A method and system for predicting electric vehicle charging load considering traffic flow

Abstract

The present invention discloses a method and system for predicting the charging load of an electric vehicle considering traffic flow, and relates to the technical field of electric vehicle charging. Firstly, an urban road is divided into a plurality of independent sections, and an environmental coefficient xct of the section is calculated by a deep machine learning algorithm, and combined with real-time traffic flow data, a traffic congestion degree zyd is generated, congested sections during peak hours are identified, and early warning instructions are issued in time, and then the use of charging stations around the marked sections is deeply analyzed, and the load saturation state of the charging stations is effectively identified by comparing the saturation coefficient xbh with the saturation threshold H. When the load is saturated, the system immediately issues an early warning, and a charging load compensation value sbc is calculated by fitting by a prediction module, and a level compensation early warning of the charging load of the charging stations around the marked sections is predicted and issued through comparative matching analysis, thereby improving the correlation between traffic flow and charging load and the accurate prediction of the charging load of electric vehicles.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicle charging, and in particular to a method and system for predicting electric vehicle charging load taking traffic flow into consideration. Background Art

[0002] With the popularity of electric vehicles, the global demand for green travel and sustainable development is increasing. Electric vehicles, with their advantages of low emissions and high efficiency, are gradually becoming an important direction for future transportation development. However, the rapid growth of electric vehicles has also brought about a sharp increase in charging demand, especially in areas with dense urban traffic. How to efficiently manage and predict the charging load of electric vehicles has become an urgent problem to be solved.

[0003] However, the existing charging load prediction methods and systems do not pay attention to the dynamic changes of traffic flow, have relatively large limitations in dealing with the impact of traffic flow, and have difficulties in coping with the volatility and complexity of charging demand. The existing charging load prediction methods and systems rely more on fixed historical data and lack flexible response to real-time traffic flow changes, resulting in a disconnect between the prediction results and the actual charging demand, and there are deficiencies in how to effectively link traffic flow with charging load. Summary of the invention

[0004] In view of the deficiencies in the prior art, the present invention provides a method and system for predicting charging load of electric vehicles taking traffic flow into consideration, which solves the problems in the above-mentioned background technology.

[0005] To achieve the above objectives, the present invention is implemented through the following technical solutions: an electric vehicle charging load prediction system considering traffic flow, including a road section division module, a data acquisition module, an environmental pre-analysis module, a traffic flow analysis module, a charging load analysis module and a prediction module;

[0006] The road segment division module is based on geographic information system technology and road network model, and divides the roads for electric vehicles in urban roads into several independent road segments according to the characteristics of geographical location and road attributes;

[0007] The data acquisition module is used to process the satellite high-definition images of several road sections through image enhancement technology to generate road section environmental status information, and is also used to collect traffic vehicle flow information of several road sections through multiple traffic cameras, and after feature extraction, to generate vehicle flow status information of different road sections in different time periods;

[0008] The environmental pre-analysis module is used to pre-analyze the environmental status information of the road section and obtain the road section environmental coefficient xct through deep machine learning calculation;

[0009] The traffic flow analysis module associates the traffic flow status information of different sections in each time period with the section environment coefficient xct, obtains N groups of traffic congestion degrees zyd by fitting, and marks the sections that are currently judged to be in the peak period after feature extraction and comparative analysis, and issues a peak warning instruction at the same time;

[0010] The charging load analysis module is based on the peak warning instruction, and is used to analyze the usage of the charging stations around the marked road section, fit the saturation coefficient xbh, and compare it with the saturation threshold H to determine whether the current charging station is in a load saturation state. If saturated, a saturation warning instruction is issued;

[0011] The prediction module is based on the saturation warning instruction, used to analyze the charging load of the marked charging station, fit the charging load compensation value SBC, and compare and match the charging load compensation value SBC, thereby predicting the level compensation warning of the charging load of the charging stations around the marked section and issuing it.

[0012] Preferably, the road segment division module includes a feature recognition unit;

[0013] The feature recognition unit is based on geographic information system technology and a road network model, and is used to perform feature recognition on roads where electric vehicles travel in urban roads. Feature recognition includes the spatial position, road type and road length of the road. After feature recognition, a target section is selected, and the road where electric vehicles travel in the target section is divided into several independent sections.

[0014] Preferably, the data collection module includes a road section environment collection unit and a traffic flow collection unit;

[0015] The road section environment acquisition unit is used to obtain road section environment status information after processing the satellite high-definition images of several road sections through image enhancement technology, and the road section environment status information includes the number of forks dls, the number of lanes dcs and the width of the road section dks;

[0016] The traffic flow collection unit is used to identify and screen the license plate colors of traffic vehicles traveling on the road section through traffic cameras, and mark the screened electric vehicles, and then collect status information of the marked electric vehicles in different sections in various time periods to obtain traffic flow status information. The traffic flow status information includes the vehicle distance dp and the vehicle speed cv in different sections in various time periods, and combined with the statistical averaging algorithm, the average vehicle distance dps and the average vehicle speed vcs are obtained respectively.

[0017] Preferably, the environmental pre-analysis module is used to analyze the environmental status information of the road section, and obtain the road section environmental coefficient xct after dimensionless processing. The road section environmental coefficient xct is calculated and obtained by the following formula:

[0018] ;

[0019] In the formula, dls represents the number of forks, dcs represents the number of lanes in the section, and dks represents the width of the section. Among them, u1, u2 and u3 represent the weight values ​​of the number of forks dls, the number of lanes in the section dcs and the width of the section dks respectively, and A represents the first correction constant.

[0020] Preferably, the traffic flow analysis module includes a traffic congestion analysis unit, a feature extraction unit and a peak period determination unit;

[0021] The traffic congestion analysis unit associates the traffic flow information of different sections in each time period with the section environment coefficient xct based on the traffic flow analysis instruction, and obtains N groups of traffic congestion degrees zyd by fitting after dimensionless processing. The calculation method of each group in the N groups of traffic congestion degrees zyd is the same. Take one group of traffic congestion degrees as an example: the traffic congestion degree of the jth section in the i-th time period is , which is obtained by the following formula:

[0022] ;

[0023] In the formula, It is expressed as the average distance between vehicles on the jth road section in the i-th time period, It is expressed as the average speed of vehicles traveling on the jth road section in the i-th time period, It is expressed as the road environment coefficient of the j-th road section, where g1, g2 and g3 are all weight values, i belongs to 1, 2, 3, ..., n, n is the number of time periods, j belongs to 1, 2, 3, ..., m, m is the number of sections, and B is the second correction constant.

[0024] Preferably, the feature extraction unit is used to classify the N groups of traffic congestion degrees zyd obtained by fitting, so as to classify them into traffic congestion degree groups at different time periods in the same road section and traffic congestion degree groups at different road sections in the same time, and to obtain the maximum traffic congestion degree at different time periods in the same road section by extracting features from the traffic congestion degree groups at different time periods in the same road section. and minimum traffic congestion .

[0025] Preferably, the peak time determination unit is based on the maximum traffic congestion degree at different time periods in the same road section. and minimum traffic congestion , obtain the reasonable traffic flow range of traffic congestion in different time periods on the same road section ;

[0026] ; ;

[0027] Where k is a constant, and the specific value is set by the user;

[0028] By making reasonable traffic flow range Compare and analyze the traffic congestion degree zyd of each period in the same road section to determine the peak period in the same road section. The specific analysis content is as follows:

[0029] If the traffic flow range is reasonable When it is less than the traffic congestion degree zyd of each period in the corresponding road section, it means that the current road section is in the traffic peak period and is marked, and a peak warning instruction is issued;

[0030] If the traffic flow range is reasonable When it is greater than or equal to the traffic congestion degree zyd in each time period of the corresponding road section, it means that the traffic flow in the current road section during this period is within a reasonable range, and no additional peak warning instructions will be issued at this time.

[0031] Preferably, the charging load analysis module includes a load state acquisition unit, a saturation analysis unit and a comparison unit;

[0032] The load status collection unit is based on the peak warning instruction, and is used to collect the usage of the charging stations around the marked section to obtain the load status data information of the charging stations around the marked section, wherein the load status data information of the charging stations includes the total number of charging piles lcd and the number of used charging piles ycd;

[0033] The saturation analysis unit is used to analyze the load status data information of the charging station around the marked road section, and obtain the saturation coefficient xbh after dimensionless processing. The saturation coefficient xbh is calculated and obtained by the following formula;

[0034] ;

[0035] In the formula, lcd represents the total number of charging piles, and ycd represents the number of charging piles used;

[0036] The comparison unit is used to compare and analyze the obtained saturation coefficient xbh with the saturation threshold H to determine whether the current charging station is in a load saturation state. If saturated, a saturation warning instruction is issued. The specific analysis content is as follows:

[0037] If the saturation coefficient xbh ≥ the saturation threshold H, it indicates that the charging stations around the current marked road section are in a load saturation state, and a saturation warning instruction is issued;

[0038] If the saturation coefficient xbh is less than the saturation threshold H, it indicates that the charging stations around the current marked road section are in an unsaturated load state, and no additional saturation warning instruction is issued.

[0039] Preferably, the prediction module is based on the saturation warning instruction, and is used to analyze the saturation coefficient xbh and the saturation threshold H of the charging stations around the marked road section, and after dimensionless processing, fit to obtain the charging load compensation value sbc, and the charging load compensation value sbc is calculated and obtained by the following formula;

[0040] ;

[0041] The prediction module is also used to compare and match the charging load compensation value SBC with the level compensation warning determination table stored in the system, thereby predicting and issuing the level compensation warning of the charging load of the charging stations around the marked road section, wherein each charging load compensation value SBC corresponds to a level compensation warning, and the level compensation warning includes a first-level compensation warning, a second-level compensation warning and a third-level compensation warning.

[0042] Preferably, a method for predicting charging load of electric vehicles considering traffic flow comprises the following steps;

[0043] Step 1: Based on geographic information system technology and road network model, and according to the characteristics of geographical location and road attributes, the roads where electric vehicles travel in urban roads are divided into several independent sections; the satellite high-definition images of several sections are processed by image enhancement technology to generate section environmental status information, and are also used to collect traffic flow information of several sections through multiple traffic cameras, and after feature extraction, to generate traffic flow status information of different sections in different time periods;

[0044] Step 2: Pre-analyze the road section environmental status information and perform deep machine learning calculation to obtain the road section environmental coefficient xct; associate the traffic flow status information of different road sections in each time period with the road section environmental coefficient xct, and obtain N groups of traffic congestion degrees zyd by fitting. After feature extraction and comparative analysis, mark the road section that is currently in the peak period, and issue a peak warning instruction at the same time;

[0045] Step 3: Based on the peak warning instruction, the usage of the charging stations around the marked road section is analyzed, the saturation coefficient xbh is obtained by fitting, and it is compared with the saturation threshold H to determine whether the current charging station is in a load saturation state. If saturated, a saturation warning instruction is issued;

[0046] Step 4: Based on the saturation warning instruction, the charging load of the marked charging station is analyzed, the charging load compensation value SBC is fitted, and the charging load compensation value SBC is compared and matched. The charging load compensation value SBC is analyzed, thereby predicting the level compensation warning of the charging load of the charging stations around the marked section and issuing it.

[0047] The present invention provides a method and system for predicting charging load of electric vehicles considering traffic flow, which has the following beneficial effects:

[0048] (1) Through refined analysis of the correlation between traffic flow and charging load, the accurate prediction of electric vehicle charging load is significantly improved. GIS technology and road network model are used to divide urban roads into multiple independent sections. Satellite high-definition images and image enhancement technology are used, combined with real-time traffic flow information obtained by multiple traffic cameras, to comprehensively capture the environmental status and traffic conditions of each section. Through deep machine learning algorithms, the environmental pre-analysis module can accurately calculate the environmental coefficient xct of the section and combine it with real-time traffic flow data to generate traffic congestion degree zyd, quickly identify congested sections during peak hours and issue early warning instructions in a timely manner. Based on this, the charging load analysis module conducts an in-depth analysis of the usage of charging stations around the marked sections. By comparing the saturation coefficient xbh with the saturation threshold H, the load saturation state of the charging station is effectively identified. When the load is saturated, the system immediately issues an early warning, and the prediction module fits and calculates the charging load compensation value sbc. Through comparison and matching analysis, the compensation demand of the charging station is more accurately predicted, and then the level compensation of the charging load is implemented. This can not only effectively cope with the dynamic changes in the charging load demand, but also improve the correlation between traffic flow and charging load and the accurate prediction of electric vehicle charging load.

[0049] (2) By combining the data acquisition module with the environmental pre-analysis module, the data collected by satellite high-definition images and traffic cameras are fully utilized. The environmental status and traffic flow of the road section are deeply analyzed through image enhancement technology and deep machine learning algorithms. The road section environmental smoothness coefficient xct and the traffic congestion degree zyd are obtained by fitting. The system can monitor and predict the traffic conditions of different road sections in real time at different time periods, providing a reliable data basis for subsequent charging load prediction.

[0050] (3) By comparing and analyzing the saturation coefficient xbh of the charging station and the saturation threshold H, the load saturation state of the charging station can be discovered in time, and the load level compensation demand of the charging station can be predicted more accurately based on the charging load compensation value sbc obtained by fitting. The corresponding charging load level compensation warning can be dynamically executed, so that the charging station can operate smoothly under high load conditions. Compared with the traditional method that relies on fixed historical data, it can effectively cope with the volatility and complexity of charging demand. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 This is a block diagram of an electric vehicle charging load prediction system considering traffic flow according to the present invention;

[0052] Figure 2 The present invention is a flow chart of a method for predicting charging load of electric vehicles taking traffic flow into consideration. DETAILED DESCRIPTION

[0053] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0054] Example 1

[0055] See also Figure 1 , the present invention provides an electric vehicle charging load prediction system considering traffic flow, including a road section division module, a data acquisition module, an environmental pre-analysis module, a traffic flow analysis module, a charging load analysis module and a prediction module;

[0056] The road segment division module is based on geographic information system technology and road network model, and divides the roads for electric vehicles in urban roads into several independent road segments according to the characteristics of geographical location and road attributes;

[0057] The data acquisition module is used to process the satellite high-definition images of several road sections through image enhancement technology to generate road section environmental status information, and is also used to collect traffic vehicle flow information of several road sections through multiple traffic cameras, and after feature extraction, to generate vehicle flow status information of different road sections in different time periods;

[0058] The environmental pre-analysis module is used to pre-analyze the environmental status information of the road section and obtain the road section environmental coefficient xct through deep machine learning calculation;

[0059] The traffic flow analysis module associates the traffic flow status information of different sections in each time period with the section environment coefficient xct, obtains N groups of traffic congestion degrees zyd by fitting, and marks the sections that are currently judged to be in the peak period after feature extraction and comparative analysis, and issues a peak warning instruction at the same time;

[0060] The charging load analysis module is based on the peak warning instruction, and is used to analyze the usage of the charging stations around the marked road section, fit the saturation coefficient xbh, and compare it with the saturation threshold H to determine whether the current charging station is in a load saturation state. If saturated, a saturation warning instruction is issued;

[0061] The prediction module is based on the saturation warning instruction, used to analyze the charging load of the marked charging station, fit the charging load compensation value SBC, and compare and match the charging load compensation value SBC, thereby predicting the level compensation warning of the charging load of the charging stations around the marked section and issuing it.

[0062] In this embodiment, through the refined analysis of the correlation between traffic flow and charging load, the prediction accuracy and management efficiency of electric vehicle charging demand are significantly improved. The system adopts GIS technology and road network model to divide urban roads into multiple independent sections. Combined with satellite high-definition images and real-time data collected by traffic cameras, deep learning analysis is performed on the environmental status and traffic flow information of each section. Through the environmental pre-analysis module, traffic flow analysis module and charging load analysis module, the system can dynamically monitor traffic congestion and the use of charging stations, issue early warning instructions in time, and accurately evaluate the charging load demand through the prediction module, and take corresponding compensation measures. The system effectively solves the shortcomings of traditional methods that are difficult to flexibly respond to changes in traffic flow, thereby avoiding overloading of charging stations during peak hours, ensuring the charging needs of electric vehicle users, and promoting the sustainable development of urban transportation.

[0063] Example 2

[0064] Please refer to Figure 1 ,Specifically: the road segment division module includes a feature recognition unit;

[0065] The feature recognition unit is based on geographic information system technology and a road network model, and is used to perform feature recognition on roads where electric vehicles travel in urban roads. Feature recognition includes the spatial position, road type and road length of the road. After feature recognition, a target section is selected, and the road where electric vehicles travel in the target section is divided into several independent sections.

[0066] It should be noted that geographic information system technology (GIS) is an existing technology used to capture, store, analyze and display information related to geographic location. GIS technology can combine geographic spatial data with attribute data, and help users understand and process complex geographic data through spatial analysis and visualization tools. Existing GIS technology has been widely used in urban planning, traffic management and environmental monitoring to provide support for various decision-making;

[0067] The road network model is a computational model used to describe and analyze the structure and function of a road system. The model is usually based on graph theory principles and represents the road system as a graph structure consisting of "nodes" (intersections, intersections) and "edges" (road segments connecting nodes). The road network model can include various attributes of the road to perform traffic flow analysis, path planning, and network optimization. Existing road network models are widely used in traffic engineering, intelligent navigation systems, and urban planning.

[0068] Specifically, the data collection module includes a road section environment collection unit and a traffic flow collection unit;

[0069] The road section environment acquisition unit is used to obtain road section environment status information after processing the satellite high-definition images of several road sections through image enhancement technology, and the road section environment status information includes the number of forks dls, the number of lanes dcs and the width of the road section dks;

[0070] The traffic flow collection unit is used to identify and screen the license plate colors of traffic vehicles traveling on the road section through a traffic camera, mark the screened electric vehicles, and then collect the status information of the marked electric vehicles in different sections in each time period to obtain traffic flow status information, the traffic flow status information includes the vehicle distance dp and the vehicle speed cv in different sections in each time period, and combines with the statistical mean algorithm to obtain the average vehicle distance dps and the average vehicle speed vcs;

[0071] It should be noted that traffic cameras identify license plate colors based on deep learning image recognition technology. First, a convolutional neural network is used to process the captured vehicle image, locate and identify the license plate area, and then the image of the license plate area is converted to the HSV color space. The license plate color is determined by analyzing the hue component. When a green hue is detected (Hue value between 60° and 180°), green-licensed vehicles are automatically identified.

[0072] In this embodiment, by introducing a feature recognition unit and a multi-dimensional data acquisition module, the accuracy and real-time performance of the electric vehicle charging load prediction are significantly improved. First, the feature recognition unit can accurately identify the spatial position, road type, and road length of urban roads based on geographic information system technology and road network models, thereby scientifically dividing independent road sections. Secondly, the data acquisition module collects and analyzes the environmental status and traffic flow status information of each road section at multiple levels and angles through the road section environment acquisition unit and the traffic flow acquisition unit. The road section environment acquisition unit uses image enhancement technology to process satellite high-definition images, comprehensively captures key environmental parameters including the number of forks, the number of lanes, and the road section width, and provides basic data support for subsequent analysis. At the same time, the traffic flow acquisition unit recognizes the color of the vehicle license plate through the traffic camera, accurately selects electric vehicles, and collects and analyzes the vehicle distance and driving speed in real time in each time period. The average vehicle distance and the average speed of the driving vehicle obtained by statistical methods can more truly reflect the traffic flow characteristics of the road section, and provide accurate basic data for further analysis and prediction of the system.

[0073] Example 3

[0074] Please refer to Figure 1 Specifically: the environmental pre-analysis module is used to analyze the environmental status information of the road section, and obtain the road section environmental coefficient xct after dimensionless processing. The road section environmental coefficient xct is calculated and obtained by the following formula;

[0075] ;

[0076] In the formula, dls represents the number of forks, dcs represents the number of lanes in the section, and dks represents the width of the section. Among them, u1, u2 and u3 represent the weight values ​​of the number of forks dls, the number of lanes in the section dcs and the width of the section dks respectively, and A represents the first correction constant.

[0077] Specifically, the traffic flow analysis module includes a traffic congestion analysis unit, a feature extraction unit and a peak period determination unit;

[0078] The traffic congestion analysis unit associates the traffic flow information of different sections in each time period with the section environment coefficient xct based on the traffic flow analysis instruction, and obtains N groups of traffic congestion degrees zyd by fitting after dimensionless processing. The calculation method of each group in the N groups of traffic congestion degrees zyd is the same. Take one group of traffic congestion degrees as an example: the traffic congestion degree of the jth section in the i-th time period is , which is obtained by the following formula:

[0079] ;

[0080] In the formula, It is expressed as the average distance between vehicles on the jth road section in the i-th time period, It is expressed as the average speed of vehicles traveling on the jth road section in the i-th time period, It is expressed as the road environment coefficient of the j-th road section, where g1, g2 and g3 are all weight values, i belongs to 1, 2, 3, ..., n, n is the number of time periods, j belongs to 1, 2, 3, ..., m, m is the number of sections, and B is the second correction constant.

[0081] In this embodiment, by introducing refined calculation of road section environment coefficient and traffic congestion analysis, the prediction accuracy and rationality of electric vehicle charging load are greatly improved. First, the environmental pre-analysis module uses key parameters such as the number of forks, the number of lanes and the road section width, combined with weight values ​​and correction constants, to construct a scientific road section environment coefficient calculation formula. This ensures comprehensive consideration of environmental characteristics of different road sections, allowing the system to more accurately evaluate the traffic conditions of each road section. Secondly, the traffic flow analysis module combines the average vehicle distance, the average speed of traveling vehicles and the road section environment coefficient through dimensionless processing and fitting calculation of multiple groups of traffic congestion degrees zyd, thereby realizing a dynamic analysis of traffic congestion conditions in each time period and each road section.

[0082] Example 4

[0083] Please refer to Figure 1 Specifically: the feature extraction unit is used to classify the N groups of traffic congestion degrees zyd obtained by fitting, so as to classify them into traffic congestion degree groups at different time periods in the same road section and traffic congestion degree groups at different road sections in the same time period, and to obtain the maximum traffic congestion degree at different time periods in the same road section by extracting features from the traffic congestion degree groups at different time periods in the same road section. and minimum traffic congestion .

[0084] Specifically, the peak time determination unit is based on the maximum traffic congestion degree at different time periods in the same road section. and minimum traffic congestion , obtain the reasonable traffic flow range of traffic congestion in different time periods on the same road section ;

[0085] ; ;

[0086] Where k is a constant, and the specific value is set by the user;

[0087] By making reasonable traffic flow range Compare and analyze the traffic congestion degree zyd of each period in the same road section to determine the peak period in the same road section. The specific analysis content is as follows:

[0088] If the traffic flow range is reasonable When it is less than the traffic congestion degree zyd of each period in the corresponding road section, it means that the current road section is in the traffic peak period and is marked, and a peak warning instruction is issued;

[0089] If the traffic flow range is reasonable When it is greater than or equal to the traffic congestion degree zyd in each time period of the corresponding road section, it means that the traffic flow in the current road section during this period is within a reasonable range, and no additional peak warning instructions will be issued at this time.

[0090] In this embodiment, the accuracy and efficiency of peak period determination are further improved through precise traffic congestion classification and feature extraction. The feature extraction unit classifies the N groups of traffic congestion degrees zyd obtained by fitting in detail, and divides them into traffic congestion degree groups of different time periods in the same road section and different road sections in the same time. The maximum and minimum traffic congestion degrees of different time periods in the same road section are extracted in particular. This classification and extraction method enables the system to deeply analyze the changes in traffic conditions in each time period in the same road section. Subsequently, the peak period determination unit uses the extracted maximum and minimum traffic congestion degrees, combined with the set reasonable traffic flow range, to compare and analyze the traffic congestion degrees of each time period, effectively screen out the traffic peak period and issue an early warning. Through this rigorous analysis and comparison mechanism, the peak period of road traffic can be discovered and marked in time.

[0091] Example 5

[0092] Please refer to Figure 1 , specifically: the charging load analysis module includes a load state acquisition unit, a saturation analysis unit and a comparison unit;

[0093] The load status collection unit is based on the peak warning instruction, and is used to collect the usage of the charging stations around the marked section to obtain the load status data information of the charging stations around the marked section, wherein the load status data information of the charging stations includes the total number of charging piles lcd and the number of used charging piles ycd;

[0094] The saturation analysis unit is used to analyze the load status data information of the charging station around the marked road section, and obtain the saturation coefficient xbh after dimensionless processing. The saturation coefficient xbh is calculated and obtained by the following formula;

[0095] ;

[0096] In the formula, lcd represents the total number of charging piles, and ycd represents the number of charging piles used;

[0097] The comparison unit is used to compare and analyze the obtained saturation coefficient xbh with the saturation threshold H to determine whether the current charging station is in a load saturation state. If saturated, a saturation warning instruction is issued. The specific analysis content is as follows:

[0098] If the saturation coefficient xbh ≥ the saturation threshold H, it indicates that the charging stations around the current marked road section are in a load saturation state, and a saturation warning instruction is issued;

[0099] If the saturation coefficient xbh is less than the saturation threshold H, it indicates that the charging stations around the current marked road section are in an unsaturated load state, and no additional saturation warning instruction is issued.

[0100] In this embodiment, through the triple analysis mechanism of the charging load analysis module, the detection accuracy and response speed of the load saturation state of the charging station are effectively improved, and the close connection between traffic flow and charging load is enhanced. First, the load state acquisition unit first accurately collects the load state data information of the charging stations around the marked section based on the peak warning instruction, ensuring the real-time and pertinence of the data. Next, the saturation analysis unit uses these data to quantitatively evaluate the load condition of the charging station by calculating the saturation coefficient xbh. Finally, the comparison unit compares and analyzes the calculated saturation coefficient xbh with the preset saturation threshold H to quickly determine whether the load state of the current charging station has reached saturation. When the saturation coefficient xbh reaches or exceeds the saturation threshold H, the system immediately issues a saturation warning instruction. When the saturation coefficient xbh is lower than the saturation threshold H, the system will not issue additional warnings, reducing unnecessary interference.

[0101] Example 6

[0102] Please refer to Figure 1 Specifically: the prediction module is based on the saturation warning instruction, and is used to analyze the saturation coefficient xbh and the saturation threshold H of the charging stations around the marked road section, and after dimensionless processing, fit to obtain the charging load compensation value sbc, and the charging load compensation value sbc is calculated and obtained by the following formula;

[0103] ;

[0104] The prediction module is also used to compare and match the charging load compensation value SBC with the level compensation warning determination table stored in the system, thereby predicting and issuing the level compensation warning of the charging load of the charging stations around the marked road section, wherein each charging load compensation value SBC corresponds to a level compensation warning, and the level compensation warning includes a first-level compensation warning, a second-level compensation warning and a third-level compensation warning.

[0105] In this embodiment, the charging load compensation analysis function of the prediction module further improves the ability to accurately predict and warn the load conditions of the charging station. First, based on the saturation warning instruction, the prediction module deeply analyzes the relationship between the saturation coefficient xbh and the saturation threshold H of the charging stations around the marked section, and calculates the charging load compensation value sbc through the formula. This compensation value represents the additional load compensation required to maintain the normal operation of the charging station under the current load state. Next, the system compares and matches the calculated charging load compensation value sbc with the pre-stored level compensation warning determination table. Based on the matching results, the system can accurately predict the load level of the charging stations around the marked section and issue corresponding warnings. Each charging load compensation value sbc corresponds to a specific warning level, covering level one, level two and level three compensation warnings. Through this mechanism, the system can not only provide warnings when the charging station is about to reach or has reached saturation, but also issue compensation warnings of corresponding levels according to different load conditions, so as to achieve dynamic balance and effective diversion of charging demand and reduce potential risks caused by overload.

[0106] Example 7

[0107] Please refer to Figure 1 and Figure 2 ,Specifically: A method for predicting charging load of electric vehicles considering ,traffic flow includes the following steps;

[0108] Step 1: Based on geographic information system technology and road network model, and according to the characteristics of geographical location and road attributes, the roads where electric vehicles travel in urban roads are divided into several independent sections; the satellite high-definition images of several sections are processed by image enhancement technology to generate section environmental status information, and are also used to collect traffic flow information of several sections through multiple traffic cameras, and after feature extraction, to generate traffic flow status information of different sections in different time periods;

[0109] Step 2: Pre-analyze the road section environmental status information and perform deep machine learning calculation to obtain the road section environmental coefficient xct; associate the traffic flow status information of different road sections in each time period with the road section environmental coefficient xct, and obtain N groups of traffic congestion degrees zyd by fitting. After feature extraction and comparative analysis, mark the road section that is currently in the peak period, and issue a peak warning instruction at the same time;

[0110] Step 3: Based on the peak warning instruction, the usage of the charging stations around the marked road section is analyzed, the saturation coefficient xbh is obtained by fitting, and it is compared with the saturation threshold H to determine whether the current charging station is in a load saturation state. If saturated, a saturation warning instruction is issued;

[0111] Step 4: Based on the saturation warning instruction, the charging load of the marked charging station is analyzed, the charging load compensation value SBC is fitted, and the charging load compensation value SBC is compared and matched. The charging load compensation value SBC is analyzed, thereby predicting the level compensation warning of the charging load of the charging stations around the marked section and issuing it.

[0112] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An electric vehicle charging load prediction system considering traffic flow, characterized in that: It includes road segmentation module, data collection module, environmental pre-analysis module, traffic flow analysis module, charging load analysis module and prediction module; The road segment division module is based on geographic information system technology and road network model, and divides the roads for electric vehicles in urban roads into several independent road segments according to the characteristics of geographical location and road attributes; The data acquisition module is used to process the satellite high-definition images of several road sections through image enhancement technology to generate road section environmental status information, and is also used to collect traffic vehicle flow information of several road sections through multiple traffic cameras, and after feature extraction, to generate vehicle flow status information of different road sections in different time periods; The environmental pre-analysis module is used to pre-analyze the environmental status information of the road section, and obtain the road section environmental coefficient xct through deep machine learning calculation. The road section environmental coefficient xct is calculated and obtained by the following formula; ； Wherein, dls represents the number of forks, dcs represents the number of lanes in the road section, dks represents the width of the road section, u1, u2 and u3 represent the weight values ​​of the number of forks dls, the number of lanes in the road section dcs and the width of the road section dks respectively, and A represents the first correction constant; The traffic flow analysis module associates the traffic flow status information of different sections in each time period with the section environment coefficient xct, obtains N groups of traffic congestion degrees zyd by fitting, and marks the sections that are currently judged to be in the peak period after feature extraction and comparative analysis, and issues a peak warning instruction at the same time; The traffic flow analysis module includes a traffic congestion analysis unit, a feature extraction unit and a peak period determination unit; The traffic congestion analysis unit associates the traffic flow information of different sections in each time period with the section environment coefficient xct based on the traffic flow analysis instruction, and obtains N groups of traffic congestion degrees zyd by fitting after dimensionless processing. The calculation method of each group in the N groups of traffic congestion degrees zyd is the same. Take one group of traffic congestion degrees as an example: the traffic congestion degree of the jth section in the i-th time period is , which is obtained by the following formula: ； In the formula, It is expressed as the average distance between vehicles on the jth road section in the i-th time period, It is expressed as the average speed of vehicles traveling on the jth road section in the i-th time period, is represented as the road environment coefficient of the jth road section, wherein g1, g2 and g3 are all weight values, i belongs to 1, 2, 3, ..., n, n is the number of time periods, j belongs to 1, 2, 3, ..., m, m is the number of road sections, and B is the second correction constant; The charging load analysis module is based on the peak warning instruction, and is used to analyze the usage of the charging stations around the marked road section, fit the saturation coefficient xbh, and compare it with the saturation threshold H to determine whether the current charging station is in a load saturation state. If saturated, a saturation warning instruction is issued; The prediction module is based on the saturation warning instruction, used to analyze the charging load of the marked charging station, fit the charging load compensation value SBC, and compare and match the charging load compensation value SBC, thereby predicting the level compensation warning of the charging load of the charging stations around the marked section and issuing it.

2. The electric vehicle charging load prediction system considering traffic flow according to claim 1 is characterized in that: The road segment division module includes a feature recognition unit; The feature recognition unit is based on geographic information system technology and a road network model, and is used to perform feature recognition on roads where electric vehicles travel in urban roads. Feature recognition includes the spatial position, road type and road length of the road. After feature recognition, a target section is selected, and the road where electric vehicles travel in the target section is divided into several independent sections.

3. The electric vehicle charging load prediction system considering traffic flow according to claim 1 is characterized in that: The data collection module includes a road section environment collection unit and a traffic flow collection unit; The road section environment acquisition unit is used to obtain road section environment status information after processing the satellite high-definition images of several road sections through image enhancement technology, and the road section environment status information includes the number of forks dls, the number of lanes dcs and the width of the road section dks; The traffic flow collection unit is used to identify and screen the license plate colors of traffic vehicles traveling on the road section through traffic cameras, and mark the screened electric vehicles, and then collect status information of the marked electric vehicles in different sections in various time periods to obtain traffic flow status information. The traffic flow status information includes the vehicle distance dp and the vehicle speed cv in different sections in various time periods, and combined with the statistical averaging algorithm, the average vehicle distance dps and the average vehicle speed vcs are obtained respectively.

4. The electric vehicle charging load prediction system considering traffic flow according to claim 1 is characterized in that: The feature extraction unit is used to classify the N groups of traffic congestion degrees zyd obtained by fitting, so as to classify them into traffic congestion degree groups at different time periods in the same road section and traffic congestion degree groups at different road sections in the same time period, and to obtain the maximum traffic congestion degree at different time periods in the same road section by extracting features from the traffic congestion degree groups at different time periods in the same road section. and minimum traffic congestion .

5. The electric vehicle charging load prediction system considering traffic flow according to claim 4 is characterized in that: The peak time determination unit is based on the maximum traffic congestion degree in different time periods in the same road section. and minimum traffic congestion , obtain the reasonable traffic flow range of traffic congestion in different time periods on the same road section ; ； ； Where k is a constant, and the specific value is set by the user; By making reasonable traffic flow range Compare and analyze the traffic congestion degree zyd of each period in the same road section to determine the peak period in the same road section. The specific analysis content is as follows: If the traffic flow range is reasonable When it is less than the traffic congestion degree zyd of each period in the corresponding road section, it means that the current road section is in the traffic peak period and is marked, and a peak warning instruction is issued; If the traffic flow range is reasonable When it is greater than or equal to the traffic congestion degree zyd in each time period of the corresponding road section, it means that the traffic flow in the current road section during this period is within a reasonable range, and no additional peak warning instructions will be issued at this time.

6. The electric vehicle charging load prediction system considering traffic flow according to claim 1 is characterized in that: The charging load analysis module includes a load state acquisition unit, a saturation analysis unit and a comparison unit; The load status collection unit is based on the peak warning instruction, and is used to collect the usage of the charging stations around the marked section to obtain the load status data information of the charging stations around the marked section, wherein the load status data information of the charging stations includes the total number of charging piles lcd and the number of used charging piles ycd; The saturation analysis unit is used to analyze the load status data information of the charging station around the marked road section, and obtain the saturation coefficient xbh after dimensionless processing. The saturation coefficient xbh is calculated and obtained by the following formula; ； In the formula, lcd represents the total number of charging piles, and ycd represents the number of charging piles used; The comparison unit is used to compare and analyze the obtained saturation coefficient xbh with the saturation threshold H to determine whether the current charging station is in a load saturation state. If saturated, a saturation warning instruction is issued. The specific analysis content is as follows: If the saturation coefficient xbh ≥ the saturation threshold H, it indicates that the charging stations around the current marked section are in a load saturation state, and a saturation warning instruction is issued; If the saturation coefficient xbh is less than the saturation threshold H, it indicates that the charging stations around the current marked road section are in an unsaturated load state, and no additional saturation warning instruction is issued.

7. The electric vehicle charging load prediction system considering traffic flow according to claim 1 is characterized in that: The prediction module is based on the saturation warning instruction, and is used to analyze the saturation coefficient xbh and the saturation threshold H of the charging stations around the marked road section, and after dimensionless processing, fit to obtain the charging load compensation value sbc, and the charging load compensation value sbc is calculated and obtained by the following formula; ； The prediction module is also used to compare and match the charging load compensation value SBC with the level compensation warning determination table stored in the system, thereby predicting and issuing the level compensation warning of the charging load of the charging stations around the marked road section, wherein each charging load compensation value SBC corresponds to a level compensation warning, and the level compensation warning includes a first-level compensation warning, a second-level compensation warning and a third-level compensation warning.

8. A method for predicting charging load of electric vehicles considering traffic flow, used to implement a system for predicting charging load of electric vehicles considering traffic flow as described in any one of claims 1 to 7, characterized in that: The steps include: Step 1: Based on geographic information system technology and road network model, and according to the characteristics of geographical location and road attributes, the roads where electric vehicles travel in urban roads are divided into several independent sections; the satellite high-definition images of several sections are processed by image enhancement technology to generate section environmental status information, and are also used to collect traffic flow information of several sections through multiple traffic cameras, and after feature extraction, to generate traffic flow status information of different sections in different time periods; Step 2: Pre-analyze the road section environmental status information and perform deep machine learning calculation to obtain the road section environmental coefficient xct; associate the traffic flow status information of different road sections in each time period with the road section environmental coefficient xct, and obtain N groups of traffic congestion degrees zyd by fitting. After feature extraction and comparative analysis, mark the road section that is currently in the peak period, and issue a peak warning instruction at the same time; Step 3: Based on the peak warning instruction, the usage of the charging stations around the marked road section is analyzed, the saturation coefficient xbh is obtained by fitting, and it is compared with the saturation threshold H to determine whether the current charging station is in a load saturation state. If saturated, a saturation warning instruction is issued; Step 4: Based on the saturation warning instruction, the charging load of the marked charging station is analyzed, the charging load compensation value SBC is fitted, and the charging load compensation value SBC is compared and matched. The charging load compensation value SBC is analyzed, thereby predicting the level compensation warning of the charging load of the charging stations around the marked section and issuing it.

Images