Charging pile safety monitoring and fault diagnosis method and system based on artificial intelligence
By building the working correlation model and fault prediction model of smart charging piles, technical faults and data security problems in smart charging pile monitoring and fault diagnosis are solved, accurate prediction and diagnosis of smart charging pile faults are achieved, and monitoring and Real-time and accuracy of troubleshooting.
Patent Information
- Application Number
- CN202411056070.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-02
- Publication Date
- 2024-11-01
- Estimated Expiration
- 2044-08-02
AI Technical Summary
During the monitoring and fault diagnosis process, smart charging piles may experience technical failures, inaccurate data or unauthorized access, resulting in inaccurate fault diagnosis and insufficient real-time performance, affecting the safety monitoring effect.
Using an artificial intelligence-based method, by obtaining the temperature, voltage and current data of the intelligent charging pile, a working correlation model is constructed, a fault prediction model is trained, the status of the next moment is predicted, and the fault probability is calculated. If the threshold is exceeded, fault diagnosis is performed.
It realizes accurate prediction and diagnosis of the faults of smart charging piles, prompt alarm and repair, and improves the safety monitoring and fault handling efficiency of charging piles.
Smart Images

Figure CN118585755B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of artificial intelligence technology, and in particular to an artificial intelligence-based charging pile safety monitoring and fault diagnosis method and system. Background Art
[0002] Smart charging piles are a kind of high-tech charging equipment that has emerged with the development of the new energy vehicle industry. They have the following features: through Internet technology, smart charging piles can remotely monitor the charging status, realize fault diagnosis and remote maintenance; smart charging piles are equipped with touch screens or buttons, and users can perform charging operations through these interfaces, such as selecting charging modes and paying fees; smart charging piles can collect charging data, including charging time, power, user usage habits, etc., to provide data support for operators to optimize services; when unsafe conditions such as overload, short circuit, leakage, etc. occur, they can automatically start protection functions to ensure the safety of the charging process; they can intelligently dispatch charging power and time according to grid load and user needs to achieve orderly charging; they can also achieve information security to prevent user data from being leaked or unauthorized access.
[0003] Although smart charging piles can remotely monitor the charging status in real time, some problems will still be encountered in actual operation. For example: the monitoring system may have technical failures, such as software defects, hardware damage or network connection problems, resulting in inaccurate monitoring data or monitoring interruption; real-time monitoring involves the transmission and storage of large amounts of data, which may face the risk of data leakage or unauthorized access; the monitoring system needs to be able to identify abnormal situations in a timely manner and issue alarms. If the alarm system responds slowly, it may not be able to handle the problem in time; severe weather or other environmental factors may affect the operation of monitoring equipment, resulting in inaccurate monitoring data or equipment damage; real-time monitoring systems are time-sensitive, and any delays may affect the monitoring effect, so it is necessary to ensure that the system has a high degree of real-time performance.
[0004] The above problems directly or indirectly lead to deviations in the monitoring results, which will cause the fault diagnosis to be inaccurate. Therefore, is it possible to combine the prediction with the monitoring of the safety of the charging pile and the diagnosis of the fault? In this way, the probability of failure in the next period can be predicted, and then corresponding measures can be taken in time in the current period. The alarm or maintenance system will be able to know the problem and the staff can deal with the problem in time. Summary of the invention
[0005] In view of the shortcomings of the prior art, the present invention provides a charging pile safety monitoring and fault diagnosis method and system based on artificial intelligence.
[0006] In order to solve the above technical problems, the present invention is solved by the following technical solutions: a charging pile safety monitoring and fault diagnosis method based on artificial intelligence, comprising the following steps: based on a preset time series, obtaining relevant data of each time series of the smart charging pile to form a historical data set, the relevant data at least including temperature data, voltage data and current data; preprocessing the historical data set to obtain a processed data set, constructing a working association model of the temperature data, voltage data and current data, and dividing it into a normal state working association mode and an abnormal state working association mode according to the working state; identifying the state working association mode based on the processed data set and calculating the corresponding associated error data based on the working association model to obtain an associated error data set; constructing a fault prediction pre-training model, training and verifying the fault prediction pre-training model through the associated error data set to obtain a fault prediction model, and predicting the associated error data of the state working association mode of the smart charging pile at the next moment based on the fault prediction model to obtain a prediction result; calculating the probability of failure of the prediction result based on a preset fault probability model, if it exceeds a preset probability threshold, analyzing the relevant data of the current time series of the smart charging pile to obtain a fault diagnosis result.
[0007] As an implementable embodiment, the construction of the working association mode of temperature data, voltage data and current data includes the following process: the temperature data is the temperature data before operation and the temperature data after operation; the voltage data is the input voltage data and the output voltage data; the current data is the input current data and the output current data; based on the input voltage data and the input current data, the input power data is obtained; based on the output voltage data and the output current data, the output power data is obtained; based on the temperature data before operation and the temperature data after operation, the temperature change data is obtained; based on the input power data and the output power data, the power loss data is obtained; according to the temperature change data, the power loss data and the time, a working association model is constructed, and based on the working association model, the association error data is obtained; the working association model is expressed as follows: in, represents the associated error data, Indicates temperature change data, Indicates Time series of power loss data, Indicates time series, represents the temperature influence factor, Indicates the range of temperature influence, , Indicates The input power data of a time series, Indicates Time series output power data.
[0008] As an implementable method, the preprocessing includes one or more of data cleaning processing, removing or filling missing data, identifying and processing abnormal data or outlier data; or time alignment processing; or data type conversion processing; or time series decomposition processing; or data resampling processing; or sliding window feature processing; or time feature engineering processing; or standardization / normalization processing; or difference processing; or data denoising processing; or data segmentation processing; or data rolling processing.
[0009] As an implementable method, the construction of the fault prediction pre-training model includes the following steps: constructing an initial fault prediction model based on multiple decision trees; obtaining the importance weight of each decision tree by associating the sum of errors of each data in the fault prediction model in the error data set; the importance weight is the importance coefficient of the decision tree corresponding to the prediction accuracy evaluation, and the importance coefficient is expressed as follows: Where N represents the number of decision trees. represents a decision tree, y represents a decision tree The sum of the prediction errors of all sample predictions, X represents the importance coefficient of the corresponding decision tree; the initial fault prediction model is updated according to the importance weight of each decision tree until the preset number of iterations is reached or the performance of the initial fault prediction model converges, thereby obtaining a fault prediction pre-training model.
[0010] As an implementable method, the fault prediction model is expressed as follows: in, represents the fault prediction model, N represents the number of decision trees, Indicates The prediction results of a decision tree, Represents the error between the predicted result and the actual data. Represents the importance coefficient.
[0011] As an implementable method, the preset fault probability model is expressed as follows: in, Represents the prediction results The probability of failure occurring, Represents the prediction results The adjustment factor, represents the mean, Represents standard deviation.
[0012] As an implementable method, the relevant data of the current time series of the smart charging pile is analyzed to obtain a fault diagnosis result, including the following steps: at least the input voltage data and output voltage data, input current data and output current data, and pre-operation temperature data and post-operation temperature data associated with the current time series are analyzed separately; if there is a problem with any one of the input voltage data and output voltage data, input current data and output current data, and pre-operation temperature data and post-operation temperature data, the corresponding fault diagnosis result is matched.
[0013] A charging pile safety monitoring and fault diagnosis system based on artificial intelligence includes a data acquisition module, a processing construction module, an identification calculation module, a construction prediction module and a judgment analysis module; the data acquisition module is used to obtain relevant data of each time series of the intelligent charging pile based on a preset time series to form a historical data set, and the relevant data at least includes temperature data, voltage data and current data; the processing construction module is used to pre-process the historical data set to obtain a processed data set, construct a working association model of temperature data, voltage data and current data, and divide it into a normal state working association mode and an abnormal state working association mode according to the working state; the identification calculation module is used to Based on the processed data set, the state-work association mode is identified and the corresponding association error data are calculated based on the work association model to obtain the association error data set; the prediction construction module is used to construct a fault prediction pre-training model, train and verify the fault prediction pre-training model through the association error data set to obtain a fault prediction model, and predict the association error data of the state-work association mode of the smart charging pile at the next moment based on the fault prediction model to obtain a prediction result; the judgment and analysis module calculates the probability of failure of the prediction result based on a preset fault probability model. If it exceeds a preset probability threshold, the relevant data of the current time series of the smart charging pile is analyzed to obtain a fault diagnosis result.
[0014] A computer-readable storage medium stores a computer program, which implements the following method when executed by a processor: based on a preset time series, obtain relevant data of each time series of a smart charging pile to form a historical data set, wherein the relevant data at least includes temperature data, voltage data and current data; preprocess the historical data set to obtain a processed data set, construct a working association model of the temperature data, voltage data and current data, and divide the data into a normal state working association mode and an abnormal state working association mode according to the working state; identify the state working association mode based on the processed data set and calculate corresponding associated error data based on the working association model to obtain an associated error data set; construct a fault prediction pre-training model, train and verify the fault prediction pre-training model through the associated error data set to obtain a fault prediction model, and predict the associated error data of the state working association mode of the smart charging pile at the next moment based on the fault prediction model to obtain a prediction result; calculate the probability of failure of the prediction result based on a preset fault probability model, and if it exceeds a preset probability threshold, analyze the relevant data of the current time series of the smart charging pile to obtain a fault diagnosis result.
[0015] A charging pile safety monitoring and fault diagnosis device based on artificial intelligence comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein the processor implements the following method when executing the computer program: based on a preset time series, obtaining relevant data of each time series of a smart charging pile to form a historical data set, wherein the relevant data at least comprises temperature data, voltage data and current data; preprocessing the historical data set to obtain a processed data set, constructing a working association model of the temperature data, voltage data and current data, and dividing the data into a normal state working association mode and an abnormal state working association mode according to the working state; identifying the state working association mode based on the processed data set and calculating corresponding associated error data based on the working association model to obtain an associated error data set; constructing a fault prediction pre-training model, training and verifying the fault prediction pre-training model through the associated error data set to obtain a fault prediction model, predicting the associated error data of the state working association mode of the smart charging pile at the next moment based on the fault prediction model to obtain a prediction result; calculating the probability of failure of the prediction result based on a preset fault probability model, and if the probability exceeds a preset probability threshold, analyzing the relevant data of the current time series of the smart charging pile to obtain a fault diagnosis result.
[0016] The present invention has significant technical effects due to the adoption of the above technical solutions: through the method and system of the present invention, a fault prediction model is designed, which can predict the probability of failure in the next time period, and then analyze various data in this time period, so as to prevent problems before they occur. The alarm or maintenance system can know the problem and the staff can deal with the problem in time. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. 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 paying creative labor.
[0018] Figure 1 It is a schematic diagram of the overall process of the method of the present invention;
[0019] Figure 2 It is a schematic diagram of the overall structure of the system of the present invention;
[0020] Figure 3 A schematic diagram of a process flow corresponding to an embodiment of the present invention. DETAILED DESCRIPTION
[0021] The present invention is further described in detail below in conjunction with embodiments. The following embodiments are for explanation of the present invention but the present invention is not limited to the following embodiments.
[0022] Embodiment 1:
[0023] A charging pile safety monitoring and fault diagnosis method based on artificial intelligence, such as Figure 1 As shown, the following steps are included:
[0024] S100, based on a preset time series, obtaining relevant data of each time series of the smart charging pile to form a historical data set, where the relevant data at least includes temperature data, voltage data, and current data;
[0025] S200, preprocessing the historical data set to obtain a processed data set, constructing a working association model of temperature data, voltage data, and current data, and dividing the data into a normal state working association mode and an abnormal state working association mode according to the working state;
[0026] S300, identifying the state work association mode based on the processed data set and calculating the corresponding association error data based on the work association model to obtain the association error data set;
[0027] S400, constructing a fault prediction pre-training model, training and verifying the fault prediction pre-training model through an associated error data set to obtain a fault prediction model, and predicting the associated error data of the state working associated mode of the smart charging pile at the next moment based on the fault prediction model to obtain a prediction result;
[0028] S500: Calculate the probability of a failure of the prediction result based on a preset failure probability model. If the probability exceeds a preset probability threshold, analyze the relevant data of the current time series of the smart charging pile to obtain a fault diagnosis result.
[0029] In step S200, the working association mode of the temperature data, the voltage data and the current data is constructed, such as Figure 3 As shown, the following process is included:
[0030] S210, the temperature data is the temperature data before operation and the temperature data after operation; the voltage data is the input voltage data and the output voltage data; the current data is the input current data and the output current data;
[0031] S220, obtaining input power data based on the input voltage data and the input current data, obtaining output power data based on the output voltage data and the output current data, and obtaining temperature change data based on the pre-operation temperature data and the post-operation temperature data;
[0032] S230, obtaining power loss data based on the input power data and the output power data;
[0033] S240, constructing a working correlation model according to the temperature change data, the power loss data and the time, and obtaining correlation error data based on the working correlation model; the working correlation model is expressed as follows: in, represents the associated error data, Indicates temperature change data, Indicates Time series of power loss data, Indicates time series, represents the temperature influence factor, Indicates the range of temperature influence, , Indicates The input power data of a time series, Indicates Time series output power data.
[0034] Here, based on the input voltage data and the input current data, the input power data is obtained, and the expression of the input power data can be expressed as , Indicates The input voltage data of a time series, Indicates Based on the input current data of a time series, the output power data is obtained based on the output voltage data and output current data. The expression of the output power data can be , Indicates Time series of output voltage data, Indicates The output current data of a time series, and the input voltage data, input current data, output voltage data and output current data can all be obtained in real time, and do not need to be obtained through relevant calculations. The real-time acquisition method can be achieved through existing technical means. Here, the specific problem to be solved in this embodiment is to perform subsequent related applications based on these data obtained in real time.
[0035] In one embodiment, the preprocessing includes one or more of data cleaning, removing or filling missing data, identifying and processing abnormal data or outlier data; or time alignment; or data type conversion; or time series decomposition; or data resampling; or sliding window feature processing; or time feature engineering; or standardization / normalization; or difference processing; or data denoising; or data segmentation; or data rolling processing. The preprocessing here can select a specific processing method of the preprocessing according to the specific situation of the data set, and finally obtain the processed data set.
[0036] Decision trees are commonly used for classification and regression tasks. Through a series of questions, data is divided into smaller and smaller subsets until the stopping condition is met, and finally a prediction result is given at each leaf node. A decision tree consists of nodes (including internal nodes and leaf nodes) and edges. Internal nodes represent attribute tests, and leaf nodes give prediction results. At each internal node, the decision tree selects a feature and a threshold to split the data. Feature selection is based on impurity measurement. The data is divided into different subsets according to the feature value, and this process is recursively performed until the stopping condition is met. The stopping condition can be that the preset maximum depth is reached, the number of samples in the node is less than a certain threshold, or further segmentation does not significantly reduce the impurity. In classification problems, leaf nodes usually give a category label; in regression problems, leaf nodes give a numerical value. Therefore, this embodiment is a fault prediction pre-training model constructed using decision trees, including the following steps: constructing an initial fault prediction model based on multiple decision trees; obtaining the importance weight of each decision tree by associating the sum of the errors of each data in the fault prediction model in the error data set; the importance weight is the importance coefficient of the decision tree corresponding to the prediction accuracy evaluation, and the importance coefficient is expressed as follows: Where N represents the number of decision trees. represents a decision tree, y represents a decision tree The sum of the prediction errors of all sample predictions, X represents the importance coefficient of the corresponding decision tree; the initial fault prediction model is updated according to the importance weight of each decision tree until the preset number of iterations is reached or the performance of the initial fault prediction model converges, thereby obtaining a fault prediction pre-training model.
[0037] In this embodiment, the fault prediction model is expressed as follows: in, represents the fault prediction model, N represents the number of decision trees, Indicates The prediction results of a decision tree, Represents the error between the predicted result and the actual data. Represents the importance coefficient.
[0038] In one embodiment of the present invention, the preset fault probability model is expressed as follows: in, Represents the prediction results The probability of failure occurring, Represents the prediction results The adjustment factor, represents the mean, Indicates standard deviation. According to the calculation, the prediction result satisfies the normal distribution within a certain range, so the possibility of failure is judged by calculating the failure probability. If it exceeds the set threshold, it means that the subsequent specific failure analysis process, that is, the subsequent operation steps, need to be performed.
[0039] In one embodiment, the analysis of the relevant data of the current time series of the smart charging pile to obtain the fault diagnosis result includes the following steps: at least the input voltage data and output voltage data, input current data and output current data, pre-operation temperature data and post-operation temperature data associated with the current time series are analyzed separately; if any one of the input voltage data and output voltage data, input current data and output current data, pre-operation temperature data and post-operation temperature data has a problem, the corresponding fault diagnosis result is matched.
[0040] In the present embodiment, it is formulated for the situation where the prediction result of the next time series does not meet the requirements. If there is a problem with the prediction result of the next time series, it means that there may be a potential problem with the relevant data of the current time series. In this case, the relevant data of the current time series will be analyzed. Of course, it is not limited to the input voltage data and the output voltage data, the input current data and the output current data, the pre-operation temperature data and the post-operation temperature data. Other relevant data may even be analyzed, such as protection technical parameters, etc. If there is a problem with these data, specific fault diagnosis results will be matched or analyzed. For example, a simple fault diagnosis table may be set in advance. If the fault is very simple, it will be directly matched to the corresponding result. If the problem is too complicated, detailed analysis and calculation will be performed based on the relevant data to finally obtain specific fault diagnosis results.
[0041] Embodiment 2:
[0042] An artificial intelligence-based charging pile safety monitoring and fault diagnosis system, such as Figure 2 As shown, it includes a data acquisition module 100, a processing and construction module 200, an identification and calculation module 300, a prediction module 400 and a judgment and analysis module 500; the data acquisition module 100 is used to obtain relevant data of each time series of the smart charging pile based on a preset time series to form a historical data set, and the relevant data at least includes temperature data, voltage data and current data; the processing and construction module 200 is used to pre-process the historical data set to obtain a processed data set, construct a working association model of temperature data, voltage data and current data, and divide it into a normal state working association mode and an abnormal state working association mode according to the working state; the identification and calculation module 300 is used to The processed data set identifies the state-work association mode and calculates the corresponding association error data based on the work association model to obtain the association error data set; the construction prediction module 400 is used to construct a fault prediction pre-training model, train and verify the fault prediction pre-training model through the association error data set to obtain a fault prediction model, and predict the association error data of the state-work association mode of the smart charging pile at the next moment based on the fault prediction model to obtain a prediction result; the judgment and analysis module 500 calculates the probability of failure of the prediction result based on a preset fault probability model. If it exceeds the preset probability threshold, the relevant data of the current time series of the smart charging pile is analyzed to obtain a fault diagnosis result.
[0043] The various embodiments in this specification are described in a progressive manner, and each embodiment focuses on the differences from other embodiments. The same or similar parts between the various embodiments can be referenced to each other.
[0044] It will be appreciated by those skilled in the art that embodiments of the present invention may be provided as methods, apparatus, or computer program products. Therefore, the present invention may take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the present invention may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.
[0045] The present invention is described with reference to the flowcharts and / or block diagrams of the method, terminal device (system), and computer program product according to the present invention. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of the processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing terminal device to generate a machine, so that the instructions executed by the processor of the computer or other programmable data processing terminal device generate instructions for implementing the process in the flowchart. Figure 1 A process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.
[0046] These computer program instructions may also be stored in a computer readable memory capable of directing a computer or other programmable data processing terminal device to operate in a specific manner, so that the instructions stored in the computer readable memory produce a manufactured product including an instruction device, which implements the process Figure 1 A process or multiple processes and / or boxes Figure 1 A function specified in one or more boxes.
[0047] These computer program instructions can also be loaded onto a computer or other programmable data processing terminal device so that a series of operating steps are executed on the computer or other programmable terminal device to produce a computer-implemented process, thereby providing instructions for executing on the computer or other programmable terminal device to implement the process. Figure 1 A process or multiple processes and / or boxes Figure 1 The steps for the functions specified in one or more boxes.
[0048] It should be noted that the "one embodiment" or "embodiment" mentioned in the specification means that the specific features, structures or characteristics described in conjunction with the embodiment are included in at least one embodiment of the present invention. Therefore, the phrases "one embodiment" or "embodiment" appearing in various places throughout the specification do not necessarily refer to the same embodiment.
[0049] In addition, it should be noted that the shapes and names of the parts and components of the specific embodiments described in this specification may be different. Any equivalent or simple changes made based on the structure, features and principles described in the patent concept of the present invention are included in the protection scope of the patent of the present invention. The technicians in the technical field of the present invention can make various modifications or supplements to the specific embodiments described or replace them in a similar manner, as long as they do not deviate from the structure of the present invention or exceed the scope defined by the claims, they should all fall within the protection scope of the present invention.
Claims
1. A charging pile safety monitoring and fault diagnosis method based on artificial intelligence, characterized in that: The method comprises the following steps: based on a preset time series, obtaining relevant data of each time series of the smart charging pile to form a historical data set, wherein the relevant data at least includes temperature data, voltage data and current data; Preprocessing the historical data set to obtain a processed data set, constructing a working association model of temperature data, voltage data, and current data, and dividing the data into a normal state working association mode and an abnormal state working association mode according to the working state; identifying the state working association mode based on the processed data set and calculating corresponding association error data based on the working association model to obtain an association error data set; Construct a fault prediction pre-training model, train and verify the fault prediction pre-training model through the associated error data set to obtain a fault prediction model, and predict the associated error data of the state working associated mode of the smart charging pile at the next moment based on the fault prediction model to obtain a prediction result; The probability of failure of the predicted result is calculated based on the preset fault probability model. If it exceeds the preset probability threshold, the relevant data of the current time series of the smart charging pile is analyzed to obtain the fault diagnosis result; wherein, the construction of the working association mode of temperature data, voltage data and current data includes the following processes: the temperature data is the temperature data before work and the temperature data after work; the voltage data is the input voltage data and the output voltage data; the current data is the input current data and the output current data; based on the input voltage data and the input current data, the input power data is obtained, based on the output voltage data and the output current data, the output power data is obtained, and based on the temperature data before work and the temperature data after work, the temperature change data is obtained; based on the input power data and the output power data, the power loss data is obtained; according to the temperature change data, the power loss data and the time, a working association model is constructed, and based on the working association model, the associated error data is obtained; the working association model is expressed as follows: in, represents the associated error data, Indicates temperature change data, Indicates Time series of power loss data, Indicates time series, represents the temperature influence factor, Indicates the range of temperature influence, , Indicates The input power data of a time series, Indicates Time series output power data.
2. The charging pile safety monitoring and fault diagnosis method based on artificial intelligence according to claim 1 is characterized in that: The preprocessing includes one or more of data cleaning, removing or filling missing data, identifying and processing abnormal data or outlier data; or time alignment; or data type conversion; or time series decomposition; or data resampling; or sliding window feature processing; or time feature engineering; or standardization / normalization; or difference processing; or data denoising; or data segmentation; or data rolling processing.
3. The charging pile safety monitoring and fault diagnosis method based on artificial intelligence according to claim 1 is characterized in that: The construction of the fault prediction pre-training model includes the following steps: constructing an initial fault prediction model based on multiple decision trees; obtaining the importance weight of each decision tree by associating the sum of errors of each data in the fault prediction model in the error data set; the importance weight is the importance coefficient of the decision tree corresponding to the prediction accuracy evaluation, and the importance coefficient is expressed as follows: Where N represents the number of decision trees. represents a decision tree, y represents a decision tree The sum of the prediction errors of all sample predictions, X represents the importance coefficient of the corresponding decision tree; the initial fault prediction model is updated according to the importance weight of each decision tree until the preset number of iterations is reached or the performance of the initial fault prediction model converges, thereby obtaining a fault prediction pre-training model.
4. The charging pile safety monitoring and fault diagnosis method based on artificial intelligence according to claim 1 is characterized in that: The fault prediction model is expressed as follows: in, represents the fault prediction model, N represents the number of decision trees, Indicates The prediction results of a decision tree, Represents the error between the predicted result and the actual data. Represents the importance coefficient.
5. The charging pile safety monitoring and fault diagnosis method based on artificial intelligence according to claim 1 is characterized in that: The preset fault probability model is expressed as follows: in, Represents the prediction results The probability of failure occurring, Represents the prediction results The adjustment factor, represents the mean, Represents standard deviation.
6. The charging pile safety monitoring and fault diagnosis method based on artificial intelligence according to claim 1 is characterized in that: The analysis of the relevant data of the current time series of the smart charging pile to obtain the fault diagnosis result includes the following steps: at least the input voltage data and output voltage data, input current data and output current data, pre-operation temperature data and post-operation temperature data associated with the current time series are analyzed respectively; if any one of the input voltage data and output voltage data, input current data and output current data, pre-operation temperature data and post-operation temperature data has a problem, the corresponding fault diagnosis result is matched.
7. An artificial intelligence-based charging pile safety monitoring and fault diagnosis system, characterized in that: It includes a data acquisition module, a processing and construction module, an identification and calculation module, a construction and prediction module, and a judgment and analysis module; the data acquisition module is used to acquire relevant data of each time series of the smart charging pile based on a preset time series to form a historical data set, and the relevant data at least includes temperature data, voltage data, and current data; The processing and construction module is used to preprocess the historical data set to obtain a processed data set, construct a working association model of temperature data, voltage data and current data, and divide it into a normal state working association mode and an abnormal state working association mode according to the working state; the identification and calculation module is used to identify the state working association mode based on the processed data set and calculate the corresponding associated error data based on the working association model to obtain an associated error data set; the construction prediction module is used to construct a fault prediction pre-training model, train and verify the fault prediction pre-training model through the associated error data set to obtain a fault prediction model, and predict the associated error data of the state working association mode of the smart charging pile at the next moment based on the fault prediction model to obtain a prediction result; The judgment and analysis module calculates the probability of failure of the prediction result based on a preset fault probability model. If the probability exceeds a preset probability threshold, the relevant data of the current time series of the smart charging pile is analyzed to obtain a fault diagnosis result; Among them, the construction of the working association mode of temperature data, voltage data and current data includes the following process: the temperature data is the temperature data before work and the temperature data after work; the voltage data is the input voltage data and the output voltage data; the current data is the input current data and the output current data; based on the input voltage data and the input current data, the input power data is obtained, based on the output voltage data and the output current data, the output power data is obtained, based on the temperature data before work and the temperature data after work, the temperature change data is obtained; based on the input power data and the output power data, the power loss data is obtained; according to the temperature change data, the power loss data and the time, the working association model is constructed, and the association error data is obtained based on the working association model; the working association model is expressed as follows: in, represents the associated error data, Indicates temperature change data, Indicates Time series of power loss data, Indicates time series, represents the temperature influence factor, Indicates the range of temperature influence, , Indicates The input power data of a time series, Indicates Time series output power data.
8. A computer-readable storage medium storing a computer program, characterized in that: When the computer program is executed by a processor, the method according to any one of claims 1 to 6 is implemented.
9. An artificial intelligence-based charging pile safety monitoring and fault diagnosis device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the method according to any one of claims 1 to 6 is implemented.
