Decentralized charging load access power grid management method based on alliance chain

By adopting alliance chain technology and deep learning models in the power grid management system, changes in charging demand and data security issues are solved, efficient transaction processing and data protection are achieved, and grid operation efficiency and user charging satisfaction are improved.

CN118982162APending Publication Date: 2024-11-19HAINAN POWER GRID CO LTD ELECTRIC POWER RES INST
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202410829939.2
Authority / Receiving Office
CN · China
Patent Type
Applications(China)
Current Assignee / Owner
Filing Date
2024-06-25
Publication Date
2024-11-19

AI Technical Summary

Technical Problem

The existing power grid management system has limitations in responding to changes in charging demand for electric vehicles, ensuring data security and user privacy protection, especially in terms of rapidly changing charging load requirements and data security.

Method used

Adopting a decentralized management method based on alliance chain, by building a multi-level consensus mechanism electric vehicle charging alliance chain network, designing fine-grained access control strategies, deploying condition-triggered charging load access smart contracts, and using deep learning algorithms to build Charging load prediction model, and developing a load scheduling strategy based on multi-objective optimization to realize automatic execution of smart contracts and real-time monitoring.

Benefits of technology

It significantly improves transaction processing speed and data security, enhances the privacy protection capabilities of the network, improves the accuracy of charging demand forecasts and grid operation efficiency, and enhances the management flexibility and emergency response capabilities of the electric vehicle charging network.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118982162A_ABST
    Figure CN118982162A_ABST
Patent Text Reader

Abstract

The invention discloses a decentralized charging load access power grid management method based on an alliance chain, and relates to the technical field of power grids, and the method comprises the steps: constructing an electric vehicle charging alliance chain network based on a multistage consensus mechanism; deploying a condition triggering type charging load access intelligent contract in the alliance chain network; analyzing the preprocessed data by using a deep learning algorithm to construct a charging load prediction model, and dynamically predicting a charging demand in a short period; a load scheduling strategy based on multi-objective optimization is developed, and load distribution of the charging stations is optimized on the premise that stable operation of the power grid is met; and the optimized charging load scheduling scheme is automatically executed by using an intelligent contract, and a real-time monitoring and adjusting mechanism is provided to deal with emergencies and abnormal conditions. According to the invention, the CP-ABE technology is adopted, and the alliance chain network realizes a fine-grained access control strategy.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of power grid technology, and in particular to a method and system for managing decentralized charging load access to a power grid based on an alliance chain. Background Art

[0002] With the widespread use of electric vehicles and the advancement of smart energy management systems, the construction and management of charging infrastructure has become one of the key factors in optimizing and updating the power grid system. The increase in demand for electric vehicle charging has put tremendous pressure on the power grid, especially during peak hours. At present, power grid management mainly relies on traditional centralized control systems, which have limitations in coping with rapidly changing charging demands and achieving efficient energy distribution. In addition, with the increase in the proportion of renewable energy, the volatility and uncertainty of the power grid have increased, and there is an urgent need to develop more flexible and efficient management strategies to ensure the stability and reliability of the power grid.

[0003] In addition, existing grid management methods face challenges in data security and privacy protection. Due to the sensitivity of charging infrastructure and grid data, it is necessary to ensure that information security and user privacy are fully protected. However, the traditional centralized management model is prone to become a target of cyber attacks. Once the data center is attacked, the security and stability of the entire system will be affected. Summary of the invention

[0004] In view of the limitations of existing power grid management systems in coping with changes in electric vehicle charging demand and ensuring data security and privacy protection, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is how to effectively respond to the rapidly changing charging load demand while ensuring the security of transaction data and the protection of user privacy, and improving the stability and efficiency of the power grid.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In the first aspect, an embodiment of the present invention provides a decentralized charging load access to the power grid management method based on an alliance chain, which includes building an electric vehicle charging alliance chain network based on a multi-level consensus mechanism, and designing a fine-grained access control strategy; deploying a condition-triggered charging load access smart contract in the alliance chain network, and automatically triggering the access of the charging load according to the conditions; collecting grid-side data and user-side data in real time through the alliance chain network nodes, and preprocessing the collected data; using a deep learning algorithm to analyze the preprocessed data to build a charging load prediction model, and dynamically predict the charging demand in the short term; developing a load scheduling strategy based on multi-objective optimization to optimize the load distribution of charging sites while meeting the stable operation of the power grid; using smart contracts to automatically execute the optimized charging load scheduling plan, and provide a real-time monitoring and adjustment mechanism to deal with emergencies and abnormal situations.

[0008] As a preferred solution of the decentralized charging load access to the power grid management method based on the alliance chain described in the present invention, wherein: building an electric vehicle charging alliance chain network based on a multi-level consensus mechanism includes the following steps: designing a layered alliance chain architecture, dividing the alliance chain network into three layers: core layer, operation layer and edge layer; based on CP-ABE technology, generating attribute keys for three types of entities: power grid companies, operators, and users, and the attribute keys include attribute private keys and subject public keys; when charging data needs to be shared and accessed, the charging pile nodes in the edge layer perform the first CP-ABE encryption on the local data and formulate a preliminary access strategy, and upload the encrypted data and the corresponding access strategy to the operation layer of the alliance chain; the operation layer uses the public keys of each entity to perform the second CP-ABE encryption on the encrypted data and access strategy from the edge layer to generate the final encrypted data and access policy ciphertext; the subject that meets the access strategy uses its own attribute private key to decrypt the encrypted data and access policy ciphertext to obtain data content and access rights; based on the blockchain identity framework, the attribute permission information of the subject is recorded and maintained, and a monitoring contract is deployed to automatically update the attribute key, revoke or restore the corresponding authority by analyzing the changes in the subject status; and corresponding cascade consensus paths are formulated for different types of transactions.

[0009] As a preferred solution of the decentralized charging load access to the power grid management method based on the alliance chain described in the present invention, wherein: formulating corresponding cascade consensus paths for different types of transactions includes the following steps: if it involves major decisions of the entire electric vehicle charging ecosystem, it is determined to be a key decision-making transaction, and the cascade consensus path is as follows: the core layer authoritative node initiates a key transaction proposal, the operation layer node pre-executes and votes on the proposal, the edge layer node provides DPoS weight for the vote of the operation layer node to which it belongs, and the smart contract collects and summarizes the voting results of each level, and decides whether to execute this transaction according to the preset high voting threshold; if it involves the daily operation and management of the alliance chain, it is determined to be a general operation transaction, and the cascade consensus path is as follows: the operation layer node initiates a transaction proposal, the edge layer charging pile node runs the PoA consensus mechanism to review and vote on the proposal, and the smart contract decides whether to execute the transaction and upload it to the chain according to the preset medium voting threshold; if only a quick confirmation is required at the edge layer, and no upper layer needs to participate in the consensus, it is determined to be a bottom-level data transaction, and the cascade consensus path is as follows: the edge layer charging pile node initiates a transaction according to the local data status, and directly executes the transaction after reaching a consensus through the PoA consensus, and uploads the result to the chain.

[0010] As a preferred solution of the decentralized charging load access power grid management method based on alliance chain described in the present invention, wherein: constructing a charging load prediction model includes the following steps: normalizing the real-time collected grid-side data and user-side data so that data of different dimensions are uniformly mapped to the [0,1] interval;

[0011] A deep belief network (DBN) is constructed by stacking multiple restricted Boltzmann machine (RBM) layers, and an unsupervised layer-by-layer greedy training strategy is adopted for the RBM. The contribution of each feature to the prediction result is analyzed, and important feature subsets are selected, while more training data is generated through data amplification. The trained DBN is used to encode the normalized data into a sequence of hidden layer feature vectors, and these sequences are used as the input of the Seq2Seq model. The DBN and Seq2Seq models are alternately trained using the mean square error loss, and dynamic load forecasting is performed by rolling the prediction time window during the test phase.

[0012] As a preferred solution of the decentralized charging load access to the power grid management method based on the alliance chain of the present invention, wherein: the unsupervised layer-by-layer greedy training strategy for RBM includes the following steps: training the first RBM layer to learn the characteristics of the input data; using the hidden layer output of the first RBM as the input of the second RBM, repeating this process until all layers are trained; introducing random noise in the training to improve the generalization ability of the model; adding a regression output layer at the top layer of DBN to predict future charging demand; the contribution calculation formula is as follows:

[0013]

[0014] Among them, f i Represents the characteristics, I(f i ) represents the contribution, y j Indicates the actual load value. represents the average value of the load, f i (x j ) indicates that the model uses feature f i The predicted value at time , N represents the total number of data points.

[0015] The encoder of the Seq2Seq model uses a bidirectional long short-term memory network, and the decoder uses a unidirectional LSTM. The specific formula for predicting sequence load elements is as follows:

[0016] h t =LSTM(h t-1 ,y t-1 ,c t )

[0017]

[0018] Among them, h t represents the hidden layer state, y t represents the value of the output sequence at time step t, c t represents the cell state of the LSTM unit, W h and b represent training parameters.

[0019] As a preferred solution of the decentralized charging load access power grid management method based on alliance chain described in the present invention, the development of a load scheduling strategy based on multi-objective optimization includes using the DBN-Seq2Seq charging load prediction model, combining the collected grid-side and user-side data and the real-time data obtained by the alliance chain to predict the charging load curve L(t) in the future T period; obtaining real-time data including renewable energy power generation power P from the electric vehicle charging alliance chain network. w (t), grid load curve P g(t), parameters of each generator, electricity price curve P(t), line and node voltage constraints; based on the charging load curve L(t) and real-time data, an improved multi-objective optimization model is established; MOEA / D-AWA is used to perform intelligent noise reduction on the load curve predicted by DBN to generate a high-quality initial population; the consensus index C(x) is introduced as an additional evaluation mechanism, and the consensus convergence speed of the solution is optimized by an adaptive variant operator based on C(x); the optimized solution is decoded to obtain the load scheduling schemes and charging station scheduling quantities X(r) at each level; the core layer decomposes X(r) into the operation layer scheduling X(m), which is issued through the alliance chain, and the operation layer decomposes X(m) into the edge layer charging pile execution instructions X(e), and the edge layer performs the actual distributed load scheduling operation according to X(e); the entire scheduling process is monitored and traced based on the alliance chain data, and the load response of each level is monitored in real time in combination with the Internet of Things technology, while the actual operation data of the power grid is collected to evaluate and adjust the scheduling strategy.

[0020] As a preferred solution of the decentralized charging load access to the power grid management method based on the alliance chain of the present invention, the specific formula of the improved multi-objective optimization model is as follows:

[0021]

[0022] Among them, f1(x) and f2(x) represent the cost-effectiveness objective function and the energy efficiency objective function, c i represents the charging cost per unit time of the i-th charging station, p i (t) represents the power of the i-th charging station at time t, Δt represents the time interval, and p gent,t represents the generated power at time t, p load,t represents the load power at time t, t represents the time step, i represents the charging station, N represents the total number of charging stations, T represents the time period, and p i,min 、p i,max Represents the minimum power and maximum power respectively, V j (t) represents the voltage of node j at time point t, V min 、V max Indicates minimum and maximum voltage.

[0023] On the second aspect, an embodiment of the present invention provides a decentralized charging load access power grid management system based on an alliance chain, which includes an access control module for building an electric vehicle charging alliance chain network based on a multi-level consensus mechanism and designing a fine-grained access control strategy; a conditional contract module for deploying a condition-triggered charging load access smart contract in the alliance chain network, and automatically triggering the access of the charging load according to the conditions; a data acquisition module for collecting grid-side data and user-side data in real time through the alliance chain network nodes, and preprocessing the collected data; a model building module for analyzing the preprocessed data using a deep learning algorithm to build a charging load prediction model, and dynamically predict the charging demand in the short term; a load scheduling module for developing a load scheduling strategy based on multi-objective optimization, and optimizing the load distribution of charging sites while meeting the stable operation of the power grid; an execution adjustment module for automatically executing the optimized charging load scheduling plan using a smart contract, and providing a real-time monitoring and adjustment mechanism to deal with emergencies and abnormal situations.

[0024] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, the steps of the decentralized charging load access to the power grid management method based on the alliance chain as described in the first aspect of the present invention are implemented.

[0025] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, the steps of the decentralized charging load access to the power grid management method based on the alliance chain as described in the first aspect of the present invention are implemented.

[0026] The beneficial effects of the present invention are as follows: the present invention significantly improves the transaction processing speed and shortens the confirmation time by adopting a hierarchical consensus mechanism. At the same time, the CP-ABE technology is used to achieve fine-grained control of data access, thereby improving the data security and privacy protection capabilities of the network. The introduction of the deep learning model makes the charging demand forecast more accurate. Combined with the multi-objective optimization model, the distribution of charging load is further optimized, thereby effectively improving the grid operation efficiency and the utilization rate of renewable energy. In addition, the conditionally triggered smart contract and real-time monitoring load scheduling execution mechanism of the present invention increase the management flexibility and emergency response capabilities of the electric vehicle charging network, enabling it to quickly adapt to changes in the grid and the market, effectively respond to emergencies and abnormal situations, and ensure the stable operation of the charging network and meet the charging needs of users. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work.

[0028] Figure 1 The framework flow chart of the decentralized charging load access to the power grid management method based on the alliance chain.

[0029] Figure 2 A consortium chain design diagram for a decentralized charging load access to the power grid management method based on a consortium chain.

[0030] Figure 3 Computer equipment diagram of a decentralized charging load access grid management method based on alliance chain. DETAILED DESCRIPTION

[0031] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0032] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0033] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0034] Example 1

[0035] Reference Figure 1 to Figure 3 , which is the first embodiment of the present invention, and provides a decentralized charging load access power grid management method based on alliance chain, including:

[0036] S1: Build an electric vehicle charging alliance chain network based on a multi-level consensus mechanism and design a fine-grained access control strategy.

[0037] Specifically, a layered alliance chain architecture is designed to divide the alliance chain network into three layers: core layer, operation layer and edge layer; based on CP-ABE (attribute-based encryption) technology, attribute keys are generated for three types of entities: power grid companies, operators and users, including attribute private keys and subject public keys; when data (such as charging data) needs to be shared and accessed, the charging pile nodes in the edge layer perform the first CP-ABE encryption on the local data and formulate a preliminary access strategy, and upload the encrypted data and the corresponding access strategy to the operation layer of the alliance chain; the operation layer uses the public keys of each entity to perform a second CP-ABE encryption on the encrypted data and access policy from the edge layer to generate the final encrypted data and access policy ciphertext; the subject that meets the access policy uses its own attribute private key to decrypt the encrypted data and access policy ciphertext to obtain the data content and access rights; based on the blockchain identity framework, the attribute permission information of the subject is recorded and maintained, and a monitoring contract is deployed to automatically update the attribute key, revoke or restore the corresponding authority by analyzing the changes in the subject status; for different types of transactions, corresponding cascade consensus paths are formulated.

[0038] Furthermore, the corresponding cascading consensus path is formulated for different types of transactions, including the following steps: If it involves major decisions of the entire electric vehicle charging ecosystem (such as modifying the alliance chain governance rules and updating technical standards), it is determined to be a key decision-making transaction, and the cascading consensus path is as follows: the core layer authoritative node initiates a key transaction proposal, the operation layer node pre-executes and votes on the proposal, the edge layer node provides DPoS weight for the vote of the operation layer node to which it belongs, and the smart contract collects and summarizes the voting results of each level, and decides whether to execute the transaction according to the preset high voting threshold; if it involves the daily operation and management of the alliance chain (charging pricing adjustment, new node admission), it is determined to be a general operation transaction, and the cascading consensus path is as follows: the operation layer node initiates a transaction proposal, the edge layer charging pile node runs the PoA consensus mechanism to review and vote on the proposal, and the smart contract decides whether to execute the transaction and upload it to the chain according to the preset medium voting threshold; if only a quick confirmation is required at the edge layer, and no upper layer needs to participate in the consensus, it is determined to be a bottom-level data transaction, and the cascading consensus path is as follows: the edge layer charging pile node initiates a transaction based on the local data status, and directly executes the transaction after reaching a consensus through the PoA consensus, and uploads the result to the chain.

[0039] It should be noted that the core layer is composed of authoritative nodes such as power grid companies and government regulatory departments, and runs the PBFT consensus algorithm to ensure the consistency and security of decision-making. The operation layer is composed of nodes of relevant stakeholders such as charging station operators and car companies, and runs the DPoS consensus algorithm to improve transaction processing efficiency. The edge layer is composed of charging pile nodes, and runs a lightweight PoA consensus mechanism to participate in the overall consensus process.

[0040] It should be noted that the preset voting threshold refers to the voting ratio threshold pre-set in the smart contract. The high voting threshold is that the proportion of votes cast in favor by the core layer authoritative nodes is not less than 70%, the proportion of votes cast in favor by the operation layer nodes is not less than 60%, and the proportion of DPoS weight support provided by the edge layer for the operation layer vote is not less than 50%; the medium voting threshold is that the number of nodes that vote in favor of the proposal shall not be less than 75% when the proposal is reviewed and voted on by running the PoA consensus mechanism.

[0041] Preferably, different consensus algorithms are run on nodes at different levels, which is conducive to balancing the security, efficiency and decentralization of the system; CP-ABE technology is used to perform multiple encryptions on data, and the blockchain identity framework is combined to dynamically manage subject attribute permissions, thereby improving the refinement of data access control; a cascading consensus path is designed, and a differentiated consensus process is adopted for different types of transactions, thereby improving the system processing efficiency and flexibility.

[0042] S2: Deploy a condition-triggered charging load access smart contract in the alliance chain network to automatically trigger the access of the charging load according to the conditions.

[0043] Specifically, the triggering conditions for charging load access are determined to be one or more of the following: renewable energy power generation reaches a preset threshold, grid load is lower than a preset threshold, within a preset specific time period, and other conditions set based on the grid operation status or management needs; the core layer authoritative node initiates a proposal to deploy a condition monitoring contract based on the determined triggering conditions, and the contract assumes the responsibility of real-time monitoring whether the triggering conditions are met; the operation layer node initiates a proposal to deploy a charging load access contract, which is associated with the condition monitoring contract and is automatically executed when the triggering conditions are monitored to be met; the condition monitoring contract continuously monitors the preset triggering conditions, and if the preset triggering conditions are monitored to be met, the condition monitoring contract automatically triggers the associated charging load access contract and executes it; after the charging load access contract is triggered, the edge layer charging pile node performs the actual charging load access operation according to the contract instructions; the edge layer charging pile node submits the result of the executed load access operation as the underlying data transaction, and uploads the result to the chain after reaching a consensus on the transaction review through the PoA consensus mechanism; utilizing the tamper-proof characteristics of blockchain data, relevant nodes and participants can query the on-chain data to monitor and trace the entire process of the charging load access operation.

[0044] S3: Collect grid-side data and user-side data in real time through alliance chain network nodes, and pre-process the collected data.

[0045] S4: Use deep learning algorithms to analyze the preprocessed data to build a charging load prediction model and dynamically predict the charging demand in the short term.

[0046] Specifically, the following steps are included:

[0047] S4.1: Normalize the real-time collected grid-side data and user-side data so that data of different dimensions are uniformly mapped to the [0,1] interval.

[0048] Furthermore, the specific formula for normalization is as follows:

[0049]

[0050] Where x represents the original grid-side data / user-side data, x' represents the normalized grid-side data / user-side data, and x min 、x max Indicates the minimum and maximum values ​​in the grid-side data set / user-side data set.

[0051] S4.2: Use multiple restricted Boltzmann machine (RBM) layers to build a deep belief network (DBN), and adopt an unsupervised layer-by-layer greedy training strategy for RBM.

[0052] Specifically, each RBM layer uses the hidden layer activations of the previous layer as the input of its visible layer for training, adopting an unsupervised layer-by-layer greedy training strategy; train the first RBM layer to learn the characteristics of the input data; use the hidden layer output of the first RBM as the input of the second RBM, and repeat this process until all layers are trained; introduce random noise during training to improve the generalization ability of the model; and add a regression output layer at the top layer of the DBN to predict future charging demand.

[0053] S4.3: Analyze the contribution of each feature to the prediction results and select important feature subsets, while generating more training data through data augmentation.

[0054] Specifically, the contribution calculation formula is as follows:

[0055]

[0056] Among them, f i Represents the characteristics, I(f i ) represents the contribution, y j represents the actual load value, y represents the average load value, and f i (x j ) indicates that the model uses feature f i The predicted value at time , N represents the total number of data points.

[0057] S4.4: Use the trained DBN to encode the normalized data into hidden feature vector sequences, and use these sequences as the input of the Seq2Seq model.

[0058] It should be noted that the encoder of the Seq2Seq (sequence to sequence) model uses a bidirectional long short-term memory network (BiLSTM) and the decoder uses a unidirectional LSTM.

[0059] Specifically, the specific formula for predicting the sequence load element is as follows:

[0060] h t =LSTM(h t-1 ,y t-1 ,c t )

[0061]

[0062] Among them, h t represents the hidden layer state, y t represents the value of the output sequence at time step t, c t represents the cell state of the LSTM unit, W h and b represent training parameters.

[0063] S4.5: DBN and Seq2Seq models are alternately trained using mean squared error loss, and dynamic load forecasting is performed by rolling the prediction time window during the test phase.

[0064] The optimal charging load prediction model combines unsupervised deep learning with supervised feature learning. DBN captures the deep features of the data, and Seq2Seq captures the temporal correlation of the data, forming an end-to-end prediction process, which improves the prediction accuracy and robustness. At the same time, it focuses on feature engineering and data enhancement to further improve the generalization ability of the model.

[0065] S5: Develop a load dispatching strategy based on multi-objective optimization to optimize the load distribution of charging stations while ensuring stable operation of the power grid.

[0066] Specifically, the DBN-Seq2Seq charging load prediction model is used to combine the collected grid-side and user measurement data with the real-time data obtained by the alliance chain to predict the charging load curve L(t) in the future T period, t = 1, 2, 3, ..., T; real-time data including renewable energy power generation power P is obtained from the electric vehicle charging alliance chain network. w (t), grid load curve P g(t), parameters of each generator (such as maximum / minimum output, ramp rate, etc.), electricity price curve P(t), line and node voltage constraints; based on the charging load curve L(t) and real-time data, an improved multi-objective optimization model is established; MOEA / D-AWA (multi-objective optimization algorithm / adaptive weight adjustment) is used to perform intelligent noise reduction on the load curve predicted by DBN to generate a high-quality initial population (this initial population is the starting point for the initial optimization of the evolutionary algorithm, and its quality directly affects the convergence speed of the algorithm and the quality of the final solution); the consensus index C(x) is introduced as an additional evaluation mechanism, and the The consensus convergence speed of the adaptive variant operator optimization solution of C(x); the optimized solution is decoded to obtain the load scheduling scheme of each level and the charging station scheduling quantity X(r); the core layer decomposes X(r) into the operation layer scheduling X(m), which is issued through the alliance chain. The operation layer decomposes X(m) into the edge layer charging pile execution instruction X(e), and the edge layer performs the actual distributed load scheduling operation according to X(e); the entire scheduling process is monitored and traced based on the alliance chain data, and the load response of each level is monitored in real time in combination with the Internet of Things technology, while collecting the actual operation data of the power grid to evaluate and adjust the scheduling strategy.

[0067] Preferably, the multi-objective optimization model includes an objective function and constraints, wherein the objective function includes cost minimization and energy efficiency maximization, and the constraints include power balance constraints, charging station power range constraints, and power grid stability constraints. The specific formula is as follows:

[0068]

[0069] Among them, f1(x) and f2(x) represent the cost-effectiveness objective function and the energy efficiency objective function, c i represents the charging cost per unit time of the i-th charging station, p i (t) represents the power of the i-th charging station at time t, Δt represents the time interval, and p gent,t represents the generated power at time t, p load,t represents the load power at time t, t represents the time step, i represents the charging station, N represents the total number of charging stations, T represents the time period, and p i,min 、p i,max Represents the minimum power and maximum power respectively, V j (t) represents the voltage of node j at time point t, V min 、V max Indicates minimum and maximum voltage.

[0070] S6: Use smart contracts to automatically execute the optimized charging load scheduling plan and provide real-time monitoring and adjustment mechanisms to deal with emergencies and abnormal situations.

[0071] S7: Put charging transaction data on the chain and introduce an intelligent audit mechanism to automatically detect and correct abnormal transactions.

[0072] Furthermore, this embodiment also provides a decentralized charging load access power grid management system based on the alliance chain, including an access control module for building an electric vehicle charging alliance chain network based on a multi-level consensus mechanism and designing a fine-grained access control strategy; a conditional contract module for deploying a condition-triggered charging load access smart contract in the alliance chain network, and automatically triggering the access of the charging load according to the conditions; a data acquisition module for collecting grid-side data and user-side data in real time through the alliance chain network nodes, and preprocessing the collected data; a model building module for analyzing the preprocessed data using a deep learning algorithm to build a charging load prediction model, and dynamically predict the charging demand in the short term; a load scheduling module for developing a load scheduling strategy based on multi-objective optimization, and optimizing the load distribution of charging sites while meeting the stable operation of the power grid; an execution adjustment module for automatically executing the optimized charging load scheduling plan using a smart contract, and providing a real-time monitoring and adjustment mechanism to deal with emergencies and abnormal situations.

[0073] This embodiment also provides a computer device, which is suitable for the decentralized charging load access to the power grid management method based on the alliance chain, including a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the decentralized charging load access to the power grid management method based on the alliance chain as proposed in the above embodiment.

[0074] The computer device may be a terminal, and the computer device includes a processor connected via a bus,

[0075] The computer device includes a memory, a communication interface, a display screen and an input device. 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 and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device housing, or an external keyboard, touchpad or mouse, etc.

[0076] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the method for implementing decentralized charging load access to the power grid management based on the alliance chain as proposed in the above embodiment is implemented.

[0077] In summary, the present invention significantly improves the transaction processing speed and shortens the confirmation time by adopting a hierarchical consensus mechanism. At the same time, the CP-ABE technology is used to achieve fine-grained control of data access, thereby improving the data security and privacy protection capabilities of the network. The introduction of the deep learning model makes the charging demand forecast more accurate. Combined with the multi-objective optimization model, the distribution of charging load is further optimized, thereby effectively improving the grid operation efficiency and the utilization rate of renewable energy. In addition, the conditionally triggered smart contract and real-time monitoring load scheduling execution mechanism of the present invention increase the management flexibility and emergency response capabilities of the electric vehicle charging network, enabling it to quickly adapt to changes in the power grid and the market, effectively respond to emergencies and abnormal situations, and ensure the stable operation of the charging network and meet the charging needs of users.

[0078] Example 2

[0079] Reference Figure 1 to Figure 3 , which is the second embodiment of the present invention, provides a decentralized charging load access to the power grid management method based on the alliance chain. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments.

[0080] Specifically, the alliance chain network includes 5 nodes in the core layer (composed of power grid companies and government regulatory departments), 10 nodes in the operation layer (composed of major charging station operators and car companies), and 50 charging pile nodes in the edge layer. 10 key decision-making transactions, 50 general operation transactions, and 200 underlying data transactions were considered. The assumptions about power data include an average daily renewable energy generation of 5,000MW, a grid load of 3,000MW, and a user-side demand of 4,500MW.

[0081] The experimental process is divided into two stages: construction and operation. In the construction stage, a three-layer alliance chain architecture was first built, and PBFT, DPoS, and PoA consensus mechanisms were configured at different layers. Then, keys were assigned using attribute-based encryption (CP-ABE) technology, and encryption and access control policies were set based on preset attributes. In the operation stage, various types of transactions were simulated, and the decision-making process was handled through a multi-level consensus mechanism. A conditionally triggered charging load access smart contract was deployed to ensure that it automatically executes related operations when the trigger conditions are met. At the same time, a deep learning model is used to predict charging demand, and the load distribution is adjusted through a multi-objective optimization algorithm based on the predicted data.

[0082] The experimental results show that the approval rate of key decision-making transactions reached 98%, the review and approval process time of general operational transactions was shortened by 30% compared with before, and the confirmation time of underlying data transactions was less than 5 seconds. When all trigger conditions are met, the charging load access contract is successfully executed, and the response time of the edge layer charging pile is 2 seconds on average. The prediction accuracy of the deep network model on the test data reached 94%. After data enhancement and feature selection, the stability of the prediction model was improved by 10%. The load distribution scheme adjusted by the optimization algorithm has improved the grid operation efficiency by 15% and the cost-effectiveness by 20%.

[0083] Preferably, the comparison indicators of the present invention and the traditional method are shown in Table 1.

[0084] Table 1 Comparison indexes of the present invention and the traditional method

[0085] index The present invention Traditional methods Transaction processing time Underlying data transaction <5s >30s Transaction approval rate 98% About 70% Smart contract execution response time Average 2s >10s Load forecast accuracy 94% About 80% System operation efficiency 15% improvement No significant changes Cost-effectiveness Increased by 20% No significant changes

[0086] Preferably, it can be clearly seen from Table 1 that the present invention brings significant improvements and advantages to the management of electric vehicle charging networks, including a significant increase in transaction processing speed, an increase in decision-making approval rate, the efficiency of smart contract execution, the accuracy of load forecasting, and the improvement of the overall operating efficiency and cost-effectiveness of the system. In addition, the new solution significantly enhances the security of the system by introducing advanced encryption and consensus mechanisms, and improves the level of data protection through fine-grained access control. These improvements make the new solution show great application potential and value in the management of modern electric vehicle charging networks.

[0087] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A decentralized charging load access power grid management method based on alliance chain, characterized by: include, Build an electric vehicle charging alliance chain network based on a multi-level consensus mechanism and design a fine-grained access control strategy; Deploy a condition-triggered charging load access smart contract in the alliance chain network to automatically trigger the access of charging load according to the conditions; Collect grid-side data and user-side data in real time through alliance chain network nodes, and pre-process the collected data; Use deep learning algorithms to analyze preprocessed data to build a charging load prediction model and dynamically predict short-term charging demand; Develop a load dispatch strategy based on multi-objective optimization to optimize the load distribution of charging stations; Smart contracts are used to automatically execute optimized charging load scheduling plans, and provide real-time monitoring and adjustment mechanisms to deal with emergencies and abnormal situations.

2. The decentralized charging load access power grid management method based on alliance chain according to claim 1, characterized in that: The construction of an electric vehicle charging alliance chain network based on a multi-level consensus mechanism includes the following steps: Design a layered alliance chain architecture, dividing the alliance chain network into three layers: core layer, operation layer, and edge layer; Based on CP-ABE technology, attribute keys are generated for three types of entities: power grid companies, operators, and users. The attribute keys include attribute private keys and entity public keys. When charging data needs to be shared and accessed, The charging pile node at the edge layer performs the first CP-ABE encryption on the local data and formulates a preliminary access strategy, and uploads the encrypted data and the corresponding access strategy to the operation layer of the alliance chain; The operation layer uses the public keys of each subject to perform a second CP-ABE encryption on the encrypted data and access policies from the edge layer to generate the final encrypted data and access policy ciphertext; The subject that meets the access policy uses its own attribute private key to decrypt the encrypted data and access policy ciphertext to obtain the data content and access rights; Based on the blockchain identity framework, the subject's attribute permission information is recorded and maintained, and a monitoring contract is deployed to automatically update the attribute key, revoke or restore the corresponding permissions by analyzing the subject's status changes; Develop corresponding cascading consensus paths for different types of transactions.

3. The decentralized charging load access power grid management method based on alliance chain as claimed in claim 2, characterized in that: The formulation of corresponding cascading consensus paths for different types of transactions includes the following steps: If it involves major decisions of the entire electric vehicle charging ecosystem, it is considered a key decision transaction. The cascading consensus path is as follows: the core layer authoritative node initiates a key transaction proposal, the operation layer node pre-executes and votes on the proposal, the edge layer node provides DPoS weight for the votes of the operation layer nodes to which it belongs, and the smart contract collects and summarizes the voting results of each layer, and decides whether to execute this transaction based on the preset high voting threshold; If it involves the daily operation and management of the alliance chain, it will be judged as a general operation transaction. The cascading consensus path is as follows: the operation layer node initiates a transaction proposal, the edge layer charging pile node runs the PoA consensus mechanism to review and vote on the proposal, and the smart contract decides whether to execute the transaction and upload it to the chain based on the preset medium voting threshold; If only quick confirmation is required at the edge layer and no upper layer is required to participate in the consensus, it is judged as a bottom-level data transaction. The cascade consensus path is as follows: the edge layer charging pile node initiates a transaction based on the local data status, and directly executes the transaction after reaching a consensus through the PoA consensus, and the result is uploaded to the chain.

4. The decentralized charging load access power grid management method based on alliance chain according to claim 1, characterized in that: The construction of the charging load prediction model comprises the following steps: Normalize the real-time collected grid-side data and user-side data so that data of different dimensions are uniformly mapped to the [0,1] interval; Use multiple restricted Boltzmann machine (RBM) layers to build a deep belief network (DBN), and use an unsupervised layer-by-layer greedy training strategy for RBM. Analyze the contribution of each feature to the prediction result and select important feature subsets, while generating more training data through data augmentation; Use the trained DBN to encode the normalized data into hidden feature vector sequences, and use these sequences as input to the Seq2Seq model; The DBN and Seq2Seq models are trained alternately using mean square error loss, and dynamic load forecasting is performed by rolling the prediction time window during the testing phase.

5. The decentralized charging load access power grid management method based on alliance chain according to claim 4, characterized in that: The unsupervised layer-by-layer greedy training strategy for RBM includes the following steps: Train the first RBM layer to learn the characteristics of the input data; The hidden layer output of the first RBM is used as the input of the second RBM, and this process is repeated until all layers are trained; Introducing random noise during training; Add a regression output layer on top of DBN to predict future charging demand; The calculation formula of the contribution is as follows: Among them, f i Represents the characteristics, I(f i ) represents the contribution, y j Indicates the actual load value. represents the average value of the load, f i (x j ) indicates that the model uses feature f i The predicted value at , N represents the total number of data points; The encoder of the Seq2Seq model adopts a bidirectional long short-term memory network, and the decoder adopts a unidirectional LSTM. The specific formula for predicting the sequence load element is as follows: h t =LSTM(h t-1 ,y t-1 ,c t ) Among them, h t represents the hidden layer state, y t represents the value of the output sequence at time step t, c t represents the cell state of the LSTM unit, W h and b represent training parameters.

6. The decentralized charging load access power grid management method based on alliance chain according to claim 1, characterized in that: The development of a load dispatch strategy based on multi-objective optimization includes: Using the DBN-Seq2Seq charging load prediction model, combined with the collected grid-side and user-side data and the real-time data obtained by the alliance chain, the charging load curve L(t) in the future T period is predicted; Obtain real-time data from the electric vehicle charging alliance chain network; Based on the charging load curve L(t) and real-time data, an improved multi-objective optimization model is established; Using MOEA / D-AWA, the load curve predicted by DBN is intelligently denoised to generate high-quality initial population; The consensus index C(x) is introduced as an additional evaluation mechanism, and the consensus convergence speed of the solution is optimized by an adaptive variant operator based on C(x); The optimized solution is decoded to obtain the load scheduling schemes at each level and the charging station scheduling quantity X(r); The core layer decomposes X(r) into the operation layer scheduling X(m), which is sent down through the alliance chain. The operation layer decomposes X(m) into the edge layer charging pile execution instruction X(e). The edge layer performs the actual distributed load scheduling operation according to X(e); Based on the alliance chain data, the entire dispatching process is monitored and traced, and the Internet of Things technology is combined to monitor the load response at all levels in real time, while collecting actual power grid operation data to evaluate and adjust the dispatching strategy.

7. The decentralized charging load access power grid management method based on alliance chain according to claim 6, characterized in that: The specific formula of the improved multi-objective optimization model is as follows: Among them, f1(x) and f2(x) represent the cost-effectiveness objective function and the energy efficiency objective function, c i represents the charging cost per unit time of the i-th charging station, p i (t) represents the power of the i-th charging station at time t, Δt represents the time interval, and p gent,t represents the generated power at time t, p load,t represents the load power at time t, t represents the time step, i represents the charging station, N represents the total number of charging stations, T represents the time period, and p i,min 、p i,max Represents the minimum power and maximum power respectively, V j (t) represents the voltage of node j at time point t, V min 、V max Indicates minimum and maximum voltage.

8. A decentralized charging load access to the power grid management system based on a consortium chain, based on the decentralized charging load access to the power grid management method based on a consortium chain according to any one of claims 1 to 7, characterized in that: Also includes, Access control module, which is used to build an electric vehicle charging alliance chain network based on a multi-level consensus mechanism and design fine-grained access control strategies; The conditional contract module is used to deploy condition-triggered charging load access smart contracts in the alliance chain network, automatically triggering the access of charging loads according to conditions; The data acquisition module is used to collect grid-side data and user-side data in real time through the alliance chain network nodes, and pre-process the collected data; A model building module is used to analyze the pre-processed data using a deep learning algorithm to build a charging load prediction model and dynamically predict short-term charging demand; Load dispatching module, which is used to develop load dispatching strategies based on multi-objective optimization and optimize the load distribution of charging stations under the premise of ensuring stable operation of the power grid; The execution adjustment module is used to automatically execute the optimized charging load scheduling plan using smart contracts, and provide real-time monitoring and adjustment mechanisms to deal with emergencies and abnormal situations.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the decentralized charging load access to the power grid management method based on the alliance chain are implemented in any one of claims 1 to 7.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the decentralized charging load access to the power grid management method based on the alliance chain according to any one of claims 1 to 7 are implemented.