Multi-objective optimization planning method, system, equipment and medium for wind-light-electricity-hydrogen storage system

Abstract

The invention discloses a multi-objective optimization planning method, system, equipment and medium for a wind-light-electricity-hydrogen storage system, and relates to the technical field of collaborative planning and operation optimization of an energy system, and the method comprises the steps: obtaining system operation data, carrying out the data preprocessing of the data, building a system model based on the preprocessed data, and building a multi-objective optimization model based on the system model. A mathematical expression and constraint conditions are defined, a hierarchical optimization strategy is adopted to solve the multi-objective optimization model, static configuration and dynamic response cooperation are carried out, an equipment capacity configuration scheme and an operation strategy are verified, performance indexes are calculated, and feedback adjustment is carried out according to a verification result. And iteratively optimizing the multi-objective optimization model and the hierarchical optimization strategy until all indexes reach the standard. According to the method, deep collaborative optimization of static configuration and dynamic response of the wind-light-electricity-hydrogen storage system is realized, and comprehensive benefit maximization of the final scheme on the premise of meeting all performance indexes is ensured.

Description

Technical Field

[0001] This invention relates to the field of collaborative planning and operation optimization technology for energy systems, specifically to multi-objective optimization planning methods, systems, equipment, and media for wind, solar, and hydrogen storage systems. Background Technology

[0002] As the global energy structure transitions towards low-carbon and clean energy, new energy power generation technologies, represented by wind and solar power, have developed. However, the inherent temporal fluctuations and intermittency of wind and solar resources pose a severe challenge to the safe and stable operation of power systems. To mitigate fluctuations in new energy output and improve energy utilization efficiency, combining wind and solar power generation systems with energy storage technology has become the mainstream technological approach. Early research focused on using short-term energy storage media such as lithium batteries for power smoothing, but such energy storage systems suffer from limited capacity, high cost, and difficulty in achieving long-term energy time shifts. In recent years, hydrogen energy, as an ideal clean energy carrier, has been regarded as a key technology for building high-proportion renewable energy power systems due to its advantages of long-term and large-scale storage.

[0003] Existing technologies still have three significant shortcomings: First, in terms of optimization objectives, existing methods mostly focus on single economic indicators, lacking quantitative constraints on key environmental and energy efficiency indicators such as carbon emission intensity and renewable energy absorption rate, making it difficult for planning schemes to meet the actual needs of multi-dimensional synergistic optimization under dual carbon objectives. Second, at the system architecture level, existing schemes typically adopt a single hydrogen energy storage architecture, failing to consider the time-series mismatch between short-term fluctuations in wind and solar power output and the slow dynamic characteristics of hydrogen energy systems. They have neither established a hybrid energy storage synergy mechanism for lithium batteries and hydrogen storage tanks nor refined modeling of the dynamic efficiency characteristics of proton exchange membrane electrolyzers and fuel cells. Finally, at the engineering implementation level, existing methods generally lack a closed-loop verification mechanism that runs through the entire planning and operation process. They have neither established a multi-dimensional verification system that includes safety constraints (hydrogen storage tank pressure threshold, lithium battery SOC limit), economic indicators (life cycle cost), and environmental indicators (carbon emission reduction), nor have they achieved iterative optimization of planning parameters and operating strategies through feedback adjustment mechanisms. Summary of the Invention

[0004] In view of the above-mentioned existing problems, the present invention provides a multi-objective optimization planning method, system, equipment and medium for wind, solar and hydrogen storage systems, in order to solve the problems of equipment capacity mismatch, operation control response lag and safety boundary violation in the prior art.

[0005] To address the aforementioned technical challenges, a multi-objective optimization planning method for wind, solar, and hydrogen storage systems is proposed, including: Acquire system operation data, preprocess the data, and construct a system model based on the preprocessed data; establish a multi-objective optimization model based on the system model, define mathematical expressions and constraints, and solve the multi-objective optimization model using a hierarchical optimization strategy, performing static configuration and dynamic response coordination; verify the equipment capacity configuration scheme and operation strategy, calculate performance indicators, and make feedback adjustments based on the verification results, iteratively optimizing the multi-objective optimization model and hierarchical optimization strategy until all indicators meet the standards.

[0006] As a preferred embodiment of the multi-objective optimization planning method for wind, solar, and hydrogen storage systems described in this invention, the data preprocessing includes: identifying and locating missing parts in the input data, filling them in using numerical interpolation methods, detecting abnormal values ​​in the input data that exceed the physical operating limits of the equipment, excluding abnormal values ​​from the dataset, and performing a standardization check on the processed data to ensure that the numerical range meets the parameter requirements of the model construction.

[0007] As a preferred embodiment of the multi-objective optimization planning method for the wind, solar, and hydrogen storage system described in this invention, the construction of the system model includes constructing an energy input unit model to characterize the mapping relationship between the output power of the wind and solar energy conversion equipment and external environmental parameters. An energy conversion unit model was constructed to simulate the correlation between equipment operating efficiency and input / output power during the conversion processes from electricity to hydrogen and from hydrogen to electricity. An energy storage unit model is constructed to describe the state evolution of long-term and short-term energy storage media during the charging and discharging process, and to integrate safe operation boundary conditions.

[0008] As a preferred embodiment of the multi-objective optimization planning method for wind, solar and hydrogen storage systems described in this invention, the establishment of the multi-objective optimization model includes constructing a multi-objective function and setting power balance constraints for the system, requiring that the total power supply of the system matches the total power demand at any given time. Set equipment operation constraints to limit the operation of each energy conversion and storage device within its rated capacity and efficiency range, and set planning capacity constraints to specify feasible upper and lower limits for the configuration capacity of each device.

[0009] As a preferred embodiment of the multi-objective optimization planning method for wind, solar, and hydrogen storage systems described in this invention, the hierarchical optimization strategy includes, at the planning layer, using an improved non-dominated sorting genetic algorithm to collaboratively optimize the capacity of wind turbine generators, photovoltaic arrays, electrolyzers, fuel cells, hydrogen storage tanks, and lithium batteries, and generating a set of Pareto optimal planning schemes that are non-dominated under the multi-objective function through algorithm iteration. At the operational level, a model predictive control method is adopted. Based on real-time collected wind and solar power output data, load demand data, lithium battery state of charge data, and hydrogen storage tank pressure data, the optimization problem in the future finite time domain is solved in each control cycle, and the input power command of the proton exchange membrane electrolyzer, the output power command of the proton exchange membrane fuel cell, the charging and discharging power command of the lithium battery, and the charging and discharging command of the hydrogen storage tank are dynamically output.

[0010] As a preferred embodiment of the multi-objective optimization planning method for wind, solar, and hydrogen storage systems described in this invention, the improved non-dominated sorting genetic algorithm includes: initializing a population, generating a set of individuals representing different combinations of equipment capacity, simulating the annual operation of each individual based on the system model, calculating its performance on multiple objective functions, namely, the total life cycle cost, annual carbon emissions, and annual renewable energy absorption rate, and performing non-dominated sorting and crowding calculation on the individuals in the population according to multiple objective function values ​​to evaluate the quality of the individuals, generating a new offspring population through selection, crossover, and mutation operations, maintaining the diversity of the solution set in the objective space using a reference point mechanism, iteratively optimizing until the termination condition is met, and outputting a set of Pareto optimal planning schemes; The model predictive control method includes, at each control moment, using an autoregressive integral moving average model to predict wind power output and photovoltaic power output for future periods, combining historical load data, adopting a load prediction model that considers intraday periodicity and inter-day correlation to predict the power load demand for the same period in the future, and using the predicted data as input to construct and solve a rolling optimization problem with the goal of minimizing daily operating costs, applying the first instruction in the solved control instruction sequence to the actual system, and repeating the prediction and optimization process when entering the next control cycle; The computational performance indicators include the calculation of new energy absorption rate, carbon emission reduction, economic efficiency, and safety. The formula for calculating the renewable energy absorption rate is as follows: in, For the new energy consumption rate, Let t be the wind power output. Photovoltaic output at time t Let t be the amount of wind and solar power curtailed. For time indexing, The total number of time steps is used to verify whether... Greater than or equal to 92%; The formula for calculating carbon emission reduction targets is as follows: in, This represents the total annual carbon emissions. Carbon emission factors from purchasing electricity from the power grid The carbon emission factor of the electrolyzer operation, As a carbon emission reduction factor for fuel cells, Let t be the power purchased by the power grid. Let be the power of the electrolytic cell at time t. Let be the output power of the fuel cell at time t. For time step; Economic indicators are expressed as follows: in, For total lifecycle cost, For initial investment costs, For operation and maintenance costs, For equipment replacement costs, The residual value of the equipment; Safety indicators are expressed as follows: in, Let t be the pressure of the hydrogen storage tank. The minimum safe pressure for the hydrogen storage tank, This is the maximum safe pressure for the hydrogen storage tank. Let t be the state of charge of the lithium battery. This represents the lowest state of charge for a lithium battery. This represents the highest state of charge of the lithium battery.

[0011] As a preferred embodiment of the multi-objective optimization planning method for wind, solar and hydrogen storage systems described in this invention, the feedback adjustment includes: when any performance index fails to reach a preset threshold, extracting data on equipment start-up and shutdown frequency, wind and solar power curtailment time distribution, and energy storage equipment status change trajectory during the operation simulation process, and feeding the data back to the improved non-dominated sorting genetic algorithm optimization in the planning layer, adjusting the weight coefficient of the corresponding unmet index in the multi-objective function, or modifying the capacity constraint boundary of the equipment; Based on the adjusted parameters, the optimization from the planning layer to the operation layer simulation is re-executed, and the adjustments are repeated until all performance indicators meet the preset requirements, and an implementable system capacity configuration and operation strategy scheme is output.

[0012] The beneficial effects of this preferred technical solution are as follows: the hierarchical strategy unifies static capacity planning and dynamic operation control within a single framework. The planning layer ensures the long-term economic efficiency and rationality of the capacity, while the operation layer MPC optimizes based on real-time data, enabling the system to proactively adapt to random fluctuations in wind and solar power output and load, thus avoiding disconnection.

[0013] As a preferred embodiment of the multi-objective optimization planning system for wind, solar, and hydrogen storage systems described in this invention, it is characterized by including a data acquisition and preprocessing module, a system modeling and simulation module, a multi-objective optimization modeling module, a hierarchical collaborative solution module, and a verification and feedback adjustment module.

[0014] The data acquisition and preprocessing module is used to collect multi-source data on wind and solar resources, load demand, equipment parameters, and economic and environmental parameters. It fills in missing data through interpolation and identifies and removes abnormal data based on the physical limits of the equipment.

[0015] The system modeling and simulation module is used to transform the physical system into a mathematical model and to simulate and predict the system's operating state under different conditions.

[0016] The multi-objective optimization modeling module is used to establish a multi-objective function with the core objectives of minimizing the total life cycle cost, minimizing carbon emissions, and maximizing the renewable energy absorption rate, and to set the power balance constraints, equipment operation constraints, and planned capacity constraints that the system must comply with.

[0017] The hierarchical collaborative solution module is used to decouple long-term planning problems from short-term operation and control problems through a hierarchical strategy, and to perform collaborative optimization.

[0018] The verification and feedback adjustment module is used to simulate the operation of the planning scheme throughout the year based on historical data, calculate the new energy absorption rate, carbon emission reduction, full life cycle cost and safety indicators, and compare them with the preset targets. When any indicator fails to meet the target, the problems exposed in the operation simulation are fed back to the optimization algorithm of the planning layer, the target weight or capacity parameter is adjusted, and the optimization iteration is restarted until the optimal scheme that meets the target is output.

[0019] A computer device includes a memory and a processor, the memory storing a computer program, and the processor executing the computer program to implement the steps of a method for multi-objective optimization planning of a wind, solar, and hydrogen storage system.

[0020] A computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of a method for multi-objective optimization planning of a wind, solar, and hydrogen storage system.

[0021] The beneficial effects of this invention are as follows: By preprocessing data and refining system modeling, this invention ensures the quality and physical rationality of input data from the source. It also significantly improves the reliability and simulation realism of the optimization scheme by utilizing precise equipment and energy storage dynamic models, avoiding planning deviations caused by data anomalies or model distortions. By constructing a multi-objective function integrating economic, environmental, and energy efficiency considerations, and combining it with a hierarchical optimization strategy of planning and operation, a Pareto optimal scheme set for scientific consideration is generated. This effectively solves the problem of a single objective and the disconnect between short-term and long-term timescale decisions, enabling the scheme to possess optimality and strong robustness in both static configuration and dynamic response. Furthermore, by introducing a verification mechanism based on quantitative indicators and closed-loop feedback adjustment, the invention ensures that key performance indicators meet the standards, enabling the system to self-correct and continuously optimize planning defects, greatly improving the success rate and overall benefits of the final scheme in engineering implementation. Attached Figure Description

[0022] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0023] Figure 1 The above is a flowchart of a multi-objective optimization planning method for a wind, solar, and hydrogen storage system provided in one embodiment of the present invention.

[0024] Figure 2 The flowchart shows the system scheme of a multi-objective optimization planning system for a wind, solar, and hydrogen storage system provided in one embodiment of the present invention. Detailed Implementation

[0025] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0026] Example 1, referring to Figure 1 As an embodiment of the present invention, a multi-objective optimization planning method for wind, solar, and hydrogen storage systems is provided, comprising: S100: Acquire system operation data, preprocess the data, and build a system model based on the preprocessed data; S200: Based on the system model, a multi-objective optimization model is established, and mathematical expressions and constraints are defined. A hierarchical optimization strategy is adopted to solve the multi-objective optimization model, and static configuration and dynamic response are coordinated. S300: Verify the equipment capacity configuration scheme and operation strategy, calculate performance indicators, and make feedback adjustments based on the verification results. Iterate and optimize the multi-objective optimization model and hierarchical optimization strategy until all indicators meet the standards.

[0027] It should be noted that data preprocessing ensured the robustness and accuracy of the model, and the construction of a refined system model ensured that the solution closely reflected actual working conditions. The establishment of a multi-objective optimization model achieved a scientific balance between economic, environmental, and technical benefits. Furthermore, the hierarchical optimization strategy achieved deep synergy between static configuration and dynamic response. Through quantitative performance indicators and feedback adjustment mechanisms, the scientific nature of the integrated wind, solar, hydrogen, and energy storage system planning, operational reliability, and overall benefits were significantly improved.

[0028] Example 2, refer to Figure 1 This is a second embodiment of the present invention, which provides a multi-objective optimization planning method for wind, solar, and hydrogen storage systems, including: In step S100, acquiring system operating data includes steps S101 to S104: S101: Obtain wind and solar resource data, specifically including hourly wind speed, solar irradiance, and ambient temperature data throughout the year.

[0029] S102: Obtain load demand data, i.e., hourly power load data throughout the year.

[0030] S103: Obtain basic equipment parameters, specifically including wind turbine cut-in / rated / cut-out wind speeds, photovoltaic module power temperature coefficient, PEMEC / PEMFC efficiency fitting coefficient, hydrogen storage tank safety pressure threshold, and lithium battery charge / discharge efficiency.

[0031] S104: Obtain economic and environmental parameters, specifically including equipment unit cost, grid electricity purchase price and carbon emission factor, green hydrogen subsidy standard, discount rate, and equipment residual value rate.

[0032] Furthermore, in this embodiment of the application, in step S100, the data preprocessing includes steps S111~S112: S111: Use linear interpolation to fill in missing data, identify missing points in time series data, and use the arithmetic mean of two consecutive valid data points to fill in the missing data to maintain data continuity. S112: Employ a threshold method based on physical limits to remove abnormal data. Compare the input data with the physical operating limits of the equipment (wind turbine cut-out wind speed 25m / s). Any data point exceeding the current threshold is judged as abnormal and directly removed to ensure that the input data is within the effective parameter range of the equipment model.

[0033] In an optional implementation, in step S100, the data preprocessing further includes calculating the feature similarity between missing data points and multiple adjacent data points in the time series, selecting the K most similar neighbors, filling them with a weighted average, and using a clustering algorithm (K-means) to group the data for the whole year. Data points falling outside any cluster center at a certain distance are considered outliers and are removed.

[0034] In another optional implementation, in step S100, the data preprocessing may further include performing wavelet decomposition on the overall data sequence containing noise and anomalies to obtain components of different frequencies, filtering out high-frequency components that are determined to be noise through threshold processing, and using the remaining low-frequency components and the processed high-frequency components to perform wavelet reconstruction to achieve data smoothing and missing value filling.

[0035] Furthermore, in this embodiment of the application, in step S100, the construction of the system model includes steps S121 to S123: S121: Construct an energy input unit model. The output of the wind turbine is calculated through the power characteristic curve. The curve defines the output power under different wind speed ranges (including cut-in wind speed, rated wind speed and cut-out wind speed). The wind turbine model is represented as follows: in, Let v be the wind speed probability density function, describing the probability distribution of wind speed v. For real-time wind speed, This is a shape parameter, with values ​​ranging from 1.5 to 3.5. For scale parameters, The average wind speed, For gamma function, The base of the exponential function; The power output model of a wind turbine is represented as follows: in, Let be the actual output power of the wind turbine at time t. For time indexing, Let be the real-time wind speed at time t. This refers to the cut-in wind speed of the wind turbine, which is the minimum wind speed at which the turbine starts generating electricity, 3 m / s. The rated wind speed of the wind turbine is 13 m / s, which is the wind speed at which the turbine reaches its rated output power. This refers to the cut-off wind speed of the wind turbine, which is the maximum wind speed at which the turbine stops operating to protect the equipment, 25 m / s. This refers to the rated power of the wind turbine.

[0036] The output of the photovoltaic array is calculated using the rated power, actual irradiance, module temperature, and power temperature coefficient under standard test conditions. The photovoltaic array model is represented as follows: in, Let be the actual output power of the photovoltaic array at time t. This refers to the rated power of the photovoltaic array under standard test conditions. Let be the actual solar irradiance projected onto the photovoltaic panel at time t. The reference irradiance under standard test conditions is 1000 W / m². Let be the operating temperature of the photovoltaic module at time t. The reference component temperature under standard test conditions is 25°C. This is the power temperature coefficient of the photovoltaic module, typically a negative value of -0.004 / ℃.

[0037] S122: Construct an energy conversion unit model. The operating efficiency of the proton exchange membrane electrolyzer is determined by the functional relationship between its input power and a set of experimental fitting coefficients, while the hydrogen production is calculated based on the operating efficiency, Faraday efficiency, and input power. The efficiency model of an electrolyzer is expressed as: in, The operating efficiency of the proton exchange membrane electrolyzer at time t is taken as 0.6-0.85. Let be the electrical power input to the electrolytic cell at time t. , and These are the fitting coefficients of the quadratic polynomial obtained through calibration using experimental data; The hydrogen production model is expressed as follows: in, Let be the volume of hydrogen produced by the electrolyzer at time t. Let be the electrical power input to the electrolytic cell at time t. For time step, This is Faraday's constant, with a value of 96485 C / mol. This is the lower heating value of hydrogen. The value represents the Faraday efficiency of the electrolytic cell, ranging from 0.9 to 0.95. The operating efficiency of a proton exchange membrane fuel cell is calculated by its output power and a coefficient characterizing efficiency decay, while hydrogen consumption is calculated based on the operating efficiency and output power. The fuel cell efficiency model is expressed as: in, The operating efficiency of the fuel cell at time t is taken as 0.4-0.6. This represents the maximum efficiency that the fuel cell can achieve under no-load or light-load conditions, with a value of 0.6. The efficiency degradation coefficient characterizes the rate at which fuel cell efficiency decreases linearly with increasing output power; a typical value is 5 × 10⁻⁶. -5 kW, Let be the electrical power output of the fuel cell at time t; The hydrogen consumption model is expressed as follows: in, Let be the volume of hydrogen consumed by the fuel cell at time t. Let be the electrical power output of the fuel cell at time t.

[0038] S123: Construct an energy storage unit model. The pressure inside the hydrogen storage tank is calculated based on the amount of hydrogen stored inside, the volume of the hydrogen storage tank, and the ambient temperature, according to the gas state equation. The dynamic model of hydrogen storage capacity is expressed as follows: in, Let be the volume of hydrogen stored in the hydrogen storage tank at the end of time t. Let be the volume of hydrogen stored in the hydrogen storage tank at the end of time t−1 (i.e., the previous time). Let be the volume of hydrogen gas produced by the electrolyzer at time t. Let be the volume of hydrogen consumed by the fuel cell at time t. Let t be the amount of hydrogen loss caused by hydrogen leakage and emission from the hydrogen storage system. Let be the gas pressure inside the hydrogen storage tank at time t. This refers to the physical volume of the hydrogen storage tank. The gas constant of hydrogen is 4.124 kJ / kg / Kelvin. The thermodynamic temperature of hydrogen gas inside the hydrogen storage tank; The pressure constraint model for the hydrogen storage tank is expressed as follows: in, Let be the gas pressure inside the hydrogen storage tank at time t. The minimum safe pressure allowed for the hydrogen storage tank. This is the maximum safe pressure allowed for the hydrogen storage tank.

[0039] The state of charge (SOC) of a lithium battery is calculated based on the SOC of the previous moment, the current charging and discharging power, charging and discharging efficiency, and self-discharge rate. The dynamic model of battery capacity is expressed as: in, Let be the state of charge of the lithium battery at time t, with a value ranging from 0.2 to 0.8. Let be the state of charge at time t−1 (i.e., the previous time). The charging efficiency of lithium batteries is typically 0.9-0.95. This refers to the discharge efficiency of lithium batteries. Let be the charging power of the lithium battery at time t. Let be the discharge power of the lithium battery at time t. This refers to the rated capacity of the lithium battery. This represents the self-discharge rate of the lithium battery, typically 0.005 per day. The charge / discharge power constraint model is expressed as follows: in, Let be the charging power of the lithium battery at time t. Let be the discharge power of the lithium battery at time t. This refers to the rated charge and discharge power of the lithium battery.

[0040] In an optional implementation, in step S100, the construction of the system model further includes collecting a large amount of historical operating data (such as wind speed, irradiance and corresponding actual output of wind turbines and photovoltaics) as a training set, constructing a deep neural network model, using environmental parameters as input and equipment output or efficiency as output for training, and establishing a black-box input-output mapping relationship.

[0041] In another optional implementation, in step S100, the construction of the system model may further include simplifying the dynamic response characteristics (such as start-up, shutdown, and power ramp-up) of devices such as electrolyzers and fuel cells into a first- or second-order transfer function through a system identification method to describe the dynamic process of output changing with input.

[0042] In this embodiment of the application, step S200, the establishment of the multi-objective optimization model includes steps S201 to S204: S201: Construct a multi-objective function. The function integration reflects the economic expenditure of the system throughout its entire life cycle, the total carbon emissions during operation, and the actual utilization level of renewable energy power generation. The formula for minimizing the life cycle cost (LCC) is expressed as: in, For initial investment costs, The objective function is one, namely the total lifecycle cost of the system. It is the collection of all devices in the system. Let be the unit capacity cost of the i-th type of equipment. Let i be the configuration capacity of the i-th type of device. For operation and maintenance costs, The system's operating cycle, i.e., the total number of years for the planning and investigation. The discount rate is... Let be the electricity purchase cost in year t. Let t be the hydrogen operation and maintenance cost in year t. The equipment maintenance cost in year t is... To reduce replacement costs, In the running cycle The number of times internal equipment needs to be replaced. The investment cost required for the nth equipment replacement. Let n be the year in which the nth replacement occurs. Residual value is the present value of the remaining equipment value at the end of the system's operating cycle. Let be the residual value rate of the i-th type of equipment, which is the proportion of the equipment's remaining value at the end of its life to the initial investment; The formula for minimizing carbon emissions is expressed as: in, The second objective function is the total carbon emissions of the system within one year. The total number of steps in a year is 8760. Carbon emission factors from purchasing electricity from the power grid Let be the electrical power purchased from the grid at time t. The carbon emission factor of the electrolyzer operation, Let be the input power of the electrolytic cell at time t. As a carbon emission reduction factor for fuel cells, Let be the output power of the fuel cell at time t. For time step; Maximizing the renewable energy absorption rate is expressed as: in, The objective function is the system's annual renewable energy consumption rate. The total number of time steps in a year. The actual output of the wind power at time t. The actual output of the photovoltaic system at time t. Let be the power of wind and solar power curtailment at time t. For time indexing.

[0043] S202: Set the power balance constraint of the system to ensure that the total power supply of the system matches the total power demand at any given time. The power balance constraint is expressed as follows: in, The actual output of the wind power at time t. The actual output of the photovoltaic system at time t. Let be the output power of the fuel cell at time t. Let be the electrical power purchased from the grid at time t. Let be the charging power of the lithium battery at time t. Let be the discharge power of the lithium battery at time t. Let be the input power of the electrolytic cell at time t. Let be the system's power load demand at time t.

[0044] S203: Set equipment operating constraints, limiting the operation of each energy conversion and storage device within its rated capacity and efficiency range. The equipment operating constraints are expressed as follows: in, This is the rated input power of the electrolytic cell. To determine the operating efficiency of the electrolyzer at time t, This represents the minimum stable operating power of the fuel cell. This refers to the rated output power of the fuel cell. To determine the operating efficiency of the fuel cell at time t, Let be the amount of hydrogen stored in the hydrogen storage tank at time t. This is the minimum allowable hydrogen storage capacity for the hydrogen storage tank. This is the maximum allowable hydrogen storage capacity of the hydrogen storage tank. Let t be the amount of hydrogen stored in the hydrogen storage tank at time t−1 (i.e., the previous time). The hydrogen production efficiency of an electrolyzer is the overall efficiency of converting input electrical energy into hydrogen energy. The power generation efficiency of a fuel cell is the overall efficiency of converting hydrogen energy into output electrical energy. Let be the amount of hydrogen lost due to leakage and emission at time t.

[0045] S204: Set planning capacity constraints, specifying feasible upper and lower limits for the configuration capacity of each device to ensure the engineering feasibility of the planning scheme. The planning capacity constraints are expressed as follows: in, Let i be the configuration capacity of the i-th type of equipment to be planned. The minimum capacity that the i-th type of device can be configured with. The maximum capacity that can be configured for the i-th type of device.

[0046] In an optional implementation, in step S200, the establishment of the multi-objective optimization model further includes selecting one of the three objectives as the primary objective (life cycle cost), converting the remaining objectives (carbon emissions, new energy absorption rate) into constraints, requiring values ​​to be better than a set threshold, and generating a uniformly distributed Pareto optimal solution by systematically adjusting the set threshold values.

[0047] In another optional implementation, in step S200, establishing the multi-objective optimization model may further include setting an ideal value to be achieved for each objective, optimizing by minimizing the deviation between the actual value and the ideal value of each objective, and seeking a compromise solution that is closest to the ideal value by assigning priority or weight to the deviations of different objectives.

[0048] Furthermore, in this embodiment of the application, in step S200, the hierarchical optimization strategy includes steps S211~S213: S211: At the planning level, an improved non-dominated sorting genetic algorithm is used to coordinate the optimization of wind turbine capacity, photovoltaic array capacity, electrolyzer capacity, fuel cell capacity, hydrogen storage tank volume and lithium battery rated capacity. The algorithm iteratively generates a set of Pareto optimal planning schemes that are non-dominated under multiple objective functions. The objective function is expressed as: in, For the total lifecycle cost of the system, For initial investment costs, For operation and maintenance costs, For equipment replacement costs, The revenue includes revenue from electricity sales and hydrogen sales. Let be the electrical power purchased from the grid at time t. The marginal carbon emission factor of the power grid represents the carbon emissions associated with purchasing one unit of electricity from the grid. This represents the system's total annual carbon emissions. For wind and solar curtailment rates, Let be the wind curtailment power at time t. Let be the power of light discarded at time t. To provide the theoretical output of wind power at time t, This contributes to the theoretical photovoltaic effect at time t.

[0049] S212: At the operation layer, model predictive control is adopted based on real-time collected wind and solar power output data, load demand data, lithium battery state of charge data, and hydrogen storage tank pressure data. The wind and solar power output prediction model is expressed as follows: in, This represents the predicted wind power output at time t+j from the current time t. This represents the predicted photovoltaic output at time t+j from the current time t. This is the historical actual power output sequence of wind power from the current time t back to time t−p+1. This is the historical actual power output sequence of photovoltaic power generation from the current time t back to time t−q+1. The coefficients of the wind power prediction autoregressive model are obtained through training with historical data. The coefficients of the photovoltaic prediction autoregressive model are obtained through training with historical data. Let this be the autoregressive order of the wind power prediction model. The autoregressive order of the photovoltaic prediction model is given. Let be the prediction error for wind power at time t+j. Let be the prediction error of photovoltaic at time t+j; The load forecasting model is expressed as follows: in, Let be the predicted load demand at time t+j from the current time t. The actual load at time t. This represents the actual load 24 hours ago (at the same time yesterday). This is a constant term (intercept), representing the baseline level of the load. This is the intraday load correlation coefficient, reflecting the degree to which the current load depends on the recent (previous hour's) load. This is the inter-day load correlation coefficient, reflecting the 24-hour periodicity of load, i.e., the impact of yesterday's load at the same time on today's forecast. The load prediction error at time t+j is denoted as .

[0050] S213: In each control cycle, the optimization problem in the future finite time domain is solved in a rolling manner, and the input power command of the proton exchange membrane electrolyzer, the output power command of the proton exchange membrane fuel cell, the charging and discharging power command of the lithium battery, and the charging and discharging command of the hydrogen storage tank are dynamically output. The rolling solution formula is expressed as: in, The optimization objective is to find the optimal vector of control variables. To minimize the objective function value, Let be the vector of control variables that needs to be solved at time t. To predict the step size index in the time domain, For the prediction time domain, that is, the total number of steps to predict the future in one optimization. This represents the power purchased by the power grid at time t+j, predicted at time t. This is a reference value for the power purchased by the power grid. This represents the hydrogen quantity in the hydrogen storage tank predicted at time t+j from time t. This is a reference value for hydrogen storage capacity. This represents the predicted state of charge (SOC) of the lithium battery at time t+j, based on the value at time t. This is a reference value for the state of charge of lithium batteries. , and These are the weighting coefficients; It should be noted that the control variable vector is represented as follows: in, Let be the input power of the electrolytic cell at time t. Let be the output power of the fuel cell at time t. Let be the charging power of the lithium battery at time t. Let be the discharge power of the lithium battery at time t. It is the transpose of the vector.

[0051] In an optional implementation, in step S200, the hierarchical optimization strategy further includes adding a mid-term scheduling layer between the planning layer and the operation layer. After the planning layer outputs the capacity plan, the mid-term scheduling layer formulates a hydrogen allocation plan (target inventory curve) for the hydrogen storage tank based on wind and solar load forecasts for the next few weeks to months. Under the guidance of this hydrogen allocation plan, the operation layer performs power optimization on a shorter time scale.

[0052] In another optional implementation, in step S200, the hierarchical optimization strategy may further include coupling capacity planning and operation control within a unified dynamic programming framework, using a value function to approximate the long-term operating costs under different capacity configurations and system states, and continuously updating and approximating the value function while optimizing capacity, thereby achieving collaborative decision-making between planning and control.

[0053] Furthermore, in step S211, the improved non-dominated sorting genetic algorithm includes steps A1 to A5: A1: Initialize the population by randomly generating a set of individuals representing different combinations of device capacities.

[0054] A2: For each individual, a full-year operation simulation is conducted based on the system model to calculate its performance on multiple objective functions, namely, the total life cycle cost, the total annual carbon emissions, and the annual renewable energy consumption rate.

[0055] A3: Based on multiple objective function values, perform non-dominated ranking and crowding calculation on individuals in the population to evaluate their quality.

[0056] A4: New offspring populations are generated through selection, crossover, and mutation operations, and the diversity of the solution set in the target space is maintained by using a reference point mechanism.

[0057] A5: Iteratively execute the optimization steps until the termination condition is met, and output the set of Pareto optimal planning solutions.

[0058] In step S212, the model predictive control method includes steps B1 to B4: B1: At each control time, the wind power output and photovoltaic power output for future periods are predicted using an autoregressive integral moving average model.

[0059] B2: Combining historical load data, a load forecasting model that considers intraday periodicity and inter-day correlation is used to predict the electricity load demand for the same period in the future.

[0060] B3: Using predicted data as input, construct and solve a rolling optimization problem with the goal of minimizing daily operating costs. The operating costs include grid power purchase costs and equipment operation and maintenance costs. Penalize the deviation of the hydrogen storage tank pressure and lithium battery state of charge from the expected values ​​through a weight matrix. B4: Apply the first instruction in the control instruction sequence obtained by the solution to the actual system, and repeat the prediction and optimization process when entering the next control cycle.

[0061] In step S300, the computational performance indicators include steps S301 to S304: S301: The formula for calculating the renewable energy consumption rate is as follows: in, For the new energy consumption rate, Let t be the wind power output. Photovoltaic output at time t Let t be the amount of wind and solar power curtailed. For time indexing, The total number of time steps is used to verify whether... .

[0062] S302: Calculation of carbon emission reduction targets, the calculation formula is as follows: in, This represents the total annual carbon emissions. Carbon emission factors from purchasing electricity from the power grid The carbon emission factor of the electrolyzer operation, As a carbon emission reduction factor for fuel cells, Let t be the power purchased by the power grid. Let be the power of the electrolytic cell at time t. Let be the output power of the fuel cell at time t. For the time step, verify whether .

[0063] S303: The economic indicators are expressed as follows: in, For total lifecycle cost, For initial investment costs, For operation and maintenance costs, For equipment replacement costs, To determine the equipment's residual value, verify whether it is reduced by 10%-15% compared to existing technologies.

[0064] S304: Safety indicators are expressed as follows: in, Let t be the pressure of the hydrogen storage tank. The minimum safe pressure for the hydrogen storage tank, This is the maximum safe pressure for the hydrogen storage tank. Let t be the state of charge of the lithium battery. This represents the lowest state of charge for a lithium battery. For the highest state of charge of the lithium battery, the hydrogen storage tank pressure should be between 0.5 and 35 or 70 MPa, and the lithium battery SOC should be between 0.2 and 0.8.

[0065] Furthermore, in step S300, the feedback adjustment includes, when any performance indicator fails to reach the preset threshold, extracting the equipment start-up and shutdown frequency, wind and solar power curtailment time distribution, and energy storage equipment status change trajectory data during the operation simulation process, and feeding the data back to the improved non-dominated sorting genetic algorithm optimization in the planning layer, adjusting the weight coefficient of the corresponding unmet indicator in the multi-objective function, or modifying the capacity constraint boundary of the equipment. Based on the adjusted parameters, the optimization from the planning layer to the operation layer simulation is re-executed, and the adjustments are repeated until all performance indicators meet the preset requirements, and an implementable system capacity configuration and operation strategy scheme is output.

[0066] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0067] Example 3, referring to Figure 2 This is the third embodiment of the present invention. This embodiment provides a multi-objective optimization planning system for wind, solar and hydrogen storage systems, including a data acquisition and preprocessing module, a system modeling and simulation module, a multi-objective optimization modeling module, a hierarchical collaborative solution module, and a verification and feedback adjustment module.

[0068] The data acquisition and preprocessing module is used to collect multi-source data on wind and solar resources, load demand, equipment parameters, and economic and environmental parameters. It fills in missing data through interpolation and identifies and removes abnormal data based on the physical limits of the equipment.

[0069] The system modeling and simulation module is used to transform the physical system into a mathematical model and to simulate and predict the system's operating state under different conditions.

[0070] The multi-objective optimization modeling module is used to establish a multi-objective function with the core objectives of minimizing the total life cycle cost, minimizing carbon emissions, and maximizing the renewable energy absorption rate, and to set the power balance constraints, equipment operation constraints, and planned capacity constraints that the system must comply with.

[0071] The hierarchical collaborative solution module is used to decouple long-term planning problems from short-term operation and control problems through a hierarchical strategy, and to perform collaborative optimization.

[0072] The verification and feedback adjustment module is used to simulate the operation of the planning scheme throughout the year based on historical data, calculate the new energy absorption rate, carbon emission reduction, full life cycle cost and safety indicators, and compare them with the preset targets. When any indicator fails to meet the target, the problems exposed in the operation simulation are fed back to the optimization algorithm of the planning layer, the target weight or capacity parameter is adjusted, and the optimization iteration is restarted until the optimal scheme that meets the target is output.

[0073] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

[0074] Example 4, the fourth embodiment of the present invention, differs from the previous three embodiments in that: If the aforementioned functions are implemented as software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0075] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0076] More specific examples of computer-readable media (a non-exhaustive list) include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which the program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0077] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

Claims

1. A multi-objective optimization planning method for wind, solar, and hydrogen storage systems, characterized by: include, Acquire system operation data, preprocess the data, and build a system model based on the preprocessed data; A multi-objective optimization model is established based on the system model, and mathematical expressions and constraints are defined. A hierarchical optimization strategy is adopted to solve the multi-objective optimization model, and static configuration and dynamic response are coordinated. Verify the equipment capacity configuration scheme and operation strategy, calculate performance indicators, and make feedback adjustments based on the verification results. Iterate and optimize the multi-objective optimization model and hierarchical optimization strategy until all indicators meet the standards.

2. The multi-objective optimization planning method for wind, solar, and hydrogen storage systems as described in claim 1, characterized in that: The data preprocessing includes identifying and locating missing parts in the input data, filling them in using numerical interpolation methods, detecting abnormal values ​​in the input data that exceed the physical operating limits of the device and excluding abnormal values ​​from the dataset, and performing a standardization check on the processed data to ensure that the numerical range meets the parameter requirements of the model construction.

3. The multi-objective optimization planning method for wind, solar, and hydrogen storage systems as described in claim 2, characterized in that: The system model construction includes constructing an energy input unit model to characterize the mapping relationship between the output power of wind and solar energy conversion equipment and external environmental parameters. An energy conversion unit model was constructed to simulate the correlation between equipment operating efficiency and input / output power during the conversion processes from electricity to hydrogen and from hydrogen to electricity. An energy storage unit model is constructed to describe the state evolution of long-term and short-term energy storage media during the charging and discharging process, and to integrate safe operation boundary conditions.

4. The multi-objective optimization planning method for wind, solar, and hydrogen storage systems as described in claim 3, characterized in that: The establishment of the multi-objective optimization model includes constructing a multi-objective function and setting power balance constraints for the system, requiring that the total power supply of the system matches the total power demand at any given time. Set equipment operation constraints to limit the operation of each energy conversion and storage device within its rated capacity and efficiency range, and set planning capacity constraints to specify feasible upper and lower limits for the configuration capacity of each device.

5. The multi-objective optimization planning method for wind, solar, and hydrogen storage systems as described in claim 4, characterized in that: The hierarchical optimization strategy includes, at the planning level, using an improved non-dominated sorting genetic algorithm to collaboratively optimize the capacity of wind turbine generators, photovoltaic arrays, electrolyzers, fuel cells, hydrogen storage tanks, and lithium batteries, and generating a set of Pareto optimal planning schemes that are non-dominated under multiple objective functions through algorithm iteration. At the operational level, a model predictive control method is adopted. Based on real-time collected wind and solar power output data, load demand data, lithium battery state of charge data, and hydrogen storage tank pressure data, the optimization problem in the future finite time domain is solved in each control cycle, and the input power command of the proton exchange membrane electrolyzer, the output power command of the proton exchange membrane fuel cell, the charging and discharging power command of the lithium battery, and the charging and discharging command of the hydrogen storage tank are dynamically output.

6. The multi-objective optimization planning method for wind, solar, and hydrogen storage systems as described in claim 5, characterized in that: The improved non-dominated sorting genetic algorithm includes: initializing the population, generating a set of individuals representing different combinations of equipment capacity, simulating the operation of each individual throughout the year based on the system model, calculating the performance on multiple objective functions, namely the total life cycle cost, annual carbon emissions, and annual renewable energy absorption rate, and performing non-dominated sorting and crowding calculation on the individuals in the population according to the values ​​of multiple objective functions to evaluate the quality of the individuals, generating a new offspring population through selection, crossover, and mutation operations, maintaining the diversity of the solution set in the objective space using a reference point mechanism, iteratively optimizing until the termination condition is met, and outputting a set of Pareto optimal planning schemes. The model predictive control method includes, at each control moment, using an autoregressive integral moving average model to predict wind power output and photovoltaic power output for future periods, combining historical load data, adopting a load prediction model that considers intraday periodicity and inter-day correlation to predict the power load demand for the same period in the future, and using the predicted data as input to construct and solve a rolling optimization problem with the goal of minimizing daily operating costs, applying the first instruction in the solved control instruction sequence to the actual system, and repeating the prediction and optimization process when entering the next control cycle; The computational performance indicators include the calculation of new energy absorption rate, carbon emission reduction, economic efficiency, and safety. The formula for calculating the renewable energy absorption rate is as follows: in, For the new energy consumption rate, Let t be the wind power output. Photovoltaic output at time t Let t be the amount of wind and solar power curtailed. For time indexing, The total number of time steps is used to verify whether... Greater than or equal to 92%; The formula for calculating carbon emission reduction targets is as follows: in, This represents the total annual carbon emissions. Carbon emission factors from purchasing electricity from the power grid The carbon emission factor of the electrolyzer operation, As a carbon emission reduction factor for fuel cells, Let t be the power purchased by the power grid. Let be the power of the electrolytic cell at time t. Let be the output power of the fuel cell at time t. For time step; Economic indicators are expressed as follows: in, For total lifecycle cost, For initial investment costs, For operation and maintenance costs, For equipment replacement costs, The residual value of the equipment; Safety indicators are expressed as follows: in, Let t be the pressure of the hydrogen storage tank. The minimum safe pressure for the hydrogen storage tank, This is the maximum safe pressure for the hydrogen storage tank. Let t be the state of charge of the lithium battery. This represents the lowest state of charge for a lithium battery. This represents the highest state of charge of the lithium battery.

7. The multi-objective optimization planning method for wind, solar, and hydrogen storage systems as described in claim 6, characterized in that: The feedback adjustment includes extracting data on equipment start-up and shutdown frequency, wind and solar power curtailment time distribution, and energy storage equipment status change trajectory during the operation simulation process when any performance indicator fails to reach the preset threshold. The data is then fed back to the improved non-dominated sorting genetic algorithm optimization in the planning layer to adjust the weight coefficients of the corresponding non-compliant indicators in the multi-objective function or modify the capacity constraint boundary of the equipment. Based on the adjusted parameters, the optimization from the planning layer to the operation layer simulation is re-executed, and the adjustments are repeated until all performance indicators meet the preset requirements, and an implementable system capacity configuration and operation strategy scheme is output.

8. A multi-objective optimization planning system for wind, solar, and hydrogen storage systems, employing the multi-objective optimization planning method for wind, solar, and hydrogen storage systems as described in any one of claims 1 to 7, characterized in that, It includes a data acquisition and preprocessing module, a system modeling and simulation module, a multi-objective optimization modeling module, a hierarchical collaborative solution module, and a verification and feedback adjustment module; The data acquisition and preprocessing module is used to collect multi-source data on wind and solar resources, load demand, equipment parameters, and economic and environmental parameters, fill in missing data through interpolation, and identify and remove abnormal data based on the physical limits of the equipment. The system modeling and simulation module is used to transform the physical system into a mathematical model and to simulate and predict the system's operating state under different conditions. The multi-objective optimization modeling module is used to establish a multi-objective function with the core objectives of minimizing the total life cycle cost, minimizing carbon emissions, and maximizing the renewable energy absorption rate, and to set the power balance constraints, equipment operation constraints, and planning capacity constraints that the system must comply with. The hierarchical collaborative solution module is used to decouple long-term planning problems from short-term operation control problems through a hierarchical strategy, and to perform collaborative optimization. The verification and feedback adjustment module is used to simulate the operation of the planning scheme throughout the year based on historical data, calculate the new energy absorption rate, carbon emission reduction, full life cycle cost and safety indicators, and compare them with the preset targets. When any indicator fails to meet the target, the problems exposed in the operation simulation are fed back to the optimization algorithm of the planning layer, the target weight or capacity parameter is adjusted, and the optimization iteration is restarted until the optimal scheme that meets the target is output.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the multi-objective optimization planning method for the wind, solar, and hydrogen storage system as described in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the multi-objective optimization planning method for the wind, solar, and hydrogen storage system as described in any one of claims 1 to 7.

Images