CN117932519B - New energy automobile charging station monitored control system - Google Patents

Abstract

The invention relates to the technical field of intelligent charging network management and new energy automobile support systems, in particular to a new energy automobile charging station monitoring system. In the invention, the energy management module adopting the convolutional neural network and the long-term and short-term memory network improves the power demand and the power grid load prediction accuracy and improves the energy utilization rate. The safety early warning module combines a support vector machine, a decision tree and an isolated forest algorithm, so that the system safety is improved. In the aspect of pricing optimization, charging prices are adjusted according to power grid loads and user demands by using an ARIMA model and a Q learning algorithm, so that fairness and market response are improved. The charging scheduling module improves and adopts linear programming and genetic algorithm, and charging efficiency is improved. The data acquisition and processing analysis module is upgraded to provide comprehensive and accurate monitoring data.

Description

New energy automobile charging station monitored control system

Technical Field

The invention relates to the technical field of intelligent charging network management and new energy automobile support systems, in particular to a new energy automobile charging station monitoring system.

Background

The intelligent charging network management and new energy automobile support system is a technical field which is focused on optimizing the charging process and infrastructure of the new energy automobile. This field encompasses a range of technologies including, but not limited to, intelligent charging station design, energy management systems, data analysis and prediction, remote monitoring and control technologies. The core aim is to realize an efficient, reliable and economical charging solution to support the wide use and popularization of new energy automobiles.

A new energy automobile charging station monitoring system is a monitoring solution specially designed for new energy automobile charging stations. It generally includes hardware and software for real-time monitoring of charging station operation, such as status monitoring, energy consumption monitoring, safety monitoring, etc. of charging piles, and is mainly aimed at improving the operation efficiency and reliability of charging stations, ensuring the safety of charging process, and providing user-friendly service experience. Through monitored control system, can real-time supervision charging process, ensure charging efficiency and security, simultaneously discern fast and respond trouble or the potential safety hazard of any facility that charges to collect and analyze charging data, with the management and the operation of optimizing the charging station, reach the function of reinforcing user experience.

The existing new energy automobile charging station monitoring system lacks high-efficiency load prediction and strategy optimization capability in the aspect of energy management, so that the power grid is overloaded and energy is wasted. In terms of safety, the traditional system does not have advanced fault prediction and abnormality detection capabilities, and increases the running risk of the system. In terms of pricing strategies, existing systems cannot dynamically adjust prices according to real-time power grids and user demands, resulting in unfair prices or insufficient market reactions. The efficiency and accuracy of the charging schedule are inadequate, resulting in an poor user experience and insufficient resource utilization. The limited ability to collect and process analysis of data affects overall system optimization and decision making. Finally, the lack of a user interface results in user difficulty in understanding and operating the system, affecting user satisfaction.

Disclosure of Invention

The invention aims to solve the defects in the prior art and provides a new energy automobile charging station monitoring system.

In order to achieve the above purpose, the present invention adopts the following technical scheme: the new energy automobile charging station monitoring system comprises an energy management module, a safety early warning module, a pricing optimization module, a charging scheduling module, a data acquisition module, a data processing analysis module and a user interface module;

The energy management module performs load prediction by adopting a convolutional neural network and a long-term and short-term memory network based on power demand and power grid load data, performs strategy optimization by combining with Monte Carlo tree search, and generates an optimized charging strategy;

The safety early warning module establishes a fault prediction model by applying a support vector machine and a decision tree based on equipment state and environment monitoring data, and uses an isolated forest algorithm to perform abnormality detection to generate safety early warning information;

the pricing optimization module is used for carrying out demand prediction by adopting an ARIMA model based on an optimized charging strategy and user data, and adjusting charging pricing by combining Q learning to generate a dynamic charging price;

The charging scheduling module performs charging scheduling by using a linear programming and genetic algorithm based on an optimized charging strategy and a dynamic charging price to generate an optimized charging schedule;

The data acquisition module is used for generating real-time acquisition data based on the sensor, the integrated equipment state and the user behavior data;

the data processing analysis module is used for carrying out association analysis by matching with a Bayesian network based on real-time acquisition data to generate an association analysis result;

The user interface module generates a user-friendly operation interface based on an optimized charging strategy, safety early warning information, dynamic charging price, optimized charging schedule and correlation analysis result by combining a back propagation neural network.

As a further scheme of the invention, the optimized charging strategy is to dynamically adjust a charging plan according to a predicted load and a user demand, the safety early warning information is to dynamically adjust a charging rate according to a grid load and a user demand, the optimized charging schedule comprises a charging sequence and time, the real-time collected data comprises a device running state, an environmental parameter and a user charging behavior, and the user-friendly operation interface comprises a charging state display, an early warning information prompt, a dynamic price display and a charging schedule display.

As a further scheme of the invention, the energy management module comprises a load prediction sub-module, a strategy optimization sub-module and a charging decision sub-module;

the safety early warning module comprises a fault prediction sub-module, an abnormality detection sub-module and an early warning generation sub-module;

the pricing optimization module comprises a charging demand prediction sub-module, a pricing strategy optimization sub-module and a pricing decision sub-module;

The charging scheduling module comprises a scheduling demand prediction sub-module, a scheduling strategy optimization sub-module and a scheduling decision sub-module;

the data acquisition module comprises an equipment state acquisition sub-module, an environment data acquisition sub-module and a user behavior acquisition integration sub-module;

The data processing analysis module comprises a data integration sub-module, a data preprocessing sub-module and a data relevance analysis sub-module;

the user interface module comprises a strategy display sub-module, an early warning display sub-module, a price display sub-module and a scheduling display sub-module.

As a further scheme of the invention, the load prediction submodule carries out power load prediction by adopting a convolutional neural network and a long-term and short-term memory network based on power demand and power grid load data to generate a power load prediction result;

The strategy optimization submodule performs charging strategy optimization by adopting Monte Carlo tree search based on the power load prediction result to generate a preliminary optimized charging strategy result;

and the charging decision sub-module performs charging decision based on the preliminary optimized charging strategy result to generate an optimized charging strategy.

As a further scheme of the invention, the fault prediction submodule adopts a support vector machine and a decision tree to conduct fault prediction based on equipment state and environment monitoring data so as to generate a fault prediction result;

The abnormality detection submodule carries out abnormality detection by adopting an isolated forest algorithm based on a fault prediction result to generate an abnormality detection result;

and the early warning generation sub-module generates safety early warning information based on the abnormal detection result.

As a further scheme of the invention, the charging demand prediction submodule adopts an autoregressive integral moving average model to predict the charging demand based on an optimized charging strategy and user data, and generates a charging demand prediction result;

the pricing strategy optimization submodule carries out charging pricing adjustment by adopting Q learning based on the charging demand prediction result to generate a pricing strategy optimization result;

And the pricing decision sub-module performs charging pricing decision based on the pricing strategy optimization result to generate dynamic charging price.

As a further scheme of the invention, the scheduling demand prediction sub-module predicts the charging scheduling demand based on an optimized charging strategy and a dynamic charging price, and generates a scheduling demand prediction result;

the scheduling strategy optimization submodule performs charging scheduling optimization by adopting a linear programming and genetic algorithm based on the scheduling demand prediction result to generate a scheduling strategy optimization result;

And the scheduling decision sub-module performs charging scheduling decision based on the scheduling policy optimization result to generate optimized charging schedule.

As a further scheme of the invention, the equipment state acquisition submodule is used for collecting real-time equipment state data based on the sensor to generate equipment state acquisition data;

the environment data acquisition submodule is used for collecting environment data based on the sensor and generating an environment data acquisition result;

The user behavior acquisition and integration submodule is used for collecting user behavior data based on the sensor and generating real-time acquisition data by combining equipment state acquisition data and environmental data acquisition results.

As a further scheme of the invention, the data integration submodule performs data integration based on real-time acquisition data to generate a data integration result;

The data preprocessing sub-module performs data cleaning and formatting processing based on the data integration result to generate a data preprocessing result;

And the data relevance analysis submodule carries out data relevance analysis by adopting a Bayesian network based on the data preprocessing result to generate a relevance analysis result.

As a further scheme of the invention, the strategy display submodule generates a strategy visual interface by adopting a back propagation neural network based on an optimized charging strategy;

the early warning display submodule generates an early warning information display interface based on the safety early warning information and the strategy visual interface by applying a data analysis method;

The price display sub-module generates a dynamic price display interface by utilizing a real-time data processing technology based on the dynamic charging price and the early warning information display interface;

the scheduling display sub-module generates a user-friendly operation interface by adopting a data visualization technology based on a dynamic price display interface, the optimized charging schedule and a correlation analysis result.

Compared with the prior art, the invention has the advantages and positive effects that:

According to the invention, by adopting the convolutional neural network and the long-term and short-term memory network, the energy management module improves the prediction accuracy of the power demand and the power grid load data, so that the charging strategy is more efficient and flexible, the power grid pressure is reduced by strategy optimization, and the energy utilization rate is improved. The safety early warning module adopts a support vector machine and a decision tree, combines an isolated forest algorithm, enhances the capability of fault prediction and anomaly detection, and improves the safety of the whole system. In the aspect of pricing optimization, through an ARIMA model and a Q learning algorithm, the charging price can be dynamically adjusted according to the power grid load and the user demand, and the price fairness and the market response capability are improved. The improvement of the charging scheduling module comprises the application of linear programming and genetic algorithm, so that the charging process is more efficient, and the waiting time of a user is reduced. The upgrade of the data acquisition and processing analysis module also enables the monitoring data to be more comprehensive and accurate, and reliable data support is provided for the optimization of the whole system. Optimization of the user interface module promotes the user experience so that the user can more easily understand and operate the system.

Drawings

FIG. 1 is a system flow diagram of the present invention;

FIG. 2 is a schematic diagram of a system framework of the present invention;

FIG. 3 is a flow chart of an energy management module according to the present invention;

FIG. 4 is a flow chart of a safety precaution module of the present invention;

FIG. 5 is a flow chart of a pricing optimization module of the present invention;

FIG. 6 is a flow chart of a charge scheduling module according to the present invention;

FIG. 7 is a flow chart of a data acquisition module according to the present invention;

FIG. 8 is a flow chart of a data processing analysis module of the present invention;

FIG. 9 is a flowchart of a user interface module of the present invention.

Detailed Description

The present invention will be described in further detail with reference to the drawings and examples, in order to make the objects, technical solutions and advantages of the present invention more apparent. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the scope of the invention.

In the description of the present invention, it should be understood that the terms "length," "width," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," and the like indicate orientations or positional relationships based on the orientation or positional relationships shown in the drawings, merely to facilitate describing the present invention and simplify the description, and do not indicate or imply that the devices or elements referred to must have a specific orientation, be configured and operated in a specific orientation, and therefore should not be construed as limiting the present invention. Furthermore, in the description of the present invention, the meaning of "a plurality" is two or more, unless explicitly defined otherwise.

Examples

Referring to fig. 1, a new energy vehicle charging station monitoring system includes an energy management module, a safety early warning module, a pricing optimization module, a charging scheduling module, a data acquisition module, a data processing analysis module and a user interface module;

The energy management module performs load prediction by adopting a convolutional neural network and a long-term and short-term memory network based on power demand and power grid load data, and performs strategy optimization by combining with Monte Carlo tree search to generate an optimized charging strategy;

The safety early warning module establishes a fault prediction model by applying a support vector machine and a decision tree based on the equipment state and the environment monitoring data, and uses an isolated forest algorithm to perform anomaly detection so as to generate safety early warning information;

the pricing optimization module is used for carrying out demand prediction by adopting an ARIMA model based on an optimized charging strategy and user data, and adjusting charging pricing by combining Q learning to generate a dynamic charging price;

the charging scheduling module performs charging scheduling by using a linear programming and genetic algorithm based on an optimized charging strategy and a dynamic charging price to generate an optimized charging schedule;

the data acquisition module is used for generating real-time acquisition data based on the sensor, the integrated equipment state and the user behavior data;

the data processing analysis module is used for carrying out association analysis by matching with a Bayesian network based on real-time acquisition data to generate an association analysis result;

the user interface module generates a user-friendly operation interface based on an optimized charging strategy, safety early warning information, dynamic charging price, optimized charging schedule and correlation analysis result by combining a back propagation neural network.

The optimized charging strategy is characterized in that a charging plan is dynamically adjusted according to predicted load and user demands, safety early warning information is a real-time alarm of abnormal fault behaviors, the dynamic charging price is a charging rate which is dynamically adjusted according to grid load and user demands, the charging schedule after optimization comprises a charging sequence and time, real-time collected data comprise equipment running states, environment parameters and user charging behaviors, and a user-friendly operation interface comprises charging state display, early warning information prompt, dynamic price display and charging schedule display.

The system is in the aspects of energy management, safety early warning, pricing optimization, charging scheduling, user interface and the like. By adopting an advanced prediction model and algorithm, the system can accurately predict the load and optimize the charging strategy, reduce the load of the power grid and improve the utilization efficiency of the power. Meanwhile, the safety early warning module based on the equipment state and the environment monitoring data can timely find faults and abnormal conditions, and the safety of the charging process is guaranteed. And the pricing optimization module dynamically adjusts the charging price according to the user demands and the power grid load, so that the economy and the user satisfaction are improved. The charging scheduling module reasonably arranges the charging sequence and time, improves the charging efficiency and reduces the waiting time of users. Finally, the user-friendly operation interface enables the user to intuitively know the charging state, the early warning information, the dynamic price and the charging scheduling condition, and improves the user experience and participation. In conclusion, the system can effectively improve the energy utilization efficiency, ensure the charging safety and improve the user satisfaction and the economic benefit.

Referring to fig. 2, the energy management module includes a load prediction sub-module, a policy optimization sub-module, and a charging decision sub-module;

the safety early warning module comprises a fault prediction sub-module, an abnormality detection sub-module and an early warning generation sub-module;

the pricing optimization module comprises a charging demand prediction sub-module, a pricing strategy optimization sub-module and a pricing decision sub-module;

the charging scheduling module comprises a scheduling demand prediction sub-module, a scheduling strategy optimization sub-module and a scheduling decision sub-module;

the data acquisition module comprises an equipment state acquisition sub-module, an environment data acquisition sub-module and a user behavior acquisition integration sub-module;

the data processing analysis module comprises a data integration sub-module, a data preprocessing sub-module and a data relevance analysis sub-module;

the user interface module comprises a strategy display sub-module, an early warning display sub-module, a price display sub-module and a scheduling display sub-module.

In the energy management module, a load prediction submodule adopts a convolutional neural network and a long-term and short-term memory network to perform load prediction based on power demand and power grid load data; the strategy optimization sub-module performs strategy optimization by combining Monte Carlo tree search to generate an optimized charging strategy; and the charging decision sub-module carries out charging decision according to the optimized charging strategy.

In the safety early warning module, a fault prediction submodule applies a support vector machine and a decision tree to establish a fault prediction model, and an isolated forest algorithm is used for abnormality detection; the abnormality detection submodule generates safety precaution information; and the early warning generation sub-module generates safety early warning information according to the abnormal detection result.

In the pricing optimization module, a charging demand prediction sub-module adopts an ARIMA model to predict demand; the pricing strategy optimization sub-module adjusts charging pricing in combination with Q learning; the pricing decision sub-module performs pricing decisions based on the dynamic charge prices.

In the charging scheduling module, a scheduling demand prediction sub-module predicts scheduling demands according to an optimized charging strategy and a dynamic charging price; the scheduling strategy optimization submodule performs charging scheduling by using linear programming and a genetic algorithm; the schedule decision sub-module generates an optimized charge schedule.

In the data acquisition module, an equipment state acquisition submodule integrates equipment state data; the environment data acquisition submodule integrates environment monitoring data; the user behavior acquisition and integration sub-module integrates user behavior data to generate real-time acquisition data.

In the data processing analysis module, a data integration sub-module integrates the real-time acquired data; the data preprocessing submodule preprocesses the integrated data; the data association analysis sub-module cooperates with the Bayesian network to carry out association analysis and generate association analysis results.

In the user interface module, a strategy display submodule displays an optimized charging strategy; the early warning display submodule displays safety early warning information; the price display sub-module displays the dynamic charging price; the schedule display sub-module displays the optimized charging schedule; the user-friendly operation interface is generated through the back propagation neural network and provided for a user to operate.

Referring to fig. 3, the load prediction sub-module performs power load prediction by adopting a convolutional neural network and a long-term and short-term memory network based on power demand and power grid load data to generate a power load prediction result;

The strategy optimization submodule performs charging strategy optimization by adopting Monte Carlo tree search based on the power load prediction result to generate a preliminary optimized charging strategy result;

And the charging decision-making submodule carries out charging decision making based on the preliminary optimized charging strategy result to generate an optimized charging strategy.

The load prediction submodule predicts the power load by adopting a convolutional neural network and a long-term and short-term memory network based on the power demand and the power grid load data. First, historical power demand and grid load data are collected and data cleansing and preprocessing is performed. And then, inputting the processed data into a convolutional neural network for feature extraction and pattern recognition to obtain a preliminary power load prediction result. And then taking the preliminary prediction result as input, and carrying out time sequence modeling and prediction through a long-short-period memory network to generate a final power load prediction result.

The strategy optimization submodule adopts Monte Carlo tree search to carry out charging strategy optimization based on the power load prediction result. First, the power demand and the available power amount for each period are determined according to the power load prediction result. Then, an objective function of charging strategy optimization is established, and factors such as charging cost, user satisfaction and the like are considered. Next, a monte carlo search algorithm is used to search and optimize the charging strategy, and the optimal charging strategy is found by continuously iterating and adjusting the strategy parameters. And finally, generating a preliminary charging strategy result.

And the charging decision sub-module is used for carrying out charging decision and generating an optimal charging strategy based on the optimal charging strategy result. Firstly, determining specific decisions such as charging site selection, charging pile distribution and the like of each time period according to a preliminary charging strategy result. And then, combining the real-time power demand and the power grid load data to dynamically adjust and optimize the charging decision. Meanwhile, the personalized charging strategy is formulated by considering factors such as charging requirements and behavior habits of users. And finally, generating an optimized charging strategy, including specific charging station selection, charging pile allocation schemes and the like.

Referring to fig. 4, the fault prediction sub-module performs fault prediction by using a support vector machine and a decision tree based on the equipment state and the environmental monitoring data to generate a fault prediction result;

the abnormality detection sub-module is used for carrying out abnormality detection by adopting an isolated forest algorithm based on a fault prediction result to generate an abnormality detection result;

the early warning generation sub-module generates safety early warning information based on the abnormal detection result.

In the fault prediction sub-module, firstly, equipment state and environment monitoring data are collected, and characteristics related to faults are extracted. The data set is then divided into a training set and a test set. And constructing a fault prediction model by using a support vector machine algorithm, and training and adjusting the super parameters. And finally, predicting the test set to generate a fault prediction result and evaluating the performance of the model.

In the abnormality detection sub-module, firstly, a fault prediction result is obtained from the fault prediction sub-module and is used as a known fault label. Data that requires anomaly detection, such as real-time equipment status and environmental monitoring data, is then collected. And based on the fault prediction result, performing abnormality detection by using an isolated forest algorithm. By constructing multiple decision trees, the path length of each sample on the tree can be used as a measure of the degree of anomaly. And determining which samples belong to the abnormality according to the set abnormality threshold. And finally, outputting an abnormality detection result, marking the abnormality as an abnormal point or generating corresponding alarm information.

In the early warning generation sub-module, firstly, an abnormality detection result is obtained from the abnormality detection sub-module. And then, generating corresponding safety early warning information according to the abnormal detection result. Different early warning levels and corresponding processing measures can be set according to different abnormality types and degrees. For example, for severe anomalies, emergency alert notification-related personnel are sent for processing. And finally, transmitting the generated safety early warning information to related personnel or systems to prompt timely measures to treat potential safety problems.

Referring to fig. 5, the charging demand prediction sub-module performs charging demand prediction by adopting an autoregressive integral moving average model based on an optimized charging strategy and user data, and generates a charging demand prediction result;

The pricing strategy optimization submodule carries out charging pricing adjustment by adopting Q learning based on the charging demand prediction result to generate a pricing strategy optimization result;

and the pricing decision sub-module performs charging pricing decision based on the pricing strategy optimization result to generate a dynamic charging price.

In the charging demand prediction sub-module, historical charging demand data and user data are collected first. The data may include a charge amount of the charging station, a charge time, user information, and the like. Features related to the charging demand are then extracted from the collected data. Time series analysis methods such as sliding windows, fourier transforms, etc. may be used to extract features. An autoregressive integral moving average model (ARIMA) is used to construct a charging demand prediction model. And carrying out stationarity test and differential processing on the historical charging demand data so as to meet the assumption condition of the ARIMA model. And (5) carrying out model training and verification according to the parameter estimation result of the ARIMA model. And predicting new data by using the trained model to generate a charging demand prediction result. And evaluating the performance of the charging demand prediction model. The accuracy and stability of the model is assessed using common evaluation criteria such as Root Mean Square Error (RMSE), mean Absolute Percentage Error (MAPE), etc.

In the pricing strategy optimization sub-module, a charging demand prediction result is obtained from the charging demand prediction sub-module. Data related to charge pricing, such as charge cost, market price, etc., is collected. Meanwhile, information such as a reward function and state transition probability required by Q learning is also required to be prepared. And establishing a Q learning model based on the charging demand prediction result and the collected data. And determining a state space, an action space and a reward function, and setting parameters such as a discount factor, a learning rate and the like according to actual conditions. The Q value table is continuously updated by interacting with the environment. Based on the current state and the selected action, a prize value is calculated and the Q value is updated based on the Belman equation. This process is repeated until convergence or a preset number of exercises is reached. And generating a pricing strategy optimization result according to the trained Q learning model. The optimal pricing strategy can be determined based on different states and actions.

In the pricing decision sub-module, first, a pricing strategy optimization result is obtained from the pricing strategy optimization sub-module.

And according to the pricing strategy optimization result, making a decision of the dynamic charging price. And determining the optimal charging price according to the current market conditions, the user requirements and other factors. This may be achieved by mapping the pricing policies onto specific pricing schemes. And feeding back an actual pricing decision result to the system, and updating the information of the dynamic charging price. This can be used for further analysis and optimization.

Referring to fig. 6, the scheduling demand prediction sub-module performs charging scheduling demand prediction based on an optimized charging strategy and a dynamic charging price, and generates a scheduling demand prediction result;

the scheduling strategy optimization submodule carries out charging scheduling optimization by adopting a linear programming and genetic algorithm based on the scheduling demand prediction result to generate a scheduling strategy optimization result;

And the scheduling decision sub-module performs charging scheduling decision based on the scheduling policy optimization result to generate optimized charging schedule.

In the scheduling demand prediction sub-module, features related to charging scheduling demands are extracted from data collected first. Time series analysis methods such as sliding windows, fourier transforms, etc. may be used to extract features. And constructing a scheduling demand prediction model by adopting a regression model (such as linear regression, support vector regression and the like). Model training is performed using the training set, and super parameters of the model are adjusted to obtain optimal predictive performance. And predicting the test set to generate a scheduling demand prediction result. And evaluating the performance of the scheduling demand prediction model. The accuracy and stability of the model is assessed using common evaluation criteria such as Root Mean Square Error (RMSE), mean Absolute Percentage Error (MAPE), etc.

In the scheduling strategy optimization sub-module, a scheduling demand prediction result is firstly obtained from the scheduling demand prediction sub-module. Information of the charging station is collected, including the number, capacity, position and the like of the charging piles. At the same time, constraints required for optimizing the objective function, such as maximum load limit of the charging station, power limit of the charging device, etc., have to be prepared. And constructing a charging schedule optimization model by adopting a Linear Programming (LP) algorithm. And according to the scheduling demand prediction result and constraint conditions, establishing an objective function and constraint conditions of the linear programming, and aiming at minimizing the charging cost or maximizing the system benefit. And solving the constructed model by using a linear programming solver to obtain an optimal charging scheduling strategy. And evaluating the performance of the optimized charging scheduling strategy. The difference between different strategies, such as charging cost, user satisfaction, etc., can be compared.

And in the scheduling decision sub-module, an optimized charging scheduling strategy result is obtained from the scheduling strategy optimization sub-module. And carrying out actual charging scheduling decision according to the optimized charging scheduling strategy result. According to the current demand conditions and the position information of the vehicles, determining which vehicles need to be charged and on which charging piles. And feeding back an actual scheduling decision result to the system, and updating the state information of the vehicle and the state information of the charging station. For further analysis and optimization.

Referring to fig. 7, the device state acquisition submodule performs device real-time state data collection based on the sensor to generate device state acquisition data;

The environmental data acquisition submodule is used for collecting environmental data based on the sensor and generating an environmental data acquisition result;

The user behavior acquisition and integration submodule is used for collecting user behavior data based on the sensor and generating real-time acquisition data by combining equipment state acquisition data and environmental data acquisition results.

In the equipment state acquisition submodule, a proper sensor is selected first and is installed on equipment to be monitored. Ensuring that the sensor is able to accurately measure real-time status data of the device. Real-time status data of the device is collected by the sensor. This may include information on the operating status of the device, temperature, humidity, etc. And transmitting the acquired equipment state data. The data may be transmitted to a data processing center or cloud storage using wired or wireless communication. And storing the collected equipment state data. A database or other suitable storage may be used to ensure the security and accessibility of the data.

In the environment data acquisition sub-module, a proper sensor is selected first and is installed in an environment to be monitored. Ensuring that the sensor is able to accurately measure various parameters of the environment. Environmental data is collected by the sensors. This may include information such as temperature, humidity, illumination intensity, etc. And transmitting the collected environmental data. The data may be transmitted to a data processing center or cloud storage using wired or wireless communication. And storing the collected environmental data. A database or other suitable storage may be used to ensure the security and accessibility of the data.

In the user behavior acquisition and integration sub-module, a proper sensor is selected first and is installed at a position where the user behavior to be monitored occurs. Ensuring that the sensor can accurately capture the behavior information of the user. Behavior data of the user is collected by the sensor. Information such as a movement trace, an operation behavior of the user, and the like may be included. And integrating the acquired user behavior data with the equipment state data and the environmental data acquisition result. The data from the different sources may be matched and correlated by a time stamp or other correlation information. And transmitting and storing the integrated data. The data may be transmitted to a data processing center or cloud storage using wired or wireless communication.

Referring to fig. 8, the data integration sub-module performs data integration based on real-time collected data to generate a data integration result;

The data preprocessing sub-module performs data cleaning and formatting processing based on the data integration result to generate a data preprocessing result;

And the data relevance analysis submodule carries out data relevance analysis by adopting a Bayesian network based on the data preprocessing result to generate a relevance analysis result.

In the data integration sub-module, real-time acquisition data is acquired from the equipment state acquisition sub-module, the environment data acquisition sub-module and the user behavior acquisition integration sub-module. The real-time acquisition data of different sources are integrated. The data of the different data sources may be matched and correlated using time stamps or other correlation information. And converting the format of the integrated data to ensure the consistency and the processibility of the data. For example, the data is converted into a unified date and time format or CSV format. And storing the integrated data. A database or other suitable storage may be used to ensure the security and accessibility of the data.

In the data preprocessing submodule, a data set required by data preprocessing is loaded from the data integration submodule, and the data is subjected to cleaning operation, including removal of repeated values, processing of missing values and the like. Data cleansing may be performed using related functions and methods in a data processing tool or programming language. And formatting the data to enable the data to meet the requirement of subsequent analysis. For example, the date and time field is converted into a numeric field, and standardized processing is performed. And selecting and extracting the characteristics of the data according to specific requirements. Feature selection and extraction may be performed using statistical analysis methods, machine learning algorithms, and the like. The data is then transformed, normalized, etc., as needed. And the performance and stability of the model are improved. And storing the preprocessed data. A database or other suitable storage may be used to ensure the security and accessibility of the data.

In the data relevance analysis submodule, a data set required by relevance analysis is loaded from the data preprocessing submodule. Based on the data preprocessing result, a Bayesian network algorithm is used for constructing a relevance analysis model. Model construction may be performed using specialized bayesian network libraries or tools. And learning parameters of the Bayesian network model through the training data set, and performing inference analysis. Parameter learning and inference can be performed using maximum likelihood estimation, expectation maximization, and the like. And generating a relevance analysis result according to the inferred result of the Bayesian network model. Information such as a relevance index, a conditional probability distribution, and the like can be output. And evaluating and optimizing the generated association analysis result. Results are analyzed and evaluated using visualization tools, statistical indicators, etc., and model adjustments and optimizations are made as needed.

Referring to fig. 9, the policy presentation submodule generates a policy visualization interface based on an optimized charging policy by using a back propagation neural network;

The early warning display sub-module generates an early warning information display interface by applying a data analysis method based on the safety early warning information and the strategy visualization interface;

The price display sub-module generates a dynamic price display interface by utilizing a real-time data processing technology based on the dynamic charging price and the early warning information display interface;

The scheduling display sub-module generates a user-friendly operation interface by adopting a data visualization technology based on a dynamic price display interface, the optimized charging schedule and the correlation analysis result.

And in the strategy display sub-module, firstly, acquiring an optimized charging strategy result from the optimized charging strategy sub-module. And constructing a strategy display model by adopting a back propagation neural network. A neural network model may be built using a deep learning framework, such as TensorFlow or PyTorch. The training data set is used to train the game display model and the super parameters of the model are adjusted to obtain the best performance. And designing a user-friendly strategy visualization interface according to the result of the strategy display model. The policy results are presented to the user in the form of charts, maps, etc., using a relational library in the data visualization tool or programming language.

And in the early warning display sub-module, acquiring the safety early warning information and data required by the strategy visual interface from the safety early warning information sub-module. And selecting a proper data analysis method, such as statistical analysis, a machine learning algorithm and the like, according to specific requirements. And processing and analyzing the safety early warning information according to the selected data analysis method, and extracting key indexes and characteristics. And designing a user-friendly early warning information display interface according to the analysis result. The pre-warning information is presented to the user in the form of a chart, table, etc., using a relational library in a data visualization tool or programming language.

And in the price display sub-module, data required by the dynamic charging price and early warning information display interface are acquired from the dynamic charging price grid module. And selecting proper real-time data processing technology, such as streaming computing, message queue and the like, according to specific requirements. The dynamic charge price is processed and calculated in real time using selected real-time data processing techniques. And designing a user-friendly dynamic price display interface according to the calculation result. Dynamic prices are presented to users in the form of charts, curves, etc., using relational libraries in data visualization tools or programming languages.

And in the scheduling display sub-module, required data is acquired from the dynamic price display interface, the optimized charging scheduling and relevance analysis result sub-module. Appropriate data visualization techniques, such as interactive charts, map visualization, etc., are selected according to particular needs. According to the data preparation and the selection of the visualization technology, a user-friendly operation interface is designed. Various interactive elements, such as buttons, drop-down menus, etc., are included so that a user can conveniently view and manipulate data. And using a proper programming language and development tools to realize the user-friendly operation interface of the scheduling display sub-module, and performing deployment and test. Ensuring the stability and usability of the interface.

The present invention is not limited to the above embodiments, and any equivalent embodiments which can be changed or modified by the technical disclosure described above can be applied to other fields, but any simple modification, equivalent changes and modification made to the above embodiments according to the technical matter of the present invention will still fall within the scope of the technical disclosure.

Claims (8)

1. A new energy automobile charging station monitored control system, its characterized in that: the new energy automobile charging station monitoring system comprises an energy management module, a safety early warning module, a pricing optimization module, a charging scheduling module, a data acquisition module, a data processing analysis module and a user interface module;

The energy management module performs load prediction by adopting a convolutional neural network and a long-term and short-term memory network based on power demand and power grid load data, performs strategy optimization by combining with Monte Carlo tree search, and generates an optimized charging strategy;

The safety early warning module establishes a fault prediction model by applying a support vector machine and a decision tree based on equipment state and environment monitoring data, and uses an isolated forest algorithm to perform abnormality detection to generate safety early warning information;

the pricing optimization module is used for carrying out demand prediction by adopting an ARIMA model based on an optimized charging strategy and user data, and adjusting charging pricing by combining Q learning to generate a dynamic charging price;

The charging scheduling module performs charging scheduling by using a linear programming and genetic algorithm based on an optimized charging strategy and a dynamic charging price to generate an optimized charging schedule;

The data acquisition module is used for generating real-time acquisition data based on the sensor, the integrated equipment state and the user behavior data;

the data processing analysis module is used for carrying out association analysis by matching with a Bayesian network based on real-time acquisition data to generate an association analysis result;

the user interface module generates a user-friendly operation interface based on an optimized charging strategy, safety early warning information, dynamic charging price, optimized charging schedule and correlation analysis result by combining a back propagation neural network;

The charging scheduling module comprises a scheduling demand prediction sub-module, a scheduling strategy optimization sub-module and a scheduling decision sub-module;

The scheduling demand prediction sub-module predicts the charging scheduling demand based on an optimized charging strategy and a dynamic charging price, and generates a scheduling demand prediction result;

the scheduling strategy optimization submodule performs charging scheduling optimization by adopting a linear programming and genetic algorithm based on the scheduling demand prediction result to generate a scheduling strategy optimization result;

And the scheduling decision sub-module performs charging scheduling decision based on the scheduling policy optimization result to generate optimized charging schedule.

2. The new energy vehicle charging station monitoring system of claim 1, wherein: the optimized charging strategy is characterized in that a charging plan is dynamically adjusted according to predicted load and user demands, the safety early warning information is specifically a real-time warning of abnormal fault behaviors, the dynamic charging price is specifically a charging rate which is dynamically adjusted according to grid load and user demands, the optimized charging schedule comprises a charging sequence and time, the real-time collected data comprise equipment running states, environment parameters and user charging behaviors, and the user-friendly operation interface comprises charging state display, early warning information prompt, dynamic price display and charging schedule display.

3. The new energy vehicle charging station monitoring system of claim 1, wherein: the energy management module comprises a load prediction sub-module, a strategy optimization sub-module and a charging decision sub-module;

the load prediction submodule performs power load prediction by adopting a convolutional neural network and a long-term and short-term memory network based on power demand and power grid load data to generate a power load prediction result;

The strategy optimization submodule performs charging strategy optimization by adopting Monte Carlo tree search based on the power load prediction result to generate a preliminary optimized charging strategy result;

and the charging decision sub-module performs charging decision based on the preliminary optimized charging strategy result to generate an optimized charging strategy.

4. The new energy vehicle charging station monitoring system of claim 1, wherein:

the safety early warning module comprises a fault prediction sub-module, an abnormality detection sub-module and an early warning generation sub-module;

The fault prediction submodule carries out fault prediction by adopting a support vector machine and a decision tree based on equipment state and environment monitoring data to generate a fault prediction result;

The abnormality detection submodule carries out abnormality detection by adopting an isolated forest algorithm based on a fault prediction result to generate an abnormality detection result;

and the early warning generation sub-module generates safety early warning information based on the abnormal detection result.

5. The new energy vehicle charging station monitoring system of claim 1, wherein: the pricing optimization module comprises a charging demand prediction sub-module, a pricing strategy optimization sub-module and a pricing decision sub-module;

The charging demand prediction submodule carries out charging demand prediction by adopting an autoregressive integral moving average model based on an optimized charging strategy and user data, and generates a charging demand prediction result;

the pricing strategy optimization submodule carries out charging pricing adjustment by adopting Q learning based on the charging demand prediction result to generate a pricing strategy optimization result;

And the pricing decision sub-module performs charging pricing decision based on the pricing strategy optimization result to generate dynamic charging price.

6. The new energy vehicle charging station monitoring system of claim 1, wherein: the data acquisition module comprises an equipment state acquisition sub-module, an environment data acquisition sub-module and a user behavior acquisition integration sub-module;

The equipment state acquisition submodule is used for collecting real-time equipment state data based on the sensor and generating equipment state acquisition data;

the environment data acquisition submodule is used for collecting environment data based on the sensor and generating an environment data acquisition result;

The user behavior acquisition and integration submodule is used for collecting user behavior data based on the sensor and generating real-time acquisition data by combining equipment state acquisition data and environmental data acquisition results.

7. The new energy vehicle charging station monitoring system of claim 1, wherein: the data processing analysis module comprises a data integration sub-module, a data preprocessing sub-module and a data relevance analysis sub-module;

The data integration submodule performs data integration based on real-time acquisition data to generate a data integration result;

The data preprocessing sub-module performs data cleaning and formatting processing based on the data integration result to generate a data preprocessing result;

And the data relevance analysis submodule carries out data relevance analysis by adopting a Bayesian network based on the data preprocessing result to generate a relevance analysis result.

8. The new energy vehicle charging station monitoring system of claim 1, wherein: the user interface module comprises a strategy display sub-module, an early warning display sub-module, a price display sub-module and a scheduling display sub-module;

The strategy display submodule generates a strategy visual interface by adopting a back propagation neural network based on an optimized charging strategy;

the early warning display submodule generates an early warning information display interface based on the safety early warning information and the strategy visual interface by applying a data analysis method;

The price display sub-module generates a dynamic price display interface by utilizing a real-time data processing technology based on the dynamic charging price and the early warning information display interface;

the scheduling display sub-module generates a user-friendly operation interface by adopting a data visualization technology based on a dynamic price display interface, the optimized charging schedule and a correlation analysis result.