Remote control method and system for charging pile

Abstract

The invention discloses a remote control method and system for a charging pile, and relates to the technical field of smart power grids and charging, and the method comprises the steps: collecting and preprocessing power grid data, and building a power grid data set according to the power grid data; constructing an adaptive charging scheduling model by using historical data, and predicting a future power grid load change trend based on power grid data; recording the current state of the power grid in real time by using an intelligent electric meter, and identifying an optimal charging strategy and recording charging data according to a prediction result and the current state of the power grid; constructing a charging fault detection model by using the charging data, detecting a charging fault, and automatically notifying operation and maintenance personnel to repair; and after the repair is completed, updating the fault and maintenance records, and optimizing the fault detection model and the self-adaptive charging scheduling model by using the fault and maintenance records. Through data integrity index calculation, quantitative evaluation of power grid data transmission quality is realized, data integrity and reliability are ensured, and system stability is improved.

Description

Technical Field

[0001] The present invention relates to the field of smart grid and charging technology, and in particular to a remote control method and system for a charging pile. Background Art

[0002] In recent years, with the popularization of electric vehicles (EVs) and the advancement of smart grid technology, intelligent management and remote control of charging piles, as the key node connecting the power system and electric vehicles, have become a research hotspot. Traditional charging piles mostly adopt local control methods, users need to manually start the charging process, and lack effective management of grid load. However, with the introduction of the concept of smart grids and the development of technologies, remote control methods have gradually become an important means to achieve efficient, stable and environmentally friendly charging. The widespread use of smart meters makes it possible to monitor the state of the grid in real time. Combined with big data analysis and machine learning algorithms, a more accurate load forecasting model can be constructed to optimize the charging scheduling strategy. In addition, the application of Internet of Things (IoT) technology and communication protocols (such as DLMS / COSEM) has further enhanced the interconnection and interoperability of charging pile systems, providing a solid technical foundation for remote monitoring and control.

[0003] Despite this, existing technologies still face many challenges in achieving remote control of charging piles. On the one hand, traditional methods are difficult to effectively handle complex and changing power grid environments, especially when facing peak loads, and cannot flexibly adjust charging strategies to avoid impacting the power grid. On the other hand, the existing fault detection mechanism is not perfect. Once a fault occurs, the operation and maintenance response time is long, affecting user experience and service reliability. Therefore, it is particularly important to develop a remote control method that can adapt to dynamically changing power grid conditions and has efficient fault detection and repair functions. Summary of the invention

[0004] In view of the above existing problems, the present invention is proposed.

[0005] Therefore, the present invention provides a remote control method for a charging pile to solve the problem that the traditional method cannot flexibly adjust the charging strategy during peak load and the fault detection mechanism is imperfect, resulting in slow operation and maintenance response.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, the present invention provides a remote control method for a charging pile, which includes collecting and preprocessing grid data, establishing a grid data set based on the grid data; using historical data to build an adaptive charging scheduling model, and based on the grid data, predicting future grid load change trends; using smart meters to record the current state of the grid in real time, and identifying the optimal charging strategy and recording the charging data based on the prediction results and the current state of the grid; using the charging data to build a charging fault detection model, detect charging faults, and automatically notify operation and maintenance personnel to repair them; updating the fault and maintenance records after the repair is completed, and using the fault and maintenance records to optimize the fault detection model and the adaptive charging scheduling model.

[0008] As a preferred solution of the remote control method of the charging pile of the present invention, the specific steps of collecting and preprocessing the power grid data and establishing a power grid data set according to the power grid data are as follows:

[0009] Select high-precision smart meters that support the DLMS / COSEM protocol and install them at key nodes of the power grid. After configuring communication parameters, perform calibration and testing, and enable the data logging function to collect voltage, current, and power.

[0010] The collected power grid data is transmitted to the central server in real time, and the data integrity is calculated using the following formula:

[0011]

[0012] Where I is the data integrity index, L is the packet loss rate, D is the average data delay, α is the impact coefficient of the packet loss rate, β is the impact coefficient of the data delay, γ is the threshold parameter, and e is the base of the natural logarithm;

[0013] Check the data structure of power grid data, remove duplicate values, process missing values, detect and correct outliers, and standardize power grid data on the central server;

[0014] Integrate the preprocessed power grid data, build a power grid data set and store it in the database.

[0015] As a preferred solution of the remote control method of the charging pile of the present invention, the specific steps of constructing the adaptive charging scheduling model are as follows:

[0016] Import the power grid data into the Python environment for preprocessing, use the ADF test to ensure data stability, and divide it into training and test sets;

[0017] Select ARIMA as the adaptive charging scheduling model, use Pandas' acf and pacf functions to generate ACF and PACF plots for stationary data, optimize parameter selection through grid search, use AIC as the evaluation metric, initialize the adaptive charging scheduling model with the selected parameters, and fit the model on the training set. Use Python's statsmodels library to implement the adaptive charging scheduling model, verify the performance of the adaptive charging scheduling model on the test set, calculate the prediction error, and analyze the residual plot;

[0018] A seasonal adjustment factor λ is introduced to optimize the adaptive charging scheduling model. The specific formula is as follows:

[0019]

[0020] in, is the predicted value of the power grid load at time t, c is a constant term, φ i is the autoregressive coefficient, θ j is the moving average coefficient, ∈ t is the error term, S t is the seasonal component, λ is the seasonal adjustment factor, p is the autoregressive order, i is the index of the autoregressive term, j is the index of the moving average term, q is the moving average order, t is the time index, and X t-i is the actual observed value in the past i.

[0021] As a preferred solution of the remote control method of the charging pile of the present invention, the following specific steps are taken to predict the future load change trend of the power grid based on the power grid data:

[0022] The grid data was verified through stationarity test, ACF and PACF plotting, parameter selection, setting seasonal adjustment factor λ and adaptive charging scheduling model. The trained adaptive charging scheduling model was used to predict the future grid load change trend. The expression is:

[0023]

[0024] in, For the future trend of grid load changes, is the predicted value of the power grid load at time t, is the predicted value of the power grid load at time t-1.

[0025] As a preferred solution of the remote control method of the charging pile of the present invention, wherein: the smart meter is used to record the current state of the power grid in real time, and the optimal charging strategy is identified and the charging data is recorded according to the prediction results and the current state of the power grid. The specific steps are as follows:

[0026] Real-time grid status information is obtained from smart meters. At the same time, based on the grid load forecast for a period of time in the future, grid load smoothing and charging efficiency optimization functions are defined. The specific formula is as follows:

[0027]

[0028] Among them, F(t) is the charging strategy fitness value at time t, X t is the actual grid load value at time t, X max is the maximum load value in historical data, E k (t) is the actual charging energy of the kth charging pile at time t, E max,k is the maximum charging energy of the kth charging pile, N is the total number of charging piles, Q1 is the weight coefficient of load smoothing, Q2 is the weight coefficient of charging efficiency, Q3 is the weight coefficient of future load change trend, is the future grid load change trend, k is the charging pile index;

[0029] Initialize the genetic algorithm parameters and generate the initial population, calculate the fitness value of each individual, generate a new generation of population through selection, crossover and mutation operations according to the fitness value of each individual, iterate and optimize until the termination condition is met, and finally identify the charging strategy with the largest fitness value as the optimal charging strategy and execute the optimal charging strategy, while recording the actual charging situation.

[0030] As a preferred solution of the remote control method of the charging pile of the present invention, wherein: the charging data is used to build a charging fault detection model, detect charging faults, and automatically notify the operation and maintenance personnel to repair them. The specific steps are as follows:

[0031] Use the Pandas library to clean and preprocess the power grid data and divide the data set into training and test sets;

[0032] Random forest is selected as the charging fault detection model. The charging fault model is trained using the training set data. A fault probability prediction factor is introduced. The specific formula is as follows:

[0033]

[0034] Among them, P f is the probability of failure, b is the bias term, ζ is the failure probability prediction factor, M is the total number of features, m is the feature index, and w m is the weight of the mth feature, x m is the value of the mth feature;

[0035] Construct a classifier based on random forest, the expression is as follows:

[0036]

[0037] Among them, R is the fault detection result, P f is the probability of failure, T is the set threshold;

[0038] The trained charging fault model is deployed to the real-time monitoring system to perform fault detection on the real-time collected data. When a charging fault is detected, an alarm is automatically sent to the operation and maintenance team, and the fault location and preliminary diagnostic information are provided.

[0039] As a preferred solution of the remote control method of the charging pile of the present invention, wherein: after the repair is completed, the fault and maintenance records are updated, and the fault detection model and the adaptive charging scheduling model are optimized by using the fault and maintenance records. The specific steps are as follows:

[0040] Record fault information in the database and record the response time and handling status of the operation and maintenance team. After the repair is completed, update the fault record, close the alarm, review the performance of the entire charging scheduling system, and collect the latest operating data from the real-time monitoring system;

[0041] The performance of the existing charging fault detection model and adaptive charging scheduling model is evaluated using the collected operating data, and the accuracy, recall rate, F1 score and root mean square error are calculated. Based on the collected data set, the charging fault detection model and the adaptive charging scheduling model are retrained, the parameters are adjusted, and a new feature selection method is introduced. The cross-validation method is used to further adjust the parameters of the charging fault detection model and the adaptive charging scheduling model, and the fault detection model and the adaptive charging scheduling model are optimized.

[0042] In a second aspect, the present invention provides a remote control system for a charging pile, comprising a data acquisition module, an adaptive charging scheduling module, an optimal charging strategy identification module, a charging fault detection module and a model optimization module; the data acquisition module is used to collect and pre-process power grid data, and establish a power grid data set according to the power grid data; the adaptive charging scheduling module is used to construct an adaptive charging scheduling model using historical data, and predict future power grid load change trends based on the power grid data; the optimal charging strategy identification module is used to use a smart meter to record the current state of the power grid in real time, and identify the optimal charging strategy and record the charging data according to the prediction results and the current state of the power grid; the charging fault detection module is used to construct a charging fault detection model using the charging data, detect charging faults, and automatically notify the operation and maintenance personnel to repair them; the model optimization module is used to update the fault and maintenance records after the repair is completed, and use the fault and maintenance records to further improve the fault detection model and optimize the adaptive charging scheduling model.

[0043] In a third aspect, the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the remote control method for a charging pile as described in the first aspect of the present invention is implemented.

[0044] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the remote control method for a charging pile as described in the first aspect of the present invention is implemented.

[0045] The beneficial effects of the present invention are as follows: by calculating the data integrity index, a quantitative evaluation of the quality of power grid data transmission is achieved, data integrity and reliability are guaranteed, and system stability is improved. By introducing a seasonal adjustment factor to optimize the adaptive charging scheduling model, a more accurate time series prediction is achieved, the prediction accuracy of the adaptive charging scheduling model is improved, and the charging strategy is optimized to achieve effective use of energy. BRIEF DESCRIPTION OF THE DRAWINGS

[0046] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0047] Figure 1 This is a flow chart of the remote control method of the charging pile in Example 1.

[0048] Figure 2 This is a schematic diagram of the remote control system of the charging pile in Example 1. DETAILED DESCRIPTION

[0049] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0050] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0051] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0052] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a remote control method for a charging pile, comprising the following steps:

[0053] S1: Collect and pre-process power grid data, and establish a power grid data set based on the power grid data;

[0054] Select high-precision smart meters that support the DLMS / COSEM protocol and install them at key nodes of the power grid. After configuring the communication parameters, perform calibration and testing, and start the data logging function to collect voltage, current, and power.

[0055] It should also be noted that: smart meters are installed at key nodes of the power grid (such as substations, important load points, etc.) to ensure that the collected data can fully reflect the status of the power grid; the communication parameters of the meter are set, including IP address, port number, communication rate, etc., to ensure a stable connection with the central server; the installed smart meter is calibrated to ensure that the parameters it measures, such as voltage, current and power, are accurate, and then a functional test is performed to confirm that the data recording function is started normally.

[0056] The collected power grid data is transmitted to the central server in real time, and the data integrity is calculated using the following formula:

[0057]

[0058] Where I is the data integrity index, L is the packet loss rate, D is the average data delay, α is the impact coefficient of the packet loss rate, β is the impact coefficient of the data delay, γ is the threshold parameter, and e is the base of the natural logarithm.

[0059] It should also be noted that: the collected power grid data is transmitted to the central server in real time through a secure and reliable communication network (such as optical fiber, 4G / 5G, Ethernet, etc.); the data integrity index I is calculated using the above formula to evaluate the quality of data transmission. The purpose of this formula is to quantify the problems that may occur during data transmission, such as packet loss and delay, and adjust the communication strategy or take remedial measures accordingly;

[0060] Check the data structure of power grid data, remove duplicate values, process missing values, detect and correct outliers, and standardize power grid data on the central server;

[0061] It should also be noted that: checking the data structure specifically includes: checking the data structure of the power grid data to confirm that the timestamps of all records are arranged in sequence;

[0062] To remove duplicate values, use the duplicated() method in the Python Pandas library to find duplicate records. For the duplicate records found, choose to keep the latest record.

[0063] To handle missing values, we use Pandas’ isnull() to identify missing values ​​and use linear interpolation to fill in missing values. If some records have too many missing values ​​and cannot be reliably estimated, we delete these records directly.

[0064] Detecting and correcting outliers is as follows: Apply Z-score to detect outliers. For a feature, if its Z-score exceeds 3, the point is considered an outlier. Replace outliers that exceed the normal range with upper and lower limits. If the outliers are caused by erroneous readings, directly remove the relevant records.

[0065] Integrate the preprocessed power grid data, build a power grid data set and store it in the database.

[0066] It should also be noted that storing in a database specifically means storing the integrated data set in a high-performance relational database (such as MySQL, PostgreSQL) or NoSQL database (such as MongoDB) to facilitate subsequent query and analysis.

[0067] S2: Use historical data to build an adaptive charging scheduling model and predict future grid load change trends based on grid data;

[0068] The power grid data is imported into the Python environment for preprocessing, the ADF test is used to ensure the data stationarity, and the data is divided into training and test sets.

[0069] Furthermore, you can use the Pandas library to read CSV files, database connections, or other data sources; use the ADF (Augmented Dickey-Fuller) test to check whether the time series is stationary. If the data is non-stationary, you need to perform a difference or other transformation; divide the data into chronological order, 80% for training and 20% for testing.

[0070] Select ARIMA as the adaptive charging scheduling model, use Pandas' acf and pacf functions to generate ACF and PACF plots for stationary data, optimize parameter selection through grid search, use AIC as the evaluation metric, initialize the adaptive charging scheduling model with the selected parameters, and fit the model on the training set. Use Python's statsmodels library to implement the adaptive charging scheduling model, verify the performance of the adaptive charging scheduling model on the test set, calculate the prediction error, and analyze the residual plot;

[0071] A seasonal adjustment factor λ is introduced to optimize the adaptive charging scheduling model. The specific formula is as follows:

[0072]

[0073] in, is the predicted value of the power grid load at time t, c is a constant term, φ i is the autoregressive coefficient, θ j is the moving average coefficient, ∈ t is the error term, S t is the seasonal component, λ is the seasonal adjustment factor, p is the autoregressive order, i is the index of the autoregressive term, j is the index of the moving average term, q is the moving average order, t is the time index, and X t-i is the actual observed value in the past i.

[0074] Furthermore, the autocorrelation function (ACF) and partial autocorrelation function (PACF) graphs are plotted using Pandas' acf and pacf functions to help determine the parameters of the ARIMA model. The best ARIMA parameter combination is found through the grid search method, and AIC (Akaike Information Criterion) is used as the evaluation metric. The ARIMA model is initialized with the selected parameters and fitted on the training set. The model performance is verified on the test set, the prediction error is calculated, and the residual graph is analyzed to ensure that no pattern remains, indicating that the model has captured the dynamics of the data well. For each seasonal cycle (e.g., weekly, monthly), the average value of each time point in the cycle is calculated as the seasonal adjustment factor for that time point.

[0075] The grid data was verified through stationarity test, ACF and PACF plotting, parameter selection, setting seasonal adjustment factor λ and adaptive charging scheduling model. The trained adaptive charging scheduling model was used to predict the future grid load change trend. The expression is:

[0076]

[0077] in, For the future trend of grid load changes, is the predicted value of the power grid load at time t, is the predicted value of the power grid load at time t-1.

[0078] S3: Use smart meters to record the current state of the power grid in real time, identify the optimal charging strategy based on the prediction results and the current state of the power grid, and record the charging data;

[0079] Real-time grid status information is obtained from smart meters. At the same time, based on the grid load forecast for a period of time in the future, grid load smoothing and charging efficiency optimization functions are defined. The specific formula is as follows:

[0080]

[0081] Among them, F(t) is the charging strategy fitness value at time t, X t is the actual grid load value at time t, X max is the maximum load value in historical data, E k (t) is the actual charging energy of the kth charging pile at time t, E max,k is the maximum charging energy of the kth charging pile, N is the total number of charging piles, Q1 is the weight coefficient of load smoothing, Q2 is the weight coefficient of charging efficiency, Q3 is the weight coefficient of future load change trend, is the future grid load change trend, k is the charging pile index;

[0082] It should also be noted that: through the installed smart meters that support the DLMS / COSEM protocol, grid parameters such as voltage, current, and power are collected in real time; the fitness value is used to evaluate the quality of different charging strategies.

[0083] Initialize the genetic algorithm parameters and generate the initial population, calculate the fitness value of each individual, generate a new generation of population through selection, crossover and mutation operations according to the fitness value of each individual, iterate and optimize until the termination condition is met, and finally identify the charging strategy with the largest fitness value as the optimal charging strategy and execute the optimal charging strategy, while recording the actual charging situation.

[0084] Furthermore, genetic algorithm parameters such as population size, crossover probability, and mutation probability are determined, and an initial population (i.e., a set of random charging strategies) is generated. For each charging strategy in the population, its fitness value is calculated, and individuals with better performance are selected as parents according to the fitness value. The selected parents are crossed to generate new offspring, and the offspring are mutated with a certain probability. New gene combinations are introduced, and selection, crossover, mutation, and evaluation are iteratively performed until the termination condition is met (such as reaching the maximum number of iterations or the fitness value is no longer significantly improved); the optimal charging strategy found is applied to the actual charging process, and the charging power of each charging pile is controlled to save the actual charging data, including start time, end time, charging amount and other information, for subsequent analysis and reporting.

[0085] S4: Use charging data to build a charging fault detection model to detect charging faults and automatically notify operation and maintenance personnel to repair them;

[0086] The Pandas library is used to clean and preprocess the power grid data and divide the data set into training and test sets.

[0087] It should also be noted that the Pandas library is used to clean and preprocess the power grid data, such as removing duplicate values, processing missing values, and standardizing features.

[0088] Random forest is selected as the charging fault detection model. The charging fault model is trained using the training set data. A fault probability prediction factor is introduced. The specific formula is as follows:

[0089]

[0090] Among them, P f is the probability of failure, b is the bias term, ζ is the failure probability prediction factor, M is the total number of features, m is the feature index, and w m is the weight of the mth feature, x m is the value of the mth feature;

[0091] Construct a classifier based on random forest, the expression is as follows:

[0092]

[0093] Among them, R is the fault detection result, P f is the probability of failure, T is the set threshold;

[0094] The trained charging fault model is deployed to the real-time monitoring system to perform fault detection on the real-time collected data. When a charging fault is detected, an alarm is automatically sent to the operation and maintenance team, and the fault location and preliminary diagnostic information are provided.

[0095] It should also be noted that the random forest classifier is trained using the training set data, R is a binary variable, R=1 indicates a fault, R=0 indicates no fault, and T is a set threshold (e.g. 0.5) used to determine whether the probability of the model output is high enough to be determined as a fault.

[0096] S5: After the repair is completed, the fault and maintenance records are updated, and the fault detection model and the adaptive charging scheduling model are optimized using the fault and maintenance records;

[0097] The fault information is recorded in the database and the response time and handling status of the operation and maintenance team are recorded. After the repair is completed, the fault record is updated, the alarm is closed, the performance of the entire charging scheduling system is reviewed, and the latest operating data is collected from the real-time monitoring system.

[0098] It should also be noted that after the fault is repaired, the alarm is closed and the fault status is updated to "solved", and the latest data is obtained from the real-time monitoring system to evaluate system performance and optimize the model.

[0099] The performance of the existing charging fault detection model and adaptive charging scheduling model is evaluated using the collected operating data, and the accuracy, recall rate, F1 score and root mean square error are calculated. Based on the collected data set, the charging fault detection model and the adaptive charging scheduling model are retrained, the parameters are adjusted, and a new feature selection method is introduced. The cross-validation method is used to further adjust the parameters of the charging fault detection model and the adaptive charging scheduling model, and the fault detection model and the adaptive charging scheduling model are optimized.

[0100] Furthermore, the existing model is used to predict the newly collected data, and the accuracy, recall, F1 score and root mean square error are calculated. The comparison between the predicted results and the actual results is displayed through graphs to help more intuitively understand the model performance. For example, ROC curves and residual graphs can be drawn. According to the results of the evaluation indicators, it is judged whether the performance of the model is satisfactory. If some indicators perform poorly, consider possible reasons, such as overfitting or underfitting. If the model performance is found to have degraded, consider retraining the model with new data, or adjust the model parameters. Random forests can be used to give an importance score for each feature. If some features are no longer important in the new data, consider removing these features. At the same time, explore whether more useful information can be extracted from the new data as new features. According to the characteristics of the new data, it may be necessary to change the model structure or algorithm to improve its generalization ability.

[0101] The present embodiment also provides a remote control system for a charging pile, including: a data acquisition module, an adaptive charging scheduling module, an optimal charging strategy identification module, a charging fault detection module and a model optimization module; the data acquisition module is used to collect and pre-process power grid data, and establish a power grid data set according to the power grid data; the adaptive charging scheduling module is used to use historical data to build an adaptive charging scheduling model, and based on the power grid data, predict future power grid load change trends; the optimal charging strategy identification module is used to use a smart meter to record the current state of the power grid in real time, and identify the optimal charging strategy and record charging data according to the prediction results and the current state of the power grid; the charging fault detection module is used to build a charging fault detection model using charging data, detect charging faults, and automatically notify operation and maintenance personnel to repair them; the model optimization module is used to update the fault and maintenance records after the repair is completed, and use the fault and maintenance records to further improve the fault detection model and optimize the adaptive charging scheduling model.

[0102] This embodiment also provides a computer device, which is suitable for the remote control method of a charging pile, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the remote control method of the charging pile proposed in the above embodiment.

[0103] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0104] The present embodiment also provides a storage medium on which a computer program is stored. When the program is executed by the processor, the remote control method of the charging pile proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0105] In summary, the present invention realizes quantitative evaluation of the quality of power grid data transmission through data integrity index calculation, ensures data integrity and reliability, and improves system stability. By introducing seasonal adjustment factors to optimize the adaptive charging scheduling model, more accurate time series prediction is achieved, the prediction accuracy of the adaptive charging scheduling model is improved, and the charging strategy is optimized to achieve effective use of energy.

[0106] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A remote control method for a charging pile, characterized in that: include, Collect and pre-process power grid data, and establish a power grid data set based on the power grid data; Use historical data to build an adaptive charging scheduling model and predict future grid load change trends based on grid data; Use smart meters to record the current state of the power grid in real time, identify the optimal charging strategy and record charging data based on the prediction results and the current state of the power grid; Use charging data to build a charging fault detection model to detect charging faults and automatically notify maintenance personnel to repair them; After the repair is completed, the fault and maintenance records are updated, and the fault detection model and the adaptive charging scheduling model are optimized using the fault and maintenance records.

2. The remote control method of the charging pile according to claim 1, characterized in that: The specific steps of collecting power grid data and preprocessing and establishing a power grid data set according to the power grid data are as follows: Select high-precision smart meters that support the DLMS / COSEM protocol and install them at key nodes of the power grid. After configuring communication parameters, perform calibration and testing, and enable the data logging function to collect voltage, current, and power. The collected power grid data is transmitted to the central server in real time, and the data integrity is calculated using the following formula: Where I is the data integrity index, L is the packet loss rate, D is the average data delay, α is the impact coefficient of the packet loss rate, β is the impact coefficient of the data delay, γ is the threshold parameter, and e is the base of the natural logarithm; Check the data structure of power grid data, remove duplicate values, process missing values, detect and correct outliers, and standardize power grid data on the central server; Integrate the preprocessed power grid data, build a power grid data set and store it in the database.

3. The remote control method of the charging pile according to claim 2, characterized in that: The specific steps of constructing the adaptive charging scheduling model are as follows: Import the power grid data into the Python environment for preprocessing, use the ADF test to ensure data stability, and divide it into training and test sets; Select ARIMA as the adaptive charging scheduling model, use Pandas' acf and pacf functions to generate ACF and PACF plots for stationary data, optimize parameter selection through grid search, use AIC as the evaluation metric, initialize the adaptive charging scheduling model with the selected parameters, and fit the model on the training set. Use Python's statsmodels library to implement the adaptive charging scheduling model, verify the performance of the adaptive charging scheduling model on the test set, calculate the prediction error, and analyze the residual plot; The seasonal adjustment factor λ is introduced to optimize the adaptive charging scheduling model. The specific formula is as follows: in, is the predicted value of the power grid load at time t, c is a constant term, φ i is the autoregressive coefficient, θ j is the moving average coefficient, ∈ t is the error term, S t is the seasonal component, λ is the seasonal adjustment factor, p is the autoregressive order, i is the index of the autoregressive term, j is the index of the moving average term, q is the moving average order, t is the time index, and X t-i is the actual observed value in the past i.

4. The remote control method of the charging pile according to claim 3, characterized in that: The specific steps of predicting the future power grid load change trend based on power grid data are as follows: The grid data was verified through stationarity test, ACF and PACF plotting, parameter selection, setting seasonal adjustment factor λ and adaptive charging scheduling model. The trained adaptive charging scheduling model was used to predict the future grid load change trend. The expression is: in, For the future trend of grid load changes, is the predicted value of the power grid load at time t, is the predicted value of the power grid load at time t-1.

5. The remote control method of the charging pile according to claim 4, characterized in that: The smart meter is used to record the current state of the power grid in real time, and the optimal charging strategy is identified and the charging data is recorded according to the prediction results and the current state of the power grid. The specific steps are as follows: Real-time grid status information is obtained from smart meters. At the same time, based on the grid load forecast for a period of time in the future, grid load smoothing and charging efficiency optimization functions are defined. The specific formula is as follows: Among them, F(t) is the charging strategy fitness value at time t, X t is the actual grid load value at time t, X max is the maximum load value in historical data, E k (t) is the actual charging energy of the kth charging pile at time t, E max,k is the maximum charging energy of the kth charging pile, N is the total number of charging piles, Q1 is the weight coefficient of load smoothing, Q2 is the weight coefficient of charging efficiency, Q3 is the weight coefficient of future load change trend, is the future grid load change trend, k is the charging pile index; Initialize the genetic algorithm parameters and generate the initial population, calculate the fitness value of each individual, generate a new generation of population through selection, crossover and mutation operations according to the fitness value of each individual, iterate and optimize until the termination condition is met, and finally identify the charging strategy with the largest fitness value as the optimal charging strategy and execute the optimal charging strategy, while recording the actual charging situation.

6. The remote control method of the charging pile according to claim 5, characterized in that: The charging data is used to construct a charging fault detection model, detect charging faults, and automatically notify the operation and maintenance personnel to repair them. The specific steps are as follows: Use the Pandas library to clean and preprocess the power grid data and divide the data set into training and test sets; Random forest is selected as the charging fault detection model. The charging fault model is trained using the training set data. The fault probability prediction factor is introduced. The specific formula is as follows: Among them, P f is the probability of failure, b is the bias term, ζ is the failure probability prediction factor, M is the total number of features, m is the feature index, and w m is the weight of the mth feature, x m is the value of the mth feature; Construct a classifier based on random forest, the expression is as follows: Among them, R is the fault detection result, P f is the probability of failure, T is the set threshold; The trained charging fault model is deployed to the real-time monitoring system to perform fault detection on the real-time collected data. When a charging fault is detected, an alarm is automatically sent to the operation and maintenance team, and the fault location and preliminary diagnostic information are provided.

7. The remote control method of the charging pile according to claim 6, characterized in that: After the repair is completed, the fault and maintenance records are updated, and the fault detection model and the adaptive charging scheduling model are optimized using the fault and maintenance records. The specific steps are as follows: Record fault information in the database and record the response time and handling status of the operation and maintenance team. After the repair is completed, update the fault record, close the alarm, review the performance of the entire charging scheduling system, and collect the latest operating data from the real-time monitoring system; The performance of the existing charging fault detection model and adaptive charging scheduling model is evaluated using the collected operating data, and the accuracy, recall rate, F1 score and root mean square error are calculated. Based on the collected data set, the charging fault detection model and the adaptive charging scheduling model are retrained, the parameters are adjusted, and a new feature selection method is introduced. The cross-validation method is used to further adjust the parameters of the charging fault detection model and the adaptive charging scheduling model, and the fault detection model and the adaptive charging scheduling model are optimized.

8. A remote control system for a charging pile, based on the remote control method for a charging pile according to any one of claims 1 to 7, characterized in that: Including data acquisition module, trend prediction module, strategy identification module, fault detection module and model optimization module; A data acquisition module is used to collect and pre-process power grid data and establish a power grid data set based on the power grid data; The trend prediction module is used to build an adaptive charging scheduling model using historical data and predict future grid load change trends based on grid data; The charging strategy identification module is used to use smart meters to record the current state of the power grid in real time, identify the optimal charging strategy and record charging data based on the prediction results and the current state of the power grid; The fault detection module is used to build a charging fault detection model using charging data, detect charging faults, and automatically notify the operation and maintenance personnel to repair them; The model optimization module is used to update the fault and maintenance records after the repair is completed, and use the fault and maintenance records to further improve the fault detection model and optimize the adaptive charging scheduling model.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the remote control method for a charging pile according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the remote control method for a charging pile according to any one of claims 1 to 7 are implemented.

Images