New energy charging pile fault prediction and maintenance system and method based on internet of things

Abstract

The application relates to the technical field of new energy charging pile fault prediction and maintenance, in particular to a new energy charging pile fault prediction and maintenance system and method based on Internet of Things, which comprises data collection and generation of charging pile multidimensional information, processing of the charging pile multidimensional information, obtaining of charging pile operation data based on Internet of Things, comparative analysis of the charging pile operation data, determination of a charging pile fault prediction method based on Internet of Things, and implementation of alarm decision based on the charging pile fault prediction method. The application discards high-frequency hard data collection, is based on relevant soft data collection, reduces the cost and difficulty of data acquisition required for charging pile fault prediction, can process and analyze soft data when the charging pile is in standby state and working state, realizes early prediction and alarm of charging pile faults, improves the freedom of self-fault prediction of the charging pile, and is favorable for early elimination of hidden dangers of the charging pile.

Description

Technical Field

[0001] This invention relates to the field of fault prediction and maintenance technology for new energy charging piles, specifically to a fault prediction and maintenance system and method for new energy charging piles based on the Internet of Things. Background Technology

[0002] With the country's strong support for the development of electric vehicles, the construction of new energy charging piles that provide power for electric vehicles has also developed rapidly. Today, many regions have built charging stations of varying sizes. At the same time, the frequency of use of charging piles in charging stations is also increasing, which gradually increases the possibility of charging pile failures and damages. Therefore, the safety management of charging piles has become particularly important.

[0003] Chinese patent CN118246750B, authorized by patent number CN118246750B, discloses an AI-based charging pile anomaly diagnosis and early warning system. This system includes a charging pile operation and management platform, an information acquisition unit, a status monitoring unit, a charging safety assessment unit, a self-test charging analysis unit, a fusion feedback unit, a fault prediction and assessment unit, and a management execution unit. In this patent document, the early warning system assesses the safety and performance of charging piles based on real-time data, diagnoses the charging piles based on the assessment results, and then issues relevant early warnings. While it is real-time, the data required for early warning is high-frequency hard data such as real-time current, real-time voltage, and real-time temperature. The cost of collecting and analyzing this hard data on each charging pile is extremely high, placing very high demands on the system's data processing capabilities, easily leading to redundancy and significant limitations. Furthermore, the provided technical support lacks the flexibility to predict faults in both standby and operating states of the charging pile, resulting in delayed detection of charging pile problems and preventing technicians from proactively identifying potential safety hazards, thus rendering the system ineffective. Summary of the Invention

[0004] The purpose of this invention is to address the problems existing in the background technology by proposing a fault prediction and maintenance system and method for new energy charging piles based on the Internet of Things.

[0005] The technical solution of this invention: a method for fault prediction and maintenance of new energy charging piles based on the Internet of Things, comprising the following steps:

[0006] S1. Collect information on the vehicle model, actual charging amount, output power of the charging pile, number of charging pile services, number of charging pile maintenance, and current health status of the charging pile to determine multi-dimensional information of the charging pile based on the Internet of Things.

[0007] S2. Process the multi-dimensional information of the charging pile to obtain charging pile operation data based on the Internet of Things;

[0008] S3. Compare and analyze the charging pile operation data to determine the charging pile fault prediction method based on the Internet of Things;

[0009] S4. Implement alarm decision-making based on the charging pile fault prediction method, and notify technical personnel to maintain the charging pile based on the implementation of the alarm decision.

[0010] Preferably, for S2, the method for processing multi-dimensional information of charging piles includes the following steps:

[0011] Store multi-dimensional information about charging piles and establish a multi-dimensional information database for charging piles.

[0012] Obtain the multi-dimensional information of the k1 most recent charging piles from the dimensional information database and establish the first sample set;

[0013] Obtain the multi-dimensional information of the k2 nearest charging piles under the unique identifier of the charging pile in the dimensional information database, and establish a second sample set. The first and second sample sets are related sample sets.

[0014] Determine whether the charging pile is in operation; if the charging pile is in operation, establish a feature vector based on the current multi-dimensional information of the charging pile, construct sample vectors corresponding to the multi-dimensional information of several charging piles in the first sample set, and construct a sample vector set based on several sample vectors.

[0015] Charging pile operation data is generated based on the first sample set, the second sample set, the feature vector, and the sample vector set.

[0016] If the charging pile is in standby mode, charging pile operation data is generated based on the first sample set and the second sample set.

[0017] Preferably, the method for comparing and analyzing the charging pile operation data for S3 is as follows:

[0018] S301. Perform anomaly identification on the first sample set to obtain anomaly identification results, and calculate the first risk parameter of the associated sample set based on the identification results;

[0019] S302. Apply conditional constraints to the associated sample set;

[0020] S303. Perform dimensional correlation analysis on the associated sample set to obtain the analysis results, and calculate the second risk parameter of the associated sample set based on the analysis results.

[0021] S304. Based on whether the charging pile operation data contains feature vectors and sample vector sets, calculate the average correlation between feature vectors and sample vector sets.

[0022] S305. The second risk parameter is weighted based on the average correlation degree. The first risk parameter, the weighted second risk parameter, and the average correlation degree are input into the fault risk prediction model, and the fault risk value is output.

[0023] Preferably, for S301, the method for identifying anomalies in the first sample set, obtaining anomaly identification results, and calculating the first risk parameter of the associated sample set based on the identification results is as follows:

[0024] The health status of charging piles in the multi-dimensional information of charging piles in the sample set is marked as a key search term. The multi-dimensional information of charging piles with faults as key search terms is marked as abnormal information. The occurrence frequency of abnormal information is calculated and used as the abnormal occurrence frequency of the sample set.

[0025] The frequency of anomalies in the first sample set is used as the risk frequency threshold. The ratio of the number of multi-dimensional information of charging piles in the second sample set to the risk frequency threshold is used to calculate the first risk parameter α1 of the associated sample set.

[0026] Preferably, for S302, conditional constraints are applied to the associated sample set, and the conditional constraints satisfy:

[0027] Constraint 1: The vehicle models contained in k1 pieces of first information must include the vehicle models contained in k2 pieces of second information;

[0028] Constraint 2: The frequency of anomalies in the first sample set is not 0;

[0029] Constraint 3: The anomaly frequency of the second sample set is 0;

[0030] Constraint 4: The number of multi-dimensional information items k1 of the charging piles in the first sample set and the number of multi-dimensional information items k2 of the charging piles in the second sample set must satisfy: k1>>k2>n, where n is the dimension value of the multi-dimensional information of the charging piles.

[0031] Preferably, for S303, the method for performing dimensional correlation analysis on the associated sample set to obtain the analysis results, and calculating the second risk parameter of the associated sample set based on the analysis results, includes the following steps:

[0032] Through formula The power loss analysis value L of the associated sample set is calculated;

[0033] In the formula, ELi is the charging coefficient corresponding to the vehicle models included in the first sample set, and ELj is the charging coefficient corresponding to the vehicle models included in the second sample set; the charging coefficient is obtained based on big data on vehicle charging; OEi is the output energy of the charging piles included in the first sample set, and OEj is the output energy of the charging piles included in the second sample set; IEi is the actual charging amount of the vehicles included in the first sample set, and IEj is the actual charging amount of the vehicles included in the second sample set; i is the multi-dimensional information number of the charging pile in the first sample set, i∈[1, k1], where k1 is a positive integer; j is the multi-dimensional information number of the charging pile in the second sample set, i∈[1, k2], where k2 is a positive integer; lc is the basic loss constant.

[0034] The total number of charging services for charging piles included in the multi-dimensional prediction information of the first and second sample sets are counted respectively and labeled as T1 and T2.

[0035] The number of maintenance times corresponding to charging piles included in the multi-dimensional prediction information in the first and second sample sets are counted respectively, and labeled as W1 and W2 respectively;

[0036] Through formula The maintenance frequency analysis value P of the associated sample set was calculated;

[0037] The second risk parameter α2 is calculated using the formula α2=β1×L+β2×P; where β1 and β2 are the risk coefficients of the power loss analysis value L and the maintenance frequency analysis value P, respectively.

[0038] Preferably, for S304, the method for calculating the average correlation between the feature vector and the sample vector set is as follows:

[0039] If the charging pile operation data contains feature vectors and sample vector sets, the cosine similarity algorithm is used to calculate the cosine similarity between the feature vector and several sample vectors in the sample vector set, and this cosine similarity is used as the correlation between the feature vector and the sample vector. The correlation between the feature vector and several sample vectors is summed and the mean is calculated. The result is used as the average correlation between the feature vector and the sample vector set.

[0040] If the charging pile operation data does not contain feature vectors and sample vector sets, no calculation will be performed.

[0041] Preferably, for S305, the expression for the fault risk prediction model is:

[0042] ;

[0043] In the formula, Ω is the average correlation between the feature vector and the sample vector set, X is the state variable, and F is the fault risk value;

[0044] The specific method for implementing alarm decision-making based on charging pile fault prediction is as follows:

[0045] Data analysis is performed on the fault risk value F. If F is less than 0, no alarm is triggered; if F is not less than 0, an alarm is triggered, and technicians are notified to perform maintenance, and the health status of the charging pile is set to fault.

[0046] This invention also discloses an Internet of Things (IoT)-based fault prediction and maintenance system for new energy charging piles, which applies the aforementioned IoT-based fault prediction and maintenance method for new energy charging piles, specifically including:

[0047] The charging operation and management module is used for monitoring and managing charging piles;

[0048] The data acquisition module is used to collect information on the vehicle model connected to the charging pile, the actual charging amount of the vehicle, the output power of the charging pile, the number of times the charging pile is serviced, the number of times the charging pile is maintained, and the current health status of the charging pile, so as to determine the multi-dimensional information of the charging pile based on the Internet of Things; the current health status of the charging pile is divided into good and faulty.

[0049] The data processing module is used to process multi-dimensional information of charging piles to obtain charging pile operation data based on the Internet of Things.

[0050] The data analysis module is used to compare and analyze the operation data of charging piles to determine the charging pile fault prediction method based on the Internet of Things.

[0051] The maintenance alarm module is used to implement alarm decisions based on the charging pile fault prediction method. Based on the implementation of the alarm decisions, the charging operation and management module notifies the technicians to maintain the charging pile.

[0052] Preferably, the charging operation and management module includes an identification unit and a monitoring unit;

[0053] The identification unit is used to add a unique identifier to each charging pile and bind the multi-dimensional data of the charging pile collected based on the unique identifier of each charging pile.

[0054] The monitoring unit is used to monitor the status of charging piles in real time and identify whether the charging pile is in operation or in standby mode.

[0055] Compared with the prior art, the above-mentioned technical solution of the present invention has the following beneficial technical effects:

[0056] This invention eliminates the need for high-frequency hard data collection such as real-time current, real-time voltage, connection confirmation voltage, and real-time temperature of the charging head for each charging pile. Instead, it focuses on collecting soft data related to charging pile services through a data acquisition module. This significantly reduces the cost and difficulty of acquiring data required for charging pile fault prediction. Based on big data analysis, it calculates the energy loss analysis value of the associated sample set for the actual charging amount of the charging pile to the vehicle and the output power of the charging pile. As the energy loss analysis value increases, it can promptly detect the increase in energy loss caused by the charging pile in the second sample set experiencing a fault leading to increased heat generation, high connection voltage of charging pile components, or high charging energy loss due to aging or damage of charging pile components. Furthermore, based on big data analysis, it calculates the maintenance frequency analysis value of the associated sample set for the number of service and maintenance times of the charging pile. This reveals that for the charging pile in the second sample set, the potential risks increase after a certain number of charging service times.

[0057] This invention also has the advantage of analyzing soft data when the charging pile is in standby or working state. By outputting a fault risk value through a charging pile fault prediction method, it enables early prediction and alarm of charging pile faults, improves the freedom of self-fault prediction of the charging pile, helps to eliminate potential problems of the charging pile in advance, and reduces the possibility of charging pile faults. Attached Figure Description

[0058] Figure 1 This is a flowchart of the method according to Embodiment 1 of the present invention;

[0059] Figure 2 This is a flowchart of the S3 method in Embodiment 1 of the present invention;

[0060] Figure 3 This is a flowchart of the modules in Embodiment 2 of the present invention. Detailed Implementation

[0061] Example 1, as Figure 1 As shown, the IoT-based fault prediction and maintenance method for new energy charging piles proposed in this invention includes the following steps:

[0062] S1. Collect information on the vehicle model, actual charging amount, output power of the charging pile, number of charging pile services, number of charging pile maintenance, and current health status of the charging pile to determine multi-dimensional information of the charging pile based on the Internet of Things.

[0063] S2. Process the multi-dimensional information of the charging pile to obtain charging pile operation data based on the Internet of Things;

[0064] The S2 method includes the following steps:

[0065] Store multi-dimensional information about charging piles and establish a multi-dimensional information database for charging piles.

[0066] Obtain the multi-dimensional information of the k1 most recent charging piles from the dimensional information database and establish the first sample set;

[0067] Obtain the multi-dimensional information of the k2 nearest charging piles under the unique identifier of the charging pile in the dimensional information database, and establish a second sample set. The first and second sample sets are related sample sets.

[0068] Determine whether the charging pile is in operation; if the charging pile is in operation, establish a feature vector based on the current multi-dimensional information of the charging pile, construct sample vectors corresponding to the multi-dimensional information of several charging piles in the first sample set, and construct a sample vector set based on several sample vectors.

[0069] Charging pile operation data is generated based on the first sample set, the second sample set, the feature vector, and the sample vector set.

[0070] If the charging pile is in standby mode, charging pile operation data is generated based on the first sample set and the second sample set.

[0071] S3. Compare and analyze the charging pile operation data to determine the charging pile fault prediction method based on the Internet of Things;

[0072] like Figure 2 As shown, the S3 method includes the following steps:

[0073] S301. Perform anomaly identification on the first sample set to obtain anomaly identification results, and calculate the first risk parameter of the associated sample set based on the identification results;

[0074] In S301, the specific implementation method is as follows: anomaly information identification is performed on the first sample set to obtain anomaly information identification results; the method for calculating the first risk parameter of the associated sample set based on the identification results is as follows:

[0075] The health status of charging piles in the multi-dimensional information of charging piles in the sample set is marked as a key search term. The multi-dimensional information of charging piles with faults as key search terms is marked as abnormal information. The occurrence frequency of abnormal information is calculated and used as the abnormal occurrence frequency of the sample set.

[0076] The frequency of anomalies in the first sample set is used as the risk frequency threshold. The ratio of the number of multi-dimensional information of charging piles in the second sample set to the risk frequency threshold is used to calculate the first risk parameter α1 of the associated sample set.

[0077] S302. Apply conditional constraints to the associated sample set;

[0078] In S302, the specific implementation method is to apply conditional constraints to the associated sample set, and the conditional constraints satisfy:

[0079] Constraint 1: The vehicle models contained in k1 pieces of first information must include the vehicle models contained in k2 pieces of second information;

[0080] Constraint 2: The frequency of anomalies in the first sample set is not 0;

[0081] Constraint 3: The anomaly frequency of the second sample set is 0;

[0082] Constraint 4: The number of multi-dimensional information items k1 of charging piles in the first sample set and the number of multi-dimensional information items k2 of charging piles in the second sample set must satisfy: k1>>k2>n, where n is the dimension value of the multi-dimensional information of the charging piles; it should be noted that the symbol ">>" means much greater than;

[0083] S303. Perform dimensional correlation analysis on the associated sample set to obtain the analysis results, and calculate the second risk parameter of the associated sample set based on the analysis results.

[0084] In S303, the specific implementation method is as follows: performing dimensional correlation analysis on the associated sample set to obtain the analysis results, and calculating the second risk parameter of the associated sample set based on the analysis results includes the following steps:

[0085] Through formula The power loss analysis value L of the associated sample set is calculated;

[0086] In the formula, ELi is the charging coefficient corresponding to the vehicle models included in the first sample set, and ELj is the charging coefficient corresponding to the vehicle models included in the second sample set; the charging coefficient is obtained based on big data on vehicle charging; OEi is the output energy of the charging piles included in the first sample set, and OEj is the output energy of the charging piles included in the second sample set; IEi is the actual charging amount of the vehicles included in the first sample set, and IEj is the actual charging amount of the vehicles included in the second sample set; i is the multi-dimensional information number of the charging pile in the first sample set, i∈[1, k1], where k1 is a positive integer; j is the multi-dimensional information number of the charging pile in the second sample set, i∈[1, k2], where k2 is a positive integer; lc is the basic loss constant.

[0087] The total number of charging services for charging piles included in the multi-dimensional prediction information of the first and second sample sets are counted respectively and labeled as T1 and T2.

[0088] The number of maintenance times corresponding to charging piles included in the multi-dimensional prediction information in the first and second sample sets are counted respectively, and labeled as W1 and W2 respectively;

[0089] Through formula The maintenance frequency analysis value P of the associated sample set was calculated;

[0090] It should be noted that the default number of maintenance cycles for charging piles is 1, meaning that maintenance is counted once when the pile is manufactured or put into operation.

[0091] The second risk parameter α2 is calculated using the formula α2=β1×L+β2×P; where β1 and β2 are the risk coefficients of the power loss analysis value L and the maintenance frequency analysis value P, respectively.

[0092] In the formula, the risk coefficients β1 and β2 are calculated as follows:

[0093] Based on the calculated multi-dimensional information of each charging pile in the first sample set, the actual charging amount of the vehicle and the output power of the charging pile are used to calculate the charging power loss ratio. Specifically, the charging power loss is obtained by calculating the difference between the output power of the charging pile and the actual charging amount of the vehicle, and then the ratio of the charging power loss to the output power of the charging pile is calculated. The calculated information is used as the charging power loss ratio of the multi-dimensional information of the charging pile. The charging power loss ratios of several charging piles in the first sample set are summed and averaged to obtain the average charging power loss ratio of the first sample set.

[0094] Accordingly, the average percentage of charging power loss for abnormal information in the first sample set is calculated and used as the average percentage of abnormal power loss.

[0095] The first coefficient parameter is obtained by calculating the ratio of the average percentage of abnormal power loss to the average percentage of power loss in the sample set.

[0096] The ratio of the total number of charging services T1 of the charging piles included in the multi-dimensional prediction information of the first sample set to the number of maintenance W1 corresponding to the T1 charging piles is calculated to obtain the maintenance frequency of the sample set.

[0097] Accordingly, the charging piles included in the abnormal information in the first sample set are marked as target charging piles. The total number of service times of the target charging piles is counted to obtain the total number of abnormal services. The total number of maintenance times of the target charging piles is counted to obtain the total number of abnormal maintenance times. The ratio of the total number of abnormal services to the total number of abnormal maintenance times is calculated to obtain the abnormal maintenance frequency.

[0098] The second coefficient parameter is obtained by calculating the ratio of the frequency of sample set maintenance to the frequency of abnormal maintenance.

[0099] The risk coefficients β1 and β2 are constrained to β1+β2=1, and the risk coefficients β1 and β2 are obtained based on the weight relationship between the first coefficient parameter and the second coefficient parameter.

[0100] Specifically, for example, if the first coefficient parameter is 0.9 and the second coefficient parameter is 0.6, then the risk coefficients β1 and β2 are 0.6 and 0.4, respectively;

[0101] The advantages of this invention are that it eliminates the need for high-frequency hard data collection such as real-time current, real-time voltage, connection confirmation voltage, and real-time temperature of the charging head for each charging pile. Instead, it collects soft data related to charging pile services based on the data acquisition module, greatly reducing the cost and difficulty of data acquisition required for charging pile fault prediction. Based on big data, it calculates the energy loss analysis value of the associated sample set for the actual charging amount of the charging pile to the vehicle and the output power of the charging pile. Based on the increase of the energy loss analysis value, it can promptly detect the increase in energy loss caused by the charging pile corresponding to the second sample set experiencing faults leading to increased heat generation, high connection voltage of charging pile components, or high charging energy loss due to aging or damage of charging pile components. Furthermore, based on big data, it calculates the maintenance frequency analysis value of the associated sample set for the number of service and maintenance times of the charging pile, which shows that for the charging pile corresponding to the second sample set, the potential risks increase after a certain number of charging service times.

[0102] S304. Based on whether the charging pile operation data contains feature vectors and sample vector sets, calculate the average correlation between feature vectors and sample vector sets.

[0103] In S304, the specific implementation method is as follows: if the charging pile operation data includes feature vectors and sample vector sets, the cosine similarity algorithm is used to calculate the cosine similarity between the feature vector and several sample vectors in the sample vector set, and this cosine similarity is used as the correlation between the feature vector and the sample vector. The correlation between the feature vector and several sample vectors is summed and the mean is calculated. The result is used as the average correlation between the feature vector and the sample vector set.

[0104] If the charging pile operation data does not contain feature vectors and sample vector sets, no calculation will be performed; the cosine similarity algorithm is existing technology and will not be elaborated on here.

[0105] S305. The second risk parameter is weighted based on the average correlation degree. The first risk parameter, the weighted second risk parameter, and the average correlation degree are input into the fault risk prediction model, and the fault risk value is output.

[0106] The specific implementation method in S305 is as follows:

[0107] The expression for the failure risk prediction model is:

[0108] ;

[0109] In the formula, Ω is the average correlation between the feature vector and the sample vector set, X is the state variable, and F is the fault risk value;

[0110] It should be noted that when the charging pile is in standby mode, feature vectors cannot be constructed, so the average correlation degree Ω does not exist at this time, and the state variable X takes the value of 1.

[0111] S4. Implement alarm decision-making based on the charging pile fault prediction method, and notify technical personnel to maintain the charging pile based on the implementation of the alarm decision;

[0112] For S4, the specific implementation plan is as follows: perform data analysis on the fault risk value F. If F is less than 0, no alarm will be issued; if F is not less than 0, an alarm will be issued, and technicians will be notified to perform maintenance and the health status of the charging pile will be set to fault.

[0113] This invention also has the advantage of analyzing soft data when the charging pile is in standby or working state. By outputting a fault risk value through a charging pile fault prediction method, it enables early prediction and alarm of charging pile faults, improves the freedom of self-fault prediction of the charging pile, helps to eliminate potential problems of the charging pile in advance, and reduces the possibility of charging pile faults.

[0114] Example 2, as Figure 3 As shown, the IoT-based fault prediction and maintenance system for new energy charging piles proposed in this invention is applied to the IoT-based fault prediction and maintenance method for new energy charging piles proposed in Embodiment 1, including:

[0115] The charging operation and management module is used for monitoring and managing charging piles;

[0116] The data acquisition module is used to collect information on the vehicle model connected to the charging pile, the actual charging amount of the vehicle, the output power of the charging pile, the number of times the charging pile is serviced, the number of times the charging pile is maintained, and the current health status of the charging pile, so as to determine the multi-dimensional information of the charging pile based on the Internet of Things; the current health status of the charging pile is divided into good and faulty.

[0117] The data processing module is used to process multi-dimensional information of charging piles to obtain charging pile operation data based on the Internet of Things.

[0118] The data analysis module is used to compare and analyze the operation data of charging piles to determine the charging pile fault prediction method based on the Internet of Things.

[0119] The maintenance alarm module is used to implement alarm decisions based on the charging pile fault prediction method. Based on the implementation of the alarm decisions, the charging operation and management module notifies the technicians to maintain the charging pile.

[0120] The charging operation and management module includes an identification unit and a monitoring unit;

[0121] The identification unit is used to add a unique identifier to each charging pile and bind the multi-dimensional data of the charging pile collected based on the unique identifier of each charging pile.

[0122] The monitoring unit is used to monitor the status of charging piles in real time and identify whether the charging pile is in operation or in standby mode.

[0123] The embodiments of the present invention have been described in detail above with reference to the accompanying drawings. However, the present invention is not limited thereto. Various changes can be made within the scope of knowledge possessed by those skilled in the art without departing from the spirit of the present invention.

Claims

1. A new energy charging pile fault prediction and maintenance method based on the Internet of Things, characterized in that, The method comprises the following steps: S1, collecting the vehicle model, the actual charging capacity of the vehicle, the output power of the charging pile, the service frequency of the charging pile, the maintenance frequency of the charging pile and the current health status of the charging pile, and determining the multi-dimensional information of the charging pile based on the Internet of Things; S2, processing the multi-dimensional information of the charging pile to obtain the operation data of the charging pile based on the Internet of Things; S3, comparing and analyzing the operation data of the charging pile to determine the fault prediction method of the charging pile based on the Internet of Things; S4, implementing the alarm decision based on the fault prediction method of the charging pile, and notifying the technical personnel to maintain the charging pile based on the implementation of the alarm decision; For S2, the method for processing the multi-dimensional information of the charging pile comprises the following steps: storing the multi-dimensional information of the charging pile, and establishing a dimension information library of the multi-dimensional information of the charging pile; obtaining the latest k1 multi-dimensional information of the charging pile in the dimension information library, and establishing a first sample set; obtaining the latest k2 multi-dimensional information of the charging pile under the unique identifier of the charging pile in the dimension information library, and establishing a second sample set, the first sample set and the second sample set are associated sample sets; determining whether the charging pile is in operation; if the charging pile is in operation, a feature vector is established based on the current multi-dimensional information of the charging pile, a sample vector corresponding to the multi-dimensional information of the charging pile in the first sample set is constructed, and a sample vector set is constructed based on the sample vectors; the operation data of the charging pile is generated based on the first sample set, the second sample set, the feature vector and the sample vector set; if the charging pile is in standby, the operation data of the charging pile is generated based on the first sample set and the second sample set; For S3, the method for comparing and analyzing the operation data of the charging pile is as follows: S301, identifying abnormal information of the first sample set to obtain an abnormal information identification result, and calculating a first risk parameter of the associated sample set based on the identification result; S302, conditionally restricting the associated sample set; S303, performing dimension association analysis on the associated sample set to obtain an analysis result, and calculating a second risk parameter of the associated sample set based on the analysis result; S304, calculating the average correlation degree of the feature vector and the sample vector set based on whether the operation data of the charging pile contains the feature vector and the sample vector set; S305, weighting the second risk parameter based on the average correlation degree, inputting the first risk parameter and the weighted second risk parameter and the average correlation degree into a fault risk prediction model, and outputting a fault risk value; For S301, the method for identifying abnormal information of the first sample set to obtain an abnormal information identification result, and calculating a first risk parameter of the associated sample set based on the identification result is as follows: marking the health status of the charging pile in the multi-dimensional information of the charging pile as a key search word, marking the multi-dimensional information of the charging pile with fault as abnormal information, calculating the occurrence frequency of the abnormal information, and taking the abnormal occurrence frequency of the sample set as the risk frequency threshold; calculating the first risk parameter a1 of the associated sample set by ratio calculation of the number of the multi-dimensional information of the charging pile in the second sample set and the risk frequency threshold. For S302, the condition constraints are applied to the associated sample set, and the condition constraints meet: constraint condition one: the vehicle models contained in the k1 first information need to include the vehicle models contained in the k2 second information; constraint condition two: the abnormal occurrence frequency of the first sample set is not 0; constraint condition three: the abnormal occurrence frequency of the second sample set is 0; constraint condition four: the number of charging pile multi-dimensional information contained in the first sample set k1 and the number of charging pile multi-dimensional information contained in the second sample set k2 need to meet: k1 >> k2 > n, n is the dimension value of the charging pile multi-dimensional information; For S303, the dimension association analysis is performed on the associated sample set to obtain the analysis result, and the method for calculating the second risk parameter of the associated sample set based on the analysis result includes the following steps: The power loss analysis value L of the associated sample set is calculated by the formula ; wherein ELi is the charging coefficient corresponding to the vehicle model contained in the first sample set, ELj is the charging coefficient corresponding to the vehicle model contained in the second sample set; the charging coefficient is obtained based on vehicle charging big data; OEi is the output power of the charging pile contained in the first sample set, OEj is the output power of the charging pile contained in the second sample set; IEi is the actual charging amount of the vehicle contained in the first sample set, IEj is the actual charging amount of the vehicle contained in the second sample set; i is the multi-dimensional information number of the charging pile in the first sample set, i∈[1, k1], k1 is a positive integer; j is the multi-dimensional information number of the charging pile in the second sample set, i∈[1, k2], k2 is a positive integer; lc is a basic loss constant; the total number of charging services of the charging piles contained in the multi-dimensional prediction information in the first sample set and the second sample set is respectively counted and marked as T1 and T2; the corresponding maintenance number of the charging piles contained in the multi-dimensional prediction information in the first sample set and the second sample set is respectively counted and marked as W1 and W2; The maintenance frequency analysis value P of the correlation sample set is calculated by the formula The second risk parameter a2 is calculated by the formula a2=β1×L+β2×P; wherein β1 and β2 are risk coefficients of the electric energy loss analysis value L and the maintenance frequency analysis value P, respectively. For S304, the method for calculating the average correlation degree of the feature vector and the sample vector set is: if the charging pile operation data contains the feature vector and the sample vector set, the cosine similarity algorithm is used to calculate the cosine similarity of the feature vector and a plurality of sample vectors in the sample vector set, and the cosine similarity is taken as the correlation degree of the feature vector and the sample vector, the correlation degrees of the feature vector and the plurality of sample vectors are summed and averaged, and the calculated result is taken as the average correlation degree of the feature vector and the sample vector set; if the charging pile operation data does not contain the feature vector and the sample vector set, no calculation is performed; For S305, the expression of the fault risk prediction model is: In the formula, Ω is the average correlation degree of the feature vector and the sample vector set, X is the state variable, and F is the fault risk value; the specific method for implementing the alarm decision based on the charging pile fault prediction method is: performing data analysis on the fault risk value F, if F is less than 0, no alarm is performed; if F is not less than 0, an alarm is performed, a technical personnel is notified to maintain, and the health status of the charging pile is set to fault.

2. The new energy charging pile fault prediction and maintenance system based on the Internet of Things applies the new energy charging pile fault prediction and maintenance method based on the Internet of Things as claimed in claim 1, characterized in that, It includes: The charging management module is used for monitoring and managing the charging pile; The data acquisition module is used for acquiring the vehicle model, the vehicle actual charging capacity, the charging pile output power, the charging pile service frequency, the charging pile maintenance frequency and the current charging pile health status of the charging pile, and determining the charging pile multi-dimensional information based on the Internet of Things; The current charging pile health status is divided into good and fault; The data processing module is used for processing the charging pile multi-dimensional information to obtain the charging pile operation data based on the Internet of Things; The data analysis module is used for comparing and analyzing the charging pile operation data to determine the charging pile fault prediction method based on the Internet of Things; The maintenance alarm module is used for implementing the alarm decision based on the charging pile fault prediction method, and the technical personnel is notified to maintain the charging pile based on the implementation of the alarm decision through the charging management module. 3.The IoT-based new energy charging pile fault prediction and maintenance system according to claim 2, characterized in that, The charging management module includes an identification unit and a supervision unit; the identification unit is used for adding a unique identification to each charging pile, and binding the charging pile multi-dimensional data collected by the charging pile based on the unique identification of each charging pile; The supervision unit is used for real-time supervision of the state of the charging pile, and identifying that the charging pile is in working or standby.

Images