Wind-solar-storage multi-target self-cruising regulation and control method for low-carbon park

Abstract

The invention discloses a wind and light storage multi-target self-cruise control method for a low-carbon park, and belongs to the technical field of new energy microgrid intelligent control. The method comprises the steps of integrating wind power generation, photovoltaic power generation, an energy storage technology and a load unit, collecting wind and light generation power data and load data, stabilizing wind / light intermittency and volatility, and improving energy utilization efficiency; and establishing a micro-grid economic operation model and a flexible interaction mechanism with the power distribution network by combining wind-light output prediction and load prediction results, and dynamically executing a multi-target self-cruise regulation and control instruction according to real-time working condition data. According to the method, the dependence of the park on a traditional power grid is remarkably reduced, the wind-solar-storage cooperative operation efficiency is remarkably improved, the energy consumption cost and carbon emission are effectively reduced, and the realization of a low-carbon target is facilitated.

Description

[0001] This invention belongs to the field of intelligent control technology for new energy microgrids, specifically relating to a multi-objective self-cruising control method for wind, solar and energy storage in low-carbon industrial parks. Background Technology

[0002] The core challenge facing low-carbon industrial park energy systems today lies in effectively coordinating the contradiction between the efficient utilization of renewable energy and the stable operation of the power grid. Traditional integrated wind-solar-storage systems generally employ fixed-ratio allocation or simple rule-based control strategies, which struggle to cope with the highly random and intermittent characteristics of wind / solar output. For example, when encountering sudden changes in wind speed or rapid cloud movement, conventional PID control methods often produce a 10%-15% power tracking deviation, forcing the system to rely on diesel backup units or emergency support from the main power grid. According to statistics from the National Energy Administration in 2023, the average curtailment rate of electricity in my country's industrial parks due to wind and solar fluctuations reached 18.7%, with over 60% of this wasted energy due to untimely control responses. Existing technologies improve control accuracy through enhanced prediction algorithms, but their single economic objective optimization models cannot simultaneously consider the carbon emission constraints of system operation, resulting in significant limitations in practical applications.

[0003] Regarding grid interaction, existing industrial park microgrids mostly adopt a static power exchange mode, lacking the flexible control capability to adapt to dynamic electricity market pricing. Typical problems include: when participating in demand response, fixed threshold charging and discharging strategies cause energy storage systems to reach their charging and discharging limits prematurely; and when switching between peak and off-peak electricity prices, traditional methods require an adjustment period of more than 30 minutes to establish a new operational balance.

[0004] The insufficient adaptive capability of existing control systems has become a major bottleneck restricting technological development. Traditional methods, such as model predictive control (MPC), require pre-establishing mathematical models for all possible operating conditions. When encountering unmodeled extreme weather or equipment failures, system performance deteriorates sharply. Experimental data shows that under load fluctuations exceeding 25%, the voltage instability probability of conventional MPC methods reaches over 35%. Existing technologies struggle to achieve synergistic optimization of the three objectives of economy, low carbon emissions, and stability, which is precisely the key technical problem that this invention aims to solve. Summary of the Invention

[0005] The purpose of this invention is to address the problems faced by low-carbon industrial parks in constructing high-proportion renewable energy systems, such as the intermittency and volatility of wind / solar power generation, difficulties in multi-objective collaborative optimization, and insufficient grid interaction capabilities. This invention provides a multi-objective self-cruising control method for wind, solar, and energy storage in low-carbon industrial parks.

[0006] The objective of this invention is achieved through the following technical solution: a multi-objective self-cruising control method for wind, solar, and energy storage in low-carbon industrial parks, comprising the following steps: By integrating wind power generation units, photovoltaic power generation units, energy storage units and load units, a wind-solar-storage complementary system is constructed to monitor and collect wind and solar power generation data and load data, thereby mitigating the intermittency and volatility of wind / solar power and improving energy utilization efficiency. Based on the collected wind and solar power generation data and load data, a short-term wind and solar power output prediction model and a load prediction model based on an improved Long Short-Term Memory (LSTM) network are constructed to predict the wind and solar power output data and load data for the next 24 hours. Based on the aforementioned wind and solar power output data and load data, an economic operation optimization model for microgrids is established with the goal of minimizing microgrid operating costs. Dynamically adjust the power exchange strategy based on electricity price signals and grid dispatch instructions, and set power exchange threshold constraints; Based on the aforementioned economic operation optimization model and power exchange strategy (flexible interaction strategy), a multi-resource coordinated regulation objective function is constructed. Based on real-time operating data, multi-target self-cruise control commands are dynamically executed.

[0007] Furthermore, the short-term wind and solar power output prediction model and load prediction model based on the improved Long Short-Term Memory (LSTM) network are specifically as follows: A dual-hidden-layer LSTM structure is adopted, with 128-256 units per layer. The number of LSTM structure units is determined by hyperparameter tuning to ensure feature extraction and prevent overfitting. An attention mechanism is introduced to optimize feature weights, thereby obtaining more key feature information, improving model performance, and reducing prediction error. Input 72 hours of historical data, with a time step of 15 minutes; The output layer uses a linear activation function to adapt to continuous prediction.

[0008] Furthermore, the microgrid economic operation optimization model aimed at minimizing microgrid operating costs is expressed in the following mathematical form: ; in, For the cost of wind and solar power generation, To cover the cost of purchasing and selling electricity, Costs associated with energy storage charging and discharging losses.

[0009] Furthermore, the power exchange strategy is expressed in the following mathematical form: ; in, Let t be the power dispatched by the distribution network. , These represent the maximum and minimum values ​​of the power dispatched by the distribution network at time t, respectively.

[0010] Furthermore, the multi-resource coordinated regulation objective function includes the following objectives: Economic objective: a function aimed at minimizing the operating cost of the microgrid; Stability target: Microgrid frequency deviation ; Low-carbon targets: carbon emission intensity constraints threshold .

[0011] Furthermore, the construction of the multi-resource coordinated regulation objective function includes: Input data includes economic operation model results, flexible interaction strategies, and real-time monitoring data; the real-time monitoring data includes SOC, frequency deviation, and carbon emission intensity. Construct an objective function based on the data; The NSGA-II multi-objective optimization algorithm is used to solve the objective function in multiple objectives. Output multi-target coordinated control commands.

[0012] Furthermore, the NSGA-II multi-objective optimization algorithm includes: Multi-objective processing: Simultaneously optimize economic, stability, and low-carbon objectives, maintain solution set diversity through non-dominated sorting and crowding distance, and generate Pareto optimal frontier; Convergence criteria: Hard constraints such as power balance, equipment capacity, and environmental limits must be met.

[0013] Furthermore, the step of dynamically executing multi-target self-cruise control commands based on real-time operating condition data includes: Establish a multi-condition scenario library that includes wind and solar power output fluctuations, load surges, and power grid faults; A rolling optimization and feedback correction mechanism is adopted to update the control instructions every 15 minutes.

[0014] This invention also provides a multi-target autonomous cruise control system for wind, solar, and energy storage, comprising: Data acquisition layer: used for real-time monitoring and acquisition of wind, solar and energy storage unit and load data; Predictive decision layer: used for wind and solar power output and load forecasting, including a short-term wind and solar power output forecasting submodule and a short-term load demand forecasting submodule based on an improved LSTM neural network. Optimization control layer: used to optimize operating costs, including an economic operation model solver and a flexible interactive strategy generator; Execution layer: Used to execute multi-target self-cruise control commands, power conversion devices, and energy storage management systems.

[0015] The beneficial effects of this invention are as follows: By integrating wind power generation, photovoltaic power generation and energy storage technologies, this invention smooths out the intermittency and volatility of wind / solar power and improves energy utilization efficiency; by combining wind and solar power output forecasting and load forecasting results, it establishes a microgrid economic operation model and a flexible interaction mechanism with the distribution network, realizing multi-objective self-cruising control commands, which is suitable for low-carbon scenarios such as industrial parks and science parks, and can significantly improve the efficiency of wind, solar and energy storage coordinated operation, helping to achieve the goal of carbon neutrality. Attached Figure Description

[0016] Figure 1 This is a flowchart of the multi-objective self-cruising control process for wind, solar and energy storage in low-carbon industrial parks according to the present invention. Figure 2 This is a flowchart of the multi-objective coordinated regulation process of the present invention; Figure 3 This is a flowchart of the core steps of the NSGA-II multi-objective optimization algorithm of this invention. Detailed Implementation

[0017] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present invention. Rather, they are merely examples of apparatuses and methods consistent with some aspects of the invention as detailed in the appended claims.

[0018] The terminology used herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The singular forms “a,” “the,” and “the” used in this invention and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used herein refers to and includes any or all possible combinations of one or more of the associated listed items.

[0019] The present invention will now be described in detail with reference to the accompanying drawings. Unless otherwise specified, the features of the following embodiments and implementations can be combined with each other.

[0020] like Figure 1 As shown, this embodiment of the invention provides a multi-objective self-cruising control method for wind, solar, and energy storage in low-carbon industrial parks, comprising the following steps: This invention has been applied to a project. Based on the actual environment of the low-carbon park, a wind-solar-storage complementary system of photovoltaic, wind turbine and energy storage was constructed with an installed capacity of 17KW (1MW wind power, 16MW photovoltaic, 5MW / 20MWh energy storage) and applied to the load in the park. At the same time, a comprehensive energy management platform was built.

[0021] Step 1: Integrate wind power generation units, photovoltaic power generation units, energy storage units, and load units to construct a wind-solar-storage complementary system. The system monitors and collects wind and solar power generation data and load data every 15 minutes through a data acquisition layer, transmitting this data in real time to the central control system. This wind-solar-storage complementarity mitigates the intermittency and volatility of wind / solar power, improving energy utilization efficiency.

[0022] Step 2: Based on the wind and solar power generation data and load data collected by the data acquisition layer, construct a short-term wind and solar power output prediction model and a load prediction model based on an improved Long Short-Term Memory (LSTM) network, and input them into the prediction decision layer to predict the wind and solar power output data and load data for the next 24 hours.

[0023] Step 3: Based on the short-term wind and solar power output forecasts and load forecasts obtained from the forecasting decision-making layer, establish an economic operation optimization model for microgrids with the goal of minimizing microgrid operating costs.

[0024] Step 4: Optimize the control layer to generate a power dispatch strategy (flexible interaction strategy) with the distribution network.

[0025] Step 5: Integrate the aforementioned economic operation optimization model with flexible interaction strategies to construct a multi-resource coordinated regulation objective function.

[0026] Step Six: Based on real-time operating data, dynamically execute multi-target self-cruise control commands through the execution layer.

[0027] In one embodiment, the short-term wind and solar power output prediction model and load prediction model based on the improved Long Short-Term Memory (LSTM) network are as follows: A dual-hidden-layer LSTM structure (128-256 units per layer) is adopted. The number of LSTM structure units is determined by hyperparameter tuning to ensure feature extraction and prevent overfitting.

[0028] To obtain more key feature information, an attention mechanism is introduced to optimize feature weights, improve model performance, and reduce prediction error.

[0029] Enter 72 hours of historical data (time step 15 minutes).

[0030] The output layer uses a linear activation function to adapt to continuous prediction.

[0031] In one embodiment, the economic operation optimization model (cost minimization) is expressed in the following mathematical form: ; in, For the cost of wind and solar power generation, To cover the cost of purchasing and selling electricity, Costs associated with energy storage charging and discharging losses.

[0032] In one embodiment, the power switching strategy is expressed in the following mathematical form: ; in, Let t be the power dispatched by the distribution network. , These represent the maximum and minimum values ​​of the power dispatched by the distribution network at time t, respectively.

[0033] In one embodiment, the multi-resource coordinated regulation objective function includes the following objective: Economic objective: a function aimed at minimizing the operating cost of the microgrid; Stability target: Microgrid frequency deviation ; Low-carbon targets: carbon emission intensity constraints Threshold 0.2 kg / kWh.

[0034] In one embodiment, such as Figure 2 As shown, step five also includes the following sub-steps: 5.1 Input data: Economic operation model results, flexible interaction strategies, and real-time monitoring data (SOC, frequency deviation, carbon emission intensity).

[0035] 5.2 Construct the objective function based on the data.

[0036] 5.3 The NSGA-II multi-objective optimization algorithm is used to solve the multi-objective problem.

[0037] 5.4 Output multi-target coordinated control commands.

[0038] In one embodiment, such as Figure 3 As shown, the NSGA-II multi-objective optimization algorithm includes: Multi-objective processing: Simultaneously optimize economic, stability, and low-carbon objectives, maintain solution set diversity through non-dominated sorting and crowding distance, and generate Pareto optimal frontier; Convergence criteria: Hard constraints such as power balance, equipment capacity, and environmental limits must be met.

[0039] In one embodiment, step six further includes the following sub-steps: 6.1 Data Acquisition and Abnormal Data Monitoring: Collect data such as wind and solar power output, load, and power grid status every 15 minutes.

[0040] 6.2 Scenario matching: Scenario matching is performed based on a multi-condition scenario library (wind and solar fluctuations, load surges, and power grid failures) to enable strategy scheduling.

[0041] 6.3. A rolling optimization and feedback correction mechanism is adopted to solve the multi-objective function, obtain the updated wind, solar and energy storage scheduling instructions, compare them with the actual operation results, and perform feedback correction on the parameters.

[0042] 6.4. Issue and dynamically execute multi-target self-cruise control commands.

[0043] Other embodiments of this application will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This application is intended to cover any variations, uses, or adaptations of this application that follow the general principles of this application and include common knowledge or customary techniques in the art not disclosed herein. The specification and embodiments are to be considered exemplary only.

[0044] It should be understood that this application is not limited to the precise structure described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope.

Claims

1. Integrate wind power generation units, photovoltaic power generation units, energy storage units, and load units to construct a wind-solar-storage complementary system, and monitor and collect wind and solar power generation data and load data; Based on the collected wind and solar power generation data and load data, a short-term wind and solar power output prediction model and a load prediction model based on an improved long short-term memory network are constructed to predict the wind and solar power output data and load data for the next 24 hours. Based on the aforementioned wind and solar power output data and load data, an economic operation optimization model for microgrids is established with the goal of minimizing microgrid operating costs. Dynamically adjust the power exchange strategy based on electricity price signals and grid dispatch instructions, and set power exchange threshold constraints; Based on the aforementioned economic operation optimization model and power exchange strategy, a multi-resource coordinated regulation objective function is constructed. Based on real-time operating data, multi-target self-cruise control commands are dynamically executed.

2. The method according to claim 1, characterized in that, The short-term wind and solar power output prediction model and load prediction model based on the improved long short-term memory network are as follows: A dual-hidden-layer LSTM structure is adopted, with 128-256 units per layer. The number of units in the LSTM structure is determined by hyperparameter tuning to ensure feature extraction and prevent overfitting. An attention mechanism is introduced to optimize feature weights, thereby obtaining more key feature information and reducing prediction error. Input 72 hours of historical data, with a time step of 15 minutes; The output layer uses a linear activation function to adapt to continuous prediction.

3. The method according to claim 1, characterized in that, The microgrid economic operation optimization model, which aims to minimize the operating cost of the microgrid, is expressed in the following mathematical form: ； in, For the cost of wind and solar power generation, To cover the cost of purchasing and selling electricity, Costs associated with energy storage charging and discharging losses.

4. The method according to claim 1, characterized in that, The power exchange strategy is expressed in the following mathematical form: ； in, Let t be the power dispatched by the distribution network. , These represent the maximum and minimum values ​​of the power dispatched by the distribution network at time t, respectively.

5. The method according to claim 1, characterized in that, The multi-resource coordinated regulation objective function includes the following objectives: Economic objective: a function aimed at minimizing the operating cost of the microgrid; Stability target: Microgrid frequency deviation ; Low-carbon targets: carbon emission intensity constraints threshold .

6. The method according to claim 1, characterized in that, The construction of the multi-resource coordinated regulation objective function includes: Input data includes economic operation model results, flexible interaction strategies, and real-time monitoring data; Construct an objective function based on the data; The NSGA-II multi-objective optimization algorithm is used to solve the objective function in multiple objectives. Output multi-target coordinated control commands.

7. The method according to claim 6, characterized in that, The NSGA-II multi-objective optimization algorithm includes: Multi-objective processing: Simultaneously optimize economic, stability, and low-carbon objectives, maintain solution set diversity through non-dominated sorting and crowding distance, and generate Pareto optimal frontier; Convergence criteria: Hard constraints such as power balance, equipment capacity, and environmental limits must be met.

8. The method according to claim 1, characterized in that, The step of dynamically executing multi-target self-cruise control commands based on real-time operating condition data includes: Establish a multi-condition scenario library that includes wind and solar power output fluctuations, load surges, and power grid faults; A rolling optimization and feedback correction mechanism is adopted to update the control instructions every 15 minutes.

9. A wind-solar-storage multi-target autonomous cruise control system for implementing the method of any one of claims 1-8, characterized in that, include: Data acquisition layer: used for real-time monitoring and acquisition of wind, solar and energy storage unit and load data; Predictive decision layer: used for wind and solar power output and load forecasting, including a short-term wind and solar power output forecasting submodule and a short-term load demand forecasting submodule based on an improved LSTM neural network. Optimization control layer: used to optimize operating costs, including an economic operation model solver and a flexible interactive strategy generator; Execution layer: Used to execute multi-target self-cruise control commands, power conversion devices, and energy storage management systems.

Images