Dynamic Optimization Scheduling Method and System for Green Energy Storage Systems

Abstract

This invention provides a dynamic optimization scheduling method and system for green energy storage systems. The method collects operational data from the energy storage system, performs standardization processing and outlier removal to generate a standardized dataset. Based on this dataset, it extracts time-varying characteristics of environmental data, constructs a wind power prediction model, and considers the nonlinear characteristics of the energy storage system's charging and discharging efficiency and dynamic capacity changes to determine a feasible operating range. Combining the prediction model and the feasible range, it establishes a collaborative optimization objective function based on maximizing grid power smoothness, maximizing wind power absorption rate, and optimizing energy storage capacity utilization. A multi-objective optimization algorithm is used to solve the objective function, constructing a constraint set and establishing a state observation equation. Based on real-time data feedback, an adaptive control method is used to dynamically update the control input to achieve optimal scheduling. This invention can significantly improve prediction accuracy and operational efficiency, achieving multi-objective collaborative optimization and possessing significant application value.

Description

Technical Field

[0001] This invention belongs to the field of energy storage system scheduling technology, specifically relating to a dynamic optimization scheduling method and system for green energy storage systems. Background Technology

[0002] With the advancement of energy structure transformation, the installed capacity of renewable energy power generation, represented by wind power, continues to grow rapidly. As an important means of regulation, energy storage systems can effectively improve the absorption capacity of wind power and smooth power fluctuations. However, the intermittent, volatile, and random characteristics of wind power pose a severe challenge to the safe and stable operation of the power system, requiring the development of more advanced energy storage system dispatching technologies.

[0003] Existing wind power forecasting methods are mainly based on static characteristic models, failing to fully consider the dynamic changes in environmental factors, resulting in insufficient forecast accuracy, especially under complex terrain and extreme weather conditions, where forecast errors increase significantly. Traditional energy storage system scheduling strategies often use simplified linear efficiency models, ignoring the nonlinear changes in charging and discharging efficiency with power and temperature, the decay characteristics of energy storage capacity with cycle count and usage time, and the impact of state of charge and depth of discharge on lifetime during actual operation. This leads to a large deviation between actual scheduling results and theoretical calculations. Current energy storage optimization scheduling methods generally adopt a single optimization objective, failing to consider multiple key indicators such as grid power smoothness, wind power absorption rate, and energy storage capacity utilization rate in a coordinated manner. Furthermore, the control strategy is fixed and cannot be adaptively adjusted according to real-time operating conditions.

[0004] Therefore, there is an urgent need for a new method for dynamic optimization and scheduling of green energy storage systems, so that the system can be flexibly adjusted according to its real-time operating status. Summary of the Invention

[0005] The purpose of this invention is to provide a dynamic optimization scheduling method and system for green energy storage systems. By constructing a wind power prediction model with time-varying characteristics, fully considering the nonlinear characteristics and dynamic constraints of the energy storage system, a multi-objective collaborative optimization framework is established, and an adaptive control strategy is adopted to achieve optimal scheduling of the energy storage system.

[0006] In a first aspect, the present invention provides a dynamic optimization scheduling method for green energy storage systems, used to adaptively optimize the grid power smoothness, wind power absorption rate, and energy storage capacity utilization rate in wind farm-based energy storage systems. The method includes the following steps:

[0007] Step 1: Collect operational data from the green energy storage system, standardize the operational data and remove outliers to generate a standardized dataset.

[0008] The operational data includes: wind farm environmental data, wind turbine output, energy storage system parameters, and grid load data; the wind farm environmental data includes: wind speed, wind direction, temperature, air pressure, and turbulence intensity.

[0009] Step 2: Based on the standardized dataset, extract the time-varying features of the environmental data and construct a wind power prediction model with time-varying parameters; considering the nonlinear characteristics of the charging and discharging efficiency of the energy storage system and the dynamic changes in capacity, determine the feasible operating range of the energy storage system according to the system safety constraints.

[0010] Step 3: Combine the wind power prediction model and the feasible operating range of the energy storage system to establish the collaborative optimization objective function of the energy storage system.

[0011] The objective function for the collaborative optimization of the energy storage system is based on three collaborative objectives: maximizing grid power smoothness, maximizing wind power absorption rate, and optimizing energy storage capacity utilization.

[0012] Step four: Use a multi-objective optimization algorithm to solve the collaborative optimization objective function of the energy storage system, construct a set of system operation constraints, establish the state observation equation of the energy storage system, and use an adaptive control method to dynamically update the control input of the energy storage system based on real-time data feedback to achieve optimal scheduling of the energy storage system.

[0013] Furthermore, step two, based on the standardized dataset, extracts time-varying features of environmental data and constructs a wind power prediction model with time-varying parameters, including:

[0014] The time-varying features of the environmental data are extracted as follows: F(t) = X(t)·M(t) + D(t), where t is the time variable in seconds; F(t) is the environmental feature vector at time t; X(t) is the environmental state matrix at time t; M(t) is the time-varying mapping matrix at time t; and D(t) is the environmental disturbance vector at time t.

[0015] The environmental state matrix X(t) at time t is expressed as: X(t) = [v(t), d(t), h(t), p(t), i(t)], where v(t) is the wind speed, in units of: d(t) is the wind direction angle, h(t) is the temperature (°C), p(t) is the air pressure (hPa), and i(t) is the turbulence intensity (%).

[0016] The expression for the transformation matrix M(t) at time t is: M(t) = M(t-1) + α·e(t), where α is the learning rate parameter, with a value range of (0, 1), and e(t) is the feature error vector.

[0017] The expression for the wind power prediction model based on time-varying characteristics is: Pr(t)=F(t)·V(t)·Pb; where Pr(t) is the predicted power, in kW.

[0018] V(t) is the time-varying characteristic influence vector, V(t)=[v1(t),v2(t),v3(t),v4(t),v5(t)], where v1(t),v2(t),v3(t),v4(t),v5(t) represent the wind speed characteristic influence coefficient, wind direction characteristic influence coefficient, temperature characteristic influence coefficient, air pressure characteristic influence coefficient, and turbulence characteristic influence coefficient, respectively.

[0019]

[0020] Where vn is the rated wind speed, d0 is the optimal wind direction angle, h0 is the reference temperature (20℃), kh is the temperature influence coefficient (range 0, 1), p0 is the standard atmospheric pressure, and ki is the turbulence influence coefficient (range 0, 1).

[0021] Furthermore, in step two, considering the nonlinear characteristics of the energy storage system's charge and discharge efficiency and dynamic capacity changes, the feasible operating range of the energy storage system is determined based on system safety constraints, including:

[0022] The nonlinear characteristics of energy storage system efficiency during charging and discharging are described as follows:

[0023]

[0024] Where η0 is the nominal efficiency, Pc(t) is the charging power at time t, Pd(t) is the discharging power at time t, Pn is the rated power, kc and kd are the power attenuation coefficients with values ​​ranging from (0 to 1); and kt is the temperature influence coefficient.

[0025] The dynamic change in the capacity of an energy storage system can be expressed as: C(t) = C0.e ((-kn·N(t))) ·[1-kd·|D(t)-D0|]; where C0 is the initial capacity, kn is the cycle lifetime decay coefficient with a value range of (0, 1), N(t) is the equivalent number of cycles, D(t) is the discharge depth at time t, and D0 is the reference discharge depth.

[0026] Based on the nonlinear efficiency characteristics and dynamic capacity changes of energy storage systems, the safety constraints that the feasible operating range of energy storage systems must meet are:

[0027] The charging and discharging power constraints are: |Pc(t)|≤Pc| and Pd(t)|≤Pd;

[0028] State of charge constraint: SOCm≤SOC(t)≤SOCM| and dSOC(t) / dt|≤SOCr;

[0029] Cyclic lifetime constraint: N(t)≤Nm and Dm≤D(t)≤DM;

[0030] Temperature constraint, hm≤h(t)≤hM;

[0031] Where Pc is the maximum charging power, Pd is the maximum discharging power, SOCm is the minimum state of charge, SOC(t) is the charge at time t, SOCM is the maximum state of charge, SOCr is the maximum rate of change of state of charge; Nm is the maximum number of cycles, Dm is the minimum depth of discharge, DM is the maximum depth of discharge; hm is the minimum operating temperature, and hM is the maximum operating temperature.

[0032] Furthermore, the collaborative optimization objective function of the energy storage system must meet the safety constraints that the feasible operating range of the energy storage system needs to satisfy. The collaborative optimization objective function is based on grid power smoothness, wind power absorption rate and energy storage capacity utilization rate.

[0033] Power grid smoothness index: Where AP(t) is the rate of change of grid power at time t; wind power absorption rate index: Where Pa(t) represents the actual wind power output at time t, in kW; energy storage capacity utilization rate index: Where S0(t) is the optimal state of charge at time t.

[0034] The objective function for coordinated optimization is: L(t)=∑[w1·S(t+k)+w2·U(t+k)+w3·E(t+k)], k=0,1,...,N; where N is the optimization time domain length, w1, w2, w3 are weight coefficients, and w1+w2+w3=1, S(t+k) is the grid power smoothness index predicted in k steps, U(t+k) is the wind power absorption rate index predicted in k steps, and E(t+k) is the energy storage capacity utilization rate index predicted in k steps.

[0035] Furthermore, a multi-objective optimization algorithm is used to solve the collaborative optimization objective function of the energy storage system, and the system operation constraint set is expressed as follows:

[0036] G(x)=g1(x), g2(x), g3(x), g4(x), g5(x)≤0

[0037] Where g1(x), g2(x), g3(x), g4(x), and g5(x) correspond to the safety constraints that the feasible operating range of the energy storage system must satisfy: g1(x): SOCm≤SOC(t)≤SOCM; g2(x): |Pc(t)|≤Pc; g3(x): |Pd(t)|≤Pd; g4(x): g5(x): hm≤h(t)≤hM;

[0038] The state observation equation for the energy storage system is: x(t+1)=A(t)·x(t)+B(t)·u(t)+K(t)·e(t)

[0039] Where x(t) is the system state vector, containing state variables S(t), Pa(t), and P(t), u(t) is the control input vector, containing control variables such as Pc(t) and Pd(t), e(t) is the observation error, K(t) is the adaptive gain matrix, and A(t) and B(t) are time-varying system matrices.

[0040] The control input of the energy storage system is dynamically updated using an adaptive control method.

[0041] u(t) = u0(t) + Δu(t);

[0042]

[0043] Where Δu(t) is the compensation control quantity, and β is the control gain coefficient, with a value range of (0, 1). Let L(t) be the gradient of the objective function with respect to the control variable.

[0044] Secondly, based on the same inventive concept, the present invention provides a dynamic optimization scheduling system for green energy storage systems, used to execute the method described in the first aspect. The scheduling system includes, in sequence, a data acquisition and preprocessing module, a wind power prediction and energy storage system constraint analysis module, an optimization target construction module, and an adaptive optimization control module.

[0045] The data acquisition and preprocessing module is used to collect the operating data of the green energy storage system, and to standardize and remove outliers from the operating data to generate a standardized dataset. The operating data includes wind farm environmental data, wind turbine output, energy storage system parameters, and grid load data. The wind farm environmental data includes wind speed, wind direction, temperature, air pressure, and turbulence intensity.

[0046] The wind power prediction and energy storage system constraint analysis module is used to extract the time-varying characteristics of environmental data based on the standardized dataset, construct a wind power prediction model with time-varying parameters, and determine the feasible operating range of the energy storage system based on the nonlinear characteristics of the charging and discharging efficiency and the dynamic changes in capacity, according to the system safety constraints.

[0047] The optimization objective construction module is used to combine the wind power prediction model and the feasible operating range of the energy storage system to establish a collaborative optimization objective function for the energy storage system based on maximizing grid power smoothness, maximizing wind power absorption rate, and optimizing energy storage capacity utilization.

[0048] The adaptive optimization control module is used to solve the collaborative optimization objective function of the energy storage system using a multi-objective optimization algorithm;

[0049] A set of system operation constraints is constructed, and the state observation equation of the energy storage system is established. Based on real-time data feedback, an adaptive control method is used to dynamically update the control input of the energy storage system, thereby achieving optimal scheduling of the energy storage system.

[0050] The beneficial effects of this invention are as follows:

[0051] This invention constructs a wind power prediction model by introducing a time-varying mapping matrix and an environmental disturbance vector, which can effectively capture the dynamic changes of environmental factors. It uses a feature error vector for online updating, significantly improving the accuracy of wind power prediction. It comprehensively considers the nonlinear characteristics of the charging and discharging efficiency and the dynamic changes in capacity of the energy storage system, and combines multi-dimensional safety constraints to determine the feasible operating range, effectively extending the service life of the energy storage system. Furthermore, it innovatively constructs a collaborative optimization objective function for grid power smoothness, wind power absorption rate, and energy storage capacity utilization rate. By introducing weighting coefficients, it achieves dynamic balance of multiple objectives and uses an adaptive control method to dynamically update the control input, ensuring the robustness and adaptability of system scheduling.

[0052] This invention can improve the smoothness of power grid by 15%-25%, increase the wind power absorption rate by 10%-20%, and increase the energy storage capacity utilization rate by 20%-30%. Attached Figure Description

[0053] Figure 1 This is a flowchart of the dynamic optimization scheduling method for the green energy storage system of the present invention;

[0054] Figure 2 This is a schematic diagram of the dynamic optimization scheduling system of the green energy storage system of the present invention. Detailed Implementation

[0055] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention are described clearly and completely below. Obviously, the described embodiments are only some embodiments of this invention, not all embodiments. Based on the embodiments of this invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0056] Example 1

[0057] like Figure 1 As shown, this invention provides a dynamic optimization scheduling method for green energy storage systems, which is used to adaptively optimize the grid power smoothness, wind power absorption rate, and energy storage capacity utilization rate in wind farm-based energy storage systems.

[0058] The method includes the following steps:

[0059] Step 1: Collect operational data from the green energy storage system, standardize the operational data and remove outliers to generate a standardized dataset.

[0060] Specifically, taking the actual operating data of a wind farm as an example, the collected operating data is processed as follows: the wind speed data is normalized, mapping the wind speed values ​​of 0-30m / s to the 0-1 range; the temperature data is standardized so that its mean is 0 and its variance is 1; outliers are removed using the 3σ criterion, that is, data points that exceed the range of the mean ± 3 times the standard deviation are marked as outliers and removed.

[0061] The operational data includes: wind farm environmental data, wind turbine output, energy storage system parameters, and grid load data; the wind farm environmental data includes: wind speed, wind direction, temperature, air pressure, and turbulence intensity.

[0062] Step 2: Based on the standardized dataset, extract the time-varying features of the environmental data and construct a wind power prediction model with time-varying parameters; considering the nonlinear characteristics of the charging and discharging efficiency of the energy storage system and the dynamic changes in capacity, determine the feasible operating range of the energy storage system according to the system safety constraints.

[0063] Based on the standardized dataset, time-varying features of environmental data are extracted, and a wind power prediction model with time-varying parameters is constructed, including:

[0064] The time-varying features of the environmental data are extracted as follows: F(t) = X(t)·M(t) + D(t), where t is the time variable in seconds; F(t) is the environmental feature vector at time t; X(t) is the environmental state matrix at time t; M(t) is the time-varying mapping matrix at time t; and D(t) is the environmental disturbance vector at time t.

[0065] The environmental state matrix X(t) at time t is expressed as: X(t) = [v(t), d(t), h(t), p(t), i(t)], where v(t) is the wind speed, in units of: d(t) is the wind direction angle, h(t) is the temperature (°C), p(t) is the air pressure (hPa), and i(t) is the turbulence intensity (%).

[0066] The expression for the transformation matrix M(t) at time t is: M(t) = M(t-1) + α·e(t), where α is the learning rate parameter, with a value range of (Θ, 1), and e(t) is the feature error vector.

[0067] The expression for the wind power prediction model based on time-varying characteristics is: Pr(t)=F(t)·V(t)·Pb; where Pr(t) is the predicted power, in kW.

[0068] V(t) is the time-varying characteristic influence vector, V(t)=[v1(t),v2(t),v3(t),v4(t),v5(t)], where v1(t),v2(t),v3(t),v4(t),v5(t) represent the wind speed characteristic influence coefficient, wind direction characteristic influence coefficient, temperature characteristic influence coefficient, air pressure characteristic influence coefficient, and turbulence characteristic influence coefficient, respectively.

[0069]

[0070] Where vn is the rated wind speed, d0 is the optimal wind direction angle, h0 is the reference temperature (20℃), kh is the temperature influence coefficient (range 0, 1), p0 is the standard atmospheric pressure, and ki is the turbulence influence coefficient (range 0, 1).

[0071] Taking actual wind farm data as an example, when the wind speed v(t) = 12 m / s and the rated wind speed vn = 15 m / s, the wind speed characteristic influence coefficient v1(t) = 0.85; when the deviation between the wind direction angle d(t) and the optimal wind direction angle d0 is 30°, the wind direction characteristic influence coefficient v2(t) = 0.866; when the temperature h(t) = 25℃ and the temperature influence coefficient kh = 0.5, the temperature characteristic influence coefficient v3(t) = 0.875.

[0072] Considering the nonlinear characteristics of the charge and discharge efficiency and the dynamic changes in capacity of the energy storage system, the feasible operating range of the energy storage system is determined based on system safety constraints, including:

[0073] The nonlinear characteristics of energy storage system efficiency during charging and discharging are described as follows:

[0074]

[0075] Where η0 is the nominal efficiency, Pc(t) is the charging power at time t, Pd(t) is the discharging power at time t, Pn is the rated power, kc and kd are the power attenuation coefficients, with values ​​ranging from (0 to 1); kt is the temperature influence coefficient. For example, the nominal efficiency of a certain energy storage system is η0 = 0.95. When the charging power is 80% of the rated power, the power attenuation coefficient kc = 0.2, the ambient temperature is 5℃ higher than the reference temperature, and the temperature influence coefficient kt = 0.01, the charging efficiency ηc(t) = 0.92.

[0076] The dynamic change in the capacity of an energy storage system can be expressed as: C(t) = C0·e ((-kn·N(t))) ·[1-kd·|D(t)-D0|]; where C0 is the initial capacity, kn is the cycle lifetime decay coefficient with a value range of (0, 1), N(t) is the equivalent number of cycles, D(t) is the discharge depth at time t, and D0 is the reference discharge depth.

[0077] Based on the nonlinear efficiency characteristics and dynamic capacity changes of energy storage systems, the safety constraints that the feasible operating range of energy storage systems must meet are:

[0078] The charging and discharging power constraints are: |Pc(t)|≤Pc| and Pd(t)|≤Pd;

[0079] State of charge constraint: SOCm≤SOC(t)≤SOCM| and dSOC(t) / dt|≤SOCr;

[0080] Cyclic lifetime constraint: N(t)≤Nm and Dm≤D(t)≤DM;

[0081] Temperature constraint, hm≤h(t)≤hM;

[0082] Where Pc is the maximum charging power, Pd is the maximum discharging power, SOCm is the minimum state of charge, SOC(t) is the charge at time t, SOCM is the maximum state of charge, SOCr is the maximum rate of change of state of charge; Nm is the maximum number of cycles, Dm is the minimum depth of discharge, DM is the maximum depth of discharge; hm is the minimum operating temperature, and hM is the maximum operating temperature.

[0083] Step 3: Combine the wind power prediction model and the feasible operating range of the energy storage system to establish the collaborative optimization objective function of the energy storage system.

[0084] The objective function for the collaborative optimization of the energy storage system is based on three collaborative objectives: maximizing grid power smoothness, maximizing wind power absorption rate, and optimizing energy storage capacity utilization.

[0085] The collaborative optimization objective function of the energy storage system must meet the safety constraints that the feasible operating range of the energy storage system must satisfy. The collaborative optimization objective function is based on grid power smoothness, wind power absorption rate and energy storage capacity utilization rate.

[0086] Power grid smoothness index: Where AP(t) is the rate of change of grid power at time t; wind power absorption rate index: Where Pa(t) represents the actual wind power output at time t, in kW; energy storage capacity utilization rate index: Where S0(t) is the optimal state of charge at time t.

[0087] The objective function for coordinated optimization is: L(t)=∑[w1·S(t+k)+w2·U(t+k)+w3·E(t+k)], k=0,1,...,N; where N is the optimization time domain length, w1, w2, w3 are weight coefficients, and w1+w2+w3=1, S(t+k) is the grid power smoothness index predicted in k steps, U(t+k) is the wind power absorption rate index predicted in k steps, and E(t+k) is the energy storage capacity utilization rate index predicted in k steps.

[0088] Step four: Use a multi-objective optimization algorithm to solve the collaborative optimization objective function of the energy storage system, construct a set of system operation constraints, establish the state observation equation of the energy storage system, and use an adaptive control method to dynamically update the control input of the energy storage system based on real-time data feedback to achieve optimal scheduling of the energy storage system.

[0089] A multi-objective optimization algorithm is used to solve the collaborative optimization objective function of the energy storage system. The system operation constraint set is expressed as follows:

[0090] G(x) = g1(x), g g2(x), g g3(x), g g4(x), g5(x)≤0

[0091] Where g1(x), g2(x), g3(x), g4(x), and g5(x) correspond to the safety constraints that the feasible operating range of the energy storage system must satisfy: g1(x): SOCm≤SOC(t)≤SOCM; g2(x): |Pc(t)|≤Pc; g3(x): |Pd(t)|≤Pd; g4(x): g5(x): hm≤h(t)≤hM;

[0092] The state observation equation for the energy storage system is: x(t+1)=A(t)·x(t)+B(t)·u(t)+K(t)·e(t)

[0093] Where x(t) is the system state vector, containing state variables S(t), Pa(t), and P(t), u(t) is the control input vector, containing control variables such as Pc(t) and Pd(t), e(t) is the observation error, K(t) is the adaptive gain matrix, and A(t) and B(t) are time-varying system matrices.

[0094] The control input of the energy storage system is dynamically updated using an adaptive control method.

[0095] u(t) = u0(t) + Δu(t);

[0096]

[0097] Where Δu(t) is the compensation control quantity, and β is the control gain coefficient, with a value range of (0, 1). Let L(t) be the gradient of the objective function with respect to the control variable.

[0098] To verify the effectiveness of this invention, a practical application test was conducted on an energy storage system from a wind farm. The wind farm has an installed capacity of 100MW, and the associated energy storage system has a capacity of 20MWh. The test period was from January 2023 to December 2023. Specific experimental data and results are as follows:

[0099] Wind farm environmental conditions: annual average wind speed 8.5 m / s, prevailing wind direction northwest, annual average temperature 15℃, average turbulence intensity 12%;

[0100] Energy storage system parameters: rated power 10MW, nominal efficiency 0.95, reference depth of discharge 0.8, cycle life 10,000 times;

[0101] System weight parameters: w 1 = 0.4, w2 = 0.3, w3 = 0.3, optimized time domain length N = 24.

[0102] By comparing the system operation data before and after adopting the method of this invention:

[0103] 1) Power grid smoothness: The original average fluctuation rate was 22.5%, which was reduced to 16.8% after adopting this method, representing an improvement of 25.3%;

[0104] 2) Wind power absorption rate: The original average absorption rate was 82.3%, which was increased to 96.8% after adopting this method, an improvement of 17.6%;

[0105] 3) Energy storage capacity utilization rate: The original average utilization rate was 65.2%, which was increased to 83.1% after adopting this method, representing an improvement of 27.5%.

[0106] Based on the above operational results, the annual wind power grid connection can be increased by approximately 18.5 million kWh, which translates to an economic benefit of approximately 5.55 million yuan at a price of 0.3 yuan / kWh. Simultaneously, due to the extended lifespan of the energy storage system, it is estimated that equipment replacement costs can be saved by approximately 2 million yuan. The above verification cases fully demonstrate the significant effects of this invention in improving grid power smoothness, wind power absorption rate, and energy storage capacity utilization.

[0107] Example 2

[0108] like Figure 2 The diagram shows the composition of the dynamic optimization scheduling system for the green energy storage system of the present invention, which is used to execute the method of Embodiment 1. The scheduling system includes, in sequence, a data acquisition and preprocessing module, a wind power prediction and energy storage system constraint analysis module, an optimization target construction module, and an adaptive optimization control module.

[0109] This module adopts a distributed data acquisition system, which includes multiple field data acquisition units with a sampling period of 1 second. It transmits data through industrial Ethernet to achieve real-time acquisition and processing of wind farm operation data.

[0110] The data acquisition and preprocessing module is used to collect the operating data of the green energy storage system, and to standardize and remove outliers from the operating data to generate a standardized dataset. The operating data includes wind farm environmental data, wind turbine output, energy storage system parameters, and grid load data. The wind farm environmental data includes wind speed, wind direction, temperature, air pressure, and turbulence intensity.

[0111] The wind power prediction and energy storage system constraint analysis module is used to extract time-varying features of environmental data based on the standardized dataset and construct a wind power prediction model with time-varying parameters. Considering the nonlinear characteristics of the energy storage system's charging and discharging efficiency and dynamic capacity changes, it determines the feasible operating range of the energy storage system according to system safety constraints. This module adopts a rolling prediction method with a prediction timescale of 48 hours and a prediction step size of 15 minutes. By updating the time-varying mapping matrix M(t) in real time, the prediction model can adapt to environmental changes. In practical applications, the prediction accuracy of this module can reach 85%-90%, providing reliable data support for the optimized scheduling of energy storage systems.

[0112] The optimization objective construction module is used to combine the wind power prediction model and the feasible operating range of the energy storage system to establish a collaborative optimization objective function for the energy storage system based on maximizing grid power smoothness, maximizing wind power absorption rate, and optimizing energy storage capacity utilization.

[0113] The adaptive optimization control module is used to solve the collaborative optimization objective function of the energy storage system using a multi-objective optimization algorithm; it constructs a set of system operating constraints, establishes the state observation equation of the energy storage system, and dynamically updates the control input of the energy storage system using an adaptive control method based on real-time data feedback to achieve optimal scheduling of the energy storage system. This module is based on the principle of model predictive control (MPC) with a control cycle of 1 minute. Within each control cycle, the current system state is first obtained based on the state observation equation, then the optimal control sequence is obtained by solving the multi-objective optimization problem, and finally the control input for the current moment is executed. For example, when the state of charge (SOC(t)) of the energy storage system deviates from the target value, the control module automatically adjusts the charging and discharging power to gradually bring the system towards its optimal operating state.

[0114] In practical applications, the weighting coefficients can be dynamically adjusted according to the grid operation requirements and the characteristics of the energy storage system. For example, when the grid fluctuates greatly, the power smoothness weight w1 can be increased; when the wind power output is high, the absorption rate weight w2 can be increased; and when the energy storage capacity is close to its limit, the capacity utilization weight w3 can be increased.

[0115] The various functional modules interact with each other via an industrial communication network, and the distributed architecture improves the system's reliability and real-time performance. The system is configured with dual-machine hot backup; when the primary controller fails, the backup controller can complete a seamless switchover within 100ms, ensuring the continuous and stable operation of the energy storage system.

[0116] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. A dynamic optimization scheduling method for green energy storage systems, used for adaptive optimization of grid power smoothness, wind power absorption rate, and energy storage capacity utilization rate in wind farm-based energy storage systems, characterized in that... The method includes the following steps: Step 1: Collect operational data from the green energy storage system, standardize the operational data and remove outliers to generate a standardized dataset; The operational data includes: wind farm environmental data, wind turbine output, energy storage system parameters, and grid load data; the wind farm environmental data includes: wind speed, wind direction, temperature, air pressure, and turbulence intensity. Step 2: Based on the standardized dataset, extract the time-varying features of the environmental data and construct a wind power prediction model with time-varying parameters; considering the nonlinear characteristics of the charging and discharging efficiency of the energy storage system and the dynamic changes in capacity, determine the feasible operating range of the energy storage system according to the system safety constraints. Extracting time-varying features from environmental data. ,in, For time variables, the unit is seconds (s). for Environmental feature vector at any given time; for The environment state matrix at time t, for The time-varying mapping matrix at time t, D(t) is The environmental disturbance vector at time; Environment state matrix at time step The expression is: ,in, Wind speed, unit: ; Wind direction angle Temperature, unit: ; Pressure is measured in atmospheres, in units of: ; Turbulence intensity, unit: ; Time-mapping matrix The expression is: ,in, The learning rate parameter takes values ​​in the range (0, 1). The feature error vector; The expression for the wind power prediction model based on time-varying characteristics is as follows: ;in, For predicted power, unit: kW; This is the time-varying feature influence vector. , , , , , These represent the influence coefficients of wind speed, wind direction, temperature, air pressure, and turbulence, respectively. ； in, Rated wind speed, For the optimal wind direction angle, For reference temperature, 20℃ is used. This is the temperature influence coefficient, with a value range of (0, 1). Standard atmospheric pressure The turbulence influence coefficient has a value range of (0, 1). Step 3: Combine the wind power prediction model and the feasible operating range of the energy storage system to establish the collaborative optimization objective function of the energy storage system; The energy storage system collaborative optimization objective function is based on three collaborative objectives: maximizing grid power smoothness, maximizing wind power absorption rate, and optimizing energy storage capacity utilization. Step four: Use a multi-objective optimization algorithm to solve the collaborative optimization objective function of the energy storage system, construct a set of system operation constraints, establish the state observation equation of the energy storage system, and use an adaptive control method to dynamically update the control input of the energy storage system based on real-time data feedback to achieve optimal scheduling of the energy storage system.

2. The dynamic optimization scheduling method for green energy storage systems according to claim 1, characterized in that, Step two considers the nonlinear characteristics of the energy storage system's charge and discharge efficiency and dynamic capacity changes. Based on system safety constraints, the feasible operating range of the energy storage system is determined as follows: The nonlinear characteristics of energy storage system efficiency during charging and discharging are described as follows: ； ； in, For nominal efficiency, The charging power at time t, Let be the discharge power at time t. Rated power, , The power attenuation coefficient has a value range of (0, 1). This is the temperature influence coefficient; The dynamic changes in the capacity of an energy storage system can be described as follows: ;in, For the initial capacity, This is the cycle life decay coefficient, with a value range of (0, 1). For the equivalent number of iterations, Let t be the discharge depth. For reference discharge depth; Based on the nonlinear efficiency characteristics and dynamic capacity changes of energy storage systems, the safety constraints that the feasible operating range of energy storage systems must meet are: Charge and discharge power constraints and ; State of charge constraints and ; Cyclic lifetime constraints and ; Temperature constraint, ; in, Maximum charging power, For maximum discharge power, At minimum state of charge, Let be the amount of charge at time t. At maximum state of charge, This represents the maximum rate of change of state of charge. The maximum number of loops. Minimum discharge depth This represents the maximum depth of discharge. Minimum operating temperature, This is the maximum operating temperature.

3. The dynamic optimization scheduling method for green energy storage systems according to claim 2, characterized in that, The collaborative optimization objective function of the energy storage system must meet the safety constraints that the feasible operating range of the energy storage system must satisfy. The collaborative optimization objective function is based on grid power smoothness, wind power absorption rate and energy storage capacity utilization rate. Power grid smoothness index: ;in, for Real-time grid power change rate; wind power absorption rate indicators: ,in, The actual wind power output at time t, in kW; energy storage capacity utilization rate index: ,in, The optimal state of charge at time t; Coordinated optimization objective function: ;in, To optimize the time domain length, , , These are the weighting coefficients, and , For the smoothness index of grid power prediction in k steps, The wind power absorption rate index is the result of k-step prediction. is the energy storage capacity utilization rate index predicted in k steps.

4. The dynamic optimization scheduling method for green energy storage systems according to claim 3, characterized in that, A multi-objective optimization algorithm is used to solve the collaborative optimization objective function of the energy storage system. The system operation constraint set is expressed as follows: ； in , , , and The safety constraints that the feasible operating range of the corresponding energy storage system must meet are: ; ; ; ; ; State observation equations for energy storage systems: ; Where x(t) is the system state vector, containing state variables S(t), Pa(t), and P(t), u(t) is the control input vector, containing control variables Pc(t) and Pd(t), e(t) is the observation error, K(t) is the adaptive gain matrix, and A(t) and B(t) are time-varying system matrices. The control input of the energy storage system is dynamically updated using an adaptive control method. ； ； Where Δu(t) is the compensation control quantity, β is the control gain coefficient with a value range of (0, 1), and ∇L(t) is the gradient of the objective function L(t) with respect to the control quantity.

5. A dynamic optimization scheduling system for green energy storage systems, used to execute the method described in any one of claims 1-4, characterized in that, The scheduling system comprises the following components connected in sequence: The data acquisition and preprocessing module is used to collect the operational data of the green energy storage system, and to standardize and remove outliers from the operational data to generate a standardized dataset. The operational data includes wind farm environmental data, wind turbine output, energy storage system parameters, and grid load data. The wind farm environmental data includes wind speed, wind direction, temperature, air pressure, and turbulence intensity. The wind power prediction and energy storage system constraint analysis module is used to extract the time-varying characteristics of environmental data based on the standardized dataset, construct a wind power prediction model with time-varying parameters, and determine the feasible operating range of the energy storage system based on the nonlinear characteristics of the charging and discharging efficiency and the dynamic changes in capacity, according to the system safety constraints. An optimization objective construction module is used to combine the wind power prediction model and the feasible operating range of the energy storage system to establish a collaborative optimization objective function for the energy storage system based on maximizing grid power smoothness, maximizing wind power absorption rate, and optimizing energy storage capacity utilization. The adaptive optimization control module is used to solve the collaborative optimization objective function of the energy storage system using a multi-objective optimization algorithm; A set of system operation constraints is constructed, and the state observation equation of the energy storage system is established. Based on real-time data feedback, an adaptive control method is used to dynamically update the control input of the energy storage system, thereby achieving optimal scheduling of the energy storage system.

Images