Consistency algorithm-based multi-type energy storage system double-layer frequency modulation control method, system, equipment and medium

Abstract

The invention relates to the technical field of multi-type energy storage system operation control, and discloses a multi-type energy storage system double-layer frequency modulation control method, system and device based on a consistency algorithm and a medium. Constructing a first frequency control model for accessing the multi-type energy storage system to the large power grid according to a division result; analyzing a frequency response process after the power grid is disturbed, designing a fuzzy controller based on the first frequency control model, determining weight factors of different control modes, and obtaining a first energy storage system control model; and based on the first energy storage system control model, state observation based on neighbor frequency deviation is introduced, a second energy storage system control model is constructed, and double-layer frequency modulation control of the energy storage system is realized. According to the invention, the multi-type energy storage system double-layer frequency modulation control model based on the consistency algorithm is constructed, and the operation control of the multi-type energy storage system can be effectively realized.

Description

Technical Field

[0001] This invention relates to the field of operation control technology for multiple types of energy storage systems, and in particular to a two-layer frequency regulation control method, system, equipment and medium for multiple types of energy storage systems based on a consensus algorithm. Background Technology

[0002] "Peaking carbon emissions" and "carbon neutrality" are important national strategic development goals, and building a new power system based on clean energy is a key measure to achieve these goals. Vigorously developing solar and wind power and promoting the grid connection and consumption of a high proportion of renewable energy have become concrete paths for my country to build a new power system.

[0003] With a large amount of distributed renewable energy being connected to the AC grid via power electronic converters, my country's future power system will exhibit "high-frequency and high-renewal" characteristics. The uncertainty of distributed photovoltaic and wind power output increases the frequency regulation demand of the grid. Furthermore, the abundance of controllable resources in the new power system exacerbates the complexity of operating conditions in high-proportion distributed AC / DC hybrid grids. The coupling and coordinated operation of numerous flexible devices such as power electronic components make distribution network operation and control more difficult, and the establishment of grid control models becomes increasingly complex, severely limiting the efficient grid connection of large-scale distributed resources, improving operational flexibility, and enhancing grid power supply capacity. Energy storage systems, with their rapid response capabilities, have become an important means of coping with frequency fluctuations. However, currently, energy storage participation in grid frequency regulation mostly adopts decentralized and independent control strategies, and different types of energy storage have different operating characteristics, resulting in low energy storage utilization and failing to fully realize the frequency regulation potential of energy storage.

[0004] Therefore, in order to fully explore its regulation potential, realize flexible interaction and coordinated regulation of multiple types of energy storage, and ensure a deep balance between supply and demand, it is urgent to study the two-layer frequency regulation control strategy of multiple types of energy storage systems based on consensus algorithms, and take the frequency regulation control of multiple types of energy storage systems as the core technology for the research of multiple types of energy storage systems. Summary of the Invention

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

[0006] Therefore, this invention provides a two-layer frequency regulation control method, system, device and medium for multi-type energy storage systems based on consensus algorithms, which solves the problems of insufficient frequency regulation accuracy and low energy storage resource utilization caused by the use of decentralized independent control strategies and failure to coordinate and utilize the complementary characteristics of multiple types of energy storage when existing energy storage systems participate in grid frequency regulation.

[0007] To solve the above-mentioned technical problems, the present invention provides the following technical solution: In a first aspect, the present invention provides a two-layer frequency regulation control method for multiple types of energy storage systems based on a consensus algorithm, comprising: The energy storage system is divided into categories using the first clustering method, and a first frequency control model for connecting multiple types of energy storage systems to the power grid is constructed based on the classification results. The frequency response process of the power grid after being disturbed is analyzed, a fuzzy controller is designed based on the first frequency control model, the weight factors of different control methods are determined, and the control model of the first energy storage system is obtained. Based on the first energy storage system control model, a second energy storage system control model is constructed by introducing state observation based on neighbor frequency deviation, thereby realizing two-layer frequency regulation control of the energy storage system.

[0008] As a preferred embodiment of the two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithms described in this invention, the analysis of the frequency response process of the power grid after disturbance includes: Based on the dynamic response curve of the power grid frequency after being disturbed, the frequency fluctuation process is divided into the first fluctuation stage, the second fluctuation stage and the third fluctuation stage. For the first fluctuation phase, a control method combining virtual inertial control and droop control is adopted; For the second fluctuation phase, a control method combining virtual negative inertia control and droop control is adopted; For the third fluctuation stage, a droop control method is adopted.

[0009] As a preferred embodiment of the two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithms described in this invention, the weighting factors for determining different control modes include: Construct a fuzzy controller; The universe of discourse and fuzzy subsets are set for the input and output variables of the fuzzy controller, respectively, and the membership relationship of each fuzzy subset is defined by the first membership function; Based on the control requirements at different stages of frequency fluctuation, a fuzzy control rule table is established between input and output variables, and the weight factors for different control methods are obtained according to the fuzzy control rule table.

[0010] As a preferred embodiment of the two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithms described in this invention, the step of obtaining the first energy storage system control model includes: Based on the weighting factors of the different control methods, the power commands generated by virtual droop control, virtual inertial control and virtual negative inertial control are weighted and fused to generate a primary control active power command. Based on the primary control active power command, a control model for the first energy storage system for primary frequency regulation is formed.

[0011] The beneficial effects of this preferred technical solution are: it enables the system to automatically switch the optimal control strategy during different frequency fluctuation stages, which not only gives full play to the technical advantages of each control mode, but also avoids the limitations of a single control mode, and greatly improves the adaptability and accuracy of primary control.

[0012] As a preferred embodiment of the two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithms described in this invention, the construction of the second energy storage system control model includes: Each energy storage unit in the system is defined as an agent. The integral of the frequency deviation of each agent's grid connection point is used as the consistency state variable. A nonlinear state predictor based on the neighbor frequency deviation is designed to predict the changing trend of each agent's state. A consensus protocol is constructed that integrates state difference measurement and nonlinear state prediction. The consensus protocol dynamically calculates the secondary frequency correction based on the state of each agent, the state of neighboring agents, and the state prediction. Based on the aforementioned consensus protocol, a distributed control model is established with the goal of eliminating frequency steady-state error and achieving frequency synchronization, serving as the control model for the second energy storage system.

[0013] The beneficial effects of this preferred technical solution are: it achieves rapid elimination of frequency deviation and system synchronization through distributed coordination, effectively overcoming the communication bottleneck and single-point failure risk of traditional centralized control.

[0014] As a preferred embodiment of the two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithm described in this invention, wherein: the division of the energy storage system using the first clustering method includes: Electrochemical energy storage is used as an energy-type energy storage solution to respond to system frequency power fluctuations. Supercapacitor energy storage and flywheel energy storage are used as power-type energy storage to jointly respond to high-frequency power fluctuations in the system.

[0015] As a preferred embodiment of the two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithms described in this invention, the construction of the first frequency control model for connecting multi-type energy storage systems to the power grid includes: A first-order inertial element is used to describe the external response characteristics of electrochemical energy storage when participating in grid frequency regulation, and a dynamic model of the electrochemical energy storage unit is established. The equivalent circuit model of the supercapacitor energy storage unit is constructed using the classical equivalent model; The power response characteristics of the flywheel energy storage unit are described by a first-order inertial element, and a system model of the flywheel energy storage unit is established. A frequency response model for thermal power units is constructed by combining the transfer functions of the governor, reheat turbine, generator, and load. The models of electrochemical energy storage, supercapacitor energy storage, flywheel energy storage units, and thermal power units are integrated to generate the first frequency control model for connecting the various types of energy storage systems to the power grid.

[0016] Secondly, this invention provides a two-layer frequency regulation control system for multiple types of energy storage systems based on a consensus algorithm, including: The frequency control model construction module is used to divide the energy storage system using the first clustering method and construct the first frequency control model for connecting multiple types of energy storage systems to the power grid based on the division results. The primary control model construction module is used to analyze the frequency response process of the power grid after disturbance, design a fuzzy controller based on the first frequency control model, determine the weight factors of different control methods, and obtain the first energy storage system control model. The secondary control model construction module is used to construct a second energy storage system control model based on the first energy storage system control model and introduce state observation based on neighbor frequency deviation, so as to realize the two-layer frequency regulation control of the energy storage system.

[0017] Thirdly, the present invention provides an electronic device, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor executes the computer-executable instructions to implement the steps of a two-layer frequency modulation control method for multi-type energy storage systems based on a consensus algorithm.

[0018] Fourthly, the present invention provides a computer-readable storage medium storing computer-executable instructions, which, when executed by a processor, implement the steps of a two-layer frequency modulation control method for multi-type energy storage systems based on a consensus algorithm.

[0019] Compared with existing technologies, the beneficial effects of this invention are as follows: This invention establishes a two-layer frequency regulation control model for multi-type energy storage systems based on a consensus algorithm. According to the response characteristics of the energy storage system, energy storage is divided into power-type energy storage and energy-type energy storage. High-frequency fluctuation components are allocated to the responses of supercapacitors and flywheel energy storage, while low-frequency fluctuation components are allocated to the electrochemical energy storage response. Each energy storage PCS adopts a control method combining virtual droop control and virtual inertial control, and the weighting factors for different control methods are determined through a fuzzy controller. Based on the consensus algorithm, the frequency deviation integral of each energy storage grid connection point is used as the state variable of each agent. A state observer is introduced to construct a secondary control model for multi-type energy storage systems based on frequency deviation integral and state prediction. This achieves coordinated control of multi-type energy storage systems, and the two-layer frequency regulation control model based on the consensus algorithm effectively realizes the operation control of multi-type energy storage systems. Attached Figure Description

[0020] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments will be briefly introduced. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0021] Figure 1 This is a schematic diagram of the overall flow logic of a two-layer frequency regulation control method for multi-type energy storage systems based on a consensus algorithm, provided in one embodiment of the present invention.

[0022] Figure 2 The flowchart illustrates the solution process for a centralized-distributed dual-layer control model of a distribution network based on a consensus algorithm for a dual-layer frequency regulation control method for multi-type energy storage systems, as provided in an embodiment of the present invention.

[0023] Figure 3 The graph shows the frequency response curve of the power grid after disturbance, based on a consensus algorithm-based two-layer frequency regulation control method for multi-type energy storage systems provided in one embodiment of the present invention. Detailed Implementation

[0024] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0025] Example 1, referring to Figure 1 As one embodiment of the present invention, a two-layer frequency regulation control method for multi-type energy storage systems based on a consensus algorithm is provided, such as... Figure 1 The specific steps shown are as follows: S100: The energy storage system is divided into categories using the first clustering method, and a first frequency control model for connecting multiple types of energy storage systems to the power grid is constructed based on the classification results. S200: Analyze the frequency response process of the power grid after disturbance, design a fuzzy controller based on the first frequency control model, determine the weight factors of different control methods, and obtain the first energy storage system control model; S300: Based on the first energy storage system control model, a second energy storage system control model is constructed by introducing state observation based on neighbor frequency deviation, thereby realizing dual-layer frequency regulation control of the energy storage system.

[0026] It should be noted that, to address the issues of insufficient frequency regulation accuracy and low energy storage resource utilization caused by the use of decentralized and independent control strategies and the failure to coordinate and utilize the complementary characteristics of multiple energy storage types when existing energy storage systems participate in grid frequency regulation, steps S100-S300 establish a two-layer frequency regulation control model for multiple types of energy storage systems based on a consensus algorithm. According to the response characteristics of the energy storage system, energy storage is divided into power-type energy storage and energy-type energy storage. High-frequency fluctuation components are allocated to the responses of supercapacitors and flywheel energy storage, while low-frequency fluctuation components are allocated to the response of electrochemical energy storage. Each energy storage PCS adopts a control method combining virtual droop control and virtual inertial control, and the weighting factors for different control methods are determined through a fuzzy controller. Based on the consensus algorithm, the frequency deviation integral of each energy storage grid connection point is used as the state variable of each agent. A state observer is introduced to construct a secondary control model for multiple types of energy storage systems based on frequency deviation integral and state prediction. This achieves coordinated control of multiple types of energy storage systems, and the two-layer frequency regulation control model based on the consensus algorithm effectively realizes the operation control of multiple types of energy storage systems.

[0027] Example 2, refer to Figure 2 and Figure 3 Based on the previous embodiment, this embodiment provides a specific implementation method for a two-layer frequency regulation control method for multi-type energy storage systems based on a consensus algorithm, in order to illustrate the technical means used in this method.

[0028] In this embodiment of the invention, step S100 includes the following sub-steps A1 and A2: In A1: The energy storage system is divided using the first clustering method; Specifically, in this embodiment, electrochemical energy storage is used as an energy-type energy storage to respond to the system's medium-frequency power fluctuations; supercapacitor energy storage and flywheel energy storage are used as power-type energy storage to jointly respond to the system's high-frequency power fluctuations.

[0029] It should be noted that, based on the characteristics of energy storage technology, energy storage can be divided into energy-type energy storage, which has high energy density and large storage capacity, and power-type energy storage, which has high power density, fast response speed, and can be frequently charged and discharged. Energy-type energy storage has a slower response speed but high energy density, making it suitable for providing energy support to the system for extended periods to cope with medium-frequency load fluctuations in the power grid. Power-type energy storage has a fast response speed but low energy density, making it suitable for providing energy support to the power grid for short periods to cope with high-frequency load fluctuations in the power grid.

[0030] In an optional embodiment, the first clustering method can also be based on the dynamic response time constant of the energy storage. By analyzing the time required for each energy storage unit to reach the target output after receiving a power command, supercapacitors and flywheel energy storage with short response times are classified into fast-response clusters, responsible for suppressing instantaneous high-frequency disturbances; while electrochemical energy storage with relatively long response times is classified into slow-response clusters, used for smooth power regulation and energy support.

[0031] In another optional embodiment, the first clustering method can also perform functional clustering based on the applicable frequency regulation scenarios of energy storage. According to the actual role of energy storage in grid frequency regulation, energy storage units suitable for primary frequency regulation and requiring rapid absorption or release of power are divided into instantaneous power support clusters; energy storage units suitable for secondary frequency regulation and capable of providing continuous power adjustment are divided into continuous power regulation clusters, thereby achieving division of labor and cooperation according to frequency regulation functional requirements.

[0032] In A2: Based on the division results, a first frequency control model for connecting multiple types of energy storage systems to the main power grid is constructed, such as... Figure 2 The following are included: A first-order inertial element is used to describe the external response characteristics of electrochemical energy storage when participating in grid frequency regulation, and a dynamic model of the electrochemical energy storage unit is established. The equivalent circuit model of the supercapacitor energy storage unit is constructed using the classical equivalent model; The power response characteristics of the flywheel energy storage unit are described by a first-order inertial element, and a system model of the flywheel energy storage unit is established. A frequency response model for thermal power units is constructed by combining the transfer functions of the governor, reheat turbine, generator, and load. By integrating models of electrochemical energy storage, supercapacitor energy storage, flywheel energy storage units, and thermal power units, a first frequency control model for connecting multiple types of energy storage systems to the large power grid is generated.

[0033] In an optional embodiment, the first frequency control model can also be constructed using a unified modeling method based on port equivalence. All types of energy storage units and thermal power units are considered as controlled current sources or power sources with specific impedance characteristics. By defining unified interface variables at common connection points, a system-wide frequency domain model described in the form of nodal admittance matrices is established.

[0034] In another optional embodiment, the first frequency control model can also be constructed as a control-oriented hierarchical aggregation model. Based on the clustering results, multiple energy storage units of the same type are electrically equivalent to a single aggregation unit. For example, all electrochemical energy storage is aggregated into an equivalent large-capacity battery, and all power energy storage is aggregated into an equivalent fast-response source. Subsequently, a simplified frequency response model is established based on the equivalent multi-energy system, comprising three main components: energy storage, power storage, and thermal power units.

[0035] Specifically, the modeling of electrochemical energy storage units includes: the electrochemical reaction mechanism of the battery during charging and discharging is difficult to express using a mathematical model. Considering the severe coupling of various parameters during dynamic processes, complex mathematical models are not suitable for optimizing multi-type energy storage coordinated control systems. Based on the output characteristics of the battery during charging and discharging, it can be simplified to a first-order inertial equivalent model, the mathematical model of which is: in, For electrochemical energy storage, a transfer function model is used. This is the battery delay response time constant; and These are the virtual droop output and virtual inertial output of electrochemical energy storage, respectively. , These are the weighting factors for virtual droop and virtual inertial control, respectively. , These are the virtual droop coefficient and virtual inertia coefficient for electrochemical energy storage, respectively. This is the low-frequency signal after frequency division processing.

[0036] Specifically, the modeling of the supercapacitor energy storage unit includes: the supercapacitor model adopts a first-order RC equivalent model. This model ignores the influence of environmental factors on the resistance value and is suitable for applications with frequent charging and discharging and low precision requirements, such as primary frequency regulation and low voltage ride-through. Its mathematical expression is: in, U This is the voltage of the supercapacitor; U c This is the equivalent capacitance voltage; i c This is the output current of the supercapacitor; P c This refers to the power of the supercapacitor. The external characteristics of supercapacitors participating in grid frequency regulation adopt a first-order inertial element, and its mathematical expression is as follows: in, , These are the changes in current and voltage of the supercapacitor. T SC It is a time constant. K v,SC For voltage-controlled gain, R eq Equivalent resistance; C This is the equivalent capacitance of a supercapacitor. U SC,0 The initial voltage; Δ P SC Δ U SC Theoretical output power and voltage change of supercapacitors.

[0037] Specifically, the flywheel energy storage model also uses a first-order inertial element to represent its system model, and the transfer function is: in, T FW The time constant of the flywheel energy storage system; flywheel energy storage SOC for: in, SOC 0 represents the initial state of charge of the flywheel energy storage. P FW For flywheel energy storage output power, MW, E This represents the total energy storage capacity of the flywheel energy storage system.

[0038] Specifically, the construction of the thermal power unit model includes: after receiving the frequency regulation signal, the thermal power unit outputs active power sequentially through the governor and the reheat turbine unit, and its frequency response model is as follows: in, The time constant of the governor of the thermal power unit; , , These are the reheater gain, heater time constant, and turbine time constant, respectively. s The Laplace transform operator is used; the transfer function model for the generator and load is as follows: in, H Let be the generator's inertial time constant. D This is the load damping coefficient.

[0039] It should be noted that step S100 above scientifically classifies multiple types of energy storage systems through clustering and constructs a unified frequency control model, which effectively solves the problem of difficulty in coordinating the heterogeneous characteristics of different types of energy storage units, realizes targeted responses to different frequency components, and significantly improves the overall response capability of energy storage systems to grid frequency fluctuations.

[0040] In this embodiment of the invention, step S200 includes the following sub-steps B1 to B3: In B1: Analyze the frequency response process of the power grid after being disturbed; Specifically, such as Figure 3 As shown in the dynamic response curve after the power grid frequency is disturbed, the frequency response process after the power grid is disturbed can be divided into the first fluctuation stage, the second fluctuation stage and the third fluctuation stage. Specifically, for the first fluctuation stage, namely the frequency deterioration stage, a control method combining virtual inertial control and droop control is adopted. Specifically, for the second fluctuation stage, namely the frequency recovery stage, a control method combining virtual negative inertia control and droop control is adopted. Specifically, for the third fluctuation stage, namely the steady-state frequency fluctuation stage, a droop control method is adopted.

[0041] It should be noted that virtual droop control is suitable for regulating steady-state frequency fluctuations, while virtual inertia and virtual complex inertia control are suitable for suppressing transient frequency fluctuations.

[0042] In B2: A fuzzy controller is designed based on the first frequency control model, and the weighting factors for different control methods are determined; the detailed steps include: Construct a fuzzy controller; The universe of discourse and fuzzy subsets are set for the input and output variables of the fuzzy controller, respectively, and the membership relationship of each fuzzy subset is defined by the first membership function. Based on the control requirements at different stages of frequency fluctuation, a fuzzy control rule table is established between input and output variables, and the weight factors for different control methods are obtained from the fuzzy control rule table.

[0043] It should be noted that a single control mode is insufficient to handle complex frequency fluctuations. Different frequency fluctuation phases require different control measures. Therefore, a mechanism is needed to coordinate and combine the outputs of these control modes, i.e., to determine the weight of each mode in the total output power. The total primary control active power command can be expressed as a weighted sum of the contributions of these three modes: in, P i ( t ), Pi,D ( t ), P i,I ( t ), P i,NI ( t ) is an energy storage unit i At any moment t The total primary control active power command, the power command generated based on virtual droop control, the power command generated based on virtual inertial control, and the power command generated based on virtual negative inertial control; μ D ( t ), μ I ( t ), μ NI ( t The virtual droop, virtual inertia, and virtual negative inertia control modes are respectively located at time [time value missing]. t The output weight.

[0044] Specifically, the fuzzy controller is based on the system frequency deviation Δ f and the rate of change of frequency deviation |dΔ f |The fuzzy control rules of the system are established as the output, and the input quantity Δ f and |dΔ f The universe of discourse of | is set to [-1, 1], and the output is... μ D , μ I、 μ NI The universe of discourse is set to [0,1], and the fuzzy subsets of the two input quantities are... M I and the fuzzy subset of the output M O They are respectively: in, N Represents negative. Z Represents zero. P Represents positive. S Represents small. M Medium L Represents "Large". V It means very.

[0045] Specifically, the membership functions of both the input and output quantities adopt the triangular membership function. Based on the analysis of the selection of weighting factors according to the changes in the high-frequency signal, the fuzzy rules are established as shown in Table 1.

[0046] Table 1: Fuzzy control rules for frequency deviation and its rate of change.

[0047] In an alternative embodiment, the first membership function can also be a Gaussian membership function, which has a smooth, continuous curve and is differentiable everywhere, and can more naturally describe fuzzy concepts such as frequency deviations of approximately zero, allowing membership to transition smoothly across the universe of discourse.

[0048] In another alternative embodiment, the first membership function may also be a trapezoidal membership function, the shape of which is between a triangle and a rectangle, having a stable region in which the membership degree remains 1.

[0049] Specifically, based on the fuzzy control rule table, the weighting factors of various control methods can be obtained, thereby determining the output under different control methods.

[0050] In an optional embodiment, dynamic allocation of weight factors can also be achieved by constructing an adaptive weight allocator based on dynamic programming. By calculating the optimal trajectory of system frequency recovery within a finite time domain in an online rolling manner, and with the goal of minimizing the integral of frequency deviation, the optimal weight sequence of each control mode at different times is solved in reverse, thereby achieving feedforward precise setting of weight factors.

[0051] In another alternative embodiment, the dynamic allocation of weight factors can also be achieved using a weight decision network based on reinforcement learning. By constructing a deep Q-network or a policy gradient model, the system can autonomously learn a weight allocation strategy through continuous interaction with the power grid environment, and output a combination of weight factors that maximizes long-term frequency regulation benefits based on multi-dimensional state information such as frequency deviation, rate of change, and energy storage status.

[0052] In B3: Construct the control model for the first energy storage system; Specifically, based on the weighting factors of different control methods, the power commands generated by virtual droop control, virtual inertial control and virtual negative inertial control are weighted and fused to generate primary control active power commands, thereby forming the first energy storage system control model for primary frequency regulation.

[0053] Specifically, considering the applicable scenarios of different control methods, a fuzzy controller is used to determine the weights of different control methods. Simultaneously, considering the SOC of the energy storage system, this paper divides the SOC into minimum values ​​(…). S min ), smaller value ( S low ), larger value ( S high ), maximum value ( Smax The coefficients for the three control methods are as follows: The charge / discharge coefficients for virtual droop control are as follows: in, K BED,max This represents the maximum value of the electrochemical energy storage droop control coefficient. S BE For the actual SOC of electrochemical energy storage, α c , β c , α d , β d This is the charge / discharge adaptive factor.

[0054] Considering that inertial response requires bidirectional power (discharging during acceleration and charging during deceleration), and that virtual negative inertia may also require bidirectional power, a unified SOC scaling function is designed. f BE ( S )∈[0,1], used to limit the inertia and negative inertia contribution of energy storage in the extreme case of SOC, the scaling function is calculated as follows: in, S i Indicates the first i The SOC scaling factor for each energy storage unit; The virtual inertia coefficient and the virtual negative inertia coefficient can be expressed as: in, K iα,M For virtual inertia coefficients, K iα,N This is a virtual negative inertia coefficient.

[0055] In an optional embodiment, the construction of the first energy storage system control model can also employ an online optimization framework based on model predictive control. By establishing a predictive model that incorporates the dynamic characteristics of energy storage and SOC constraints, the optimal power allocation problem in the finite time domain is solved in each control cycle, directly calculating the multi-objective optimization command that takes into account both frequency regulation requirements and energy storage operating status, thereby constructing a first energy storage system control model with feedforward optimization capabilities.

[0056] In another alternative embodiment, the first energy storage system control model can also be a distributed decision-making model based on multi-agent reinforcement learning. Each energy storage unit is treated as an independent agent, and a joint reward function is designed with the objectives of minimizing frequency deviation and achieving SOC balance. This allows each agent to autonomously learn the optimal cooperative control strategy under a decentralized architecture, ultimately forming a distributed primary frequency control model capable of adapting to complex operating conditions.

[0057] It should be noted that step S200 above innovatively achieves dynamic adaptive adjustment of the weight factors of different control modes by introducing a fuzzy controller design. It can adjust the output ratio of virtual droop control, virtual inertia control and virtual negative inertia control in real time according to the frequency deviation and its rate of change, so that the system can automatically switch the optimal control strategy at different frequency fluctuation stages. This not only gives full play to the technical advantages of each control mode, but also avoids the limitations of a single control mode, and greatly improves the adaptability and accuracy of primary control.

[0058] In this embodiment of the invention, step S300, based on the first energy storage system control model, introduces state observation based on neighbor frequency deviation to construct a second energy storage system control model, thereby realizing dual-layer frequency regulation control of the energy storage system, including the following sub-steps C1~C3: In C1: Each energy storage unit in the system is defined as an agent. The integral of the frequency deviation of each agent's grid connection point is used as the consistency state variable. A nonlinear state predictor based on the neighbor frequency deviation is designed to predict the changing trend of each agent's state. In C2: Construct a consensus protocol that integrates state difference measurement and nonlinear state prediction. The consensus protocol dynamically calculates the secondary frequency correction based on the state of each agent, the state of neighboring agents, and the state prediction. In C3: Based on the consensus protocol, a distributed control model is established with the goal of eliminating frequency steady-state error and achieving frequency synchronization, serving as the control model for the second energy storage system.

[0059] Specifically, under primary control, the system frequency will deviate from the rated value due to load fluctuations, resulting in steady-state error. The objectives of secondary control are: ① Frequency recovery: to restore the frequency of all energy storage units. f i Accurately restore to rated frequency f 0, that is ② Frequency synchronization: Ensure that the frequency of all energy storage units remains consistent, i.e. .

[0060] Specifically, to eliminate the frequency deviation generated during a single control process, the following key error state variables are defined: Frequency deviation: ; Frequency deviation integral: ; Secondary frequency correction amount: x i ( t )=Δ f ci ( t ); Specifically, the proposed secondary control strategy is defined by... x i ( t The dynamic equations of ) enable it to adapt to frequency deviations. e i ( t ) and its integral Adjustments are made to the information exchange with neighbors. The calculation formula is: in, K P,x and K I,x These are proportional and integral gains, which directly affect the frequency deviation. e i ( t ) and its integral ϕ i ( t ).

[0061] Specifically, basic consensus protocols can achieve frequency recovery and synchronization, but their dynamic response speed may not be fast enough, especially when faced with frequent disturbances. To further improve the system's dynamic performance and convergence speed, the concept of state prediction is introduced, combined with a nonlinear function to enhance the controller's effectiveness under different error magnitudes. The nonlinear function is of the following form: in, c >0 is an adjustable gain that determines the strength of the nonlinear term; 0< α <1 is an adjustable exponential parameter. When the error | z When | is large, sig( z The linear part of the function z It dominates and provides stable linear convergence properties. However, when the error | z As it approaches zero, the nonlinear term... c | z | α sgn( z The convergence speed of the inequality term is much faster than that of the linear term. This is because for 0 < α <1, when | z |→0,| z | αThe rate of descent is equal to | z Slow, which means... z | α / | z |=| z | (α-1) The value tends towards infinity, thus providing a relatively stronger control. This allows the system to quickly bring the error to zero when the error is small, significantly accelerating the convergence process, especially near the equilibrium point.

[0062] Specifically, based on the above sig( z The function defines the nonlinear state predictor. y i ( t ): It should be noted that by using intelligent agents i State prediction y i ( t ) and its neighbors' state predictions y j ( t The difference terms between the states are introduced into the protocol. Each agent not only considers the difference between its own current state and that of its neighbors, but can also adjust its own state in advance by predicting the possible changing trends of its neighbors, thereby accelerating the uniform convergence of the entire system.

[0063] Specifically, a consensus protocol that integrates nonlinear state prediction: in, γ The gain of the state prediction term. a ij This represents the communication weighting coefficient between energy storage systems.

[0064] In an optional embodiment, the control model of the second energy storage system can also be constructed using a distributed observer design based on an event-triggered mechanism. By designing a local state observer for each agent and introducing event-triggered conditions based on a frequency error threshold, neighbor communication and state updates are only performed when the system state changes exceed a set tolerance. This significantly reduces the system communication burden while ensuring frequency recovery accuracy, enhancing its practicality in communication-constrained scenarios.

[0065] In another alternative embodiment, the control model of the second energy storage system can also be constructed based on an adaptive gain-adjusted sliding mode consistency controller. By designing a sliding surface that includes the frequency deviation and its integral, and introducing an adaptive gain adjustment law that is robust to system uncertainties and disturbances, each agent can quickly achieve frequency synchronization and steady-state error elimination even in the presence of modeling errors and external disturbances.

[0066] It should be noted that step S300 treats each energy storage unit as an intelligent agent, achieving rapid elimination of frequency deviation and system synchronization through distributed coordination, effectively overcoming the communication bottlenecks and single-point failure risks of traditional centralized control. Secondary control complements primary control, together forming a complete two-layer frequency regulation system, ultimately achieving coordinated and optimized operation of multiple types of energy storage and high-quality restoration of grid frequency.

[0067] Example 3: This example provides a two-layer frequency regulation control system for multiple types of energy storage systems based on a consensus algorithm, including: The frequency control model construction module is used to divide the energy storage system using the first clustering method and construct the first frequency control model for connecting multiple types of energy storage systems to the power grid based on the division results. The primary control model construction module is used to analyze the frequency response process of the power grid after disturbance, design a fuzzy controller based on the first frequency control model, determine the weight factors of different control methods, and obtain the first energy storage system control model. The secondary control model construction module is used to construct a second energy storage system control model based on the first energy storage system control model and introduce state observation based on neighbor frequency deviation, so as to realize the two-layer frequency regulation control of the energy storage system.

[0068] It should be noted that the technical solution of the two-layer frequency regulation control system for multi-type energy storage systems based on consensus algorithm is based on the same concept as the technical solution of the two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithm described above. For details not described in detail in the technical solution of the two-layer frequency regulation control system for multi-type energy storage systems based on consensus algorithm described above, please refer to the description of the technical solution of the two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithm described above.

[0069] The above-mentioned unit modules can be embedded in the processor of the electronic device in hardware form or independent of it, or they can be stored in the memory of the electronic device in software form, so that the processor can call and execute the corresponding operations of the above modules.

[0070] This embodiment also provides an electronic device, which includes a processor, a memory, a communication interface, a display screen, and an input device connected via a system bus. The processor provides computing and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores an operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs in the non-volatile storage medium. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a two-layer frequency modulation control method for multi-type energy storage systems based on a consensus algorithm. The display screen can be a liquid crystal display (LCD) or an e-ink display. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0071] This embodiment also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the method proposed in the above embodiments.

[0072] The storage medium proposed in this embodiment belongs to the same inventive concept as the method proposed in the above embodiments. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0073] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory, random access memory, flash memory, hard disk, or optical disk, and includes several instructions to cause an electronic device (which may be a personal computer, server, or network device, etc.) to execute the method of the embodiments of the present invention.

[0074] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithm, characterized in that, include: The energy storage system is divided into categories using the first clustering method, and a first frequency control model for connecting multiple types of energy storage systems to the power grid is constructed based on the classification results. The frequency response process of the power grid after being disturbed is analyzed, a fuzzy controller is designed based on the first frequency control model, the weight factors of different control methods are determined, and the control model of the first energy storage system is obtained. Based on the first energy storage system control model, a second energy storage system control model is constructed by introducing state observation based on neighbor frequency deviation, thereby realizing two-layer frequency regulation control of the energy storage system.

2. The two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithm as described in claim 1, characterized in that, The analysis of the frequency response process of the power grid after being disturbed includes: Based on the dynamic response curve of the power grid frequency after being disturbed, the frequency fluctuation process is divided into the first fluctuation stage, the second fluctuation stage and the third fluctuation stage. For the first fluctuation phase, a control method combining virtual inertial control and droop control is adopted; For the second fluctuation phase, a control method combining virtual negative inertia control and droop control is adopted; For the third fluctuation stage, a droop control method is adopted.

3. The two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithm as described in claim 2, characterized in that, The weighting factors for determining different control methods include: Construct a fuzzy controller; The universe of discourse and fuzzy subsets are set for the input and output variables of the fuzzy controller, respectively, and the membership relationship of each fuzzy subset is defined by the first membership function; Based on the control requirements at different stages of frequency fluctuation, a fuzzy control rule table is established between input and output variables, and the weight factors for different control methods are obtained according to the fuzzy control rule table.

4. The two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithm as described in claim 3, characterized in that, The acquisition of the first energy storage system control model includes: Based on the weighting factors of the different control methods, the power commands generated by virtual droop control, virtual inertial control and virtual negative inertial control are weighted and fused to generate a primary control active power command. Based on the primary control active power command, a control model for the first energy storage system for primary frequency regulation is formed.

5. The two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithm as described in claim 4, characterized in that, The construction of the control model for the second energy storage system includes: Each energy storage unit in the system is defined as an agent. The integral of the frequency deviation of each agent's grid connection point is used as the consistency state variable. A nonlinear state predictor based on the neighbor frequency deviation is designed to predict the changing trend of each agent's state. A consensus protocol is constructed that integrates state difference measurement and nonlinear state prediction. The consensus protocol dynamically calculates the secondary frequency correction based on the state of each agent, the state of neighboring agents, and the state prediction. Based on the aforementioned consensus protocol, a distributed control model is established with the goal of eliminating frequency steady-state error and achieving frequency synchronization, serving as the control model for the second energy storage system.

6. The two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithm as described in claim 1, characterized in that, The process of dividing the energy storage system using the first clustering method includes: Electrochemical energy storage is used as an energy-type energy storage solution to respond to system frequency power fluctuations. Supercapacitor energy storage and flywheel energy storage are used as power-type energy storage to jointly respond to high-frequency power fluctuations in the system.

7. The two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithm as described in claim 6, characterized in that, The first frequency control model for connecting multiple types of energy storage systems to the power grid includes: A first-order inertial element is used to describe the external response characteristics of electrochemical energy storage when participating in grid frequency regulation, and a dynamic model of the electrochemical energy storage unit is established. The equivalent circuit model of the supercapacitor energy storage unit is constructed using the classical equivalent model; The power response characteristics of the flywheel energy storage unit are described by a first-order inertial element, and a system model of the flywheel energy storage unit is established. A frequency response model for thermal power units is constructed by combining the transfer functions of the governor, reheat turbine, generator, and load. The models of electrochemical energy storage, supercapacitor energy storage, flywheel energy storage units, and thermal power units are integrated to generate the first frequency control model for connecting the various types of energy storage systems to the power grid.

8. A two-layer frequency regulation control system for multi-type energy storage systems based on a consensus algorithm, wherein the two-layer frequency regulation control method for multi-type energy storage systems based on a consensus algorithm as described in any one of claims 1 to 7 is characterized in that, include: The frequency control model construction module is used to divide the energy storage system using the first clustering method and construct the first frequency control model for connecting multiple types of energy storage systems to the power grid based on the division results. The primary control model construction module is used to analyze the frequency response process of the power grid after disturbance, design a fuzzy controller based on the first frequency control model, determine the weight factors of different control methods, and obtain the first energy storage system control model. The secondary control model construction module is used to construct a second energy storage system control model based on the first energy storage system control model and introduce state observation based on neighbor frequency deviation, so as to realize the two-layer frequency regulation control of the energy storage system.

9. An electronic device comprising a memory and a processor, characterized in that: The memory is used to store computer-executable instructions, and when the processor executes the computer-executable instructions, it implements the steps of the two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithms as described in any one of claims 1 to 7.

10. A computer-readable storage medium having computer-executable instructions stored thereon, characterized in that: When the computer-executable instructions are executed by the processor, they implement the steps of the two-layer frequency regulation control method for multi-type energy storage systems based on consensus algorithms as described in any one of claims 1 to 7.

Images