A smart charging pile with anti-electricity theft function

Abstract

The intelligent charging pile with anti-electricity theft function of the present invention includes four key modules: charging interface module, data acquisition and processing module, data analysis and control module, and user interaction module; A, charging interface module: responsible for connecting two-wheeled electric vehicles, with built-in current and voltage sensors, and real-time monitoring of charging status; B, data acquisition and processing module: collects the original data of the charging interface module, including but not limited to current, voltage, and power; C, data analysis and control module: core component, sets the charging gear according to the initial power, and automatically upgrades the gear once the power rise is detected to avoid electricity theft; D, user interaction module: supports multiple payment methods, such as swiping cards or scanning codes, and provides real-time feedback on charging status and fees. The present invention effectively prevents users from stealing electricity by rotating equipment and other methods through integrated data acquisition and analysis functions, thereby protecting the economic interests of operators.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicle charging management, and in particular to an intelligent charging pile with an anti-electricity theft function. Background Art

[0002] With the acceleration of urbanization, more and more people choose electric vehicles as a means of transportation. The number of two-wheeled electric vehicles in my country exceeds 350 million, and one in four Chinese people owns a two-wheeled electric vehicle. At the same time, the charging problem has gradually become the focus of people's attention. Electric vehicle smart charging piles use low-power chargers, which can provide DC, AC and other charging methods, are compatible with most electric vehicles on the market, and can solve various problems of electric vehicle charging. Technological progress and innovation have brought opportunities to the electric vehicle charging pile market. With the continuous innovation of charging technology, new technologies such as fast charging and smart charging are gradually being applied to the electric vehicle charging pile market, making charging faster, safer and more intelligent.

[0003] However, with the popularization of electric vehicle charging piles, many "electricity theft" phenomena have also emerged, causing great trouble for the operation and maintenance party. Summary of the invention

[0004] In order to make up for the deficiencies of the prior art, the present invention provides an intelligent charging pile with an anti-electricity theft function designed specifically for two-wheeled electric vehicles.

[0005] The present invention is achieved through the following technical solutions:

[0006] An intelligent charging pile with an anti-electricity theft function is characterized by comprising four interrelated key modules: a charging interface module, a data acquisition and processing module, a data analysis and control module, and a user interaction module;

[0007] A. Charging interface module: responsible for connecting two-wheeled electric vehicles, with built-in current and voltage sensors to monitor the charging status in real time;

[0008] B. Data acquisition and processing module: collects raw data from the charging interface module, including but not limited to current, voltage and power;

[0009] C. Data analysis and control module: core component, sets the charging gear according to the initial power, and automatically upgrades the gear once the power rises to avoid power theft;

[0010] D. User interaction module: supports multiple payment methods, including swiping cards or scanning codes, and provides real-time feedback on charging status and fees.

[0011] Further preferably, the specific functions of each module of the smart charging pile with anti-electricity theft function include:

[0012] A. Charging interface module: It directly establishes a physical connection with the charging port of the two-wheeled electric vehicle, responsible for the transmission of power and monitoring of the charging process; it has built-in high-precision current and voltage sensors to monitor the charging current in real time, ensure the current stability during the charging process, avoid overcurrent or undercurrent, and provide accurate current data for the data analysis and control module, continuously monitor the charging voltage, ensure that the voltage level is suitable for the charging requirements of the electric vehicle battery, and avoid overvoltage damage to the battery; it is used to monitor the current and voltage parameters during the charging process in real time and transmit them to the data acquisition and processing module in real time;

[0013] B. Data acquisition and processing module: Receive, analyze and process the raw data transmitted in real time from the charging interface module, and perform real-time processing and analysis through built-in algorithms to determine the changes in charging power;

[0014] The data acquisition and processing module has a built-in high-performance microprocessor and uses a real-time calculation algorithm to calculate the charging power in real time based on the received current and voltage data; the power calculation formula is usually:

[0015] , (1)

[0016] Where P(t) represents the instantaneous power at time t, V(t) and I(t) represent the voltage and current values ​​at the same time respectively;

[0017] C. Data analysis and control module: processes the real-time data transmitted by the data acquisition and processing module, conducts in-depth analysis, and makes intelligent decisions based on the analysis results to optimize the charging process and prevent electricity theft;

[0018] D. User Interaction Module: Provides an intuitive, convenient and secure interactive experience, supports multiple payment methods such as swiping cards or scanning codes, displays real-time charging status and fees, and enhances user satisfaction and ease of use; the screen displays the charging status in real time, including key information such as the amount charged, remaining charging time, and current power, so that users can see the charging process at a glance; the system accurately calculates the fee based on the real-time electricity price and charging amount, and updates it in real time on the screen to ensure transparency and fairness; when the system detects any abnormality, the interface will immediately display the fault code and a brief description, and provide preliminary solution suggestions or a shortcut to contact customer service.

[0019] Further preferably, in B. data acquisition and processing module, in the smart charging pile, in order to prevent the current and voltage measurements from being interfered by various external factors, including electromagnetic noise and line impedance changes, which will directly affect the accuracy of power calculation; in order to ensure that accurate power can be obtained, a Kalman filter algorithm is introduced into the data acquisition and processing module to realize real-time filtering of these measurement data, thereby obtaining purer current and voltage values;

[0020] First, based on the state estimation at the previous moment and the dynamic model of the system, the current and voltage states at the current moment are predicted. Based on the dynamic model of the system, the optimal state estimation value at the previous moment is used. and the control input uk of the system (if any), predict the current state , calculation method:

[0021] (2)

[0022] Where, Fk is the state transfer matrix; Bk is the control input matrix;

[0023] Prediction covariance update: Update the covariance matrix of the state estimate , to reflect the uncertainty of the forecast;

[0024] (3)

[0025] Where Qk is the process noise covariance matrix; is the predicted state covariance matrix; is the transposed matrix of the state transfer matrix;

[0026] Then receive the actual current and voltage values ​​from the sensor and calculate the Kalman gain Kk:

[0027] (4)

[0028] Hk is the observation matrix, which maps the true state space to the observation space; It is the transposed matrix of the observation matrix; Rk is the covariance matrix of the observation noise;

[0029] The Kalman gain Kk determines the weight of the new measurement value when updating the state estimate; finally, the Kalman gain, the predicted value and the measured value are used to update the state estimate at the current moment, where the state is updated:

[0030] (5)

[0031] is the state estimate at the current moment after the update; yk is the difference between the measured value and the predicted value, also called the residual;

[0032] Covariance update:

[0033] (6)

[0034] I represents the identity matrix; Pk|k is the updated covariance matrix.

[0035] The data after filtering will be purer and can be used for power calculation. Due to the real-time and accuracy of the Kalman filter, it can significantly improve the power calculation accuracy of the charging pile, thereby optimizing the charging process and improving the user experience.

[0036] Further preferably, the data analysis and control module specifically includes the following functions:

[0037] (C-1). Through continuous monitoring and calculation of power data, the data analysis and control module can analyze the trend of power changes over time; this step is crucial for identifying abnormal charging behavior. When a sudden and significant increase in power is detected, it may mean that additional loads are connected, which is an important sign of power theft;

[0038] (C-2). The data acquisition and processing module dynamically sets the threshold of power change based on historical data and current charging mode; when the real-time power exceeds this threshold, the early warning mechanism is triggered to notify the data analysis and control module for further processing;

[0039] (C-3). Label the data and classify the electricity theft into multiple levels (such as minor, moderate, and severe) according to the severity. This requires manual review and other methods to make preliminary judgments. First, use the decision tree algorithm to preliminarily classify the electricity theft. The decision tree divides the data set by recursively selecting the optimal features. Common selection criteria include information gain (based on entropy), information gain ratio, Gini index, etc. Information gain:

[0040] (7)

[0041] Where IG(T, a) represents the information gain of feature a, Entropy(T) represents the entropy of data set T, Tv represents the subset when feature a takes the value v; ∑v∈values ​​(a) represents the summation operation of all possible values ​​v of feature a; here values ​​(a) refers to the set of all possible values ​​of feature a.

[0042] (8)

[0043] K is the total number of categories, and pk is the proportion of the kth category in T.

[0044] The feature with the largest information gain is selected as the split feature of the current node, and then this process is recursively repeated for each subset until the stopping condition is met (such as reaching the preset maximum depth, the minimum number of leaf nodes, or the information gain is lower than the threshold). In order to overcome the overfitting problem that may exist in a single decision tree and improve the generalization ability of the model, the random forest algorithm can also be used; for classification problems, the random forest takes the mode of the output of all decision trees as the final prediction result; that is, for a given input sample, each decision tree will give a category prediction, and the random forest will select the category with the most occurrences as the final prediction; each decision tree will randomly select some features and samples during training to increase the diversity of the model.

[0045] (C-4). The data analysis and control module has a built-in anomaly detection algorithm, which continuously monitors real-time power data and identifies any power fluctuations that exceed the preset normal range by analyzing historical trends and current status. Such fluctuations are caused by a variety of factors, including equipment failure, illegal operations such as power theft, external interference, or internal system anomalies. The anomaly detection algorithm sets the highest power threshold for each power level. When the anomaly detection algorithm repeatedly or continuously captures power fluctuations that exceed the threshold, it determines the level of power theft and quickly initiates a series of automated response procedures, including but not limited to: the system automatically adjusts the power supply according to the severity of the power theft. For minor power theft, the module evaluates the current power consumption. If the detected anomaly is caused by a sudden increase in power demand, it will automatically adjust the charging level to a higher level and send a warning notification to match the actual power consumption, preventing the power theft from causing economic losses to the operator. For moderate power theft, the charging level is further adjusted and the monitoring frequency is increased. For severe power theft, safety protection measures are initiated, taking into account the potential safety hazards caused by overload, including damage caused by overheating of the equipment or potential fire risks. The abnormal detection algorithm also integrates a safety protection mechanism; once it is found that the power fluctuation exceeds the overload threshold and may cause overload, overload protection is adopted, and the power supply is automatically cut off and reported to the relevant departments when necessary to prevent accidents, ensure personnel safety and equipment integrity, and prevent equipment damage or potential fire risks; record abnormal events, in order to ensure that all key stakeholders can understand the operating status of the system in a timely manner, especially when abnormal events occur, the module is designed with detailed event recording and alarm functions; after each abnormal detection is triggered, the system will record the nature, time, location and possible causes of the event in detail, and generate corresponding alarm information; in addition to traditional screen display and SMS notification, this information uses modern communication technologies such as the Internet or mobile networks to add instant messaging (such as WeChat, APP push) functions to ensure that users and relevant maintenance personnel can receive power theft warnings or processing results in the first time; ensure that they can take prompt action to evaluate and handle abnormal situations; there is also an interactive feedback mechanism: users can provide feedback through the interactive module, including complaints, confirmation of power theft, etc., to enhance user experience and system transparency.

[0046] The beneficial effect of the present invention is that the intelligent charging pile system can detect the "electricity theft" behavior by itself, and can formulate different charging levels at any time according to the changes in parameters such as power, so as to effectively avoid the impact of the "electricity theft" behavior on the operation and maintenance party.

[0047] The present invention is an intelligent charging pile system designed for two-wheeled electric vehicles. Through the integrated data collection and analysis function, it effectively prevents users from stealing electricity by rotating equipment, etc., and protects the economic interests of operators. It solves various drawbacks of traditional charging piles and greatly expands its application scope.

[0048] The data analysis and control module sets an initial charging gear according to the initial charging power, and automatically upgrades the charging gear to a higher power level when a power increase is detected to ensure accurate charging.

[0049] The user interaction module supports swiping a card, scanning a code, and entering a number in a mini program to start the charging service, while displaying real-time charging status and cost information, charging time, charging power and other information to enhance user transparency.

[0050] The charging pile is equipped with an intelligent identification system to prevent external devices from accessing, so as to further enhance the anti-electricity theft function. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 It is a schematic diagram of the workflow of the present invention. DETAILED DESCRIPTION

[0052] The accompanying drawings show a specific embodiment of the present invention.

[0053] The intelligent charging pile with anti-electricity theft function of the present invention includes four interrelated key modules: a charging interface module, a data acquisition and processing module, a data analysis and control module, and a user interaction module;

[0054] A. Charging interface module: It directly establishes a physical connection with the charging port of the two-wheeled electric vehicle and is responsible for the transmission of power and monitoring of the charging process. It has built-in high-precision current and voltage sensors to monitor the charging current in real time, ensure the current stability during the charging process, avoid overcurrent or undercurrent, and provide accurate current data to the data analysis and control module, continuously monitor the charging voltage, ensure that the voltage level is suitable for the charging requirements of the electric vehicle battery, and avoid overvoltage damage to the battery. It is used to monitor the current, voltage and other parameters during the charging process in real time, and transmit them to the data acquisition and processing module in real time;

[0055] B. Data acquisition and processing module: Receive, analyze and process the raw data transmitted in real time from the charging interface module, and perform real-time processing and analysis through built-in algorithms to determine the changes in charging power.

[0056] The module has a built-in high-performance microprocessor and uses a real-time calculation algorithm to calculate the charging power in real time based on the received current and voltage data. The power calculation formula is usually:

[0057] , (1)

[0058] Among them, P(t) represents the instantaneous power at time t, V(t) and I(t) represent the voltage and current values ​​at the same time, respectively.

[0059] For example, if at a certain moment (t), the voltage V(t) measured by the smart charging pile is 220V and the current I(t) is 10A, then the instantaneous power P(t) at this moment is calculated as: P(t) = V(t)×I(t)= 2200W.

[0060] Further preferably, in the smart charging pile, the measurement of current and voltage is often interfered by various external factors, such as electromagnetic noise, line impedance changes, etc., which directly affect the accuracy of power calculation. In order to ensure that accurate power can be obtained, the Kalman filter algorithm is introduced into the data acquisition and processing module to realize real-time filtering of these measurement data, thereby obtaining purer current and voltage values.

[0061] First, based on the state estimation at the previous moment and the dynamic model of the system, the current and voltage states at the current moment are predicted. Based on the dynamic model of the system, the optimal state estimation value at the previous moment is used. and the control input uk of the system (if any), predict the current state , calculation method:

[0062] (2)

[0063] Among them, Fk is the state transfer matrix; Bk is the control input matrix.

[0064] For example, the optimal state estimate of the current at the previous moment is is 10A, the state transfer matrix Fk is the unit matrix I, and the control input uk of the system is 0, that is, there is no external control input, then the predicted optimal state estimate is:

[0065] , still 10A.

[0066] Prediction covariance update: Update the covariance matrix of the state estimate , to reflect the uncertainty of the forecast;

[0067] (3)

[0068] Where Qk is the process noise covariance matrix; is the predicted state covariance matrix; is the transposed matrix of the state transfer matrix;

[0069] Assuming the process noise covariance matrix Qk is 0.5, the predicted state covariance matrix is 1, then the covariance matrix of the updated state estimate is: .

[0070] Then receive the actual current and voltage values ​​from the sensor and calculate the Kalman gain Kk:

[0071] (4)

[0072] Hk is the observation matrix, which maps the true state space to the observation space; It is the transposed matrix of the observation matrix; Rk is the covariance matrix of the observation noise;

[0073] Assuming that the covariance matrix Rk of the observation noise is 0.1, the Kalman gain can be calculated:

[0074] .

[0075] The Kalman gain Kk determines the weight of the new measurement value when updating the state estimate; finally, the Kalman gain, the predicted value and the measured value are used to update the state estimate at the current moment, where the state is updated:

[0076] (5)

[0077] is the state estimate at the current moment after the update; yk is the difference between the measured value and the predicted value, also called the residual;

[0078] Covariance update:

[0079] (6)

[0080] I represents the identity matrix; Pk|k is the updated covariance matrix.

[0081] Assuming that the current measured this time is 9.8A, calculate the last updated current value based on the above assumed data:

[0082] .

[0083] The data after filtering will be purer and can be used for power calculation. Due to the real-time and accuracy of the Kalman filter, it can significantly improve the power calculation accuracy of the charging pile, thereby optimizing the charging process and improving the user experience.

[0084] C. Data analysis and control module: processes the real-time data transmitted by the data acquisition and processing module, conducts in-depth analysis, and makes intelligent decisions based on the analysis results to optimize the charging process and prevent electricity theft.

[0085] C-1. Through continuous monitoring and calculation of power data, the module can analyze the trend of power changes over time. This step is crucial to identifying abnormal charging behavior. For example, when a sudden and significant increase in power is detected, it may mean that additional loads are connected, which is an important sign of power theft.

[0086] C-2. The data acquisition and processing module dynamically sets the power change threshold based on historical data and the current charging mode. When the real-time power exceeds this threshold, the early warning mechanism is triggered to notify the data analysis and control module for further processing.

[0087] C-3. Label the data and classify the electricity theft into multiple levels (such as minor, moderate, and severe) according to the severity. This requires manual review and other methods to make preliminary judgments. First, use the decision tree algorithm to preliminarily classify the electricity theft. The decision tree divides the data set by recursively selecting the optimal features. Common selection criteria include information gain (based on entropy), information gain ratio, Gini index, etc. Information gain:

[0088] (7)

[0089] Where Entropy(T) represents the entropy of the data set T, IG(T, a) represents the information gain of feature a, Tv represents the subset when feature a takes the value v; ∑v∈values ​​(a) represents the summation operation of all possible values ​​v of feature a; here values ​​(a) refers to the set of all possible values ​​of feature a.

[0090] Suppose we have a data set T about electricity theft, which contains two types of samples: electricity theft (positive class) and non-electricity theft (negative class). The information entropy Entropy(T) of data set T can be calculated as follows:

[0091] (8)

[0092] K is the total number of categories, and pk is the proportion of the kth category in T.

[0093] Assume that there are 100 samples in the data set T, of which 60 are electricity theft (positive class) and 40 are non-electricity theft (negative class);

[0094] but .

[0095] Now consider a feature a (e.g. “electricity usage”), which has two possible values: high (v=1) and low (v=2);

[0096] Assume that according to the value of feature a, the data set T is divided into two subsets T1 and T2. T1 (high power consumption) contains 30 power theft samples and 10 non-power theft samples. T2 (low power consumption) contains 30 power theft samples and 30 non-power theft samples; therefore,

[0097]

[0098]

[0099] Next calculate

[0100]

[0101] This information gain value indicates that after using feature a (electricity consumption) to divide the data set D, the information entropy is reduced by about 0.047.

[0102] The feature with the maximum information gain is selected as the split feature of the current node, and then this process is recursively repeated for each subset until the stopping condition is met (such as reaching the preset maximum depth, the minimum number of leaf nodes, or the information gain is lower than the threshold). In order to overcome the overfitting problem that may exist in a single decision tree and improve the generalization ability of the model, the random forest algorithm can also be used. For classification problems, the random forest takes the mode of the output of all decision trees as the final prediction result. That is, for a given input sample, each decision tree will give a category prediction, and the random forest will select the category with the most occurrences as the final prediction. Each decision tree will randomly select some features and samples during training to increase the diversity of the model.

[0103] C-4. The data analysis and control module has a built-in anomaly detection algorithm that can identify power fluctuations beyond the normal range. Once an anomaly is detected, the module will immediately initiate a response procedure, including but not limited to: adjusting the charging gear to a higher level to match the actual power consumption to prevent power theft from causing economic losses to the operator, initiating safety protection measures such as overload protection, cutting off the power supply to prevent equipment damage or potential fire risks, recording abnormal events, generating alarm information, and sending it to the remote monitoring center or the user's mobile phone APP through the network communication module to promptly notify relevant personnel. The details are as follows.

[0104] The data analysis and control module has a built-in anomaly detection algorithm, which continuously monitors real-time power data and identifies any power fluctuations that exceed the preset normal range by analyzing historical trends and current status. Such fluctuations may be caused by a variety of factors, including equipment failure, illegal operations such as power theft, external interference, or internal system anomalies. The anomaly detection algorithm sets the highest power threshold for each power level. When the anomaly detection algorithm repeatedly or continuously captures power fluctuations that exceed the threshold, it determines the level of power theft. It quickly initiates a series of automated response procedures, including but not limited to: Depending on the severity of the power theft, the system can automatically adjust the power supply. For example, for minor power theft, the module will evaluate the current power consumption. If the detected anomaly is caused by a sudden increase in power demand, it will automatically adjust the charging level to a higher level and send a warning notification to match the actual power consumption to prevent the power theft from causing economic losses to the operator; for moderate power theft, the charging level is further adjusted and the monitoring frequency is increased; for severe power theft, safety protection measures are initiated. Considering the potential safety hazards caused by overload, such as damage caused by overheating of the equipment or potential fire risks, the anomaly detection algorithm also integrates a safety protection mechanism. Once it is found that the power fluctuation exceeds the overload threshold and may cause overload, overload protection is taken, and the power supply is automatically cut off and reported to the relevant departments when necessary. , prevent accidents from happening, ensure personnel safety and equipment integrity, and prevent equipment damage or potential fire risks; record abnormal events. In order to ensure that all key stakeholders can understand the operating status of the system in a timely manner, especially when abnormal events occur, the module is designed with detailed event recording and alarm functions. After each abnormal detection is triggered, the system will record the nature, time, location and possible causes of the event in detail, and generate corresponding alarm information. In addition to traditional screen display and SMS notifications, these information uses modern communication technologies such as the Internet or mobile networks to add instant messaging (such as WeChat, APP push) functions to ensure that users and relevant maintenance personnel can receive power theft warnings or processing results as soon as possible. Ensure that they can take prompt action to evaluate and handle abnormal situations. There is also an interactive feedback mechanism: users can provide feedback through the interactive module, including complaints, confirmation of power theft, etc., to enhance user experience and system transparency.

[0105] D. User interaction module: Provides an intuitive, convenient and safe interactive experience, supports multiple payment methods such as swiping cards or scanning codes, displays real-time charging status and fees, and enhances user satisfaction and convenience. The screen displays the charging status in real time, including key information such as the amount charged, remaining charging time, and current power, so that users can see the charging process at a glance. The system accurately calculates the fee based on the real-time electricity price and charging amount, and updates it in real time on the screen to ensure transparency and fairness. When the system detects any abnormality, the interface will immediately display the fault code and a brief description, and provide preliminary solution suggestions or a shortcut to contact customer service.

[0106] Workflow: The user starts the charging pile by swiping a card or scanning a QR code, and the system immediately records the initial power. The user or system selects the appropriate charging gear according to the vehicle type and charging needs. The data analysis and control module continuously monitors the power changes during the charging process. If a significant increase in power is detected, the system automatically upgrades the gear and adjusts the billing standard to ensure that the cost matches the actual power resources consumed to prevent electricity theft. The user interaction module updates and displays charging status information in real time, including but not limited to the amount charged, remaining charging time, current power and accumulated costs, so that users can keep abreast of the charging progress and enhance service transparency. It also provides notification functions for abnormal situations, such as system failures, fee changes or emergency power outages, to ensure that users can respond to various situations in a timely manner and reduce the inconvenience caused by uncertainty.

Claims

1. A smart charging pile with anti-electricity theft function, characterized by: It includes four interrelated key modules: charging interface module, data acquisition and processing module, data analysis and control module, and user interaction module; A. Charging interface module: responsible for connecting two-wheeled electric vehicles, with built-in current and voltage sensors to monitor the charging status in real time; B. Data acquisition and processing module: collects the raw data of the charging interface module, including current, voltage and power; C. Data analysis and control module: core component, sets the charging gear according to the initial power, and automatically upgrades the gear once the power rises to avoid power theft; D. User interaction module: supports multiple payment methods, including swiping cards or scanning codes, and provides real-time feedback on charging status and fees; Among them, B, data acquisition and processing module, specifically includes the following functions: In smart charging piles, in order to prevent the current and voltage measurements from being interfered by various external factors, including electromagnetic noise and line impedance changes, which will directly affect the accuracy of power calculation; in order to ensure that accurate power can be obtained, the Kalman filter algorithm is introduced into the data acquisition and processing module to achieve real-time filtering of these measurement data, thereby obtaining purer current and voltage values; First, based on the state estimation at the previous moment and the dynamic model of the system, the current and voltage states at the current moment are predicted. Based on the dynamic model of the system, the optimal state estimation value at the previous moment is used. and the system control input u k , predict the current state Calculation method: Among them, F k is the state transfer matrix; B k is the control input matrix; Prediction covariance update: Update the covariance matrix P of the state estimate k|k-1 , to reflect the uncertainty of the forecast; Among them, Q k is the process noise covariance matrix; P k-1|k-1 is the predicted state covariance matrix; Yes Status The transpose of the transfer matrix; Then receive the actual current and voltage values ​​from the sensor and calculate the Kalman gain K k : H k is the observation matrix, which maps the true state space to the observation space; It is the transposed matrix of the observation matrix; R k is the covariance matrix of the observation noise; Kalman gain K k It determines the weight of the new measurement value in updating the state estimate, and finally uses the Kalman gain, the predicted value and the measured value to update the state estimate at the current moment, where the state is updated: is the state estimate at the current time after update; y k is the difference between the measured and predicted values, also called the residual; Covariance update: P k|k =(I-K k H k )P k|k-1 , (6) I represents the identity matrix; P k|k It is the updated covariance matrix; The data after filtering will be purer and can be used for power calculation. Due to the real-time and accuracy of the Kalman filter, it can significantly improve the power calculation accuracy of the charging pile, thereby optimizing the charging process and improving the user experience.

2. The smart charging pile with anti-electricity theft function according to claim 1 is characterized in that: The specific functions of each module include: A. Charging interface module: It directly establishes a physical connection with the charging port of the two-wheeled electric vehicle, responsible for the transmission of power and monitoring of the charging process; it has built-in high-precision current and voltage sensors to monitor the charging current in real time, ensure the current stability during the charging process, avoid overcurrent or undercurrent, and provide accurate current data for the data analysis and control module, continuously monitor the charging voltage, ensure that the voltage level is suitable for the charging requirements of the electric vehicle battery, and avoid overvoltage damage to the battery; it is used to monitor the current and voltage parameters during the charging process in real time and transmit them to the data acquisition and processing module in real time; B. Data acquisition and processing module: Receive, analyze and process the raw data transmitted in real time from the charging interface module, and perform real-time processing and analysis through built-in algorithms to determine the changes in charging power; The data acquisition and processing module has a built-in high-performance microprocessor and uses a real-time calculation algorithm to calculate the charging power in real time based on the received current and voltage data; the power calculation formula is: P(t)=V(t)×I(t), (1) Where P(t) represents the instantaneous power at time t, V(t) and I(t) represent the voltage and current values ​​at the same time, respectively; C. Data analysis and control module: processes the real-time data transmitted by the data acquisition and processing module, conducts in-depth analysis, and makes intelligent decisions based on the analysis results to optimize the charging process and prevent electricity theft; D. User Interaction Module: Provides an intuitive, convenient and secure interactive experience, supports multiple payment methods such as swiping cards or scanning codes, displays real-time charging status and fees, and enhances user satisfaction and ease of use; the screen displays the charging status in real time, including key information such as the amount charged, remaining charging time, and current power, so that users can see the charging process at a glance. The system accurately calculates the fee based on the real-time electricity price and charging amount, and updates it in real time on the screen to ensure transparency and fairness. When the system detects any abnormality, the interface will immediately display the fault code and a brief description, and provide preliminary solution suggestions or a shortcut to contact customer service.

3. The smart charging pile with anti-electricity theft function according to claim 2 is characterized in that: Data analysis and control module, including the following functions: (C-1). Through continuous monitoring and calculation of power data, the data analysis and control module can analyze the trend of power changes over time; this step is crucial for identifying abnormal charging behavior. When a sudden and significant increase in power is detected, it means that additional loads are connected, which is an important sign of power theft; (C-2). The data acquisition and processing module dynamically sets the threshold of power change based on historical data and the current charging mode; when the real-time power exceeds this threshold, the early warning mechanism is triggered to notify the data analysis and control module for further processing; (C-3). Label the data and classify the electricity theft into mild, medium and severe levels according to the severity. This requires manual review for preliminary judgment. First, use the decision tree algorithm to preliminarily classify the electricity theft. The decision tree divides the data set by recursively selecting the optimal features. Common selection criteria include information gain based on entropy, information gain ratio, and Gini index. Information gain: Among them, IG(T, a) represents the information gain of feature a, Entropy(T) represents the entropy of data set T, and T v represents the subset when the feature a takes the value v; ∑ v∈values(a) represents the summation operation of all possible values ​​v of feature a; here values(a) refers to the set of all possible values ​​of feature a; K is the total number of categories, p k is the proportion of the kth category in T; The feature with the largest information gain is selected as the split feature of the current node, and then this process is recursively repeated for each subset until the stopping condition is met. If the preset maximum depth, minimum number of leaf nodes or information gain is lower than the threshold, in order to overcome the overfitting problem of a single decision tree and improve the generalization ability of the model, the random forest algorithm can also be used; for classification problems, the random forest takes the mode of the output of all decision trees as the final prediction result; that is, for a given input sample, each decision tree will give a category prediction, and the random forest will select the category with the most occurrences as the final prediction; each decision tree will randomly select some features and samples during training, thereby increasing the diversity of the model; (C-4). The data analysis and control module has a built-in anomaly detection algorithm that continuously monitors real-time power data and identifies any power fluctuations that exceed the preset normal range by analyzing historical trends and current status. Such fluctuations are caused by a variety of factors, including equipment failure, power theft due to illegal operation, external interference or internal system anomalies. The anomaly detection algorithm sets the highest power threshold for each power level. When the anomaly detection algorithm repeatedly or continuously captures power fluctuations that exceed the threshold, it determines the level of power theft and quickly initiates a series of automated response procedures, including: the system automatically adjusts the power supply according to the severity of the power theft; (1) For minor power theft, the module will evaluate the current power consumption. If the detected anomaly is caused by a sudden increase in power demand, it will automatically adjust the charging level to a higher level and send a warning notification to match the actual power consumption, preventing the power theft from causing economic losses to the operator; (2) For moderate power theft, the charging level is further adjusted and the monitoring frequency is increased; (3) For severe power theft, safety protection measures are initiated, taking into account the safety hazards caused by overload, including equipment overheating. In order to prevent damage or potential fire risks, the anomaly detection algorithm also integrates a safety protection mechanism; once it is found that the power fluctuation exceeds the overload threshold and causes overload, overload protection is adopted, the power supply is automatically cut off and reported to the relevant departments to prevent accidents, ensure personnel safety and equipment integrity, so as to prevent equipment damage or potential fire risks; record abnormal events. In order to ensure that all key stakeholders can understand the operating status of the system in a timely manner, when an abnormal event occurs, the module is designed with detailed event recording and alarm functions; after each abnormal detection is triggered, the system will record the nature, time, location and cause of the event in detail, and generate corresponding alarm information; in addition to traditional screen display and SMS notifications, these information uses modern communication technology Internet or mobile network to add instant messaging functions, including WeChat push and APP push, to ensure that users and relevant maintenance personnel can receive power theft warnings or processing results in the first time; ensure that they can take prompt action to evaluate and handle abnormal situations; there is also an interactive feedback mechanism: users can provide feedback through the interactive module, including complaints and confirmation of power theft, to enhance user experience and system transparency.

4. A charging method for a smart charging pile with an anti-electricity theft function according to claim 1, characterized in that: The following steps are involved: The user starts the charging station by swiping a card or scanning a QR code, and the system immediately records the initial power. The user or system selects the appropriate charging gear according to the vehicle type and charging needs; the data analysis and control module continuously monitors the power changes during the charging process; if a significant increase in power is detected, the system automatically upgrades the gear and adjusts the billing standard to ensure that the cost matches the actual power resources consumed to prevent electricity theft; the user interaction module updates and displays the charging status information in real time, including the charged amount, remaining charging time, current power and accumulated costs, so that users can understand the charging progress at any time and enhance service transparency. It also provides notification functions for abnormal situations, including system failures, fee changes or emergency power outages, to ensure that users can respond to various situations in a timely manner and reduce the inconvenience caused by uncertainty.

Images