An ai optimization-based multi-objective collaborative control method for a wind-solar complementary calcium cycle thermochemical energy storage system

Abstract

The application discloses a kind of wind and light complementary calcium cycle thermochemical energy storage system multi-objective collaborative control method based on AI optimization, it is related to renewable energy storage technical field, the reversible carbonation reaction of calcium-based material (CaO / CaCO3) is used to realize the high-density storage and release of heat energy, the space-time prediction model of wind power, photovoltaic output is constructed by random forest algorithm, and the global optimization of system multi-objective is realized by combining improved particle swarm optimization algorithm.Compared with existing electrochemical energy storage and physical energy storage technology, the calcium cycle thermochemical energy storage of the application has the significant advantages of low energy storage cost, long energy storage time, long cycle life, abundant raw materials and the like.The results show that the annual power generation of the system after optimization reaches 223.67GWh, the cost of electricity per kilowatt-hour is reduced to 0.35 yuan / kWh, the annual carbon emission reduction is 167800 tons, and the investment recovery period is 7.34 years.The application provides an economical and efficient long-time energy storage solution for high-proportion renewable energy grid connection.

Description

Technical Field

[0001] This invention relates to the field of renewable energy storage technology, specifically to a multi-objective collaborative control method for a wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization. Background Technology

[0002] With the global energy structure transitioning towards low-carbon development and the deepening of the "dual-carbon" strategy, the installed capacity of renewable energy sources such as wind and solar power continues to grow rapidly. However, the inherent intermittency, volatility, and unpredictability of wind and solar resources pose severe challenges to the safe and stable operation of the power system and its high-proportion grid integration. Energy storage technology, as a key support for mitigating renewable energy fluctuations and improving grid flexibility and reliability, is becoming increasingly important. Currently, mainstream energy storage technologies mainly include electrochemical energy storage (such as lithium-ion batteries and flow batteries), physical energy storage (such as pumped hydro storage and compressed air energy storage), and traditional thermal energy storage (such as molten salt thermal energy storage). Among these, electrochemical energy storage has advantages such as fast response speed and high energy density, but suffers from high cost, limited cycle life, and thermal runaway safety risks; physical energy storage, while technologically mature and with large capacity, is severely constrained by geographical conditions, has a long construction cycle, and relatively low efficiency; traditional thermal energy storage generally suffers from low storage temperature, low energy density, and high medium cost. In recent years, calcium cycle thermochemical energy storage technology has gradually become a research hotspot in the field of long-term energy storage due to its advantages such as high energy density (theoretical energy density can reach 3.1 MJ / kg), wide availability and low price of materials, long cycle life (more than 20,000 cycles), and environmental friendliness.

[0003] Although some optimized control methods for wind-solar-storage complementary systems have been proposed in existing technologies, the following technical shortcomings still exist: First, existing methods generally use hydropower, lithium batteries, supercapacitors, or hydrogen energy storage as the core regulation body, without deep coupling with calcium cycle thermochemical energy storage systems. They lack system modeling methods for the endothermic and exothermic reaction characteristics of calcium-based materials and the thermo-electro-chemical energy conversion mechanism, making it difficult to fully leverage the high-density and long-term energy storage advantages of calcium cycle energy storage. Second, existing technologies lack an integrated collaborative control framework between wind and solar power output prediction and system operation optimization, often employing static or offline optimization strategies. This makes it difficult to cope with rapid dynamic changes in wind and solar resources and load demand, resulting in lag in real-time regulation and insufficient wind and solar power absorption capacity. Furthermore, calcium cycle thermochemical energy storage systems involve the coupling of multiple parameters such as decomposition reaction temperature, carbonation reaction temperature, CO2 partial pressure, waste heat recovery rate, and cycle decay. Existing single-objective or simplified multi-objective optimization methods struggle to achieve a globally optimal trade-off between multiple conflicting economic and technical indicators such as investment payback period, energy utilization rate, cost per kilowatt-hour, and carbon emission reduction. The overall economic efficiency and energy efficiency of the system still have significant room for improvement. Therefore, there is an urgent need to develop an integrated collaborative control method that can combine high-precision wind and solar forecasting, dynamic modeling of calcium cycle thermochemical energy storage, and multi-objective intelligent optimization to solve the comprehensive technical bottlenecks of existing wind, solar, and energy storage systems in terms of cost, efficiency, lifespan, and dynamic response. Summary of the Invention

[0004] The purpose of this invention is to provide a multi-objective collaborative control method for a wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization, so as to solve the problems existing in the prior art mentioned in the background.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A multi-objective collaborative control method for a wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization includes the following steps: S1: Collect wind and solar resource data and load demand data to construct a training dataset; the wind and solar resource data includes wind speed, solar radiation intensity, ambient temperature and relative humidity; the load demand data includes grid-side load curves, industrial electricity demand and residential electricity demand; S2: A wind and solar power output prediction model is constructed using the random forest algorithm; S3: Establish a mathematical and physical model of a calcium cycle thermochemical energy storage system, wherein the calcium cycle thermochemical energy storage system uses calcium-based materials CaO and CaCO3 as energy storage media and realizes the storage and release of thermal energy through a reversible carbonation reaction; S4: Construct a multi-objective optimization function, the objective function expression is:

[0006] in, To optimize the variable vector, For the investment payback period target, For energy efficiency targets, To achieve the cost per kilowatt-hour target, To achieve carbon emission reduction targets; S5: An improved particle swarm optimization algorithm is used to globally optimize system parameters. The optimization variables include: wind power ratio, unit investment, energy storage efficiency, energy storage duration, reaction temperature, and waste heat recovery rate. S6: Outputs the optimal parameter configuration scheme, including the ratio of wind and solar installed capacity, the capacity of the calcium cycle energy storage system, the reaction temperature setting, the operation strategy, and generates a system performance evaluation report.

[0007] Preferably, in step S2, the method for constructing the wind and solar power output prediction model specifically includes the following steps: S21: Preprocess the collected wind and solar resource data, including outlier removal, missing value filling, normalization, and outlier detection using the Z-score method; S22: Use the Bootstrap autosampler to randomly sample K subsets from the original dataset; S23: Construct a CART decision tree for each subset of data, using the Gini coefficient (Gini Impurity) as the splitting criterion when splitting nodes.

[0008] in, The proportion of category i in node D; S24: Integrate the prediction results of K CART decision trees, and use the average value method for regression problems:

[0009] in, This represents the prediction result for the i-th tree; S25: Use 5-fold cross-validation to evaluate model performance, and calculate the coefficient of determination R² and root mean square error RMSE.

[0010]

[0011] Where N is the total number of samples involved in the calculation.

[0012] Preferably, in step S3, the steps for establishing the mathematical and physical model of the calcium cycle thermochemical energy storage system are as follows: S31: Decomposition reaction kinetic model:

[0013] in, Let A be the pre-exponential factor, representing the decomposition reaction rate. Let R be the activation energy, R be the gas constant, and T be the reaction temperature. Here, n represents the conversion rate, and n represents the reaction order. S32: Kinetic model of carbonation reaction:

[0014] in, The carbonation reaction rate, Pre-exponential factor, For activation energy, The reaction temperature, Here, m represents the partial pressure of CO2, and m is the reaction order. Conversion rate; S33: Energy storage efficiency calculation model:

[0015] in, As a benchmark efficiency, This is a temperature correction factor. For sulfur resistance coefficient, This is the cyclic decay coefficient; S34: Temperature correction factor model, based on third-order polynomial fitting:

[0016] Where T is the reaction temperature. ; S35: Reaction heat calculation model:

[0017] in, The heat of reaction, This represents the number of moles of calcium carbonate. H For the enthalpy of reaction, This refers to the reaction efficiency.

[0018] Preferably, in step S4, the specific process of constructing the multi-objective optimization function is as follows: S41: Investment recovery period target:

[0019] in, For the investment recovery period, For the minimum investment payback period, To maximize the investment recovery period, all items are normalized. Formula for calculating payback period:

[0020] in, Annual power generation For the cost per kilowatt-hour, Unit operating cost d is the initial investment, and d is the discount rate; S42: Energy efficiency target:

[0021] in, For the overall energy utilization rate of the system; Energy efficiency calculation:

[0022] in, This represents the actual amount of electricity generated. To the extent of wasted electricity, Available power generation resources; S43: Cost per kilowatt-hour target:

[0023] Where LCOE is the cost per kilowatt-hour; Cost per kilowatt-hour calculation:

[0024] Where t represents the year, ranging from 1 to 25; S44: Carbon emission reduction targets:

[0025] in, This refers to the annual carbon emission reduction. Carbon emission reduction calculation:

[0026] in, Total power generation For coal-fired power plant emission coefficients, This represents the system emission coefficient.

[0027] Preferably, in step S5, the improved particle swarm optimization algorithm includes: S51: Speed ​​Update Formula:

[0028] in, Let be the velocity of particle i in the t-th iteration. For location, For inertial weights, The acceleration constant, A random number in the range [0,1]. For the individual's historical best position, The globally optimal position; S52: Position Update Formula:

[0029] in, Let i be the position of particle i in the (t+1)th iteration; S53: Linear decreasing inertia weight strategy:

[0030] in, =1.1, =0.3, where t is the current iteration number. This represents the maximum number of iterations. S54: Adaptive Mutation Operation: When the random perturbation is less than the mutation probability, the iteration position changes as follows:

[0031] in, The random perturbation is a Gaussian distribution. For variable asynchronous length; S55: Convergence Criterion: The iteration stops when any of the following conditions are met: (1) Current iteration number ; (2) The global optimal fitness remains unchanged for 50 consecutive iterations, i.e. ; (3) The population standard deviation is less than the threshold, i.e., std(x) < 0.01.

[0032] Preferably, in step S6, the generated system performance evaluation report includes the following indicators: Technical performance indicators: annual power generation, annual curtailment rate, energy utilization rate, energy storage efficiency, and system reliability; Economic performance indicators: total investment cost, cost per kilowatt-hour, payback period, internal rate of return, and net present value; Environmental benefit indicators: annual carbon emission reduction, emission reduction costs, and carbon trading revenue.

[0033] A wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization includes the following modules: Wind and solar power generation module: Composed of a wind turbine generator and a photovoltaic array, used to convert wind energy and solar energy into electrical energy; Calcium cycle thermochemical energy storage module: It consists of a decomposition furnace, a carbonation reactor, a heat storage tank, a cold storage tank, and a waste heat recovery system, and is used for the conversion and storage of electrical energy and thermal energy. The intelligent decision control module consists of a data acquisition unit, a prediction unit, an optimization unit, and a decision unit. It is used to collect data in real time, predict power generation output, optimize operating parameters, and generate control commands. Energy management module: Composed of power converter, energy router, grid connection interface and load dispatching system, used for power distribution, conversion, grid connection and load management.

[0034] Preferably, the configuration parameters of the wind and solar power generation module are as follows: Installed capacity is ,in The optimization range is 0.3-0.7 for the proportion of wind power; the cut-in wind speed is... Rated wind speed is Cut-out wind speed is The capacity factor is The power curve model is ,in The air density is 1.225 kg / m³, and A is the swept area. The wind energy utilization factor is 0.4; Photovoltaic modules: installed capacity is The standard test conditions are STC = 1000 W / m², 25℃; the photoelectric conversion efficiency is... Temperature coefficient is The capacity factor is .

[0035] Preferably, the core equipment parameters of the calcium cycle thermochemical energy storage module are: The decomposition furnace adopts a circulating fluidized bed, and its processing capacity can reach 500 t / d CaCO3; the operating temperature is set to... Operating pressure is Decomposition efficiency is Energy consumption is ; The carbonation reactor is a moving bed reactor with a processing capacity of 500 t / d CaO; the operating temperature is set to... Operating pressure is Carbonation efficiency is The heat of reaction is released as ; The thermal storage tank and cold storage tank are selected as insulated storage tanks; the energy storage medium is CaO / CaCO3 particles; the energy storage density is... Energy storage capacity is The heat loss rate is ; The waste heat recovery system originates from exhaust gas from the decomposition furnace and heat dissipation from the carbonation reactor; the waste heat recovery rate is... Waste heat power generation efficiency is Waste heat power generation is .

[0036] Compared with the prior art, the beneficial effects of the present invention are: 1. This invention constructs a high-precision wind and solar power output prediction model using the random forest algorithm, and combines it with an improved particle swarm optimization algorithm to perform multi-objective global optimization of the calcium cycle thermochemical energy storage system, thereby achieving efficient absorption of wind and solar resources and dynamic matching of the energy storage system. Compared with traditional electrochemical or physical energy storage schemes, this invention makes full use of the high energy storage density characteristics of the reversible carbonation reaction of calcium-based materials, effectively solving the spatiotemporal mismatch between the intermittency of wind and solar power output and the grid load demand.

[0037] 2. This invention constructs a multi-objective optimization function that includes investment payback period, energy utilization rate, cost per kilowatt-hour, and carbon emission reduction. By improving the global optimization capability of the particle swarm optimization algorithm, the optimal balance between system economy and environmental benefits is achieved. At the same time, the cost of calcium cycle energy storage medium is only 1 / 10 of that of lithium battery, and the cycle life is 5-10 times that of lithium battery, which greatly reduces the total life cycle cost of the system. Attached Figure Description

[0038] Figure 1 This is a diagram illustrating the overall technical roadmap of the present invention.

[0039] Figure 2 This is a detailed flowchart of the calcium cycle energy storage system of the present invention.

[0040] Figure 3 This is a detailed calculation flowchart of an embodiment of the present invention.

[0041] Figure 4 This is a bar chart comparing the power generation of different modules in this invention.

[0042] Figure 5 This is a comparison chart of annual seasonal power generation according to the present invention.

[0043] Figure 6 A figure showing the performance comparison of the system before and after optimization.

[0044] Figure 7 Two figures show the performance comparison of the system before and after optimization.

[0045] Figure 8 This is a diagram showing the composition of the total investment in this invention.

[0046] Figure 9 This is a temperature-efficiency curve of the calcium cycle energy storage system of the present invention. Detailed Implementation

[0047] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0048] Please see Figure 1-9 The present invention provides the following technical solutions: A multi-objective collaborative control method for a wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization includes the following steps: S1: Collect wind and solar resource data and load demand data to build a training dataset; wind and solar resource data include wind speed, solar radiation intensity, ambient temperature and relative humidity; load demand data includes grid-side load curves, industrial electricity demand and residential electricity demand; data collection frequency is 1 hour, and the time span covers at least one full year.

[0049] S2: A wind and solar power generation output prediction model was constructed using the random forest algorithm. The model parameters were set as follows: number of decision trees N=100, minimum number of leaf node samples S_min=10, feature sampling method "all", and maximum depth D_max=20. The model performance was evaluated using 5-fold cross-validation, with a determination coefficient R²>0.85 and a root mean square error RMSE<0.06.

[0050] The method for constructing a wind and solar power output prediction model specifically includes the following steps: S21: Preprocess the collected wind and solar resource data, including outlier removal, missing value imputation, and normalization. Use the Z-score method for outlier detection, with a threshold set to... ; S22: Use the Bootstrap method to randomly sample K=100 subsets from the original dataset, each subset containing 63.2% of the original dataset samples; S23: Construct a CART decision tree for each subset of data, using the Gini Impurity coefficient as the splitting criterion when splitting nodes:

[0051] in, The proportion of category i in node D; S24: Integrate the prediction results of 100 CART decision trees, and use the average method for regression problems:

[0052] in, This represents the prediction result for the i-th tree; S25: Use 5-fold cross-validation to evaluate model performance, and calculate the coefficient of determination R² and root mean square error RMSE.

[0053]

[0054] Where N is the total number of samples involved in the calculation.

[0055] S3: Establish a mathematical and physical model for a calcium cycle thermochemical energy storage system. This system uses calcium-based materials CaO and CaCO3 as energy storage media, and achieves the storage and release of thermal energy through a reversible carbonation reaction. The core reaction equation is:

[0056] The forward reaction (calcium carbonate decomposition) is an endothermic reaction that occurs in a decomposition furnace at a temperature of 850-900℃; the reverse reaction (calcium oxide carbonation) is an exothermic reaction that occurs in a carbonation reactor at a temperature of 650-750℃.

[0057] The steps for establishing the mathematical and physical model of the calcium cycle thermochemical energy storage system are as follows: S31: Decomposition reaction kinetic model:

[0058] in, Let A be the decomposition reaction rate (mol / (kg·s)) and A be the pre-exponential factor. , where is the activation energy (J / mol), R is the gas constant 8.314 J / (mol·K), and T is the reaction temperature (K). Here, n represents the conversion rate, and n represents the reaction order. S32: Kinetic model of carbonation reaction:

[0059] in, The carbonation reaction rate, Pre-exponential factor, For activation energy, The reaction temperature. Here, m represents the partial pressure of CO2, and m is the reaction order. Conversion rate; S33: Energy storage efficiency calculation model:

[0060] in, As a benchmark efficiency, This is a temperature correction factor. For sulfur resistance coefficient, This is the cyclic decay coefficient; S34: Temperature correction factor model, based on third-order polynomial fitting:

[0061] Where T is the reaction temperature (°C). ; S35: Reaction heat calculation model:

[0062] in, The heat of reaction, This represents the number of moles of calcium carbonate. H For the enthalpy of reaction, This represents the reaction efficiency.

[0063] S4: Construct a multi-objective optimization function, the objective function expression is:

[0064] in, To optimize the variable vector, For the investment payback period target, For energy efficiency targets, To achieve the cost per kilowatt-hour target, The target is to reduce carbon emissions.

[0065] The specific process of constructing a multi-objective optimization function is as follows: S41: Investment recovery period target:

[0066] in, For the investment recovery period, =5 years, =10 years, and all items are normalized; Formula for calculating payback period:

[0067] in, Annual power generation For the cost per kilowatt-hour, Unit operating cost d represents the initial investment, and d is the discount rate of 8%. S42: Energy efficiency target:

[0068] in, For the overall energy utilization rate of the system; Energy efficiency calculation:

[0069] in, This represents the actual amount of electricity generated. To the extent of wasted electricity, Available power generation resources; S43: Cost per kilowatt-hour target:

[0070] Wherein, LCOE is the cost per kilowatt-hour (RMB / kWh); Cost per kilowatt-hour calculation:

[0071] Where t represents the year, ranging from 1 to 25; S44: Carbon emission reduction targets:

[0072] in, This refers to the annual carbon emission reduction. Carbon emission reduction calculation:

[0073] in, Total power generation =0.8 kg / kWh is the emission coefficient for coal-fired power plants. =0.05 kg / kWh is the system emission factor.

[0074] S5: An improved particle swarm optimization algorithm is used to globally optimize system parameters. Optimization variables include: wind power ratio, unit investment (yuan / kW), energy storage efficiency, energy storage duration (h), reaction temperature (°C), and waste heat recovery rate. Algorithm parameters are set as follows: population size M=200, maximum number of iterations T_max=600, and inertia weight. The acceleration constant is 1.5.

[0075] Improved particle swarm optimization algorithms include: S51: Speed ​​Update Formula:

[0076] in, Let be the velocity of particle i in the t-th iteration. For location, For inertial weights, The acceleration constant, A random number in the range [0,1]. For the individual's historical best position, The globally optimal position; S52: Position Update Formula:

[0077] in, Let i be the position of particle i in the (t+1)th iteration; S53: Linear decreasing inertia weight strategy:

[0078] in, =1.1, =0.3, where t is the current iteration number. This represents the maximum number of iterations. S54: Adaptive Mutation Operation: When the random perturbation is less than the mutation probability, the iteration position changes as follows:

[0079] in, The random perturbation is a Gaussian distribution. For variable asynchronous length; S55: Convergence Criterion: The iteration stops when any of the following conditions are met: (1) Current iteration number ; (2) The global optimal fitness remains unchanged for 50 consecutive iterations, i.e. ; (3) The population standard deviation is less than the threshold, i.e., std(x) < 0.01.

[0080] S6: Outputs the optimal parameter configuration scheme, including the ratio of wind and solar installed capacity, the capacity of the calcium cycle energy storage system, the reaction temperature setting, the operation strategy, and generates a system performance evaluation report.

[0081] The generated system performance evaluation report includes the following metrics: Technical performance indicators: annual power generation, annual curtailment rate, energy utilization rate, energy storage efficiency, and system reliability; Economic performance indicators: total investment cost, cost per kilowatt-hour, payback period, internal rate of return, and net present value; Environmental benefit indicators: annual carbon emission reduction, emission reduction costs, and carbon trading revenue.

[0082] This invention also provides an AI-optimized wind-solar hybrid calcium cycle thermochemical energy storage system, comprising the following modules: Wind and solar power generation module: Composed of a wind turbine generator and a photovoltaic array, used to convert wind energy and solar energy into electrical energy; Calcium cycle thermochemical energy storage module: It consists of a decomposition furnace, a carbonation reactor, a heat storage tank, a cold storage tank, and a waste heat recovery system, and is used for the conversion and storage of electrical energy and thermal energy. The intelligent decision control module consists of a data acquisition unit, a prediction unit, an optimization unit, and a decision unit. It is used to collect data in real time, predict power generation output, optimize operating parameters, and generate control commands. Energy management module: Composed of power converter, energy router, grid connection interface and load dispatching system, used for power distribution, conversion, grid connection and load management.

[0083] The configuration parameters for the wind and solar power generation module are as follows: Installed capacity is ,in The optimization range is 0.3-0.7 for the proportion of wind power; the cut-in wind speed is... Rated wind speed is Cut-out wind speed is The capacity factor is (Average annual utilization hours: 2200h); Power curve model is: ,in The air density is 1.225 kg / m³, and A is the swept area. The wind energy utilization factor is 0.4; Photovoltaic modules: installed capacity is The standard test conditions are STC = 1000 W / m², 25℃; the photoelectric conversion efficiency is... (N-type TOPCon battery); temperature coefficient is The capacity factor is (Average annual utilization hours: 1300h).

[0084] The power model is expressed as:

[0085] Where G is the solar radiation intensity (W / m²) and T is the component temperature (°C).

[0086] The core equipment parameters of the calcium cycle thermochemical energy storage module are as follows: The decomposition furnace adopts a circulating fluidized bed, with a processing capacity of 500 t / d CaCO3; the operating temperature is set to... (Optimized value); Operating pressure is (Atmospheric pressure); Decomposition efficiency is Energy consumption is ; The carbonation reactor uses a moving bed and has a processing capacity of 500 t / d CaO; the operating temperature is set to... Operating pressure is (Pressurized carbonation); carbonation efficiency is The heat of reaction is released as ; The thermal storage tank and cold storage tank are selected as insulated storage tanks; the energy storage medium is CaO / CaCO3 particles; the energy storage density is... Energy storage capacity is (Energy storage duration 11.11 hours, optimized value); Heat loss rate is ; The waste heat recovery system originates from the exhaust gas from the decomposition furnace and the heat dissipation from the carbonation reactor; the waste heat recovery rate is... (Optimized value); Waste heat power generation efficiency is Waste heat power generation is .

[0087] This invention uses calcium-based materials (CaO / CaCO3) as the energy storage medium. Compared with existing electrochemical energy storage (lithium batteries, flow batteries) and physical energy storage (pumped hydro storage, compressed air), it has the following significant advantages: the energy storage cost of calcium-based materials is extremely low, generally between 0.15-0.25 yuan / kWh, only 1 / 10 of that of lithium batteries; its energy storage duration is extremely long, reaching over 72 hours; at the same time, its cycle life is extremely long, exceeding 20,000 cycles, 5-10 times that of lithium batteries; the raw materials are abundant and readily available, with abundant and inexpensive limestone reserves; it has high safety, with no risk of fire or explosion, and is environmentally friendly.

[0088] A temperature-efficiency coupled optimization strategy was adopted, with the decomposition reaction temperature optimized within the range of 850-900℃ (balancing reaction rate and energy consumption); the carbonation reaction temperature optimized within the range of 650-750℃ (balancing reaction efficiency and equipment durability); a waste heat recovery strategy was implemented: waste heat from the high-temperature stage was used for steam power generation, waste heat from the medium-temperature stage was used for preheating raw materials, and waste heat from the low-temperature stage was used for district heating. A high-porosity CaO adsorbent was prepared using the extrusion spheroidization method, exhibiting a high sulfur resistance coefficient. Adding MgO, ZnO and other additives improves the material's cycle stability; a segmented carbonation strategy is adopted to delay material degradation.

[0089] Example Using Chuzhou, Anhui Province as the site and a 100MW wind-solar hybrid calcium cycle energy storage system as the simulation model, the technical solutions in the embodiments of this invention are clearly and completely described. Obviously, the described embodiments are only a part of the embodiments of this invention, not all of them. 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.

[0090] During the data acquisition phase, the meteorological data collection location for this invention was Chuzhou City, Anhui Province (32.3°N, 118.3°E); the collection period was from January 1, 2018 to December 31, 2022, a total of 5 years; the collection frequency was 1 hour, with a total of 43,800 data points; the data sources were the Anhui Provincial Meteorological Bureau and the NASA meteorological database. The collection parameters are as follows: wind speed (10-meter height): average 3.2 m / s, Weibull distribution parameters k=2.5, c=3.6 m / s; solar radiation: annual total radiation 4600 MJ / m², Beta distribution parameters... =2, =5, Ambient temperature: annual average temperature 15.8℃, summer average 28.5℃, winter average 4.2℃, relative humidity: annual average 75%. Load data includes grid-side load curve: peak-to-valley ratio 1.8, daily peak load occurs between 10-11 am and 6-8 pm; industrial electricity demand: accounts for 65% of the total load, relatively stable; residential electricity demand: accounts for 35% of the total load, with significant seasonal fluctuations.

[0091] In the preprocessing stage, outlier detection and removal are performed on the data first. For missing values, linear interpolation is used to fill them. Then, the data is normalized and the seasons are coded as follows: 1-Spring, 2-Summer, 3-Autumn, 4-Winter.

[0092] In the construction phase of the random forest prediction model, the random forest model training data was used, and the model performance was evaluated after 5-fold cross-validation. The wind power prediction results were R² = 0.9215 and RMSE = 0.0423, and the photovoltaic prediction results were R² = 0.8976 and RMSE = 0.0518.

[0093] In the modeling stage of the calcium cycle thermochemical energy storage system, the reaction rate equation is listed, and a temperature correction factor is introduced to determine the overall energy storage efficiency of the system.

[0094] A multi-objective optimization function is constructed, the objective function is listed, the optimization variables are extracted, and the annual power generation, cost per kilowatt-hour, investment payback period and carbon emission reduction are calculated. The comprehensive optimal value is calculated by multi-objective weighted summation.

[0095] The particle swarm optimization algorithm is invoked to perform the adaptive mutation operation.

[0096] Regarding the optimization results and analysis, the parameter comparison before and after optimization is shown in Table 1 below: Table 1.

[0097] The system performance is compared in Table 2 below: Table 2.

[0098] According to this embodiment, the system performance evaluation report includes the following core indicators: Technical performance indicators: Annual power generation is GWh; Annual curtailment rate Energy utilization rate Energy storage efficiency is The system reliability is (Annual available hours: 8743h); Economic performance indicators: Total investment cost is CAPEX = 703 million yuan; LCOE is 0.35 yuan / kWh; Payback period is PBP = 7.34 years; Internal rate of return is IRR = 12.87%; Net present value (NPV) (25 years) is NPV = 852 million yuan; Environmental benefit indicators: Annual carbon emission reduction is ΔCO2 = 167,800 tons / year; emission reduction cost is... CO2; Carbon trading revenue (based on 60 yuan / ton) is 10,000 yuan / year; Compared with the baseline plan: annual power generation increases by 3.61%; cost per kilowatt-hour decreases by 7.89%; investment payback period shortens by 0.95 years; annual carbon emission reduction increases by 3.64%.

[0099] In summary, the present invention has the following significant advantages: The calcium cycle thermochemical energy storage technology integrates AI optimization algorithms to achieve low-cost, high-density, and long-term energy storage. The energy storage cost is only 0.15-0.25 yuan / kWh, the energy storage density is as high as 3.1 MJ / kg, and the cycle life exceeds 20,000 cycles. The technical indicators are at the international leading level.

[0100] The AI ​​optimization framework combining random forest and improved particle swarm optimization achieves multi-objective global optimization, improving computational efficiency by more than 80% and optimization accuracy by 35% compared to traditional methods.

[0101] The optimized system generates 223.67 GWh of electricity annually (+3.61%), reduces the cost per kilowatt-hour to 0.35 yuan / kWh (-7.89%), shortens the investment payback period to 7.34 years (-0.95 years), and reduces annual carbon emissions by 167,800 tons (+3.64%), demonstrating a significant improvement in overall performance.

[0102] This invention is applicable to wind-solar hybrid systems of varying scales (10MW-1000MW) and can be extended to wind- and solar-rich areas nationwide, providing a replicable solution for high-proportion renewable energy grid integration. It not only achieves efficient utilization of wind and solar resources but also generates significant carbon emission reduction benefits, aligning with the national "dual-carbon" strategy and sustainable development goals. Therefore, this invention provides an innovative technical path for solving the challenges of source-grid-load-storage coordination in new power systems, possessing significant theoretical value and broad engineering application prospects.

[0103] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A multi-objective collaborative control method for a wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization, characterized in that, Includes the following steps: S1: Collect wind and solar resource data and load demand data to construct a training dataset; the wind and solar resource data includes wind speed, solar radiation intensity, ambient temperature and relative humidity; the load demand data includes grid-side load curves, industrial electricity demand and residential electricity demand; S2: A wind and solar power output prediction model is constructed using the random forest algorithm; S3: Establish a mathematical and physical model of a calcium cycle thermochemical energy storage system, wherein the calcium cycle thermochemical energy storage system uses calcium-based materials CaO and CaCO3 as energy storage media and realizes the storage and release of thermal energy through a reversible carbonation reaction; S4: Construct a multi-objective optimization function, the objective function expression is: in, To optimize the variable vector, For the investment payback period target, For energy efficiency targets, To achieve the cost per kilowatt-hour target, To achieve carbon emission reduction targets; S5: An improved particle swarm optimization algorithm is used to globally optimize system parameters. The optimization variables include: wind power ratio, unit investment, energy storage efficiency, energy storage duration, reaction temperature, and waste heat recovery rate. S6: Outputs the optimal parameter configuration scheme, including the ratio of wind and solar installed capacity, the capacity of the calcium cycle energy storage system, the reaction temperature setting, the operation strategy, and generates a system performance evaluation report.

2. The multi-objective collaborative control method for a wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization according to claim 1, characterized in that, In step S2, the method for constructing the wind and solar power output prediction model specifically includes the following steps: S21: Preprocess the collected wind and solar resource data, including outlier removal, missing value filling, normalization, and outlier detection using the Z-score method; S22: Use the Bootstrap autosampler to randomly sample K subsets from the original dataset; S23: Construct a CART decision tree for each subset of data, using the Gini Impurity coefficient as the splitting criterion when splitting nodes. in, The proportion of category i in node D; S24: Integrate the prediction results of K CART decision trees, and use the average method for regression problems: in, This represents the prediction result for the i-th tree; S25: Use 5-fold cross-validation to evaluate model performance, and calculate the coefficient of determination R² and root mean square error RMSE. Where N is the total number of samples involved in the calculation.

3. The multi-objective collaborative control method for a wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization according to claim 1, characterized in that, In step S3, the steps for establishing the mathematical and physical model of the calcium cycle thermochemical energy storage system are as follows: S31: Decomposition reaction kinetic model: in, Let A be the pre-exponential factor, representing the decomposition reaction rate. Let R be the activation energy, R be the gas constant, and T be the reaction temperature. Here, n represents the conversion rate, and n represents the reaction order. S32: Kinetic model of carbonation reaction: in, The carbonation reaction rate, Pre-exponential factor, For activation energy, The reaction temperature, Here, m represents the partial pressure of CO2, and m is the reaction order. Conversion rate; S33: Energy storage efficiency calculation model: in, As a benchmark efficiency, This is a temperature correction factor. For sulfur resistance coefficient, This is the cyclic decay coefficient; S34: Temperature correction factor model, based on third-order polynomial fitting: Where T is the reaction temperature. ; S35: Reaction heat calculation model: in, The heat of reaction, This represents the number of moles of calcium carbonate. H For the enthalpy of reaction, This refers to the reaction efficiency.

4. The multi-objective collaborative control method for a wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization according to claim 3, characterized in that, In step S4, the specific process of constructing the multi-objective optimization function is as follows: S41: Investment recovery period target: in, For the investment recovery period, For the minimum investment payback period, To maximize the investment recovery period, all items are normalized. Formula for calculating payback period: in, Annual power generation For the cost per kilowatt-hour, Unit operating cost d is the initial investment, and d is the discount rate; S42: Energy efficiency target: in, For the overall energy utilization rate of the system; Energy efficiency calculation: in, This represents the actual amount of electricity generated. To the extent of wasted electricity, Available power generation resources; S43: Cost per kilowatt-hour target: Where LCOE is the cost per kilowatt-hour; Cost per kilowatt-hour calculation: Where t represents the year, ranging from 1 to 25; S44: Carbon emission reduction targets: in, This refers to the annual carbon emission reduction. Carbon emission reduction calculation: in, Total power generation For coal-fired power plant emission coefficients, This represents the system emission coefficient.

5. The multi-objective collaborative control method for a wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization according to claim 4, characterized in that, In step S5, the improved particle swarm optimization algorithm includes: S51: Speed ​​Update Formula: in, Let be the velocity of particle i in the t-th iteration. For location, For inertial weights, The acceleration constant, A random number in the range [0,1]. For the individual's historical best position, The globally optimal position; S52: Position Update Formula: in, Let i be the position of particle i in the (t+1)th iteration; S53: Linear decreasing inertia weight strategy: in, =1.1, =0.3, where t is the current iteration number. This represents the maximum number of iterations. S54: Adaptive Mutation Operation: When the random perturbation is less than the mutation probability, the iteration position changes as follows: in, The random perturbation is a Gaussian distribution. For variable asynchronous length; S55: Convergence Criterion: The iteration stops when any of the following conditions are met: (1) Current iteration number ; (2) The global optimal fitness remains unchanged for 50 consecutive iterations, i.e. ; (3) The population standard deviation is less than the threshold, i.e., std(x) < 0.

01.

6. The multi-objective collaborative control method for a wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization according to claim 1, characterized in that, In step S6, the system performance evaluation report content is generated. Including the following indicators: Technical performance indicators: annual power generation, annual curtailment rate, energy utilization rate, energy storage efficiency, and system reliability; Economic performance indicators: total investment cost, cost per kilowatt-hour, payback period, internal rate of return, and net present value; Environmental benefit indicators: annual carbon emission reduction, emission reduction costs, and carbon trading revenue.

7. A wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization, characterized in that, Includes the following modules: Wind and solar power generation module: Composed of a wind turbine generator and a photovoltaic array, used to convert wind energy and solar energy into electrical energy; Calcium cycle thermochemical energy storage module: It consists of a decomposition furnace, a carbonation reactor, a heat storage tank, a cold storage tank, and a waste heat recovery system, and is used for the conversion and storage of electrical energy and thermal energy. The intelligent decision control module consists of a data acquisition unit, a prediction unit, an optimization unit, and a decision unit. It is used to collect data in real time, predict power generation output, optimize operating parameters, and generate control commands. Energy management module: Composed of power converter, energy router, grid connection interface and load dispatching system, used for power distribution, conversion, grid connection and load management.

8. The AI-optimized wind-solar hybrid calcium cycle thermochemical energy storage system according to claim 7, characterized in that, The configuration parameters of the wind and solar power generation module are as follows: Installed capacity is ,in The optimization range is 0.3-0.7 for the proportion of wind power; the cut-in wind speed is... Rated wind speed is Cut-out wind speed is The capacity factor is The power curve model is ,in The air density is 1.225 kg / m³, and A is the swept area. The wind energy utilization factor is 0.4; Photovoltaic modules: installed capacity is ; The standard test conditions are STC = 1000 W / m², 25℃; the photoelectric conversion efficiency is... Temperature coefficient is The capacity factor is .

9. A wind-solar hybrid calcium cycle thermochemical energy storage system based on AI optimization according to claim 7, characterized in that, The core equipment parameters of the calcium cycle thermochemical energy storage module are as follows: The decomposition furnace adopts a circulating fluidized bed, and its processing capacity can reach 500 t / d CaCO3; the operating temperature is set to... Operating pressure is Decomposition efficiency is Energy consumption is ; The carbonation reactor is a moving bed reactor with a processing capacity of 500 t / d CaO; the operating temperature is set to... Operating pressure is ; Carbonation efficiency is The heat of reaction is released as ; The thermal storage tank and cold storage tank are selected as insulated storage tanks; the energy storage medium is CaO / CaCO3 particles; the energy storage density is... Energy storage capacity is The heat loss rate is ; The waste heat recovery system originates from exhaust gas from the decomposition furnace and heat dissipation from the carbonation reactor; the waste heat recovery rate is... Waste heat power generation efficiency is Waste heat power generation is .

Images