A method and system for user-side small household energy storage to participate in system frequency regulation
By adopting the two-layer model predictive control strategy of game theory in the power system, combining non-cooperative game and cooperative game models, the frequency response of household energy storage is optimized, and the problem of missing control framework for household energy storage participating in frequency adjustment is solved, and the problem of improving the The system's frequency modulation capability and frequency stability are achieved.
Patent Information
- Application Number
- CN202410552296.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-07
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2044-05-07
AI Technical Summary
In the prior art, household energy storage participates in the frequency regulation of the power system lacks a perfect control framework and frequency regulation strategy, resulting in insufficient frequency regulation capabilities of the system and unstable frequency.
A two-layer model prediction control strategy based on game theory is proposed, combining non-cooperative games and cooperative game models to build an upper and lower control framework, and using the frequency response characteristics of thermal power units and household energy storage, optimize the frequency regulation process and ensure the stability of the system frequency sex.
It improves the system frequency regulation capability, enhances frequency stability, takes into account the user needs of household energy storage and system control accuracy, and is suitable for power system scheduling with increased penetration rate of new energy.
Smart Images

Figure CN118487292B_ABST
Abstract
Description
Technical Field
[0001] The present invention belongs to the technical field of power system frequency control, and specifically relates to a method and system for user-side small household energy storage to participate in system frequency regulation. Background Art
[0002] Frequency control is of great significance to achieve system frequency stability and ensure power quality. my country's renewable energy power generation technology is developing rapidly. New energy generally has the characteristics of volatility and uncertainty. The increase in the penetration rate of renewable energy reduces the frequency regulation capacity of the system, bringing new challenges to the frequency regulation of the power system. At present, some scholars have conducted detailed research on the participation of renewable energy power generation in power system frequency regulation. In addition to strengthening the frequency regulation capability of the system on the power supply side, the use of flexible loads that can be adjusted on the user side to participate in the system automatic generation control (automatic generation control, AGC) has opened up a new idea for improving the frequency regulation capability of the power system.
[0003] Under the influence of relevant factors, the configuration demand for small household energy storage is gradually increasing, and it is expected to become an important part of user-side energy storage in the future. Household energy storage usually refers to energy storage devices with a charging and discharging power of less than 5kW and a capacity of less than 13.5kW·h. At present, major companies at home and abroad have developed application products for household energy storage. The energy storage converter can be used to realize the power supply of energy storage equipment to the power grid. In theory, the reasonable guidance of household energy storage to participate in system frequency regulation can effectively improve the frequency regulation capability of the system. However, the following issues should be noted when household energy storage participates in frequency regulation: 1) Household energy storage is a distributed structure in space, with small single-unit capacity and huge number, and a suitable control framework is required to manage household energy storage; 2) The essential attribute of household energy storage is electricity load. When controlling household energy storage to participate in system frequency regulation, the user's usage needs should be guaranteed first; 3) In the frequency regulation process, household energy storage, as a user-side resource, does not need to participate in the system frequency regulation throughout the process. However, there is currently little research on household energy storage participating in system frequency regulation, and there is a lack of perfect control framework research and corresponding frequency regulation strategy research. Summary of the invention
[0004] In view of the shortcomings of the prior art, the present invention proposes a method and system for user-side small household energy storage to participate in system frequency regulation. Based on game theory, a two-layer model predictive control (MPC) strategy for household energy storage to participate in power system frequency regulation is proposed, which fully considers user needs, improves the system frequency regulation capability, and enhances frequency stability.
[0005] To achieve the above object, the present invention provides the following solutions:
[0006] A method for user-side small household energy storage to participate in system frequency regulation, comprising the following steps:
[0007] Taking into account the load characteristics and frequency response characteristics of household energy storage, a two-layer control framework for household energy storage to participate in frequency regulation is constructed, wherein the two-layer control framework includes: an upper layer and a lower layer; the upper layer includes: thermal power units, upper dispatching centers, and intermediate agents, and the lower layer includes: intermediate agents and all household energy storage connected to the grid;
[0008] Based on the two-layer control framework, a two-layer game model prediction control strategy for household energy storage participating in frequency regulation is constructed, and the two-layer game model prediction control strategy includes: capacity reporting method, thermal power unit frequency response model, household energy storage frequency response model and grid connection and off-grid mechanism;
[0009] Based on the thermal power unit frequency response model and the household energy storage frequency response model, an upper system equation group is constructed, and according to the upper system equation group, combined with the proposed improved non-cooperative game, a non-cooperative game model prediction control strategy is constructed;
[0010] Based on the household energy storage SOC model, a lower-level system equation group is constructed, and according to the lower-level system equation group, combined with the proposed cooperative game that introduces the concept of user level, a cooperative game model prediction control strategy is constructed;
[0011] In the upper-level control, the upper-level dispatching center uses the non-cooperative game model predictive control strategy to solve the output reference values of the thermal power units and household energy storage clusters in the control time domain online according to the operating status of the thermal power units at the previous moment and the frequency regulation capacity reported by the intermediate agent, and outputs the solution results as control instructions;
[0012] In the lower-level control, the intermediate agent uses the proposed cooperative game model predictive control strategy to solve the frequency regulation task of each household energy storage, and predicts the SOC value at the next moment according to the power response of the household energy storage;
[0013] Based on the control instruction and the SOC value at the next moment, a reference power signal of each frequency modulation unit in the system at the next moment is obtained; after receiving the reference power signal, each frequency modulation unit adjusts the operating state of the thermal power unit to achieve the optimization of system frequency stability.
[0014] Preferably, the non-cooperative game model predictive control strategy includes: an upper-level prediction model, an upper-level objective function and an upper-level constraint condition;
[0015] The upper system equations are:
[0016]
[0017] Where ΔP h is the output change of the thermal power unit, ΔP ref.his the change in the reference value of the thermal power unit output, G gov (s) is the speed governor model of thermal power unit, G tur (s) is the turbine model of the thermal power unit, R is the primary frequency modulation coefficient of the system, Δf is the system frequency deviation, H is the inertia time constant, ΔP RESs is the output change of the household energy storage cluster, is the transfer function of the household energy storage cluster, ΔP ref.RESs is the change in the output reference value of the household energy storage cluster, H is the load damping constant; ΔP load is the system load fluctuation, ΔP new The power fluctuation of new energy in the system;
[0018] The upper prediction model established according to the upper system equations is:
[0019]
[0020] In the formula, A, B, E, and C are the state matrix, control input matrix, interference input matrix, and control output matrix of the system respectively; x(t) is the system state variable, u(t) is the system input variable, L(t) is the system disturbance variable, and y(t) is the system output variable;
[0021] The upper objective function is:
[0022]
[0023] Where F1 is the upper objective function, M and N are the weight coefficients of the upper objective function; μ is called the game factor, and its size is determined by Δf, A ACE Provides automatic power generation control signal for the system;
[0024] The upper-level constraints include: upper and lower limits of thermal power unit output, ramp rate constraints, and upper and lower limits of household energy storage capacity for frequency regulation.
[0025] Preferably, the cooperative game model predictive control strategy includes: a lower-level prediction model, a lower-level objective function, and lower-level constraints;
[0026] The lower layer prediction model is:
[0027]
[0028] Where, T s is the discrete time constant, P RES.i , i∈1,2,…,n is the frequency regulation task undertaken by the i-th household energy storage; SOC i , E bess.i, i∈1,2,…,n are the SOC value of the i-th household energy storage and the capacity of the household energy storage battery respectively;
[0029] The lower layer objective function is:
[0030]
[0031] Where F2 is the lower layer objective function, P ref.RESi is the frequency regulation task undertaken by the i-th household energy storage, Y i , i∈1,2,…,n is the user level of the i-th household energy storage, U and V are the weight coefficients of the lower layer objective function;
[0032] The lower-level constraints include: the output of all household energy storage is equal to the power reference value constraint, and the upper and lower limit constraints of the frequency regulation of each household energy storage.
[0033] Preferably, the method in which each frequency modulation unit adjusts the operating state of the thermal power unit after receiving the reference power signal to achieve the most stable and optimized system frequency includes:
[0034] S1: Building upper and lower layers of game model prediction controllers, wherein the game model prediction controller is provided with an objective function, a prediction model and constraints;
[0035] S2: Based on each game model prediction controller, the operating status of each frequency modulation unit at the current moment is obtained, the operating status is calculated through the prediction model, the ACE signal at the current moment is generated, and the objective function is solved through the Gurobi solver according to the ACE signal;
[0036] S3: outputting the first set of parameters of the solution result as the real-time control instruction, and updating the reference power signal of each frequency modulation unit in the system at the next moment based on the real-time control instruction;
[0037] S4: After receiving the new reference power signal, each frequency modulation unit in the system takes action to change the operating state of the system and returns to S2.
[0038] The present invention also provides a system for user-side small household energy storage to participate in system frequency regulation, comprising: a first building module, a second building module, a third building module, a fourth building module, an upper control module, a lower control module and an adjustment module;
[0039] The first construction module is used to comprehensively consider the load characteristics and frequency response characteristics of household energy storage, and construct a two-layer control framework for household energy storage to participate in frequency regulation, wherein the two-layer control framework includes: an upper layer and a lower layer; the upper layer includes: thermal power units, upper dispatching centers, and intermediate agents, and the lower layer includes: intermediate agents and all grid-connected household energy storage;
[0040] The second construction module is used to construct a double-layer game model prediction control strategy for household energy storage participating in frequency regulation based on the double-layer control framework, and the double-layer game model prediction control strategy includes: capacity reporting method, thermal power unit frequency response model, household energy storage frequency response model and grid connection and off-grid mechanism;
[0041] The third building module is used for constructing an upper system equation group based on the thermal power unit frequency response model and the household energy storage frequency response model, and constructing a non-cooperative game model prediction control strategy based on the upper system equation group and the proposed improved non-cooperative game;
[0042] The fourth building module is used to construct a lower-level system equation group based on the household energy storage SOC model, and to construct a cooperative game model prediction control strategy based on the lower-level system equation group and in combination with the cooperative game that introduces the concept of user level;
[0043] The upper control module is used for the upper dispatch center to solve the output reference values of the thermal power units and the household energy storage cluster in the control time domain online according to the operating status of the thermal power units at the previous moment and the frequency regulation capacity reported by the intermediate agent by using the non-cooperative game model predictive control strategy in the upper control, and output the solution results as control instructions;
[0044] The lower-level control module is used for the intermediate agent to solve the frequency regulation task of each household energy storage by using the proposed cooperative game model prediction control strategy in the lower-level control, and predict the SOC value at the next moment according to the power of the household energy storage response;
[0045] The adjustment module is used to obtain a reference power signal of each frequency modulation unit in the system at the next moment based on the control instruction and the SOC value at the next moment; after receiving the reference power signal, each frequency modulation unit adjusts the operating state of the thermal power unit to achieve the optimization of system frequency stability.
[0046] Preferably, the non-cooperative game model predictive control strategy includes: an upper-level prediction model, an upper-level objective function and an upper-level constraint condition;
[0047] The upper system equations are:
[0048]
[0049] In the formula, ΔP h is the output change of the thermal power unit, ΔP ref.h is the change in the reference value of the thermal power unit output, G gov (s) is the speed governor model of thermal power unit, G tur(s) is the turbine model of the thermal power unit, R is the primary frequency modulation coefficient of the system, Δf is the system frequency deviation, H is the inertia time constant, ΔP RESs is the output change of the household energy storage cluster, is the transfer function of the household energy storage cluster, ΔP ref.RESs is the change in the output reference value of the household energy storage cluster, H is the load damping constant; ΔP load is the system load fluctuation, ΔP new The power fluctuation of new energy in the system;
[0050] The upper prediction model established according to the upper system equations is:
[0051]
[0052] In the formula, A, B, E, and C are the state matrix, control input matrix, interference input matrix, and control output matrix of the system respectively; x(t) is the system state variable, u(t) is the system input variable, L(t) is the system disturbance variable, and y(t) is the system output variable;
[0053] The upper objective function is:
[0054]
[0055] Where F1 is the upper objective function, M and N are the weight coefficients of the upper objective function; μ is called the game factor, and its size is determined by Δf, A ACE Provides automatic power generation control signal for the system;
[0056] The upper-level constraints include: upper and lower limits of thermal power unit output, ramp rate constraints, and upper and lower limits of household energy storage capacity for frequency regulation.
[0057] Preferably, the cooperative game model predictive control strategy includes: a lower-level prediction model, a lower-level objective function, and lower-level constraints;
[0058] The lower layer prediction model is:
[0059]
[0060] Where, T s is the discrete time constant, P RES.i , i∈1,2,…,n is the frequency regulation task undertaken by the i-th household energy storage; SOC i , E bess.i , i∈1,2,…,n are the SOC value of the i-th household energy storage and the capacity of the household energy storage battery respectively;
[0061] The lower layer objective function is:
[0062]
[0063] Where F2 is the lower layer objective function, P ref.RESi is the frequency regulation task undertaken by the i-th household energy storage, Y i , i∈1,2,…,n is the user level of the i-th household energy storage, U and V are the weight coefficients of the lower layer objective function;
[0064] The lower-level constraints include: the output of all household energy storage is equal to the power reference value constraint, and the upper and lower limit constraints of the frequency regulation of each household energy storage.
[0065] Preferably, the adjustment module includes: a construction unit, a solution unit, an update unit and an adjustment unit;
[0066] The building unit is used to build the upper and lower layers of the game model prediction controller, wherein the game model prediction controller is provided with an objective function, a prediction model and constraint conditions;
[0067] The solving unit is used to predict the controller based on each game model, obtain the operating status of each frequency modulation unit at the current moment, calculate the operating status through the prediction model, generate the ACE signal at the current moment, and solve the objective function through the Gurobi solver according to the ACE signal;
[0068] The updating unit is used to output the first set of parameters of the solution result as the real-time control instruction, and based on the real-time control instruction, update the reference power signal of each frequency modulation unit in the system at the next moment;
[0069] The regulating unit is used for each frequency modulation unit in the system to take action after receiving a new reference power signal, change the operating state of the system, and return to the solving unit.
[0070] Compared with the prior art, the present invention has the following beneficial effects:
[0071] In view of the impact of the increasing penetration rate of new energy on the frequency regulation capability of the power system, the present invention proposes a strategy for household energy storage to participate in system frequency regulation based on double-layer model predictive control. The technical effects that can be achieved are as follows:
[0072] 1) Controlling household energy storage to participate in system frequency regulation, although it reduces the charging rate of household energy storage to a certain extent, it enhances the frequency regulation capability of the system and improves the system frequency stability.
[0073] 2) The improved non-cooperative game model predictive control strategy proposed in the upper-level control of the present invention takes into account both the frequency regulation effect and the user attributes of household energy storage, and the comprehensive frequency regulation effect is better than the traditional MPC structure. In addition, this strategy is very suitable for the situation where user-side flexible loads participate in frequency regulation, and can also be used in research such as power system scheduling and planning.
[0074] 3) The cooperative game model predictive control strategy containing the user level concept proposed in the lower-level control of the present invention not only ensures the overall charging rate of household energy storage, but also takes into account the electricity demand of different household energy storage.
[0075] 4) The present invention adopts the MPC control structure, which can effectively deal with the constraints involved in the process of household energy storage participating in the system frequency regulation compared with other control methods, and ensure the system control accuracy. BRIEF DESCRIPTION OF THE DRAWINGS
[0076] In order to more clearly illustrate the technical solution of the present invention, the following briefly introduces the drawings required for use in the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative labor.
[0077] Figure 1 It is a structural diagram of the electric vehicle participating in frequency modulation control constructed by the present invention;
[0078] Figure 2 It is a solution flow chart of the present invention;
[0079] Figure 3 is the system frequency fluctuation result obtained by simulation of the present invention;
[0080] Figure 4 It is a flow chart of a method for user-side small household energy storage to participate in system frequency regulation according to the present invention. DETAILED DESCRIPTION
[0081] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0082] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the present invention is further described in detail below with reference to the accompanying drawings and specific embodiments.
[0083] Embodiment 1
[0084] like Figure 4As shown, the present invention provides a method for user-side small household energy storage to participate in system frequency regulation, comprising the following steps:
[0085] Taking into account the load characteristics and frequency response characteristics of household energy storage, a two-layer control framework for household energy storage to participate in frequency regulation is constructed. The two-layer control framework includes: an upper layer and a lower layer; the upper layer includes: thermal power units, upper dispatching centers, and intermediate agents; the lower layer includes: intermediate agents and all household energy storage connected to the grid;
[0086] Based on the two-layer control framework, a two-layer game model predictive control strategy for household energy storage participating in frequency regulation is constructed. The two-layer game model predictive control strategy includes: capacity reporting method, thermal power unit frequency response model, household energy storage frequency response model and grid-connection and off-grid mechanism;
[0087] Based on the frequency response model of thermal power units and household energy storage frequency response model, the upper system equations are constructed. Based on the upper system equations and the proposed improved non-cooperative game, a non-cooperative game model predictive control strategy is constructed.
[0088] Based on the household energy storage SOC model, the lower system equations are constructed, and according to the lower system equations, combined with the proposed cooperative game that introduces the concept of user level, a cooperative game model predictive control strategy is constructed;
[0089] In the upper-level control, the upper-level dispatching center uses the non-cooperative game model predictive control strategy to solve the output reference values of the thermal power units and household energy storage clusters in the control time domain online according to the operating status of the thermal power units at the previous moment and the frequency regulation capacity reported by the intermediate agent, and outputs the solution results as control instructions;
[0090] In the lower-level control, the intermediate agent uses the proposed cooperative game model predictive control strategy to solve the frequency regulation task of each household energy storage and predicts the SOC value at the next moment based on the power response of the household energy storage;
[0091] Based on the control instruction and the SOC value at the next moment, a reference power signal of each frequency modulation unit in the system at the next moment is obtained; after receiving the reference power signal, each frequency modulation unit adjusts the operating state of the thermal power unit to achieve the optimization of system frequency stability.
[0092] In this embodiment, the process of constructing a two-layer control framework for household energy storage to participate in frequency regulation includes: an upper layer and a lower layer; wherein the upper layer includes thermal power units, an upper dispatching center, and upper intermediate agents, and the lower layer includes lower intermediate agents and all grid-connected household energy storage;
[0093] The tasks of the middle agents can be summarized as “controlling the flow of information from top to bottom”;
[0094] "Upload" means that before the frequency response, the capacity of household energy storage in the jurisdiction that can be used for frequency regulation is counted according to the status of household energy storage in the jurisdiction, and reported to the superior dispatching center through the information exchange channel;
[0095] "Issuance" means that after receiving the AGC signal from the dispatch center through the upper information exchange channel, the intermediate agent calculates the charging power or discharging power of household energy storage within the jurisdiction according to relevant constraints, and issues power control instructions through the lower information exchange channel;
[0096] "Central control" means that the intermediary agents should regularly monitor the various states of household energy storage within their jurisdiction to ensure better and more timely completion of "upload" and "download" tasks;
[0097] The superior dispatching center is responsible for calculating the output reference values of the generator sets and household energy storage according to the current frequency fluctuations, unit status and the capacity reported by the intermediate agent, and communicating them through the upper-level information interaction channel.
[0098] In this embodiment, considering the needs of users, the present invention divides the status of grid-connected household energy storage in detail to better complete capacity reporting:
[0099] State 1: Household energy storage SOC is less than SOC min , SOC min In order to meet the minimum SOC value required by users, household energy storage does not participate in frequency regulation in this state;
[0100] State 2: Household energy storage SOC is between SOC min To SOC th Between, SOC th is the threshold close to full charge. In this state, the charging power of household energy storage is P e , P e Slightly less than the maximum charging power P max , household energy storage participates in the frequency regulation of the power grid by increasing or decreasing the charging power;
[0101] State 3: Household energy storage SOC is greater than SOC th In this state, the charging mode of household energy storage is the same as that in state 2. In this state, household energy storage can participate in frequency regulation by increasing or decreasing charging power, as well as by discharging in reverse to the grid.
[0102] Frequency response model of household energy storage cluster:
[0103]
[0104] Where ΔP RESs is the output change of household energy storage cluster, G RES (s) is the transfer function of the household energy storage cluster, ΔPref.RESs is the change in the output reference value of the household energy storage cluster; K, T RESs It is the coefficient of the transfer function of the household energy storage cluster.
[0105] Household energy storage charging and discharging model:
[0106] The SOC value will change when household energy storage participates in system frequency regulation. The power loss during the charging and discharging process of household energy storage is borne by the household energy storage. Its SOC model is as follows:
[0107] SOC=SOC0+ΔSOC
[0108] Where SOC0 is the initial value of the household energy storage SOC, and ΔSOC is the offset of the household energy storage SOC during the frequency modulation process.
[0109] Household energy storage grid-connected and off-grid mechanisms:
[0110] The present invention further regulates the grid-connection and off-grid connection of household energy storage: in the strategy proposed in the present invention, the intermediate agent counts the capacity of all household energy storage that can be used for frequency regulation every 30 minutes for capacity reporting. When the household energy storage that is not connected to the grid is connected to the grid for charging, it will not be connected to the grid immediately, but at the next capacity statistical moment of the intermediate agent; when the grid-connected user energy storage is off-grid, it will not be off-grid immediately, but at the next capacity statistical moment of the intermediate agent.
[0111] In this embodiment, the non-cooperative game model predictive control strategy is: frequency fluctuation is introduced as a third party in the game process between household energy storage and thermal power units, and the output of household energy storage and thermal power units is determined by frequency fluctuation. When the frequency fluctuation is small, household energy storage does not participate in frequency modulation, but participates in frequency modulation when the frequency fluctuation is large.
[0112] Including: upper-level prediction model, upper-level objective function and upper-level constraint conditions;
[0113] The upper system equations are:
[0114]
[0115] Where ΔP h is the output change of the thermal power unit, ΔP ref.h is the change in the reference value of the thermal power unit output, G gov (s) is the speed governor model of thermal power unit, G tur (s) is the turbine model of the thermal power unit, R is the primary frequency modulation coefficient of the system, Δf is the system frequency deviation, H is the inertia time constant, ΔP RESs is the output change of the household energy storage cluster, is the transfer function of the household energy storage cluster, ΔP ref.RESsis the change in the output reference value of the household energy storage cluster, H is the load damping constant; ΔP load is the system load fluctuation, ΔP new The power fluctuation of new energy in the system;
[0116] The upper prediction model established according to the upper system equations is:
[0117]
[0118] In the formula, A, B, E, and C are the state matrix, control input matrix, interference input matrix, and control output matrix of the system respectively; x(t) is the system state variable, u(t) is the system input variable, L(t) is the system disturbance variable, and y(t) is the system output variable.
[0119] The upper objective function is:
[0120]
[0121] Where F1 is the upper objective function, M and N are the weight coefficients of the upper objective function; μ is called the game factor, and its size is determined by Δf, A ACE Provides automatic power generation control signal for the system;
[0122] The upper-level constraints include: upper and lower limits of thermal power unit output, ramp rate constraints, and upper and lower limits of household energy storage capacity for frequency regulation.
[0123] In this embodiment, the cooperative game model predictive control strategy is: in the lower-level control, the SOC model of household energy storage is incorporated into the prediction model, and the cooperative game model of the intermediate agent and household energy storage is combined with MPC to construct the lower-level cooperative game model predictive control strategy;
[0124] The change of household energy storage SOC value is predicted within the interval of the intermediate agent's capacity statistics, and the capacity statistics every 30 minutes are used as a feedback link to correct the household energy storage SOC value.
[0125] Including: lower-level prediction model, lower-level objective function and lower-level constraint conditions;
[0126] The lower layer prediction model is:
[0127]
[0128] Where, T s is the discrete time constant, P RES.i , i∈1,2,…,n is the frequency regulation task undertaken by the i-th household energy storage; SOC i , E bess.i, i∈1,2,…,n are the SOC value of the i-th household energy storage and the capacity of the household energy storage battery respectively;
[0129] The lower layer objective function is:
[0130]
[0131] Where F2 is the lower layer objective function, P ref.RESi is the frequency regulation task undertaken by the i-th household energy storage, Y i , i∈1,2,…,n is the user level of the i-th household energy storage, U and V are the weight coefficients of the lower layer objective function;
[0132] The lower-level constraints include: the output of all household energy storage is equal to the power reference value constraint, and the upper and lower limit constraints of the frequency regulation of each household energy storage.
[0133] In this embodiment, the method in which each frequency modulation unit adjusts the operating state of the thermal power unit after receiving the reference power signal to achieve the most stable and optimized system frequency includes:
[0134] S1: Building upper and lower layers of game model prediction controllers, wherein the game model prediction controller is provided with an objective function, a prediction model and constraints;
[0135] S2: Based on each game model prediction controller, the operating status of each frequency modulation unit at the current moment is obtained, the operating status is calculated through the prediction model, the ACE signal at the current moment is generated, and the objective function is solved through the Gurobi solver according to the ACE signal;
[0136] S3: outputting the first set of parameters of the solution result as the real-time control instruction, and updating the reference power signal of each frequency modulation unit in the system at the next moment based on the real-time control instruction;
[0137] S4: After receiving the new reference power signal, each frequency modulation unit in the system takes action to change the operating state of the system and returns to S2.
[0138] Embodiment 2
[0139] This embodiment provides the following technical solution: a method for electric vehicles to participate in system frequency control, comprising:
[0140] Construct an interconnected system with different areas, including traditional units, wind turbines and energy storage power stations;
[0141] Firstly, a two-layer control framework and capacity reporting method for household energy storage participating in system frequency regulation are proposed, and based on the proposed control framework, an upper and lower layer model predictive control strategy is proposed.
[0142] In the upper-level control, a predictive control strategy based on an improved non-cooperative game model is adopted. Thermal power units and household energy storage clusters are regarded as the two parties in the game, and the output of the two parties is determined by the third party, frequency fluctuation. The load properties of household energy storage are fully considered, and household energy storage is controlled to participate in frequency regulation when the frequency fluctuation is large.
[0143] At the lower level, a cooperative game model predictive control strategy is adopted to allocate power among household energy storages. The concept of user level is added to the objective function, and the two indicators of SOC value and user level are used as reference standards for power allocation, fully considering the differences in demand among household energy storages.
[0144] (1) Control framework for household energy storage to participate in frequency regulation
[0145] If the grid dispatch center controls each household energy storage separately, it will be difficult and inefficient. Referring to the control framework of electric vehicles participating in system frequency regulation, the present invention proposes a two-layer control framework consisting of an intermediate agent, a superior dispatch center and household energy storage. Figure 1 As shown, each household energy storage is managed by an intermediate agent, which reduces the difficulty of control. Through the aggregation of each household energy storage by the intermediate agent, the power used for frequency regulation reaches a volume that can affect the active balance of the system, so that the household energy storage can effectively participate in the system frequency regulation.
[0146] After the introduction of the intermediate agent, the way household energy storage participates in system frequency regulation is similar to that of the generator set on the power supply side, which is to reduce or increase the charging power of household energy storage by receiving AGC signals. The upper layer of the proposed control structure consists of the dispatch center, the generator set and the intermediate agent; the lower layer consists of the intermediate agent and all household energy storage in the jurisdiction. Among them, the tasks of the intermediate agent can be summarized as "upload and issue central control". "Upload" means that before the frequency response, the capacity of household energy storage in the jurisdiction that can be used for frequency regulation is counted according to the status of household energy storage in the jurisdiction, and reported to the upper dispatch center through the information interaction channel. "Issuance" means that after the intermediate agent receives the AGC signal from the dispatch center through the upper information interaction channel, it calculates the charging power or discharge power of household energy storage in the jurisdiction according to relevant constraints, and issues power control instructions through the lower information interaction channel. "Central control" means that the intermediate agent must regularly monitor the various states of household energy storage in the jurisdiction to ensure better and more timely completion of the "upload" and "issue" tasks. The superior dispatching center is responsible for calculating the output reference values of the generator sets and household energy storage according to the current frequency fluctuations, unit status and the capacity reported by the intermediate agent, and communicating them through the upper-level information interaction channel.
[0147] The frequency response model of thermal power units is as follows:
[0148]
[0149] Where ΔP h is the output change of the thermal power unit, ΔP ref.h G is the change of the reference value of the thermal power unit output; gov (s) is the speed governor model of thermal power unit, T g is the speed governor coefficient of the thermal power unit; G tur (s) is the turbine model of thermal power unit, T CH is the turbine coefficient of thermal power unit.
[0150] Household energy storage capacity reporting method:
[0151] The intermediate agent needs to report the capacity of household energy storage that can be used for frequency modulation to the superior dispatching center to ensure the accuracy of control. For this purpose, it is necessary to study a suitable capacity reporting method. The working mode of household energy storage is relatively simple, that is, it discharges when the user uses it, and charges when the user is not using it. When the household energy storage is in use, it is off-grid. At this time, the household energy storage will not participate in the system frequency modulation, and the intermediate agent will no longer control this household energy storage. Only when the user is not in use will the intermediate agent control the household energy storage to make it participate in frequency modulation. Taking into account user needs, the present invention makes a detailed division of the status of grid-connected household energy storage:
[0152] State 1: Household energy storage SOC is less than SOC min , SOC min In order to meet the minimum SOC value required by users, household energy storage does not participate in frequency regulation in this state.
[0153] State 2: Household energy storage SOC is between SOC min To SOC th Between, SOC th is the threshold close to full charge. In this state, the charging power of household energy storage is P e , P e Slightly less than the maximum charging power P max , household energy storage participates in the frequency regulation of the power grid by increasing or decreasing the charging power.
[0154] State 3: Household energy storage SOC is greater than SOC th In this state, the charging mode of household energy storage is the same as that in state 2. In this state, household energy storage can participate in frequency regulation by increasing or decreasing the charging power, as well as by discharging in reverse to the grid.
[0155] Frequency response model of household energy storage cluster:
[0156] The charging and discharging characteristics of household energy storage are similar to those of power system energy storage devices. Therefore, the present invention uses a first-order inertia link to represent the control and communication delay effects of household energy storage clusters. The equivalent frequency response model is:
[0157]
[0158] Where ΔP RESs is the output change of household energy storage cluster, G RES (s) is the transfer function of the household energy storage cluster, ΔP ref.RESs is the change in the output reference value of the household energy storage cluster; K, T RESs It is the coefficient of the transfer function of the household energy storage cluster.
[0159] Household energy storage charging and discharging model:
[0160] The SOC value will change when household energy storage participates in system frequency regulation. The power loss during the charging and discharging process of household energy storage is borne by the household energy storage. Its SOC model is as follows:
[0161] SOC=SOC0+ΔSOC
[0162] Where SOC0 is the initial value of the household energy storage SOC, and ΔSOC is the offset of the household energy storage SOC during the frequency modulation process.
[0163] The expression of ΔSOC when charging household energy storage is:
[0164]
[0165] Where P in is the household energy storage charging power; η in Charging efficiency for household energy storage; E bess Battery capacity for household energy storage.
[0166] The expression of ΔSOC when household energy storage is discharged is:
[0167]
[0168] Where P out is the household energy storage discharge power; η out Discharge efficiency of household energy storage.
[0169] Household energy storage grid-connected and off-grid mechanisms:
[0170] Household energy storage will cause SOC changes during the charging process. The accurate capacity reporting of the intermediate agent to the superior dispatch center can make the control more accurate. However, the number of household energy storage in the jurisdiction is huge. The capacity statistics of household energy storage clusters at each sampling time will increase the workload of the intermediate agents. In addition, there are many uncertain factors such as communication delays in the capacity statistics process, which may cause the intermediate agents to fail to report the capacity in time, thus deteriorating the control effect.
[0171] In order to make the supervision and control of household energy storage by the intermediate agent more accurate, the present invention further regulates the grid connection and off-grid connection of household energy storage: in the strategy proposed by the present invention, the intermediate agent counts the capacity of all household energy storage that can be used for frequency regulation every 30 minutes for capacity reporting. When the household energy storage that is not connected to the grid is connected to the grid for charging, it will not be connected to the grid immediately, but at the next capacity statistics time of the intermediate agent; when the grid-connected user energy storage is off-grid, it will not be off-grid immediately, but at the next capacity statistics time of the intermediate agent. This specification avoids the problem of inaccurate control caused by the grid connection and off-grid connection of household energy storage during the sampling interval.
[0172] (2) Upper-level MPC based on non-cooperative game theory
[0173] In the issue of household energy storage participating in the frequency regulation of the power system, there are many contradictions between the needs of users and the system for household energy storage. Compared with household energy storage, the power regulation range of thermal power units is wider, but the response speed is slower, and changing the output will cause losses to thermal power units; compared with thermal power units, household energy storage has a fast response speed, but a narrow power regulation range, and its essence is the power load, and its charging needs should be met first. Out of consideration for their own interests, both household energy storage and thermal power units hope that the other party can take on more frequency regulation tasks. Traditional MPC generally uses a fixed weight method to deal with such optimization problems. This method cannot take into account the demands of thermal power units and household energy storage. Moreover, when the frequency fluctuates within a small range, household energy storage does not need to participate in the frequency regulation task. When using the traditional weight method MPC control, the final decision cannot guarantee that household energy storage does not participate at this time, and thus cannot guarantee the user needs of household energy storage.
[0174] Non-cooperative game is a process in which each participant maximizes his or her own interests and thus promotes cooperation among the participants, thereby maximizing the collective interests. Combining non-cooperative game with MPC improves the speed of solving optimization problems, takes into account the demands of each participant, and obtains the optimal solution according to the constraints. However, unlike traditional non-cooperative game problems, the system frequency changes all the time during the system frequency regulation process, and the game environment of household energy storage and thermal power units is different each time. The ideal situation for household energy storage to participate in frequency regulation is not to participate in frequency regulation when the frequency fluctuation is small, and to participate in frequency regulation when the frequency fluctuation is large. In order to realize the working method described in the previous invention, the present invention proposes an improved non-cooperative game model predictive control strategy: frequency fluctuation is introduced as a third party in the game process between household energy storage and thermal power units, and the output of household energy storage and thermal power units is determined by frequency fluctuations.
[0175] (3) Lower-level MPC based on cooperative game theory
[0176] Cooperative game refers to a game played by some participants in the form of alliance and cooperation. After receiving the AGC instruction from the superior dispatching center, the intermediate agent should distribute power to the household energy storage in the jurisdiction in response to the frequency change. There is a cooperative game relationship between the intermediate agent and the household energy storage. It is very suitable to use the cooperative game model to solve the problem of power distribution between household energy storage.
[0177] The demands of various household energy storages are not the same. If the power distribution among household energy storages is performed only based on the SOC value, the demand differences among users will not be taken into account, which will eventually affect the use of users with higher demands. In order to meet the needs of users with high demands, the present invention introduces a user rating system. When allocating power, the intermediate agent not only refers to the SOC value, but also takes the user level of each household energy storage into the reference standard. On the premise that each household energy storage participates in frequency modulation, the charging needs of high-level users are met first.
[0178] As mentioned above, the intermediate agent of the strategy proposed by the present invention performs capacity statistics every 30 minutes, and introducing MPC into the control strategy has another advantage: the SOC change of household energy storage in the interval time can be predicted according to the prediction model, so as to obtain the capacity of household energy storage that can be used for frequency regulation at this time. In the interval time, the upper-level dispatching center uses the predicted frequency regulation capacity to make AGC power decisions, which can make the control effect better. Therefore, the present invention incorporates the SOC model of household energy storage into the prediction model in the lower-level control, combines the cooperative game model of the intermediate agent and household energy storage with MPC to construct a lower-level cooperative game model prediction control strategy, predicts the change of household energy storage SOC value in the interval time of the intermediate agent capacity statistics, and uses the capacity statistics every 30 minutes as a feedback link to correct the household energy storage SOC value.
[0179] Embodiment 3
[0180] This embodiment provides the following technical solutions:
[0181] Step one: Build a framework for household energy storage to participate in frequency regulation.
[0182] Step 2: Construct the upper-level improved non-cooperative game model predictive control strategy.
[0183] (1) Upper prediction model:
[0184]
[0185] In the formula, R is the primary frequency modulation coefficient of the system, Δf is the system frequency deviation, H is the inertia time constant, ΔP load is the system load fluctuation, ΔP new It is the power fluctuation of new energy in the system.
[0186] According to the system equations, the following system prediction model is established:
[0187]
[0188] Where A, B, E, and C are the state matrix, control input matrix, interference input matrix, and control output matrix of the system, respectively.
[0189] T s is the sampling time for household energy storage to participate in frequency regulation, expressed as T s The prediction model is discretized for intervals, and the prediction model after processing is:
[0190]
[0191] Where A_d, B_d, E_d, and C_d are the system state matrix, control input matrix, interference input matrix, and control output matrix after discretization, respectively.
[0192] (2) Upper layer objective function:
[0193] The interconnected system AGC requires that both the interconnected system frequency offset and the tie line power offset be minimized as much as possible. For a single system, the AGC only requires that the system frequency offset be minimized. ACE is A ACE , which is defined as:
[0194] A ACE =βΔf
[0195] The objective function expression of the upper control is:
[0196]
[0197] Where F1 is the upper objective function, M and N are the weight coefficients of the upper objective function; μ is called the game factor, and its size is determined by Δf, A ACE It is the automatic power generation control signal for the system.
[0198] When the frequency fluctuates greatly downward, the μ value is very small, and the household energy storage is controlled to reduce the charging power as much as possible to provide frequency increase service; when the frequency fluctuation is near 0, the μ value is large, and the household energy storage is controlled not to participate in the frequency regulation service; when the frequency fluctuates greatly upward, the μ value takes the maximum value, and the household energy storage is controlled to increase the charging power to provide frequency decrease service. Under the action of this objective function, the overall charging power of the household energy storage cluster increases as the frequency increases in the range of (-0.2Hz, 0.2Hz).
[0199] (3) Upper-level constraints:
[0200] Constraints of thermal power units:
[0201]
[0202] Where P h.max , P h.min They are the upper and lower limits of thermal power unit output respectively; P h is the power generation of thermal power units; P h.up , P h.down They are the upper and lower limits of the climbing rate of thermal power units respectively.
[0203] Constraints of household energy storage clusters:
[0204] P RESs.d ≤P RESs ≤P RESs.u
[0205] Where P RESs.u , P RESs.d are the power used by household energy storage clusters to provide frequency increase and decrease services respectively; P RESs Frequency regulation tasks undertaken by household energy storage.
[0206] Step three, construct the lower-level cooperative game model to predict the control strategy.
[0207] (1) Lower layer prediction model:
[0208] The discretized prediction model of the lower-level control is as follows:
[0209]
[0210] Where, T s is the discrete time constant, P RES.i (i∈1,2,…,n) is the frequency regulation task undertaken by the i-th household energy storage; SOC i , E bess.i (i∈1,2,…,n) are the SOC value of the i-th household energy storage and the capacity of the household energy storage battery respectively.
[0211] The state variable matrix and control output matrix of the lower control layer are both unit matrices of dimension i; the dimension of its control input matrix is also i, and the elements on the main diagonal are T s / E bess.i .
[0212] (2) Lower layer objective function:
[0213] After the introduction of user levels, the frequency regulation task undertaken by each household energy storage is determined by the user level and SOC value. The lower objective function is as follows:
[0214]
[0215] Where F2 is the lower layer objective function, P ref.RESi Y is the frequency regulation task undertaken by the i-th household energy storage;i (i∈1,2,…,n) is the user level of the i-th household energy storage; U and V are the weight coefficients of the lower layer objective function.
[0216] (3) Lower level constraints:
[0217] The household energy storage within the jurisdiction of the intermediate agent jointly bears the AGC instructions, so the equality constraint that the lower-level control needs to satisfy is:
[0218]
[0219] Household energy storage in state 1 does not participate in frequency regulation and has no constraints. Household energy storage in states 2 and 3 is subject to the constraints of its upward and downward regulation capacities, expressed as:
[0220] P d.i ≤P ref.RESi ≤P u.i
[0221] Where P u.i , P d.i (i∈1,2,…,n) are the upward and downward adjustment capacities of the i-th household energy storage.
[0222] Step 4: Solve the model.
[0223] The upper layer improved non-cooperative game MPC strategy and the lower layer cooperative game MPC strategy proposed in the present invention cooperate with each other to complete the household energy storage participating in the system frequency regulation. The specific solution process is as follows: Figure 2 shown.
[0224] In the upper-level control, the upper-level dispatch center uses the proposed non-cooperative game model predictive control strategy to solve the output reference values of the two in the control time domain online according to the operating status of the thermal power units at time k-1 and the frequency regulation capacity reported by the intermediate agent, and outputs the solution results as control instructions. In the lower-level control, the intermediate agent uses the proposed cooperative game model predictive control strategy to solve the frequency regulation task of each household energy storage, and predicts the SOC value at the next moment according to the power response of the household energy storage, which is used for capacity reporting work within the capacity statistics interval.
[0225] Embodiment 4
[0226] The present invention also provides a system for user-side small household energy storage to participate in system frequency regulation, comprising: a first building module, a second building module, a third building module, a fourth building module, an upper control module, a lower control module and an adjustment module;
[0227] The first construction module is used to comprehensively consider the load characteristics and frequency response characteristics of household energy storage, and construct a two-layer control framework for household energy storage to participate in frequency regulation, wherein the two-layer control framework includes: an upper layer and a lower layer; the upper layer includes: thermal power units, upper dispatching centers, and intermediate agents, and the lower layer includes: intermediate agents and all grid-connected household energy storage;
[0228] The second construction module is used to construct a double-layer game model prediction control strategy for household energy storage participating in frequency regulation based on the double-layer control framework, and the double-layer game model prediction control strategy includes: capacity reporting method, thermal power unit frequency response model, household energy storage frequency response model and grid connection and off-grid mechanism;
[0229] The third building module is used for constructing an upper system equation group based on the thermal power unit frequency response model and the household energy storage frequency response model, and constructing a non-cooperative game model prediction control strategy based on the upper system equation group and the proposed improved non-cooperative game;
[0230] The fourth construction module is used to construct a lower-level system equation group based on the household energy storage SOC model, and to construct a cooperative game model prediction control strategy based on the lower-level system equation group and the cooperative game that introduces the concept of user level;
[0231] The upper control module is used for the upper control, in which the upper dispatch center uses the non-cooperative game model predictive control strategy to solve the output reference values of the thermal power units and the household energy storage cluster in the control time domain online according to the operating status of the thermal power units at the previous moment and the frequency regulation capacity reported by the intermediate agent, and outputs the solution results as control instructions;
[0232] The lower-level control module is used for the intermediate agent to solve the frequency regulation task of each household energy storage by using the proposed cooperative game model prediction control strategy in the lower-level control, and predict the SOC value at the next moment according to the power of the household energy storage response;
[0233] The adjustment module is used to obtain a reference power signal of each frequency modulation unit in the system at the next moment based on the control instruction and the SOC value at the next moment; after receiving the reference power signal, each frequency modulation unit adjusts the operating state of the thermal power unit to achieve the optimization of system frequency stability.
[0234] In this embodiment, the non-cooperative game model predictive control strategy includes: an upper-level prediction model, an upper-level objective function, and an upper-level constraint condition;
[0235] The upper system equations are:
[0236]
[0237] In the formula, ΔP h is the output change of the thermal power unit, ΔP ref.h is the change in the reference value of the thermal power unit output, G gov (s) is the speed governor model of thermal power unit, G tur (s) is the turbine model of the thermal power unit, R is the primary frequency modulation coefficient of the system, Δf is the system frequency deviation, H is the inertia time constant, ΔP RESs is the output change of the household energy storage cluster, is the transfer function of the household energy storage cluster, ΔP ref.RESs is the change in the output reference value of the household energy storage cluster, H is the load damping constant; ΔP load is the system load fluctuation, ΔP new The power fluctuation of new energy in the system;
[0238] The upper prediction model established according to the upper system equations is:
[0239]
[0240] In the formula, A, B, E, and C are the state matrix, control input matrix, interference input matrix, and control output matrix of the system respectively; x(t) is the system state variable, u(t) is the system input variable, L(t) is the system disturbance variable, and y(t) is the system output variable.
[0241] The upper objective function is:
[0242]
[0243] Where F1 is the upper objective function, M and N are the weight coefficients of the upper objective function; μ is called the game factor, and its size is determined by Δf, A ACE It is the automatic power generation control signal for the system.
[0244] The upper-level constraints include: upper and lower limits of thermal power unit output, ramp rate constraints, and upper and lower limits of household energy storage capacity for frequency regulation.
[0245] In this embodiment, the cooperative game model predictive control strategy includes: a lower-level prediction model, a lower-level objective function, and a lower-level constraint condition;
[0246] The lower layer prediction model is:
[0247]
[0248] Where, T s is the discrete time constant, P RES.i , i∈1,2,…,n is the frequency regulation task undertaken by the i-th household energy storage; SOC i, E bess.i , i∈1,2,…,n are the SOC value of the i-th household energy storage and the capacity of the household energy storage battery respectively;
[0249] The lower layer objective function is:
[0250]
[0251] Where F2 is the lower layer objective function, P ref.RESi is the frequency regulation task undertaken by the i-th household energy storage, Y i , i∈1,2,…,n is the user level of the i-th household energy storage, U and V are the weight coefficients of the lower layer objective function;
[0252] The lower-level constraints include: the output of all household energy storage is equal to the power reference value constraint, and the upper and lower limit constraints of the frequency regulation of each household energy storage.
[0253] In this embodiment, the adjustment module includes: a construction unit, a solution unit, an update unit and an adjustment unit;
[0254] The building unit is used to build the upper and lower layers of the game model prediction controller, wherein the game model prediction controller is provided with an objective function, a prediction model and constraint conditions;
[0255] The solving unit is used to predict the controller based on each game model, obtain the operating status of each frequency modulation unit at the current moment, calculate the operating status through the prediction model, generate the ACE signal at the current moment, and solve the objective function through the Gurobi solver according to the ACE signal;
[0256] The updating unit is used to output the first set of parameters of the solution result as the real-time control instruction, and based on the real-time control instruction, update the reference power signal of each frequency modulation unit in the system at the next moment;
[0257] The adjustment unit is used for each frequency modulation unit in the system to take action after receiving a new reference power signal, change the operating state of the system, and return to the solution unit.
[0258] Embodiment 5
[0259] In order to verify the effect of the household energy storage participating in the system frequency regulation strategy proposed in the present invention, a single-area system model was built in MATLAB / Simulink for verification.
[0260] Due to the lack of real household energy storage data, the operating data of household energy storage in the present invention is randomly generated. The total number of household energy storages in this area is 4,000; the capacity data of household energy storages satisfies the uniform distribution U (10, 13.5); household energy storage is generally connected to the grid for charging when the battery is low after use by the user, so the present invention sets the SOC state of household energy storage when connected to the grid to satisfy the uniform distribution U (0.2, 0.4); the charging time of household energy storage connected to the grid once satisfies the normal distribution probability N (3, 0.5); the number of grid connections per day for each household energy storage satisfies the uniform distribution U (1, 4); the time periods for household energy storage to be connected to the grid are concentrated at around 9 am, around 2 pm and around 11 pm, and the data from the time period from 0 am to 3 am is used for simulation experiments. Figure 3 shown.
[0261] The system AGC sampling interval is set to 1 minute. The control time domain of the upper layer is 3 minutes, the control time domain of the lower layer is 1 minute, the prediction time domain of the upper and lower layers is 30 minutes, and the total simulation time is 4 hours. The new energy penetration rate of the system is 40%. The system parameters are converted to per-unit values based on 500MW as the benchmark value, and the detailed data of thermal power units and household energy storage are shown in Table 1. A set of data is randomly generated between [-0.04pu, 0.04pu] at intervals of 1 minute as the load fluctuation of the system.
[0262] Table 1
[0263]
[0264] Under the control of the strategy proposed in the present invention, the maximum value of the system frequency fluctuation is 0.1976Hz, and the minimum value is -0.1515Hz. The proposed strategy makes the frequency fluctuation at all times within the allowable safety range. The proposed strategy also solves the limitation of the traditional control method that it cannot take into account the interests of thermal power units and household energy storage when controlling household energy storage to participate in frequency modulation, and is more suitable for the situation where household energy storage participates in frequency modulation. Under the control of the proposed strategy, the overall SOC increase rate of household energy storage is 31.5%, which is lower than 42.14% when household energy storage does not participate in frequency modulation. However, it must be noted that controlling household energy storage to participate in frequency modulation will inevitably reduce the charging amount of household energy storage. However, the penetration rate of new energy in the system has increased recently. The participation of household energy storage in the system AGC has enhanced the frequency modulation ability of the system in the face of extreme situations, keeping the system frequency fluctuation within a safe range, which is more positive than the missing charging amount of household energy storage. After adding the concept of user level to the lower objective function, the charging rate of high-level household energy storage is 42.07%. The proposed strategy not only improves the system frequency regulation capability but also fully guarantees the overall and individual usage needs of household energy storage.
[0265] The embodiments described above are only descriptions of the preferred embodiments of the present invention and are not intended to limit the scope of the present invention. Without departing from the design spirit of the present invention, various modifications and improvements made to the technical solutions of the present invention by ordinary technicians in this field should all fall within the protection scope determined by the claims of the present invention.
Claims
1. A method for user-side small household energy storage to participate in system frequency regulation, characterized in that: The following steps are involved: Taking into account the load characteristics and frequency response characteristics of household energy storage, a two-layer control framework for household energy storage to participate in frequency regulation is constructed, wherein the two-layer control framework includes: an upper layer and a lower layer; the upper layer includes: thermal power units, upper dispatching centers, and intermediate agents, and the lower layer includes: intermediate agents and all household energy storage connected to the grid; Based on the two-layer control framework, a two-layer game model prediction control strategy for household energy storage participating in frequency regulation is constructed, and the two-layer game model prediction control strategy includes: capacity reporting method, thermal power unit frequency response model, household energy storage frequency response model and grid connection and off-grid mechanism; Based on the thermal power unit frequency response model and the household energy storage frequency response model, an upper system equation group is constructed, and according to the upper system equation group, combined with the proposed improved non-cooperative game, a non-cooperative game model prediction control strategy is constructed; Based on the household energy storage SOC model, a lower-level system equation group is constructed, and according to the lower-level system equation group, combined with the proposed cooperative game that introduces the concept of user level, a cooperative game model prediction control strategy is constructed; In the upper-level control, the upper-level dispatching center uses the non-cooperative game model predictive control strategy to solve the output reference values of the thermal power units and household energy storage clusters in the control time domain online according to the operating status of the thermal power units at the previous moment and the frequency regulation capacity reported by the intermediate agent, and outputs the solution results as control instructions; In the lower-level control, the intermediate agent uses the proposed cooperative game model predictive control strategy to solve the frequency regulation task of each household energy storage, and predicts the SOC value at the next moment according to the power response of the household energy storage; Based on the control instruction and the SOC value at the next moment, a reference power signal of each frequency modulation unit in the system at the next moment is obtained; after receiving the reference power signal, each frequency modulation unit adjusts the operating state of the thermal power unit to achieve the optimization of system frequency stability; The non-cooperative game model predictive control strategy includes: an upper-level prediction model, an upper-level objective function and an upper-level constraint condition; The upper system equations are: Where ΔP h is the output change of the thermal power unit, ΔP ref.h is the change in the reference value of the thermal power unit output, G gov (s) is the speed governor model of thermal power unit, G tur (s) is the turbine model of the thermal power unit, R is the primary frequency modulation coefficient of the system, Δf is the system frequency deviation, H is the inertia time constant, ΔP RESs is the output change of the household energy storage cluster, is the transfer function of the household energy storage cluster, ΔP ref.RESs is the change in the output reference value of the household energy storage cluster, ΔP load is the system load fluctuation, ΔP new The power fluctuation of new energy in the system; The upper prediction model established according to the upper system equations is: In the formula, A, B, E, and C are the state matrix, control input matrix, interference input matrix, and control output matrix of the system respectively; x(t) is the system state variable, u(t) is the system input variable, L(t) is the system disturbance variable, and y(t) is the system output variable; The upper objective function is: F1=min{MA ACE +N(ΔP h 2 +μΔP RESs )}, Where F1 is the upper objective function, M and N are the weight coefficients of the upper objective function; μ is called the game factor, and its size is determined by Δf, A ACE Provides automatic power generation control signal for the system; The upper-level constraints include: upper and lower limits of thermal power unit output, ramp rate constraints, and upper and lower limits of household energy storage capacity for frequency regulation; The cooperative game model predictive control strategy includes: a lower-level prediction model, a lower-level objective function, and lower-level constraints; The lower layer prediction model is: Where, T s is the discrete time constant, P RES.i , i∈1,2,···,n is the frequency regulation task undertaken by the i-th household energy storage; SOC i , E bess.i , i∈1,2,···,n are the SOC value of the i-th household energy storage and the capacity of the household energy storage battery respectively; The lower layer objective function is: Where F2 is the lower layer objective function, P ref.RESi is the frequency regulation task undertaken by the i-th household energy storage, Y i , i∈1,2,···,n is the user level of the i-th household energy storage, U and V are the weight coefficients of the lower layer objective function; The lower-level constraints include: the output of all household energy storage is equal to the power reference value constraint, and the upper and lower limit constraints of the frequency regulation of each household energy storage.
2. The method for user-side small household energy storage to participate in system frequency regulation according to claim 1, characterized in that: The method in which each frequency modulation unit adjusts the operating state of the thermal power unit after receiving the reference power signal to achieve the most stable and optimized system frequency includes: S1: Building upper and lower layers of game model prediction controllers, wherein the game model prediction controller is provided with an objective function, a prediction model and constraints; S2: Based on each game model prediction controller, the operating status of each frequency modulation unit at the current moment is obtained, the operating status is calculated through the prediction model, the ACE signal at the current moment is generated, and the objective function is solved through the Gurobi solver according to the ACE signal; S3: outputting the first set of parameters of the solution result as the real-time control instruction, and updating the reference power signal of each frequency modulation unit in the system at the next moment based on the real-time control instruction; S4: After receiving the new reference power signal, each frequency modulation unit in the system takes action to change the operating state of the system and returns to S2.
3. A system in which small household energy storage at the user side participates in system frequency regulation, characterized in that: include: A first building module, a second building module, a third building module, a fourth building module, an upper control module, a lower control module and an adjustment module; The first construction module is used to comprehensively consider the load characteristics and frequency response characteristics of household energy storage, and construct a two-layer control framework for household energy storage to participate in frequency regulation, wherein the two-layer control framework includes: an upper layer and a lower layer; the upper layer includes: thermal power units, upper dispatching centers, and intermediate agents, and the lower layer includes: intermediate agents and all grid-connected household energy storage; The second construction module is used to construct a double-layer game model prediction control strategy for household energy storage participating in frequency regulation based on the double-layer control framework, and the double-layer game model prediction control strategy includes: capacity reporting method, thermal power unit frequency response model, household energy storage frequency response model and grid connection and off-grid mechanism; The third building module is used for constructing an upper system equation group based on the thermal power unit frequency response model and the household energy storage frequency response model, and constructing a non-cooperative game model prediction control strategy based on the upper system equation group and the proposed improved non-cooperative game; The fourth building module is used to construct a lower-level system equation group based on the household energy storage SOC model, and to construct a cooperative game model prediction control strategy based on the lower-level system equation group and in combination with the cooperative game that introduces the concept of user level; The upper control module is used for the upper dispatch center to solve the output reference values of the thermal power units and the household energy storage cluster in the control time domain online according to the operating status of the thermal power units at the previous moment and the frequency regulation capacity reported by the intermediate agent by using the non-cooperative game model predictive control strategy in the upper control, and output the solution results as control instructions; The lower-level control module is used for the intermediate agent to solve the frequency regulation task of each household energy storage by using the proposed cooperative game model prediction control strategy in the lower-level control, and predict the SOC value at the next moment according to the power of the household energy storage response; The adjustment module is used to obtain a reference power signal of each frequency modulation unit in the system at the next moment based on the control instruction and the SOC value at the next moment; after receiving the reference power signal, each frequency modulation unit adjusts the operating state of the thermal power unit to achieve the optimization of system frequency stability; The non-cooperative game model predictive control strategy includes: an upper-level prediction model, an upper-level objective function and an upper-level constraint condition; The upper system equations are: Where ΔP h is the output change of the thermal power unit, ΔP ref.h is the change in the reference value of the thermal power unit output, G gov (s) is the speed governor model of thermal power unit, G tur (s) is the turbine model of the thermal power unit, R is the primary frequency modulation coefficient of the system, Δf is the system frequency deviation, H is the inertia time constant, ΔP RESs is the output change of the household energy storage cluster, is the transfer function of the household energy storage cluster, ΔP ref.RESs is the change in the output reference value of the household energy storage cluster, ΔP load is the system load fluctuation, ΔP new The power fluctuation of new energy in the system; The upper prediction model established according to the upper system equations is: In the formula, A, B, E, and C are the state matrix, control input matrix, interference input matrix, and control output matrix of the system respectively; x(t) is the system state variable, u(t) is the system input variable, L(t) is the system disturbance variable, and y(t) is the system output variable; The upper objective function is: Where F1 is the upper objective function, M and N are the weight coefficients of the upper objective function; μ is called the game factor, and its size is determined by Δf, A ACE Provides automatic power generation control signal for the system; The upper-level constraints include: upper and lower limits of thermal power unit output, ramp rate constraints, and upper and lower limits of household energy storage capacity for frequency regulation; The cooperative game model predictive control strategy includes: a lower-level prediction model, a lower-level objective function, and lower-level constraints; The lower layer prediction model is: Where, T s is the discrete time constant, P RES.i , i∈1,2,···,n is the frequency regulation task undertaken by the i-th household energy storage; SOC i , E bess.i , i∈1,2,···,n are the SOC value of the i-th household energy storage and the capacity of the household energy storage battery respectively; The lower layer objective function is: Where F2 is the lower layer objective function, P ref.RESi is the frequency regulation task undertaken by the i-th household energy storage, Y i , i∈1,2,···,n is the user level of the i-th household energy storage, U and V are the weight coefficients of the lower layer objective function; The lower-level constraints include: the output of all household energy storage is equal to the power reference value constraint, and the upper and lower limit constraints of the frequency regulation of each household energy storage.
4. The system for user-side small household energy storage to participate in system frequency regulation according to claim 3 is characterized in that: The adjustment module includes: a building unit, a solving unit, an updating unit and an adjusting unit; The building unit is used to build the upper and lower layers of the game model prediction controller, wherein the game model prediction controller is provided with an objective function, a prediction model and constraint conditions; The solving unit is used to predict the controller based on each game model, obtain the operating status of each frequency modulation unit at the current moment, calculate the operating status through the prediction model, generate the ACE signal at the current moment, and solve the objective function through the Gurobi solver according to the ACE signal; The updating unit is used to output the first set of parameters of the solution result as real-time control instructions, and based on the real-time control instructions, update the reference power signal of each frequency modulation unit in the system at the next moment; the adjusting unit is used for each frequency modulation unit in the system to take action after receiving the new reference power signal, change the operating state of the system, and return to the solving unit.
