Power distribution network hydrogen scheduling method, device, equipment, medium and program product

Abstract

The invention relates to a power distribution network hydrogen scheduling method and device, computer equipment, a computer readable storage medium and a computer program product. The method comprises the steps that after it is detected that a hydrogen scheduling operation starting condition is met at the current moment, a current receiver needing hydrogen in a power distribution network and a current supplier providing hydrogen are determined, wherein the current receiver and the current supplier are involved in the current hydrogen scheduling operation; obtaining multi-dimensional data of a current recipient at the current moment, wherein the multi-dimensional data at least comprises dimension data corresponding to a weather dimension, a traffic dimension and an electricity-hydrogen load dimension; based on the multi-dimensional data, the hydrogenation demand of the current recipient is determined through the trained hydrogen load prediction model; based on the hydrogenation demand, determining the hydrogen mutual aid amount between the current recipient and the current supplier, and determining a target transportation route between the current recipient and the current supplier by taking the minimum transportation time as an optimization target; and executing the current hydrogen scheduling operation based on the hydrogen mutual aid amount and the target transportation route, thereby improving the hydrogen scheduling accuracy of the power distribution network.

Description

Technical Field

[0001] This application relates to the field of hydrogen dispatching technology in power distribution networks, and in particular to a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for hydrogen dispatching in power distribution networks. Background Technology

[0002] With the development of science and technology, distributed renewable energy (DRE), represented by wind power and photovoltaics, is being integrated into the power system on a large scale. At the distribution network level, the inherent uncertainties of DRE lead to increasingly prominent issues regarding its absorption. By integrating into an Electric-Hydrogen Integrated Energy System (EHIES), the distribution network can meet the hydrogen demand of hydrogen loads and improve its capacity to absorb DRE by utilizing adjustable hydrogen-producing electricity loads. However, considering that the hydrogen supply capacity of hydrogen refueling stations located within EHIES is affected by the uncertainties of wind and solar power output, if reasonable operating strategies are not formulated based on the hydrogen demand corresponding to EHIES, it will not only be difficult to tap the potential of the distribution network to absorb DRE, but it may also increase the peak-shaving pressure on the distribution network and affect its normal operation. Therefore, predicting the hydrogen demand of hydrogen refueling stations located within EHIES is an important prerequisite for optimizing the coordinated operation of the distribution network's electricity and hydrogen systems.

[0003] In related technologies, hydrogen demand is typically predicted based on time series analysis of historical data, such as autoregressive moving average and multiple linear regression. However, this method suffers from significant prediction errors, thus failing to accurately manage hydrogen distribution among refueling stations. Summary of the Invention

[0004] Therefore, it is necessary to provide a method, apparatus, computer equipment, computer-readable storage medium, and computer program product for hydrogen dispatching in power distribution networks that can improve the accuracy of hydrogen dispatching in power distribution networks, in order to address the above-mentioned technical problems.

[0005] Firstly, this application provides a method for hydrogen dispatching in a power distribution network, including:

[0006] After detecting that the conditions for starting the hydrogen dispatch operation are met at the current moment, determine the current recipients of hydrogen in the distribution network and the current suppliers of hydrogen involved in the current hydrogen dispatch operation.

[0007] Obtain multi-dimensional data of the current recipient at the current moment, wherein the multi-dimensional data includes at least the corresponding dimension data of weather dimension, traffic dimension and electric hydrogen load dimension;

[0008] Based on the multi-dimensional data, the hydrogen loading prediction model is used to determine the current hydrogen demand of the recipient.

[0009] Based on the hydrogen refueling demand, determine the amount of hydrogen exchange between the current recipient and the current supplier, and determine the target transportation route between the current recipient and the current supplier with the goal of minimizing transportation time.

[0010] Based on the stated hydrogen resupply amount and the target transportation route, execute the current hydrogen dispatch operation.

[0011] Secondly, this application also provides a hydrogen dispatching device for a power distribution network, comprising:

[0012] The hydrogen supplier and recipient determination module is used to determine the current recipients of hydrogen in the distribution network and the current suppliers of hydrogen involved in the current hydrogen dispatch operation after detecting that the conditions for starting the hydrogen dispatch operation are met at the current time.

[0013] The multi-dimensional data acquisition module is used to acquire the multi-dimensional data of the current recipient at the current moment. The multi-dimensional data includes at least the dimension data corresponding to the weather dimension, traffic dimension and electric hydrogen load dimension.

[0014] The hydrogen demand determination module is used to determine the current hydrogen demand of the recipient based on the multi-dimensional data and through a trained hydrogen load prediction model.

[0015] The transportation route determination module is used to determine the amount of hydrogen exchange between the current recipient and the current supplier based on the hydrogen refueling demand, and to determine the target transportation route between the current recipient and the current supplier with the goal of minimizing transportation time.

[0016] The transportation operation execution module is used to execute the current hydrogen scheduling operation based on the hydrogen mutual aid quantity and the target transportation route.

[0017] Thirdly, this application also provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to perform the following steps:

[0018] After detecting that the conditions for starting the hydrogen dispatch operation are met at the current moment, determine the current recipients of hydrogen in the distribution network and the current suppliers of hydrogen involved in the current hydrogen dispatch operation.

[0019] Obtain multi-dimensional data of the current recipient at the current moment, wherein the multi-dimensional data includes at least the corresponding dimension data of weather dimension, traffic dimension and electric hydrogen load dimension;

[0020] Based on the multi-dimensional data, the hydrogen loading prediction model is used to determine the current hydrogen demand of the recipient.

[0021] Based on the hydrogen refueling demand, determine the amount of hydrogen exchange between the current recipient and the current supplier, and determine the target transportation route between the current recipient and the current supplier with the goal of minimizing transportation time.

[0022] Based on the stated hydrogen resupply amount and the target transportation route, execute the current hydrogen dispatch operation.

[0023] Fourthly, this application also provides a computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, performs the following steps:

[0024] After detecting that the conditions for starting the hydrogen dispatch operation are met at the current moment, determine the current recipients of hydrogen in the distribution network and the current suppliers of hydrogen involved in the current hydrogen dispatch operation.

[0025] Obtain multi-dimensional data of the current recipient at the current moment, wherein the multi-dimensional data includes at least the corresponding dimension data of weather dimension, traffic dimension and electric hydrogen load dimension;

[0026] Based on the multi-dimensional data, the hydrogen loading prediction model is used to determine the current hydrogen demand of the recipient.

[0027] Based on the hydrogen refueling demand, determine the amount of hydrogen exchange between the current recipient and the current supplier, and determine the target transportation route between the current recipient and the current supplier with the goal of minimizing transportation time.

[0028] Based on the stated hydrogen resupply amount and the target transportation route, execute the current hydrogen dispatch operation.

[0029] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, performs the following steps:

[0030] After detecting that the conditions for starting the hydrogen dispatch operation are met at the current moment, determine the current recipients of hydrogen in the distribution network and the current suppliers of hydrogen involved in the current hydrogen dispatch operation.

[0031] Obtain multi-dimensional data of the current recipient at the current moment, wherein the multi-dimensional data includes at least the corresponding dimension data of weather dimension, traffic dimension and electric hydrogen load dimension;

[0032] Based on the multi-dimensional data, the hydrogen loading prediction model is used to determine the current hydrogen demand of the recipient.

[0033] Based on the hydrogen refueling demand, determine the amount of hydrogen exchange between the current recipient and the current supplier, and determine the target transportation route between the current recipient and the current supplier with the goal of minimizing transportation time.

[0034] Based on the stated hydrogen resupply amount and the target transportation route, execute the current hydrogen dispatch operation.

[0035] The aforementioned hydrogen dispatching method, apparatus, computer equipment, computer-readable storage medium, and computer program product for power distribution networks, after detecting that the conditions for initiating a hydrogen dispatching operation are met at the current moment, determine the current recipients and suppliers of hydrogen in the power distribution network involved in the current hydrogen dispatching operation; acquire multi-dimensional data of the current recipients at the current moment, including at least weather, traffic, and electricity-hydrogen load dimensions; and determine the current recipients' hydrogen refueling demand based on the multi-dimensional data and a trained hydrogen load prediction model. This allows for accurate prediction of the actual current hydrogen refueling demand by combining multi-dimensional detailed information, improving the accuracy of hydrogen refueling demand prediction. Then, based on the hydrogen refueling demand, the amount of hydrogen exchange between the current recipients and suppliers is determined in real time. To further ensure dispatching accuracy, the target transportation route between the current recipients and suppliers is optimized with the goal of minimizing transportation time. Finally, based on highly accurate hydrogen exchange amounts and target transportation routes, a more precise current hydrogen dispatching operation can be executed, thereby greatly improving the accuracy of hydrogen dispatching in the power distribution network. Attached Figure Description

[0036] To more clearly illustrate the technical solutions in the embodiments of this application or related technologies, the drawings used in the description of the embodiments of this application or related technologies will be briefly introduced below. Obviously, the drawings described below are only some embodiments of this application. For those skilled in the art, other related drawings can be obtained based on these drawings without creative effort.

[0037] Figure 1 This is an application environment diagram of the hydrogen dispatching method for a power distribution network in one embodiment;

[0038] Figure 2 This is a flowchart illustrating a hydrogen dispatching method for a power distribution network in one embodiment;

[0039] Figure 3 This is a schematic diagram of an optimization model for the coordinated operation of electricity and hydrogen in a power distribution network, as shown in one embodiment.

[0040] Figure 4 This is a schematic diagram of a rolling optimization framework for a dual-hydrogen layer in a power distribution network, as shown in one embodiment.

[0041] Figure 5This is a schematic diagram illustrating the specific steps involved in determining the transportation route in one embodiment;

[0042] Figure 6 This is a schematic diagram of the technical route of the hydrogen dispatch strategy for the distribution network in one embodiment;

[0043] Figure 7 This is a structural block diagram of a hydrogen dispatching device for a power distribution network in one embodiment;

[0044] Figure 8 This is an internal structural diagram of a computer device in one embodiment. Detailed Implementation

[0045] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0046] It should be noted that the terms "first," "second," etc., used in this application can be used to describe various elements, but these elements are not limited by these terms. These terms are only used to distinguish the first element from the second element. The terms "comprising" and "having," and any variations thereof, used in this application, are intended to cover non-exclusive inclusion. The term "multiple" used in this application refers to two or more. The term "and / or" used in this application refers to one of the embodiments, or any combination of multiple embodiments.

[0047] The hydrogen dispatching method for power distribution networks provided in this application can be applied to, for example... Figure 1 In the application environment shown, terminal 102 communicates with server 104 via a network. A data storage system can store the data that server 104 needs to process. The data storage system can be integrated onto server 104, or it can be located in the cloud or on another network server.

[0048] In one embodiment, each distribution network has a corresponding server 104. After the server 104 detects that the conditions for starting a hydrogen dispatch operation are met at the current moment, it determines the current recipient of hydrogen and the current supplier of hydrogen in the distribution network involved in the current hydrogen dispatch operation. The server 104 acquires multi-dimensional data of the current recipient collected by the terminal 102 deployed in the distribution network at the current moment. The multi-dimensional data includes at least the corresponding dimension data of weather, traffic, and electricity-hydrogen load. Based on the multi-dimensional data, the server 104 determines the hydrogen refueling demand of the current recipient through a trained hydrogen load prediction model. Based on the hydrogen refueling demand, the server 104 determines the amount of hydrogen exchange between the current recipient and the current supplier. With minimizing transportation time as the optimization objective, the server 104 determines the target transportation route between the current recipient and the current supplier. Based on the amount of hydrogen exchange and the target transportation route, the server 104 executes the current hydrogen dispatch operation.

[0049] The terminal 102 is used to collect relevant data from the current recipient in the distribution network, such as data corresponding to weather, traffic, and electricity / hydrogen load dimensions. The terminal 102 can be, but is not limited to, various personal computers, laptops, smartphones, tablets, etc. The server 104 can be a standalone physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing cloud computing services.

[0050] In one exemplary embodiment, such as Figure 2 As shown, a hydrogen dispatching method for a power distribution network is provided, which can be applied to... Figure 1 Taking server 104 as an example, the explanation includes the following steps 202 to 210. Wherein:

[0051] Step 202: After detecting that the conditions for starting the hydrogen dispatch operation are met at the current time, determine the current recipients of hydrogen in the distribution network and the current suppliers of hydrogen involved in the current hydrogen dispatch operation.

[0052] The hydrogen dispatch operation initiation conditions refer to the conditions under which a hydrogen dispatch operation is initiated. The current hydrogen dispatch operation refers to the hydrogen dispatch operation within the distribution network at the current moment, specifically the operation of dispatching (transporting) hydrogen from one hydrogen refueling station to another. The current recipient refers to the hydrogen refueling station that needs hydrogen in the current hydrogen dispatch operation, and the current supplier refers to the hydrogen refueling station that provides hydrogen in the current hydrogen dispatch operation. It should be noted that each hydrogen refueling station in the distribution network provides hydrogen to hydrogen fuel cell vehicles (HFVs) to ensure their normal operation. Each hydrogen refueling station deploys a corresponding EHIES (Electronic Hydrogen Energy Systems). Hydrogen dispatch is essentially the dispatch of EHIES among these stations. Therefore, the current supplier can also be understood as the EHIES deployed by the hydrogen refueling station that provides hydrogen in the current hydrogen dispatch operation, and the current recipient refers to the EHIES deployed by the hydrogen refueling station that needs hydrogen in the current hydrogen dispatch operation.

[0053] For example, the hydrogen scheduling operation can be initiated when the time between the previous hydrogen scheduling operation and the current time meets a preset time threshold, such as starting the hydrogen scheduling operation every day or every month to initiate hydrogen scheduling on a rolling basis. Alternatively, upon receiving a hydrogen scheduling request from a client at the current time, the system can automatically determine that the current time meets the hydrogen scheduling operation initiation conditions in response to the request.

[0054] Step 204: Obtain multi-dimensional data of the current recipient at the current moment. The multi-dimensional data includes at least the corresponding dimension data of weather, traffic and electricity-hydrogen load.

[0055] Multidimensional data reflects data from multiple dimensions, each with a different data source. Therefore, multidimensional data can be understood as multi-source data. For example, weather data could be the current weather information of the current recipient, traffic data could be the current traffic flow of the current recipient, and electricity / hydrogen load data could be the current electricity / hydrogen load data of the current recipient, such as price.

[0056] Step 206: Based on multi-dimensional data, determine the current hydrogen refueling demand of the recipient using a trained hydrogen load prediction model.

[0057] Among them, the trained hydrogen load prediction model is used to predict hydrogen demand, which indicates the amount of hydrogen that the current recipient needs to obtain from the outside to ensure that the current recipient can provide enough hydrogen to the HFV located at the corresponding hydrogen refueling station.

[0058] It should be noted that, to achieve safe, economical, and sustainable operation, the power distribution network must ensure a precise spatiotemporal match between energy supply and load demand within its coverage area. Therefore, each EHIES should avoid an imbalance between hydrogen production and consumption. Consumption refers to the energy usage behavior of HFVs refueling at EHIES. The production plan of EHIES needs to be adjusted in real time based on the hydrogen demand forecasts of HFVs to maximize the utilization of wind and solar resources. However, the stochastic consumption behavior of HFVs poses a challenge to accurately predicting the energy demand of new energy vehicles. Therefore, hydrogen load forecasting models can accurately predict the corresponding hydrogen refueling demand.

[0059] In one embodiment, the trained hydrogen load prediction model includes a trained data filtering model and a trained hydrogen demand prediction model. The data filtering model is used to select data of high importance. It should be noted that, considering the redundancy of data, which may slow down the model building process and hinder further improvement in prediction accuracy, a data filtering model can be used to select data strongly correlated with the dynamic characteristics of hydrogen load for subsequent prediction. The hydrogen demand prediction model is a neural network model used to predict actual hydrogen demand.

[0060] In one embodiment, the current hydrogen refueling demand of the recipient is determined based on multi-dimensional data and a trained hydrogen load prediction model, including: preprocessing the multi-dimensional data to obtain processed data; using the processed data and a trained data filtering model to filter out data whose importance is higher than a certain threshold; and using the filtered data and a trained hydrogen refueling demand prediction model to determine the current hydrogen refueling demand of the recipient.

[0061] Data preprocessing includes at least the Z-score method. In addition, data preprocessing may also include filling, smoothing, and outlier removal. Thus, data preprocessing ensures the quality and consistency of the data.

[0062] For example, the data filtering model can be XGBoost (eXtreme Gradient Boosting), an efficient machine learning algorithm and an optimized implementation of the gradient boosting framework. For example, processed data is input into a trained XGBoost to identify data in at least one dimension whose importance exceeds a certain threshold, thus obtaining the filtered data.

[0063] XGBoost is an enhancement and extension of Gradient Boosting. It integrates multiple weak classifiers into a strong classifier, thereby measuring the importance of each feature in the original feature variable data (processed data) (i.e., the importance of each dimension) and ranking them by importance. The specific steps are as follows:

[0064] First, ensemble prediction: by summing the predictions from all decision trees, the results of multiple weak classifiers (i.e., individual decision trees) are integrated to form the final prediction of the strong classifier.

[0065]

[0066] Where, x i It is an eigenvector; y i It is the eigenvector x i The actual output value; It is the output value of the decision tree model, i.e., x i The final predicted value. K is the number of decision trees, f k (·) is the mapping function of the k-th decision tree.

[0067] Then, the objective function is defined. The objective function of a decision tree can be expressed as:

[0068]

[0069] In the formula, It is a loss function used to measure y i and The difference is that the error of the model prediction in the regression task is quantified using the squared loss function; This is the regularization penalty term; n is the number of training samples; γ and λ represent hyperparameters; w j γ is the weight of leaf node j; γT is the penalty for the number of leaf nodes T, to avoid making the model too complex; w represents the weight coefficient for leaf node j. j L2 regularization is applied to limit the range of weight values ​​and prevent overfitting.

[0070] Then, the model is trained, and the prediction value of the decision tree model in the t-th iteration is the prediction value f of the current tree. t (x i The predicted values ​​of the first t-1 trees and the first t-1 trees. The sum of these, therefore equation (2) can be reformulated as:

[0071]

[0072] Because in equation (5), the decision tree f in the current iteration t (x iChanges in y will lead to changes in the loss function L(y) i ,x i The change in y(y) and the discrete structure of the decision tree itself make it very difficult to directly solve equation (5). Therefore, the second-order Taylor series approximation of L(y) is used. i ,x i After removing irrelevant constant terms, the new objective function is as follows:

[0073]

[0074] I j ={i|q(x i )=j} (9)

[0075] In the formula, G j and H j These are the sums of the first and second derivatives of all loss functions at leaf node j; I j It is the sample set of the j-th leaf node; q(x i ) = j represents sample x i The leaf node it belongs to is j.

[0076] Next, the optimal leaf node weight is calculated, i.e.: F ob2 It's about w j The optimal solution and optimal value of the quadratic function are as follows:

[0077]

[0078] Then, the node splitting gain is calculated, which involves calculating all possible splitting points according to the greedy criterion and selecting the feature with the maximum gain value for splitting, as follows:

[0079]

[0080] In the formula, G L and H L Let G represent the sum of the first and second derivatives of the loss function for all samples at the left sub-leaf node after the split, respectively; R and H R Let represent the sum of the first and second derivatives of the loss function for all samples at the right sub-leaf node after splitting, respectively; and These represent the objective function values ​​of the left and right cotyledons after splitting; The objective function value F represents the value of the unsplit objective function. ob3 .

[0081] Next, an importance assessment is performed. A negative Gain value indicates that the optimal decision tree has been constructed, the optimal split has been established, and further splitting has ceased. Therefore, the model contribution Gain for feature A is...A It can be represented as:

[0082]

[0083] In the formula, It is the square of the gain value of the t-th split node in the k-th decision tree.

[0084] Finally, sorting is performed, that is, sorting each feature variable according to its importance. This determines the importance of each dimension. Dimensions with importance above the threshold are selected as target dimensions to obtain data of the target dimensions from the processed data, which is the filtered data mentioned earlier.

[0085] The above process is the data filtering process. The following describes the hydrogen demand forecasting process. For example, the hydrogen demand forecasting model can be a BiLSTM. To this end, the filtered data is input into the BiLSTM, and the current hydrogen demand of the recipient is output by capturing historical and future context information in both directions.

[0086] BiLSTM is derived from the basic LSTM, which uses the forget gate f t Input gate i t and output gate o t The LSTM mechanism selectively memorizes or forgets information through gating, effectively modeling long-term dependencies in time series. The relevant formulas for the LSTM gating mechanism are as follows:

[0087] f t =σ(W f ·[h t-1 ,x t ]+b f (14)

[0088] i t =σ(W i ·[h t-1 ,x t ]+b i (15)

[0089]

[0090] o t =σ(W o ·[h t-1 ,x t ]+b o (18)

[0091]

[0092] In the formula, x t and h tRepresent the input state and hidden state during time interval t, respectively; C t and Represent the candidate cell state and the current cell state, respectively; σ(·) and tanh(·) are the activation functions, respectively; W f W i W o W C and b f b i b o b C These are the weight matrix and bias matrix for the corresponding gates, respectively.

[0093] While LSTM can effectively model long-term dependencies in time series, its unidirectional propagation characteristic limits the model's global perception of contextual information. To address this, BiLSTM introduces a bidirectional temporal processing mechanism on top of LSTM. It simultaneously extracts historical and future contextual features through forward and backward LSTMs and fuses bidirectional hidden states to enhance temporal representation capabilities. The relevant computational process is as follows:

[0094]

[0095] In the formula, where and These are the output results of the forward LSTM and backward LSTM at time t, respectively; LSTM(·) represents the propagation process of formulas (14)-(19); concat(·) represents the concatenation operation of the forward LSTM and backward LSTM.

[0096] It should be noted that the above-mentioned hydrogen demand forecasting process is actually based on a data-driven hydrogen load forecasting model.

[0097] In this embodiment, by preprocessing multi-dimensional data to ensure the quality of the data used for prediction, and then by using a data filtering model, high-importance data can be used to predict hydrogen demand, ensuring the accuracy of the prediction and thus helping to improve the accuracy of hydrogen dispatch.

[0098] Step 208: Based on the hydrogen refueling demand, determine the amount of hydrogen exchange between the current recipient and the current supplier, and determine the target transportation route between the current recipient and the current supplier with the goal of minimizing transportation time.

[0099] The hydrogen reciprocity amount refers to the amount of hydrogen provided by the current supplier to the current recipient. It should be noted that if the number of current suppliers is 1, the corresponding hydrogen reciprocity amount is the amount of hydrogen involved in the hydrogen refueling demand. If the number of current suppliers is greater than 1, the sum of the hydrogen reciprocity amounts corresponding to each current supplier is the amount of hydrogen involved in the hydrogen refueling demand.

[0100] Optionally, the server determines the amount of hydrogen exchange between the current recipient and each current supplier based on the hydrogen refueling demand. For each current supplier, a target transportation route between the current recipient and that current supplier is determined with the goal of minimizing transportation time.

[0101] Before introducing the steps for determining the hydrogen mutual aid quantity and the target transportation route, it is necessary to introduce the distribution network and the Energy Hub (EH) model obtained by modeling EHIES in the distribution network: The distribution network has distributed photovoltaic (PV) and wind turbine (WT) generators connected to it, which can collect wind and solar resources within the coverage area of ​​the distribution network to produce green electricity. The distribution network is also equipped with energy storage power stations, which use the charging and discharging of energy storage to compensate for the randomness and volatility of wind and solar power output. This can, to a certain extent, mitigate the source-load power deviation caused by the dual uncertainties of renewable energy generation and electricity demand, thereby improving the system's operational stability. The distribution network is connected to distributed EHIES that integrate PV, WT, electrolyzers (ELZs), compressors (CP), electricity energy storage (EES), and hydrogen storage tanks (HST) to break down the barriers between electricity and hydrogen energy and improve operational flexibility. EHIES utilizes green electricity generated by PV and WT, along with grid electricity, to drive ELZs electrolyzers to produce hydrogen to meet the hydrogen refueling needs of local HFVs. ESS and HST together serve as energy buffers within EHIES, mitigating the uncertainty of wind and solar power output and enhancing the operational flexibility of EHIES.

[0102] Each EHIES is modeled as an EH to describe its internal electro-hydrogen flow coupling, mutual influence, and hydrogen exchange with other EHIES. At the input port of the EH, PV and WT collect local DRE endowments to produce green electricity. ELZs (electrolyzers) consume green electricity or grid electricity, using alkaline water / polymer electrolyte membranes to electrolyze water to produce hydrogen. EHIES can directly inject the produced hydrogen into the HFV using hydrogen dispensers, or store excess hydrogen in a high-pressure HST (hydrogen storage tank) of 35-70 MPa for release when hydrogen production is insufficient to meet the HFV's hydrogen demand. Due to investment costs, the HST capacity configured within each EHIES is limited, and its own hydrogen production capacity / reserves are insufficient to fully cover the local HFV's hydrogen demand. In this scenario, the distribution network fills the local hydrogen supply gap by dispatching excess hydrogen from other EHIES to hydrogen-deficient EHIES, ensuring hydrogen supply throughout the region. Furthermore, the fluctuations in wind and solar power output have numerous negative impacts on the operation of ELZs: fluctuating electrical power input to ELZs leads to intermittent hydrogen production and additional purification costs, forces ELZs to frequently deviate from their rated operating conditions, reduces hydrogen production efficiency, and causes equipment components to age, resulting in a shortened equipment lifespan. Therefore, ELZs have a limited tolerance range for input power fluctuations; for example, the power fluctuation tolerance range for polymer electrolyzers is 0%–160%, while that for alkaline electrolyzers is only 0%–110%. To address this technical limitation, EHIES are equipped with ESS (Electrolysis Energy Storage) to suppress electrolysis power fluctuations. It is evident that coordinating the internal hydrogen energy management and external hydrogen supply within EHIES through the distribution network can improve the reliability and operational flexibility of EHIES hydrogen supply, thereby promoting the absorption of distributed generation equipment (DRE) by the distribution network.

[0103] In order to accurately describe and quantify the production, conversion, storage and interaction with external hydrogen energy during the operation of EHIES, an EH model considering the effect of hydrogen transportation delay is established, as shown in equations (23)-(27):

[0104] O i,t =M i,t ·V i,t (twenty three)

[0105]

[0106] In the formula, O i,t Let be the energy output matrix of the EH model for the i-th EHIES, where Let be the on-grid power of the i-th EHIES during time period t. Let t be the delay time for transporting hydrogen from the i-th EHIES to the j-th EHIES during time interval t. Let Ω be the mass of hydrogen transported from the i-th EHIES to the j-th EHIES during time period t. EHIESFor the set of EHIES connected within the distribution network, The basic time required for a trailer to travel through a cell at free-flow speed; h represents the maximum flow capacity of cell c during time period t. c This represents the actual traffic flow in cell c. Describe the impact of traffic congestion on travel time; To be related to cell c and vehicle direction f w The relevant correction factor.

[0107] M i,t Let be the coupling matrix of the EH model energy of the i-th EHIES, which contains the connection topology information and steady-state energy conversion efficiency of various electro-hydrogen devices within the EHIES, such as β. i PV β i WT and β i DN These represent the PV operating status, WT operating status, and connection status with the distribution network within the i-th EHIES, respectively. and Let Q be the charge / discharge efficiency coefficient of the EES and the charge / discharge efficiency coefficient of the HST within the i-th EHIES, respectively. hg V represents the calorific value of hydrogen. i,t Let be the energy input matrix of the EH model for the i-th EHIES, where and Let be the curtailment rates of solar and wind power for the i-th EHIES, respectively. Let be the actual PV output power of the i-th EHIES in time period t. Let WT be the actual output power of the i-th EHIES in time period t. and Let be the mass flow rate of hydrogen produced by the i-th EHIES during time period t, and the mass flow rate of the ESS during charging / discharging. and These represent the charging / discharging power of EES, the electrolysis power of ELZs, the power consumed by CP, and the electricity purchased from the grid within the i-th EHIES during time period t. To ensure that the i-th EHIES in time period t meets the hydrogenation requirements of HFV, the following can be obtained from equations (23)-(27). and This serves as the basis for the coordinated scheduling of electricity and hydrogen in the power distribution network.

[0108] The steady-state operation model of the electro-hydrogen production, storage, and conversion equipment within EHIES is shown below:

[0109]

[0110]

[0111] In the formula, and These represent the hydrogen production efficiency, Faraday efficiency, and voltage efficiency during time period t, respectively; P c,i,t Let u be the electrolytic power of a single ELZ in the ELZs of the i-th EHIES during time period t; c The standard electrolysis voltage for ELZ; A ELZ The cross-sectional area of ​​the electrode; and All are temperature characteristic parameters of the electrolytic cell; T ELZ This refers to the actual operating temperature of ELZs; These are the standard temperature variation parameters for ELZ; This is the standard reference temperature for ELZ; N c,i Let be the number of ELZs in the i-th EHIES; This represents the maximum electrolysis power of ELZs. and These represent the energy state of HST, the lower limit of HST energy state, and the upper limit of HST energy state in the i-th EHIES during time period t, respectively. and These are the rated mass and rated volume of hydrogen that the HST in the i-th EHIES can store, respectively; and These are the inflation efficiency coefficient and deflation efficiency coefficient of HST in the i-th EHIES, respectively; and These represent the inflation and deflation masses of HST in the i-th EHIES during time period t, respectively; Δt up The time interval optimized for the rolling time domain; and Let be the operating gas pressure of HST in the i-th EHIES during time period t, and let be the lower and upper limits of the operating gas pressure, respectively; R is the ideal gas constant; T HST The rated operating temperature of HST; M hg is the molar mass of hydrogen gas; and C represents the upper limit of the HST charge / discharge mass flow rate in the i-th EHIES during time period t. hg is the specific heat capacity constant of hydrogen; The temperature of the hydrogen gas injected into the compressor; K hg is the isentropic index of hydrogen; Let be the efficiency coefficient of CP in the i-th EHIES; and These are the inlet and outlet pressures of CP in the i-th EHIES during time period t; and Let be the lower and upper limits of the CP inlet and outlet pressures in the i-th EHIES during time period t; This represents the upper limit of power consumption for the CP in the i-th EHIES during time period t. and Let $\frac{1}{2}$ be the state of charge of the $EES$ in the $i$-th $EHIES$ during time period $t$, and $\frac{1}{2}$ be the lower and upper limits of the state of charge of the $EES$. Let be the rated power capacity of the EES in the i-th EHIES; and This represents the upper limit of the charging / discharging power of the EES within the i-th EHIES during a scheduling period. This represents the upper limit of the electricity that the i-th EHIES can purchase from the distribution network. Let be the upper limit of PV output in the i-th EHIES during time period t. Let be the upper limit of WT's output in the i-th EHIES during time period t.

[0112] Equations (28)-(33) represent the operating constraints of ELZs in EHIES. It can be seen from equations (31) and (32) that by adjusting the electrolytic power input to ELZs... Can be changed This allows for flexible control of hydrogen production in ELZs; Equation (33) indicates that the electrolysis power of ELZs should be less than the rated electrolysis power to avoid equipment overload. Equations (34)-(39) are the operating constraints of HST in EHIES. Equations (36) and (37) indicate that when HST is running, its and They should be kept separately by and To prevent equipment damage, the safe operating boundary is formed. Similarly, equations (38) and (39) show that the charge and discharge mass of HST during time period t should be less than their respective upper limits. Equation (40) indicates that the mass of HST charged and discharged during time period t is... The electrical energy consumed in compressing hydrogen into the HST. Equation (41) indicates that the outlet pressure of the CP should be limited to a safe operating range. Equation (42) indicates that the power consumption of the CP should be less than the rated power to avoid overload operation. Similar to HST, the state of charge of the EES... It should be limited to To avoid excessively high SOC damaging electrode materials or excessively low SOC causing an imbalance in the internal chemical reaction of the battery, thus extending the lifespan of the EES. Equations (45) and (46) indicate that the charging / discharging power of the EES during time period t should not exceed the rated charging / discharging power to avoid overcharging / over-discharging. Equation (47) indicates that the power transmitted from the grid to the EHIES should not exceed the rated power of the tie line to avoid line overload. Equations (48) and (49) indicate that the wind and solar power output in the i-th EHIES during time period t should not exceed its maximum output limit.

[0113] Based on equations (23)-(49), the coupling and interaction mechanism of electro-hydrogen energy within EHIES is revealed, and the energy production, conversion, storage and use processes within each EHIES are described and quantified, serving as boundary conditions for determining the optimal hydrogen mutual aid quantity among EHIES.

[0114] Based on this, after determining the EH model and corresponding constraints and parameters for each current supplier based on the above equations (23)-(49), in one embodiment, determining the amount of hydrogen mutual assistance between the current recipient and the current supplier based on the hydrogen demand includes: constructing an objective function with the goal of maximizing the renewable energy absorption rate, and establishing current constraints and voltage constraints related to the distribution network; based on the objective function, solving for the amount of hydrogen mutual assistance provided by each current supplier to the current recipient through the hydrogen demand, objective function, current constraints, and voltage constraints.

[0115] like Figure 3 The diagram shown is a schematic of an optimization model for the coordinated operation of electricity and hydrogen in a power distribution network in one embodiment. Figure 3 The Distributed Network Operator (DNO), as the manager of electrical equipment within the entire distribution network, is responsible for scheduling all adjustable resources within the network, including the hydrogen electro-hydrogen equipment within the EHIES (Electronic Hydrogen Equipment). The real-time decision-making process of the entire distribution network during operation is modeled as a rolling optimization problem to handle the dual uncertainties of DRE output and HFV hydrogen demand within the power supply area. Specifically:

[0116] For example, after determining the hydrogen demand, the coupling matrix of the EH model energy corresponding to the current recipient is obtained, the corresponding first matrix parameters are obtained from the coupling matrix, the energy input matrix of the EH model corresponding to the current recipient is obtained, the corresponding second matrix parameters are obtained from the energy input matrix, and the first constant related to PV and the second constant related to WT in the distribution network are obtained. Based on the first matrix parameters, the second matrix parameters, the first constant and the second constant, the optimization objective is constructed according to the following formula (50):

[0117]

[0118] In the formula, Ω DRE The DRE set in the distribution network includes Ω PV (PV set), Ω WT (WT set), Ω EHIES (EHIES set); t0 is the initial time point of the rolling time domain; T up This represents the length of the scrolling time domain range. For example, the first matrix parameter is as mentioned above. and The parameters of the second matrix are those mentioned earlier. t represents the current time. The first constant includes... (The running status of the p-th PV) (At time t, the abandonment rate of the p-th PV); the second constant includes, (The running status of the p-th WT) (Wind curtailment rate of the p-th PV at time t); Δt up This represents the length of the rolling time difference.

[0119] The operating strategy formulated by the Distribution Network Management Office (DNO) should meet system power flow constraints, hydrogen-electricity power balance constraints, and equipment operation constraints to ensure the feasibility of the strategy. To ensure accuracy while reducing computational complexity, the classic Dist-Flow power flow model is used to calculate the power flow of the distribution network. The branch currents and node voltages of the distribution network should meet the following safety constraints:

[0120]

[0121] In the formula, Ω T For time sets; N DN For the set of distribution network nodes; E DN For distribution network branch collection; l nm,t v is the square of the current amplitude nm in the distribution network branch during time period t; m,t The square of the voltage amplitude at node m in the distribution network during time period t; I nm and The l of the branch nm respectively nm,t Lower limit and upper limit; U i and v i,t The lower and upper limits. Formulas (51) and (52) above are the current constraint and voltage constraint, respectively. The distributed resources directly connected to the nodes in the distribution network are mainly PV, WT and EES. These devices should all meet their own operating constraints. Considering that these resources are also installed in EHIES, the relevant constraints will not be repeated. Please refer to the EH model section of EHIES below.

[0122] Furthermore, the distribution network must ensure the balance of electricity and hydrogen energy across the entire power supply area during each time period. Electricity power balance is already reflected in the Dist-Flow power flow model. The dynamic balance of hydrogen production, storage, and consumption within a single EHIES has been rigorously modeled in its respective EH model; it is only necessary to ensure the amount of hydrogen supplied by the EHIES during each time period. The amount of hydrogen accepted by the recipient EHIES A sum of zero is sufficient to ensure hydrogen energy balance across the entire power distribution network coverage area, as follows:

[0123]

[0124] It should be noted that equation (54) can also be used as a constraint condition for hydrogen energy balance to solve for the amount of hydrogen mutual assistance. For this purpose, for example, the obtained hydrogen demand is filled into equation (54) to replace Based on the constraints of formulas (51)-(53) and the constraints of (51) obtained after substitution, formula (50) can be solved by the solver, and the amount of hydrogen exchange between each current supplier and the current recipient can be accurately calculated.

[0125] In this embodiment, with the goal of maximizing the renewable energy absorption rate, an objective function is constructed. Based on the objective function, hydrogen demand, current constraints, and voltage constraints, the amount of hydrogen supplied by the current supplier to the current recipient can be solved automatically.

[0126] In one embodiment, determining the target transportation route between the current recipient and the current supplier with the goal of minimizing transportation time includes: dividing each segment of the traffic network between the current recipient and the current supplier into multiple discrete units; constructing traffic flow density constraints and traffic flow speed constraints corresponding to each discrete unit with the goal of minimizing transportation time; and solving for the target transportation route between the current recipient and the current supplier using a sparrow search algorithm based on the optimization goal and the traffic flow density constraints and traffic flow speed constraints.

[0127] It should be noted that before determining the target transportation route, the server needs to first obtain the transportation network of the environment shared by the current recipient and the current supplier, and then perform spatiotemporal division of the transportation network, as follows:

[0128] DNO discretizes the hydrogen transport journey time T into small time intervals, as follows:

[0129] T={0,Δt,2Δt,...,MΔt} (55)

[0130] In the formula, Δt is a short time interval during which traffic conditions do not change significantly, and M is a sufficiently large integer.

[0131] Therefore, the server can divide each segment of the traffic network into multiple discrete units of directional segments within each Δt, thus obtaining multiple discrete units. The traffic state S of each discrete unit r in time period t is... r (t) can be represented as follows:

[0132] S r (t)=(ρ r (t),q r (t),v r (t)) (56)

[0133] In the formula, ρ r (t), q r (t) and v r (t) represents the traffic flow density, flow rate, and speed in discrete unit r during time period t, respectively.

[0134] After spatiotemporal partitioning, the transportation network between the current recipient and the current supplier is abstracted into a topological structure composed of nodes (different from the nodes in formulas (1)-(23) above, where nodes refer to network nodes, but in this embodiment, they refer to actual transportation points in the transportation network) and discrete units, using a directed graph G. T It is expressed as follows:

[0135] G T =(V T A T (57)

[0136] In the formula, V T and A T V represents the set of nodes and the set of discrete units that make up all roads, respectively. T It includes all starting points, discrete unit nodes, and ending points.

[0137] Then, for each discrete unit r, ρ r (t), q r (t) and v r (t) is used for updating.

[0138] Based on the ρ of discrete unit r during time interval t r (t) and q r (t) The change in t determines the v of this unit in time period t+1. r (t+1), specifically as follows:

[0139]

[0140] q r (t)=ρ r (t)v r (t) (59)

[0141]

[0142] In the formula, Let V(ρ) represent the traffic flow into and out of discrete unit r during time period t. r (t) represents the expected speed of traffic flow during time period t, indicating the expectation of the traffic flow equilibrium state during the journey; v f ρ represents the free-flow velocity. cr The critical density of road segment r is represented by α; α represents V(ρ) r (t) and traffic density ρ r The nonlinear parameters representing the monotonically decreasing relationship between (t) are: υ, τ, and κ, which are model parameters representing the expected constant, driver adjustment time constant, and compensation coefficient, respectively.

[0143] Then, considering traffic flow conservation and METANET model constraints, a dynamic traffic path optimization model is established with the objective of minimizing hydrogen transport time between EHIES, as shown below:

[0144]

[0145] 0≤ρ r (t)≤ρ jam (63)

[0146] v min ≤v r (t)≤v free (64)

[0147] q r (t)=ρ r (t)v r (t)≤Q max (65)

[0148]

[0149] In the formula, f ht Let Ω be the objective function of the route planning model, T be the set of hydrogen transportation periods, and Ω be the objective function. HT For trailer collection, x hr (t) is a Boolean variable representing whether trailer h selects discrete unit r during time period t. If x hr (t) = 0 indicates that h does not choose r in time interval t; otherwise, it indicates that r is chosen. jam v represents the upper limit of traffic flow density for discrete unit r; min and v free Q represents the lower limit / upper limit of traffic flow speed in discrete unit r during time period t, respectively; max This represents the upper limit of traffic flow in discrete unit r during time period t; and These represent the traffic flow from discrete unit r to node n during time period t; L out (n) and L in (n) represents the set of discrete cells for each outflow / inflow node n. ΔL is the length of the discrete cell r.

[0150] Equations (63)-(65) indicate that the density, flow rate, and speed of each discrete unit at any time period must meet the physical boundary constraints to ensure that the traffic flow parameters do not exceed the road capacity limit. That is, equations (63) and (64) respectively represent the traffic flow density constraint and the traffic flow speed constraint, and equation (66) is the node flow conservation constraint, which ensures the spatiotemporal continuity of traffic flow distribution through the balance of inflow and outflow.

[0151] Initially, all trailers are assumed to be located at the starting point of the EHIES. The initial density and flow of vehicles on other road segments are based on historical traffic data from the same period. It is worth noting that every time interval Δt, the DNO (Dynamic Traffic Management Unit) re-solves the lower-level planning model to determine the optimal hydrogen transport route, thereby minimizing the arrival time of the transport vehicles. After solving the dynamic traffic path optimization model, the optimal routes and arrival times for all trailers are determined. The trailers transport hydrogen to the corresponding EHIES, completing one hydrogen resupply cycle, i.e., the hydrogen dispatch operation.

[0152] After introducing the optimization objective and constraints involved in this embodiment above, based on this, by way of example, after determining the traffic flow density constraints and traffic flow speed constraints represented by equations (63) and (64) respectively based on the above process, constraints of (65) and (66) can also be added. Based on all constraints (equations (63)-(66)) and optimization objective (equation (62)), the target transportation route between the current recipient and the current supplier is determined by the sparrow search algorithm. The target transportation route satisfies the optimization objective of minimizing transportation time.

[0153] Among them, the Sparrow Search Algorithm (SSA) is a novel swarm intelligence optimization algorithm that simulates the foraging, vigilance, and early warning behaviors of sparrows. It demonstrates strong global search capabilities and adaptability to dynamic changes in path planning problems. The solution process is as follows:

[0154] Step 1: Population initialization: Based on EHIES spatial location and traffic network topology, several initial paths are randomly generated as sparrow population individuals. The path encoding covers the discrete unit sequence traversed by each trailer.

[0155] Step 2 Fitness Calculation: Based on the state variables such as traffic flow density and speed of each road segment in the METANET traffic model at the current time period, calculate the total transportation time required for the path as the individual fitness.

[0156] Step 3: Location Update: Referring to the foraging and early warning behavior models in SSA, local perturbation and global search updates are performed on the individual location (i.e., path) to enhance the diversity and convergence performance of the population.

[0157] Step 4 Boundary verification and constraint processing: For individuals that do not meet the constraints (63)-(66), they are corrected or eliminated by the penalty function;

[0158] Step 5 Rolling Update Mechanism: To adapt to the real-time evolution of traffic flow, SSA is re-called and the path planning results are updated in each rolling time domain to achieve dynamic adaptation with traffic conditions;

[0159] Step 6: Output the optimal transport route (target transport route) and corresponding transport time for the trailer. The optimal transport route satisfies equation (62), meaning that the corresponding transport time is minimized. The transport time corresponding to the target transport route can be understood as the maximum transport time required when transporting along that target route. Therefore, the transport time corresponding to the target transport route can be used as a transport time threshold.

[0160] It should be noted that the steps for determining the hydrogen mutual aid quantity and the target transportation route mentioned above are essentially implemented by the knowledge-driven distribution network dual-electric hydrogen layer rolling optimization framework, referring to... Figure 4 As shown, Figure 4 The diagram shows a schematic of a rolling optimization framework for a dual-layer hydrogen distribution network in one embodiment. In the upper-layer operation optimization framework, equations (50)-(54) are used to determine the hydrogen mutual aid quantity. Then, the hydrogen mutual aid quantity is distributed to the lower-layer path planning framework, where equations (62)-(66) are used to determine the target transportation route. The optimal transportation route is dynamically formulated through rolling time-domain optimization based on continuously updated traffic conditions. Once the receiving EHIES receives transportation delay information from the trailer, it will no longer accept hydrogen from the same supplier EHIES to avoid further delivery delays.

[0161] In one embodiment, such as Figure 5The diagram shows the specific steps for determining the transportation route in one embodiment. First, the traffic network marked with the IES location (i.e., the traffic network between the current recipient and the current supplier) is obtained. A road network topology consisting of nodes and directional road segment units is constructed to discretize the traffic network into discrete units. All trailers are set to be located at the starting point of the IES. The density, flow, and speed parameters are initialized based on historical traffic data of the same period. The time period is discretized into multiple small time intervals, and an initial time period is set. The target transportation route optimization dynamic route optimization framework part is determined by equations (62)-(66). The optimal path of the trailer in the next time interval is determined. The trailer moves forward one time interval along the optimal path. The flow, density, and speed parameters of all road segment units are updated by equations (58)-(61). It is determined whether the trailer has arrived. If it has not arrived, the time period is updated, i.e., the initial time period is added by one time interval. The steps of determining the target transportation route optimization dynamic route optimization framework part by equations (62)-(66) are returned to continue execution. If it has arrived, the travel time of the trailer is counted, and the target transportation route is output.

[0162] It should be noted that if the formulas for the first part of the above-mentioned prediction of hydrogen demand (i.e., formulas (1)-(22)) have repeated letters in the second part (formulas (23)-(66)) which are related to the amount of hydrogen and the target transportation route, the repeated letters have different meanings in different parts. To make it easier to distinguish, the repeated letters in the first part can be marked with a subscript 1, and the repeated letters in the second part can be marked with a subscript 2. For example, i and j in formulas (1)-(22) have different meanings than i and j in formulas (23)-(54). To make them different, i and j in formulas (1)-(22) can be replaced with i1 and j1 respectively, and i and j in formulas (23)-(54) can be replaced with i2 and j2 respectively.

[0163] In this embodiment, by determining the traffic network between the current recipient and the current supplier, each road segment is divided into multiple discrete units. The optimization objective is to minimize the transportation time, and constraints are imposed by traffic flow density and traffic flow speed. This ensures that the optimal target transportation route with the minimum transportation time can be solved, thus ensuring the efficient operation and execution of hydrogen dispatch.

[0164] Step 210: Based on the hydrogen mutual aid amount and the target transportation route, execute the current hydrogen dispatch operation.

[0165] For example, the server sends the hydrogen refueling amount and the target transport route to the trailer so that the trailer can complete the current hydrogen dispatch according to the hydrogen refueling amount and the target transport route.

[0166] In the aforementioned hydrogen dispatching method for the distribution network, after detecting that the conditions for initiating a hydrogen dispatching operation are met at the current moment, the method identifies the current recipients and suppliers of hydrogen within the distribution network involved in the current hydrogen dispatching operation. It then acquires multi-dimensional data of the current recipients, including at least weather, traffic, and electricity-hydrogen load dimensions. Based on this multi-dimensional data, a trained hydrogen load prediction model is used to determine the current recipients' hydrogen refueling needs. This approach allows for accurate prediction of actual hydrogen refueling needs by combining multi-dimensional detailed information, improving the accuracy of hydrogen refueling demand prediction. Subsequently, based on the hydrogen refueling demand, the amount of hydrogen exchanged between the current recipients and suppliers is determined in real time. To further ensure dispatching accuracy, the target transportation route between the current recipients and suppliers is optimized with the goal of minimizing transportation time. Finally, based on highly accurate hydrogen exchange amounts and target transportation routes, a more precise current hydrogen dispatching operation can be executed, significantly improving the accuracy of hydrogen dispatching in the distribution network.

[0167] In one embodiment, the method further includes: determining a transportation time threshold corresponding to the target transportation route; when the transportation time of the trailer reaches the transportation time threshold, if the trailer has not reached the current recipient, verifying whether the moment when the transportation time threshold is reached meets the hydrogen dispatch operation start condition; if not, allowing the trailer to continue transportation.

[0168] For example, if the transport time threshold for the target transport route is 1 hour, and the trailer has not arrived at the current recipient when the transport time reaches 1 hour, and if the conditions for initiating the next hydrogen dispatch operation have not been met at this time, the trailer is allowed to continue transporting hydrogen to the current recipient. Of course, if the conditions are met, the transport to the current recipient will be stopped.

[0169] In this embodiment, when transporting according to the target transportation route, once the transportation time exceeds the transportation time threshold, in order to ensure that the transported hydrogen can reach the current recipient as much as possible, it is further verified whether the time when the transportation time threshold is reached meets the hydrogen scheduling operation start condition. If it is not met, it means that the time when the transportation time threshold is reached has not yet reached the time to start the next hydrogen scheduling operation, and transportation to the current recipient can continue.

[0170] In one embodiment, the current supplier determination step includes: obtaining idle hydrogen refueling stations in the distribution network to which the current recipient belongs at the current time; and selecting at least one idle hydrogen refueling station from multiple idle hydrogen refueling stations based on the hydrogen supply capacity of each idle hydrogen refueling station, and designating at least one idle hydrogen refueling station as the current supplier.

[0171] In this context, "idle state" refers to a hydrogen refueling station that has completed the previous hydrogen dispatch operation, or an EHIES deployed at the hydrogen refueling station. That is, after the trailer involved in the previous hydrogen dispatch operation has transported hydrogen from the previous supplier to the previous recipient, the status of the previous supplier is set to idle state. Of course, the status of the previous recipient can also be set to idle state.

[0172] Hydrogen supply capacity reflects the amount of hydrogen supplied by an idle hydrogen refueling station; the greater the amount of hydrogen supplied, the higher the hydrogen supply capacity. For example, a hydrogen supply capacity threshold is obtained, and at least one idle hydrogen refueling station with a hydrogen supply capacity greater than or equal to the threshold is selected from multiple idle hydrogen refueling stations. This selected at least one idle hydrogen refueling station is then designated as the current supplier. This hydrogen supply capacity threshold can be determined based on the historical hydrogen supply demand of the current recipient. For example, idle hydrogen refueling stations are sorted from highest to lowest hydrogen supply capacity, and the idle hydrogen refueling station with the highest sequence number is selected as the current supplier; alternatively, a predetermined number of idle hydrogen refueling stations with the highest sequence numbers are selected as the current suppliers.

[0173] Once the current supplier has been identified, and the hydrogen refueling demand has been determined, it can be further verified whether the current supplier's hydrogen supply capacity meets the hydrogen refueling demand. If so, proceed to step 205.

[0174] In this embodiment, based on the idle hydrogen refueling stations in the distribution network to which the current recipient belongs at the current time, and the hydrogen supply capacity of each idle hydrogen refueling station, at least one idle hydrogen refueling station is selected from multiple idle hydrogen refueling stations. Therefore, the selected at least one idle hydrogen refueling station can be used as the current supplier to ensure the normal operation of subsequent hydrogen dispatch.

[0175] In one specific embodiment, such as Figure 6 The diagram illustrates a technical approach to hydrogen dispatching strategies in a power distribution network, as shown in one embodiment. This involves data-driven prediction of hydrogen demand, followed by knowledge-driven determination of target transportation routes. The implementation is described using a server as the primary implementation entity, and the specific process is as follows:

[0176] In the hydrogen demand forecasting section: First, after detecting that the conditions for initiating a hydrogen dispatch operation are met at the current moment, the current recipients and suppliers of hydrogen in the distribution network involved in the current hydrogen dispatch operation are determined. Then, the Z-score method is used to preprocess all input data (multi-dimensional data) to obtain processed data. Next, XGBoost is used to evaluate the processed data to filter out data whose importance is higher than a certain threshold. Finally, the filtered data is input into BiLSTM to obtain the short-term hydrogen demand of the HFV.

[0177] In the target transportation route determination section: After determining the hydrogen refueling demand, the solution is obtained using Equation (50) as the objective function and Equation (54) as the constraint condition, and the solution is obtained using the Gurobi solver. The solver outputs the amount of hydrogen exchange between the current recipient and the current supplier by processing the variables, constraints and objective function in the model. Then, with the goal of minimizing the hydrogen transportation time between EHIES (current recipient and current supplier), the optimal dynamic hydrogen transportation route is obtained using the sparrow search algorithm, thus obtaining the target transportation route. Based on the amount of hydrogen exchange and the target transportation route, the current hydrogen scheduling operation is performed. Specifically, the idle hydrogen refueling stations in the distribution network to which the current recipient belongs are obtained at the current time; based on the hydrogen supply capacity of each idle hydrogen refueling station, at least one idle hydrogen refueling station is selected from multiple idle hydrogen refueling stations, and at least one idle hydrogen refueling station is selected as the current supplier. Finally, the transportation time threshold corresponding to the target transportation route is determined; when the transportation time of the trailer reaches the transportation time threshold, if the trailer has not reached the current recipient, it is checked whether the time when the transportation time threshold is reached meets the hydrogen dispatch operation start condition; if not, the trailer is allowed to continue transportation.

[0178] In this embodiment, after detecting that the conditions for initiating a hydrogen dispatch operation are met at the current moment, the current recipients and suppliers of hydrogen in the distribution network involved in the current hydrogen dispatch operation are determined. Multi-dimensional data of the current recipients at the current moment is acquired, including at least weather, traffic, and electricity-hydrogen load dimensions. Based on this multi-dimensional data, a trained hydrogen load prediction model is used to determine the current recipients' hydrogen refueling needs. This allows for accurate prediction of the actual current hydrogen refueling needs by combining multi-dimensional detailed information, improving the accuracy of hydrogen refueling demand prediction. Then, based on the hydrogen refueling needs, the amount of hydrogen exchange between the current recipients and suppliers is determined in real time. To further ensure dispatch accuracy, the target transportation route between the current recipients and suppliers is optimized with the goal of minimizing transportation time. Finally, based on the highly accurate hydrogen exchange amount and target transportation route, a more precise current hydrogen dispatch operation can be executed, thereby greatly improving the accuracy of hydrogen dispatch in the distribution network.

[0179] It should be understood that although the steps in the flowcharts of the embodiments described above are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the embodiments described above may include multiple steps or multiple stages. These steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least some of the steps or stages in other steps. It is understood that the steps in different embodiments can be freely combined as needed, and all non-contradictory solutions formed by such combinations are within the scope of protection of this application.

[0180] Based on the same inventive concept, this application also provides a distribution network hydrogen dispatching device for implementing the above-mentioned distribution network hydrogen dispatching method. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more distribution network hydrogen dispatching device embodiments provided below can be found in the limitations of the distribution network hydrogen dispatching method described above, and will not be repeated here.

[0181] In one exemplary embodiment, such as Figure 7 As shown, a hydrogen dispatching device 700 for a power distribution network is provided, including: a hydrogen supplier / recipient determination module 702, a multi-dimensional data acquisition module 704, a hydrogen refueling demand determination module 706, a transportation route determination module 708, and a transportation operation execution module 710, wherein:

[0182] The hydrogen supplier and recipient determination module 702 is used to determine the current recipient of hydrogen in the distribution network and the current supplier of hydrogen after detecting that the hydrogen dispatch operation start conditions are met at the current time.

[0183] The multi-dimensional data acquisition module 704 is used to acquire the multi-dimensional data of the current recipient at the current moment. The multi-dimensional data includes at least the corresponding dimension data of weather dimension, traffic dimension and electric hydrogen load dimension.

[0184] The hydrogen demand determination module 706 is used to determine the current hydrogen demand of the recipient based on multi-dimensional data and a trained hydrogen load prediction model.

[0185] The transportation route determination module 708 is used to determine the amount of hydrogen exchange between the current recipient and the current supplier based on the hydrogen refueling demand, and to determine the target transportation route between the current recipient and the current supplier with the goal of minimizing transportation time.

[0186] The transportation operation execution module 710 is used to execute the current hydrogen scheduling operation based on the hydrogen mutual aid amount and the target transportation route.

[0187] In one embodiment, the trained hydrogen load prediction model includes a trained data filtering model and a trained hydrogen refueling demand prediction model. The hydrogen refueling demand determination module 706 is used to preprocess the multi-dimensional data to obtain processed data; based on the processed data, the trained data filtering model filters out data whose importance is higher than a certain threshold; based on the filtered data, the trained hydrogen refueling demand prediction model determines the current hydrogen refueling demand of the recipient.

[0188] In one embodiment, the transportation route determination module 708 is used to construct an objective function with the goal of maximizing the renewable energy absorption rate, and to establish current and voltage constraints related to the power distribution network; based on the hydrogen demand, objective function, current constraints, and voltage constraints, the solver calculates the amount of hydrogen provided by each current supplier to the current recipient.

[0189] In one embodiment, the transportation route determination module 708 is used to divide each segment of the transportation network between the current recipient and the current supplier into multiple discrete units; with the goal of minimizing transportation time, traffic flow density constraints and traffic flow speed constraints are constructed for each discrete unit; based on the optimization goal and the traffic flow density constraints and traffic flow speed constraints, the target transportation route between the current recipient and the current supplier is solved by a sparrow search algorithm.

[0190] In one embodiment, the device further includes a condition judgment module for determining the transportation time threshold corresponding to the target transportation route; when the transportation time of the trailer reaches the transportation time threshold, if the trailer has not reached the current recipient, it verifies whether the moment when the transportation time threshold is reached meets the hydrogen dispatch operation start condition; if not, the trailer is allowed to continue transportation.

[0191] In one embodiment, the apparatus further includes a supplier determination module, which is used to obtain idle hydrogen refueling stations in the distribution network to which the current recipient belongs at the current time; and select at least one idle hydrogen refueling station from multiple idle hydrogen refueling stations based on the hydrogen supply capacity of each idle hydrogen refueling station, and designate at least one idle hydrogen refueling station as the current supplier.

[0192] Each module in the aforementioned hydrogen dispatching device for the power distribution network can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in or independent of the processor in a computer device, or stored in the memory of a computer device as software, so that the processor can call and execute the corresponding operations of each module.

[0193] In one exemplary embodiment, a computer device is provided, which may be a server, and its internal structure diagram may be as follows: Figure 8 As shown, this computer device includes a processor, memory, input / output (I / O) interfaces, and a communication interface. The processor, memory, and I / O interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the I / O interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system, computer programs, and a database. The internal memory provides the environment for the operating system and computer programs stored in the non-volatile storage media to run. The I / O interfaces are used for exchanging information between the processor and external devices. The communication interface is used for communicating with external terminals via a network connection. When the computer program is executed by the processor, it implements a hydrogen dispatching method for a power distribution network.

[0194] Those skilled in the art will understand that Figure 8 The structure shown is merely a block diagram of a portion of the structure related to the present application and does not constitute a limitation on the computer device to which the present application is applied. Specific computer devices may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0195] In one embodiment, a computer device is also provided, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps in the above method embodiments.

[0196] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0197] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0198] It should be noted that the user information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to data used for analysis, data stored, data displayed, etc.) involved in this application are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of the relevant data must comply with relevant regulations.

[0199] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this application can include at least one of non-volatile memory and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this application may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this application may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, artificial intelligence (AI) processors, etc., and are not limited to these.

[0200] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this application.

[0201] The embodiments described above are merely illustrative of several implementation methods of this application, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of this patent application. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this application should be determined by the appended claims.

Claims

1. A method for hydrogen dispatching in a power distribution network, characterized in that, The method includes: After detecting that the conditions for starting the hydrogen dispatch operation are met at the current moment, determine the current recipients of hydrogen in the distribution network and the current suppliers of hydrogen involved in the current hydrogen dispatch operation. Obtain multi-dimensional data of the current recipient at the current moment, wherein the multi-dimensional data includes at least the corresponding dimension data of weather dimension, traffic dimension and electric hydrogen load dimension; Based on the multi-dimensional data, the hydrogen loading prediction model is used to determine the current hydrogen demand of the recipient. Based on the hydrogen refueling demand, determine the amount of hydrogen exchange between the current recipient and the current supplier, and determine the target transportation route between the current recipient and the current supplier with the goal of minimizing transportation time. Based on the stated hydrogen resupply amount and the target transportation route, execute the current hydrogen dispatch operation.

2. The method according to claim 1, characterized in that, The trained hydrogen load prediction model includes a trained data filtering model and a trained hydrogen refueling demand prediction model. The process of determining the current recipient's hydrogen refueling demand based on the multi-dimensional data and the trained hydrogen load prediction model includes: After performing data preprocessing on the multi-dimensional data, the processed data is obtained; Based on the processed data, a trained data filtering model is used to filter out data whose importance is higher than the importance threshold. Based on the selected data, the current hydrogen demand of the recipient is determined using a trained hydrogen demand prediction model.

3. The method according to claim 1, characterized in that, The determination of the hydrogen refueling volume between the current recipient and the current supplier based on the hydrogen demand includes: An objective function is constructed with the goal of maximizing the renewable energy absorption rate, and current and voltage constraints related to the distribution network are established. Based on the hydrogen demand, the objective function, the current constraint, and the voltage constraint, the solver calculates the amount of hydrogen supplied by each current supplier to the current recipient.

4. The method according to claim 1, characterized in that, The step of determining the target transportation route between the current recipient and the current supplier with the optimization objective of minimizing transportation time includes: Divide each segment of the traffic network between the current recipient and the current supplier into multiple discrete units; The optimization objective is to minimize the transportation time, and traffic flow density constraints and traffic flow velocity constraints are constructed for each discrete unit. Based on the optimization objective and the constraints of traffic flow density and traffic flow speed, the target transportation route between the current recipient and the current supplier is solved using the sparrow search algorithm.

5. The method according to claim 1, characterized in that, The method further includes: Determine the transportation time threshold corresponding to the target transportation route; When the transport time of the trailer reaches the transport time threshold, if the trailer has not reached the current recipient, it is checked whether the time when the transport time threshold is reached meets the hydrogen dispatch operation start condition. If the conditions are not met, the trailer may continue transporting goods.

6. The method according to any one of claims 1 to 5, characterized in that, The current supplier determination step includes: Obtain available hydrogen refueling stations that are currently idle in the distribution network to which the recipient belongs at the current time; Based on the hydrogen supply capacity of each idle hydrogen refueling station, at least one idle hydrogen refueling station is selected from multiple idle hydrogen refueling stations and designated as the current supplier.

7. A hydrogen dispatching device for a power distribution network, characterized in that, The device includes: The hydrogen supplier and recipient determination module is used to determine the current recipients of hydrogen in the distribution network and the current suppliers of hydrogen involved in the current hydrogen dispatch operation after detecting that the conditions for starting the hydrogen dispatch operation are met at the current time. The multi-dimensional data acquisition module is used to acquire the multi-dimensional data of the current recipient at the current moment. The multi-dimensional data includes at least the dimension data corresponding to the weather dimension, traffic dimension and electric hydrogen load dimension. The hydrogen demand determination module is used to determine the current hydrogen demand of the recipient based on the multi-dimensional data and through a trained hydrogen load prediction model. The transportation route determination module is used to determine the amount of hydrogen exchange between the current recipient and the current supplier based on the hydrogen refueling demand, and to determine the target transportation route between the current recipient and the current supplier with the goal of minimizing transportation time. The transportation operation execution module is used to execute the current hydrogen scheduling operation based on the hydrogen mutual aid quantity and the target transportation route.

8. 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 method according to any one of claims 1 to 6.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

10. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by a processor, it implements the steps of the method according to any one of claims 1 to 6.

Images