A monitoring and early warning method and system for electric vehicle charging port

Abstract

The application relates to the technical field of monitoring and early warning, and discloses a monitoring and early warning method and system for electric vehicle charging ports, which comprises the following steps: acquiring a port image of each port monitoring point according to a unit time of a collection time period, determining whether each port monitoring point has temperature rise aging based on the port image and a twin neural aging model, judging whether each target port monitoring point has an anomaly based on a port waveform, verifying and determining an abnormal monitoring point based on historical data of each target port monitoring point, establishing a time-temperature rise early warning graph based on a unit time and the number of abnormalities, determining a temperature rise change trend according to the change of the number of abnormalities in the time-temperature rise early warning graph, and determining an early warning level of a to-be-monitored charging port based on the temperature rise change trend and the total number of abnormalities in the time-temperature rise early warning graph. The application ensures the reliability of monitoring and early warning of the charging ports.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of monitoring and early warning, in particular to a monitoring and early warning method and system for an electric vehicle charging port. BACKGROUND

[0002] With the popularity of electric vehicles, the DC fast charging system has become an important configuration for public charging scenarios and home high-power charging because of its high charging efficiency and the ability to quickly replenish battery capacity in a short time. As a connecting component of the DC fast charging system, the charging port bears the function of energy transmission between the battery pack and the charging pile, and its reliability directly determines the safety and stability of the charging process. However, during the DC charging process of the electric vehicle, the heat effect of the current will cause heat to be generated at the charging socket and the charging port. Both long charging time and large charging current will cause temperature accumulation at the charging port. This temperature accumulation will cause certain aging of the charging port. In addition, the wear, oxidation or deformation of the material at the charging port after long-term use will continuously increase the contact resistance, causing the charging port to further lose its insulation protection ability in a high-temperature environment, thereby causing an increase in the risk of charging port leakage and even causing a charging safety accident.

[0003] Therefore, it is necessary to design a monitoring and early warning method and system for an electric vehicle charging port to solve the problems existing in the current technology. SUMMARY

[0004] In view of this, the present application provides a monitoring and early warning method and system for an electric vehicle charging port, aiming to solve the problem of lack of monitoring of the charging port, thereby causing an increase in the risk of charging port leakage and ultimately causing a charging safety accident.

[0005] In one aspect, the present application provides a monitoring and early warning method for an electric vehicle charging port, comprising:

[0006] dividing the charging port to be monitored into a plurality of port monitoring points, acquiring a port image of each port monitoring point according to the unit time of the collection time period, determining whether each port monitoring point has temperature rise aging based on the port image and a twin neural aging model, and establishing a monitoring identifier for the charging port to be monitored;

[0007] based on the monitoring identifier, reducing a plurality of port monitoring points to determine target port monitoring points, determining a port waveform according to the target port monitoring points, judging whether each target port monitoring point has an abnormality based on the port waveform, and verifying the judgment result based on the historical data of each target port monitoring point to determine an abnormal monitoring point;

[0008] obtaining the number of abnormalities of the abnormal monitoring points, establishing a time-temperature rise early warning map based on the unit time and the number of abnormalities, and determining a temperature rise change trend according to the change of the number of abnormalities in the time-temperature rise early warning map;

[0009] determining the early warning level of the to-be-monitored charging port based on the temperature rise change trend and the total number of abnormalities in the time-temperature rise early warning map.

[0010] Further, when determining whether there is temperature rise aging of each port monitoring point based on the port image and the twin neural aging model, it includes:

[0011] obtaining a historical image set of the to-be-monitored charging port, and dividing the historical image set into a training set and a test set, using a twin neural network model as a model architecture, and setting a binary classification output layer in a full connection layer;

[0012] training the twin neural network model using the training set, and testing the trained twin neural network model using the test set, finally determining that the input is a port image and the output is the result of whether there is temperature rise aging of each port monitoring point.

[0013] Further, when establishing a monitoring identifier for the to-be-monitored charging port, it includes:

[0014] The monitoring identifier includes an interface standard monitoring identifier, an interface inconsistency monitoring identifier, and an interface tentative monitoring identifier;

[0015] If all port monitoring points do not have temperature rise aging, the interface standard monitoring identifier is generated for the to-be-monitored charging port;

[0016] If one or more port monitoring points have temperature rise aging and one or more port monitoring points do not have temperature rise aging, the interface tentative monitoring identifier is generated for the to-be-monitored charging port;

[0017] If all port monitoring points have temperature rise aging, the interface inconsistency monitoring identifier is generated for the to-be-monitored charging port.

[0018] Further, when determining target port monitoring points by reducing a plurality of port monitoring points based on the monitoring identifier, and determining a port waveform according to the target port monitoring points, and judging whether each target port monitoring point has an abnormality based on the port waveform, it includes:

[0019] If the interface tentative monitoring identifier is generated for the to-be-monitored charging port, the port monitoring points without temperature rise aging are deleted, and each remaining port monitoring point is determined as the target port monitoring point;

[0020] acquiring a standard port waveform corresponding to the target port monitoring point, and comparing the port waveform with the standard port waveform;

[0021] if the port waveform is the same as the standard port waveform, determining that the target port monitoring point does not exist abnormity;

[0022] if the port waveform is not the same as the standard port waveform, determining that the target port monitoring point exists abnormity.

[0023] Further, when determining the abnormal monitoring point by verifying the determination result based on the historical data of each target port monitoring point, comprising:

[0024] acquiring the historical determination result of each target port monitoring point;

[0025] if it is determined that the target port monitoring point does not exist abnormity, and is consistent with the historical determination result, the target port monitoring point is deleted;

[0026] if it is determined that the target port monitoring point does not exist abnormity, and is not consistent with the historical determination result, the target port monitoring point is retained;

[0027] if it is determined that the target port monitoring point exists abnormity, and is not consistent with the historical determination result, the target port monitoring point is retained;

[0028] if it is determined that the target port monitoring point exists abnormity, and is consistent with the historical determination result, the target port monitoring point is retained;

[0029] each target port monitoring point retained is determined as the abnormal monitoring point.

[0030] Further, when acquiring the number of abnormities of the abnormal monitoring point, and establishing the time-temperature rise warning graph based on the unit time and the number of abnormities, comprising:

[0031] establishing a rectangular coordinate system by taking time as the X axis and the number of abnormities as the Y axis;

[0032] transforming all unit times in the collection time period and the number of abnormities of the corresponding abnormal monitoring points into coordinate points;

[0033] substituting all coordinate points into the rectangular coordinate system, and connecting adjacent coordinate points in sequence to determine the time-temperature rise warning graph.

[0034] Further, when determining the temperature rise trend according to the change of the number of abnormities in the time-temperature rise warning graph, comprising:

[0035] obtaining a maximum number of anomalies and a minimum number of anomalies in the time-temperature rise early warning diagram, obtaining a difference between the maximum number of anomalies and the minimum number of anomalies, and determining a ratio of the difference to a number of intervals per unit time in the collection time period.

[0036] Further, in determining the temperature rise change trend according to the change in the number of anomalies in the time-temperature rise early warning diagram, the method further comprises:

[0037] when the ratio is greater than 0, determining that the temperature rise change trend is an upward trend;

[0038] when the ratio is equal to 0, determining that the temperature rise change trend is a stable trend.

[0039] Further, in determining the early warning level of the to-be-monitored charging port based on the temperature rise change trend and the total number of anomalies in the time-temperature rise early warning diagram, the method further comprises:

[0040] determining a sum of the total number of anomalies in the time-temperature rise early warning diagram;

[0041] when the temperature rise change trend is an upward trend and the sum is greater than or equal to a sum threshold, determining the early warning level of the to-be-monitored charging port as a first-level early warning;

[0042] when the temperature rise change trend is an upward trend and the sum is less than the sum threshold, or when the temperature rise change trend is a stable trend and the sum is greater than or equal to the sum threshold, determining the early warning level of the to-be-monitored charging port as a second-level early warning;

[0043] when the temperature rise change trend is a stable trend and the sum is less than the sum threshold, determining the early warning level of the to-be-monitored charging port as a third-level early warning;

[0044] the first-level early warning, the second-level early warning, and the third-level early warning have emergency levels decreasing in turn.

[0045] Compared with the prior art, the present application has the beneficial effects that: by dividing the to-be-monitored charging port into a plurality of port monitoring points and collecting port images per unit time, the problem of missing local temperature rise aging caused by single-point and static monitoring is avoided, thereby realizing dynamic dead-angle-free monitoring of the to-be-monitored charging port, combining the twin neural aging model to compare the port images with the characteristics of the normal state to accurately identify aging characteristics such as material discoloration and slight deformation caused by temperature rise, deleting risk-free port monitoring points based on the monitoring identifier to determine target port monitoring points, effectively filtering redundant data, avoiding meaningless data from occupying analysis resources, focusing monitoring and analysis on the risk area, improving the overall monitoring efficiency, judging the electrical abnormality condition through the port waveform of the target port monitoring point, making up for the limitation that the twin neural aging model can only identify material aging and cannot judge functional risks, in combination with historical data to verify and exclude transient disturbances such as power grid fluctuations, ensuring the reliability of the abnormal monitoring point determination, and establishing a time-temperature rise warning graph based on the unit time and the number of abnormalities, converting the discretized data into an intuitive trend curve, determining the warning level in combination with the temperature rise change trend and the total number of abnormalities, realizing the quantitative grading of risks, ensuring the safe operation of the to-be-monitored charging port, and timely intervening in the temperature rise aging problem, thereby delaying the wear and oxidation process of the to-be-monitored charging port, and providing strong support for the reliable operation of the direct current fast charging system.

[0046] In another aspect, the present application also provides a monitoring and warning system for an electric vehicle charging port, for applying the above-mentioned monitoring and warning method for the electric vehicle charging port, comprising:

[0047] The acquisition and analysis module is configured to divide the to-be-monitored charging port into a plurality of port monitoring points, acquire a port image of each port monitoring point per unit time according to the unit time of the acquisition time period, determine whether each port monitoring point has temperature rise aging based on the port image and the twin neural aging model, and establish a monitoring identifier for the to-be-monitored charging port;

[0048] The analysis and judgment module is configured to delete a plurality of port monitoring points based on the monitoring identifier to determine target port monitoring points, determine a port waveform according to the target port monitoring points, judge whether each target port monitoring point has an abnormality based on the port waveform, and verify the judgment result based on the historical data of each target port monitoring point to determine an abnormal monitoring point;

[0049] The graph establishment module is configured to acquire the number of abnormalities of the abnormal monitoring point, establish a time-temperature rise warning graph based on the unit time and the number of abnormalities, and determine a temperature rise change trend according to the change of the number of abnormalities in the time-temperature rise warning graph;

[0050] The monitoring and early warning module is configured to determine an early warning level of the to-be-monitored charging port based on the temperature rise change trend and the total number of abnormalities in the time-temperature rise early warning map.

[0051] It can be understood that the above-mentioned monitoring and early warning method and system for the charging port of an electric vehicle have the same beneficial effects, which will not be described here again. BRIEF DESCRIPTION OF DRAWINGS

[0052] Various other advantages and benefits will become apparent to those of ordinary skill in the art upon reading the following detailed description of the preferred embodiments. The accompanying drawings are included to provide a description of the preferred embodiments and are not intended to limit the scope of the application. Moreover, the same reference numerals are used throughout the same figures. In the drawings:

[0053] Figure 1 A flowchart of a monitoring and early warning method for a charging port of an electric vehicle is provided for an embodiment of the present application;

[0054] Figure 2 A functional block diagram of a monitoring and early warning system for a charging port of an electric vehicle is provided for an embodiment of the present application. DETAILED DESCRIPTION

[0055] Exemplary embodiments of the present disclosure will be described in detail with reference to the accompanying drawings. Although exemplary embodiments of the present disclosure are shown in the drawings, it should be understood that the present disclosure can be implemented in various forms and should not be limited by the embodiments set forth herein. Rather, these embodiments are provided so that the present disclosure can be more thoroughly understood and the scope of the present disclosure can be accurately conveyed to those skilled in the art. It should be noted that the embodiments in the present application and the features in the embodiments can be combined with each other without conflict. The present application will be described in detail below with reference to the accompanying drawings and in conjunction with the embodiments.

[0056] In some embodiments of the present application, referring to Figure 1 A monitoring and early warning method for a charging port of an electric vehicle is provided, which includes:

[0057] S100: dividing the to-be-monitored charging port into a plurality of port monitoring points, acquiring a port image of each port monitoring point according to a unit time of a collection time period, determining whether there is temperature rise aging in each port monitoring point based on the port image and a twin neural aging model, and establishing a monitoring identifier for the to-be-monitored charging port.

[0058] S200: determining target port monitoring points based on the monitoring identification, determining port waveforms according to the target port monitoring points, judging whether each target port monitoring point is abnormal based on the port waveforms, and verifying and determining an abnormal monitoring point based on the historical data of each target port monitoring point.

[0059] S300: obtaining the number of abnormalities of the abnormal monitoring point, establishing a time-temperature rise warning map based on the unit time and the number of abnormalities, and determining the temperature rise change trend according to the change of the number of abnormalities in the time-temperature rise warning map.

[0060] S400: determining the warning level of the to-be-monitored charging port based on the temperature rise change trend and the total number of abnormalities in the time-temperature rise warning map.

[0061] Specifically, the to-be-monitored charging port is divided into a plurality of port monitoring points. Due to the differences in current distribution and other conditions in different areas of the charging port, a single monitoring point is easy to miss local temperature rise aging problems. The layout of multiple port monitoring points can achieve global coverage of the to-be-monitored charging port, ensuring that there is no monitoring blind area. The number of specific port monitoring points is dynamically divided according to the size of the to-be-monitored charging port, which is not limited in the embodiment. The port image of each port monitoring point is obtained according to the unit time of the collection time period, so as to dynamically capture the port state change, avoid the static monitoring that cannot timely discover the instantaneous or gradual temperature rise aging phenomenon, and ensure the timeliness of the monitoring. The collection time period is determined according to the charging time length of the to-be-monitored charging port. The collection time period of the embodiment is preferably 5 minutes, and the unit time is 30 seconds. The temperature rise aging condition is determined based on the port image and the twin neural aging model. The twin neural aging model can accurately compare the normal charging heating condition of the to-be-monitored charging port with the features of the real-time collected port image, so as to quickly identify the aging features such as port material discoloration and slight deformation caused by temperature rise. Compared with artificial detection, the judgment of the model has a certain objectivity and accuracy, effectively avoiding the misjudgment of human experience. According to the judgment result of the model, a monitoring mark is established for the to-be-monitored charging port, in order to clearly distinguish the normal port monitoring point from the port monitoring point with potential risks, and lay a foundation for subsequent focusing on high-risk areas. Based on the monitoring mark, a plurality of port monitoring points are deleted to determine the target port monitoring point. The risk-free port monitoring point is filtered, the interference of redundant data on subsequent analysis is reduced, and the efficiency of the monitoring is improved. The port waveform is determined according to the target port monitoring point. The port waveform can directly reflect the electrical operation state of the target port monitoring point, such as abnormal features such as waveform distortion and peak value deviation, which can directly correspond to problems such as poor port contact and increased resistance, and provide a quantitative basis for abnormal judgment. Whether each target port monitoring point is abnormal is judged based on the port waveform, and then verified in combination with historical data, so as to rule out temporary waveform abnormalities of the target port monitoring point caused by power grid fluctuations, instantaneous current and other conditions, and ensure the reliability of the abnormal monitoring point judgment. The number of abnormal monitoring points is obtained, and a time-temperature rise warning graph is established in combination with the unit time. The number of abnormalities at a single time point cannot reflect the risk development trend of the to-be-monitored charging port. The time-temperature rise warning graph established according to the time dimension can clearly present the change law of the number of abnormalities with time, so as to determine the temperature rise change trend. The warning level is determined based on the temperature rise change trend and the total number of abnormalities. The temperature rise change trend reflects the dynamic development characteristics of the risk (such as rapid deterioration), and the total number of abnormalities reflects the influence range of the risk (such as local abnormality or global abnormality). The combination of the two can comprehensively evaluate the overall risk degree of the to-be-monitored charging port, and avoid the one-sidedness of the evaluation caused by only relying on a single indicator.

[0062] It can be understood that the combination of the layout of several port monitoring points and the image acquisition in unit time solves the limitations of traditional single-point monitoring and static monitoring, avoids the problem of missing local temperature rise aging, can timely capture the state change of the port, realizes dynamic monitoring of the whole life cycle of the monitored charging port, and the twin neural aging model is free from the dependence on subjective experience of artificial detection, effectively avoids the interference of light, environment and other factors on the detection result, improves the accuracy of temperature rise aging judgment, and further verifies the abnormal judgment result by the historical data, ensures the reliability of the determination of the abnormal monitoring point, avoids the waste of monitoring resources caused by misjudgment or the safety hidden danger caused by missed judgment, the time-temperature rise warning graph converts the abstract temperature rise risk change into an intuitive visual curve, so that relevant personnel can quickly master the risk development trend, and the warning level determined based on the temperature rise change trend and the total number of abnormalities ensures the reliability of resource investment, effectively reduces the safety accidents caused by the leakage of the monitored charging port, thereby ensuring the safety and stability of the charging process. At the same time, timely discovery and intervention of the temperature rise aging problem can delay the aging speed of the monitored charging port, and provide strong support for the long-term reliable operation of the direct current fast charging system of the electric vehicle.

[0063] In some embodiments of the present application, when determining whether each port monitoring point has temperature rise aging based on the port image and the twin neural aging model, the following steps are included: obtaining a historical image set of the monitored charging port, and dividing the historical image set into a training set and a test set; using the twin neural network model as the model architecture, and setting a binary classification output layer in the full connection layer; training the twin neural network model using the training set, and testing the trained twin neural network model using the test set; finally determining that the input is the port image, and the output is the result of whether each port monitoring point has temperature rise aging.

[0064] Specifically, a historical image set of the to-be-monitored charging port is acquired, the historical image set contains charging normal images of different time expectations to monitor the charging port (the to-be-monitored charging port does not exist in the aging case, which is determined by the laboratory or the to-be-monitored charging port when it is shipped), and the twin neural aging model learns the image feature rules of the temperature rise aging relying on the real historical image set. The historical image set is divided into a training set and a test set. The training set is used to provide learning samples for the model, so that the model can master the differences between the to-be-monitored charging port in the normal charging and the temperature rise aging in the image features such as material color and subtle deformation through repeated iterations. The test set is used to verify the training effect of the model, to avoid overfitting of the model to the training data, and to ensure that the model can still stably judge the port images that have not been seen. The twin neural network model is used as the architecture because its core is to realize judgment by comparing the similarity of the input features, so as to accurately capture the subtle feature changes of the port monitoring points caused by the temperature rise, which are difficult to distinguish with the naked eye, such as local slight discoloration and slight deformation of the material. The two-class output layer is set in the fully connected layer because the judgment result of each port monitoring point only needs to be "temperature rise aging exists" and "temperature rise aging does not exist". The two-class structure can directly correspond to the two results, which simplifies the output logic of the model while ensuring the stability of the results. The training process of the twin neural network model is as follows: input the data in the training set into the twin neural network model, the twin neural network model extracts the feature vector of the image through the feature extraction layer, so as to adjust the parameters of the model to reduce the judgment error. After the training is completed, the test set is used to verify the accuracy of the model. If it does not meet the standard, the parameters (learning rate, etc.) of the model are adjusted and retrained until the accuracy of the model meets the requirements of the judgment, so as to determine the twin neural aging model. The twin neural aging model can input the port image to output whether the port monitoring point exists temperature rise aging, which improves the accuracy and efficiency of the temperature rise aging judgment.

[0065] In some embodiments of the present application, when the monitoring identifier of the to-be-monitored charging port is established, the monitoring identifier includes an interface standard monitoring identifier, an interface inconsistency monitoring identifier, and an interface tentative monitoring identifier. If all port monitoring points do not exist temperature rise aging, the interface standard monitoring identifier of the to-be-monitored charging port is generated. If one or more port monitoring points exist temperature rise aging and one or more port monitoring points do not exist temperature rise aging, the interface tentative monitoring identifier of the to-be-monitored charging port is generated. If all port monitoring points exist temperature rise aging, the interface inconsistency monitoring identifier of the to-be-monitored charging port is generated.

[0066] Specifically, if the interface standard monitoring mark is generated for the to-be-monitored charging port, it indicates that all port monitoring points do not exist temperature rise aging condition, and the to-be-monitored charging port can continue to be used, if the interface inconsistent monitoring mark is generated for the to-be-monitored charging port, it indicates that all port monitoring points exist temperature rise aging condition, then immediately issue a warning, the level of the warning is a first-level warning, the emergency priority of the first-level warning is the highest. If the interface monitoring mark is formulated for the to-be-monitored charging port, it indicates that one or more port monitoring points exist temperature rise aging, and one or more port monitoring points do not exist temperature rise aging condition, targeted analysis is needed, by establishing the monitoring mark, the different risk states of the port monitoring points can be quickly and directly distinguished, and the order and efficiency of the monitoring are improved.

[0067] In some embodiments of the present application, when the target port monitoring points are determined by reducing a plurality of port monitoring points based on the monitoring mark, and the port waveform is determined according to the target port monitoring points, and whether each target port monitoring point exists abnormality is judged based on the port waveform, if the interface monitoring mark is formulated for the to-be-monitored charging port, the port monitoring points that do not exist temperature rise aging are deleted, and each remaining port monitoring point is determined as a target port monitoring point, the standard port waveform corresponding to the target port monitoring point is obtained, and the port waveform and the standard port waveform are compared, if the port waveform is the same as the standard port waveform, it is determined that the target port monitoring point does not exist abnormality, if the port waveform is different from the standard port waveform, it is determined that the target port monitoring point exists abnormality.

[0068] Specifically, the twin neural aging model identifies temperature rise aging through port image, which focuses on the "material state" of the port, and determines whether there is material discoloration, slight deformation, local oxidation and other aging manifestations caused by temperature rise according to image features. This determination can only confirm "whether aging occurs", and cannot directly associate whether aging has affected the electrical function of the monitored charging port. For example, a certain port monitoring point may have slight discoloration due to slight temperature rise (the model determines that it is temperature rise aging), but at this time the contact resistance of the monitored charging port is still within the normal range, and the energy transmission is not disturbed. If only the model is used for judgment, it may mistakenly classify "aging without functional risk" as a warning. Waveform determination is to analyze the electrical state through port waveform, which focuses on the "functional state" of the monitored charging port. The port waveform can directly reflect the current and voltage transmission law of each target port monitoring point, such as waveform distortion, peak shift, frequency anomaly and other characteristics, which directly corresponds to the problems of poor contact, increased resistance and insulation failure of the monitored charging port, thereby supplementing the blank of the model in determining "functional risk verification". The standard port waveform corresponding to the target port monitoring point is obtained because the standard port waveform is the electrical characteristic benchmark of the target port monitoring point in the normal operating state. Only by taking the standard port waveform as a reference can the actual port waveform be accurately judged to deviate from the normal state. By comparing the port waveform with the standard port waveform, it is determined whether the target port monitoring point is abnormal. If the two are completely consistent in waveform form, peak value, period and other key characteristics, that is, the waveforms are consistent, it means that the target port monitoring point has temperature rise aging, but the electrical function is not abnormal. If the waveforms are inconsistent, it means that the temperature rise aging has affected the electrical performance. The comparison method based on the standard port waveform avoids the subjectivity of relying on human experience judgment, and ensures the objectivity and accuracy of the abnormality determination result.

[0069] In some embodiments of the present application, when determining the abnormal monitoring point based on the historical data of each target port monitoring point, the historical judgment result of each target port monitoring point is obtained. If it is determined that the target port monitoring point has no abnormality and is consistent with the historical judgment result, the target port monitoring point is deleted. If it is determined that the target port monitoring point has no abnormality and is inconsistent with the historical judgment result, the target port monitoring point is retained. If it is determined that the target port monitoring point has an abnormality and is inconsistent with the historical judgment result, the target port monitoring point is retained. If it is determined that the target port monitoring point has an abnormality and is consistent with the historical judgment result, the target port monitoring point is retained. Each target port monitoring point retained is determined as an abnormal monitoring point.

[0070] Specifically, the historical judgment results of each target port monitoring point in the past (i.e., the abnormal or normal results previously obtained by the port monitoring point based on the port waveform) are called, and the judgment result currently obtained based on the port waveform is compared with the historical judgment result one by one, so as to exclude the judgment deviation caused by power grid fluctuation, transient current impact and other non-faults, and ensure that the finally determined abnormal monitoring point only contains the port monitoring point with real functional risk or continuous abnormality, avoiding the accidental deletion of real abnormal monitoring points or the retention of risk-free monitoring points due to single judgment. If it is currently determined that the target port monitoring point does not exist, and the conclusion is consistent with the historical judgment result, it means that the target port monitoring point is in a stable normal state for a long time, and there is no potential risk, so it is deleted. If it is currently determined that there is no abnormality, but it is inconsistent with the historical judgment result (such as the historical abnormality), it may be a transient interference that leads to the current misjudgment, so it is retained for further observation. If it is currently determined that there is an abnormality, and it is inconsistent with the historical judgment result, it may be a new abnormal situation, so it needs to be retained to track the risk change. If it is currently determined that there is an abnormality, and it is consistent with the historical judgment result, it means that the target port monitoring point has a continuous abnormality, so it needs to be retained. Finally, each target port monitoring point retained is determined as an abnormal monitoring point. By comparing with the historical data, single misjudgment caused by temporary interference is effectively filtered, and new or continuous real abnormality is avoided, the reliability of the abnormal monitoring point judgment is improved, accurate data basis is provided for subsequent establishment of the time-temperature rise warning graph, and the reliability of the overall monitoring and warning is further improved.

[0071] In some embodiments of the present application, when the number of abnormalities of the abnormal monitoring points is obtained, and the time-temperature rise warning graph is established based on the unit time and the number of abnormalities, the following steps are included: taking time as the X-axis and the number of abnormalities as the Y-axis to establish a rectangular coordinate system, converting all unit times and the number of abnormalities of the corresponding abnormal monitoring points in the collection time period into coordinate points, substituting all coordinate points into the rectangular coordinate system, and connecting adjacent coordinate points in sequence to determine the time-temperature rise warning graph.

[0072] Specifically, the number of abnormalities of the abnormal monitoring points in the unit time is discrete data, and it is difficult to intuitively present the change rule of the number of abnormalities with time by separately listing these data. Taking time as the X-axis and the number of abnormalities as the Y-axis to establish a rectangular coordinate system can visualize the corresponding relationship between the two, ensuring that the position of each coordinate point in the coordinate system matches the actual unit time and the number of abnormalities. Finally, adjacent coordinate points in the rectangular coordinate system are connected in sequence by lines to form a complete time-temperature rise warning graph that can reflect the trajectory of the change of the number of abnormalities with time, providing an intuitive graphical basis for subsequent determination of the temperature rise trend, avoiding the one-sidedness of trend judgment caused by relying only on discrete data, and further improving the reliability of monitoring and warning.

[0073] In some embodiments of the present application, when determining the temperature rise change trend according to the change of the number of anomalies in the time-temperature rise early warning diagram, the following steps are included: obtaining the maximum number of anomalies and the minimum number of anomalies in the time-temperature rise early warning diagram, obtaining the anomaly number difference of the maximum number of anomalies and the minimum number of anomalies, and determining the ratio of the anomaly number difference to the interval number per unit time in the collection time period.

[0074] In some embodiments of the present application, when determining the temperature rise change trend according to the change of the number of anomalies in the time-temperature rise early warning diagram, the following steps are included: when the ratio is greater than 0, it is determined that the temperature rise change trend is an upward trend; and when the ratio is equal to 0, it is determined that the temperature rise change trend is a stable trend.

[0075] Specifically, the number of anomalies in the time-temperature rise early warning diagram has certain local fluctuations, and it is difficult to accurately grasp the overall temperature rise change trend only by the changes of scattered coordinate points. The maximum number of anomalies is the maximum value of the number of anomalies corresponding to all unit times in the collection time period, and the minimum number of anomalies is the minimum value of the number of anomalies corresponding to all unit times in the collection time period. The maximum number of anomalies and the minimum number of anomalies can reflect the extreme range of the number of anomalies in the collection time period, and the difference between the two can reflect the overall change amplitude. By calculating the ratio in combination with the interval number per unit time, the change of the number of anomalies can be associated with the time dimension, thereby quantifying the overall change tendency and avoiding misjudgment of the temperature rise change trend due to local fluctuations or time span differences. The interval number per unit time is determined according to the collection time period and the unit time. The previously determined collection time period is 5 minutes, and the unit time is 30 seconds, so the interval number per unit time is 10. At the same time, the temperature rise change trend is determined by the ratio. When the ratio is greater than 0, it means that the maximum number of anomalies is greater than the minimum number of anomalies, which means that the number of anomalies in the collection time period is increasing as a whole, thereby determining that the temperature rise change trend is an upward trend. When the ratio is equal to 0, it means that the maximum number of anomalies is equal to the minimum number of anomalies, which means that the number of anomalies in the collection time period has no increase or decrease, thereby determining that the temperature rise change trend is a stable trend. By focusing on the correlation between the overall extreme value and the time interval, the interference of short-term local fluctuations is effectively filtered, and the reliability of overall monitoring is ensured.

[0076] In some embodiments of the present application, when determining the warning level of the to-be-monitored charging port based on the temperature rise change trend and the total number of all abnormalities in the time-temperature rise warning graph, the following steps are included: determining the sum of the total number of all abnormalities in the time-temperature rise warning graph, when the temperature rise change trend is an upward trend, and the sum is greater than or equal to a sum threshold, then the warning level of the to-be-monitored charging port is determined as a first-level warning, when the temperature rise change trend is an upward trend, and the sum is less than the sum threshold, or when the temperature rise change trend is a stable trend, and the sum is greater than or equal to the sum threshold, then the warning level of the to-be-monitored charging port is determined as a second-level warning, when the temperature rise change trend is a stable trend, and the sum is less than the sum threshold, then the warning level of the to-be-monitored charging port is determined as a third-level warning, and the emergency degree of the first-level warning, the second-level warning and the third-level warning decreases in turn.

[0077] Specifically, the temperature rise change trend reflects the dynamic development of the risk of the to-be-monitored charging port, and the sum of the total number of all abnormalities in the time-temperature rise warning graph reflects the overall impact range of the risk. A single dimension cannot comprehensively evaluate the emergency degree of the risk. If only the upward trend is considered, the case of small risk range may be ignored, and if only the sum is considered, the case of deteriorating risk development may be ignored. The combination of the two can achieve accurate grading according to the risk development and the risk range, ensuring that the warning level matches the actual risk. Through the warning grading of the dynamic temperature rise change trend and the impact range of the double dimensions, the warning deviation caused by a single index is avoided, the warning level is more in line with the actual risk situation, the safety operation of the to-be-monitored charging port is effectively ensured, and the safety accidents caused by insufficient response are avoided, and the waste of resources caused by excessive response is avoided.

[0078] In summary, the application has the beneficial effects that: by dividing the to-be-monitored charging port into a plurality of port monitoring points and collecting port images per unit time, the problem of missing local temperature rise aging caused by single-point and static monitoring is avoided, thereby realizing dynamic dead-angle-free monitoring of the to-be-monitored charging port, comparing the port images with the characteristics of the normal state by the twin neural aging model to accurately identify aging characteristics such as material discoloration and slight deformation caused by temperature rise, deleting risk-free port monitoring points based on the monitoring identifier to determine target port monitoring points, effectively filtering redundant data, avoiding meaningless data from occupying analysis resources, focusing monitoring and analysis on the risk area, improving the overall monitoring efficiency, judging the electrical abnormality condition through the port waveform of the target port monitoring point, making up for the limitation that the twin neural aging model can only identify material aging and cannot judge functional risks, in combination with historical data to verify and exclude transient disturbances such as power grid fluctuations, ensuring the reliability of the abnormal monitoring point determination, establishing a time-temperature rise warning graph based on the unit time and the number of abnormalities, converting the discretized data into an intuitive trend curve, determining the warning level in combination with the temperature rise change trend and the total number of abnormalities, realizing the quantification and grading of risks, ensuring the safe operation of the to-be-monitored charging port, and timely intervening in the temperature rise aging problem, thereby delaying the wear and oxidation process of the to-be-monitored charging port, and providing strong support for the reliable operation of the direct current fast charging system.

[0079] In another preferred mode based on the above embodiment, referring to Figure 2 The present embodiment provides a monitoring and warning system for an electric vehicle charging port for applying the above-mentioned monitoring and warning method for an electric vehicle charging port, comprising:

[0080] The acquisition and analysis module is configured to divide the to-be-monitored charging port into a plurality of port monitoring points, acquire the port image of each port monitoring point per unit time according to the unit time of the acquisition time period, determine whether each port monitoring point has temperature rise aging based on the port image and the twin neural aging model, and establish a monitoring identifier for the to-be-monitored charging port.

[0081] The analysis and judgment module is configured to delete a plurality of port monitoring points based on the monitoring identifier to determine target port monitoring points, determine the port waveform according to the target port monitoring points, judge whether each target port monitoring point has an abnormality based on the port waveform, and verify the judgment result based on the historical data of each target port monitoring point to determine an abnormal monitoring point.

[0082] The graph establishment module is configured to acquire the number of abnormalities of the abnormal monitoring point, establish a time-temperature rise warning graph based on the unit time and the number of abnormalities, and determine the temperature rise change trend according to the change of the number of abnormalities in the time-temperature rise warning graph.

[0083] The monitoring and early warning module is configured to determine an early warning level of the to-be-monitored charging port based on the temperature rise change trend and the total number of abnormalities in the time-temperature rise early warning diagram.

[0084] Those skilled in the art will appreciate that embodiments of the application can be provided as methods, systems or computer program products. Accordingly, the application can be embodied in the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware aspects. Furthermore, the application can be embodied in the form of a computer program product on one or more computer readable storage media (including, but not limited to, disk memory, CD-ROMs, optical storage media, etc.) having computer usable program code embodied thereon.

[0085] The application is described with reference to the flowchart and / or block diagrams of the methods, apparatus (systems) and computer program products according to embodiments of the application. It should be understood that each flow and / or block in the flowchart and / or block diagrams, and a combination of flows and / or blocks in the flowchart and / or block diagrams, 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 apparatus to produce a machine, so that the instructions, which are executed via the processor of the computer or other programmable data processing apparatus, generate a means for implementing the functions specified in the flowchart and / or block diagrams of the flowchart and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus for implementing functions specified in one or more flows and / or blocks.

[0086] These computer program instructions can also be stored in a computer readable memory that can direct a computer or other programmable data processing apparatus to work in a specific manner, so that the instructions stored in the computer readable memory produce a manufacture product including instruction apparatus, which implements the functions specified in the flowchart and / or block diagrams of the flowchart and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus for implementing functions specified in one or more flows and / or blocks.

[0087] These computer program instructions can also be loaded onto a computer or other programmable data processing apparatus, so that a series of operation steps are performed on the computer or other programmable data processing apparatus to produce a computer implemented process, so that the instructions executed on the computer or other programmable data processing apparatus provide a means for implementing the functions specified in the flowchart and / or block diagrams of the flowchart and / or block diagrams. Figure 1 one or more flows and / or blocks Figure 1 an apparatus for implementing functions specified in one or more flows and / or blocks.

[0088] It should be pointed out finally that the above embodiments are only used to illustrate the technical solutions of the present application but not to limit it. Although the present application has been described in detail with reference to the above embodiments, it should be understood by those skilled in the art that the specific embodiments of the present application can be modified or replaced equivalently without departing from the spirit and scope of the present application, and any modification or equivalent replacement without departing from the spirit and scope of the present application should be covered in the protection scope of the claims of the present application.

Claims

1. A monitoring and early warning method for electric vehicle charging ports, characterized in that, include: The charging port to be monitored is divided into several port monitoring points. The port image of each port monitoring point is obtained according to the unit time of the collection period. Based on the port image and the twin neural aging model, it is determined whether there is temperature rise aging at each port monitoring point, and a monitoring mark is established for the charging port to be monitored. Based on the monitoring identifier, several port monitoring points are reduced to determine the target port monitoring point, and the port waveform is determined according to the target port monitoring point. Based on the port waveform, it is determined whether there is an anomaly at each target port monitoring point, and the judgment result is verified based on the historical data of each target port monitoring point to determine the abnormal monitoring point. The number of anomalies at the anomaly monitoring points is obtained, and a time-temperature rise early warning chart is established based on the unit time and the number of anomalies. The temperature rise trend is determined according to the change in the number of anomalies in the time-temperature rise early warning chart. The warning level of the charging port to be monitored is determined based on the temperature rise trend and the total number of anomalies in the time-temperature rise early warning graph. When establishing a monitoring identifier for the charging port to be monitored, the following steps are included: The monitoring identifiers include interface standard monitoring identifiers, interface incompatibility monitoring identifiers, and interface proposed monitoring identifiers. If none of the port monitoring points show any temperature rise or aging, then the interface standard monitoring identifier is generated for the charging port to be monitored. If one or more port monitoring points exhibit temperature rise aging, and one or more port monitoring points do not exhibit temperature rise aging, then a proposed monitoring identifier for the charging port to be monitored is generated. If all port monitoring points show signs of temperature rise and aging, an interface mismatch monitoring identifier will be generated for the charging port to be monitored. When determining target port monitoring points by eliminating several port monitoring points based on the monitoring identifier, determining port waveforms based on the target port monitoring points, and judging whether there are any abnormalities at each target port monitoring point based on the port waveforms, the process includes: If a monitoring identifier is generated for the charging port to be monitored, then port monitoring points that do not have temperature rise aging will be deleted, and each remaining port monitoring point will be determined as the target port monitoring point. Obtain the standard port waveform corresponding to the target port monitoring point, and compare the port waveform with the standard port waveform; If the waveform of the port is the same as that of the standard port, then it is determined that there is no abnormality at the target port monitoring point; If the waveform of the port is different from that of the standard port, it is determined that there is an anomaly at the target port monitoring point.

2. The monitoring and early warning method for electric vehicle charging ports according to claim 1, characterized in that, When determining whether temperature rise aging exists at each port monitoring point based on the port images and twin neuron aging model, the following steps are included: The historical image set of the charging port to be monitored is obtained, and the historical image set is divided into a training set and a test set. A twin neural network model is used as the model architecture, and a binary classification output layer is set in the fully connected layer. The training set is used to train the twin neural network model, and the test set is used to test the trained twin neural network model. Finally, the input is determined to be the port image, and the output is the result of whether there is temperature rise and aging at each port monitoring point.

3. The monitoring and early warning method for electric vehicle charging ports according to claim 2, characterized in that, When verifying the judgment results and identifying abnormal monitoring points based on historical data from each target port monitoring point, the following steps are included: Obtain the historical judgment results for each target port monitoring point; If it is determined that there is no abnormality in the target port monitoring point and the result is consistent with the historical judgment, then the target port monitoring point will be deleted. If it is determined that there is no abnormality at the target port monitoring point and the result is inconsistent with the historical judgment, then the target port monitoring point will be retained. If the target port monitoring point is determined to be abnormal and is inconsistent with the historical judgment results, the target port monitoring point will be retained. If the target port monitoring point is determined to be abnormal and the result is consistent with the historical judgment, the target port monitoring point will be retained. Each target port monitoring point that is retained is identified as the abnormal monitoring point.

4. The monitoring and early warning method for electric vehicle charging ports according to claim 3, characterized in that, When acquiring the number of anomalies at the aforementioned anomaly monitoring points and establishing a time-temperature rise early warning chart based on the unit time and the number of anomalies, the following steps are included: Establish a rectangular coordinate system with time as the X-axis and the number of anomalies as the Y-axis; Convert all unit time points and the number of anomalies at the corresponding anomaly monitoring points within the data collection period into coordinate points. Substitute all the coordinate points into the rectangular coordinate system, and connect adjacent coordinate points in sequence to determine the time-temperature rise early warning chart.

5. The monitoring and early warning method for electric vehicle charging ports according to claim 4, characterized in that, When determining the temperature rise trend based on the changes in the number of anomalies in the time-temperature rise early warning graph, the following steps are included: Obtain the maximum and minimum number of anomalies in the time-temperature rise early warning graph, obtain the difference between the maximum and minimum number of anomalies, and determine the ratio of the difference between the number of anomalies to the number of intervals per unit time within the collection period.

6. The monitoring and early warning method for electric vehicle charging ports according to claim 5, characterized in that, When determining the temperature rise trend based on the changes in the number of anomalies in the time-temperature rise early warning graph, the method further includes: When the ratio is greater than 0, the temperature rise trend is determined to be an upward trend; When the ratio is equal to 0, the temperature rise trend is determined to be a stable trend.

7. The monitoring and early warning method for electric vehicle charging ports according to claim 6, characterized in that, When determining the warning level of the charging port to be monitored based on the temperature rise trend and the total number of anomalies in the time-temperature rise warning graph, the following steps are included: Determine the sum of all anomalies in the time-temperature rise early warning graph; When the temperature rise trend is upward and the sum is greater than or equal to the sum threshold, the warning level of the charging port to be monitored is determined to be a Level 1 warning. When the temperature rise trend is upward and the sum is less than the sum threshold, or when the temperature rise trend is stable and the sum is greater than or equal to the sum threshold, the warning level of the charging port to be monitored is determined to be a level two warning. When the temperature rise trend is a stable trend and the sum is less than the sum threshold, the warning level of the charging port to be monitored is determined to be a level three warning. The urgency level of the Level 1, Level 2, and Level 3 early warnings decreases sequentially.

8. A monitoring and early warning system for electric vehicle charging ports, used to apply the monitoring and early warning method for electric vehicle charging ports as described in any one of claims 1-7, characterized in that, include: The data acquisition and analysis module is configured to divide the charging port to be monitored into several port monitoring points, acquire port images of each port monitoring point according to the unit time of the data acquisition period, determine whether there is temperature rise aging at each port monitoring point based on the port images and the twin neural aging model, and establish a monitoring label for the charging port to be monitored. The analysis and judgment module is configured to reduce a number of port monitoring points based on the monitoring identifier to determine the target port monitoring point, determine the port waveform based on the target port monitoring point, determine whether there is an anomaly at each target port monitoring point based on the port waveform, and verify the judgment result based on the historical data of each target port monitoring point to determine the abnormal monitoring point. The graph creation module is configured to acquire the number of anomalies at the anomaly monitoring points, create a time-temperature rise early warning graph based on the unit time and the number of anomalies, and determine the temperature rise trend based on the change in the number of anomalies in the time-temperature rise early warning graph. The monitoring and early warning module is configured to determine the early warning level of the charging port to be monitored based on the temperature rise trend and the total number of anomalies in the time-temperature rise early warning graph.

Images