claim 7, characterized in that, The obtaining the actual power information of the multilevel topology structure in the simulated operation state by simulating the operation of the multilevel topology structure includes: Simulate the operation of the multilevel topology structure, analyze several switching modes of the multilevel topology structure, and respectively extract the real-time output voltage waveforms of each switching mode; Perform carrier PWM modulation on the real-time output voltage waveforms, and perform filtering processing based on the Fourier transform function to generate the real-time drive signal waveforms of each switched-capacitor converter in the multilevel topology structure; Based on the modulation parameters used in the carrier PWM modulation, calculate the port power ratio of each switched-capacitor converter in the multilevel topology structure; Integrate the real-time drive signal waveforms and port power ratios of all switched-capacitor converters to generate the actual power information of the corresponding switched-capacitor converters, and then generate the actual power information of the multilevel topology structure.

9. The leveling method of the energy storage converter with a multi-level topology structure according to claim 1, characterized in that The processing the level imbalance phenomenon based on the multilevel power fuzzy equalization strategy to generate an equalization leveling instruction includes: Generate a double-layer switched-inductor equalization circuit topology; According to the power characteristic information of the power supply system connected to the multilevel topology structure, define the average power fuzzy linguistic variable, maximum difference power fuzzy linguistic variable, and output power fuzzy linguistic variable of the multilevel power fuzzy controller; Set the membership functions of the average power fuzzy linguistic variable, maximum difference power fuzzy linguistic variable, and output power fuzzy linguistic variable, and convert the output value of the multilevel power fuzzy controller into an equalization current instruction based on the membership functions; Input the equalization current instruction and the maximum difference in inductor current in the power characteristic information into a PID controller to obtain the maximum duty cycle; According to the maximum duty cycle and the adopted multilevel power fuzzy equalization strategy, obtain the duty cycle of each switch in the double-layer switched-inductor equalization circuit topology; Perform PWM modulation on the double-layer switched inductor equalization circuit topology according to the duty cycle of each switching tube to generate an equalization drive signal, and generate an equalization leveling instruction according to the equalization drive signal.

10. A leveling system for an energy storage converter with a multi-level topology, characterized in that, The multi-level topology energy storage converter leveling system is used to implement the multi-level topology energy storage converter leveling method according to any one of claims 1-9. The system includes: A multi-level topology generation module, which is used to obtain the working circuit information of the energy storage converter to be leveled and generate a corresponding multi-level topology; A theoretical power information acquisition module, which is used to collect the power characteristic information of the power supply system connected to the multi-level topology and analyze and obtain the theoretical power information of the multi-level topology based on the power characteristic information; An equalization leveling instruction generation module, which is used to judge the level imbalance phenomenon in the multi-level topology based on the theoretical power information, process the level imbalance phenomenon based on the multi-level power fuzzy equalization strategy, and generate an equalization leveling instruction; A leveling evaluation module, which is used to level the multi-level topology based on the equalization leveling instruction and evaluate the leveling effect.

Images