Charging station operation maintenance method based on multi-dimensional data system

Abstract

The invention discloses a charging station operation maintenance method based on a multi-dimensional data system, and relates to the field of intelligent optimization, and the method comprises the steps: collecting the operation, environment, user and power grid data of a charging pile, preprocessing the data into JSON, carrying out the real-time analysis and classification through Kafka-Flume, storing the data into a distributed database, building a composite index, and carrying out the retrieval through Presto; after cleaning, combining GraphQL and a graph query language to construct a space-time atlas, and extracting a multi-dimensional feature vector; a digital twinborn model is generated, multi-source data are fused through virtual-real mapping, and residual correction and reinforcement learning optimization are combined; fusing multi-modal data by using a hypergraph neural network, and optimizing topology in combination with an entropy regularization multi-target strategy and meta-learning; power is dynamically distributed, LSTM-GAN predicts and positions faults, and PID regulates and controls output; and the block chain contract automatically triggers equipment freezing and power limiting, and on-chain data feedback optimizes scheduling and load balancing. According to the invention, through multi-dimensional data fusion and a dynamic optimization mechanism, the operation and maintenance accuracy and response efficiency of the charging station are significantly improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent optimization technology, and in particular to a charging station operation and maintenance method based on a multi-dimensional data system. Background Art

[0002] Current charging station operation and maintenance technologies mainly rely on IoT sensors to collect charging pile operation data, and perform offline analysis through batch processing frameworks such as Hadoop, such as statistical modeling and regularized early warning based on historical data. Although existing solutions can achieve basic status monitoring, the integration efficiency of multi-source heterogeneous data is low. Traditional ETL processes rely on static data table associations and lack real-time parsing and dynamic classification capabilities. Especially in cross-dimensional data fusion scenarios, spatiotemporal correlation analysis is limited by a single indexing mechanism, which makes it difficult to explore the coupling relationship between environmental factors and user behavior on power grid load, forming a data island effect.

[0003] The main drawback of existing technologies is the lack of real-time integration and dynamic association capabilities for multi-source heterogeneous data. Traditional architectures cannot efficiently process high-concurrency streaming data and lack composite index support, resulting in high data retrieval latency and coarse cross-dimensional analysis granularity. For example, when the grid load suddenly changes, it is impossible to associate user charging behavior characteristics with environmental parameters in real time, causing the power allocation strategy to lag behind and making it difficult to balance charging efficiency and grid stability. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a charging station operation and maintenance method based on a multi-dimensional data system to solve the problems of low efficiency in dynamic integration of multi-source heterogeneous data of charging stations and weak real-time control deviation and fault prediction generalization ability caused by insufficient adaptive optimization of digital twin models.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions: In a first aspect, the present invention provides a charging station operation and maintenance method based on a multi-dimensional data system, which includes collecting multi-dimensional data of the charging station, packaging it into a JSON format data packet and pre-processing it, wherein the multi-dimensional data includes charging pile operation data, environmental data, user data and power grid load data; The Kafka-Flume combined architecture is used to parse and classify JSON format data packets, store them in the database and create a composite index, and Presto queries are used to perform joint data retrieval and output the retrieval result set; Perform multi-dimensional standardization on the search result set, use graph query language to dynamically associate multi-dimensional data, build a spatiotemporal association graph, extract multi-dimensional data relationship features, and generate a multi-dimensional feature vector; Based on multi-dimensional feature vectors, computer vision is combined to generate a three-dimensional model of the charging station. The virtual-real mapping equation is used to dynamically fuse multi-dimensional data into the three-dimensional model to build a digital twin model. Calculate the residual error between the digital twin model and the real data of the charging station in real time and calibrate the digital twin model; Combined with reinforcement learning algorithm to optimize interaction efficiency, the digital twin model is dynamically adjusted through adaptive gain. Use hypergraph neural network to fuse multi-dimensional feature vectors, combine entropy regularized multi-objective MDP update strategy and use meta-learning framework to quickly adapt to different tasks and dynamically optimize the hypergraph topology structure; Power is allocated through a dynamic optimization algorithm, and LSTM-GAN fusion is used to predict and locate faults, providing feedback to optimize resource scheduling and load balancing across scenarios.

[0007] As a preferred solution of the charging station operation and maintenance method based on the multi-dimensional data system of the present invention, the packaging into a JSON format data packet and pre-processing includes: The Kalman filter algorithm is used to remove noise, the Z-Score method is used to detect anomalies, and the linear interpolation algorithm is used for completion.

[0008] As a preferred solution of the charging station operation and maintenance method based on the multi-dimensional data system of the present invention, wherein: the Kafka-Flume combined architecture performs parsing and classification, stores in the database and establishes a composite index, which means, Through Kafka, JSON format data packets are parsed in real time and dynamically classified into charging pile operation data, environmental data, user data, power grid load data and original multi-dimensional data of charging stations based on JSON fields; Flume is used to store charging pile operation data, user data and grid load data in a distributed relational database, and time series index, user identification index and geographic location index are established. Charging stations are partitioned by timestamp and device ID and stored in a distributed file system.

[0009] As a preferred solution of the charging station operation and maintenance method based on the multi-dimensional data system of the present invention, the method of dynamically associating multi-dimensional data using a graph query language, constructing a spatiotemporal association graph and extracting multi-dimensional data relationship features refers to: Based on the multi-dimensional data of charging stations, the predefined node types are charging pile nodes, user nodes, environment nodes and power grid nodes, and the edge relationship types are defined as user-charging pile charging relationship, charging pile-environment location relationship and charging pile-power grid impact relationship; The graph is updated in real time through the graph query language. When a new user charging event is added, the environmental data and the grid load are dynamically associated to generate a spatiotemporal association path, and a joint index of timestamp and geo-hash is created in the spatiotemporal association graph. A spatiotemporal clustering algorithm is used to extract user behavior regional hotspots. Through dynamic relationship weight calculation, the characteristics of charging piles and environment-grid coupling strength are extracted to generate a multi-dimensional feature vector.

[0010] As a preferred solution of the charging station operation and maintenance method based on the multi-dimensional data system of the present invention, the method of dynamically fusing multi-dimensional data into a three-dimensional model using a virtual-real mapping equation to construct a digital twin model refers to: The mean absolute error residual is calculated by comparing the real-time multi-dimensional data of the charging station with the predicted status of the three-dimensional model item by item; By using error residual feedback adjustment, the real-time multi-dimensional data of the charging station collected is compared item by item with the predicted status of the real-time multi-dimensional data of the charging station of the three-dimensional model, the heat dissipation coefficient of the charging pile, the power attenuation factor of the charging pile and the influence weight of user behavior are calculated, the average output error residual is used, the virtual-reality mapping equation is reversely optimized, and a digital twin model is constructed.

[0011] As a preferred solution of the charging station operation and maintenance method based on a multi-dimensional data system of the present invention, wherein: the use of a meta-learning framework to quickly adapt to different tasks means, Based on historical environmental data, the common features of multimodal data are extracted through the hypergraph neural network, the initial policy parameters are optimized, and cross-scenario universal policy parameters are generated; Use gradient back propagation to calculate task-specific losses, update initial strategy parameters and deploy them to the digital twin model to adjust charging pile power and user diversion guidance in real time.

[0012] As a preferred solution of the charging station operation and maintenance method based on the multi-dimensional data system of the present invention, the power allocation by the dynamic optimization algorithm includes: Based on the dynamic optimization hypergraph topology, the particle swarm optimization algorithm is adopted to input the real-time power demand of charging piles, grid load data and user behavior data, and output the optimized power allocation instructions for each charging pile.

[0013] As a preferred solution of the charging station operation and maintenance method based on a multi-dimensional data system of the present invention, the method of using LSTM-GAN fusion to predict and locate faults includes: Based on LSTM-GAN learning of equipment status timing rules, predicting future failure probability, generating adversarial network simulations of radiator overload and power module attenuation abnormal data, and enhancing training samples; By comparing the feature differences between the actual data and the abnormal data generated by GAN, the fault is located by combining the GPS coordinates of the charging pile, and the fault type, location coordinates and maintenance priority are output.

[0014] In a second aspect, 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 is executed by the processor, it implements any step of the charging station operation and maintenance method based on a multi-dimensional data system as described in the first aspect of the present invention.

[0015] In a third aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the charging station operation and maintenance method based on a multi-dimensional data system as described in the first aspect of the present invention.

[0016] The beneficial effects of the present invention are as follows: the present invention significantly improves the accuracy and response efficiency of charging station operation and maintenance through multi-dimensional data fusion and dynamic optimization mechanism. Through the virtual-real mapping equation and the residual error feedback mechanism, the parameters of the digital twin model are dynamically corrected, and the adaptive gain is optimized in combination with the reinforcement learning algorithm to reduce the prediction error of the digital twin model. Furthermore, the LSTM-GAN fusion model uses a generative adversarial network to expand abnormal samples, enhances the ability to identify rare faults (such as power module attenuation), and improves positioning accuracy. At the same time, the hypergraph neural network based on the meta-learning framework quickly adapts to different task scenarios, optimizes the power allocation strategy, and realizes the automatic execution of equipment freezing and power limiting through blockchain smart contracts, ensuring the stability of the power grid under load fluctuations, and achieving the global optimization of charging station resource scheduling, fault response and power grid coordination. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 This is a flow chart of multi-dimensional data collection and preprocessing in Example 1.

[0019] Figure 2 This is a flow chart of dynamic data integration and digital twin modeling in Example 1.

[0020] Figure 3 This is a flow chart of intelligent optimization and fault response in Example 1.

[0021] Figure 4 This is a flow chart of resource scheduling and execution feedback in Example 1. DETAILED DESCRIPTION

[0022] 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.

[0023] 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.

[0024] 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.

[0025] Example 1, reference Figure 1 to Figure 4 , which is the first embodiment of the present invention, and provides a charging station operation and maintenance method based on a multi-dimensional data system, comprising the following steps: S1. Collect multi-dimensional data of the charging station, package it into a JSON format data packet and pre-process it, wherein the multi-dimensional data includes charging pile operation data, environmental data, user data and grid load data; Specifically, during the operation and maintenance of the charging station, the charging pile operation data (voltage, current, power, equipment temperature), environmental data (temperature, humidity, air quality), user data and grid load data of the charging station are collected in real time, and multi-sensor time synchronization is achieved through the NTP protocol.

[0026] In this embodiment, the device operation data, environmental data, user data and grid load data are collected in real time by sensors, and the sampling frequency is 1 time / second.

[0027] S1.1. Preprocessing of multi-dimensional data means using an IoT gateway device to package the collected raw data into JSON format, using the Kalman filter algorithm to process the raw data, removing noise introduced by sensor errors or environmental factors, detecting outliers through the Z-Score method, and eliminating data that does not conform to the normal range of variation. If data is missing, a linear interpolation algorithm is used to supplement it, and the data packet is transmitted to the cloud platform through the 5G network.

[0028] S2, using the Kafka-Flume combined architecture to parse and classify JSON format data packets, store them in the database and create a composite index, use Presto query to perform data joint retrieval, and output the retrieval result set, It includes the following steps: S2.1. Parse the JSON format data packets transmitted to the cloud platform through Kafka (streaming data processing platform). After the parsing is completed, the parsed pre-processed charging station multi-dimensional data is dynamically classified into charging pile operation data, environmental data, user data, power grid load data and charging station original multi-dimensional data according to JSON fields.

[0029] S2.2. Use Flume (distributed data collection) to store charging pile operation data, user data and grid load data in a distributed relational database. During the storage process, a time series index, user identification index and geographic location index are established. The original multi-dimensional data of the charging station is stored in a distributed file system according to timestamp and device ID partitions.

[0030] Among them, the distributed relational database uses AES-256 encryption, the key is managed securely by HSM hardware, and RBAC is used to hierarchically control data access.

[0031] S2.3. Use the Presto query engine to implement efficient SQL query and streaming analysis, support fast retrieval of composite indexes, and output the retrieval results as result sets.

[0032] S3, perform multi-dimensional standardization on the result set, use graph query language to dynamically associate multi-dimensional data, build a spatiotemporal association graph and extract multi-dimensional data relationship features to generate a multi-dimensional feature vector, It includes the following steps: S3.1. Based on the search result set, the local anomaly factor algorithm is used to detect cross-dimensional anomalies. The time series data of charging pile power and grid load are supplemented with missing values ​​using the Lagrange interpolation method. The user data is based on the KNN algorithm to fill in the missing fields (charging time) according to the similar user data.

[0033] S3.2. Use the MinHash (similar hash) algorithm to calculate the similar signature of multi-dimensional data, detect the sameness between data, and combine the SimHash (minimum hash) algorithm to convert multi-dimensional data into hash values ​​and calculate the similarity of these hash values ​​to streamline data storage and remove redundant information. Use Z-score standardization to map unified dimensional data to standard normal distribution.

[0034] S3.3. Predefined node types are defined as charging pile nodes, user nodes, environment nodes, and power grid nodes, and edge relationship types are defined as user-charging pile charging relationship, charging pile-environment location relationship, and charging pile-power grid impact relationship. New events are dynamically associated through graph query language. When new user charging events are added, environmental data and power grid loads are dynamically associated. After generating the spatiotemporal association path, a spatiotemporal association graph is constructed, and a joint index of timestamp and geographic hash is created in the spatiotemporal association graph.

[0035] S3.4. Use the spatiotemporal clustering algorithm to extract user behavior regional hotspots, extract the charging pile and environment-grid coupling strength characteristics, and use the graph attention network to aggregate the node features and edge weights to generate a multi-dimensional feature vector.

[0036] Preferably, the shortcomings of single-dimensional processing can be solved through cross-dimensional correlation analysis, intelligent completion and multimodal standardization, providing a high-quality data foundation for global state analysis and dynamic optimization of charging stations. S4. Based on multi-dimensional feature vectors, a three-dimensional model of the charging station is generated in combination with computer vision, and the virtual-real mapping equation is used to dynamically fuse multi-dimensional data into the three-dimensional model to build a digital twin model, which includes the following steps: S4.1. Generate a three-dimensional model of the charging station through computer vision technology, and dynamically map the operating status of the charging pile, user data and environmental data to the three-dimensional model.

[0037] S4.2. Use virtual-real mapping equations to establish a mathematical mapping relationship between the physical world and the three-dimensional model, and synchronize the charging pile operation status, user data, grid load data and environmental data (serialize and transmit standardized data through the Protobuf protocol) to the three-dimensional model in real time.

[0038] Among them, the virtual-real mapping equation is expressed as: ; In the formula, For time 3D model charging station multi-dimensional data, For time, It is expressed as the weighted sum in the multi-dimensional data fusion process, is the number of multi-dimensional data features, is the weighting coefficient for each data, For time Mapping functions for different data, For time Real charging station operation status data, It means to dynamically adjust the topological difference between the multi-dimensional data of the three-dimensional model charging station and the multi-dimensional data of the real charging station at each moment. To calculate the spatiotemporal gradient function of the multi-dimensional data of the three-dimensional model charging station and the multi-dimensional data of the actual charging station, It is the actual real-time multi-dimensional data of charging stations. It is a three-dimensional model of real-time multi-dimensional data of charging stations.

[0039] in, The function definition is expressed as: ; In the formula, is the residual error adaptive adjustment, To constrain the spatiotemporal continuity of the digital twin model state, It is the L2 norm square error between the real charging station data and the 3D model charging station prediction.

[0040] in, The spatiotemporal smoothing term is expressed as: ; In the formula, is the data dimension index, For time of 3D model charging station status component, is the L2 norm square error of the same state component in adjacent time steps, For the previous time.

[0041] S4.3. Through computational vision modeling and dynamic data fusion with the virtual-real Silver Snake equation, and combined with reinforcement learning to optimize parameters, a high-precision digital twin model of synchronized physical entities is constructed.

[0042] S5. Calculate the residual error between the digital twin model and the real data of the charging station in real time, calibrate the digital twin model and optimize the interaction efficiency by combining the reinforcement learning algorithm, and dynamically adjust the digital twin model through adaptive gain, which includes the following steps: S5.1. Verify the data is identical by calculating the difference between the digital twin model data and the real charging station status data, and use error feedback to adjust the weights and topology preservation items in the mapping formula. When the difference exceeds the set synchronization error judgment threshold, the correction mechanism is automatically triggered.

[0043] The charging pile state data and environmental data output by the virtual-to-real mapping equation are input into the variational autoencoder for encoding, and anomaly detection is performed through the residual error calculated from the charging pile state data and environmental data.

[0044] Among them, the residual error calculation is expressed as: ; In the formula, For time The difference between the multi-dimensional data of the digital twin model and the multi-dimensional data of the real charging station.

[0045] S5.2. Based on the operation data of the charging station for 30 consecutive days, the residual error distribution of the digital twin model is statistically analyzed, the mean and standard deviation are calculated, and the synchronization error judgment threshold is set using the 3σ principle. .

[0046] Set the synchronization error judgment threshold to ; like > , it indicates that error correction is triggered and the digital twin model is corrected.

[0047] like ≤ When , it is represented as , no correction is required and the digital twin model maintains its state unchanged.

[0048] The output user data and real-time grid load data are optimized through reinforcement learning using a deep deterministic policy gradient algorithm. The efficiency of virtual-reality interaction is optimized by dynamically adjusting the constraint factors and weights of the digital twin model.

[0049] S5.3. Adaptive gain control is used to input multi-dimensional data of the charging station, and the gain of the digital twin model is dynamically adjusted through real-time data.

[0050] Among them, the adaptive gain control is expressed as: ; In the formula, is the gain adjustment factor, is the regulating factor, is the response function of different data, For multi-dimensional data input, is the minimum protection constant.

[0051] Preferably, the three-dimensional geometric information of the charging station can be accurately obtained through computer vision modeling, so that physical objects such as charging piles, power grid equipment, and distribution boxes can be mapped in the digital twin model according to their actual spatial positions, thereby providing an accurate spatial topological basis. On this basis, the topologically aware virtual-real mapping equation introduces a spatiotemporal gradient compensation mechanism, which can adjust the topological structure of the digital twin model in real time according to the changes in the operating status of the charging station, ensuring that the digital twin model is highly consistent with the actual charging station status. However, during the dynamic mapping process, the operating status and environmental data of the charging pile may be affected by external factors, resulting in the generation of outliers or noise data. The anomaly detection method based on variational autoencoder is adopted to accurately identify abnormal situations by calculating the residual error of virtual-real data, and trigger automatic correction when the error exceeds the set synchronization error judgment threshold, further improving the reliability of the data and the stability of the model. The adaptive gain control is used to optimize the response of the digital twin model, so that the digital twin model can be continuously optimized according to the different operating environments of the charging station, thereby improving the efficiency of virtual-real interaction and the overall intelligence level.

[0052] S6. Use hypergraph neural network to fuse multimodal data, combine entropy regularized multi-objective MDP update strategy and use meta-learning framework to quickly adapt to different tasks, dynamically optimize the hypergraph topology to maximize data association, which includes the following steps: S6.1, the neural network is expressed as: ; In the formula, The topology diagram shows the interaction between the devices in the charging station. For all the multi-dimensional data in the charging station, It is the multiple interactive relationships between charging pile status data, user data, grid load data, and environmental data. is the weight of all data relations.

[0053] The hypergraph Laplacian matrix is ​​used to perform regularization and calculate the fused eigenvector.

[0054] Among them, the hypergraph Laplacian matrix is ​​expressed as: ; In the formula, is the fused feature vector, is the regularized hypergraph data, is the feature dimension of multi-dimensional data, The bias term adjusts the linear output. is the Sigmoid function.

[0055] The fused feature vector is input into the multi-objective Markov decision process algorithm to optimize the charging station strategy and balance the relationship between different objectives through entropy regularization.

[0056] S6.2, the entropy regularized multi-objective MDP update strategy is expressed as: ; In the formula, The total return of the strategy, To define time The strategy for selecting charging stations under environmental data, To calculate the average return during multiple executions of the strategy, To measure the effectiveness of current strategic actions, is the discount factor, To express the influence of the entropy regularization term on the objective function, Expressed as At time, based on the multi-dimensional data of the digital twin model, the entropy of the probability distribution of possible actions selected by the charging station under the environmental data at that moment.

[0057] S6.3. Further utilize the digital twin model to quickly adapt to different tasks and scenarios of charging stations through meta-learning optimization strategy (the meta-learning strategy outputs hyperedge weight adjustment instructions to trigger dynamic update of the hypergraph adjacency matrix), and use topology graph optimization to calculate the optimal device interaction strategy.

[0058] The optimization strategy is expressed as: ; In the formula, are the optimized strategy parameters, is the initial strategy parameter, It is a collection of charging station operation and maintenance environments. For the task index, For the The environment corresponding to each task, is the probability distribution of the charging station operation and maintenance task environment, For the task The loss function on For the task Local constraints on is the initial strategy parameter The output of the following, is the meta-learning rate control parameter update step size, is the gradient operator.

[0059] Among them, the topology optimization is expressed as: ; In the formula, is the optimal topology graph, is a topological diagram, For different data sources in the charging station, For data and data The interaction weights, For data nodes With data nodes The connection between them.

[0060] S7, power allocation is performed through dynamic optimization algorithm, LSTM-GAN fusion is used to predict and locate faults, and feedback is provided to optimize resource scheduling and load balancing across scenarios, which includes the following steps: S7.1. Use the particle swarm optimization algorithm (PSO) with the core goal of compressing the peak-to-valley difference of the power grid, and balance the global search and local convergence capabilities through a dynamic weight adjustment strategy. The algorithm input includes the real-time power demand of the charging pile, the power grid load and the user behavior data, and the output is the optimized power allocation instruction for each charging pile. The constraint conditions set the safe range of the charging pile power and the user satisfaction.

[0061] S7.2. Real-time collection of multi-dimensional data of charging stations (sampling frequency 1 time / second), constructing a data sequence with a time window of 1 hour, using a two-layer LSTM network (128 hidden units), the input layer receives standardized time series data, and the output layer calculates the probability of failure in the next hour through the Sigmoid function. The cross entropy loss function and Adam optimizer are used during training, and the accuracy is measured. The generator receives a random noise vector (dimension 100) and outputs simulated abnormal data); the discriminator is based on Wasserstein distance optimization (gradient penalty coefficient λ=10) to enhance the authenticity of the generated data, and the cosine similarity compares the feature vectors of the actual data and the generated abnormal data. Combined with the GPS coordinates of the charging pile, the fault location (such as the radiator of the charging pile CZ-001) is output.

[0062] S7.3. Based on reinforcement learning (DDPG algorithm), the proportional coefficient, integral coefficient and differential coefficient are dynamically adjusted. By setting the initial value, the reward function is the negative inverse of the power deviation. The target power (from the dynamic optimization algorithm) and the actual power are input, the deviation is calculated, and the abnormal state is output.

[0063] S7.4. When in an abnormal state, a shutdown command is triggered and the abnormal event is recorded. The charging function of the faulty equipment is disabled. A maintenance work order is generated and pushed to the nearest operation and maintenance personnel. The execution record is stored through the Hyperledger Fabric consortium chain, using the PBFT consensus algorithm (4 Orderer nodes), and tampering risk (Monte Carlo attack simulation verification).

[0064] S7.5. Based on the historical data stored in blockchain (10 charging station scenarios), train a general strategy (MAML algorithm), learn common features across scenarios (such as the morning peak load pattern of urban stations), and use the Dijkstra algorithm to calculate the optimal scheduling path for mobile charging piles. The weights are the geographical distance, the road congestion coefficient, and the scheduling response time.

[0065] This embodiment also provides a computer device, which is suitable for the case of a charging station operation and maintenance method based on a multi-dimensional data system, 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 charging station operation and maintenance method based on a multi-dimensional data system as proposed in the above embodiment.

[0066] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected through a system bus. Among them, the processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system 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 may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covered on the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0067] The present embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the charging station operation and maintenance method based on a multi-dimensional data system proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0068] In summary, the present invention significantly improves the accuracy and response efficiency of charging station operation and maintenance through: multi-dimensional data fusion and dynamic optimization mechanism. Through the virtual-real mapping equation and the residual error feedback mechanism, the parameters of the digital twin model are dynamically corrected, and the adaptive gain is optimized in combination with the reinforcement learning algorithm to reduce the prediction error of the digital twin model. Furthermore, the LSTM-GAN fusion model uses a generative adversarial network to expand abnormal samples, enhances the ability to identify rare faults (such as power module attenuation), and improves positioning accuracy. At the same time, the hypergraph neural network based on the meta-learning framework quickly adapts to different task scenarios, optimizes the power allocation strategy, and realizes the automatic execution of equipment freezing and power limiting through blockchain smart contracts, ensuring the stability of the power grid under load fluctuations, and achieving the global optimization of charging station resource scheduling, fault response and power grid coordination.

[0069] 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 charging station operation and maintenance method based on a multi-dimensional data system, characterized in that: include, Collect multi-dimensional data of the charging station, package it into JSON format data packets and pre-process it, where the multi-dimensional data includes charging pile operation data, environmental data, user data and grid load data; The Kafka-Flume combined architecture is used to parse and classify JSON format data packets, store them in the database and create a composite index, and Presto queries are used to perform joint data retrieval and output the retrieval result set; Perform multi-dimensional standardization on the search result set, use graph query language to dynamically associate multi-dimensional data, build a spatiotemporal association graph, extract multi-dimensional data relationship features, and generate a multi-dimensional feature vector; Based on multi-dimensional feature vectors, computer vision is combined to generate a three-dimensional model of the charging station. The virtual-real mapping equation is used to dynamically fuse multi-dimensional data into the three-dimensional model to build a digital twin model. Calculate the residual error between the digital twin model and the real data of the charging station in real time and calibrate the digital twin model; Combined with reinforcement learning algorithm to optimize interaction efficiency, the digital twin model is dynamically adjusted through adaptive gain. Use hypergraph neural network to fuse multi-dimensional feature vectors, combine entropy regularized multi-objective MDP update strategy and use meta-learning framework to quickly adapt to different tasks and dynamically optimize the hypergraph topology structure; Power is allocated through a dynamic optimization algorithm, and LSTM-GAN fusion is used to predict and locate faults, providing feedback to optimize resource scheduling and load balancing across scenarios.

2. The charging station operation and maintenance method based on a multi-dimensional data system according to claim 1, characterized in that: The packaging into JSON format data packets and preprocessing include: The Kalman filter algorithm is used to remove noise, the Z-Score method is used to detect anomalies, and the linear interpolation algorithm is used for completion.

3. The charging station operation and maintenance method based on a multi-dimensional data system according to claim 1, characterized in that: The Kafka-Flume combined architecture performs parsing and classification, stores data in the database, and creates a composite index, which means that: Through Kafka, JSON format data packets are parsed in real time and dynamically classified into charging pile operation data, environmental data, user data, power grid load data and original multi-dimensional data of charging stations based on JSON fields; Flume is used to store charging pile operation data, user data and grid load data in a distributed relational database, and time series index, user identification index and geographic location index are established. Charging stations are partitioned by timestamp and device ID and stored in a distributed file system.

4. The charging station operation and maintenance method based on a multi-dimensional data system according to claim 1, characterized in that: The use of graph query language to dynamically associate multi-dimensional data, construct a spatiotemporal association graph and extract multi-dimensional data relationship features refers to: Based on the multi-dimensional data of charging stations, the predefined node types are charging pile nodes, user nodes, environment nodes and power grid nodes; the edge relationship types are defined as user-charging pile charging relationship, charging pile-environment location relationship and charging pile-power grid impact relationship; The graph is updated in real time through the graph query language. When a new user charging event is added, the environmental data and the grid load are dynamically associated to generate a spatiotemporal association path, and a joint index of timestamp and geo-hash is created in the spatiotemporal association graph. A spatiotemporal clustering algorithm is used to extract user behavior regional hotspots. Through dynamic relationship weight calculation, the characteristics of charging piles and environment-grid coupling strength are extracted to generate a multi-dimensional feature vector.

5. The charging station operation and maintenance method based on a multi-dimensional data system according to claim 1, characterized in that: The method of dynamically fusing multi-dimensional data into a three-dimensional model using a virtual-real mapping equation to construct a digital twin model means: The mean absolute error residual is calculated by comparing the real-time multi-dimensional data of the charging station with the predicted status of the three-dimensional model item by item; By using error residual feedback adjustment, the real-time multi-dimensional data of the charging station collected is compared item by item with the predicted status of the real-time multi-dimensional data of the charging station of the three-dimensional model, the heat dissipation coefficient of the charging pile, the power attenuation factor of the charging pile and the influence weight of user behavior are calculated, the average output error residual is used, the virtual-reality mapping equation is reversely optimized, and a digital twin model is constructed.

6. The charging station operation and maintenance method based on a multi-dimensional data system according to claim 1, characterized in that: The use of a meta-learning framework to quickly adapt to different tasks means that Based on historical environmental data, the common features of multimodal data are extracted through the hypergraph neural network, the initial policy parameters are optimized, and cross-scenario universal policy parameters are generated; Use gradient back propagation to calculate task-specific losses, update initial strategy parameters and deploy them to the digital twin model to adjust charging pile power and user diversion guidance in real time.

7. The charging station operation and maintenance method based on a multi-dimensional data system according to claim 1, characterized in that: The power allocation by dynamic optimization algorithm includes: Based on the dynamic optimization hypergraph topology, the particle swarm optimization algorithm is adopted to input the real-time power demand of charging piles, grid load data and user behavior data, and output the optimized power allocation instructions for each charging pile.

8. The charging station operation and maintenance method based on a multi-dimensional data system according to claim 1, characterized in that: The use of LSTM-GAN fusion to predict faults and locate them includes: Based on LSTM-GAN learning of equipment status timing rules, predicting future failure probability, generating adversarial network simulations of radiator overload and power module attenuation abnormal data, and enhancing training samples; By comparing the characteristic differences between the actual charging station data and the abnormal data, the fault is located by combining the GPS coordinates of the charging pile, and the fault type, location coordinates and maintenance priority are output.

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 charging station operation and maintenance method based on a multi-dimensional data system as described in any one of claims 1 to 8 are implemented.

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 charging station operation and maintenance method based on a multi-dimensional data system described in any one of claims 1 to 8 are implemented.

Images