Charging pile intelligent energy-saving management method and system based on big data

Abstract

The invention discloses a charging pile intelligent energy-saving management method and system based on big data, and relates to the technical field of charging pile energy-saving management, and the method comprises the steps: collecting the parameter data of a charging pile and a charging device, forming a data set, carrying out the preprocessing, constructing an input feature matrix, inputting a trained charging power prediction deep learning model, and obtaining a charging power prediction model; outputting the predicted charging power, performing target optimization, and adjusting the charging power of the charging pile to perform energy-saving management; during charging, charging power is compensated according to the actual battery electric quantity and the calculated target battery electric quantity at the next time interval; and monitoring the charging pile to generate a health state coefficient, and when the health state coefficient is smaller than a preset health threshold value, generating a charging pile early warning signal for maintenance. Through power prediction, health state monitoring and charging power compensation, dynamic optimization and stable control of the charging process are realized, energy consumption is effectively reduced, the service life of equipment is prolonged, the charging stability is improved, and the charging process is ensured to be highly matched with an actual demand.

Description

Technical Field

[0001] The present invention relates to the technical field of energy-saving management of charging piles, and in particular to an intelligent energy-saving management method and system for charging piles based on big data. Background Art

[0002] With the popularity of electric vehicles, the demand for charging piles has increased dramatically. Charging piles not only need to provide efficient and stable charging services, but also face the challenges of energy saving and intelligent management. In actual use, the power management of charging piles, the health monitoring of charging equipment, and the optimization and scheduling of the charging process are all key factors affecting the operating efficiency and equipment life of charging piles. Traditional charging pile management systems can often only provide basic charging control functions, lacking functions such as refined regulation of charging power, energy-saving management, and equipment health status monitoring, which to a certain extent affects the efficiency of charging piles and user experience.

[0003] In order to improve the efficiency of charging piles and enhance their energy-saving and intelligent management capabilities, some management solutions based on data analysis and artificial intelligence have gradually been applied to charging pile management systems. For example, by collecting real-time data from charging piles and using big data analysis technology to predict and optimize charging demand, it is possible to achieve a reasonable allocation of charging power while meeting charging demand. In addition, based on the health monitoring data of charging equipment, combined with deep learning algorithms, the health status of the equipment can also be evaluated and predicted, so that preventive maintenance can be performed before a failure occurs, extending the service life of the equipment.

[0004] However, the existing big data and deep learning applications still have some shortcomings in the field of intelligent energy-saving management of charging piles. For example, the existing technology lacks a complete charging power prediction and optimization method to achieve refined control of the charging power of charging piles. At the same time, there is a lack of a system that can monitor the health status of charging piles in real time and dynamically compensate for power, resulting in performance degradation or even failure of charging equipment after long-term operation. Summary of the invention

[0005] Based on the above-mentioned shortcomings of the prior art, the purpose of the present invention is to provide a charging pile intelligent energy-saving management method and system based on big data to solve the above-mentioned technical problems.

[0006] To achieve the above purpose, the present invention provides the following technical solution: a charging pile intelligent energy-saving management method based on big data, comprising: Collect parameter data of charging piles and connected charging equipment, form data sets and perform data preprocessing; constructing an input feature matrix according to the data set, inputting the input feature matrix into a trained charging power prediction deep learning model, and outputting a predicted charging power at the next moment; Optimize the target according to the predicted charging power at the next moment and adjust the charging power of the charging pile for energy-saving management; When charging, the target battery power for the next time interval is calculated based on the current charging power. When the next time interval is reached, the actual battery power is obtained, and the charging power is compensated according to the actual battery power and the target battery power. After charging is completed, the health status of the charging pile is monitored and a charging pile health status coefficient is generated. When the charging pile health status coefficient is less than a preset health threshold, a charging pile warning signal is generated and the charging pile is inspected and maintained.

[0007] The present invention is further configured such that the parameter data includes timestamp, charging power, charging load, battery power, charging device identification, battery temperature, battery capacitance and battery internal resistance, and the data preprocessing includes cleaning the data set and removing outliers and missing values.

[0008] The present invention is further configured to construct a charging feature vector according to the data set, including: constructing charging features according to the data set, the charging features including charging power features, charging load features, battery power features, charging conversion features, battery health features, and temperature sensitivity features; A charging feature vector is constructed based on the charging characteristics. The timestamp is used as a sequence. An input feature matrix is ​​constructed based on the charging feature vector. The charging power feature is defined as the target output. Data alignment is performed. The data alignment predicts the charging power at the next moment by shifting the input feature matrix and the target output by one bit.

[0009] The present invention is further configured such that the calculation logic of the charging conversion characteristic is: ,in, For the The charging conversion characteristics of the connected charging device, for The amount of battery power increase over time, is the input charging power; The calculation logic of the battery health feature is: in, For the Battery health characteristics of each connected charging device, is the actual capacity of the current battery, is the nominal capacity of the battery, is the current internal resistance of the battery, is the standard battery internal resistance, and is the weight coefficient, and and are greater than zero; The calculation logic of the temperature sensitive feature is: ,in, For the The temperature-sensitive characteristics of the connected charging device is the current battery temperature, For optimal battery charging temperature, is the temperature sensitivity coefficient.

[0010] The present invention is further configured that the training of the charging power prediction deep learning model includes: The acquired data set is divided into a training set and a test set, and an input feature matrix and a target output are constructed according to the training set; The input feature matrix and the target output are passed to the charging power prediction deep learning model for training, wherein the charging power prediction deep learning model includes a charging power prediction calculation unit, an output unit, a loss function unit and a model evaluation unit, wherein the charging power prediction calculation unit is used to update the cell state and the hidden state through the forgetting gate, the input gate, the candidate state generation and the output gate, and the output unit is used to perform a linear transformation according to the weight matrix and the bias of the hidden state and the output layer to obtain the target output, the loss function unit is used to define the loss function and update the parameters of the model according to the loss function, and the model evaluation unit is used to evaluate the performance of the model and check the degree of fit of the model to the data; The training is performed until the maximum iteration is reached or the evaluation index meets the preset requirements, and the final parameters are set as model parameters to complete the training of the charging power prediction deep learning model.

[0011] The present invention is further configured to perform target optimization according to the predicted charging power at the next moment, wherein a target optimization function is defined: , To solve the charging power, the objective function minimize, is the number of connected charging devices, For the connected charging devices at the time The optimal charging power, For the The target battery level of each connected charging device, For the connected charging devices in The battery level at and is the start time and end time of the time window, and is the weight coefficient, which is used to adjust the balance between energy consumption and power deviation; The constraints of the objective optimization function include: , ; ; ,in, and For the The minimum and maximum power ratings of the connected charging devices, It is the smoothing amount of charging power, which is used to introduce adjustment space during charging power control to avoid frequent power fluctuations. For the connected charging devices at the time The predicted charging power, For the connected charging devices in The battery level at For the The connected charging device corresponds to The charging efficiency of a charging pile, the value range is (0,1), For the moment The total power constraint of the power grid.

[0012] The present invention is further configured that when solving the above-mentioned target optimization function, the target optimization function contains integral and time continuous variables. Since the obtained parameter data is discrete data with timestamps as a sequence, the time interval [ , ] is discretized into K hourly steps, converting the continuous time optimization into a discrete optimization, and dividing the time interval into Small step length: ; Use discrete charging power values Approximately replace continuous ; Convert the integral to sum form: ; ; After discretization, the objective function becomes: ; Solve the discretized optimization problem through numerical optimization algorithms, including gradient descent method, Lagrange multiplier method, dynamic programming and genetic algorithm; Solve to get the optimal charging power According to the optimal charging power The collection of connected charging devices is charged.

[0013] The present invention is further configured to include: When charging, based on the optimal charging power Calculate the target battery power for the next time interval. When the next moment is reached, obtain the actual battery power. Compensate the charging power according to the actual battery power and the target battery power. The calculation logic of the target battery power at the next moment is: ,in, For the next time interval The target battery level after For the connected charging devices in The battery power at the moment; the charging power is compensated according to the actual battery power and the target battery power. The calculation logic of the compensation amount is: ,in, for The charging power compensation amount at the moment is used to compensate Charging power at the moment, for The actual battery level at the moment, for Target battery level at the moment, is the proportional coefficient, which is used to control the compensation strength.

[0014] The present invention is further configured to include: monitoring the health status of the charging pile, generating a health status coefficient of the charging pile, and when the health status coefficient of the charging pile is less than a preset health threshold, repairing the charging pile, wherein the calculation logic of the health status coefficient of the charging pile is: ,in, For the The health status coefficient of the charging pile of each charging pile, for The current at the moment, For the The rated current of each charging pile, For and Take the maximum value, For the Charging piles in the time period The average current, and is the weight coefficient, which is used to adjust the impact of current integral and current fluctuation on the health index.

[0015] The present invention also provides a charging pile intelligent energy-saving management system based on big data, the system comprising: Acquisition module: collects parameter data of charging piles and connected charging equipment, forms a data set and performs data preprocessing; Prediction module: construct an input feature matrix according to the data set, input the input feature matrix into the trained charging power prediction deep learning model, and output the predicted charging power at the next moment; Energy-saving module: optimize the target according to the predicted charging power at the next moment, and adjust the charging power of the charging pile for energy-saving management; Compensation module: When charging, the target battery power of the next time interval is calculated based on the current charging power. When the next time interval is reached, the actual battery power is obtained, and the charging power is compensated according to the actual battery power and the target battery power. Monitoring module: After charging is completed, the health status of the charging pile is monitored and the health status coefficient of the charging pile is generated. When the health status coefficient of the charging pile is less than the preset health threshold, a charging pile warning signal is generated and the charging pile is inspected and maintained.

[0016] The present invention provides a charging pile intelligent energy-saving management method and system based on big data. The method collects parameter data of the charging pile and the connected charging equipment to form a data set and perform data preprocessing; constructs an input feature matrix according to the data set, inputs the input feature matrix into a trained charging power prediction deep learning model, and outputs the predicted charging power at the next moment; performs target optimization according to the predicted charging power at the next moment, and adjusts the charging power of the charging pile for energy-saving management; when charging, calculates the target battery power of the next time interval based on the current charging power, obtains the actual battery power when the next time interval is reached, and compensates the charging power according to the actual battery power and the target battery power; after charging is completed, monitors the health status of the charging pile, generates a charging pile health status coefficient, and generates a charging pile early warning signal when the charging pile health status coefficient is less than a preset health threshold, and repairs the charging pile, and the beneficial effects produced include: Improve charging efficiency and reduce energy consumption: This invention collects real-time data from charging piles and charging equipment, builds a deep learning model to predict charging power, and optimizes the target based on the predicted charging power at the next moment, dynamically adjusts the charging power output, and makes the charging process more compatible with actual needs. This optimized power allocation strategy can effectively reduce energy consumption during charging and achieve energy-saving effects; Extending the service life of equipment: The present invention monitors the health status of the charging pile in real time during the charging process, generates a health status coefficient of the charging pile, and issues a warning signal when the health status coefficient is lower than a preset threshold, performs maintenance and repairs on the charging pile, and avoids equipment overload or loss caused by long-term high-load operation, thereby extending the service life of the equipment; Charging power compensation to improve charging stability: During the charging process, the target battery power for the next time interval is calculated based on the current charging power, and the actual battery power is obtained in real time. The charging power is compensated according to the difference between the actual battery power and the target battery power. This ensures that a stable charging power output is maintained during the charging process, improves the stability of the charging process, and reduces power fluctuations.

[0017] The above description is only an overview of the technical solution of the present application. In order to more clearly understand the technical means of the present application, it can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. In the drawings: Figure 1 A flowchart of an intelligent energy-saving management method for charging piles based on big data is shown as an exemplary embodiment of the present invention; Figure 2 The schematic diagram of the structure of an intelligent energy-saving management system for charging piles based on big data is shown as an exemplary embodiment of the present invention. DETAILED DESCRIPTION

[0019] The following will describe the embodiments of the present invention with reference to the accompanying drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the contents disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments, and the details in this specification can also be modified or changed in various ways based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for illustrating the present invention, not for limiting the scope of protection of the present invention.

[0020] It should be noted that the illustrations provided in the following embodiments are only schematic illustrations of the basic concept of the present invention, and thus the drawings only show components related to the present invention rather than being drawn according to the number, shape and size of components in actual implementation. In actual implementation, the type, quantity and proportion of each component may be changed arbitrarily, and the component layout may also be more complicated.

[0021] In the following description, numerous details are discussed to provide a more thorough explanation of the embodiments of the present invention. However, it is obvious to those skilled in the art that the embodiments of the present invention can be implemented without these specific details. In other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail to avoid making the embodiments of the present invention difficult to understand. Embodiment 1

[0022] Intelligent energy-saving management method of charging piles based on big data, such as Figure 1 As shown, including: Collect parameter data of charging piles and connected charging equipment, form data sets and perform data preprocessing; constructing an input feature matrix according to the data set, inputting the input feature matrix into a trained charging power prediction deep learning model, and outputting a predicted charging power at the next moment; Optimize the target according to the predicted charging power at the next moment and adjust the charging power of the charging pile for energy-saving management; When charging, the target battery power for the next time interval is calculated based on the current charging power. When the next time interval is reached, the actual battery power is obtained, and the charging power is compensated according to the actual battery power and the target battery power. After charging is completed, the health status of the charging pile is monitored and a charging pile health status coefficient is generated. When the charging pile health status coefficient is less than a preset health threshold, a charging pile warning signal is generated and the charging pile is inspected and maintained.

[0023] Parameter data is collected from the charging pile and the connected charging equipment in real time to form a data set D, which contains N records. In each record, the parameter data includes timestamp, charging power, charging load, battery power, charging equipment identification, battery temperature, battery capacitance and battery internal resistance. The data set is cleaned to remove abnormal values ​​and missing values. Specifically, the timestamp is a mark that records the data collection time and is used to identify the specific collection time of each data. Charging power refers to the power provided by the charging pile to the battery. The charging power directly affects the charging speed and energy consumption. Controlling the charging power can adjust the efficiency and stability of the charging process. The charging load represents the current load of the charging device, that is, the number of charging devices connected to the charging pile and the load characteristics during the charging process. The electrical load data is used to evaluate the load condition of the current system; the battery charge indicates the amount of energy currently stored in the battery. In the present invention, it is not expressed in percentage but in watt-hours; the charging device identification is a number or label used to uniquely identify each charging device connected to the charging pile. The charging device identification helps the system distinguish the data of different devices when multiple devices are connected; the battery temperature is the temperature state of the battery during the charging process; the battery capacitance is the ability of the battery to store charge. The battery capacitance reflects the actual storage capacity of the battery and can be used to evaluate the health status and capacity decay of the battery; the battery internal resistance is the resistance to the flow of current inside the battery. The internal resistance will affect the charging and discharging efficiency and heat generation of the battery. An increase in internal resistance will cause the battery to heat up and reduce the charging efficiency.

[0024] The present invention is further configured to construct a charging feature vector according to the data set, including: constructing charging features according to the data set, the charging features including charging power features, charging load features, battery power features, charging conversion features, battery health features, and temperature sensitivity features; Construct a charging feature vector based on the charging characteristics, use the timestamp as a sequence, construct an input feature matrix based on the charging feature vector, define the charging power feature as the target output, and perform data alignment. Data alignment predicts the charging power at the next moment by shifting the input feature matrix and the target output by one bit. The charging characteristics are constructed based on the parameter data. Furthermore, the charging characteristics include charging power characteristics, battery power characteristics, charging load characteristics, charging conversion characteristics, battery health characteristics and temperature sensitivity characteristics; Specifically, the calculation logic of the charging conversion feature is: ,in, For the The charging conversion characteristics of the connected charging device, for The amount of battery power increase over time, is the input charging power; the above calculation logic is used to calculate the charging efficiency of the charging pile for the battery per unit time. Specifically, the value of the charging conversion characteristic is used to measure the increase in the battery power caused by the input charging power per unit time, thereby characterizing the energy conversion efficiency during the charging process; The calculation logic of the battery health feature is: in, For the Battery health characteristics of each connected charging device, is the actual capacity of the current battery, is the nominal capacity of the battery, is the current internal resistance of the battery, is the standard battery internal resistance, and is the weight coefficient, and and are all greater than zero; Specifically, the above calculation logic is used to evaluate the health status of the rechargeable battery, which is specifically characterized by the health characteristics of the battery. By combining the actual capacity and internal resistance of the battery, the current health status of the battery is comprehensively evaluated and compared with the new battery or nominal state to determine the degree of battery aging. The higher the value, the better the health status of the battery, otherwise it indicates that the battery is aged or damaged; The calculation logic of the temperature sensitive feature is: ,in, For the The temperature-sensitive characteristics of the connected charging device is the current battery temperature, For optimal battery charging temperature, is the temperature sensitivity coefficient. Specifically, the above calculation logic is used to calculate the temperature sensitivity characteristics of the battery during the charging process, which is specifically characterized by the temperature sensitivity coefficient. The temperature sensitivity coefficient reflects the degree to which the current battery temperature deviates from the optimal charging temperature, which can help the system adjust the charging power or charging strategy to avoid the impact of excessively high or low temperatures on battery performance and life. The higher the value, the more the current temperature deviates from the optimal temperature, and the greater the impact of the charging process on the battery; the closer the value is to 1, the closer the temperature is to the optimal state, and the smaller the impact of the charging process on the battery.

[0025] The present invention is further configured that the training of the charging power prediction deep learning model includes: The acquired data set is divided into a training set and a test set, and an input feature matrix and a target output are constructed according to the training set; specifically, a charging feature vector is constructed according to charging power features, battery power features, load information features, charging conversion features, battery health features, and temperature sensitivity features, and an input feature matrix X is constructed according to the feature vector with timestamps as a sequence. The charging power feature is defined as the target output, and data alignment is performed. Data alignment is to predict the charging power at the next moment by shifting the input feature and the target output by one bit; The input feature matrix and the target output are passed to the charging power prediction deep learning model for training, wherein the charging power prediction deep learning model includes a charging power prediction calculation unit, an output unit, a loss function unit and a model evaluation unit, wherein the charging power prediction calculation unit is used to update the cell state and the hidden state through the forgetting gate, the input gate, the candidate state generation and the output gate, and the output unit is used to perform a linear transformation according to the weight matrix and the bias of the hidden state and the output layer to obtain the target output, the loss function unit is used to define the loss function and update the parameters of the model according to the loss function, and the model evaluation unit is used to evaluate the performance of the model and check the degree of fit of the model to the data; The training is performed until the maximum iteration is reached or the evaluation index meets the preset requirements, and the final parameters are set as model parameters to complete the training of the charging power prediction deep learning model.

[0026] Specifically, in the charging power prediction deep learning model, the construction logic of the charging power prediction calculation unit is: ; ; ; ; ; ,in, , and Represent the activation values ​​of the forget gate, input gate, and output gate respectively; , , and are the weight matrices from hidden state to gate and candidate state, respectively. , , and are the weight matrices input to the gate and candidate states, respectively. , , and are the bias vectors of each gate and candidate state respectively; is an element-wise multiplication operation, is the Sigmoid activation function, is the hyperbolic tangent activation function.

[0027] Specifically, the forget gate Determine how much of the previous cell state to keep ; Input gate Determine the current input How much information is written into the cell state? Determines cell state How much information is output to the hidden state ; Candidate cell states Generate a new candidate cell state through the current input and the previous hidden state; cell state update Combine the information of the forget gate and the input gate to update the cell state; hide the state Used for prediction of the output layer and recursive calculation of the next moment; input features and the hidden state at the previous moment Input charging power prediction calculation unit to calculate the hidden state at the current moment and cell status ; The calculation logic of the output unit is: ,in, is the predicted charging power at the next moment, is the output layer weight matrix, is the output layer bias vector.

[0028] The calculation logic of the loss function is: , is the loss function, for The weight of the moment, without any pre-setting, for , is the actual charging power at the next moment; the parameter update logic is: ; ; ; ; ; Among them, the first-order moment estimate : The exponentially weighted moving average of the gradient, representing the expectation of the gradient; second-order moment estimate : The exponentially weighted moving average of the square of the gradient, which represents the expectation of the gradient variance; Bias correction: Since the initial estimate is too small, bias correction is performed to obtain and ; Parameter update: Use the corrected gradient estimate to update the model parameters ; In each training iteration, the gradient of the loss function is calculated and the model parameters are updated; is the set of model parameters, including all weights and biases; is the number of iterations, and is the exponential decay rate of the first- and second-order moment estimates, , ; is the learning rate; To prevent small numbers with zero denominators, ; In model evaluation, the model is evaluated by setting evaluation indicators, among which the evaluation indicator - determination coefficient The calculation logic is: , where the coefficient of determination Measures the degree of fit of the model to the data. The value range is [0,1]. The closer to 1, the stronger the explanatory power of the model. The numerator is the sum of squares of the prediction errors. The denominator is the sum of squares of the true value and its mean. is the number of test set samples; is the actual charging power; is the charging power predicted by the model; is the average value of the actual charging power.

[0029] The final obtained parameters are set as model parameters, and the prediction of charging power at the next moment is output by inputting the parameter data at the current moment in real time; The present invention is further configured to perform target optimization according to the predicted charging power at the next moment, wherein a target optimization function is defined: , To solve the charging power, the objective function minimize, is the number of connected charging devices, For the connected charging devices at the time The optimal charging power, For the The target battery level of each connected charging device, For the connected charging devices in The battery level at and is the start time and end time of the time window, and is a weight coefficient used to adjust the balance between energy consumption and power deviation; the target optimization function is used in the intelligent energy-saving management of charging piles to achieve the best power allocation of charging equipment during the charging process by dynamically adjusting the charging power; the goal is to minimize the deviation from the target battery power during the charging process, ensuring that the charging process achieves energy saving while meeting the battery requirements. Weight coefficient is the power balance coefficient, which is used to adjust the balance between battery energy consumption and power deviation; weight coefficient is the deviation adjustment coefficient, which is used to adjust the weight of the deviation between the actual power and the target power, and .

[0030] The constraints of the objective optimization function include: , ; ; ,in, and For the The minimum and maximum power ratings of the connected charging devices, It is the smoothing amount of charging power, which is used to introduce adjustment space during charging power control to avoid frequent power fluctuations. For the connected charging devices at the time The predicted charging power, For the connected charging devices in The battery level at For the The connected charging device corresponds to The charging efficiency of a charging pile, the value range is (0,1), For the moment Specifically, the objective optimization function is used to determine the target of charging power distribution, which aims to minimize power consumption and power deviation and ensure high efficiency and energy saving of charging.

[0031] Specifically, the optimization objective function is to minimize the total charging energy during the charging process and the deviation between the terminal battery power and the target power, that is, the optimization variable The value of The value of is the smallest. Under the constraint conditions, the above objective function is solved. When solving the above objective optimization function, the objective optimization function contains integral and time continuous variables. Since the obtained parameter data is discrete data with timestamps as the sequence, the time interval [ , ] is discretized into K hourly steps, converting the continuous time optimization into a discrete optimization, and dividing the time interval into Small step length: ; Use discrete charging power values Approximately replace continuous ; Convert the integral to sum form: ; ; After discretization, the objective function becomes: ; The discretized optimization problem is solved by numerical optimization algorithms, including gradient descent method, Lagrange multiplier method, dynamic programming and genetic algorithm. The specific algorithm is not limited here. Solve to get the optimal charging power A time series set; according to the optimal charging power The time series collection charges the charging device.

[0032] The present invention is further configured to include: when charging, based on the optimal charging power Calculate the target battery power for the next time interval. When the next moment is reached, obtain the actual battery power. Compensate the charging power according to the actual battery power and the target battery power. The calculation logic of the target battery power at the next moment is: ,in, For the next time interval The target battery level after For the connected charging devices in The battery power at the moment; the charging power is compensated according to the actual battery power and the target battery power. The calculation logic of the compensation amount is: ,in, for The charging power compensation amount at the moment is used to compensate Charging power at the moment, for The actual battery level at the moment, for Target battery level at the moment, is a proportional coefficient used to control the compensation strength. Specifically, the above logic is used to dynamically compensate the power during the charging process to ensure that the actual battery power is consistent with the target power. By calculating the difference between the target battery power and the actual battery power after each time interval, the charging power can be adjusted to compensate to reduce the error and ensure that the battery can reach the predetermined power level with high accuracy during the charging process. The compensation logic dynamically adjusts the charging power to make the actual battery power as close to the target power as possible to ensure that the charging process proceeds as expected. This precise power control helps optimize the charging process and improve charging efficiency.

[0033] The present invention is further configured to include: monitoring the health status of the charging pile, generating a health status coefficient of the charging pile, and when the health status coefficient of the charging pile is less than a preset health threshold, repairing the charging pile, wherein the calculation logic of the health status coefficient of the charging pile is: ,in, For the The health status coefficient of the charging pile of each charging pile, for When adjusting the charging power, the voltage remains unchanged, so the charging power is adjusted by adjusting the current. Therefore, the health status of the charging pile can be directly reflected by the current. For the The rated current of each charging pile, For and The maximum value is taken, which means that Greater than When the overload current accumulates, the cumulative amount of the overload current is recorded. The health status coefficient of the charging pile decays exponentially with the accumulation of the overload current. For the Charging piles in the time period The average current, and is a weight coefficient used to adjust the impact of current integral and current volatility on the health index. When the health status coefficient of the charging pile is less than the preset health threshold, it means that the current of the charging pile is in an overload state for a long time or the current fluctuation is large. At this time, it is necessary to optimize and restrict the energy-saving strategy or overhaul the charging pile. Specifically, the calculation logic is used to evaluate the health status of the charging pile, and the working status and cumulative loss of the charging pile during the charging process are reflected by calculating the health status coefficient. The health status coefficient of the charging pile is calculated based on the current deviation and current fluctuation. By evaluating the integral and volatility of the current deviation, this method can monitor the load status, overload condition and operating stability of the charging pile, so as to carry out timely overhaul and maintenance, extend the service life of the equipment, and ensure the safety of the charging process. Embodiment 2

[0034] See also Figure 2 , the exemplary charging pile intelligent energy-saving management system based on big data includes: Acquisition module: collects parameter data of charging piles and connected charging equipment, forms a data set and performs data preprocessing; Prediction module: construct an input feature matrix according to the data set, input the input feature matrix into the trained charging power prediction deep learning model, and output the predicted charging power at the next moment; Energy-saving module: optimize the target according to the predicted charging power at the next moment, and adjust the charging power of the charging pile for energy-saving management; Compensation module: When charging, the target battery power of the next time interval is calculated based on the current charging power. When the next time interval is reached, the actual battery power is obtained, and the charging power is compensated according to the actual battery power and the target battery power. Monitoring module: After charging is completed, the health status of the charging pile is monitored and the health status coefficient of the charging pile is generated. When the health status coefficient of the charging pile is less than the preset health threshold, a charging pile warning signal is generated and the charging pile is inspected and maintained.

[0035] It should be noted that the intelligent energy-saving management system for charging piles based on big data provided in the above embodiment and the intelligent energy-saving management method for charging piles based on big data provided in the above embodiment belong to the same concept, wherein the specific manner in which each module and unit performs operations has been described in detail in the method embodiment and will not be repeated here. In actual applications, the intelligent energy-saving management system for charging piles based on big data provided in the above embodiment can allocate the above functions to different functional modules as needed, that is, divide the internal structure of the system into different functional modules to complete all or part of the functions described above, and this is not limited here.

[0036] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented by software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state hard disk.

[0037] It should be understood that the term "and / or" in this article is only a description of the association relationship of associated objects, indicating that there can be three relationships. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. A and B can be singular or plural. In addition, the character " / " in this article generally indicates that the associated objects before and after are in an "or" relationship, but it may also indicate an "and / or" relationship. Please refer to the context for specific understanding.

[0038] In this application, "at least one" means one or more, and "more than one" means two or more. "At least one of the following" or similar expressions refers to any combination of these items, including any combination of single or plural items. For example, at least one of a, b, or c can be represented by: a, b, c, ab, ac, bc, or abc, where a, b, c can be single or multiple.

[0039] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0040] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0041] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0042] In the several embodiments provided in the present application, it should be understood that the disclosed system can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0043] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0044] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0045] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: various media that can store program codes, such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks or optical disks.

[0046] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. Intelligent energy-saving management method for charging piles based on big data, characterized in that: include: Collect parameter data of charging piles and connected charging equipment, form data sets and perform data preprocessing; constructing an input feature matrix according to the data set, inputting the input feature matrix into a trained charging power prediction deep learning model, and outputting a predicted charging power at the next moment; Optimize the target according to the predicted charging power at the next moment and adjust the charging power of the charging pile for energy-saving management; When charging, the target battery power for the next time interval is calculated based on the current charging power. When the next time interval is reached, the actual battery power is obtained, and the charging power is compensated according to the actual battery power and the target battery power. After charging is completed, the health status of the charging pile is monitored and a charging pile health status coefficient is generated. When the charging pile health status coefficient is less than a preset health threshold, a charging pile warning signal is generated and the charging pile is inspected and maintained.

2. The intelligent energy-saving management method for charging piles based on big data according to claim 1 is characterized in that: The parameter data includes timestamp, charging power, charging load, battery power, charging device identification, battery temperature, battery capacitance and battery internal resistance. Data preprocessing includes cleaning the data set and removing outliers and missing values.

3. The intelligent energy-saving management method for charging piles based on big data according to claim 2 is characterized in that: Constructing a charging feature vector according to the data set, including: constructing charging features according to the data set, the charging features including charging power features, charging load features, battery power features, charging conversion features, battery health features, and temperature sensitivity features; A charging feature vector is constructed based on the charging characteristics. The timestamp is used as a sequence. An input feature matrix is ​​constructed based on the charging feature vector. The charging power feature is defined as the target output. Data alignment is performed. The data alignment predicts the charging power at the next moment by shifting the input feature matrix and the target output by one bit.

4. The intelligent energy-saving management method for charging piles based on big data according to claim 3 is characterized in that: The calculation logic of the charging conversion characteristic is: ,in, For the The charging conversion characteristics of the connected charging device, for The amount of battery power increase over time, is the input charging power; The calculation logic of the battery health feature is: in, For the Battery health characteristics of each connected charging device, is the actual capacity of the current battery, is the nominal capacity of the battery, is the current internal resistance of the battery, is the standard battery internal resistance, and is the weight coefficient, and and are greater than zero; The calculation logic of the temperature sensitive feature is: ,in, For the The temperature-sensitive characteristics of the connected charging device is the current battery temperature, For optimal battery charging temperature, is the temperature sensitivity coefficient.

5. The intelligent energy-saving management method for charging piles based on big data according to claim 3 is characterized in that: The training of the charging power prediction deep learning model includes: The acquired data set is divided into a training set and a test set, and an input feature matrix and a target output are constructed according to the training set; The input feature matrix and the target output are passed to the charging power prediction deep learning model for training, wherein the charging power prediction deep learning model includes a charging power prediction calculation unit, an output unit, a loss function unit and a model evaluation unit, wherein the charging power prediction calculation unit is used to update the cell state and the hidden state through the forgetting gate, the input gate, the candidate state generation and the output gate, and the output unit is used to perform a linear transformation according to the weight matrix and the bias of the hidden state and the output layer to obtain the target output, the loss function unit is used to define the loss function and update the parameters of the model according to the loss function, and the model evaluation unit is used to evaluate the performance of the model and check the degree of fit of the model to the data; The training is performed until the maximum iteration is reached or the evaluation index meets the preset requirements, and the final parameters are set as model parameters to complete the training of the charging power prediction deep learning model.

6. The intelligent energy-saving management method for charging piles based on big data according to claim 1 is characterized in that: The target optimization is performed according to the predicted charging power at the next moment, wherein the target optimization function is defined as: , To solve the charging power, the objective function minimize, is the number of connected charging devices, For the connected charging devices at the time The optimal charging power, For the The target battery level of each connected charging device, For the connected charging devices in The battery level at and is the start time and end time of the time window, and is the weight coefficient, which is used to adjust the balance between energy consumption and power deviation; The constraints of the objective optimization function include: , ; ; ,in, and For the The minimum and maximum power ratings of the connected charging devices, It is the smoothing amount of charging power, which is used to introduce adjustment space during charging power control to avoid frequent power fluctuations. For the connected charging devices at the time The predicted charging power, For the connected charging devices in The battery level at For the The connected charging device corresponds to The charging efficiency of a charging pile, the value range is (0,1), For the moment The total power constraint of the power grid.

7. The intelligent energy-saving management method for charging piles based on big data according to claim 6 is characterized in that: When solving the above objective optimization function, the objective optimization function contains integral and time continuous variables. Since the obtained parameter data is discrete data with timestamps as the sequence, the time interval [ , ] is discretized into K hourly steps, converting the continuous time optimization into a discrete optimization, and dividing the time interval into Small step length: ; Use discrete charging power values Approximately replace continuous ; Convert the integral to sum form: ; ; After discretization, the objective function becomes: ; Solve the discretized optimization problem through numerical optimization algorithms, including gradient descent method, Lagrange multiplier method, dynamic programming and genetic algorithm; Solve to get the optimal charging power According to the optimal charging power The collection of charging devices connected to it is charged.

8. The intelligent energy-saving management method for charging piles based on big data according to claim 1 is characterized in that: Also includes: When charging, based on the optimal charging power Calculate the target battery power for the next time interval. When the next moment is reached, obtain the actual battery power. Compensate the charging power according to the actual battery power and the target battery power. The calculation logic of the target battery power at the next moment is: ,in, For the next time interval The target battery level after For the connected charging devices in The battery power at the moment; the charging power is compensated according to the actual battery power and the target battery power. The calculation logic of the compensation amount is: ,in, for The charging power compensation amount at the moment is used to compensate Charging power at the moment, for The actual battery level at the moment, for Target battery level at the moment, is the proportional coefficient, which is used to control the compensation strength.

9. The intelligent energy-saving management method for charging piles based on big data according to claim 1 is characterized in that: Also includes: The health status of the charging pile is monitored to generate a health status coefficient of the charging pile. When the health status coefficient of the charging pile is less than a preset health threshold, the charging pile is repaired. The calculation logic of the health status coefficient of the charging pile is: ,in, For the The health status coefficient of the charging pile of each charging pile, for The current at the moment, For the The rated current of each charging pile, For and Take the maximum value, For the Charging piles in the time period The average current, and is the weight coefficient, which is used to adjust the impact of current integral and current fluctuation on the health index.

10. A charging pile intelligent energy-saving management system based on big data, used to implement the charging pile intelligent energy-saving management method based on big data according to any one of claims 1 to 9, characterized in that: include: Acquisition module: collects parameter data of charging piles and connected charging equipment, forms a data set and performs data preprocessing; Prediction module: construct an input feature matrix according to the data set, input the input feature matrix into the trained charging power prediction deep learning model, and output the predicted charging power at the next moment; Energy-saving module: optimize the target according to the predicted charging power at the next moment, and adjust the charging power of the charging pile for energy-saving management; Compensation module: When charging, the target battery power of the next time interval is calculated based on the current charging power. When the next time interval is reached, the actual battery power is obtained, and the charging power is compensated according to the actual battery power and the target battery power. Monitoring module: After charging is completed, the health status of the charging pile is monitored and the health status coefficient of the charging pile is generated. When the health status coefficient of the charging pile is less than the preset health threshold, a charging pile warning signal is generated and the charging pile is inspected and maintained.

Images