Intelligent new energy vehicle energy dynamic scheduling method and system based on multi-source data fusion

Abstract

An embodiment of the present invention provides a method and system for dynamic energy scheduling of intelligent new energy vehicles based on multi-source data fusion. The method includes obtaining initial energy storage information of historical vehicles, analyzing the consistency of sensors based on the initial energy storage information to obtain data correlation between sensors; classifying sensors based on the data correlation to obtain a target sensor set; obtaining target energy storage information, current position information, and future path information of the target vehicle; determining the state of the road on which the target vehicle is located based on the current position information and future path information; obtaining relevant energy storage information from the initial energy storage information based on the target sensor set and the target energy storage information; determining an initial movement length based on the relevant energy storage information, and adjusting the initial movement length based on the road state to obtain a target movement length; determining an initial charging position based on the target movement length and future path information; and performing dynamic energy scheduling based on the initial charging position to obtain a target scheduling result.

Description

Technical Field

[0001] The present invention relates to the field of data processing technology, and in particular to a method and system for dynamic energy scheduling of intelligent new energy vehicles based on multi-source data fusion. Background Art

[0002] With the rapid development of the new energy vehicle market and the growing demand for charging from various smart devices, the accuracy of charging location scheduling has become increasingly critical, directly impacting the charging experience and convenience of new energy vehicle users. Accurate charging location scheduling allows users to quickly and efficiently find suitable charging locations, saving time and effort and ensuring the normal use of their devices. However, existing charging location scheduling often relies heavily on traditional algorithms, most of which are based on relatively simple logic and fixed rules. For example, in the case of new energy vehicle charging location scheduling, traditional algorithms may simply determine and assign charging locations based on the vehicle's current remaining battery level and the straight-line distance from the charging station. Because existing technologies fail to fully account for these complex practical factors, they struggle to provide the most accurate and reasonable charging location scheduling solutions. This causes significant inconvenience for users, reducing user satisfaction and increasing user anxiety. Summary of the Invention

[0003] The main purpose of the embodiments of the present invention is to provide a method and system for dynamic energy scheduling of intelligent new energy vehicles based on multi-source data fusion, aiming to solve the problem that it is difficult to provide accurate and reasonable scheduling solutions when performing charging location scheduling in related technologies, thereby causing many inconveniences to users and reducing user satisfaction.

[0004] In a first aspect, an embodiment of the present invention provides a method for dynamic energy scheduling of intelligent new energy vehicles based on multi-source data fusion, comprising:

[0005] Obtaining initial energy storage information corresponding to historical vehicles, and analyzing the data consistency of different target sensors based on the initial energy storage information to obtain the corresponding data correlation between any two target sensors;

[0006] Classifying the target sensors according to the data association degree to obtain multiple target sensor sets;

[0007] Obtaining target energy storage information corresponding to a target vehicle, as well as current position information and future path information corresponding to the target vehicle;

[0008] Determining a road state corresponding to the road where the target vehicle is located according to the current position information and the future path information;

[0009] Obtaining relevant energy storage information corresponding to the target vehicle from the initial energy storage information according to the target sensor set and the target energy storage information;

[0010] Determining an initial moving length corresponding to the target vehicle according to the relevant energy storage information, and adjusting the initial moving length according to the road conditions to obtain a target moving length;

[0011] Determining an initial charging position corresponding to the target vehicle according to the target movement length and the future path information;

[0012] Energy dynamic scheduling is performed according to the initial charging position to obtain a target scheduling result.

[0013] In a second aspect, an embodiment of the present invention provides an intelligent new energy vehicle energy dynamic scheduling system based on multi-source data fusion, comprising:

[0014] A data analysis module is used to obtain initial energy storage information corresponding to historical vehicles, and analyze the data consistency of different target sensors based on the initial energy storage information to obtain the corresponding data correlation between any two target sensors;

[0015] a relationship determination module, configured to classify the target sensors according to the data association degree to obtain a plurality of target sensor sets;

[0016] A data acquisition module is used to obtain target energy storage information corresponding to a target vehicle, as well as current position information and future path information corresponding to the target vehicle;

[0017] a state determination module, configured to determine a road state corresponding to the road where the target vehicle is located based on the current position information and the future path information;

[0018] a data determination module, configured to obtain relevant energy storage information corresponding to the target vehicle from the initial energy storage information according to the target sensor set and the target energy storage information;

[0019] a length calculation module, configured to determine an initial moving length corresponding to the target vehicle according to the relevant energy storage information, and adjust the initial moving length according to the road conditions to obtain a target moving length;

[0020] a position determination module, configured to determine an initial charging position corresponding to the target vehicle according to the target movement length and the future path information;

[0021] The scheduling determination module is used to perform dynamic energy scheduling according to the initial charging position to obtain a target scheduling result.

[0022] An embodiment of the present invention provides a method and system for dynamic energy scheduling of intelligent new energy vehicles based on multi-source data fusion. The method includes: obtaining initial energy storage information corresponding to historical vehicles, and analyzing the data consistency of different target sensors based on the initial energy storage information to obtain the corresponding data correlation between any two target sensors. By analyzing the data consistency of the different target sensors, the corresponding data correlation between any two target sensors is obtained, thereby providing a more reliable basis for subsequent decision-making, and then classifying the target sensors based on the data correlation to obtain multiple target sensor sets; obtaining target energy storage information corresponding to the target vehicle and current position information and future path information corresponding to the target vehicle; determining the road state corresponding to the road where the target vehicle is located based on the current position information and the future path information; obtaining relevant energy storage information corresponding to the target vehicle from the initial energy storage information based on the target sensor set and the target energy storage information; determining the initial movement length corresponding to the target vehicle based on the relevant energy storage information, thereby more accurately estimating the initial movement length of the vehicle during future driving; and then adjusting the initial movement length based on the road state to more accurately calculate the target movement length of the target vehicle. Then, determining the initial charging position corresponding to the target vehicle based on the target movement length and the future path information, thereby planning a more reasonable charging plan for the vehicle. This prevents vehicles from breaking down mid-route due to low battery levels, reduces unnecessary charging times, and improves vehicle efficiency. Finally, dynamic energy scheduling based on the initial charging location yields the target scheduling result, enabling rational energy allocation and optimized utilization. This provides accurate charging recommendations and a reliable energy supply, alleviating user anxiety. This also addresses the difficulty in providing accurate and reasonable scheduling solutions for charging location scheduling in related technologies, which can cause significant inconvenience to users and reduce user satisfaction. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following is a brief introduction to the drawings required for use in the description of the embodiments. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0024] Figure 1 A schematic diagram of a flow chart of a method for dynamic energy scheduling of intelligent new energy vehicles based on multi-source data fusion provided by an embodiment of the present invention;

[0025] Figure 2 A schematic diagram of the module structure of an intelligent new energy vehicle energy dynamic scheduling system based on multi-source data fusion provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0026] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of them. All other embodiments obtained by ordinary technicians in this field based on the embodiments of the present invention without making any creative efforts shall fall within the scope of protection of the present invention.

[0027] The flowcharts shown in the accompanying drawings are for illustrative purposes only and do not necessarily include all contents and operations / steps, nor must they be executed in the order described. For example, some operations / steps may be decomposed, combined, or partially merged, so the actual execution order may vary depending on the actual situation.

[0028] It should be understood that the terms used in this specification are only for the purpose of describing particular embodiments and are not intended to limit the present invention. As used in the specification and appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms unless the context clearly indicates otherwise.

[0029] Embodiments of the present invention provide a method and system for dynamic energy scheduling for intelligent new energy vehicles based on multi-source data fusion. This method can be applied to terminal devices, such as tablet computers, laptop computers, desktop computers, personal digital assistants, and wearable devices. The terminal device can also be a server or a server cluster.

[0030] The following embodiments of the present invention are described in detail with reference to the accompanying drawings. In the absence of conflict, the following embodiments and features in the embodiments may be combined with each other.

[0031] Please refer to Figure 1 , Figure 1 A flow chart of a method for dynamic energy scheduling of intelligent new energy vehicles based on multi-source data fusion provided in an embodiment of the present invention.

[0032] like Figure 1 As shown, the method for dynamic energy scheduling of intelligent new energy vehicles based on multi-source data fusion includes steps S101 to S108.

[0033] Step S101: obtaining initial energy storage information corresponding to historical vehicles, and analyzing the data consistency of different target sensors based on the initial energy storage information to obtain the corresponding data correlation between any two target sensors.

[0034] For example, to achieve efficient energy management and coordinated charging, multiple vehicles simultaneously broadcast their precise energy storage information using advanced wireless communication technology. Car A accurately transmits this information to a nearby charging station. The charging station, serving as the core hub for this information exchange, is equipped with high-performance distributed data processing and storage equipment. Upon receiving the energy storage information broadcast from Car A and other vehicles, the charging station rapidly processes this massive amount of information using distributed algorithms. This distributed processing model distributes data processing tasks across multiple computing nodes, significantly improving processing speed and efficiency while enhancing system reliability and fault tolerance. After processing, the information is stored in an organized manner in its own database.

[0035] For example, initial energy storage information corresponding to historical vehicles is obtained from a database. This initial energy storage information includes, but is not limited to, energy storage information detected by different target sensors. For example, different types of target sensors, such as battery level sensors, voltage sensors, and current sensors, can be appropriately installed on the vehicle to ensure comprehensive detection of the vehicle's energy storage status. Different types of sensors monitor different parameters related to energy storage, thereby obtaining energy storage information from multiple dimensions.

[0036] Exemplarily, statistical analysis methods such as correlation analysis and covariance analysis are used to evaluate the data correlation between any two target sensors, thereby quantifying the linear relationship or degree of coordinated variation between the sensor data.

[0037] In some embodiments, the target sensor includes at least a first sensor and a second sensor, and the initial energy storage information includes at least first energy storage information corresponding to the first sensor and second energy storage information corresponding to the second sensor, and the data consistency of different target sensors is analyzed according to the initial energy storage information to obtain the corresponding data correlation between any two target sensors, including: obtaining a first critical value corresponding to the first energy storage information, and determining a first acquisition time corresponding to the first sensor; determining a first segmentation length corresponding to data segmentation of the first energy storage information according to the first acquisition time combined with the first critical value, obtaining a second critical value corresponding to the second energy storage information, and determining a second acquisition time corresponding to the second sensor; determining a second segmentation length corresponding to data segmentation of the second energy storage information according to the second acquisition time combined with the second critical value; performing data segmentation on the first energy storage information according to the first segmentation length to obtain a first segmentation length corresponding to the first energy storage information a first connection relationship corresponding to the segmentation unit and the first segmentation unit; smoothing the first sub-data in each first segmentation unit according to the first segmentation unit and the first connection relationship to obtain a first processing result; performing data segmentation on the second energy storage information according to the second segmentation length to obtain the second segmentation unit corresponding to the second energy storage information and the second connection relationship corresponding to the second segmentation unit; smoothing the second sub-data in each second segmentation unit according to the second segmentation unit and the second connection relationship to obtain a second processing result; obtaining a first center corresponding to the first processing result and a second center corresponding to the second processing result, and determining a target correlation factor according to the first processing result and the second processing result; determining the data consistency characterization value corresponding to the first sensor and the second sensor according to the first center and the second center combined with the target correlation factor; determining the data correlation degree corresponding to the first sensor and the second sensor according to the data consistency characterization value.

[0038] Exemplarily, different target sensors include at least a first sensor and a second sensor, and first energy storage information corresponding to the first sensor and second energy storage information corresponding to the second sensor are obtained from a database.

[0039] For example, statistical methods are used to comprehensively analyze and sort the data of the first energy storage information. By calculating various statistical indicators of the data, such as the mean, median, maximum, and minimum values, and taking into account the data distribution characteristics and actual application requirements, a first critical value corresponding to the first energy storage information is determined. In some cases, the maximum value in the first energy storage information can be used as the first critical value. Similarly, the same method is used to perform statistical analysis on the second energy storage information to determine its corresponding second critical value, and the maximum value in the second energy storage information can also be used as the second critical value.

[0040] For example, the operating characteristics of a sensor and the frequency of data changes are important factors influencing data collection and analysis. Therefore, the first acquisition duration corresponding to the first sensor needs to be determined based on the operating characteristics and frequency of data changes. The first sensor may have different sampling frequencies and response times, and the frequency of data changes may also vary depending on the monitored object. Similarly, the second acquisition duration corresponding to the second sensor is determined based on the operating characteristics and frequency of data changes.

[0041] Exemplarily, a first minimum value and a first maximum value corresponding to the first energy storage information are determined, and a first length is determined based on the difference between the first minimum value and the first maximum value. Furthermore, first area information corresponding to the first energy storage information is determined based on the first acquisition duration and the first length. The first acquisition duration is then squared to obtain a first square value, and a first segmentation factor is determined. The first segmentation factor is multiplied by the first area information, divided by a first critical value, and then squared to obtain first data. The first square value is then divided by the first data to obtain a first segmentation length. Similarly, a second minimum value and a second maximum value corresponding to the second energy storage information are determined, and a second length is determined based on the difference between the second minimum value and the second maximum value. Furthermore, second area information corresponding to the second energy storage information is determined based on the second acquisition duration and the second length. The second acquisition duration is then squared to obtain a second square value, and a second segmentation factor is determined. The second segmentation factor is multiplied by the second area information, divided by a second critical value, and then squared to obtain second data. The second square value is then divided by the second data to obtain a second segmentation length.

[0042] Exemplarily, after obtaining the first segmentation length and the second segmentation length, the first energy storage information and the second energy storage information are data segmented. The first energy storage information is segmented using the determined first segmentation length to divide it into multiple first segmentation units. Each first segmentation unit contains the first energy storage information within a certain time period, which are relatively independent and interrelated data segments. During the segmentation process, it is necessary to carefully record the sequence and association between each first segmentation unit, that is, the first connection relationship. This connection relationship reflects the temporal continuity and logical association of the data, which is of great significance for subsequent data analysis and processing. Similarly, the second energy storage information is segmented using the second segmentation length to obtain the second segmentation unit and the second connection relationship.

[0043] Exemplarily, according to the first connection relationship corresponding to the first segmentation unit and the first segmentation unit, the first related unit and the first segmentation unit under the first connection relationship are locally connected to obtain first fusion data, and then the first parameter corresponding to the local Gaussian weight function is determined according to the first fusion data, and then the first sub-data in the first segmentation unit is smoothed according to the local Gaussian weight function under the first parameter to obtain the first processing result corresponding to the first sub-data.

[0044] Exemplarily, according to the second connection relationship corresponding to the second segmentation unit and the second segmentation unit, the second related unit and the second segmentation unit under the second connection relationship are locally connected to obtain second fusion data, and then the second parameter corresponding to the local Gaussian weight function is determined according to the second fusion data, and then the second sub-data in the second segmentation unit is smoothed according to the local Gaussian weight function under the second parameter to obtain a second processing result corresponding to the second sub-data.

[0045] Exemplarily, a first processing result corresponding to each first sub-data in each first segmentation unit is obtained, and then a center calculation is performed on all the first processing results in the first segmentation unit to obtain a corresponding first center. Furthermore, a second processing result corresponding to each second sub-data in the second segmentation unit is obtained, and then a center calculation is performed on all the second processing results in the second segmentation unit to obtain a corresponding second center.

[0046] For example, to evaluate the degree of correlation between the first processing result and the second processing result, a correlation analysis algorithm is used to measure the relationship between the first processing result and the second processing result by calculating an indicator such as a correlation coefficient, thereby determining a target correlation factor between the first processing result and the second processing result. This target correlation factor reflects the degree of correlation between the first processing result and the second processing result; a larger value indicates a closer correlation between the two. Data fusion is then performed based on the first and second centers combined with the target correlation factor to obtain a data consistency representation value corresponding to the first sensor and the second sensor. The data consistency representation value comprehensively considers the central characteristics and degree of correlation of the data, and can more comprehensively reflect the consistency between the two sensor data.

[0047] Exemplarily, the data correlation between the first sensor and the second sensor is determined based on the data consistency characterization value combined with a pre-set threshold range. The pre-set threshold range is determined based on actual application scenarios and experience, and it divides the data correlation into different levels. For example, if the data consistency characterization value is higher than a certain threshold, it is considered that the data correlation between the two sensors is high, which means that the data collected by the two sensors have strong consistency and correlation, and they may be monitoring the same physical quantity or have a close causal relationship; conversely, if the data consistency characterization value is lower than the threshold, it is considered that the data correlation between the two sensors is low.

[0048] In some embodiments, determining the data consistency characterization values corresponding to the first sensor and the second sensor based on the first center and the second center in combination with the target association factor includes: obtaining first neighborhood data corresponding to each third sub-data from the first processing result, and determining the first filtering feature corresponding to the third sub-data based on the first neighborhood data; obtaining second neighborhood data corresponding to each fourth sub-data from the second processing result, and determining the second filtering feature corresponding to the fourth sub-data based on the second neighborhood data; determining the target association factor corresponding to the first processing result and the second processing result based on the first filtering feature and the second filtering feature in combination with the third sub-data and the fourth sub-data; calculating the minimum and maximum values between the third sub-data and the fourth sub-data, and determining the relative distance between the third sub-data and the fourth sub-data based on the minimum and the maximum values; calculating the center distance between the first center and the second center, and determining the unit consistency characterization value corresponding to the first processing result and the second processing result under the segmentation unit based on the center distance and the relative distance in combination with the target association factor; and fusing multiple unit consistency characterization values to obtain the data consistency characterization value corresponding to the first sensor and the second sensor.

[0049] For example, each third sub-data point is obtained from the first processing result, and the data within a certain range around the third sub-data point is selected as the first neighborhood data. For example, in time series data, several data points before and after the third sub-data point can be selected as the first neighborhood data. Similarly, for each fourth sub-data point in the second processing result, the corresponding second neighborhood data is obtained.

[0050] For example, the first neighborhood data is filtered using an algorithm such as mean filtering or median filtering to obtain a value representing the neighborhood feature as the first filtering feature corresponding to the third sub-data. For the second neighborhood data, the same filtering method is used to determine the second filtering feature corresponding to the fourth sub-data.

[0051] Exemplarily, the first filtering feature corresponding to each third sub-data and the second filtering feature corresponding to each fourth sub-data are obtained to calculate the distance information between the third sub-data and the fourth sub-data, and thus determine the target correlation factor corresponding to the first processing result and the second processing result based on the first filtering feature and the second filtering feature combined with the distance information.

[0052] For example, for each pair of third and fourth sub-data, the minimum and maximum values between them are found, and then the minimum value is divided by the maximum value to obtain the relative distance. The center distance between the first center and the second center is then calculated based on the Euclidean distance. The center distance and the relative distance are then multiplied and divided by the target correlation factor to obtain the corresponding unit consistency representation value between the first processing result and the second processing result under the segmentation unit. Finally, the unit consistency representation values under multiple segmentation units are summed and divided by the number of units to obtain the data consistency representation value corresponding to the first sensor and the second sensor.

[0053] Exemplarily, the center distance and the relative distance are multiplied and then divided by the target correlation factor calculated previously to obtain the corresponding unit consistency characterization value between the first processing result and the second processing result under the segmentation unit. The unit consistency characterization value can comprehensively consider the relative distance between the sub-data, the center distance and the degree of correlation between the processed data to which they belong, and more comprehensively reflect the consistency of the two processed data under the segmentation unit. Therefore, after completing the calculation of the unit consistency characterization value under each segmentation unit, the unit consistency characterization values under multiple segmentation units are summed. Then, the sum result is divided by the number of units, and the average value obtained is the data consistency characterization value corresponding to the first sensor and the second sensor. The data consistency characterization value can reflect the degree of consistency of the data collected by the first sensor and the second sensor as a whole, and provides an important reference basis for evaluating the correlation of sensor data. In addition, a comprehensive and in-depth analysis of the data collected by different sensors can be performed to better understand the correlation and consistency between the data, and provide strong support for subsequent data applications and decision-making.

[0054] Specifically, by considering the first filtering feature and the second filtering feature, local features of the data can be extracted and analyzed to reduce the impact of noise and outliers. The unit consistency characterization value is calculated by combining multiple factors such as the target correlation factor, relative distance and center distance, which integrates the degree of correlation, numerical difference and center position relationship of the data, making the evaluation of data consistency more comprehensive and accurate. Calculating the unit consistency characterization value under each segmentation unit separately can capture the changes in data in different local areas. Different segmentation units may reflect data in different time periods or different feature intervals. The fusion of multiple unit consistency characterization values can better reflect the overall consistency changes of the data and have a more keen perception of small changes in the data. The data consistency characterization value can be used as an important indicator for evaluating the performance of the first sensor and the second sensor.

[0055] Step S102: Classify the target sensors according to the data association degree to obtain multiple target sensor sets.

[0056] Exemplarily, the preset correlation degree is a predetermined threshold value, which is set according to specific application scenarios and requirements. When the data correlation degree between any two sensors is greater than or equal to the preset correlation degree, it means that the data collected by the two sensors have a high similarity or correlation, and they may have a certain degree of synergy in function, and can jointly provide similar information about the vehicle's energy storage. Therefore, the two sensors are determined to be an initial sensing set. If the data correlation degree between any two sensors is less than the preset correlation degree, it means that the data collected by the two sensors are quite different, and they may reflect different physical phenomena or processes, and lack functional synergy. Therefore, the two sensors cannot be determined as an initial sensing set.

[0057] For example, the initial sensor set is the result of a preliminary grouping of sensors. The sensors within each set may share similarities or correlations in some way. After obtaining multiple initial sensor sets, to further optimize the sensor grouping, these initial sensor sets need to be subjected to intersection processing. The goal of the intersection processing is to identify the common components between the initial sensor sets and merge sets with shared sensors to obtain a more comprehensive and representative target sensor set.

[0058] For example, initial sensor set 1: {sensor a, sensor b}; initial sensor set 2: {sensor a, sensor c}; initial sensor set 3: {sensor b, sensor c}; and initial sensor set 4: {sensor a, sensor d}. Initial sensor sets 1, 2, and 3 share common sensors. Initial sensor sets 1 and 2 both contain sensor a, initial sensor sets 1 and 3 both contain sensor b, and initial sensor sets 2 and 3 both contain sensor c. By merging these three initial sensor sets, we obtain a new set {sensor a, sensor b, sensor c}. This new set is one of the target sensor sets obtained after the intersection process. Initial sensor set 4, however, shares only sensor a with the previously merged target sensor set {sensor a, sensor b, sensor c}. However, sensor a alone cannot form a meaningful set, and set 4 also contains the unique sensor d. Therefore, set 4 may be considered a separate target sensor set. Through this intersection processing method, multiple initial sensor sets can be integrated to obtain multiple target sensor sets. These target sensor sets can more accurately reflect the internal connections between sensors.

[0059] Step S103: Obtain target energy storage information corresponding to the target vehicle, as well as current position information and future path information corresponding to the target vehicle.

[0060] For example, the target vehicle is a currently operating new energy vehicle. Leveraging advanced wireless communication technologies such as 5G and vehicle-to-everything (V2X), these vehicles can communicate in real time with surrounding infrastructure, other vehicles, and cloud platforms. The target energy storage information includes key data such as the target vehicle's current battery remaining charge and range.

[0061] For example, while broadcasting the target energy storage information, the target vehicle also acquires and broadcasts its current location and future path information. Current location information is acquired in real time using technologies such as the high-precision Global Positioning System (GPS) and inertial navigation systems, accurately displaying the vehicle's specific location on a map. Future path information, mapped by the vehicle's navigation system, includes the vehicle's projected route and the locations of charging stations along the way.

[0062] Step S104: Determine the road state corresponding to the road where the target vehicle is located according to the current position information and the future path information.

[0063] For example, based on the current position information of the target vehicle, geographic information system technology is used to accurately match the vehicle position to a specific first road section, and the future path information of the target vehicle is also mapped to the map using geographic information system technology to determine the second road sections that the vehicle is about to pass. After matching the current position of the target vehicle to the first road section, it is necessary to obtain relevant traffic information of the section. The relevant traffic information mainly includes traffic flow, road capacity and vehicle speed, and then determine the first congestion index corresponding to the first road section based on the first neural network prediction model combined with the relevant traffic information. Similarly, for each second road section that the target vehicle is about to pass, it is also necessary to obtain its corresponding traffic flow, road capacity and vehicle speed and other information. When calculating the second congestion index, since the vehicle has not yet arrived at these sections, it is also necessary to predict the traffic flow changes of these sections in the future period based on historical traffic data and real-time traffic conditions, so as to calculate the second congestion index based on the relevant traffic information corresponding to the second road section and the traffic flow changes combined with the second neural network prediction model. After obtaining the first congestion index and the second congestion index, the road state corresponding to the road where the target vehicle is located is jointly determined, and the road state is any one of unobstructed, lightly congested, moderately congested and severely congested.

[0064] For example, the first congestion index and the second congestion index are weightedly fused to obtain the road congestion index corresponding to the road where the target vehicle is located, and then the road congestion index is compared with a pre-set congestion index threshold, so as to determine the road state corresponding to the first road section where the target vehicle is currently located based on the comparison result.

[0065] In some embodiments, the determining of the road state corresponding to the road where the target vehicle is located based on the current position information and the future path information includes: determining the target road based on the current position information and the future path information, and obtaining historical traffic parameters corresponding to the target road from a database; determining index weight information corresponding to each traffic index based on the historical traffic parameters, and determining sample weight information corresponding to the historical traffic parameters; performing data clustering on the historical traffic parameters based on the index weight information and the sample weight information to obtain a target clustering result; determining the cluster center corresponding to each sub-cluster in the target clustering result, and determining the traffic state distribution corresponding to the sub-cluster; obtaining the target road. The method comprises the following steps: determining the current road parameters corresponding to the road, and determining the associated cluster corresponding to the current road parameters according to the current road parameters and the cluster center; determining the first state distribution corresponding to the current road parameters according to the associated cluster combined with the traffic state distribution; merging the associated cluster and the current road parameters to obtain a target cluster, performing parameter analysis on the target cluster to obtain the target distribution position corresponding to the target cluster; determining the second state distribution corresponding to the current road parameters according to the target distribution position; fusing the first state distribution and the second state distribution to determine the target state distribution corresponding to the current road parameters; and determining the road state corresponding to the road where the target vehicle is located according to the target state distribution.

[0066] Exemplarily, a geographic information system is used to determine the corresponding target road according to the current position information and future path information of the target vehicle. That is, the target road includes the road where the target vehicle is currently located and the road that the target vehicle is about to pass.

[0067] For example, after determining the target road, the query function of the database is used to filter out historical traffic parameters corresponding to the target road based on the road identification information (such as road number, name, etc.). The historical traffic parameters may include traffic volume, average speed, congestion duration, etc.

[0068] Exemplarily, relevant traffic data corresponding to each traffic indicator in the historical traffic parameters is obtained, thereby obtaining entropy information corresponding to the relevant traffic data, and then determining the indicator weight information corresponding to each traffic indicator in the historical traffic parameters based on the entropy information of each traffic indicator.

[0069] Exemplarily, the degree of membership between each sub-data in the historical traffic parameters and different traffic categories is obtained, and the first distance between each sub-data and the central data corresponding to the traffic category is calculated, and the second distance is obtained by weighting according to the indicator weight information corresponding to each traffic indicator and the first distance, so as to obtain the target value by weighted summation according to the degree of membership and the second distance according to different traffic categories, and then the sample weight information corresponding to the historical traffic parameters is determined according to the target value combined with the exponential function.

[0070] Exemplarily, historical traffic parameters are clustered according to sample weight information and indicator weight information in combination with a clustering algorithm to obtain target clustering results. That is, in the clustering process, the indicator weight information of different traffic indicators and the sample weight information of each sample are considered, so that through iterative calculation, the historical traffic parameter data are divided into different subclass clusters, and finally the target clustering results are obtained.

[0071] Exemplarily, for each sub-cluster in the target clustering result, its corresponding cluster center is calculated, and the traffic state type corresponding to each data in the sub-cluster is obtained from the database, so as to obtain the traffic state distribution corresponding to the sub-cluster by performing statistics based on all traffic state types of the sub-cluster.

[0072] Exemplarily, the current road parameters corresponding to the target road are obtained from the database, and then the distance information between the current road parameters and each cluster center is calculated, so that the sub-cluster corresponding to the cluster center when the distance information is minimum is determined as the associated cluster corresponding to the current road parameters.

[0073] Exemplarily, distribution information corresponding to the associated cluster is obtained from the traffic state distribution, and then the distribution information is determined as the first state distribution corresponding to the current road parameter.

[0074] For example, the current road parameters are added as a new sample to the associated cluster to obtain a target cluster. The degree of dispersion of each traffic metric in the target cluster (such as traffic volume, average speed, and congestion duration) is then calculated. A histogram is then plotted based on the degree of dispersion to visually display the distribution of each traffic metric. The histogram divides the data into several intervals, each corresponding to a column. The height of the column represents the frequency of the data within that interval. By observing the shape of the histogram, it is possible to determine whether the data follows a near-normal distribution, a skewed distribution, or some other special distribution. The distribution characteristics of multiple traffic metrics are then integrated to determine the target distribution position of the target cluster under different traffic conditions. For example, the target distribution position may be in one of the following categories: unimpeded, lightly congested, moderately congested, or severely congested. Based on the target distribution position and in combination with pre-defined traffic state classification criteria, the second state distribution corresponding to the current road parameters is determined.

[0075] For example, based on the fused eigenvalues or principal component scores, the target cluster is judged to be more likely to be in which traffic state. For example, if the combined eigenvalue falls within a pre-set threshold for a lightly congested state, the target cluster is considered to have a corresponding distribution location in the lightly congested state. The number or proportion of samples in the target cluster that meet different traffic states can be counted to more accurately describe its distribution under different traffic states. Based on the target distribution location and in combination with pre-set traffic state classification criteria, the second state distribution corresponding to the current road parameters can be determined.

[0076] For example, the weights are determined based on the reliability of the historical data on which the first state distribution is based and the timeliness of the current real-time data considered by the second state distribution, and then the first state distribution and the second state distribution are weightedly summed using a weighted fusion method to obtain the target state distribution corresponding to the current road parameters.

[0077] For example, based on the target state distribution, the traffic state with the highest probability of occurrence is found and determined as the road state corresponding to the target vehicle's road. For example, if the probability of the "congested" state is the highest in the target state distribution, then the target vehicle's road is determined to be in a congested state.

[0078] In some embodiments, the parameter analysis of the target cluster to obtain the target distribution position corresponding to the target cluster includes: spatially and temporally dividing the sub-road parameters corresponding to the target cluster to obtain spatial information and temporal information corresponding to each sub-road parameter; determining the corresponding association parameters between any two sub-road parameters based on the spatial information and the temporal information; normalizing the sub-road parameters to obtain the normalized parameters corresponding to the sub-road parameters; determining the adjacent spatial standard values corresponding to the sub-road parameters based on the association parameters and the normalized parameters; determining the target position quadrant corresponding to the sub-road parameter based on the normalized parameters and the adjacent spatial standard values; and counting the target position quadrant to obtain the target distribution position corresponding to the target cluster.

[0079] Exemplarily, the target cluster is a data set formed by merging the associated cluster and the current road parameters, and then the spatial information and time information corresponding to each sub-road parameter in the target cluster are obtained, so that the association parameter between any two sub-road parameters is calculated based on the spatial information and time information. For example, if the spatial information between any two sub-road parameters is adjacent and the time information is also adjacent, the association parameter is determined to be 1, otherwise the association parameter is 0.

[0080] Exemplarily, the average road parameter corresponding to the target cluster is obtained, and then each sub-road parameter is normalized according to the average road parameter to obtain a normalized parameter, and then the adjacent space standard value corresponding to the sub-road parameter is determined according to the association parameter and the normalized parameter. For example, the adjacent sub-road parameters with a high degree of association with each sub-road parameter are screened out according to the association parameter, and the sum is then divided by the number of data corresponding to the adjacent sub-road parameters to obtain the adjacent space standard value corresponding to the sub-road parameter.

[0081] For example, a two-dimensional coordinate system is established with the normalized parameter as one coordinate axis and the adjacent space standard value as the other coordinate axis. This allows the coordinate plane to be divided into four quadrants based on actual traffic conditions and data analysis requirements. For example, the coordinate plane can be divided into four regions, each defined as a different quadrant, using the mean of the normalized parameter and the adjacent space standard value as the dividing line. The normalized parameter and adjacent space standard value of each sub-road parameter are then used as coordinate points to locate their corresponding positions in the established coordinate system, thereby determining the target quadrant corresponding to that sub-road parameter.

[0082] For example, the number or ratio of all sub-road parameters in the target cluster in each target location quadrant is counted. Based on the quadrant statistics, the distribution of the target cluster in different quadrants is analyzed to determine the target distribution location corresponding to the target cluster. For example, the quadrant with the largest number of sub-road parameters represents the main distribution location of the target cluster. Different quadrants represent different traffic conditions.

[0083] Step S105: Obtain relevant energy storage information corresponding to the target vehicle from the initial energy storage information according to the target sensor set and the target energy storage information.

[0084] Exemplarily, the first sub-energy storage information corresponding to different sensors in the target energy storage information is obtained, and then the distance between this first sub-energy storage information and the second sub-energy storage information for the sensor in the initial energy storage information is calculated to obtain the first similar information corresponding to the target vehicle for that sensor. Similarly, the first similar information corresponding to each sensor in the target energy storage information is obtained. Then, based on the target sensor set, the second similar information corresponding to the relevant sensor in the target sensor set is filtered from all the first similar information. The filtered second similar information is then matched with a database, and the information identifier corresponding to each second similar information is searched in the database. The information identifiers of the second similar information corresponding to each sensor in the target sensor set are then counted, and the number of occurrences of each information identifier is counted. When the information identifiers of the second similar information for most sensors in the target sensor set are the same, the data corresponding to the information identifier is determined as the relevant energy storage information corresponding to the target vehicle.

[0085] For example, by determining the relevant energy storage information corresponding to the target vehicle based on the target sensor set, errors caused by large data gaps when directly using all sensor data can be avoided. During the calculation process, the data from the sensors in the target sensor set is used to eliminate interference from the data of a specific sensor in the target sensor set, thereby making the calculation result closer to the actual energy storage situation. This method can also prevent the data collected by the sensor from deviating significantly from the actual value when a sensor in the target sensor set has a measurement anomaly. Including this abnormal data in the calculation of the relevant energy storage information would have a significant impact on the calculation result. Using the target sensor set can effectively reduce this impact. The sensors in the target sensor set are screened to form a set of sensors with strong correlations, which can, to a certain extent, avoid the occurrence of measurement anomalies. Even if individual sensors experience measurement anomalies, because the target sensor set is a combination of sensors, the data from other normal sensors can still provide relatively accurate information, thereby reducing the error caused by measurement anomalies in the calculation of relevant energy storage information.

[0086] In some embodiments, the step of obtaining the relevant energy storage information corresponding to the target vehicle from the initial energy storage information according to the target sensor set and the target energy storage information includes: obtaining a third sensor and a fourth sensor associated with each subset according to the target sensor set; obtaining first information corresponding to the third sensor and second information corresponding to the fourth sensor from the target energy storage information; obtaining third information corresponding to the third sensor and fourth information corresponding to the fourth sensor from the initial energy storage information; performing data clustering according to the first information and the third information to obtain a first clustering result, and obtaining a first cluster associated with the first information according to the first clustering result; and obtaining a first cluster associated with the first information according to the second information and the third information. The fourth information is clustered to obtain a second clustering result, and a second cluster associated with the second information is obtained based on the second clustering result; the first cluster and the second cluster are unioned to obtain the associated energy storage information corresponding to the third sensor and the fourth sensor; an initial data identifier corresponding to the associated energy storage information is obtained from a database, and the initial data identifier corresponding to each subset in the target sensor set is determined; the initial data identifier corresponding to each subset is intersectioned using the target sensor set to obtain a target data identifier corresponding to the target sensor set; and the relevant energy storage information corresponding to the target vehicle is obtained from the database based on the target data identifier.

[0087] Exemplarily, the third and fourth sensors associated with each subset are obtained based on the target sensor set. Then, based on the sensor identifiers, the first information corresponding to the third sensor and the second information corresponding to the fourth sensor are extracted from the target energy storage information. For example, if the third sensor is a temperature sensor, the first information is the temperature data collected by the temperature sensor at the current moment; similarly, the second information is the voltage data collected by the fourth sensor (e.g., a voltage sensor).

[0088] Exemplarily, third information corresponding to the third sensor and fourth information corresponding to the fourth sensor are obtained from the initial energy storage information also based on the sensor identifier.

[0089] For example, the first and third information are combined to perform a data clustering operation, resulting in a first clustering result. Then, based on the first clustering result, the cluster to which the first information belongs is determined, i.e., the first cluster. This means that the first information has a high degree of similarity with other data in the cluster. Similarly, using the same method as the first clustering process, the second and fourth information are clustered to obtain a second clustering result, and a second cluster associated with the second information is determined.

[0090] Exemplarily, the first cluster and the second cluster are subjected to a union process, where the union process refers to finding data elements belonging to the first cluster or the second cluster, thereby forming the associated energy storage information corresponding to the third sensor and the fourth sensor.

[0091] For example, the obtained associated energy storage information is matched with a database, and the initial data identifier corresponding to the associated energy storage information is searched in the database. The database pre-stores the correspondence between various energy storage information and initial data identifiers, which can be obtained through a simple query operation.

[0092] Exemplarily, the above steps are repeated for each subset in the target sensor set to determine the initial data identifier corresponding to each subset, and then the target sensor set is used to perform intersection processing on the initial data identifier corresponding to each subset to obtain the target data identifier corresponding to the target sensor set, that is, the target data identifier exists in the associated energy storage information corresponding to each subset, and then according to the target data identifier, the corresponding information is searched from the database, and this information is the relevant energy storage information corresponding to the target vehicle.

[0093] For example, the above method can avoid the situation where the data collected by the sensor deviates seriously from the actual value when a sensor in the subset has a measurement abnormality, thereby causing the calculated related energy storage information to have a large difference from the target energy storage information, thereby reducing the accuracy of subsequent calculations. This method considers strongly correlated sensors in the subset and determines the data as associated energy storage information as long as there is a type of sensor that meets similar conditions. This can avoid the problem of abnormal calculation of related energy storage information when a certain sensor has an abnormality to a certain extent. Even if an individual sensor has a measurement abnormality, since the target sensor set is a combination of a group of sensors, the data of other normal sensors can still provide relatively accurate information, thereby reducing the error in the calculation of related energy storage information caused by measurement abnormality.

[0094] In some embodiments, clustering the data according to the first information and the third information to obtain a first clustering result includes: clustering the first information and the third information using a clustering algorithm to obtain a third clustering result; performing statistical calculation on each third cluster in the third clustering result to obtain a first mean and a first variance, and performing statistical calculation on the first remaining clusters other than the third cluster in the third clustering result to obtain a second mean and a second variance; determining a first difference value between the third cluster and the first remaining cluster based on the first mean, the first variance, the second mean, and the second variance; performing distance calculation on the third cluster to obtain first distance information corresponding to the third cluster; determining a first weight corresponding to the third cluster based on the first difference value and the first distance information; and re-clustering the third clustering result based on the first weight to obtain the first clustering result; wherein the first difference value is obtained according to the following formula;

[0095]

[0096] in, represents the first difference value between the i-th third cluster and the first remaining cluster corresponding to the i-th third cluster, represents the first variance corresponding to the i-th third cluster, represents the second variance corresponding to the first remaining cluster corresponding to the i-th third cluster, e represents an exponential function, represents the first mean corresponding to the i-th third cluster, represents the second mean corresponding to the first remaining cluster corresponding to the i-th third cluster, Represents a constant.

[0097] Exemplarily, clustering analysis is performed on the data obtained by integrating the first information and the third information using a clustering algorithm such as K-means clustering, hierarchical clustering, DBSCAN, etc., so as to obtain a third clustering result.

[0098] For example, for each third cluster in the third clustering result, the values of all data points within the cluster are collected. A statistical method is used to calculate a first mean for these data points, which is the sum of all data point values divided by the number of data points. Simultaneously, a first variance is calculated. The variance measures the dispersion of the data points relative to the mean and is calculated by averaging the squared differences between each data point and the mean.

[0099] Exemplarily, for the first remaining clusters except the third cluster in the third clustering result, the values of all data points therein are also collected, and the second mean and second variance are calculated in the same way as the mean and variance of the third cluster are calculated.

[0100] Exemplarily, the first difference value between the third cluster and the first remaining cluster is determined using the first mean, the first variance, the second mean, and the second variance according to the following formula:

[0101]

[0102] in, represents the first difference value between the i-th third cluster and the first remaining cluster corresponding to the i-th third cluster, represents the first variance corresponding to the i-th third cluster, represents the second variance of the first remaining cluster corresponding to the i-th third cluster, e represents the exponential function, represents the first mean corresponding to the i-th third cluster, represents the second mean of the first remaining cluster corresponding to the i-th third cluster, Represents a constant.

[0103] For example, the distance between each data point in the third cluster and the cluster center is calculated, and then the average of the distances from all data points to the cluster center is calculated as the first distance information corresponding to the third cluster. The first weight corresponding to the third cluster is determined based on the ratio between the first difference value and the first distance information. A larger first difference value indicates a more distinct cluster from other clusters, potentially indicating a higher importance. The first distance information reflects the compactness of the data within the cluster; a closer distance may indicate more stable data within the cluster.

[0104] For example, the third clustering result is re-clustered according to the first weight. During the re-clustering process, the first weight is incorporated into the calculation of the clustering algorithm, so that the third cluster with a higher weight is more attractive to the data points during the clustering process, or is given a higher priority when dividing the data points. For example, in k-means clustering, the centroid update method can be adjusted according to the first weight so that the centroid is more inclined to move toward the cluster with a higher weight. Through re-clustering, the final first clustering result is obtained, which more accurately reflects the distribution characteristics of the data and the differences between clusters.

[0105] In some embodiments, clustering the data according to the second information and the fourth information to obtain the second clustering result includes: clustering the second information and the fourth information using the clustering algorithm to obtain a fourth clustering result; performing statistical calculation on each fourth cluster in the fourth clustering result to obtain a third mean and a third variance, and performing statistical calculation on the second remaining clusters other than the fourth cluster in the fourth clustering result to obtain a fourth mean and a fourth variance; determining a second difference value between the fourth cluster and the second remaining clusters according to the third mean, the third variance, the fourth mean, and the fourth variance; performing distance calculation on the fourth cluster to obtain second distance information corresponding to the fourth cluster; determining a second weight corresponding to the fourth cluster according to the second difference value and the second distance information; and re-clustering the fourth clustering result according to the second weight to obtain the second clustering result; wherein the second difference value is obtained according to the following formula;

[0106]

[0107] in, represents the second difference value between the j-th fourth cluster and the second remaining cluster corresponding to the j-th fourth cluster, represents the third difference corresponding to the jth fourth cluster, represents the fourth variance corresponding to the second remaining cluster corresponding to the jth fourth cluster, e represents the exponential function, represents the third mean corresponding to the j-th fourth cluster, represents the fourth mean corresponding to the second remaining cluster corresponding to the jth fourth cluster, Represents a constant.

[0108] Exemplarily, clustering analysis is performed on the data obtained by integrating the second information and the fourth information using a clustering algorithm such as K-means clustering, hierarchical clustering, DBSCAN, etc., so as to obtain a fourth clustering result.

[0109] For example, for each fourth cluster in the fourth clustering result, the values of all data points within that cluster are collected. A statistical method is used to calculate the third mean of these data points, which is the sum of all data point values divided by the number of data points. Simultaneously, the third variance is calculated. The variance measures the dispersion of the data points relative to the mean and is calculated by averaging the squares of the differences between each data point and the mean.

[0110] For example, for the second remaining clusters except the fourth cluster in the fourth clustering result, the values of all data points therein are also collected, and the fourth mean and fourth variance are calculated in the same way as the mean and variance of the fourth cluster are calculated.

[0111] Exemplarily, the second difference value between the fourth cluster and the second remaining cluster is determined using the third mean, the third variance, the fourth mean, and the fourth variance according to the following formula:

[0112]

[0113] in, represents the second difference value between the j-th fourth cluster and the second remaining cluster corresponding to the j-th fourth cluster, represents the third difference corresponding to the jth fourth cluster, represents the fourth variance corresponding to the second remaining cluster corresponding to the jth fourth cluster, e represents the exponential function, represents the third mean corresponding to the jth fourth cluster, represents the fourth mean of the second remaining cluster corresponding to the jth fourth cluster, Represents a constant.

[0114] For example, the distance between each data point in the fourth cluster and the center of the cluster is calculated, and then the average of the distances from all data points to the cluster center is calculated as the second distance information corresponding to the fourth cluster. The second weight corresponding to the fourth cluster is then determined based on the ratio between the second difference value and the second distance information. A larger second difference value indicates a more distinct cluster from other clusters, potentially indicating a higher importance. The second distance information reflects the compactness of the data within the cluster; a closer distance may indicate more stable data within the cluster.

[0115] For example, the fourth clustering result is re-clustered according to the second weight. During the re-clustering process, the second weight is incorporated into the calculation of the clustering algorithm, so that the fourth cluster with a higher weight is more attractive to the data points during the clustering process, or is given a higher priority when dividing the data points. For example, in k-means clustering, the centroid update method can be adjusted according to the second weight so that the centroid is more inclined to move toward the cluster with a higher weight. By re-clustering, the final second clustering result is obtained, which more accurately reflects the distribution characteristics of the data and the differences between clusters.

[0116] Step S106: determining an initial moving length corresponding to the target vehicle according to the relevant energy storage information, and adjusting the initial moving length according to the road conditions to obtain a target moving length.

[0117] Exemplarily, the relevant moving length corresponding to each relevant energy storage information is obtained from the database according to the relevant energy storage information, that is, the travel length corresponding to the relevant energy storage information under historical records is obtained from the database, and then the initial moving length corresponding to the target vehicle is obtained by statistics based on the relevant moving lengths.

[0118] For example, the loss rate corresponding to the initial moving length is determined according to the road state and the mapping table, and then the target moving length is obtained by multiplying the loss rate and the initial moving length.

[0119] Step S107: Determine an initial charging position corresponding to the target vehicle according to the target moving length and the future path information.

[0120] Exemplarily, the charging positions involved in the future path information are determined, and then the charging positions involved in the future path information are filtered according to the target movement length to obtain the initial charging position corresponding to the target vehicle. The initial charging position is the charging position involved in the future path information of the target vehicle under the target movement length.

[0121] Step S108: Dynamically dispatch energy according to the initial charging position to obtain a target dispatch result.

[0122] Exemplarily, an initial charging position corresponding to each target vehicle is obtained, and then an optimization algorithm (such as a genetic algorithm, a simulated annealing algorithm, etc.) is used to dynamically schedule energy for the target vehicle based on the initial charging position, thereby obtaining a target scheduling result.

[0123] In some embodiments, after obtaining the target scheduling result, the method further includes: determining a target charging pile, and determining a target associated vehicle corresponding to the target charging pile based on the target scheduling result; obtaining vehicle energy storage information corresponding to the target associated vehicle, and determining a target charging sequence and target charging power corresponding to the target associated vehicle based on the target charging pile and the vehicle energy storage information.

[0124] Exemplarily, a target charging pile is determined, and then a target-associated vehicle corresponding to the target charging pile is obtained from the target scheduling result. That is, the target-associated vehicle is a vehicle that needs to be charged at the target charging pile.

[0125] For example, by establishing a communication connection with the on-board equipment of the target-associated vehicle, the vehicle's real-time energy storage information, such as current battery level, battery health, and remaining range, is obtained. This information can reflect the vehicle's energy needs and charging urgency, thereby assessing the charging urgency of each vehicle based on the real-time energy storage information of the target-associated vehicle. For example, vehicles with low battery levels and short remaining ranges should be charged first, and then the charging order should be reasonably arranged with reference to the estimated arrival time of the vehicles in the target scheduling results. If multiple vehicles arrive at the same time, they are sorted according to the charging urgency; if the vehicles arrive in a certain order, the vehicles that arrive first are given priority. During the charging process, the charging status of the vehicle and the status of the following vehicles are monitored in real time, and the charging order is dynamically adjusted according to the actual situation.

[0126] For example, the maximum charging power a target vehicle can accept is determined based on its energy storage information and battery characteristics. For example, some vehicles can accept a higher charging power when their batteries are low, but require a lower charging power to protect the battery when they are nearly fully charged. By comprehensively considering the charging station capacity and vehicle needs, an appropriate target charging power is assigned to each target vehicle. If multiple vehicles are charging simultaneously at the target charging station, the charging power needs to be properly allocated to improve overall charging efficiency.

[0127] See also Figure 2 , Figure 2An intelligent new energy vehicle energy dynamic scheduling system 200 based on multi-source data fusion is provided in an embodiment of the present application. The intelligent new energy vehicle energy dynamic scheduling system 200 based on multi-source data fusion includes a data analysis module 201, a relationship determination module 202, a data acquisition module 203, a state determination module 204, a data determination module 205, a length calculation module 206, a position determination module 207, and a scheduling determination module 208. The data analysis module 201 is used to obtain the initial energy storage information corresponding to the historical vehicle, and analyze the data consistency of different target sensors according to the initial energy storage information to obtain the corresponding data correlation between any two target sensors; the relationship determination module 202 is used to classify the target sensors according to the data correlation to obtain multiple target sensor sets; the data acquisition module 203 is used to obtain the target vehicle corresponding to the target sensor. The target energy storage information, as well as the current position information and future path information corresponding to the target vehicle; a state determination module 204 is used to determine the road state corresponding to the road where the target vehicle is located according to the current position information and the future path information; a data determination module 205 is used to obtain the relevant energy storage information corresponding to the target vehicle from the initial energy storage information according to the target sensor set and the target energy storage information; a length calculation module 206 is used to determine the initial moving length corresponding to the target vehicle according to the relevant energy storage information, and adjust the initial moving length according to the road state to obtain the target moving length; a position determination module 207 is used to determine the initial charging position corresponding to the target vehicle according to the target moving length and the future path information; a scheduling determination module 208 is used to perform dynamic energy scheduling according to the initial charging position to obtain a target scheduling result.

[0128] In some embodiments, the intelligent new energy vehicle energy dynamic scheduling system 200 based on multi-source data fusion can be applied to terminal devices.

[0129] It should be noted that technical personnel in the relevant field can clearly understand that for the convenience and conciseness of description, the specific working process of the intelligent new energy vehicle energy dynamic scheduling system 200 based on multi-source data fusion described above can refer to the corresponding process in the aforementioned embodiment of the intelligent new energy vehicle energy dynamic scheduling method based on multi-source data fusion, and will not be repeated here.

[0130] An embodiment of the present invention also provides a storage medium for computer-readable storage, wherein the storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement any step of the method for dynamic energy scheduling of intelligent new energy vehicles based on multi-source data fusion as provided in the description of the embodiment of the present invention.

[0131] The storage medium may be an internal storage unit of the terminal device in the aforementioned embodiment, such as a hard disk or memory of the terminal device. The storage medium may also be an external storage device of the terminal device, such as a plug-in hard disk equipped on the terminal device, a smart memory card (SMC), a secure digital (SD) card, a flash card, etc.

[0132] Those skilled in the art will appreciate that all or some of the steps, systems, and functional modules / units in the methods, systems, and devices disclosed above may be implemented as software, firmware, hardware, or any combination thereof. In hardware embodiments, the division between functional modules / units described above does not necessarily correspond to the division between physical components; for example, a physical component may have multiple functions, or a function or step may be performed collaboratively by several physical components. Some or all of the physical components may be implemented as software executed by a processor, such as a central processing unit, a digital signal processor, or a microprocessor, or as hardware, or as an integrated circuit, such as an application-specific integrated circuit. Such software may be distributed on computer-readable media, which may include computer storage media (or non-transitory media) and communication media (or transient media). As is well known to those skilled in the art, the term computer storage media encompasses both volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technology, CD-ROM, digital versatile disks (DVD) or other optical disk storage, magnetic cassettes, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, as is well known to those skilled in the art, communication media typically embodies computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transport mechanism, and may include any information delivery media.

[0133] It should be understood that the term "and / or" used in the present specification and the appended claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations. It should be noted that, in this article, the terms "include", "comprise" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system that includes a series of elements includes not only those elements, but also includes other elements that are not explicitly listed, or also includes elements that are inherent to such process, method, article or system. In the absence of further limitations, an element defined by the sentence "including a..." does not exclude the presence of other identical elements in the process, method, article or system that includes the element.

[0134] The serial numbers of the above embodiments of the present invention are for descriptive purposes only and do not represent the superiority or inferiority of the embodiments. The above are only specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily think of various equivalent modifications or replacements within the technical scope disclosed by the present invention, and these modifications or replacements should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be based on the scope of protection of the claims.

Claims

1. A method for dynamic energy scheduling of intelligent new energy vehicles based on multi-source data fusion, characterized in that: The method comprises: Obtaining initial energy storage information corresponding to historical vehicles, and analyzing the data consistency of different target sensors based on the initial energy storage information to obtain the corresponding data correlation between any two target sensors; Classifying the target sensors according to the data association degree to obtain multiple target sensor sets; Obtaining target energy storage information corresponding to a target vehicle, as well as current position information and future path information corresponding to the target vehicle; Determining a road state corresponding to the road where the target vehicle is located according to the current position information and the future path information; Obtaining relevant energy storage information corresponding to the target vehicle from the initial energy storage information according to the target sensor set and the target energy storage information; Determining an initial moving length corresponding to the target vehicle according to the relevant energy storage information, and adjusting the initial moving length according to the road conditions to obtain a target moving length; Determining an initial charging position corresponding to the target vehicle according to the target movement length and the future path information; Performing dynamic energy scheduling according to the initial charging position to obtain a target scheduling result; The determining, based on the current position information and the future path information, a road state corresponding to the road where the target vehicle is located, includes: Determine a target road according to the current location information and the future path information, and obtain historical traffic parameters corresponding to the target road from a database; Determining indicator weight information corresponding to each traffic indicator based on the historical traffic parameters, and determining sample weight information corresponding to the historical traffic parameters; Performing data clustering on the historical traffic parameters according to the indicator weight information and the sample weight information to obtain a target clustering result; Determining the cluster center corresponding to each sub-cluster in the target clustering result, and determining the traffic state distribution corresponding to the sub-cluster; Obtaining current road parameters corresponding to the target road, and determining associated clusters corresponding to the current road parameters based on the current road parameters and the cluster centers; Determine a first state distribution corresponding to the current road parameter according to the associated cluster and the traffic state distribution; Merging the associated cluster and the current road parameter to obtain a target cluster, and performing parameter analysis on the target cluster to obtain a target distribution position corresponding to the target cluster; Determining a second state distribution corresponding to the current road parameter according to the target distribution position; fusing the first state distribution and the second state distribution to determine a target state distribution corresponding to the current road parameter; The road state corresponding to the road where the target vehicle is located is determined according to the target state distribution.

2. The method according to claim 1, characterized in that The target sensor includes at least a first sensor and a second sensor, the initial energy storage information includes at least first energy storage information corresponding to the first sensor and second energy storage information corresponding to the second sensor, and the analyzing the data consistency of different target sensors according to the initial energy storage information to obtain the corresponding data association between any two target sensors includes: Obtaining a first critical value corresponding to the first energy storage information, and determining a first acquisition time corresponding to the first sensor; Determine a first segmentation length corresponding to data segmentation of the first energy storage information according to the first acquisition duration and the first critical value; Obtaining a second critical value corresponding to the second energy storage information, and determining a second collection time corresponding to the second sensor; Determine a second segmentation length corresponding to data segmentation of the second energy storage information according to the second collection time and the second critical value; Performing data segmentation on the first energy storage information according to the first segmentation length to obtain a first segmentation unit corresponding to the first energy storage information and a first connection relationship corresponding to the first segmentation unit; performing smoothing processing on the first sub-data in each first segmentation unit according to the first segmentation unit and the first connection relationship to obtain a first processing result; Performing data segmentation on the second energy storage information according to the second segmentation length to obtain a second segmentation unit corresponding to the second energy storage information and a second connection relationship corresponding to the second segmentation unit; performing smoothing processing on the second sub-data in each second segmentation unit according to the second segmentation unit and the second connection relationship to obtain a second processing result; Obtaining a first center corresponding to the first processing result and a second center corresponding to the second processing result, and determining a target correlation factor based on the first processing result and the second processing result; Determining data consistency representation values corresponding to the first sensor and the second sensor based on the first center and the second center in combination with the target correlation factor; The data association degree corresponding to the first sensor and the second sensor is determined according to the data consistency characterization value.

3. The method according to claim 2, characterized in that The determining, based on the first center and the second center in combination with the target association factor, data consistency representation values corresponding to the first sensor and the second sensor includes: Obtaining first neighborhood data corresponding to each third sub-data from the first processing result, and determining a first filtering feature corresponding to the third sub-data based on the first neighborhood data; Obtaining second neighborhood data corresponding to each fourth sub-data from the second processing result, and determining a second filtering feature corresponding to the fourth sub-data based on the second neighborhood data; determining a target correlation factor corresponding to the first processing result and the second processing result according to the first filtering feature and the second filtering feature in combination with the third sub-data and the fourth sub-data; calculating a minimum value and a maximum value between the third sub-data and the fourth sub-data, and determining a relative distance between the third sub-data and the fourth sub-data according to the minimum value and the maximum value; Calculating a center distance between the first center and the second center, and determining a corresponding unit consistency representation value between the first processing result and the second processing result under the segmentation unit according to the center distance and the relative distance combined with the target association factor; The data consistent characterization values corresponding to the first sensor and the second sensor are obtained by fusing the plurality of the unit consistent characterization values.

4. The method according to claim 1, wherein The performing parameter analysis on the target cluster to obtain the target distribution position corresponding to the target cluster includes: Performing spatial and temporal division on the sub-road parameters corresponding to the target cluster to obtain spatial information and temporal information corresponding to each sub-road parameter; Determine a corresponding association parameter between any two of the sub-road parameters according to the spatial information and the time information; Normalizing the sub-road parameters to obtain normalized parameters corresponding to the sub-road parameters; Determining an adjacent space standard value corresponding to the sub-road parameter according to the association parameter and the normalization parameter; Determining a target position quadrant corresponding to the sub-road parameter according to the normalized parameter and the adjacent space standard value; The target position quadrant is counted to obtain the target distribution position corresponding to the target cluster.

5. The method according to claim 1, wherein The obtaining relevant energy storage information corresponding to the target vehicle from the initial energy storage information according to the target sensor set and the target energy storage information includes: Obtaining a third sensor and a fourth sensor associated with each subset according to the target sensor set; Obtaining first information corresponding to the third sensor and second information corresponding to the fourth sensor from the target energy storage information; Obtaining third information corresponding to the third sensor and fourth information corresponding to the fourth sensor from the initial energy storage information; Performing data clustering according to the first information and the third information to obtain a first clustering result, and obtaining a first cluster associated with the first information according to the first clustering result; Performing data clustering according to the second information and the fourth information to obtain a second clustering result, and obtaining a second cluster associated with the second information according to the second clustering result; Performing a union process on the first cluster and the second cluster to obtain associated energy storage information corresponding to the third sensor and the fourth sensor; Obtaining an initial data identifier corresponding to the associated energy storage information from a database, and determining the initial data identifier corresponding to each of the subsets in the target sensor set; Using the target sensor set, performing intersection processing on the initial data identifier corresponding to each of the subsets to obtain a target data identifier corresponding to the target sensor set; The relevant energy storage information corresponding to the target vehicle is obtained from the database according to the target data identifier.

6. The method according to claim 5, characterized in that The performing data clustering according to the first information and the third information to obtain a first clustering result includes: Performing data clustering on the first information and the third information using a clustering algorithm to obtain a third clustering result; Performing statistical calculation on each third cluster in the third clustering result to obtain a first mean and a first variance, and performing statistical calculation on the first remaining clusters in the third clustering result except the third cluster to obtain a second mean and a second variance; determining a first difference value between the third cluster and the first remaining cluster according to the first mean, the first variance, the second mean, and the second variance; Performing distance calculation on the third cluster to obtain first distance information corresponding to the third cluster; determining a first weight corresponding to the third cluster according to the first difference value and the first distance information; The third clustering result is re-clustered according to the first weight to obtain the first clustering result.

7. The method according to claim 6, characterized in that The performing data clustering according to the second information and the fourth information to obtain a second clustering result includes: Performing data clustering on the second information and the fourth information using the clustering algorithm to obtain a fourth clustering result; Performing statistical calculation on each fourth cluster in the fourth clustering result to obtain a third mean and a third variance, and performing statistical calculation on the second remaining clusters in the fourth clustering result except the fourth cluster to obtain a fourth mean and a fourth variance; determining a second difference value between the fourth cluster and the second remaining clusters according to the third mean, the third variance, the fourth mean, and the fourth variance; Performing distance calculation on the fourth cluster to obtain second distance information corresponding to the fourth cluster; Determine a second weight corresponding to the fourth cluster according to the second difference value and the second distance information; The fourth clustering result is re-clustered according to the second weight to obtain the second clustering result.

8. The method according to any one of claims 1 to 7, characterized in that After obtaining the target scheduling result, the method further includes: Determine a target charging pile, and determine a target associated vehicle corresponding to the target charging pile according to the target scheduling result; The vehicle energy storage information corresponding to the target associated vehicle is obtained, and a target charging sequence and a target charging power corresponding to the target associated vehicle are determined according to the target charging pile and the vehicle energy storage information.

9. An intelligent new energy vehicle energy dynamic scheduling system based on multi-source data fusion, characterized in that: include: A data analysis module is used to obtain initial energy storage information corresponding to historical vehicles, and analyze the data consistency of different target sensors based on the initial energy storage information to obtain the corresponding data correlation between any two target sensors; a relationship determination module, configured to classify the target sensors according to the data association degree to obtain a plurality of target sensor sets; A data acquisition module is used to obtain target energy storage information corresponding to a target vehicle, as well as current position information and future path information corresponding to the target vehicle; a state determination module, configured to determine a road state corresponding to the road where the target vehicle is located based on the current position information and the future path information; a data determination module, configured to obtain relevant energy storage information corresponding to the target vehicle from the initial energy storage information according to the target sensor set and the target energy storage information; a length calculation module, configured to determine an initial moving length corresponding to the target vehicle according to the relevant energy storage information, and adjust the initial moving length according to the road conditions to obtain a target moving length; a position determination module, configured to determine an initial charging position corresponding to the target vehicle according to the target movement length and the future path information; a scheduling determination module, configured to dynamically schedule energy according to the initial charging position to obtain a target scheduling result; The determining, based on the current position information and the future path information, a road state corresponding to the road where the target vehicle is located, includes: Determine a target road according to the current location information and the future path information, and obtain historical traffic parameters corresponding to the target road from a database; Determining indicator weight information corresponding to each traffic indicator based on the historical traffic parameters, and determining sample weight information corresponding to the historical traffic parameters; Performing data clustering on the historical traffic parameters according to the indicator weight information and the sample weight information to obtain a target clustering result; Determining the cluster center corresponding to each sub-cluster in the target clustering result, and determining the traffic state distribution corresponding to the sub-cluster; Obtaining current road parameters corresponding to the target road, and determining associated clusters corresponding to the current road parameters based on the current road parameters and the cluster centers; Determine a first state distribution corresponding to the current road parameter according to the associated cluster and the traffic state distribution; Merging the associated cluster and the current road parameter to obtain a target cluster, and performing parameter analysis on the target cluster to obtain a target distribution position corresponding to the target cluster; Determining a second state distribution corresponding to the current road parameter according to the target distribution position; fusing the first state distribution and the second state distribution to determine a target state distribution corresponding to the current road parameter; The road state corresponding to the road where the target vehicle is located is determined according to the target state distribution.

Images