Electricity-hydrogen production and charging integrated station planning method and system based on artificial intelligence and medium

Abstract

The invention discloses an electricity-hydrogen production and charging integrated station planning method and system based on artificial intelligence and a medium, which are used in the field of electricity-hydrogen production and charging integrated station planning, and the electricity-hydrogen production and charging integrated station planning method comprises the following steps: obtaining energy climate data of each traffic node in an electricity-hydrogen production and charging integrated station; pre-constructing a neural network fusion model, and inputting the energy climate data into the neural network fusion model to obtain a space-time distribution prediction model of the electric hydrogen energy demand; analyzing the energy flow of the electricity-hydrogen production and charging integrated station, and constructing a multi-target operation planning model of coupling of a power grid and a traffic network; and utilizing the spatial-temporal distribution prediction model to predict the spatial-temporal distribution of the electro-hydrogen energy charging demand, and fusing the prediction result with the output data of the multi-target operation planning model to obtain the optimal site and equipment capacity planning of the electro-hydrogen production and charging integrated station. According to the invention, management and maintenance of data of the electro-hydrogen production and charging integrated station are realized, and efficient operation and management of the integrated energy station are promoted.

Description

Technical Field

[0001] The present invention relates to the field of planning of an integrated electric hydrogen production and filling station, and in particular to an artificial intelligence-based integrated electric hydrogen production and filling station planning method, system and medium. Background Art

[0002] As a key force in promoting green transformation in the transportation sector, new energy vehicles are leading the new trend of low-carbon travel. In order to achieve this transformation, the development of integrated electric hydrogen production and filling stations has become a crucial support point, aiming to promote the efficient use of clean energy by integrating electric vehicle charging and hydrogen fuel vehicle refueling functions. However, the development of this field faces challenges such as lagging construction of charging piles and hydrogen refueling station infrastructure, difficulty in predicting electric hydrogen demand, and complexity in energy station planning.

[0003] The concept of an integrated electric hydrogen production and filling station integrates the production, conversion and supply processes of electricity and hydrogen, aiming to reduce transportation costs and safety risks through on-site hydrogen production, and reduce operating costs through shared infrastructure. This is especially true for the current market environment where electric vehicles are dominant but the development of hydrogen fuel cell vehicles needs to be taken into account. It not only requires accurate prediction of the temporal and spatial demand for electric and hydrogen by electric vehicles and hydrogen fuel cell vehicles, but also involves complex system design, including efficient conversion, storage and flexible allocation of energy to optimize site selection and scale setting.

[0004] Therefore, the research focus has shifted to developing innovative predictive models to accurately grasp the future dynamics of electric hydrogen demand, while exploring how to efficiently integrate multiple links such as grid power supply, electric vehicle charging, on-site hydrogen preparation, compression storage and refueling to form an electric hydrogen supply system that is both economical and safe. This is not only an innovation of existing energy facilities, but also a key path to achieving dual carbon goals and deep decarbonization in the transportation sector.

[0005] Currently, no effective solution has been proposed for the problems in the related technologies. Summary of the invention

[0006] In order to overcome the above problems, the present invention aims to propose an artificial intelligence-based electric hydrogen production and filling integrated station planning method, system and medium, with the aim of exploring how to efficiently integrate multiple links such as grid power supply, electric vehicle charging, on-site hydrogen preparation, compression storage and filling to form an electric hydrogen supply system that is both economical and safe.

[0007] To this end, the specific technical solution adopted by the present invention is as follows:

[0008] According to a first aspect of an embodiment of the present invention, a method for planning an integrated hydrogen production and filling station based on artificial intelligence is provided.

[0009] In one embodiment, the artificial intelligence-based electric hydrogen filling integrated station planning method includes the following steps:

[0010] S1. Obtain energy climate data of each traffic node in the electric hydrogen production and filling integrated station;

[0011] S2. Pre-build a neural network fusion model, and input energy climate data into the neural network fusion model to obtain a spatiotemporal distribution prediction model for electric hydrogen energy demand;

[0012] S3. Analyze the energy flow of the electric hydrogen production and filling station, and build a multi-objective operation planning model coupled with the power grid and transportation network;

[0013] S4. Use the spatiotemporal distribution prediction model to predict the spatiotemporal distribution of electric hydrogen charging demand, and integrate the prediction results with the output data of the multi-objective operation planning model to obtain the optimal site and equipment capacity planning of the electric hydrogen production and charging integrated station;

[0014] S5. Use smart contracts to record the optimal site and equipment capacity planning of the electric hydrogen filling station into the blockchain.

[0015] In one embodiment, obtaining energy climate data of each traffic node in the electric hydrogen production and filling integrated station includes the following steps:

[0016] S11. Obtain the historical annual ownership data of electric vehicles and hydrogen fuel vehicles, daily charging demand data and meteorological data at each traffic node in the electric hydrogen production and filling integrated station;

[0017] S12. Calculate the Pearson correlation coefficient of the annual ownership data of historical electric vehicles and hydrogen fuel cell vehicles, daily charging demand data, and meteorological data;

[0018] S13. According to the calculation result of the Pearson correlation coefficient, the meteorological factors that do not meet the preset threshold value in the Pearson correlation coefficient are eliminated.

[0019] In one embodiment, the formula for calculating the Pearson correlation coefficient of the annual ownership data of historical electric vehicles and hydrogen fuel vehicles, daily charging demand data, and meteorological data is:

[0020]

[0021] In the formula, t represents a certain moment; T represents the time series P t , Q t The total number of data; P t , Q t Represents the value of a time series at a certain moment t; Represent the time series P t , Q t The average value of XY It indicates the correlation between variable x and variable y.

[0022] In one embodiment, a neural network fusion model is pre-built, and energy climate data is input into the neural network fusion model to obtain a spatiotemporal distribution prediction model of electric hydrogen energy demand, including the following steps:

[0023] S21. Constructing the convolution layer, forget gate, input gate and output gate of the network fusion model in sequence based on the convolutional neural network model and the long short-term memory network model;

[0024] S22, normalizing the annual ownership data of historical electric vehicles and hydrogen fuel vehicles, daily charging demand data, and meteorological data after correlation analysis;

[0025] S23, input the normalized historical annual ownership data of electric vehicles and hydrogen fuel vehicles, daily charging demand data and meteorological data into the network fusion model, and establish a spatiotemporal distribution prediction model for electric hydrogen energy demand;

[0026] S24. Input relevant historical data and predicted real-time data into the spatiotemporal distribution prediction model, and combine it with the adaptive learning mechanism to build a dynamic spatiotemporal distribution prediction model for electric hydrogen energy demand.

[0027] In one embodiment, relevant historical data and predicted real-time data are input into the spatiotemporal distribution prediction model, and combined with the adaptive learning mechanism, the dynamic spatiotemporal distribution prediction model of electric hydrogen energy demand is constructed, including the following steps:

[0028] S241, inputting relevant historical data and predicted real-time data into the spatiotemporal distribution prediction model;

[0029] S242, using an adaptive learning mechanism to update parameters of the spatiotemporal distribution prediction model in real time;

[0030] S243. Construct a dynamic spatiotemporal distribution prediction model of electric hydrogen energy demand based on real-time updated parameters, and conduct real-time dynamic prediction of regional electric hydrogen energy demand.

[0031] In one embodiment, analyzing the energy flow of the electric hydrogen production and filling integrated station and constructing a multi-objective operation planning model coupled with the power grid and the transportation network includes the following steps:

[0032] S31. Based on the energy flow relationship of the electric hydrogen production and filling station, design the coupling conditions at the power grid and transportation network levels, and establish a multi-objective function model with economic indicators integrated;

[0033] S32. Based on the energy flow relationship and multi-objective function model of the electric hydrogen production and filling integrated station, a multi-objective operation planning model of the electric hydrogen production and filling integrated station with constraints is constructed;

[0034] S33. Use the improved particle swarm optimization algorithm to optimize the nonlinear constraints of the multi-objective operation planning model.

[0035] In one embodiment, based on the energy flow relationship of the electric hydrogen production and filling integrated station, the coupling conditions at the power grid and transportation network levels are designed, and a multi-objective function model integrating economic indicators is established, including the following steps:

[0036] S311. Construct the operation planning framework of the electric hydrogen production and filling integrated station;

[0037] S312. Analyze the relationship between the distribution network and the transportation network, and between electricity and hydrogen energy in the operation planning framework;

[0038] S313. Based on the analysis results, the objectives of the power distribution network and the transportation network in the planning area are coupled, and the economic indicators of the integrated hydrogen production and filling station are minimized as the planning goal, and a multi-objective function model covering various economic indicators is constructed.

[0039] In one embodiment, the economic indicators include investment cost, operating cost and grid level targets.

[0040] In one embodiment, the constraints include: equipment constraints, traffic network constraints, and power system flow constraints.

[0041] In one embodiment, optimizing the nonlinear constraints of a multi-objective operation planning model using an improved particle swarm optimization algorithm includes the following steps:

[0042] S331, initializing the distribution of electricity and hydrogen loads, land price distribution and grid structure of the transportation network in the multi-objective operation planning model;

[0043] S332, selecting a preset number of electric hydrogen production and filling integrated stations as the initial positions of the particle group;

[0044] S333, obtaining charging load data of the electric hydrogen production and charging integrated station captured by each station according to the initial site selection, and obtaining a fitness function;

[0045] S334, repeating the steps of S332-S333, updating the fitness function through iteration of the particle swarm to obtain the next generation of particles, and outputting the fitness function when the fitness function no longer changes or reaches the maximum number of iterations.

[0046] In one embodiment, the spatiotemporal distribution of the electric hydrogen charging demand is predicted using a spatiotemporal distribution prediction model, and the prediction results are integrated with the output data of the multi-objective operation planning model to obtain the optimal site and equipment capacity planning of the electric hydrogen production and filling integrated station, including the following steps:

[0047] S41, integrating the prediction results of the real-time dynamic electric hydrogen energy demand prediction model with the optimized multi-objective operation planning model;

[0048] S42. Use the subjective weighting method to determine the fused weights and determine the optimal site and equipment capacity planning for the electric hydrogen production and filling integrated station.

[0049] In one embodiment, the subjective weighting method is used to determine the fused weights, and determining the optimal site and equipment capacity planning of the electric hydrogen production and filling integrated station includes the following steps:

[0050] S421, weighted fusion of each prediction result and optimization result data through a neural network fusion model based on a subjective weighting method, and determining their respective weights;

[0051] S422. Determine the optimal site and equipment capacity planning of the electric hydrogen production and filling integrated station based on the weight ratio.

[0052] In one embodiment, the formula for weighted fusion of each prediction result and optimization result data through a neural network fusion model based on a subjective weighting method is:

[0053] F=w d ×D+w o ×O;

[0054] In the formula, F represents the fusion data result, w d represents the weight of the prediction result determined by the subjective weighting method, w o It represents the weight of the optimization result determined by the subjective weighting method, D represents the prediction result data, and O represents the optimization result data.

[0055] According to a second aspect of an embodiment of the present invention, an artificial intelligence-based electric hydrogen production and filling integrated station planning system is provided.

[0056] In one embodiment, the electric hydrogen production and filling integrated station planning system includes: a data acquisition module, a model building and prediction module, an energy flow and planning module, a planning integration module and a blockchain security module;

[0057] A data acquisition module is used to obtain energy and climate data of each traffic node in the electric hydrogen production and filling integrated station;

[0058] The model building and prediction module is used to pre-build a neural network fusion model and input energy climate data into the neural network fusion model to obtain a spatiotemporal distribution prediction model for electric hydrogen energy demand;

[0059] Energy flow and planning module, which is used to analyze the energy flow of the electric hydrogen filling station and build a multi-objective operation planning model coupled with the power grid and transportation network;

[0060] The planning integration module is used to predict the spatiotemporal distribution of electric hydrogen charging demand using the spatiotemporal distribution prediction model, and integrate the prediction results with the output data of the multi-objective operation planning model to obtain the optimal site and equipment capacity planning of the electric hydrogen production and charging integrated station;

[0061] The blockchain security module is used to record the optimal site and equipment capacity planning of the electric hydrogen filling station into the blockchain using smart contracts.

[0062] According to a third aspect of an embodiment of the present invention, a computer-readable storage medium is provided.

[0063] In one embodiment, a computer program is stored on a computer-readable storage medium, and when the computer program is executed by a processor, the steps of the above method are implemented.

[0064] Compared with the prior art, the present application has the following beneficial effects: In view of the fact that the charging demand of electric vehicles and hydrogen fuel cell vehicles is affected by the layout planning of basic charging facilities, the present invention proposes a planning method for an electric hydrogen production and filling integrated station combined with an artificial intelligence algorithm, collects relevant historical data on electric vehicles, hydrogen fuel cell vehicles and meteorological factors in the area, uses the Pearson correlation coefficient to perform correlation analysis on the above historical data in the area, eliminates meteorological factors with low correlation coefficients, and then builds a prediction model for the spatiotemporal distribution of electric hydrogen charging demand based on a deep learning network; constructs a dynamic electric hydrogen energy demand prediction model based on real-time demand data feedback and updates, and coordinates the production and consumption of electric hydrogen energy; secondly, analyzes the energy production, transportation and consumption structure of the electric hydrogen production and filling integrated station, constructs the energy flow relationship of the electric hydrogen energy coupling, and analyzes the integrated energy The operation planning framework of the source station is established; then, a multi-objective function model of the regional electric hydrogen production and filling integrated station is established with the goal of minimizing economic costs such as investment, operation, and network loss, taking into account constraints such as equipment constraints, transportation network constraints, and power system flow constraints, a multi-objective planning model of the energy station is constructed, and the improved particle swarm optimization algorithm is used to solve the model; then, combined with the regional real-time dynamic electric hydrogen energy demand forecast results and the integrated energy station operation optimization results, combined with the fusion model based on the subjective weighting method, the optimal site and equipment capacity planning results of the electric hydrogen production and filling integrated station in the region are obtained, which promotes the use of stations to bring cars, promotes the development of electric vehicles and hydrogen fuel cell vehicles, and improves the optimal utilization of electric hydrogen energy types. Finally, combined with the existing blockchain technology, the management and maintenance of the electric hydrogen production and filling station data is realized, and the efficient operation and management of the integrated energy station is promoted. BRIEF DESCRIPTION OF THE DRAWINGS

[0065] The above characteristics, features and advantages of the present invention and their implementation methods and methods will become more clearly understood in conjunction with the following description of the embodiments, which are described in detail in conjunction with the accompanying drawings. Herein, a schematic diagram is shown:

[0066] Figure 1 is a flow chart of a method for planning an electric hydrogen production and filling integrated station based on artificial intelligence according to an embodiment of the present invention;

[0067] Figure 2 is a principle block diagram of an artificial intelligence-based electric hydrogen production and filling integrated station planning system according to an embodiment of the present invention;

[0068] Figure 3 The figure is a schematic diagram showing the structure of a computer device according to an exemplary embodiment.

[0069] In the figure:

[0070] 1. Data acquisition module; 2. Model building and prediction module; 3. Energy flow and planning module; 4. Planning integration module; 5. Blockchain security module. DETAILED DESCRIPTION

[0071] In order to enable those skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are only part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of this application.

[0072] According to an embodiment of the present invention, a method, system and medium for planning an integrated hydrogen production and filling station based on artificial intelligence are provided.

[0073] The present invention will now be further described with reference to the accompanying drawings and specific embodiments. Figure 1 An embodiment of the artificial intelligence-based electric hydrogen production and filling integrated station planning method of the present invention is shown.

[0074] In this optional embodiment, the method for planning an integrated hydrogen production and filling station includes the following steps:

[0075] S1. Obtain energy climate data of each traffic node in the electric hydrogen filling station.

[0076] In this optional embodiment, obtaining energy climate data of each traffic node in the electric hydrogen production and filling integrated station includes the following steps:

[0077] S11. Obtain the historical annual ownership data of electric vehicles and hydrogen fuel vehicles, daily charging demand data and meteorological data at each traffic node in the electric hydrogen production and filling integrated station;

[0078] S12. Calculate the Pearson correlation coefficient of the annual ownership data of historical electric vehicles and hydrogen fuel cell vehicles, daily charging demand data, and meteorological data;

[0079] S13. According to the calculation result of the Pearson correlation coefficient, the meteorological factors that do not meet the preset threshold value in the Pearson correlation coefficient are eliminated.

[0080] In this optional embodiment, the formula for calculating the Pearson correlation coefficient of the annual ownership data of historical electric vehicles and hydrogen fuel vehicles, daily charging demand data, and meteorological data is:

[0081]

[0082] In the formula, t represents a certain moment; T represents the time series P t , Q t The total number of data; P t , Q t Represents the value of a time series at a certain moment t; Represent the time series P t , Q t The average value of XY It indicates the correlation between variable x and variable y.

[0083] It should be explained that the historical data on the annual ownership of electric vehicles and hydrogen fuel vehicles in the cities where the planning area is located are extracted, including: the annual ownership data of electric vehicles, the annual ownership of hydrogen fuel vehicles, etc.;

[0084] Extract historical meteorological data of the planning area, including temperature, humidity, air pressure, wind speed, etc.;

[0085] Extract historical daily charging demand data of electric vehicles and hydrogen fuel vehicles at different traffic nodes in the planning area, including: historical daily charging demand of electric vehicles at different traffic nodes, daily charging demand of hydrogen fuel vehicles at different traffic nodes, etc.

[0086] A network fusion model (CNN-LSTM model) is constructed based on the convolutional neural network model and the long short-term memory network model.

[0087] In this optional embodiment, constructing a network fusion model based on a convolutional neural network model and a long short-term memory network model includes: sequentially constructing a convolutional layer, a forgetting gate, an input gate, and an output gate of the network fusion model.

[0088] It should be explained that the network fusion model is a CNN-LSTM model, and the convolution layer of the CNN-LSTM model is constructed;

[0089]

[0090] Where: X t is the input feature of the CNN-LSTM model; F t For X t Features extracted by the convolutional layer; is the convolution operation; W g is the weight matrix of the convolutional layer; β is the bias vector; f Relu (x) is the Relu activation function.

[0091] Construct the forget gate of the CNN-LSTM model;

[0092] f t =σ(W f *[F t ,H t-1 ] T +B f );

[0093] Where: F t represents the input value of the hidden layer at time t; H t-1 represents the output value of the hidden layer at time t-1; σ is the activation function sigmoid, and the output value is between 0 and 1; W f is the weight matrix of the forget gate; B f is the forget gate bias term; f t is the output value of the forget gate at time t.

[0094] Construct the input gate of the CNN-LSTM model;

[0095]

[0096] Where tanh represents the hyperbolic tangent activation function, and the output value is between -1 and 1; W i is the weight matrix of the input gate; B i is the input gate bias term; i t is the output value of the input gate at time t; C′ t is the temporary state of the memory unit at time t; B m is the memory unit bias term; ⊙ represents the corresponding multiplication operation of matrix elements; C t is the output of the memory unit at time t, C t-1 is the output of the memory unit at time t-1.

[0097] Construct the output gate of the CNN-LSTM model;

[0098]

[0099] Where: W o is the weight matrix of the output gate; B o is the output gate bias term; O t is the output value of the output gate at time t; H t Represents the output value of the hidden layer at time t.

[0100] S2. Pre-build a neural network fusion model (CNN-LSTM model), and input energy climate data into the neural network fusion model to obtain a spatiotemporal distribution prediction model for electric hydrogen energy demand.

[0101] In this optional embodiment, a neural network fusion model is pre-built, and energy climate data is input into the neural network fusion model to obtain a spatiotemporal distribution prediction model for electric hydrogen energy demand, including the following steps:

[0102] S21. Constructing the convolution layer, forget gate, input gate and output gate of the network fusion model in sequence based on the convolutional neural network model and the long short-term memory network model;

[0103] S22, normalizing the annual ownership data of historical electric vehicles and hydrogen fuel vehicles, daily charging demand data, and meteorological data after correlation analysis;

[0104] S23, input the normalized historical annual ownership data of electric vehicles and hydrogen fuel vehicles, daily charging demand data and meteorological data into the network fusion model, and establish a spatiotemporal distribution prediction model for electric hydrogen energy demand;

[0105] S24. Input relevant historical data and predicted real-time data into the spatiotemporal distribution prediction model, and combine it with the adaptive learning mechanism to build a dynamic spatiotemporal distribution prediction model for electric hydrogen energy demand.

[0106] In this optional embodiment, relevant historical data and predicted real-time data are input into the spatiotemporal distribution prediction model, and combined with the adaptive learning mechanism, the dynamic spatiotemporal distribution prediction model of electric hydrogen energy demand is constructed, including the following steps:

[0107] S241. Input relevant historical data and predicted real-time data into the spatiotemporal distribution prediction model. The relevant historical data includes the ownership data of electric vehicles and hydrogen fuel vehicles in typical years, daily charging demand data, etc. The data is obtained by consulting relevant documents in the transportation field of the statistical area. The predicted real-time data is obtained by fitting and predicting the spatiotemporal distribution model of electric hydrogen energy demand on typical days based on the relevant historical data. The historical data and real-time predicted data provide a basis for the dynamic adjustment of subsequent model parameters.

[0108] S242, using an adaptive learning mechanism to update parameters of the spatiotemporal distribution prediction model in real time;

[0109] It should be noted that the real-time updating of the parameters of the spatiotemporal distribution prediction model using the adaptive learning mechanism includes the following steps:

[0110] The input data is cleaned and preprocessed, and missing values, outliers and noise data are processed to ensure data quality. At the same time, adaptive learning mechanisms are used, such as online learning, which is updated in real time during runtime. When new data arrives, the model is trained online using the new data, or a sliding window mechanism is used to define a fixed window size. When new data arrives, the window slides to move the earliest data out of the window, retaining the latest data, and re-learning the data in the window. Then, based on the data characteristics learned by the model, the parameters of the prediction model, such as the learning rate, regularization coefficient, etc., are updated in real time to provide more optimized and adaptive model support for the prediction of electric hydrogen energy demand in subsequent time periods.

[0111] S243. Construct a dynamic spatiotemporal distribution prediction model of electric hydrogen energy demand based on real-time updated parameters, and conduct real-time dynamic prediction of regional electric hydrogen energy demand.

[0112] In addition, it should be noted that the real-time dynamic forecast of regional electric hydrogen energy demand includes the following steps:

[0113] Collect regional automobile-related historical data and meteorological data, select data based on the degree of correlation, and then clean and preprocess the data. Input the processed data into the dynamic spatiotemporal distribution prediction model for initial training. At the same time, apply an adaptive learning mechanism to the model to achieve online real-time updating of the model. Finally, use the model to predict the electric hydrogen energy demand for the corresponding time period.

[0114] It should be explained that the daily charging demand data, ownership data and meteorological data of electric vehicles and hydrogen fuel vehicles are normalized;

[0115]

[0116] In the formula: z represents the data after normalization; Indicates the actual value of a certain type of data; and Indicates the minimum and maximum values ​​of a certain type of data.

[0117] The normalized data is input into CNN-LSTM to establish a spatiotemporal distribution prediction model for the charging demand of electric vehicles and hydrogen fuel cell vehicles. The model input includes the meteorological factors at time t on the predicted day, the historical electric vehicle and hydrogen fuel cell vehicle ownership data for the month in which the day is predicted, and the charging demand and hydrogen filling demand of a certain traffic node at time t 7 days before the day. The model output is the predicted charging demand and hydrogen filling demand of a certain traffic node at time t on the predicted day.

[0118] Integrate relevant historical data and predicted real-time data into the prediction model, and combine it with an adaptive learning mechanism to allow the model to update parameters when receiving new data, so that the model can learn gradually without retraining, and automatically adjust model parameters such as learning rate and regularization coefficient based on model performance feedback, so that the model has the ability to self-learn and predict and evaluate new scenarios, thereby realizing real-time dynamic demand forecasting of regional electric hydrogen energy.

[0119] In the early stages of iteration, the population mainly learns from excellent individuals and can quickly move closer to the optimal individual. As the iteration proceeds, δ gradually increases, the ability of individuals to maintain their own state increases, slowing down the speed of approaching the optimal individual, and improving their ability to maintain their own state and learn.

[0120] X′ i (t) = δ × X i (t)+rand(0,1)×(X t (t)-β×X m (t));

[0121] In the formula, δ is an adaptive nonlinear factor, t is the current iteration number, and X t (t) is the optimal individual of the population, X i (t) is the original state of the individual, X′ i (t) is the individual update status.

[0122] The electric hydrogen production and filling station is a multifunctional energy supply station that integrates power distribution, electric vehicle charging, hydrogen production and storage, and hydrogen filling. This energy station not only provides a comprehensive energy service, but also, as a key node in energy transmission, connects the energy production and consumption ends, realizing the coordinated operation of electricity and hydrogen energy. Under this architecture, effective coupling is achieved between the distribution network and the transportation network, and between electricity and hydrogen energy, which not only promotes the conversion between electricity and hydrogen energy, but also supports the direct application of electricity, meeting the daily energy supply needs of electric vehicles and hydrogen fuel cell vehicles. In this way, the electric hydrogen production and filling station provides strong support for the efficient use of energy and the promotion of clean energy.

[0123] S3. Analyze the energy flow of the integrated hydrogen production and filling station, and build a multi-objective operation planning model that couples the power grid and transportation network.

[0124] In this optional embodiment, analyzing the energy flow of the electric hydrogen production and filling integrated station and constructing a multi-objective operation planning model coupled with the power grid and the transportation network includes the following steps:

[0125] S31. Based on the energy flow relationship of the electric hydrogen production and filling station, design the coupling conditions at the power grid and transportation network levels, and establish a multi-objective function model with economic indicators integrated;

[0126] S32. Based on the energy flow relationship and multi-objective function model of the electric hydrogen production and filling integrated station, a multi-objective operation planning model of the electric hydrogen production and filling integrated station with constraints is constructed;

[0127] S33. Use the improved particle swarm optimization algorithm to optimize the nonlinear constraints of the multi-objective operation planning model.

[0128] In this optional embodiment, based on the energy flow relationship of the electric hydrogen production and filling integrated station, the coupling conditions at the power grid and transportation network levels are designed, and a multi-objective function model integrating economic indicators is established, including the following steps:

[0129] S311. Construct the operation planning framework of the electric hydrogen production and filling integrated station;

[0130] It should be noted that the electric hydrogen production and filling station is a multifunctional energy supply station that integrates power distribution, electric vehicle charging, hydrogen production and storage, and hydrogen filling. The multifunctional energy supply station not only provides a comprehensive energy service, but also, as a key node in energy transmission, connects the energy production and consumption ends, and realizes the coordinated operation of electricity and hydrogen energy. The operation planning framework of the electric hydrogen production and filling station includes the flow of electric hydrogen energy in the electric hydrogen station, the optimal site selection of the electric hydrogen station, the equipment capacity planning and data management of the electric hydrogen station, etc. To construct the operation planning framework of the electric hydrogen production and filling station, it is necessary to analyze the specific implementation requirements and contents of each link in the integrated station, such as analyzing the overall flow of electric hydrogen energy. Electric energy is provided by the corresponding distribution network nodes, and hydrogen energy is transmitted between various devices (electrolyzer to compressor to hydrogen storage equipment and finally enters hydrogen fuel cell vehicles), as well as optimizing the site selection of the integrated station according to the temporal and spatial distribution of electric hydrogen demand in the analysis area. The various links and implementation requirements of the integrated station are studied to construct the operation planning framework of the electric hydrogen production and filling station.

[0131] S312. Analyze the relationship between the distribution network and the transportation network, and between electricity and hydrogen energy in the operation planning framework;

[0132] S313. Based on the analysis results, the objectives of the power distribution network and the transportation network in the planning area are coupled, and the economic indicators of the integrated hydrogen production and filling station are minimized as the planning goal, and a multi-objective function model covering various economic indicators is constructed.

[0133] In this optional embodiment, the economic indicators include investment cost, operating cost and grid-level targets.

[0134] In this optional embodiment, the constraints include: equipment constraints, transportation network constraints, and power system flow constraints.

[0135] In this optional embodiment, optimizing the nonlinear constraints of the multi-objective operation planning model using the improved particle swarm optimization algorithm includes the following steps:

[0136] S331, initializing the distribution of electricity and hydrogen loads, land price distribution and grid structure of the transportation network in the multi-objective operation planning model;

[0137] S332, selecting a preset number of electric hydrogen production and filling integrated stations as the initial positions of the particle group;

[0138] S333, obtaining charging load data of the electric hydrogen production and charging integrated station captured by each station according to the initial site selection, and obtaining a fitness function;

[0139] S334, repeating the steps of S332-S333, updating the fitness function through iteration of the particle swarm to obtain the next generation of particles, and outputting the fitness function when the fitness function no longer changes or reaches the maximum number of iterations.

[0140] It needs to be explained that based on the operation planning framework of the integrated energy station, various factors such as energy production, transportation, storage and consumption are taken into consideration, the relevant goals of the regional power network and transportation network are coupled, and the minimum economic cost is taken as the planning goal. A multi-objective function model is constructed that covers economic indicators such as energy station investment cost, operating cost, and network loss.

[0141] The economic indicators of the integrated hydrogen production and filling station are expressed as: investment cost, operating cost and grid level target; among which,

[0142] 1. Investment costs include:

[0143] The investment cost W1 includes the cost of various equipment and the cost of construction land. The planning area is divided into residential area, commercial area and industrial area.

[0144]

[0145] w m,cdz =α cdz ·N m,cdz

[0146] w m,zqj =α zqj ·N m,zqj ;

[0147] Where: M is the planned number of energy stations; τ is the discount rate; T0 is the operation planning period; w m.AEL 、w m.press , w m.land 、w m.cdz 、w m.zqjare the investment and construction costs of the electrolyzer, compressor, hydrogen storage equipment, land, charging pile, and hydrogen injection machine of the mth charging station; c AEL 、c press , α zq j, α cdz They are the construction costs of electrolyzers, compressors, hydrogen storage equipment, hydrogen injection machines and charging piles per unit capacity or quantity respectively. They are respectively the rated installed capacities of the electrolyzer, compressor and hydrogen storage equipment of the mth electric hydrogen production and filling integrated station. is the land price of the functional area of ​​the city where the mth energy station belongs; S ref , They are the reference floor space and reference daily hydrogenation load of the energy station; is the hydrogenation load of the energy station during period t; T is the total number of periods within a day; N m,cdz , N m,zqj are the number of charging piles and hydrogen injection machines in the mth energy station respectively.

[0148] 2. Operating costs include:

[0149] Operating costs W2 include equipment costs, electricity and water costs, and worker costs.

[0150]

[0151] w m.op =Mc op ;

[0152] Where: w m.el 、w m.w 、w m.op They are electricity, water costs and worker costs; Time-of-use electricity price; are the electric power of the electrolyzer and the compressor at time t; c w is the industrial water price; η w.AEL is the coefficient of purified water for electrolytic cell; c op The wage cost of workers at a single energy station.

[0153] 3. Grid-level goals include:

[0154] The grid-level objectives include voltage deviation and grid loss costs, which are expressed as follows. The grid evaluation takes the mean of voltage and grid loss.

[0155]

[0156] Where: S is the voltage deviation rate of the grid node; W3 is the grid loss cost; is the voltage value of grid node n at time t; V0 is the average value of grid voltage; N is the number of nodes; T is the total number of time periods; I ij,t is the current between branches ij during time period t; R ij is the resistance of branch ij; u(i) is the set of all nodes connected to node i and located downstream of node i; σ L is the unit network loss cost of the system; P loss,ij,t is the active power loss of branch ij in time period t; D is the planned number of days.

[0157] According to the multi-objective function model of the energy station established above, the energy flow relationship of the electric hydrogen production and filling integrated station, the transportation and distribution of electricity, and the production, compression, storage and filling of hydrogen are considered, combined with the equipment used in the key links (i.e., the electrolyzer participates in the production of hydrogen, the compressor is used for hydrogen compression, the hydrogen storage equipment is used for hydrogen storage, etc.), and the coupling constraints of the transportation network and the power network are taken into account. A multi-objective planning model integrating equipment constraints, transportation network constraints, power system flow constraints and other constraints is constructed.

[0158] The device constraints can be expressed as:

[0159] 1. The electrolytic cell model can be expressed as:

[0160]

[0161] Where: is the electric power of the electrolytic cell at time t; is the electric power corresponding to the hydrogen production of the electrolyzer at time t; is the hydrogen production of the electrolyzer at time t; η AEL is the electrolysis efficiency of the electrolytic cell; is the calorific value of hydrogen.

[0162] 2. The compressor model can be expressed as:

[0163]

[0164] Where: is the power of the compressor at time t; is the amount of hydrogen flowing into the compressor at time t; F is the power consumed by the compressor at the reference working pressure; press , F are the pressure of the compressor under normal working pressure and reference working pressure respectively; ini is standard atmospheric pressure. press is the hydrogen dissipation rate from electrolyzer to compressor. is the hydrogen production of the electrolyzer at time t.

[0165] 3. The hydrogen storage equipment model can be expressed as:

[0166]

[0167] Where: are the hydrogen capacity in the hydrogen storage device at time t and time t+1 respectively; is the hydrogen dissipation rate from the compressor to the hydrogen storage device; is the amount of hydrogen flowing into the compressor at time t; is the amount of hydrogen compressed into the hydrogen storage device at time t; is the hydrogenation load at time t.

[0168] The traffic network constraints can be expressed as:

[0169] 1. The distance constraint between the electric hydrogen filling stations can be expressed as:

[0170] L min <L<L Z ;

[0171] Where: L is the actual distance between two energy stations; L min is the minimum distance between two charging stations; L Z The driving range of electric vehicles and hydrogen fuel cell vehicles with the minimum remaining fuel.

[0172] 2. The overlapping area constraints of the electric hydrogen production and filling station can be expressed as:

[0173] In the layout of the transportation network, there is an overlap in the service range of adjacent energy stations. The node charging load in the overlapping area will be captured repeatedly, and the number of repeatedly captured nodes is limited as shown in the formula.

[0174] ΔM≤ΔM max ;

[0175] Where: ΔM max The maximum number of hydrogenation load nodes repeatedly captured in the overlapping service areas of energy stations; ΔM is the number of hydrogenation load nodes repeatedly captured in the overlapping service areas of energy stations.

[0176] 3. The land occupation restriction constraint can be expressed as:

[0177] A certain safety distance is required between the hydrogen production station, compression station and hydrogen storage station of an energy station. Therefore, as the number of planned energy stations increases, the area of ​​a single station cannot be reduced indefinitely, and the minimum floor space constraint is as follows.

[0178] S ref.min ≤S;

[0179] Where: S ref.min is the reference minimum floor area of ​​the energy station; S is the actual floor area of ​​the energy station.

[0180] The power system flow constraint can be expressed as:

[0181] U t,i,min ≤U t,i ≤U t,i,max ;

[0182] I ij,min ≤|I ij |≤I ij,max ;

[0183] Where: U t,i,min , U t,i,max are the minimum voltage and maximum voltage allowed at node i at time t; I ij,min ,I ij,max are the minimum and maximum currents allowed in branch ij respectively.

[0184] S4. Use the spatiotemporal distribution prediction model to predict the spatiotemporal distribution of electric hydrogen charging demand, and integrate the prediction results with the output data of the multi-objective operation planning model to obtain the optimal site and equipment capacity planning of the electric hydrogen production and filling integrated station.

[0185] In this optional embodiment, the spatiotemporal distribution of the electric hydrogen charging demand is predicted using the spatiotemporal distribution prediction model, and the prediction results are integrated with the output data of the multi-objective operation planning model to obtain the optimal site and equipment capacity planning of the electric hydrogen production and filling integrated station, including the following steps:

[0186] S41, integrating the prediction results of the real-time dynamic electric hydrogen energy demand prediction model with the optimized multi-objective operation planning model;

[0187] S42. Use the subjective weighting method to determine the fused weights and determine the optimal site and equipment capacity planning for the electric hydrogen production and filling integrated station.

[0188] In this optional embodiment, the subjective weighting method is used to determine the fused weights, and determining the optimal site and equipment capacity planning of the electric hydrogen production and filling integrated station includes the following steps:

[0189] S421, weighted fusion of each prediction result and optimization result data through a neural network fusion model based on a subjective weighting method, and determining their respective weights;

[0190] In addition, it should be noted that the neural network fusion model based on the subjective weighting method fuses the data according to the weights, that is, it averages the results according to the weights to obtain comprehensive prediction and optimization results, but the weights of the data are determined by the subjective weighting method.

[0191] The subjective weighting method refers to a method that uses a qualitative approach, in which professionals use their personal subjective experience to weight different indicators through scoring and grading. Therefore, the determination of the weight of each result requires subjective analysis and judgment.

[0192] S422. Determine the optimal site and equipment capacity planning of the electric hydrogen production and filling integrated station based on the weight ratio.

[0193] In this optional embodiment, the formula for weighted fusion of each prediction result and optimization result data through a neural network fusion model based on a subjective weighting method is:

[0194] F=w d ×D+w o ×O;

[0195] In the formula, F represents the fusion data result, w d represents the weight of the prediction result determined by the subjective weighting method, w o It represents the weight of the optimization result determined by the subjective weighting method, D represents the prediction result data, and O represents the optimization result data.

[0196] It should be explained that the real-time dynamic electric hydrogen energy demand forecasting model is used to predict the temporal and spatial distribution of the electric hydrogen charging demand in the region. After the model is trained with historical data, the typical daily demand results are predicted. The results can be fed back into the model as input, and combined with previous historical data to participate in the training of the model, and the model parameters are corrected and updated for subsequent predictions to achieve real-time dynamic predictions. According to the real-time electric hydrogen energy demand forecast, the coordinated output of electric hydrogen energy and energy production demand in the region can be effectively regulated.

[0197] Based on the regional real-time dynamic electric hydrogen energy demand forecast results and the integrated energy operation optimization results of the integrated energy station, combined with the fusion model based on the subjective weighting method, the different forecast results and optimization result data are weighted by the subjective weighting method to determine their respective weights, and the fusion model is used to weighted fuse the data from different sources, that is, to average the weights of each result according to the weight, and obtain the comprehensive forecast and optimization results. Based on the processed fusion data results, the optimal site selection and equipment capacity planning of the integrated energy station are carried out to determine the optimal site and equipment capacity planning of the electric hydrogen production and filling integrated station. The main planning contents include the address selection of the electric hydrogen production and filling integrated station, the capacity determination of the regional integrated energy station, and the determination of the capacity and number of each equipment (electrolyzer, hydrogen storage equipment, hydrogen injection machine, etc.).

[0198] Subjective weighting method refers to a method that uses a qualitative approach, where professionals use their personal subjective experience to weight different indicators through scoring and grading. Weighted average fusion takes a weighted average of each result according to the weight to obtain the fused data result.

[0199] S5. Use smart contracts to record the optimal site and equipment capacity planning of the electric hydrogen filling station into the blockchain.

[0200] It needs to be explained that, in combination with the existing blockchain technology, the data of the integrated electric hydrogen filling station can be managed and maintained, and all kinds of data of the integrated energy station (such as production data, storage data, consumption data, etc. of electricity and hydrogen energy) can be recorded on the blockchain through smart contracts. Each data change generates a unique block and is confirmed by the consensus of the nodes in the entire network. All nodes retain a complete copy of the data ledger, and no node can tamper with the data alone. At the same time, smart contracts are used to automate the data recording and verification process. Smart contracts can set rules, and when conditions are met, the recording operation is automatically executed and published on the blockchain. The production and consumption of electricity and hydrogen energy at the energy station are recorded through the blockchain to ensure the transparency and immutability of the data, monitor and record the flow of energy, and optimize the operation and management of the energy station.

[0201] Blockchain is a distributed ledger technology that allows multiple participants to jointly record and verify transactions without relying on a central authority or intermediary. The core features of this technology include decentralization, immutability, transparency and security. In a blockchain network, data is added to a growing chain in the form of blocks. Each block contains a set of transaction records and is linked to the previous block through cryptographic methods to form an unchangeable and unforgeable chain.

[0202] Based on the existing blockchain platform and technical framework, choose the appropriate technical route, such as Ethereum, Hyperledger Fabric, Corda, EOSIO, etc.

[0203] Based on the coupling of new energy vehicles, transportation network and distribution network, taking economic factors as indicators, the annual comprehensive total cost such as annual investment and construction cost, operating cost and network loss cost related to the optimization configuration of the electric hydrogen production and filling integrated station is minimized as the objective function, and the electric hydrogen production and filling integrated station is planned. In order to simplify the analysis, three typical schemes are selected from the candidate schemes for comparison with the optimal scheme. The site selection results of the electric hydrogen production and filling integrated station in the transportation network and distribution network are shown in Tables 1 and 2.

[0204] Table 1 Annual economic cost of electric hydrogen filling station at different nodes

[0205]

[0206] Table 2 Capacity planning of each device under different nodes

[0207]

[0208] Combined with the actual market situation, hydrogen fuel cell vehicles have a small market share and relatively low demand for hydrogen filling; while electric vehicles have a high demand for charging. Therefore, when planning the electric hydrogen production and filling integrated station, for areas with low demand for hydrogen fuel cell vehicles, only charging piles are considered. Three planning schemes are set up. Scheme 1 sets up three energy stations as electric hydrogen production and filling integrated stations; Scheme 2 sets up two of the stations as electric hydrogen production and filling integrated stations, and one station is a separate charging station; Scheme 3 sets up one of the stations as an electric hydrogen production and filling integrated station, and two stations are separate charging stations. The comparison results of the annualized comprehensive economic costs under different planning schemes are shown in Table 3.

[0209] Table 3 Comparison of annualized comprehensive economic costs of different schemes

[0210]

[0211] Since the energy station can capture all electric vehicles and hydrogen fuel cell vehicles with charging needs, the total amount of electricity and hydrogen demand is certain. The operating costs of the charging stations in the three planning schemes are mainly different from the water cost and worker cost of the integrated energy station in the three schemes, but the difference is not large; the equipment investment cost and land investment cost both increase with the increase of hydrogenation equipment; in Scheme 3, the electricity demand generated by hydrogen fuel cell vehicles due to hydrogen charging demand is concentrated in one node, resulting in an increase in network loss costs; considering the annualized comprehensive economic cost, Scheme 3 should be selected. The capacity planning results of each device at different nodes in Scheme 3 are shown in Table 4.

[0212] Table 4 Capacity planning of each device under different nodes in scheme 3

[0213]

[0214] Figure 2 An embodiment of the artificial intelligence-based electric hydrogen production and filling integrated station planning system of the present invention is shown.

[0215] In this optional embodiment, the electric hydrogen production and filling integrated station planning system includes: a data acquisition module 1, a model building and prediction module 2, an energy flow and planning module 3, a planning integration module 4 and a blockchain security module 5;

[0216] Data acquisition module 1, used to acquire energy climate data of each traffic node in the electric hydrogen production and filling integrated station;

[0217] Model building and prediction module 2 is used to pre-build a neural network fusion model and input energy climate data into the neural network fusion model to obtain a spatiotemporal distribution prediction model of electric hydrogen energy demand;

[0218] Energy flow and planning module 3 is used to analyze the energy flow of the electric hydrogen filling station and build a multi-objective operation planning model coupled with the power grid and transportation network;

[0219] Planning integration module 4 is used to predict the spatiotemporal distribution of electric hydrogen charging demand using the spatiotemporal distribution prediction model, and integrate the prediction results with the output data of the multi-objective operation planning model to obtain the optimal site and equipment capacity planning of the electric hydrogen production and charging integrated station;

[0220] Blockchain security module 5 is used to record the optimal site and equipment capacity planning of the electric hydrogen filling station into the blockchain using smart contracts.

[0221] Figure 3 An embodiment of a computer device of the present invention is shown. The computer device may be a server, and its internal structure diagram may be as shown in FIG. Figure 3 As shown. The computer device includes a processor, a memory and a network interface connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system, a computer program and a database. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The database of the computer device is used to store static information and dynamic information data. The network interface of the computer device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, the steps in the above method embodiment are implemented.

[0222] Those skilled in the art will understand that Figure 3 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present invention, and does not constitute a limitation on the computer device to which the solution of the present invention is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0223] In addition, the present invention also provides a computer device, including a memory and a processor, wherein a computer program is stored in the memory, and the processor implements the steps in the above method embodiment when executing the computer program.

[0224] In addition, the present invention also provides a computer-readable storage medium on which a computer program is stored. When the computer program is executed by a processor, the steps in the above method embodiment are implemented.

[0225] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory or optical memory, etc. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM).

[0226] To sum up, with the help of the above-mentioned technical scheme of the present invention, the present invention aims at the impact of the layout planning of basic charging facilities on the charging demand of electric vehicles and hydrogen fuel cell vehicles, and proposes a planning method for an electric hydrogen production and filling integrated station combined with an artificial intelligence algorithm. The relevant historical data of electric vehicles, hydrogen fuel cell vehicles and meteorological factors in the area are collected, and the Pearson correlation coefficient is used to perform correlation analysis on the above-mentioned historical data in the area, and the meteorological factors with low correlation coefficients are eliminated, so as to build a prediction model for the spatiotemporal distribution of electric hydrogen charging demand based on a deep learning network; a dynamic electric hydrogen energy demand prediction model based on real-time demand data feedback is constructed to coordinate the production and consumption of electric hydrogen energy; secondly, the energy production, transportation and consumption structure of the electric hydrogen production and filling integrated station is analyzed, the energy flow relationship of the electric hydrogen energy coupling is constructed, and the integrated energy is analyzed. The operation planning framework of the station is established; then, a multi-objective function model of the regional electric hydrogen production and filling integrated station is established with the goal of minimizing the economic costs such as investment, operation and network loss, taking into account the constraints such as equipment constraints, transportation network constraints and power system flow constraints, and constructing a multi-objective planning model for the energy station, and using the improved particle swarm optimization algorithm to solve the model; then, combined with the regional real-time dynamic electric hydrogen energy demand forecast results and the integrated energy station operation optimization results, combined with the fusion model based on the subjective weighting method, the optimal site and equipment capacity planning results of the electric hydrogen production and filling integrated station in the region are obtained, which promotes the use of stations to drive vehicles, promotes the development of electric vehicles and hydrogen fuel cell vehicles, and improves the optimal utilization of electric hydrogen energy types. Finally, combined with the existing blockchain technology, the management and maintenance of the electric hydrogen production and filling station data is realized, and the efficient operation and management of the integrated energy station is promoted.

[0227] Although the present invention has been disclosed as above with preferred embodiments, the embodiments are merely examples for the convenience of description and are not intended to limit the present invention. Those skilled in the art may make several changes and modifications without departing from the spirit and scope of the present invention. The scope of protection claimed by the present invention shall be based on the claims.

Claims

1. The planning method of electric hydrogen filling station based on artificial intelligence is characterized by: The method for planning an integrated hydrogen production and filling station includes the following steps: S1. Obtain energy climate data of each traffic node in the electric hydrogen production and filling integrated station; S2. Pre-build a neural network fusion model, and input energy climate data into the neural network fusion model to obtain a spatiotemporal distribution prediction model for electric hydrogen energy demand; S3. Analyze the energy flow of the electric hydrogen production and filling station, and build a multi-objective operation planning model coupled with the power grid and transportation network; S4. Use the spatiotemporal distribution prediction model to predict the spatiotemporal distribution of electric hydrogen charging demand, and integrate the prediction results with the output data of the multi-objective operation planning model to obtain the optimal site and equipment capacity planning of the electric hydrogen production and charging integrated station; S5. Use smart contracts to record the optimal site and equipment capacity planning of the electric hydrogen filling station into the blockchain.

2. The method for planning an integrated hydrogen production and filling station based on artificial intelligence according to claim 1 is characterized in that: The acquisition of energy climate data of each traffic node in the electric hydrogen production and filling integrated station includes the following steps: S11. Obtain the historical annual ownership data of electric vehicles and hydrogen fuel vehicles, daily charging demand data and meteorological data at each traffic node in the electric hydrogen production and filling integrated station; S12. Calculate the Pearson correlation coefficient of the annual ownership data of historical electric vehicles and hydrogen fuel cell vehicles, daily charging demand data, and meteorological data; S13. According to the calculation result of the Pearson correlation coefficient, the meteorological factors that do not meet the preset threshold value in the Pearson correlation coefficient are eliminated.

3. The method for planning an integrated hydrogen production and filling station based on artificial intelligence according to claim 2 is characterized in that: The formula for calculating the Pearson correlation coefficient of the annual ownership data of historical electric vehicles and hydrogen fuel vehicles, daily charging demand data and meteorological data is: In the formula, t represents a certain moment; T represents the time series P t , Q t The total number of data; P t , Q t Represents the value of a time series at a certain moment t; Represent the time series P t , Q t The average value of XY It indicates the correlation between variable x and variable y.

4. The method for planning an integrated hydrogen production and filling station based on artificial intelligence according to claim 3 is characterized in that: The method of pre-building a neural network fusion model and inputting energy climate data into the neural network fusion model to obtain a spatiotemporal distribution prediction model of electric hydrogen energy demand includes the following steps: S21. Constructing the convolution layer, forget gate, input gate and output gate of the network fusion model in sequence based on the convolutional neural network model and the long short-term memory network model; S22, normalizing the annual ownership data of historical electric vehicles and hydrogen fuel vehicles, daily charging demand data, and meteorological data after correlation analysis; S23, input the normalized historical annual ownership data of electric vehicles and hydrogen fuel vehicles, daily charging demand data and meteorological data into the network fusion model, and establish a spatiotemporal distribution prediction model for electric hydrogen energy demand; S24. Input relevant historical data and predicted real-time data into the spatiotemporal distribution prediction model, and combine it with the adaptive learning mechanism to build a dynamic spatiotemporal distribution prediction model for electric hydrogen energy demand.

5. The method for planning an integrated hydrogen production and filling station based on artificial intelligence according to claim 4 is characterized in that: The inputting of relevant historical data and predicted real-time data into the spatiotemporal distribution prediction model and combining the adaptive learning mechanism to construct a dynamic spatiotemporal distribution prediction model for electric hydrogen energy demand includes the following steps: S241, inputting relevant historical data and predicted real-time data into the spatiotemporal distribution prediction model; S242, using an adaptive learning mechanism to update parameters of the spatiotemporal distribution prediction model in real time; S243. Construct a dynamic spatiotemporal distribution prediction model of electric hydrogen energy demand based on real-time updated parameters, and conduct real-time dynamic prediction of regional electric hydrogen energy demand.

6. The method for planning an integrated hydrogen production and filling station based on artificial intelligence according to claim 1 is characterized in that: The analysis of the energy flow of the electric hydrogen production and filling integrated station and the construction of a multi-objective operation planning model coupled with the power grid and the transportation network include the following steps: S31. Based on the energy flow relationship of the electric hydrogen production and filling station, design the coupling conditions at the power grid and transportation network levels, and establish a multi-objective function model with economic indicators integrated; S32. Based on the energy flow relationship and multi-objective function model of the electric hydrogen production and filling integrated station, a multi-objective operation planning model of the electric hydrogen production and filling integrated station with constraints is constructed; S33. Use the improved particle swarm optimization algorithm to optimize the nonlinear constraints of the multi-objective operation planning model.

7. The method for planning an integrated hydrogen production and filling station based on artificial intelligence according to claim 1 is characterized in that: The energy flow relationship of the electric hydrogen production and filling integrated station is used to design the coupling conditions at the power grid and transportation network levels, and establish a multi-objective function model integrating economic indicators, including the following steps: S311. Construct the operation planning framework of the electric hydrogen production and filling integrated station; S312. Analyze the relationship between the distribution network and the transportation network, and between electricity and hydrogen energy in the operation planning framework; S313. Based on the analysis results, the objectives of the power distribution network and the transportation network in the planning area are coupled, and the economic indicators of the integrated hydrogen production and filling station are minimized as the planning goal, and a multi-objective function model covering various economic indicators is constructed.

8. The method for planning an integrated hydrogen production and filling station based on artificial intelligence according to claim 7 is characterized in that: The economic indicators include investment cost, operating cost and grid-level targets.

9. The method for planning an integrated hydrogen production and filling station based on artificial intelligence according to claim 8 is characterized in that: The constraints include: equipment constraints, traffic network constraints and power system flow constraints.

10. The method for planning an integrated hydrogen production and filling station based on artificial intelligence according to claim 9, characterized in that: The method of optimizing the nonlinear constraints of the multi-objective operation planning model by using the improved particle swarm optimization algorithm comprises the following steps: S331, initializing the distribution of electricity and hydrogen loads, land price distribution and grid structure of the transportation network in the multi-objective operation planning model; S332, selecting a preset number of electric hydrogen production and filling integrated stations as the initial positions of the particle group; S333, obtaining charging load data of the electric hydrogen production and charging integrated station captured by each station according to the initial site selection, and obtaining a fitness function; S334, repeating the steps of S332-S333, updating the fitness function through iteration of the particle swarm to obtain the next generation of particles, and outputting the fitness function when the fitness function no longer changes or reaches the maximum number of iterations.

11. The method for planning an integrated hydrogen production and filling station based on artificial intelligence according to claim 1, characterized in that: The method of using the spatiotemporal distribution prediction model to predict the spatiotemporal distribution of the electric hydrogen charging demand, and fusing the prediction results with the output data of the multi-objective operation planning model to obtain the optimal site and equipment capacity planning of the electric hydrogen production and charging integrated station includes the following steps: S41, integrating the prediction results of the real-time dynamic electric hydrogen energy demand prediction model with the optimized multi-objective operation planning model; S42. Use the subjective weighting method to determine the fused weights and determine the optimal site and equipment capacity planning for the electric hydrogen production and filling integrated station.

12. The artificial intelligence-based electric hydrogen filling integrated station planning method according to claim 11 is characterized in that: The method of using the subjective weighting method to determine the fused weights and determine the optimal site and equipment capacity planning of the electric hydrogen production and filling integrated station includes the following steps: S421, weighted fusion of each prediction result and optimization result data through a neural network fusion model based on a subjective weighting method, and determining their respective weights; S422. Determine the optimal site and equipment capacity planning of the electric hydrogen production and filling integrated station based on the weight ratio.

13. The method for planning an integrated hydrogen production and filling station based on artificial intelligence according to claim 1, characterized in that: The formula for weighted fusion of each prediction result and optimization result data through a neural network fusion model based on the subjective weighting method is: F=w d ×D+w o ×O; In the formula, F represents the fusion data result, w d represents the weight of the prediction result determined by the subjective weighting method, w o It represents the weight of the optimization result determined by the subjective weighting method, D represents the prediction result data, and O represents the optimization result data.

14. An artificial intelligence-based integrated hydrogen production and filling station planning system, used to implement the artificial intelligence-based integrated hydrogen production and filling station planning method according to any one of claims 1 to 13, characterized in that: The electric hydrogen filling station planning system includes: data acquisition module, model building and prediction module, energy flow and planning module, planning integration module and blockchain security module; A data acquisition module is used to obtain energy and climate data of each traffic node in the electric hydrogen production and filling integrated station; The model building and prediction module is used to pre-build a neural network fusion model and input energy climate data into the neural network fusion model to obtain a spatiotemporal distribution prediction model for electric hydrogen energy demand; Energy flow and planning module, which is used to analyze the energy flow of the electric hydrogen filling station and build a multi-objective operation planning model coupled with the power grid and transportation network; The planning integration module is used to predict the spatiotemporal distribution of electric hydrogen charging demand using the spatiotemporal distribution prediction model, and integrate the prediction results with the output data of the multi-objective operation planning model to obtain the optimal site and equipment capacity planning of the electric hydrogen production and charging integrated station; The blockchain security module is used to record the optimal site and equipment capacity planning of the electric hydrogen filling station into the blockchain using smart contracts.

15. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 13 are implemented.

Images