A method and system for generating a power compensation path based on big data

Abstract

The application belongs to the technical field of path generation, and particularly relates to a big data-based energy supplement path generation method and system, which comprises the following steps: when a vehicle needs to supplement energy, all candidate charging stations are screened out; a trained Gaussian process regression model is used to obtain the expected waiting time mean and standard deviation of each candidate charging station, and the waiting time uncertainty factor of each candidate charging station is calculated; the risk influence coefficient of each candidate charging station is calculated in combination with the task time margin of the vehicle; the risk penalty factor of each candidate charging station is calculated; the basic recommendation score of each candidate charging station is calculated by the analytic hierarchy process based on the expected waiting time mean; the basic recommendation score of each candidate charging station is corrected by the risk penalty factor; the candidate charging station with the maximum final comprehensive recommendation score is taken as the optimal energy supplement path point, and is planned into the driving route of the vehicle. The application improves the reliability and safety of energy supplement decision.

Description

Technical Field

[0001] This invention relates to the field of path generation technology. More specifically, this invention relates to a method and system for generating energy replenishment paths based on big data. Background Technology

[0002] In modern logistics systems, the application of new energy electric vehicles is becoming increasingly widespread, and their on-the-go charging planning has become a key factor affecting operational efficiency: when a vehicle needs to recharge, it needs to select the optimal one from multiple candidate charging stations.

[0003] The Analytic Hierarchy Process (AHP) is a commonly used multi-criteria decision analysis method. It can quantify multiple complex criteria such as the distance to charging stations, charging prices, and charging speeds. By establishing a hierarchical structure model and calculating the weight scores of each option, it provides decision support for the selection of charging stations.

[0004] However, when applying the Analytic Hierarchy Process (AHP), existing technologies typically use a single deterministic value output by various prediction models as the decision input, such as "expected waiting time". This expected waiting time is a predicted value obtained by analyzing historical and real-time data, and it itself has a certain degree of volatility and uncertainty.

[0005] Inputting a predicted value with uncertainty into an analytic hierarchy process (AHP) model can prevent the model from assessing the risks hidden behind different options: Option A has an estimated waiting time of 10 minutes but the predicted result fluctuates greatly, while Option B has an estimated waiting time of 12 minutes but the predicted result is very stable. Existing technology will prioritize Option A, which has a higher risk, because 10 minutes is less than 12 minutes, thus bringing delay risks to actual logistics and delivery tasks and affecting the reliability and security of decision-making. Summary of the Invention

[0006] To address the aforementioned technical problem of decision-making risks arising from the failure to consider the uncertainty of predicted input values, the present invention provides solutions in several aspects.

[0007] In a first aspect, the present invention provides a method for generating charging routes based on big data, comprising: when a vehicle needs to recharge, selecting all candidate charging stations within the vehicle's reachable range from a charging station database based on the vehicle's current GPS coordinates and estimated remaining range; obtaining the mean and standard deviation of the estimated waiting time for each candidate charging station using a trained Gaussian process regression model; calculating a basic recommendation score for each candidate charging station based on the mean of the estimated waiting time using the analytic hierarchy process (AHP); using the ratio of the standard deviation of the estimated waiting time to the mean of the estimated waiting time for each candidate charging station as the waiting time uncertainty factor for each candidate charging station, and calculating the risk impact coefficient for each candidate charging station in conjunction with the vehicle's task time margin; calculating a risk penalty factor for each candidate charging station using the risk impact coefficient; correcting the basic recommendation score for each candidate charging station using the risk penalty factor to obtain a final comprehensive recommendation score for each candidate charging station; and selecting the candidate charging station with the highest final comprehensive recommendation score as the optimal charging route point and planning it into the vehicle's driving route.

[0008] This invention obtains both the mean and standard deviation of waiting time using a Gaussian process regression model, which represents the uncertainty of the prediction results. By combining this uncertainty with the vehicle's task time margin, a dynamic risk penalty mechanism is used to correct the basic recommendation score derived from the traditional analytic hierarchy process. This makes the recommendation of the charging route no longer a static selection of the charging station with the shortest theoretical waiting time, but an adaptive adjustment based on the urgency of the task. This effectively avoids the delay risk that may be caused by existing technologies ignoring prediction volatility, and significantly improves the reliability and safety of charging decisions in complex real-world application scenarios.

[0009] Preferably, the training process of the Gaussian process regression model includes: obtaining the feature data of each candidate charging station for each day in each training cycle and the average waiting time for each day to form a training set, wherein the average waiting time for each day is equal to the mean of the waiting time of all vehicles, and the waiting time of each vehicle is equal to the difference between the time of starting charging and the time of arrival at the charging station; using the feature data of each candidate charging station for each day in the training set as input and the average waiting time of each candidate charging station for each day in the training set as output, training a Gaussian process regression model, wherein the Gaussian process regression model learns the distribution therein to construct a probability distribution model about the input-output relationship, including the mean and standard deviation of the expected waiting time.

[0010] This invention uses a Gaussian process regression model to predict waiting time. It can not only predict the mean of the waiting time, but also output the standard deviation of the predicted value to represent the confidence or uncertainty of the prediction result, providing data input for subsequent risk assessment.

[0011] Preferably, the daily characteristic data of the candidate charging station in each training cycle includes: peak charging time and off-peak charging time of the day, whether it is a holiday, total number of charging guns, station type, maximum charging power of the charging station, and number of idle charging guns per hour of the day.

[0012] Preferably, the step of calculating the basic recommendation score for each candidate charging station using the analytic hierarchy process (AHP) includes: establishing a hierarchical model, which includes a target layer, a criterion layer, and a solution layer. The target layer refers to selecting the optimal charging station; the criterion layer includes time cost, economic cost, and convenience; and the solution layer includes all selected candidate charging stations. A criterion layer judgment matrix is ​​constructed. From the criterion layer judgment matrix, the quantified weight of each criterion is calculated and a consistency check is performed. The average expected waiting time, maximum charging power, electricity pricing strategy, and detour distance of each candidate charging station are used as inputs to calculate the score of each candidate charging station under each criterion. The scores of each candidate charging station under each criterion are weighted and summed using the quantified weight of each criterion to obtain the basic recommendation score for each candidate charging station.

[0013] Preferably, the criteria layer includes time cost, economic cost, and convenience, specifically: time cost equals the sum of the average expected waiting time, detour time, and expected charging time; economic cost equals the charging price, obtained based on the expected charging time and the charging station's electricity pricing strategy; convenience is used to measure the ease of reaching the charging station, and convenience equals the reciprocal of the detour distance.

[0014] Preferably, the vehicle's task time margin includes: locating the last task point requiring delivery from the vehicle's task sequence, and using the latest time point of the delivery time window of the last task point requiring delivery in the vehicle's task sequence as the planned latest delivery time for all subsequent delivery points of the vehicle. Using the multi-waypoint route calculation function of the route planning service API, an optimal driving route is planned for all task points in the task sequence connecting the vehicles, and the total driving time required to complete the route is given as the predicted remaining driving time for all subsequent delivery points. The product of the number of all task points in the vehicle's task sequence and the average service time is used as the predicted remaining service time for all subsequent delivery points of the vehicle. Vehicle task time margin ,in, Indicates the current time.

[0015] This invention obtains the current time margin available for handling additional delays by using the latest delivery time of the last task point and the estimated driving and service time, thereby enabling real-time monitoring of the urgency of the task and providing dynamically changing input parameters for calculating the risk impact coefficient. This closely links the refueling decision with the transportation task objectives.

[0016] Preferably, the calculation of the risk impact coefficient for each candidate charging station, taking into account the vehicle's task time margin, includes: In the formula, For the first Risk impact coefficient of each candidate charging station; For the first The uncertainty factor of waiting time for each candidate charging station; For the vehicle's task time margin; This is the adjustment constant; It is a natural exponential function.

[0017] This invention dynamically combines the uncertainty factor of the waiting time at charging stations with the task time margin of vehicles to calculate the risk impact coefficient. The system's sensitivity to risk automatically increases as the urgency of the task increases, making risk assessment closely related to specific task scenarios and greatly improving the scenario adaptability of decision-making.

[0018] Preferably, the step of calculating the risk penalty factor for each candidate charging station using the risk impact coefficient includes: In the formula, For the first Risk penalty factors for each candidate charging station; For the first Risk impact coefficient of each candidate charging station.

[0019] This invention uses a smooth nonlinear function to transform the risk impact coefficient into a risk penalty factor. When the risk impact is small, the recommendation score is minimally affected; however, when the risk impact increases, the penalty intensifies rapidly, reducing the weight of options with high base scores but also high potential risks. This filters out false optimal solutions in the final ranking, ensuring that the final recommended solution to the user is a robust choice that balances benefits and risks.

[0020] Preferably, the step of filtering all candidate charging stations within the vehicle's reachable range from the charging station database based on the vehicle's current GPS coordinates and estimated remaining range includes: performing a fast circular area query in the charging station database with the vehicle's current GPS coordinates as the center and the estimated remaining range as the radius to obtain a coarse list of candidate charging stations; traversing each candidate charging station in the coarse list, using the vehicle's current GPS coordinates as the starting point and the GPS coordinates of each candidate charging station as the ending point, calling the route planning service API to estimate the driving navigation distance from the vehicle to each candidate charging station; multiplying the vehicle's current GPS coordinates and estimated remaining range by a safety factor to obtain the vehicle's estimated remaining safe driving range; and retaining all candidate charging stations whose driving navigation distance is less than or equal to the estimated remaining safe driving range.

[0021] Secondly, the present invention provides a big data-based power replenishment path generation system, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned big data-based power replenishment path generation method is implemented.

[0022] By adopting the above technical solution, a computer program is generated from the above-mentioned big data-based energy replenishment path generation method and stored in a memory so that it can be loaded and executed by a processor. In this way, a terminal device can be made based on the memory and the processor for convenient use.

[0023] The beneficial effects of this invention are as follows:

[0024] This invention obtains both the mean and standard deviation of waiting time using a Gaussian process regression model, which represents the uncertainty of the prediction results. By combining this uncertainty with the vehicle's task time margin, a dynamic risk penalty mechanism is used to correct the basic recommendation score derived from the traditional analytic hierarchy process. This makes the recommendation of the charging route no longer a static selection of the charging station with the shortest theoretical waiting time, but an adaptive adjustment based on the urgency of the task. This effectively avoids the delay risk that may be caused by existing technologies ignoring prediction volatility, and significantly improves the reliability and safety of charging decisions in complex real-world application scenarios. Attached Figure Description

[0025] Figure 1 This is a flowchart illustrating a big data-based energy replenishment path generation method according to the present invention;

[0026] Figure 2 This is a schematic diagram illustrating step S4.

[0027] Figure 3 This is a schematic diagram showing the comparison of the recommended scores of each candidate charging station before and after correction when the vehicle's task time margin is 20 minutes.

[0028] Figure 4 This is a schematic diagram showing the comparison of the recommended scores of each candidate charging station before and after correction when the vehicle's task time margin is 5 minutes. Detailed Implementation

[0029] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0030] The specific embodiments of the present invention will now be described in detail with reference to the accompanying drawings.

[0031] This invention discloses a method for generating energy replenishment paths based on big data, referring to... Figure 1 This includes steps S1-S5:

[0032] S1: When the vehicle needs to recharge, based on the vehicle's current GPS coordinates and estimated remaining range, all candidate charging stations within the vehicle's reachable range are selected from the charging station database.

[0033] It should be noted that in order to make comprehensive decisions, all basic data related to the decision criteria must first be obtained. This includes not only traditional decision factors, but also key data for assessing decision risk, namely the volatility of the waiting time prediction results.

[0034] Specifically, when a vehicle needs recharging, based on the vehicle's current GPS coordinates and estimated remaining range, all candidate charging stations within the vehicle's reach are selected from the charging station database. This process includes two stages: coarse screening and fine screening, balancing computational efficiency and accuracy. The specific method is as follows:

[0035] (1) Rapid coarse screening based on straight-line distance includes: in the charging station database, with the current GPS coordinates of the vehicle as the center and the estimated remaining driving range as the radius, a rapid circular area query is performed. All charging stations falling within this circular area are initially selected to form a candidate charging station coarse screening list; it can quickly eliminate a large number of obviously too far charging stations, reducing the number of candidate charging stations from tens of thousands to tens or hundreds, thus reducing the burden on the next stage of accurate calculation.

[0036] (2) Precise screening based on road network navigation: Traverse each candidate charging station in the coarse screening list of candidate charging stations, take the current GPS coordinates of the vehicle as the starting point and the GPS coordinates of each candidate charging station as the ending point, call the route planning service API, the route planning service API will estimate the driving navigation distance from the vehicle to each candidate charging station based on real-time traffic conditions; multiply the current GPS coordinates of the vehicle and the estimated remaining range by a safety factor of 0.8 to cope with the additional power consumption risks such as uphill and traffic jams, and obtain the estimated remaining safe driving range of the vehicle; retain all candidate charging stations that meet the condition that the driving navigation distance is less than or equal to the estimated remaining safe driving range of the vehicle, and the list of charging stations selected in this way is the final candidate charging station that the vehicle can safely reach with the current power.

[0037] S2: Using a trained Gaussian process regression model, obtain the mean and standard deviation of the expected waiting time for each candidate charging station.

[0038] Specifically, the feature data of each candidate charging station over the past 24 hours is obtained and input into a trained Gaussian process regression model. The model outputs two key pieces of information for each candidate charging station: the mean of the expected waiting time and the standard deviation of the expected waiting time. The mean of the expected waiting time is the core output of the model and is the prediction result of the expected waiting time of the candidate charging station. The standard deviation of the expected waiting time is the confidence level of the prediction result.

[0039] The daily characteristic data of the charging station includes: peak charging time and off-peak charging time of the day, whether it is a holiday, the total number of charging guns, the station type, the maximum charging power of the charging station, and the number of idle charging guns per hour of the day; whether it is a holiday is divided into weekdays, weekends and public holidays, and is coded with unique hot codes; the station type includes logistics parks, highway service areas and public parking lots, and is coded with unique hot codes.

[0040] The training process of the Gaussian process regression model is as follows: The feature data and average waiting time of each candidate charging station for each day in each training period are obtained to form a training set. The average waiting time is equal to the mean of the waiting times of all vehicles, and the waiting time of each vehicle is equal to the difference between the start time of charging and the arrival time of each vehicle at the charging station. The training period is one month. The feature data of each candidate charging station in the training set is used as input, and the average waiting time of each candidate charging station in the training set is used as output to train a Gaussian process regression model. The Gaussian process regression model learns the distribution within the model to construct a probability distribution model of the input-output relationship, including the mean and standard deviation of the expected waiting time.

[0041] S3: Based on the average expected waiting time, calculate the basic recommendation score for each candidate charging station using the analytic hierarchy process.

[0042] It should be noted that a basic score is calculated using the traditional analytic hierarchy process without considering risk. This score reflects the merits of the solution under ideal forecasting conditions and can serve as a benchmark to facilitate targeted risk adjustments in subsequent steps.

[0043] The basic recommendation score for each candidate charging station is calculated using the analytic hierarchy process. The entire process can be broken down into the following five specific operational steps:

[0044] 1. Establish a hierarchical structure model.

[0045] The decision problem of selecting the optimal charging station is decomposed into an objective layer, a criterion layer, and a solution layer, which structures and hierarchically presents the complex decision problem. The objective layer refers to the final decision goal, namely, selecting the optimal charging station. The criterion layer sets several key criteria for evaluating the merits of the objective, including time cost, economic cost, and convenience. The solution layer includes all the objects to be decided, namely all the candidate charging stations selected in step S1.

[0046] At the criteria level: Time cost is used to comprehensively measure the total time consumed in the charging process, and time cost is equal to the sum of the average expected waiting time, detour time, and expected charging time. The expected charging time is obtained based on the vehicle's remaining battery power, target battery power, and the charging station's maximum charging power; Economic cost is used to comprehensively measure the direct costs incurred in the charging process, and economic cost is equal to the charging price, which is obtained based on the expected charging time and the charging station's electricity pricing strategy; Convenience is used to measure the ease of reaching the charging station, and convenience is equal to the reciprocal of the detour distance.

[0047] 2. Construct the criterion layer judgment matrix.

[0048] By quantifying the relative importance of different criteria, a basis for calculating weights is provided. The construction of the criterion-level judgment matrix is ​​based on the experience of the fleet manager, who compares each criterion in the criterion layer pairwise to determine the relative importance of each pair of criteria, with values ​​ranging from [1, 9]. Here, 1 indicates that the two criteria are equally important, 3 indicates that the former is slightly more important than the latter, 5 indicates that the former is significantly more important than the latter, 7 indicates that the former is strongly more important than the latter, and 9 indicates that the former is extremely more important than the latter. 2, 4, 6, and 8 represent the intermediate values ​​between the above adjacent judgments. If the scale of the relative importance of criterion 1 and criterion 2 is 3, then the scale of the relative importance of criterion 2 and criterion 1 is... .

[0049] For example, fleet managers, based on experience, consider time cost to be slightly more important than economic cost; therefore, the relative importance scale of time cost to economic cost is 2. Time cost is significantly more important than convenience; therefore, the relative importance scale of time cost to convenience is 4. Economic cost is slightly more important than convenience; therefore, the relative importance scale of economic cost to convenience is 3. Thus, a criterion-level judgment matrix can be constructed: The horizontal axis of the matrix represents time cost, economic cost, and convenience, respectively, while the vertical axis represents time cost, economic cost, and convenience, respectively.

[0050] 3. From the criterion-level judgment matrix, calculate the quantification weight of each criterion and perform a consistency check.

[0051] 4. Using the average expected waiting time, maximum charging power, electricity pricing strategy, and detour distance of each candidate charging station obtained in S1 as input, calculate the score of each candidate charging station under each criterion.

[0052] 5. By using the quantitative weight of each criterion, the scores of each candidate charging station under each criterion are weighted and summed to obtain the basic recommendation score for each candidate charging station.

[0053] S4: Based on the mean and standard deviation of the expected waiting time for each candidate charging station, calculate the waiting time uncertainty factor for each candidate charging station, and combine it with the vehicle's task time margin to calculate the risk impact coefficient for each candidate charging station, which is used to calculate the risk penalty factor for each candidate charging station.

[0054] It should be noted that the uncertainty of the predicted data is used as a risk penalty, and the idealized base score obtained in S2 is corrected through the risk penalty, so that the decision result can effectively avoid risks.

[0055] Refer to the flowchart for step S4 Figure 2 The process includes steps S401 to S404, specifically as follows:

[0056] S401: Calculate the waiting time uncertainty factor for each candidate charging station based on the mean and standard deviation of the expected waiting time.

[0057] It should be noted that, in order to make horizontal comparisons between waiting times of different durations, this invention constructs a waiting time uncertainty factor based on the mean and standard deviation of the predicted waiting time to measure the degree of uncertainty.

[0058] Specifically, based on the mean and standard deviation of the expected waiting time for each candidate charging station obtained in S1, the waiting time uncertainty factor for each candidate charging station is calculated. The specific calculation formula is as follows:

[0059] ;

[0060] In the formula, For the first The uncertainty factor of waiting time for each candidate charging station; For the first Standard deviation of the expected waiting time for each candidate charging station; For the first The average expected waiting time for each candidate charging station.

[0061] This formula transforms absolute volatility into relative volatility by calculating the ratio of the standard deviation to the mean: when the expected waiting time is average... When it increases, even the standard deviation Unchanged, waiting time uncertainty factor This will also decrease, reflecting that for a longer waiting time, the relative risk of fluctuations of the same absolute duration is lower.

[0062] It should be noted that this factor achieves a normalized measure of the risk of waiting time prediction at different charging stations, making it possible to compare the two scenarios of "waiting for 5 minutes with a fluctuation of 2 minutes" and "waiting for 30 minutes with a fluctuation of 2 minutes" side-by-side.

[0063] S402: Calculate the vehicle's task time margin.

[0064] It should be noted that the actual impact of the risks brought about by uncertainty on the task is closely related to the time requirements of the task itself. A time-pressured task has a very low tolerance for any uncertainty.

[0065] Specifically, the task time margin of the vehicle is calculated as follows: Locate the last task point requiring delivery from the vehicle's task sequence, and use the latest time point of the delivery time window for that task point as the planned latest delivery time for all subsequent delivery points of the vehicle. Using the multi-waypoint route calculation function of the route planning service API, an optimal driving route is planned for all task points in the task sequence connecting the vehicles, and the total driving time required to complete the route is given as the predicted remaining driving time for all subsequent delivery points. The product of the number of all task points in the vehicle's task sequence and the average service time is used as the predicted remaining service time for all subsequent delivery points of the vehicle. The average service time refers to the average time spent on unloading, handover, and signing, which is set to 15 minutes.

[0066] Furthermore, based on the latest planned delivery time for all subsequent delivery points of the vehicle. Predicting remaining driving time Predicting remaining service time To obtain the vehicle's task time margin, the vehicle's task time margin is... ,in, This indicates the current time; the obtained task time margin indicates the maximum time the vehicle can withstand without affecting the final on-time delivery. Additional delays, such as those for charging and dealing with sudden traffic jams, if the total time spent on detours, waiting, and charging for a charging solution exceeds [a certain threshold]. This will inevitably lead to task delays.

[0067] S403: Calculate the risk impact coefficient of each candidate charging station by combining the waiting time uncertainty factor of each candidate charging station and the vehicle's task time margin.

[0068] Specifically, the risk impact coefficient of each candidate charging station is calculated by combining the waiting time uncertainty factor of each candidate charging station and the vehicle's task time margin. The specific calculation formula is as follows:

[0069] ;

[0070] In the formula, For the first Risk impact coefficient of each candidate charging station; For the first The uncertainty factor of waiting time for each candidate charging station; The vehicle's task time margin, in minutes; This is an adjustment constant used to control the rate at which the impact of the task time margin on the risk decays; It is a natural exponential function.

[0071] in, This is an adjustment constant used to control the rate at which the impact of task time margin on risk decays; it represents the system's sensitivity to time margin. The larger the value, the faster the risk impact will be attenuated, even with a small time margin, and the system will exhibit risk tolerance. The smaller the value, the less sensitive the system is to time margin. Even with a long time margin, the impact of risk is still significant, and the system exhibits risk aversion. The value range is set to [0.05, 0.5]. In this invention, the value range is... Set to 0.1.

[0072] The calculation of the risk impact coefficient of candidate charging stations involves associating the risk with the task scenario: when the task time margin... When it is very large, the exponent term Approaching 0, the overall risk impact coefficient... It also approaches 0, indicating that the impact of uncertainty is negligible; conversely, when the task time margin is high... When the value is very small or even negative, the exponential term approaches 1, and the risk impact coefficient is mainly determined by the uncertainty factor of the waiting time. A decision indicates that uncertainty will have a significant impact.

[0073] It should be noted that the risk impact coefficient enables the risk assessment to adapt to different scenarios. For tasks with different levels of urgency, the system can dynamically assess different levels of risk caused by the uncertainty of the same charging station.

[0074] S404: Calculate the risk penalty factor for each candidate charging station based on the risk impact coefficient.

[0075] Specifically, based on the risk impact coefficient of each candidate charging station, a risk penalty factor is calculated for each candidate charging station. The specific calculation formula is as follows:

[0076] ;

[0077] In the formula, For the first Risk penalty factors for each candidate charging station; For the first Risk impact coefficient of each candidate charging station.

[0078] The risk penalty factor is calculated as a smooth function with a range of (0,1): when the risk impact coefficient... When it is 0, the risk penalty factor A value of 1 indicates no penalty; as the risk impact coefficient increases, the risk penalty factor decreases smoothly and approaches 0, indicating that the penalty applied is becoming increasingly severe.

[0079] It should be noted that the risk penalty factor provides a basis for subsequent revision of the base recommendation score. The magnitude of the risk penalty factor directly reflects the degree of discount on the recommendation caused by uncertainty.

[0080] S5: By using risk penalty factors, the basic recommendation score of each candidate charging station is adjusted to obtain the final comprehensive recommendation score of each candidate charging station. The candidate charging station with the highest final comprehensive recommendation score is selected as the optimal charging route point and planned into the vehicle's driving route.

[0081] It should be noted that the final decision is derived by combining the basic recommendation score of the candidate charging stations with the risk assessment.

[0082] Specifically, the basic recommendation score of each candidate charging station is adjusted by using the risk penalty factor of each candidate charging station. That is, the product of the risk penalty factor and the basic recommendation score of each candidate charging station is used as the final comprehensive recommendation score of each candidate charging station.

[0083] Furthermore, the final comprehensive recommendation scores of all candidate charging stations are ranked, and the candidate charging station with the highest score is selected as the optimal charging route point and planned into the vehicle's driving route.

[0084] It should be noted that by multiplying the base recommendation score by the risk penalty factor, the scores of schemes with high uncertainty in waiting time prediction and tight vehicle task time are significantly reduced, while those schemes with stable prediction and low risk are retained or their rankings are improved. The final recommended charging station is the optimal choice after combining expected benefits and risk control.

[0085] It should be noted that the energy replenishment path recommendation method of the present invention can adaptively adjust according to the urgency of the task: when there is sufficient time for the task, a certain degree of uncertainty can be tolerated in exchange for better cost or time; when the task is urgent, charging stations with more stable prediction results and lower risk are given priority, thereby effectively avoiding the delay risk that may be caused by the prior art ignoring the volatility of prediction, and significantly improving the reliability and safety of energy replenishment decision-making in complex real-world application scenarios.

[0086] For example, the candidate charging stations for the vehicle include candidate charging station A, candidate charging station B, candidate charging station C, and candidate charging station D. Candidate charging station A has a short expected waiting time but extremely high uncertainty, making it a high-risk option; candidate charging station B has a slightly longer expected waiting time but extremely low uncertainty, making it a very safe low-risk option; candidate charging station C has the lowest charging price but the longest waiting and detour times, with moderate uncertainty; candidate charging station D has the shortest detour distance and waiting time, but also high uncertainty, making it another high-risk option. The specific analysis process is as follows:

[0087] 1. The basic recommendation score for each candidate charging station is calculated using the analytic hierarchy process. The basic recommendation scores for candidate charging stations A to D are 0.741, 0.269, 0.4, and 0.76, respectively. Candidate charging station D has the highest basic recommendation score because it has the lowest total time cost, followed closely by candidate charging station A.

[0088] 2. When the vehicle's task time margin is 20 minutes: Candidate charging stations A and D have high expected waiting time standard deviations, resulting in correspondingly high calculated waiting time uncertainty factors and ultimately lower risk penalty factors, leading to significant score penalties. Conversely, candidate charging stations B and C have relatively stable expected waiting time standard deviations, with risk penalty factors close to 1, and their final comprehensive recommended scores remain essentially unchanged. Although candidate charging stations A and D have high base scores, multiplying them by the risk penalty factor significantly lowers their final comprehensive recommended scores. While candidate charging station D still has the highest final comprehensive recommended score, the gap with candidate charging station A has narrowed. A comparison chart of the recommended scores of each candidate charging station before and after correction is shown below. Figure 3 As shown, the final comprehensive recommendation scores for candidate charging stations A to D are 0.665, 0.264, 0.389, and 0.675, respectively, and the final optimal charging route point is candidate charging station D.

[0089] 3. The optimal charging route recommendation model of this invention has strong scene adaptability. When the vehicle's task time margin is 5 minutes: the risk penalty factors of candidate charging stations A and D are smaller, resulting in heavier penalties. The comparison chart of the recommendation scores of each candidate charging station before and after correction is shown in the figure. Figure 4 As shown, the final comprehensive recommendation scores for candidate charging stations A to D are 0.456, 0.249, 0.354, and 0.447, respectively, and the final optimal charging route point is candidate charging station A.

[0090] This invention also discloses a big data-based power replenishment path generation system, including a processor and a memory. The memory stores computer program instructions, and when the computer program instructions are executed by the processor, a big data-based power replenishment path generation method according to this invention is implemented.

[0091] The system also includes other components well known to those skilled in the art, such as communication buses and communication interfaces, the settings and functions of which are known in the art and will not be described in detail here.

Claims

1. A method for generating energy replenishment paths based on big data, characterized in that, include: When a vehicle needs to recharge, based on the vehicle's current GPS coordinates and estimated remaining range, all candidate charging stations within the vehicle's reach are selected from the charging station database. The mean and standard deviation of the expected waiting time for each candidate charging station are obtained by using a trained Gaussian process regression model. Based on the average expected waiting time, the basic recommendation score for each candidate charging station is calculated using the analytic hierarchy process (AHP). This includes: establishing a hierarchical model comprising a target layer, a criterion layer, and a solution layer. The target layer refers to selecting the optimal charging station; the criterion layer includes time cost, economic cost, and convenience; and the solution layer includes all selected candidate charging stations. A criterion layer judgment matrix is ​​constructed. From the criterion layer judgment matrix, the quantified weight of each criterion is calculated and a consistency check is performed. Using the average expected waiting time, maximum charging power, electricity pricing strategy, and detour distance of each candidate charging station as inputs, the score of each candidate charging station under each criterion is calculated. Finally, using the quantified weight of each criterion, the scores of each candidate charging station under each criterion are weighted and summed to obtain the basic recommendation score for each candidate charging station. The ratio of the standard deviation of the expected waiting time to the mean of the expected waiting time for each candidate charging station is used as the waiting time uncertainty factor for each candidate charging station. Combined with the vehicle's task time margin, the risk impact coefficient for each candidate charging station is calculated, including: In the formula, For the first Risk impact coefficient of each candidate charging station; For the first The uncertainty factor of waiting time for each candidate charging station; For the vehicle's task time margin; This is the adjustment constant; The natural exponential function is used; the risk penalty factor for each candidate charging station is calculated using the risk impact coefficient, including: In the formula, For the first Risk penalty factor for each candidate charging station; by adjusting the basic recommendation score of each candidate charging station based on the risk penalty factor, the final comprehensive recommendation score of each candidate charging station is obtained. The candidate charging station with the highest final comprehensive recommendation score is selected as the optimal charging route point and planned into the vehicle's driving route.

2. The method for generating a replenishment path based on big data according to claim 1, characterized in that, The training process of the Gaussian process regression model includes: The training set is formed by obtaining the daily feature data and average waiting time of each candidate charging station for each training cycle. The average waiting time is equal to the mean of the waiting times of all vehicles, and the waiting time of each vehicle is equal to the difference between the start time of charging and the time of arrival at the charging station. The daily feature data of each candidate charging station in the training set is used as input, and the average waiting time of each candidate charging station in the training set is used as output. A Gaussian process regression model is trained. The Gaussian process regression model learns the distribution to construct a probability distribution model of the input-output relationship, including the mean and standard deviation of the expected waiting time.

3. The method for generating a power replenishment path based on big data according to claim 2, characterized in that, The feature data of the candidate charging stations for each day in each training cycle includes: The data includes peak and off-peak charging times during the day, whether it is a holiday, the total number of charging guns, the type of charging station, the maximum charging power of the charging station, and the number of idle charging guns per hour during the day.

4. The method for generating a replenishment path based on big data according to claim 1, characterized in that, The criteria layer includes time cost, economic cost, and convenience, specifically: Time cost equals the sum of the average expected waiting time, detour time, and expected charging time; Economic cost equals charging price, which is determined based on the expected charging time and the charging station's electricity pricing strategy; Convenience measures how easily one can reach a charging station; convenience is equal to the reciprocal of the detour distance.

5. The method for generating a replenishment path based on big data according to claim 1, characterized in that, The vehicle's task time margin includes: From the vehicle's task sequence, locate the last delivery point that needs to be delivered. Use the latest delivery time window of this last delivery point as the planned latest delivery time for all subsequent delivery points. Using the multi-waypoint route calculation function of the route planning service API, an optimal driving route is planned for all task points in the task sequence connecting the vehicles, and the total driving time required to complete the route is given as the predicted remaining driving time for all subsequent delivery points. The product of the number of all task points in the vehicle's task sequence and the average service time is used as the predicted remaining service time for all subsequent delivery points of the vehicle. Vehicle task time margin ,in, Indicates the current time.

6. The method for generating a replenishment path based on big data according to claim 1, characterized in that, Based on the vehicle's current GPS coordinates and estimated remaining range, the system filters out all candidate charging stations within the vehicle's reachable range from a charging station database, including: In the charging station database, a fast circular area query is performed with the vehicle's current GPS coordinates as the center and the estimated remaining driving range as the radius to obtain a preliminary list of candidate charging stations. Iterate through each candidate charging station in the preliminary list, starting from the vehicle's current GPS coordinates and ending at the GPS coordinates of each candidate charging station. Call the route planning service API to estimate the driving navigation distance from the vehicle to each candidate charging station. Multiply the vehicle's current GPS coordinates and the estimated remaining range by a safety factor to obtain the vehicle's estimated remaining safe driving range. Retain all candidate charging stations whose driving navigation distance is less than or equal to the estimated remaining safe driving range.

7. A big data-based energy replenishment path generation system, characterized in that, include: A processor and a memory, wherein the memory stores computer program instructions that, when executed by the processor, implement a big data-based energy replenishment path generation method according to any one of claims 1-6.

Images