Method and system for virtual power plant to participate in AGC frequency modulation

Abstract

The invention discloses a virtual power plant participation AGC frequency modulation method and system, and the method comprises the steps: receiving a power grid AGC frequency modulation instruction, and screening adjustable resources meeting conditions according to the actual control constraint, adjustment characteristic requirements and configuration conditions of the adjustable resources to form a standby resource pool; verifying the running state of the system and the adjusting capacity range of the standby resource pool; constructing a virtual power plant participation AGC frequency modulation net profit model; constructing constraint conditions including an adjustment range constraint, an energy storage charge state constraint and an AGC control constraint; a virtual power plant participation AGC frequency modulation strategy optimization model is constructed, an interior point method is adopted for solving, and an optimal adjustment strategy is obtained; executing the optimization strategy, monitoring the adjustment deviation in real time, and starting real-time deviation correction operation based on the adjustment deviation; according to the method, the calculation efficiency and the solving precision are remarkably improved, the quick response to the frequency modulation requirement of the power grid is ensured through a real-time correction mechanism, and high-precision power regulation control is realized.

Description

Technical Field

[0001] The present invention belongs to the field of electricity management, and in particular relates to a method and system for a virtual power plant to participate in AGC frequency modulation. Background Art

[0002] With the large-scale integration of renewable energy into the power system, grid frequency stability faces new challenges. Automatic generation control (AGC), a key means of maintaining grid frequency stability, requires fast and precise frequency regulation resource response. Traditional thermal power units, limited by their large mechanical inertia and slow regulation speed, struggle to adapt to the rapid frequency regulation required by high renewable energy access rates. In this context, virtual power plants (VPPs) are becoming an important technological development direction for improving AGC frequency regulation performance, as they can aggregate multiple flexible resources such as distributed power sources, energy storage systems, and adjustable loads, and participate in grid frequency regulation through collaborative optimization and control.

[0003] Currently, virtual power plants participating in AGC frequency regulation primarily rely on centralized or distributed optimization control methods. Existing technologies typically employ model predictive control, multi-objective optimization algorithms, or hierarchical coordinated control strategies. These technologies optimize the scheduling of multiple resource types within a virtual power plant by establishing mathematical models that incorporate objectives such as power balance, energy storage charging and discharging constraints, and market returns.

[0004] However, when participating in AGC frequency regulation, virtual power plants need to deal with high-dimensional, nonlinear, and multi-constrained optimization problems. Traditional optimization algorithms often face problems such as long calculation time and easy falling into local optimality when solving such complex models, resulting in poor real-time performance of frequency regulation instructions and difficulty in taking into account multiple goals such as economy, energy storage life, and grid frequency regulation accuracy. Summary of the Invention

[0005] Purpose of the invention: The purpose of the present invention is to provide a method for virtual power plants to participate in AGC frequency regulation with high efficiency and high solution accuracy; on the other hand, to provide a system for virtual power plants to participate in AGC frequency regulation.

[0006] Technical solution: The frequency modulation method of the present invention comprises the following steps:

[0007] (1) When receiving the AGC frequency regulation command from the power grid, the system selects qualified adjustable resources according to the actual control constraints, regulation characteristics and configuration of the adjustable resources to form a backup resource pool, ensuring that the resources participating in the frequency regulation have the ability to respond quickly while avoiding regulation delays or invalid scheduling caused by resource mismatch;

[0008] (2) Verify the system operation status and the regulation capacity range of the backup resource pool to determine whether the current frequency regulation requirements are met, so as to prevent frequency regulation failure or grid stability from being affected due to insufficient regulation capacity;

[0009] (3) Based on the total revenue from grid subsidies obtained by virtual power plants participating in frequency regulation and the total cost of virtual power plants participating in frequency regulation, a net profit model for virtual power plants participating in AGC frequency regulation is constructed to provide an economic basis for subsequent strategy optimization and ensure that virtual power plants maximize their revenue while meeting frequency regulation needs;

[0010] (4) Constructing constraints, including regulation range constraints, energy storage state of charge constraints, and AGC control constraints, to limit the minimum / maximum adjustable power of each resource, avoid overload or invalid regulation, ensure that the SOC of the energy storage is within a safe range during the frequency regulation process, prevent overcharging / over-discharging from affecting its life, meet the grid's response time to frequency regulation instructions, regulation accuracy, and other requirements, and ensure the feasibility and safety of the optimization strategy;

[0011] (5) Based on the net profit model, an optimization model for the virtual power plant participating in the AGC frequency regulation strategy is constructed and solved using the interior point method according to the constraints to obtain the optimal regulation strategy. The global optimal or approximate optimal solution can be quickly found while ensuring convergence, thereby obtaining the optimal power allocation strategy for each adjustable resource, taking into account both economic and dynamic response performance;

[0012] (6) Execute the optimization strategy and monitor the regulation deviation in real time, initiate real-time correction operations based on the regulation deviation, significantly improve the frequency regulation accuracy and system robustness, and ensure that the virtual power plant can reliably participate in grid auxiliary services.

[0013] Preferably, the actual control constraints described in step 1 are that the resource is in remote control mode, the operating status is normal, and the data quality is normal; the regulation characteristic requirement is that the historical frequency regulation performance index is higher than the preset threshold K0; the configuration situation is that it is allowed to participate in regulation and is not in the power protection period; the system can accurately screen out high-reliability, high-responsiveness, high-quality frequency regulation resources; this screening mechanism performs a three-dimensional evaluation from five dimensions: control authority, real-time health, data credibility, historical performance, and scheduling compliance, to ensure that each device in the formed backup resource pool has millisecond-level remote controllability, stable frequency regulation performance records, and conflict-free scheduling windows, thereby eliminating the risk of frequency regulation failure caused by device offline, communication anomalies, substandard performance, or scheduling conflicts from the source, and significantly reducing the invalid iterations of subsequent optimization calculations.

[0014] Preferably, the system operating state in step 2 is that the system is in normal operating state; the adjustment capacity range is the preset ratio th0 in which the target adjustment amount is less than the sum of the maximum adjustment amounts of the backup resource pool. This mechanism establishes dual protection from the two dimensions of system stability and adjustment margin. This dynamic capacity verification not only avoids the failure of frequency regulation instructions due to system abnormalities, but also effectively responds to uncertain factors such as actual resource response deviation and grid demand fluctuation by retaining the adjustment margin, thereby reducing the safety risk of the frequency regulation process, and at the same time, the capacity utilization rate of the backup resource pool is always in the optimal range, which not only prevents the reduction of adjustment accuracy caused by resource overload, but also avoids the economic loss caused by idle capacity. The preset ratio can be dynamically adjusted through online learning algorithms according to different application scenarios to further adapt to the flexibility requirements of new energy high penetration power grids.

[0015] Preferably, the construction of the net profit model of the virtual power plant participating in AGC frequency regulation in step 3 includes:

[0016] (31) Calculate the total revenue R of the virtual power plant from participating in frequency regulation and obtaining grid subsidies in :

[0017] R in =R ina +R ina

[0018] R ina =(|P PV -P PV0 |×Y PV ×K Z1 )+(P CL ×Y CL ×K Z2 )+(|P ESS -P ESS0 |×Y ESS ×K Z3 )

[0019] R inb =(S PV ×T PV ×M PV )+(S CL ×T CL ×M CL )+(S ESS ×T ESS ×M ESS )

[0020] Where, R ina is the total income of the virtual power plant participating in the frequency regulation mileage subsidy, R inb is the total income from the virtual power plant participating in frequency regulation capacity subsidy, P PV is the actual output value of the photovoltaic unit, P PV0is the initial output value of the photovoltaic unit when responding to the instruction, Y PV is the settlement price per unit frequency regulation mileage of photovoltaic units, K Z1 is the comprehensive frequency regulation performance index of photovoltaic units, P CL Y is the controllable load downward adjustment amount, CL is the settlement price per unit frequency regulation mileage of controllable load, K Z2 is the comprehensive frequency regulation performance index of controllable load, P ESS is the actual output value of the energy storage system, P ESS0 Y is the initial output value of the energy storage system when responding to the command, ESS is the settlement price per unit frequency regulation mileage of the energy storage system, K Z3 is the comprehensive frequency regulation performance index of the energy storage system, S PV is the frequency regulation capacity of the photovoltaic unit, T PV Frequency regulation service time for photovoltaic units, M PV is the compensation price per unit capacity of photovoltaic units, S CL To control the load to participate in frequency regulation, adjust the capacity downward, T CL Frequency regulation service duration for controllable load, M CL Adjust the capacity compensation price for the controllable load unit, S ESS is the frequency regulation capacity of the energy storage system, T ESS Frequency regulation service duration for energy storage system, M ESS The compensation price per unit capacity of the energy storage system;

[0021] The comprehensive frequency modulation performance index K Zi The calculation formula is as follows:

[0022] K Zi =(k v / k N )×(T N / T d )×(ΔP N / ΔP a )

[0023] Where k v is the actual regulation rate of the frequency regulating unit, k N is the standard regulation rate of the frequency modulation unit, T N is the standard response time of the frequency modulation unit, T d is the actual response time of the frequency regulating unit, ΔP a is the actual adjustment deviation of the frequency regulating unit, ΔP N The allowable adjustment deviation for the frequency regulating unit;

[0024] (32) Calculate the total cost C of the virtual power plant participating in frequency regulation cost :

[0025]

[0026] Where, P PV is the actual output value of the photovoltaic unit, P PVn is the planned output value of the photovoltaic unit, c MT1 is the unit photovoltaic output operation and maintenance cost coefficient, c q is the penalty coefficient for abandoned photovoltaic power per unit output, c clbn P is the compensation price for controllable load participating in frequency regulation at each stage. CL is the amount of downward adjustment of the controllable load, G1, G2, G3, ..., G n is the downward adjustment amount of the controllable load at each stage, C MT2 is the unit energy storage output operation and maintenance cost coefficient, P ESS is the actual output value of energy storage;

[0027] (33) The formula for the net profit of a virtual power plant participating in AGC frequency regulation is:

[0028] P r =R in -C cost .

[0029] The model innovatively divides frequency regulation benefits into two dimensions: mileage subsidies and capacity subsidies, and introduces comprehensive frequency regulation performance indicators to achieve differentiated benefit distribution, effectively incentivizing various types of resources to improve frequency regulation performance; in terms of cost accounting, the model fully considers key factors such as photovoltaic curtailment losses, energy storage operation and maintenance costs, and load compensation costs, especially for controllable loads. A time-of-use compensation mechanism is designed. This refined modeling method can not only accurately reflect the economic benefits of the frequency regulation process, but also guide resources to optimize their own operating characteristics, thereby significantly improving the overall economic benefits and resource utilization efficiency of the virtual power plant while ensuring the grid's frequency regulation needs.

[0030] Preferably, the adjustment range constraint expression in step 4 is:

[0031]

[0032] Where, P PVmax 、P CLmax 、P ESSmax 、P PVmin 、P CLmin 、P ESSmin They are the upper and lower limits of photovoltaic, controllable load and energy storage output respectively;

[0033] The energy storage state of charge constraint expression is:

[0034]

[0035] Where, SOC0, SOC tare the initial value and the end value of the energy storage charge state, E is the energy storage capacity, η c ,η d are charging and discharging efficiency, S c 、S d They are energy storage charging and discharging states, and can be 0 or 1, SOC t SOC is the state of charge of the energy storage during period t. max , SOC min are the upper and lower limits of the energy storage state of charge respectively;

[0036] The AGC control constraint expression is:

[0037] P AGC =P PV -P CL +P ESS ×S c -P ESS ×S d

[0038] Where, P AGC Issue AGC control command values ​​to the upper-level power grid dispatcher.

[0039] By constructing a multi-dimensional constraint system, a reliable and safe operating boundary is provided for virtual power plants to participate in AGC frequency regulation; among them, the regulation range constraint clarifies the upper and lower limits of the output of various resources such as photovoltaics, controllable loads and energy storage, ensuring that the frequency regulation instructions are executed within the physical tolerance of the equipment; the energy storage charge state constraint adopts a dynamic energy balance equation, comprehensively considering the charging and discharging efficiency and operating status, effectively maintaining the energy balance of energy storage in the frequency regulation process, and avoiding overcharging and over-discharging; the AGC control constraint strictly ensures the consistency of the overall output of the virtual power plant with the grid dispatching instructions; these constraints together constitute a complete operation safety protection network, which ensures that all participating resources are always in a safe, stable and controllable operating state while pursuing the economy of frequency regulation, laying a solid foundation for the reliable participation of virtual power plants in grid auxiliary services.

[0040] Preferably, the objective function expression of the AGC frequency regulation strategy optimization model for the virtual power plant in step 5 is:

[0041]

[0042] The net profit model of virtual power plants participating in AGC frequency regulation is directly used as the optimization objective function, achieving an organic unity of economic efficiency and frequency regulation performance. The model is guided by maximizing the net profit of frequency regulation and intelligently coordinates the regulation behavior of heterogeneous resources such as photovoltaics, energy storage, and controllable loads while meeting various operational constraints. This design enables the optimization algorithm to automatically balance the relationship between frequency regulation benefits and operating costs, ensuring the timeliness and accuracy of responding to the power grid's AGC instructions while fully considering the regulation costs and operational limitations of different resources.

[0043] Preferably, the interior point method is used to solve the problem in step 5 to obtain the optimal adjustment strategy, which includes:

[0044] (51) Initialize the solution parameters: set the slack variables l and u to ensure [l,u] T > 0, set the Lagrange multipliers z, w, y to satisfy [z > 0, w < 0, y ≠ 0] T > 0, set the initial values ​​of the variables in the optimization problem, take the central parameter σ∈(0,1), and give the calculation accuracy ε=10 6 , the initial value of the number of iterations k = 0, the maximum number of iterations K max =50, solve the dual gap Gap=l T zu T w, solve the disturbance factor μ=Gap / 2r>0, r is the number of slack variables;

[0045] (52) Calculate the coefficient matrix of the correction equation:

[0046] The variable x is expressed as follows:

[0047] x={P PV P CL P ESS S c S d SOC t}

[0048] Jacobian matrix of equality constraints▽ x The expression of h(x) is as follows:

[0049]

[0050] Where h i (x) are all the equations in the above constraints, and h i (x) = 0;

[0051] Inequality constraint Jacobian matrix ▽ x The expression of g(x) is as follows:

[0052]

[0053] Where g i (x) are respectively the upper and lower bounds of the output value of photovoltaic or controllable load and energy storage participating in AGC frequency regulation, the upper and lower bounds of energy storage charge state, and the energy storage charging and discharging constraints in the above constraints;

[0054] The diagonal matrix expression is as follows:

[0055]

[0056] The Heisenberg matrix H' is expressed as follows:

[0057]

[0058] Where, is the Heisenberg matrix of the target function, is the equality constraint Heisenberg matrix, is the Heisenberg matrix of the inequality constraints;

[0059] Heisenberg matrix of the objective function The expression is as follows:

[0060]

[0061] Equality Constrained Heisenberg Matrix The product expression with the Lagrange multiplier y is as follows:

[0062]

[0063] Inequality Constrained Heisenberg Matrix The product expression with the Lagrange multiplier z+w is as follows:

[0064]

[0065] (53) According to the coefficient matrix and objective function of the correction equations, the expression of the correction equations is as follows:

[0066]

[0067] (54) All variables and constants are brought into the correction equation to obtain the correction parameters, and the original variables and Lagrange multipliers are updated. If the number of iterations is less than the maximum number of iterations and the complementary gap satisfies the gap parameter, the optimal solution is obtained and the calculation is stopped. The optimal photovoltaic unit output, the controllable load downward adjustment amount and the energy storage output are obtained, and the virtual power plant participating in the AGC frequency regulation strategy is obtained.

[0068] By systematically applying the interior point method to solve the optimization model, efficient calculation and accurate solution of the virtual power plant frequency regulation strategy are achieved; this solution method can effectively handle large-scale nonlinear constraint problems, overcome the defect that traditional optimization algorithms are prone to falling into local optimality, ensure the acquisition of global optimal or near-global optimal regulation strategies, and provide accurate power allocation solutions for the coordinated frequency regulation of photovoltaic units, controllable loads and energy storage systems, so that the overall operation of the virtual power plant is at a balance point between technical feasibility and economic optimality.

[0069] Preferably, the adjustment deviation in step 6 is the difference between the current real-time load of all resources involved in the regulation and the target load specified by the regulation demand. If the adjustment deviation is greater than a preset threshold of the regulation target, a real-time correction operation is initiated;

[0070] The resource screening constraints in the real-time correction operation are that the resource has adjustment margin in the direction of compensating the difference, the time since the last adjustment is greater than the control cycle, it has a second-level response capability, and the cumulative number of adjustments is less than the upper limit of the adjustment frequency.

[0071] By establishing a closed-loop feedback regulation mechanism, precise control of the virtual power plant's participation in the AGC frequency regulation process is achieved; the system monitors the deviation between the actual output of each regulation resource and the target instruction in real time. This dynamic compensation mechanism not only ensures the accurate tracking of the frequency regulation instruction, but also avoids excessive use of resources or frequent actions by setting reasonable regulation constraints. It maintains the stability of the frequency regulation process and effectively extends the service life of the equipment. This correction strategy significantly improves the overall frequency regulation quality of the virtual power plant, enabling the system to continuously meet the grid's stringent requirements for frequency regulation.

[0072] The frequency modulation system of the present invention comprises:

[0073] The resource screening module is used to screen eligible adjustable resources according to the actual control constraints, regulation characteristics and configuration of the adjustable resources to form a backup resource pool when receiving the grid AGC frequency regulation command;

[0074] A demand verification module is used to verify the system operation status and the adjustment capacity range of the backup resource pool to determine whether the current frequency regulation demand is met;

[0075] A net profit model building module is used to build a net profit model for virtual power plants participating in AGC frequency regulation based on the total revenue from grid subsidies and the total cost of virtual power plants participating in frequency regulation.

[0076] Constraint construction module, used to construct regulation range constraints, energy storage state of charge constraints and AGC control constraints;

[0077] An optimization solution module is used to construct an optimization model for the virtual power plant participating in the AGC frequency regulation strategy based on the net profit model and solve it using the interior point method according to the constraint conditions to obtain the optimal regulation strategy;

[0078] The execution feedback module is used to execute the optimization strategy and monitor the adjustment deviation in real time, and start the real-time correction operation based on the adjustment deviation.

[0079] A computer-readable storage medium having a computer program stored thereon, characterized in that when the computer program is executed by a processor, the method of virtual power plant participating in AGC frequency modulation is implemented.

[0080] Beneficial effects: Compared with the existing technology, the present invention has the following significant advantages: 1. It significantly improves the computing efficiency and solution accuracy, ensures rapid response to the grid frequency regulation needs through a real-time correction mechanism, and realizes high-precision power regulation control; 2. By unifying the modeling of frequency regulation benefits, costs and physical constraints, while optimizing economic benefits and system operation indicators, and introducing a comprehensive performance evaluation system, it realizes the coordinated and optimized operation of various distributed resources; 3. It establishes a resource screening system based on multi-dimensional evaluation standards, dynamically adjusts the resource combination involved in frequency regulation, effectively improves resource utilization efficiency, and reduces ineffective scheduling. BRIEF DESCRIPTION OF THE DRAWINGS

[0081] Figure 1 is a flow chart of the method of the present invention;

[0082] Figure 2 This is a flow chart of the interior point method for solving the optimization model. DETAILED DESCRIPTION

[0083] The technical solution of the present invention will be further described below with reference to the accompanying drawings.

[0084] like Figure 1 As shown, this embodiment provides a method for a virtual power plant to participate in AGC frequency modulation, comprising the following steps:

[0085] (1) After receiving the AGC frequency regulation command from the power grid, the system selects the qualified adjustable resources according to the actual control constraints, regulation characteristics and configuration of the adjustable resources to form a backup resource pool.

[0086] The actual control constraints are that the resource is in remote control mode, the operating status is normal, and the data quality is normal; the regulation characteristic requirement is that the historical frequency regulation performance index is higher than the preset threshold K0 (which can be configured according to the actual situation on site); the configuration is that participation in regulation is allowed and it is not in the power protection period.

[0087] (2) Verify the system operation status and the adjustment capacity range of the backup resource pool to determine whether the current frequency regulation requirements are met.

[0088] The system operation state is that the system is in normal operation; the adjustment capacity range is the preset ratio th0 (which can be configured according to actual needs on site) in which the target adjustment amount is less than the sum of the maximum adjustment amounts of the backup resource pool.

[0089] (3) Based on the total income from grid subsidies obtained by virtual power plants participating in frequency regulation and the total cost of virtual power plants participating in frequency regulation, a net profit model for virtual power plants participating in AGC frequency regulation is constructed. The specific steps are as follows:

[0090] (31) Calculate the total revenue R of the virtual power plant from participating in frequency regulation and obtaining grid subsidies in :

[0091] R in =R ina +R ina

[0092] R ina =(|P PV -P PV0 |×Y PV ×K Z1 )+(P CL ×Y CL ×K Z2 )+(|P ESS -P ESS0 |×Y ESS ×K Z3 )

[0093] R inb =(S PV ×T PV ×M PV )+(S CL ×T CL ×M CL )+(S ESS ×T ESS ×M ESS )

[0094] Where, R ina is the total income of the virtual power plant participating in the frequency regulation mileage subsidy, R inb is the total income from the virtual power plant participating in frequency regulation capacity subsidy, P PV is the actual output value of the photovoltaic unit, P PV0 is the initial output value of the photovoltaic unit when responding to the instruction, Y PV is the settlement price per unit frequency regulation mileage of photovoltaic units, K Z1 is the comprehensive frequency regulation performance index of photovoltaic units, P CL Y is the controllable load downward adjustment amount, CL is the settlement price per unit frequency regulation mileage of controllable load, K Z2 is the comprehensive frequency regulation performance index of controllable load, P ESS is the actual output value of the energy storage system, PESS0 Y is the initial output value of the energy storage system when responding to the command, ESS is the settlement price per unit frequency regulation mileage of the energy storage system, K Z3 is the comprehensive frequency regulation performance index of the energy storage system, S PV is the frequency regulation capacity of the photovoltaic unit, which is 7.5% of the photovoltaic unit capacity. PV Frequency regulation service time for photovoltaic units, M PV is the compensation price per unit capacity of photovoltaic units, S CL The controllable load participates in frequency regulation and adjusts the capacity downward, which is 7.5% of the regulation capacity. CL Frequency regulation service duration for controllable load, M CL Adjust the capacity compensation price for the controllable load unit, S ESS is the frequency regulation capacity of the energy storage system, which is 7.5% of the energy storage system capacity. ESS Frequency regulation service duration for energy storage system, M ESS The compensation price per unit capacity of the energy storage system;

[0095] Comprehensive frequency modulation performance index K Zi The calculation formula is as follows:

[0096] K Zi =(k v / k N )×(T N / T d )×(ΔP N / ΔP a )

[0097] Where k v is the actual regulation rate of the frequency regulating unit, k N is the standard regulation rate of the frequency modulation unit, T N is the standard response time of the frequency modulation unit, T d is the actual response time of the frequency regulating unit, ΔP a is the actual adjustment deviation of the frequency regulating unit, ΔP N The allowable adjustment deviation for the frequency regulating unit;

[0098] (32) Calculate the total cost C of the virtual power plant participating in frequency regulation cost :

[0099]

[0100] Where, P PV is the actual output value of the photovoltaic unit, P PVn is the planned output value of the photovoltaic unit, c MT1 is the unit photovoltaic output operation and maintenance cost coefficient, c q is the penalty coefficient for abandoned photovoltaic power per unit output, c clbnP is the compensation price for controllable load participating in frequency regulation at each stage. CL is the amount of downward adjustment of the controllable load, G1, G2, G3, ..., G n is the downward adjustment amount of the controllable load at each stage, C MT2 is the unit energy storage output operation and maintenance cost coefficient, P ESS is the actual output value of energy storage;

[0101] (33) The formula for the net profit of a virtual power plant participating in AGC frequency regulation is:

[0102] P r =R in -C cost .

[0103] (4) Construct constraints, including regulation range constraints, energy storage state of charge constraints, and AGC control constraints. The specific steps are as follows:

[0104] The adjustment range constraint expression is:

[0105]

[0106] Where, P PVmax 、P CLmax 、P ESSmax 、P PVmin 、P CLmin 、P ESSmin They are the upper and lower limits of photovoltaic, controllable load and energy storage output respectively;

[0107] The energy storage state of charge constraint expression is:

[0108]

[0109] Where, SOC0, SOC t are the initial value and the end value of the energy storage charge state, E is the energy storage capacity, η c ,η d are charging and discharging efficiency, S c 、S d They are energy storage charging and discharging states, and can be 0 or 1, SOC t SOC is the state of charge of the energy storage during period t. max , SOC min are the upper and lower limits of the energy storage state of charge respectively;

[0110] The AGC control constraint expression is:

[0111] P AGC =P PV -P CL +P ESS ×S c -P ESS×S d

[0112] Where, P AGC Issue AGC control command values ​​to the upper-level power grid dispatcher.

[0113] (5) Construct an optimization model for the virtual power plant to participate in the AGC frequency regulation strategy and use the interior point method to solve it to obtain the optimal regulation strategy.

[0114] Among them, the objective function expression of the AGC frequency regulation strategy optimization model for virtual power plants is:

[0115] f(x) max =P r =R in -C cost

[0116] =(|P PV -P PV0 |×Y PV ×K Z1 +P CL ×Y CL ×K Z2 +|P ESS -P ESS0 |×Y ESS ×K Z3 )

[0117] +(S PV ×T PV ×M PV +S CL ×T CL ×M CL +S ESS ×T ESS ×M ESS )

[0118] -(c MT1 ×P PV +c q ×max(0,P PVn -P PV )+c clb ×P CL +C MT2 ×P ESS )

[0119] like Figure 2 As shown, the specific steps of using the interior point method to solve and obtain the optimal adjustment strategy are as follows:

[0120] (51) Initialize the solution parameters: set the slack variables l and u to ensure [l,u] T > 0, set the Lagrange multipliers z, w, y to satisfy [z > 0, w < 0, y ≠ 0] T> 0, set the initial values ​​of the variables in the optimization problem, take the central parameter σ∈(0,1), and give the calculation accuracy ε=10 6 , the initial value of the number of iterations k = 0, the maximum number of iterations K max =50, solve the dual gap Gap=l T zu T w, solve the disturbance factor μ=Gap / 2r>0, r is the number of slack variables;

[0121] (52) Calculate the coefficient matrix of the correction equation:

[0122] The variable x is expressed as follows:

[0123] x={P PV P CL P ESS S c S d SOC t}

[0124] Jacobian matrix of equality constraints▽ x The expression of h(x) is as follows:

[0125]

[0126] Where h i (x) are all the equations in the above constraints, and h i (x) = 0;

[0127] Inequality constraint Jacobian matrix ▽ x The expression of g(x) is as follows:

[0128]

[0129] Where g i (x) are respectively the upper and lower bounds of the output value of photovoltaic or controllable load and energy storage participating in AGC frequency regulation, the upper and lower bounds of energy storage charge state, and the energy storage charging and discharging constraints in the above constraints;

[0130] The diagonal matrix expression is as follows:

[0131]

[0132] The Heisenberg matrix H' is expressed as follows:

[0133]

[0134] Where, is the Heisenberg matrix of the target function, is the equality constraint Heisenberg matrix, is the Heisenberg matrix of the inequality constraints;

[0135] Heisenberg matrix of the objective function The expression is as follows:

[0136]

[0137] Equality Constrained Heisenberg Matrix The product expression with the Lagrange multiplier y is as follows:

[0138]

[0139] Inequality Constrained Heisenberg Matrix The product expression with the Lagrange multiplier z+w is as follows:

[0140]

[0141] (53) According to the coefficient matrix and objective function of the correction equations, the expression of the correction equations is as follows:

[0142]

[0143] (54) All variables and constants are brought into the correction equation to obtain the correction parameters, and the original variables and Lagrange multipliers are updated. If the number of iterations is less than the maximum number of iterations and the complementary gap satisfies the gap parameter, the optimal solution is obtained and the calculation is stopped. The optimal photovoltaic unit output, the controllable load downward adjustment amount and the energy storage output are obtained, and the virtual power plant participating in the AGC frequency regulation strategy is obtained.

[0144] (6) Executing the optimization strategy and monitoring the adjustment deviation in real time, and initiating a real-time correction operation based on the adjustment deviation.

[0145] The adjustment deviation is the difference between the current real-time load of all resources involved in the regulation and the target load specified by the regulation demand. If the adjustment deviation is greater than the preset threshold of the regulation target (for example, 5%), the real-time correction operation is initiated;

[0146] The resource screening constraints in real-time correction operations are that the resource has adjustment margin in the direction of compensating for the difference, the time since the last adjustment is greater than the control cycle, it has a response capability in seconds, and the cumulative number of adjustments is less than the upper limit of the adjustment frequency.

[0147] Based on the above frequency modulation method, the corresponding frequency modulation system includes:

[0148] The resource screening module is used to receive the grid AGC frequency regulation command and select qualified adjustable resources according to the actual control constraints, regulation characteristics and configuration of the adjustable resources to form a backup resource pool;

[0149] The demand verification module is used to verify the system operation status and the adjustment capacity range of the backup resource pool to determine whether the current frequency regulation demand is met;

[0150] A net profit model building module is used to build a net profit model for virtual power plants participating in AGC frequency regulation based on the total revenue from grid subsidies and the total cost of virtual power plants participating in frequency regulation.

[0151] Constraint construction module, used to construct regulation range constraints, energy storage state of charge constraints and AGC control constraints;

[0152] The optimization solution module is used to build an optimization model for the virtual power plant to participate in the AGC frequency regulation strategy and use the interior point method to solve it to obtain the optimal regulation strategy;

[0153] The execution feedback module is used to execute the optimization strategy and monitor the adjustment deviation in real time, and start the real-time correction operation based on the adjustment deviation.

[0154] The embodiment of the present invention also discloses a computer-readable storage medium.

[0155] Specifically, a computer-readable storage medium is used to store a computer program, and when the computer program is executed by a processor, the method in the above-mentioned method implementation is implemented. Those skilled in the art will understand that the implementation of all or part of the process in the above-mentioned embodiment method of the present application can be completed by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it may include the process of the implementation of each of the above-mentioned methods. Among them, the storage medium may be a magnetic disk, an optical disk, a read-only memory (ROM), a random access memory (RAM), a flash memory (Flash Memory), a hard disk (Hard Disk Drive, abbreviated: HDD) or a solid-state drive (SSD), etc.; the storage medium may also include a combination of the above-mentioned types of memories.

Claims

1. A method for virtual power plant to participate in AGC frequency modulation, characterized in that: include: When receiving the AGC frequency regulation command from the power grid, the system selects the qualified adjustable resources according to the actual control constraints, regulation characteristics and configuration of the adjustable resources to form a backup resource pool; Verify the system operating status and the adjustment capacity range of the backup resource pool to determine whether the current frequency regulation requirements are met; Based on the total revenue from grid subsidies obtained by virtual power plants participating in frequency regulation and the total cost of virtual power plants participating in frequency regulation, a net profit model for virtual power plants participating in AGC frequency regulation is constructed; Construct constraints, including regulation range constraints, energy storage state of charge constraints, and AGC control constraints; Based on the net profit model, an optimization model for the virtual power plant participating in the AGC frequency regulation strategy is constructed and solved using the interior point method according to the constraints to obtain the optimal regulation strategy; Execute the optimization strategy and monitor the adjustment deviation in real time, and initiate a real-time correction operation based on the adjustment deviation.

2. The frequency modulation method according to claim 1, wherein: The actual control constraints include at least one of the following: the resource is in remote control mode, the operating status is normal, and the data quality is normal; the regulation characteristic requirement is that the historical frequency regulation performance index is higher than the preset threshold K0; the configuration situation is that participation in regulation is allowed and it is not in the power protection period.

3. The frequency modulation method according to claim 1, wherein: The system operation state is that the system is in a normal operation state; the adjustment capacity range is a preset ratio th0 in which the target adjustment amount is less than the sum of the maximum adjustment amounts of the backup resource pool.

4. The frequency modulation method according to claim 1, wherein: The net profit model of constructing a virtual power plant participating in AGC frequency regulation includes: Calculate the total income R of the virtual power plant participating in frequency regulation and obtaining grid subsidies in : R in =R ina +R ina R ina =(|P PV -P PV0 |×Y PV ×K Z1 )+(P CL ×Y CL ×K Z2 )+(|P ESS -P ESS0 |×Y ESS ×K Z3 ) R inb =(S PV ×T PV ×M PV )+(S CL ×T CL ×M CL )+(S ESS ×T ESS ×M ESS ) Where, R ina is the total income of the virtual power plant participating in the frequency regulation mileage subsidy, R inb is the total income from the virtual power plant participating in frequency regulation capacity subsidy, P PV is the actual output value of the photovoltaic unit, P PV0 is the initial output value of the photovoltaic unit when responding to the instruction, Y PV is the settlement price per unit frequency regulation mileage of photovoltaic units, K Z1 is the comprehensive frequency regulation performance index of photovoltaic units, P CL Y is the controllable load downward adjustment amount, CL is the settlement price per unit frequency regulation mileage of controllable load, K Z2 is the comprehensive frequency regulation performance index of controllable load, P ESS is the actual output value of the energy storage system, P ESS0 Y is the initial output value of the energy storage system when responding to the command, ESS is the settlement price per unit frequency regulation mileage of the energy storage system, K Z3 is the comprehensive frequency regulation performance index of the energy storage system, S PV is the frequency regulation capacity of the photovoltaic unit, T PV Frequency regulation service time for photovoltaic units, M PV is the compensation price per unit capacity of photovoltaic units, S CL To control the load to participate in frequency regulation, adjust the capacity downward, T CL Frequency regulation service duration for controllable load, M CL Adjust the capacity compensation price for the controllable load unit, S ESS is the frequency regulation capacity of the energy storage system, T ESS Frequency regulation service duration for energy storage system, M ESS The compensation price per unit capacity of the energy storage system; The comprehensive frequency modulation performance index K Zi The calculation formula is as follows: K Zi =(k v / k N )×(T N / T d )×(ΔP N / ΔP a ) Where k v is the actual regulation rate of the frequency regulating unit, k N is the standard regulation rate of the frequency modulation unit, T N is the standard response time of the frequency modulation unit, T d is the actual response time of the frequency regulating unit, ΔP a is the actual adjustment deviation of the frequency regulating unit, ΔP N The allowable adjustment deviation for the frequency regulating unit; Calculate the total cost C of virtual power plant participating in frequency regulation cost : Where, P PV is the actual output value of the photovoltaic unit, P PVn is the planned output value of the photovoltaic unit, c MT1 is the unit photovoltaic output operation and maintenance cost coefficient, c q is the penalty coefficient for abandoned photovoltaic power per unit output, c clbn P is the compensation price for controllable load participating in frequency regulation at each stage. CL is the amount of downward adjustment of the controllable load, G1, G2, G3, ..., G n is the downward adjustment amount of the controllable load at each stage, C MT2 is the unit energy storage output operation and maintenance cost coefficient, P ESS is the actual output value of energy storage; The formula for the net profit of the virtual power plant participating in AGC frequency regulation is: P r =R in -C cost 。 5. The frequency modulation method according to claim 1, wherein: The adjustment range constraint expression is: Where, P PVmax 、P CLmax 、P ESSmax 、P PVmin 、P CLmin 、P ESSmin They are the upper and lower limits of photovoltaic, controllable load and energy storage output respectively; The energy storage state of charge constraint expression is: Where, SOC0, SOC t are the initial value and the end value of the energy storage charge state, E is the energy storage capacity, η c ,η d are charging and discharging efficiency, S c 、S d They are energy storage charging and discharging states, taking 0 or 1, SOC t SOC is the state of charge of the energy storage during period t max , SOC min They are the upper and lower limits of energy storage state of charge respectively; The AGC control constraint expression is: P AGC =P PV -P CL +P ESS ×S c -P ESS ×S d Where, P AGC Issue AGC control command values ​​to the upper-level power grid dispatcher.

6. The frequency modulation method according to claims 4 and 5, characterized in that The objective function expression of the AGC frequency regulation strategy optimization model for the virtual power plant is:

7. The frequency modulation method according to claim 6, characterized in that: The interior point method is used to solve and obtain the optimal adjustment strategy, which includes: Initialize the solution parameters: set the slack variables l and u to ensure [l,u] T > 0, set the Lagrange multipliers z, w, y to satisfy [z > 0, w < 0, y ≠ 0] T > 0, set the initial values ​​of the variables in the optimization problem, take the central parameter σ∈(0,1), and give the calculation accuracy ε=10 6 , the initial value of the number of iterations k = 0, the maximum number of iterations K max =50, solve the dual gap Gap=l T zu T w, solve the disturbance factor μ=Gap / 2r>0, r is the number of slack variables; Compute the coefficient matrix of the correction equation: The variable x is expressed as follows: x={P PV P CL P ESS S c S d SOC t } Jacobian matrix of equality constraints▽ x The expression of h(x) is as follows: Where h i (x) are all the equations in the above constraints, and h i (x) = 0; Inequality constraint Jacobian matrix ▽ x The expression of g(x) is as follows: Where g i (x) are respectively the upper and lower bounds of the output value of photovoltaic or controllable load and energy storage participating in AGC frequency regulation, the upper and lower bounds of energy storage charge state, and the energy storage charging and discharging constraints in the above constraints; The diagonal matrix expression is as follows: The Heisenberg matrix H' is expressed as follows: Where, is the Heisenberg matrix of the target function, is the equality constraint Heisenberg matrix, is the Heisenberg matrix of the inequality constraints; Heisenberg matrix of the objective function The expression is as follows: Equality Constrained Heisenberg Matrix The product expression with the Lagrange multiplier y is as follows: Inequality Constrained Heisenberg Matrix The product expression with the Lagrange multiplier z+w is as follows: According to the coefficient matrix and objective function of the correction equations above, the expression of the correction equations is as follows: All variables and constants are substituted into the correction equation to obtain the correction parameters, and the original variables and Lagrange multipliers are updated. If the number of iterations is less than the maximum number of iterations and the complementary gap satisfies the requirement of being less than the gap parameter, the optimal solution is obtained and the calculation is stopped. The optimal photovoltaic unit output, controllable load downward adjustment amount and energy storage output are obtained, and the virtual power plant participating in the AGC frequency regulation strategy is obtained.

8. The frequency modulation method according to claim 1, wherein: The adjustment deviation is the difference between the current real-time load of all resources involved in the regulation and the target load specified by the regulation demand. If the adjustment deviation is greater than the preset threshold of the regulation target, the real-time deviation correction operation is initiated; The resource screening constraints in the real-time correction operation are that the resource has adjustment margin in the direction of compensating the difference, the time since the last adjustment is greater than the control cycle, it has a second-level response capability, and the cumulative number of adjustments is less than the upper limit of the adjustment frequency.

9. A virtual power plant participates in an AGC frequency modulation system, characterized in that: include: The resource screening module is used to screen eligible adjustable resources according to the actual control constraints, regulation characteristics and configuration of the adjustable resources to form a backup resource pool when receiving the grid AGC frequency regulation command; A demand verification module is used to verify the system operation status and the adjustment capacity range of the backup resource pool to determine whether the current frequency regulation demand is met; A net profit model building module is used to build a net profit model for virtual power plants participating in AGC frequency regulation based on the total revenue from grid subsidies and the total cost of virtual power plants participating in frequency regulation. Constraint construction module, used to construct regulation range constraints, energy storage state of charge constraints and AGC control constraints; The optimization solution module is used to build an optimization model for the virtual power plant participating in the AGC frequency regulation strategy based on the prime number net profit model and use the interior point method to solve it according to the constraints to obtain the optimal regulation strategy; The execution feedback module is used to execute the optimization strategy and monitor the adjustment deviation in real time, and start the real-time correction operation based on the adjustment deviation.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the method for a virtual power plant to participate in AGC frequency regulation according to any one of claims 1 to 8 is implemented.

Images