Abnormal early warning method and early warning system for mobile energy storage charging pile

Abstract

The invention provides an abnormity early warning method and early warning system of a mobile energy storage charging pile, and relates to the technical field of abnormity early warning of the mobile energy storage charging pile, and the method comprises the steps: collecting the current, voltage and charging state values of the charging pile in the historical charging process, and calculating the health state value of the charging pile; secondly, performing state marking on historical monitoring data by utilizing an expert scoring method, and constructing an anomaly detection model based on a random forest algorithm; and acquiring current, voltage, charging state value and health state value data at the current moment, and acquiring a current state mark through the model. And if the state is marked as an abnormal state, calculating a danger index at the current moment by combining the temperature and humidity data, and comparing the danger index with a set threshold value to judge whether an early warning needs to be sent out or not. The method can effectively identify the abnormal condition of the charging pile, and gives out early warning in advance, thereby reducing the potential safety hazard and improving the user experience.

Description

Technical Field

[0001] The present invention relates to the technical field of abnormal warning of mobile energy storage charging piles, and specifically to an abnormal warning method of a mobile energy storage charging pile and an abnormal warning method and warning system of a mobile energy storage charging pile. Background Art

[0002] With the increasing awareness of environmental protection and the advancement of technology, the use of electric vehicles is becoming increasingly popular. Governments around the world have introduced a series of policies to encourage the promotion of electric transportation in order to reduce greenhouse gas emissions and reduce dependence on fossil fuels. With the surge in the number of electric vehicles, the demand for charging infrastructure has also risen rapidly. Traditional fixed charging piles often cannot meet the charging needs of users in different locations, while mobile energy storage charging piles can flexibly provide services in different places, especially in public places, temporary events and remote areas. With the advancement of battery technology, the energy density, charging and discharging efficiency and safety of mobile energy storage systems have been significantly improved. This makes mobile energy storage charging piles more reliable in performance and able to meet the needs of a variety of usage scenarios.

[0003] However, existing charging piles often face a variety of technical problems during operation, such as battery overheating, abnormal charging current, and excessive humidity. These problems not only affect charging efficiency, but may also cause equipment damage and safety hazards. In addition, abnormal monitoring in existing technologies often relies on a single parameter and lacks comprehensive data analysis and intelligent early warning mechanisms, making it difficult to detect and handle abnormal situations in a timely manner. Therefore, there is an urgent need for an abnormal early warning system that integrates multi-parameter monitoring and intelligent analysis to improve the safety and reliability of charging piles.

[0004] The prior art has the following deficiencies:

[0005] Many traditional charging piles lack real-time monitoring capabilities and cannot continuously track key parameters during the charging process, such as current, voltage, and temperature. This lack of real-time monitoring makes it difficult to detect abnormal situations in a timely manner, increasing the possibility of equipment failure and safety risks. In addition, existing anomaly detection methods mostly rely on simple threshold judgments and lack intelligent analysis capabilities, which can easily lead to false alarms or missed alarms, and cannot fully reflect the actual operating status of the charging pile, affecting the efficiency of operation and maintenance. Secondly, the existing system has a relatively slow response speed when a fault occurs, and has failed to establish an effective early warning mechanism. It often handles the fault only after it occurs. This delay may lead to more serious consequences and even affect the user's safety experience. In addition, many monitoring systems have poor adaptability and cannot automatically adjust monitoring parameters according to different usage environments and conditions, which makes the monitoring effect in a specific environment unsatisfactory and cannot achieve effective risk prevention and control.

[0006] The above information disclosed in this Background section is only for enhancement of understanding of the background of the present disclosure and therefore it may contain information that does not constitute the prior art that is already known to one of ordinary skill in the art. Summary of the invention

[0007] The object of the present invention is to provide an abnormal warning method and a warning system for a mobile energy storage charging pile to solve the problems raised in the above background technology.

[0008] To achieve the above object, the present invention provides the following technical solutions:

[0009] A method for abnormal warning of a mobile energy storage charging pile, the specific steps comprising:

[0010] Step 1: Collect the current, voltage and charging state values ​​at multiple historical collection times during the historical charging process of the charging pile, calculate the health state value of the charging pile at the corresponding historical collection time, and integrate the current, voltage, charging state value and health state value as historical monitoring data;

[0011] Step 2: Based on the expert scoring method, the status of the charging piles at each moment in the historical monitoring data is marked, and the historical monitoring data is used as input, and the status mark of the charging pile at the corresponding moment is used as output. An anomaly detection model is constructed based on the random forest algorithm, and the status mark is normal state and abnormal state;

[0012] Step 3: Collect the current, voltage and charging state values ​​at the current moment, calculate the health state value at the current moment, and input them into the trained anomaly detection model to obtain the state mark at the current moment;

[0013] Step 4: If the charging pile is marked as abnormal at the current moment, the danger index of the charging pile at the current moment is calculated based on the monitoring data at the current moment and combined with the temperature and humidity data of the charging pile at the current moment;

[0014] Step 5: Compare the danger index of the charging pile at the current moment with the preset danger threshold to determine whether the charging pile needs to issue an early warning at the current moment.

[0015] Furthermore, the specific logic for collecting historical monitoring data of charging piles is:

[0016] A multifunctional battery monitor is installed at the power input port of the charging pile to collect the current, voltage and charging status value data of the charging pile. A temperature sensor and a humidity sensor are installed at the power module position to collect the temperature and humidity data of the charging pile. The formula for obtaining the health status value of the charging pile is:

[0017]

[0018] Among them, HI represents the health status value, R is the internal impedance of the charging pile battery, and R max is the maximum normal impedance of the charging pile battery determined by testing, L is the number of charging cycles of the battery, and L max is the maximum number of cycles expected for the battery in normal use, SOC is the state of charge value, ω1, ω2, ω3 are the weights of the corresponding items, 0<ω3<ω2<ω1<1, and ω1+ω2+ω3=1;

[0019] The current, voltage and charging status value data collected at multiple historical collection times during the historical charging process of the charging pile, as well as the calculated health status value, are standardized using the Z-score method, and the preprocessed data are stored as a sample data set.

[0020] Furthermore, the specific logic for building anomaly detection models is as follows:

[0021] In normal state, mark A=1, and in abnormal state, mark A=0. The sample data set is divided into training set and test set with a division ratio of 7:3. The historical monitoring data in the training set is used as input, and the state mark is used as the label. The random forest model in machine learning is selected to build the anomaly detection model, and then the historical monitoring data in the test set is input to evaluate the performance of the model. The evaluation indicators include accuracy, precision, and recall. If all changes in the evaluation indicators are less than 0.01 in five iterations, the model training is considered to be completed.

[0022] Furthermore, the specific logic for calculating the danger index of the charging pile at the current moment is:

[0023] The formula for obtaining the danger index of the charging pile at the current moment is:

[0024]

[0025] Where P represents the danger index of the charging pile at the current moment, I is the current collected by the sensor at the current moment, V is the voltage collected by the sensor at the current moment, T is the temperature collected by the sensor at the current moment, H is the humidity collected by the sensor at the current moment, and I max is the maximum current allowed by the charging pile, V max is the maximum voltage allowed by the charging pile, is the optimal operating temperature of the charging pile. is the optimal working humidity of the charging pile, β1, β2, β3, β4, β5, and β6 are the weight coefficients of their respective items, 0<β5=β6<β4<β3<β1=β2<1, and β1+β2+β3+β4+β5+β6=1.

[0026] Furthermore, the specific logic for determining whether the charging pile needs to issue an early warning at the current moment is:

[0027] Compare the danger index of the charging pile at the current moment with the preset danger threshold. If the danger index of the charging pile at the current moment is lower than the preset danger threshold, no warning will be issued temporarily. If the danger index of the charging pile at the current moment is higher than the preset danger threshold, a warning needs to be issued to remind that there is an abnormality in the charging of the charging pile at the current moment.

[0028] The present invention further provides an abnormal warning system for a mobile energy storage charging pile, wherein the abnormal warning system for a mobile energy storage charging pile is used to implement the abnormal warning method for the mobile energy storage charging pile, comprising:

[0029] The data acquisition module is used to collect the current, voltage and charging status values ​​at multiple historical collection moments during the historical charging process of the charging pile, calculate the health status value of the charging pile at the corresponding historical collection moment, and integrate the current, voltage, charging status value and health status value into historical monitoring data;

[0030] The anomaly detection module builds a module for marking the status of the charging piles at each moment in the historical monitoring data based on the expert scoring method, and takes the historical monitoring data as input and the status mark of the charging piles at the corresponding moment as output, and builds an anomaly detection model based on the random forest algorithm, wherein the status mark is a normal state and an abnormal state;

[0031] The real-time monitoring and status recognition module is used to collect the current current, voltage and charging status values, calculate the health status value at the current moment, and input them into the trained anomaly detection model to obtain the current status mark;

[0032] A danger index acquisition module is used to calculate the danger index of the charging pile at the current moment based on the monitoring data at the current moment and in combination with the temperature and humidity data of the charging pile at the current moment if the charging pile is marked as an abnormal state at the current moment;

[0033] The early warning judgment and response module is used to compare the danger index of the charging pile at the current moment with the preset danger threshold to determine whether the charging pile needs to issue an early warning at the current moment.

[0034] In the above technical solution, the technical effects and advantages provided by the present invention are:

[0035] This solution achieves comprehensive monitoring of mobile energy storage charging piles by establishing an intelligent multi-parameter monitoring and adaptive early warning system. The system integrates multiple important parameters such as temperature, humidity, current, voltage, etc., and uses machine learning algorithms for anomaly detection, with the ability to adaptively adjust the early warning threshold. Through this innovative method, it can effectively reduce the equipment failure rate, improve charging safety, and realize real-time monitoring and remote maintenance to ensure the efficient and stable operation of the charging pile, thereby providing users with a safer and more reliable charging experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0036] Figure 1 It is a schematic diagram of the overall method flow of the present invention;

[0037] Figure 2 Schematic diagram of the system structure of the present invention. DETAILED DESCRIPTION

[0038] In order to make the objectives, technical solutions and advantages of the present invention more clearly understood, the present invention is further described in detail below in conjunction with specific embodiments.

[0039] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in the present invention should be understood by people with ordinary skills in the field to which the present invention belongs. The words "first", "second" and similar words used in the present invention do not indicate any order, quantity or importance, but are only used to distinguish different components. "Include" or "comprise" and similar words mean that the elements or objects appearing before the word include the elements or objects listed after the word and their equivalents, without excluding other elements or objects. "Connect" or "connected" and similar words are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. "Up", "down", "left", "right" and the like are only used to indicate relative positional relationships. When the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0040] Example:

[0041] See also Figure 1 , the present invention provides a technical solution:

[0042] A method for abnormal warning of a mobile energy storage charging pile, the specific steps comprising:

[0043] Step 1: Collect the current, voltage and charging state values ​​at multiple historical collection times during the historical charging process of the charging pile, calculate the health state value of the charging pile at the corresponding historical collection time, and integrate the current, voltage, charging state value and health state value as historical monitoring data;

[0044] In this embodiment, the specific logic for collecting the historical monitoring data of the charging pile is:

[0045] A multifunctional battery monitor is installed at the power input port of the charging pile to collect the current, voltage and charging status value data of the charging pile. A temperature sensor and a humidity sensor are installed at the power module position to collect the temperature and humidity data of the charging pile. The formula for obtaining the health status value of the charging pile is:

[0046]

[0047] Among them, HI represents the health status value, R is the internal impedance of the charging pile battery, and R max is the maximum normal impedance of the charging pile battery determined by testing, L is the number of charging cycles of the battery, and L max is the maximum number of cycles expected for the battery in normal use, SOC is the state of charge value, ω1, ω2, ω3 are the weights of the corresponding items, 0<ω3<ω2<ω1<1, and ω1+ω2+ω3=1;

[0048] The current, voltage and charging status value data collected at multiple historical collection times during the historical charging process of the charging pile, as well as the calculated health status value, are standardized using the Z-score method, and the preprocessed data are stored as a sample data set.

[0049] By collecting current, voltage, temperature, and humidity data at multiple historical collection moments during the historical charging process of the charging pile, we can fully understand the operating status of the charging pile. These data provide a basis for subsequent analysis and can help identify potential abnormal conditions. For example, abnormal fluctuations in current and voltage may indicate a fault in the charging pile, while temperature and humidity data can help determine the impact of environmental factors on charging efficiency and equipment safety. Secondly, obtaining charging status values ​​and health status values ​​are important indicators for evaluating the performance of charging piles. By comparing the internal impedance and the number of charging cycles with their respective maximum normal values, these indicators can be converted into relative values. This standardization process allows the various parameters to maintain consistency in the calculation formula, making the comparison between parameters more intuitive. The formula and In the form of a graph, the deviation from the normal range is highlighted. In particular, when the internal impedance increases or the number of charge cycles approaches its maximum value, the health status value will decrease accordingly. This design helps to draw attention and prompt measures to extend the service life of the device.

[0050] The formula takes into account several important variables, including the state of charge value, the internal impedance of the battery, and the number of battery charging cycles. As a direct indicator of the current charge level of the battery, SOC is a direct indicator of the current charge level of the battery, which affects the available capacity and performance of the battery. The higher the state of charge, the more available power the battery has, which is directly related to the health of the charging pile and the user experience. Therefore, SOC is given the highest weight ω1 in the health state value, reflecting its significant impact on the health state of the charging pile. Internal impedance is an important indicator of battery performance and aging. An increase in impedance usually means a decrease in the energy transfer efficiency of the battery, which may lead to problems such as reduced charging efficiency and heat generation. Therefore, although internal impedance is inferior to the state of charge, it is still a key factor, and it is reasonable to give it a medium weight ω2. The number of charging cycles refers to the battery's usage history and directly affects the remaining life and performance of the battery. Although the number of cycles has a significant impact on the battery health state, compared with the immediate state of charge and internal impedance, its impact is usually indirect and gradual. Therefore, it is reasonable to give it the lowest weight ω3, reflecting the auxiliary role of the number of cycles in the health state assessment.

[0051] SOC can reflect the charging progress of the battery, while HI takes into account the internal impedance and cycle times of the battery, providing a quantitative analysis of the battery's health status. Through these two indicators, the charging pile management system can monitor the battery status in real time and issue warnings in time when abnormalities occur to avoid possible safety hazards.

[0052] The preprocessing step of implementing Z-score standardization further enhances the comparability and accuracy of the data. Through standardization, the indicators in the data set can be converted to the same scale, eliminating the impact of different dimensions and orders of magnitude on the analysis results. In this way, in subsequent anomaly detection and data analysis, the model can more accurately identify abnormal behaviors and improve the accuracy and reliability of early warnings.

[0053] Step 2: Based on the expert scoring method, the status of the charging piles at each moment in the historical monitoring data is marked, and the historical monitoring data is used as input, and the status mark of the charging pile at the corresponding moment is used as output. An anomaly detection model is constructed based on the random forest algorithm, and the status mark is normal state and abnormal state;

[0054] In this embodiment, the specific logic for constructing the anomaly detection model is:

[0055] In normal state, mark A=1, and in abnormal state, mark A=0. The sample data set is divided into training set and test set with a division ratio of 7:3. The historical monitoring data in the training set is used as input, and the state mark is used as the label. The random forest model in machine learning is selected to build the anomaly detection model, and then the historical monitoring data in the test set is input to evaluate the performance of the model. The evaluation indicators include accuracy, precision, and recall. If all changes in the evaluation indicators are less than 0.01 in five iterations, the model training is considered to be completed.

[0056] The expert scoring method is used to mark the status of historical charging data, so that the operating status of the charging pile is systematically defined. Through the professional knowledge of experts, the voltage and current data at each moment are marked as normal state (A=1) or abnormal state (A=0), providing a reliable training basis for the subsequent machine learning model. This labeling not only increases the interpretability of the data, but also helps the model to learn to identify the feature differences between normal and abnormal states more accurately. It is highly reasonable to use the random forest algorithm to build anomaly detection models. Random forest is an ensemble learning method that can improve the accuracy and stability of the model by building multiple decision trees and combining their outputs. The algorithm has strong robustness to data noise and overfitting, and is very suitable for processing application scenarios such as charging piles with complex features and noisy data. Finally, by monitoring the change amplitude of the evaluation index in five iterations, the completion of the model training is determined by the standard of less than 0.01, which ensures the convergence and reliability of the model. This iterative verification method not only effectively prevents the overfitting of the model, but also ensures the feasibility and efficiency of the model in practical applications, thereby improving the accuracy, precision and recall of anomaly detection.

[0057] Step 3: Collect the current, voltage and charging state values ​​at the current moment, calculate the health state value at the current moment, and input them into the trained anomaly detection model to obtain the state mark at the current moment;

[0058] In this embodiment, the current, voltage and charging status value data of the charging pile at the current moment are collected in real time, the health status value of the charging pile at this moment is calculated, and the standardization processing is consistent with the historical monitoring data. The Z-score is used to standardize the current current, voltage, charging status value and health status value data at the current moment. After standardization, the current, voltage, charging status value and health status value are input into the trained anomaly detection model to obtain the state mark predicted by the model.

[0059] Step 4: If the charging pile is marked as abnormal at the current moment, the danger index of the charging pile at the current moment is calculated based on the monitoring data at the current moment and combined with the temperature and humidity data of the charging pile at the current moment;

[0060] In this embodiment, the specific logic for calculating the danger index of the charging pile at the current moment is:

[0061] The formula for obtaining the danger index of the charging pile at the current moment is:

[0062]

[0063] Where P represents the danger index of the charging pile at the current moment, I is the current collected by the sensor at the current moment, V is the voltage collected by the sensor at the current moment, T is the temperature collected by the sensor at the current moment, H is the humidity collected by the sensor at the current moment, and I max is the maximum current allowed by the charging pile, V max is the maximum voltage allowed by the charging pile, is the optimal operating temperature of the charging pile. is the optimal working humidity of the charging pile, β1, β2, β3, β4, β5, and β6 are the weight coefficients of their respective items, 0<β5=β6<β4<β3<β1=β2<1, and β1+β2+β3+β4+β5+β6=1.

[0064] By combining the current, voltage, temperature and humidity data of the charging pile and calculating the hazard index (P), the operating status and safety of the charging pile can be comprehensively evaluated. This method not only focuses on a single indicator, but also fully considers multiple influencing factors, thereby improving the accuracy of risk assessment. and The current and voltage are standardized to between 0 and 1 for easy comparison. If the current or voltage is close to the maximum value, the corresponding risk index increases. The absolute values ​​of the current and voltage may vary depending on the characteristics of the equipment, and using the absolute values ​​directly may lead to misjudgment. The standardized relative values ​​can more accurately reflect the proximity of the operating status of the charging pile to the safety limit. The deviation of temperature and humidity is calculated as and If the current temperature or humidity is far from its optimal value, the hazard index will increase significantly. The square form makes the deviation of temperature and humidity amplified. This means that even a small deviation will have a large impact on the hazard index, which emphasizes the importance of equipment in optimal working conditions. In addition, the impact of temperature and humidity on equipment is usually nonlinear. Too high or too low temperature, too high or too low humidity will have a significant impact on the performance of the charging pile. The square term can effectively reflect this nonlinear relationship. The charging state value reflects the working state of the charging pile, and the health state value indicates its reliability. The charging state value and the health state value are inversely proportional to the hazard index. The higher the charging state value and the health state value, the lower the hazard index, and the lower the charging state value and the health state value, the higher the hazard index. When the SOC or HI is low, that is, when the battery is in poor condition, the value of the exponential function will be higher, indicating that the hazard is significantly increased. This nonlinear reaction can more intuitively reflect the relationship between the battery state and the safety risk, and enhance the sensitivity to changes in SOC and HI.

[0065] The current and voltage of the charging pile are important parameters that directly affect safe operation. Excessive current or voltage may cause safety hazards such as equipment damage and fire, so it is logical to give them higher weights (same and higher), reflecting their key role in assessing danger. Temperature and humidity also have significant effects on equipment, but compared with current and voltage, the impact is usually indirect and continuous. Their weights are slightly lower, reflecting that although they are important, their impact on the safety of the equipment is slightly weaker than that of current and voltage, so they are set to be higher but lower than the weights of current and voltage. SOC and HI provide indirect information on the health and performance evaluation of the battery. The impact weights of these two items are the same and lower, mainly because they reflect the long-term health trend of the equipment, while in emergency situations, the instantaneous state of current and voltage will be more critical. In addition, temperature and humidity are important physical parameters in the operating environment of the charging pile, which directly affect the performance and safety of the equipment. Extreme temperature and humidity conditions can cause equipment overheating, short circuit or material aging, leading to failures or safety accidents. Therefore, fluctuations in temperature and humidity have a direct and obvious impact on the safety of the charging pile in the short term, and their weights are higher than the charging state and health state values.

[0066] Step 5: Compare the danger index of the charging pile at the current moment with the preset danger threshold to determine whether the charging pile needs to issue an early warning at the current moment.

[0067] In this embodiment, the specific logic for determining whether the charging pile needs to issue an early warning at the current moment is:

[0068] Compare the danger index of the charging pile at the current moment with the preset danger threshold. If the danger index of the charging pile at the current moment is lower than the preset danger threshold, no warning will be issued temporarily. If the danger index of the charging pile at the current moment is higher than the preset danger threshold, a warning needs to be issued to remind that there is an abnormality in the charging of the charging pile at the current moment.

[0069] By comparing the danger index with the preset danger threshold, the system can realize real-time monitoring of the operating status of the charging pile. When the danger index is higher than the preset threshold, a warning is issued in time, which can effectively remind operators and users to take necessary measures to avoid potential safety accidents. This timeliness is crucial in ensuring safety. The threshold is set as the average value of the danger index in the historical monitoring data where there are abnormal conditions and warnings are issued. A reasonable danger threshold, combined with the danger index calculated in real time, can provide managers with a scientific decision-making basis. The setting of the threshold should be based on historical data analysis and expert opinions to ensure its scientificity and rationality, which will help improve management efficiency and enhance responsiveness.

[0070] See also Figure 2 The present invention further provides an abnormal warning system for a mobile energy storage charging pile, wherein the abnormal warning system for a mobile energy storage charging pile is used to implement the abnormal warning method for a mobile energy storage charging pile, comprising:

[0071] The data acquisition module is used to collect the current, voltage and charging status values ​​at multiple historical collection moments during the historical charging process of the charging pile, calculate the health status value of the charging pile at the corresponding historical collection moment, and integrate the current, voltage, charging status value and health status value into historical monitoring data;

[0072] The anomaly detection module builds a module for marking the status of the charging piles at each moment in the historical monitoring data based on the expert scoring method, and takes the historical monitoring data as input and the status mark of the charging piles at the corresponding moment as output, and builds an anomaly detection model based on the random forest algorithm, wherein the status mark is a normal state and an abnormal state;

[0073] The real-time monitoring and status recognition module is used to collect the current current, voltage and charging status values, calculate the health status value at the current moment, and input them into the trained anomaly detection model to obtain the current status mark;

[0074] A danger index acquisition module is used to calculate the danger index of the charging pile at the current moment based on the monitoring data at the current moment and in combination with the temperature and humidity data of the charging pile at the current moment if the charging pile is marked as an abnormal state at the current moment;

[0075] The early warning judgment and response module is used to compare the danger index of the charging pile at the current moment with the preset danger threshold to determine whether the charging pile needs to issue an early warning at the current moment.

[0076] The above formulas are all dimensionless and numerical calculations. The formula is a formula for the most recent real situation obtained by collecting a large amount of data and performing software simulation. The preset parameters in the formula are set by technicians in this field according to actual conditions.

[0077] The above embodiments may be implemented in whole or in part by software, hardware, firmware or any other combination thereof. When implemented by software, the above embodiments may be implemented in whole or in part in the form of a computer program product. Those skilled in the art may appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein may be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed by hardware or software methods depends on the specific application and design constraints of the technical solution.

[0078] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, and may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0079] The above description is only a specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any technician familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application.

Claims

1. A method for abnormal warning of a mobile energy storage charging pile, characterized in that: The specific steps include: Step 1: Collect the current, voltage and charging status values ​​at multiple historical collection times during the historical charging process of the charging pile, calculate the health status value of the charging pile at the corresponding historical collection time, and integrate the current, voltage, charging status value and health status value as historical monitoring data: Step 2: Based on the expert scoring method, the status of the charging piles at each moment in the historical monitoring data is marked, and the historical monitoring data is used as input, and the status mark of the charging pile at the corresponding moment is used as output. An anomaly detection model is constructed based on the random forest algorithm, and the status mark is normal state and abnormal state; Step 3: Collect the current, voltage and charging state values ​​at the current moment, calculate the health state value at the current moment, and input them into the trained anomaly detection model to obtain the state mark at the current moment; Step 4: If the charging pile is marked as abnormal at the current moment, the danger index of the charging pile at the current moment is calculated based on the monitoring data at the current moment and combined with the temperature and humidity data of the charging pile at the current moment; Step 5: Compare the danger index of the charging pile at the current moment with the preset danger threshold to determine whether the charging pile needs to issue an early warning at the current moment.

2. The abnormal warning method for a mobile energy storage charging pile according to claim 1 is characterized in that: The specific logic for collecting historical monitoring data of charging piles is: A multifunctional battery monitor is installed at the power input port of the charging pile to collect the current, voltage and charging status value data of the charging pile. A temperature sensor and a humidity sensor are installed at the power module position to collect the temperature and humidity data of the charging pile. The formula for obtaining the health status value of the charging pile is: Among them, HI represents the health status value, R is the internal impedance of the charging pile battery, and R max is the maximum normal impedance of the charging pile battery determined by testing, L is the number of charging cycles of the battery, and L max is the maximum number of cycles expected for the battery in normal use, SOC is the state of charge value, ω1, ω2, ω3 are the weights of the corresponding items, 0<ω3<ω2<ω1<1, and ω1+ω2+ω3=1; The current, voltage and charging status value data collected at multiple historical collection times during the historical charging process of the charging pile, as well as the calculated health status value, are standardized using the Z-score method, and the preprocessed data are stored as a sample data set.

3. The abnormal warning method for a mobile energy storage charging pile according to claim 2 is characterized in that: The specific logic for building anomaly detection models is: In normal state, mark A=1, and in abnormal state, mark A=0. The sample data set is divided into training set and test set with a division ratio of 7:

3. The historical monitoring data in the training set is used as input, and the state mark is used as the label. The random forest model in machine learning is selected to build the anomaly detection model, and then the historical monitoring data in the test set is input to evaluate the performance of the model. The evaluation indicators include accuracy, precision, and recall. If all changes in the evaluation indicators are less than 0.01 in five iterations, the model training is considered to be completed.

4. The abnormal warning method for a mobile energy storage charging pile according to claim 3 is characterized in that: The specific logic for calculating the danger index of the charging pile at the current moment is: The formula for obtaining the danger index of the charging pile at the current moment is: Where P represents the danger index of the charging pile at the current moment, I is the current collected by the sensor at the current moment, V is the voltage collected by the sensor at the current moment, T is the temperature collected by the sensor at the current moment, H is the humidity collected by the sensor at the current moment, and I max is the maximum current allowed by the charging pile, V max is the maximum voltage allowed by the charging pile, is the optimal operating temperature of the charging pile. is the optimal working humidity of the charging pile, β1, β2, β3, β4, β5, and β6 are the weight coefficients of their respective items, 0<β5=β6<β4<β3<β1=β2<1, and β1+β2+β3+β4+β5+β6=1.

5. The abnormal warning method for a mobile energy storage charging pile according to claim 4 is characterized in that: The specific logic for judging whether the charging pile needs to issue an early warning at the current moment is: Compare the danger index of the charging pile at the current moment with the preset danger threshold. If the danger index of the charging pile at the current moment is lower than the preset danger threshold, no warning will be issued temporarily. If the danger index of the charging pile at the current moment is higher than the preset danger threshold, a warning needs to be issued to remind that there is an abnormality in the charging of the charging pile at the current moment.

6. An abnormal warning system for a mobile energy storage charging pile, characterized in that: The abnormal warning system of a mobile energy storage charging pile is used to implement the abnormal warning method of a mobile energy storage charging pile according to any one of claims 1 to 5, comprising: The data acquisition module is used to collect the current, voltage and charging status values ​​at multiple historical collection moments during the historical charging process of the charging pile, calculate the health status value of the charging pile at the corresponding historical collection moment, and integrate the current, voltage, charging status value and health status value into historical monitoring data; The anomaly detection module builds a module for marking the status of the charging piles at each moment in the historical monitoring data based on the expert scoring method, and takes the historical monitoring data as input and the status mark of the charging piles at the corresponding moment as output, and builds an anomaly detection model based on the random forest algorithm, wherein the status mark is a normal state and an abnormal state; The real-time monitoring and status recognition module is used to collect the current current, voltage and charging status values, calculate the health status value at the current moment, and input them into the trained anomaly detection model to obtain the current status mark; A danger index acquisition module is used to calculate the danger index of the charging pile at the current moment based on the monitoring data at the current moment and in combination with the temperature and humidity data of the charging pile at the current moment if the charging pile is marked as an abnormal state at the current moment; The early warning judgment and response module is used to compare the danger index of the charging pile at the current moment with the preset danger threshold to determine whether the charging pile needs to issue an early warning at the current moment.

Images