Vehicle battery replacement management method and system applying cloud side cooperation technology

Abstract

The invention provides a vehicle battery replacement management method and system applying a cloud-side collaboration technology, and relates to the field of cloud-side collaboration, and the method comprises the steps: determining a related road of each battery replacement station based on historical battery replacement demand data of a management region, and then determining a battery replacement station optimization scheme of the management region; based on the battery swap station optimization scheme of the management area, optimizing a plurality of battery swap stations of the management area; on the basis of the related road of each battery swap station, real-time battery swap influence data of the management area is obtained; the cloud platform predicts future power conversion loads of the plurality of power conversion stations based on the real-time power conversion load data of the plurality of power conversion stations and the real-time power conversion influence data of the management area, and issues the future power conversion loads to the edge computing device; the cloud platform receives a battery replacement request initiated by a user and allocates an edge computing device; and the edge calculation device generates the optimal battery replacement strategy corresponding to the battery replacement request based on the future battery replacement loads of the plurality of battery replacement stations, and the method has the advantages of improving the battery replacement efficiency of the vehicle and the user experience.

Description

Technical Field

[0001] This invention relates to the field of cloud-edge collaboration, and in particular to a vehicle battery swapping management method and system that applies cloud-edge collaboration technology. Background Technology

[0002] In recent years, with the depletion of oil resources and severe air pollution, the development of electric vehicles has become an inevitable trend in the automotive industry. Vigorously developing electric vehicles is an important choice for ensuring my country's energy security and achieving sustainable development of the automotive industry.

[0003] Currently, pure electric vehicles primarily utilize two energy replenishment modes: vehicle charging and battery swapping. Charging an electric vehicle via AC is slow and time-consuming, and parking is limited. DC fast charging offers high power and short charging times, but it puts a significant strain on the power grid and reduces battery lifespan. Battery swapping, on the other hand, allows for coordinated charging through grid interaction, enabling peak-shaving and energy storage of the power load, improving the overall efficiency of power equipment. It provides rapid energy replenishment for electric vehicles, reducing user waiting time, without compromising battery lifespan.

[0004] As battery swapping stations gain increasing acceptance, more and more automakers are joining the development process. Simultaneously, the increase in users has led to a gradual rise in the number of vehicles that can be swapped per unit of time at these stations. Some stations are operating at full capacity during the day and still experience queues. In this situation, in addition to continuously improving the mechanical efficiency of battery swapping, effective management of swapping vehicles, scheduling of swaps, and ensuring the orderly and reliable operation of battery swapping stations are essential measures.

[0005] Therefore, there is a need to provide vehicle battery swapping management methods and systems that utilize cloud-edge collaborative technologies to improve vehicle battery swapping efficiency and user experience. Summary of the Invention

[0006] This invention provides a vehicle battery swapping management method using cloud-edge collaborative technology, comprising: acquiring historical battery swapping demand data for a management area; determining the relevant roads for each battery swapping station based on the historical battery swapping demand data for the management area; determining an optimization scheme for the battery swapping stations in the management area based on the historical battery swapping demand data for the management area and the relevant roads for each battery swapping station; optimizing multiple battery swapping stations in the management area based on the optimization scheme for the battery swapping stations in the management area; acquiring real-time battery swapping impact data for the management area based on the relevant roads for each battery swapping station; the cloud platform predicting the future battery swapping load of multiple battery swapping stations based on the real-time battery swapping load data of multiple battery swapping stations and the real-time battery swapping impact data of the management area, and distributing the predictions to edge computing devices; the cloud platform receiving battery swapping requests initiated by users and allocating edge computing devices; and the edge computing devices generating the optimal battery swapping strategy corresponding to the battery swapping request based on the future battery swapping load of multiple battery swapping stations.

[0007] Further, the historical battery swap demand data of the management area includes battery swap loads of the plurality of battery swap stations and quantities of battery swap demand vehicles of the plurality of roads in a plurality of historical time periods; and the relevant roads of each battery swap station are determined based on the historical battery swap demand data of the management area, including: for each battery swap station and each road, a first correlation coefficient of the battery swap station and the road is calculated based on the battery swap loads of the battery swap station and the quantities of battery swap demand vehicles of the road in the plurality of historical time periods; and the relevant roads of each battery swap station are determined based on the first correlation coefficients of each battery swap station and each road.

[0008] Further, the battery swap station optimization scheme of the management area is determined based on the historical battery swap demand data of the management area and the relevant roads of each battery swap station, including: for each relevant road of the battery swap station, a demand delay time corresponding to the battery swap station and the relevant road is determined based on the battery swap loads of the battery swap station and the quantities of battery swap demand vehicles of the relevant road in the plurality of historical time periods; and the battery swap station optimization scheme of the management area is determined based on the battery swap loads of the plurality of battery swap stations, the quantities of battery swap demand vehicles of the plurality of roads and the demand delay times corresponding to each battery swap station and the relevant roads in the plurality of historical time periods.

[0009] Further, the demand delay time corresponding to the battery swap station and the relevant road is determined based on the battery swap loads of the battery swap station and the quantities of battery swap demand vehicles of the relevant road in the plurality of historical time periods, including: a plurality of candidate demand delay times are determined based on the battery swap loads of the battery swap station and the quantities of battery swap demand vehicles of the relevant road in the plurality of historical time periods; for each candidate demand delay time, a quantity sequence of battery swap demand vehicles corresponding to the candidate demand delay time is determined based on the candidate demand delay time and the quantities of battery swap demand vehicles of the relevant road in the plurality of historical time periods, a second correlation coefficient of the battery swap station and the relevant road corresponding to the candidate demand delay time is calculated based on the battery swap loads of the battery swap station and the quantity sequence of battery swap demand vehicles of the relevant road corresponding to the candidate demand delay time; and the demand delay time corresponding to the battery swap station and the relevant road is determined based on the second correlation coefficients of the battery swap station and the relevant road corresponding to each candidate demand delay time.

[0010] Further, the plurality of candidate demand delay times are determined based on the battery swap loads of the battery swap station and the quantities of battery swap demand vehicles of the relevant road in the plurality of historical time periods, including: the battery swap loads of the battery swap station in the plurality of historical time periods are subjected to variational mode decomposition, and key load features of the battery swap station are extracted from a variational mode decomposition result; the quantities of battery swap demand vehicles of the relevant road in the plurality of historical time periods are subjected to variational mode decomposition, and key quantity features of the relevant road are extracted from a variational mode decomposition result; and the plurality of candidate demand delay times are determined based on the key load features of the battery swap station and the key quantity features of the relevant road.

[0011] Further, based on the battery swap load of the plurality of battery swap stations in a plurality of historical time periods, the number of battery swap demand vehicles of the plurality of roads, and the demand delay time corresponding to each battery swap station and the related road, a battery swap station optimization scheme of the management area is determined, including: establishing a constraint condition, wherein the constraint condition at least includes the maximum number of battery swap positions of each battery swap station and the position constraint of the newly added battery swap station; generating a plurality of candidate battery swap station optimization schemes, wherein the candidate battery swap station optimization scheme includes the battery swap position of each battery swap station, the position and battery swap position of each newly added battery swap station; based on the battery swap load of the plurality of battery swap stations in a plurality of historical time periods, the number of battery swap demand vehicles of the plurality of roads, and the demand delay time corresponding to each battery swap station and the related road, a plurality of battery swap demand scenarios are established; according to the plurality of candidate battery swap station optimization schemes and the plurality of battery swap demand scenarios, the battery swap station optimization scheme of the management area is determined.

[0012] Further, the real-time battery swap influence data of the management area includes the real-time number of battery swap demand vehicles of the plurality of roads; the cloud platform predicts the future battery swap load of the plurality of battery swap stations based on the real-time battery swap load data of the plurality of battery swap stations and the real-time battery swap influence data of the management area, including: for each battery swap station, based on the real-time battery swap load data of the battery swap station, the real-time number of battery swap demand vehicles of the related road, and the demand delay time corresponding to the battery swap station and the related road, the future battery swap load of the battery swap station is predicted.

[0013] Further, based on the real-time battery swap load data of the battery swap station, the real-time number of battery swap demand vehicles of the related road, and the demand delay time corresponding to the battery swap station and the related road, the future battery swap load of the battery swap station is predicted, including: based on the battery swap load of the battery swap station in a plurality of historical time periods, the number of battery swap demand vehicles of each related road, and the demand delay time corresponding to the battery swap station and each related road, a load prediction model corresponding to the battery swap station is established; through the load prediction model corresponding to the battery swap station, based on the real-time battery swap load data of the battery swap station, the real-time number of battery swap demand vehicles of the related road of the battery swap station, and the demand delay time corresponding to the battery swap station and the related road, the future battery swap load of the battery swap station is predicted.

[0014] Further, the edge computing device generates an optimal battery swap strategy corresponding to the battery swap request based on the future battery swap load of the plurality of battery swap stations, including: establishing a multi-index strategy evaluation function, wherein the multi-index strategy evaluation function is at least related to a battery swap station load balancing index and a battery swap time cost index; based on the future battery swap load of the plurality of battery swap stations and the multi-index strategy evaluation function, an optimal battery swap strategy corresponding to the battery swap request is generated.

[0015] The application provides a vehicle battery replacement management system using cloud edge collaboration technology, and a vehicle battery replacement management method using the cloud edge collaboration technology.

[0016] Compared with the prior art, the vehicle battery replacement management method and system using cloud edge collaboration technology provided by the application have at least the following beneficial effects: 1. By analyzing the historical battery replacement demand data of the management area, the spatio-temporal law of battery replacement behavior is mined, data support is provided for battery replacement station optimization, and resource waste is avoided.

[0017] 2. The cloud platform centrally processes global data, the edge computing device focuses on local real-time information, and the two work together to realize a prediction mode of "global planning + local precision", which significantly improves the prediction accuracy (such as reducing the error by more than 30%). Based on the future battery replacement load prediction result, the edge computing device generates an optimal battery replacement path (such as recommending the nearest battery replacement station with the shortest queuing time) in real time by combining user location, vehicle power, and battery replacement station idle state information, thereby reducing the blindness of users searching for battery replacement stations. Through cloud edge collaboration, the system can real-time perceive battery replacement station load changes (such as sudden peaks), dynamically adjust battery replacement strategies (such as guiding users to shunt to low-load sites), avoid local congestion, and shorten the average waiting time of users by more than 50%.

[0018] 3. The cloud platform uniformly manages global data and algorithm models, and the edge computing device distributes local computing, thereby reducing data transmission volume (such as uploading only key indicators) and reducing network bandwidth cost by more than 40%. The system architecture does not need to be reconstructed when a new battery replacement station or edge node is seamlessly accessed, and the system is convenient for large-area (such as city-level) popularization. BRIEF DESCRIPTION OF DRAWINGS

[0019] The present specification will be further illustrated in the manner of exemplary embodiments, which will be described in detail with reference to the accompanying drawings. These embodiments are not restrictive, and in these embodiments, the same numbers represent the same structures, in which: Figure 1 is a flowchart of a vehicle battery swap management method applying cloud edge collaboration technology according to some embodiments of the present specification; Figure 2 is a flowchart of key load characteristics of a battery swap station and key quantity characteristics of related roads according to some embodiments of the present specification; Figure 3 is a flowchart of determining an optimization scheme of a battery swap station in a management area according to some embodiments of the present specification; Figure 4 is a module diagram of a vehicle battery swap management system applying cloud edge collaboration technology according to some embodiments of the present specification. DETAILED DESCRIPTION

[0020] In order to more clearly illustrate the technical solutions of the embodiments of the present specification, the drawings needed in the embodiment description will be briefly introduced below. Obviously, the drawings in the following description are only some examples or embodiments of the present specification, and for those skilled in the art, the present specification can be applied to other similar scenarios without creative labor. Unless it is obvious from the language environment or otherwise stated, the same reference numbers in the drawings represent the same structure or operation.

[0021] Figure 1 is a flowchart of a vehicle battery swap management method applying cloud edge collaboration technology according to some embodiments of the present specification, as Figure 1 indicated, the vehicle battery swap management method applying cloud edge collaboration technology can include the following steps.

[0022] S110, historical battery swap demand data of a management area is obtained.

[0023] The historical battery swap demand data of the management area includes the battery swap load of a plurality of battery swap stations and the number of battery swap demand vehicles of a plurality of roads in a plurality of historical time periods (such as every hour, every day, every month). For example, the battery swap load can be represented by the number of battery swaps, and the number of battery swaps of battery swap station A from 8:00 to 9:00 on August 1, 2024 is 45, i.e. the battery swap load of battery swap station A is 45. The battery swap demand vehicles of the road in the historical time period can be vehicles that travel on the road in the historical time period and have a remaining battery level less than a remaining battery level threshold (for example, 20%, 25%, etc.) as battery swap demand vehicles.

[0024] S120, based on the historical battery swap demand data of the management area, the related roads of each battery swap station are determined.

[0025] Specifically, the method comprises the following steps: For each battery swap station and each road, a first correlation coefficient between the battery swap station and the road is calculated based on the battery swap load of the battery swap station and the number of battery swap demand vehicles of the road in a plurality of historical time periods. Based on the first correlation coefficient between each battery swap station and each road, the relevant road of each battery swap station is determined.

[0026] Specifically, the battery swap load of the battery swap station and the number of battery swap demand vehicles of the road in a plurality of historical time periods can be taken as two variables, and substituted into a correlation coefficient (for example, Pearson correlation coefficient, etc.) calculation formula to calculate the first correlation coefficient between the battery swap station and the road.

[0027] The road with an absolute value of the first correlation coefficient greater than a first threshold value (for example, 0.5) can be taken as the relevant road of the battery swap station.

[0028] S130, based on the historical battery swap demand data of the management area and the relevant road of each battery swap station, a battery swap station optimization scheme of the management area is determined.

[0029] Specifically, the method comprises the following steps: For each relevant road of the battery swap station, a demand delay time corresponding to the battery swap station and the relevant road is determined based on the battery swap load of the battery swap station and the number of battery swap demand vehicles of the relevant road in a plurality of historical time periods. Based on the battery swap load of a plurality of battery swap stations, the number of battery swap demand vehicles of a plurality of roads and the demand delay time corresponding to each battery swap station and the relevant road in a plurality of historical time periods, a battery swap station optimization scheme of the management area is determined.

[0030] In some embodiments, the demand delay time corresponding to the battery swap station and the relevant road is determined based on the battery swap load of the battery swap station and the number of battery swap demand vehicles of the relevant road in a plurality of historical time periods, comprising: A plurality of candidate demand delay times are determined based on the battery swap load of the battery swap station and the number of battery swap demand vehicles of the relevant road in a plurality of historical time periods. For each candidate demand delay time, a sequence of the number of battery swap demand vehicles corresponding to the candidate demand delay time of the relevant road is determined based on the candidate demand delay time and the number of battery swap demand vehicles of the relevant road in a plurality of historical time periods, and a second correlation coefficient between the battery swap station and the relevant road corresponding to the candidate demand delay time is calculated based on the battery swap load of the battery swap station and the sequence of the number of battery swap demand vehicles of the relevant road corresponding to the candidate demand delay time in a plurality of historical time periods. The demand delay time corresponding to the battery swap station and the relevant road is determined based on the second correlation coefficient between the battery swap station and the relevant road corresponding to each candidate demand delay time.

[0031] Specifically, the peak of the battery swap load of the battery swap station may lag behind the peak of the vehicle battery swap demand of the related road, and the reasons include: Geographical distance: the driving time from the vehicle generating the battery swap demand on the road to the battery swap station (e.g., 10 minutes for a vehicle in a logistics park to reach the battery swap station); Behavioral inertia: the driver may prefer to continue driving to the next battery swap station or wait until the battery is low before swapping; Traffic congestion: traffic congestion during peak hours delays the arrival of vehicles at the battery swap station.

[0032] Through historical data analysis, the time delay relationship between the battery swap load of the battery swap station and the vehicle demand of the related road is quantified to provide a basis for dynamic scheduling.

[0033] A plurality of candidate demand delay times can be determined based on the battery swap load of the battery swap station and the number of battery swap demand vehicles of the related road in a plurality of historical time periods in any manner, for example, a demand delay time range is determined, for example, 0 minutes to 120 minutes, and 10 minutes is taken as a time step, so the set of a plurality of candidate demand delay times can be {0, 10, 20, …, 120}.

[0034] For each candidate delay time, the number of battery swap demand vehicles of the related road in a plurality of historical time periods is shifted forward by the candidate delay time, and the number sequence of battery swap demand vehicles of the related road corresponding to the candidate demand delay time is obtained. From the battery swap load of the battery swap station in a plurality of historical time periods, the battery swap load sequence that is time-aligned with the number sequence is cut from the battery swap load of the battery swap station in a plurality of historical time periods. The number sequence of battery swap demand vehicles of the related road corresponding to the candidate demand delay time and the battery swap load sequence are taken as two variables, and substituted into the calculation formula of the correlation coefficient (e.g., Pearson correlation coefficient, etc.) to calculate the second correlation coefficient of the battery swap station and the related road corresponding to the candidate demand delay time.

[0035] The candidate demand delay time corresponding to the maximum second correlation coefficient can be taken as the demand delay time corresponding to the battery swap station and the related road.

[0036] Preferably, based on the battery swap load of the battery swap station and the number of battery swap demand vehicles of the related road in a plurality of historical time periods, a plurality of candidate demand delay times are determined, including: Perform variational mode decomposition on the battery swap load of the battery swap station in a plurality of historical time periods, and extract the key load characteristics of the battery swap station from the variational mode decomposition result; Perform variational mode decomposition on the number of battery swap demand vehicles of the related road in a plurality of historical time periods, and extract the key number characteristics of the related road from the variational mode decomposition result; Based on the key load characteristics of the battery swap station and the key number characteristics of the related road, a plurality of candidate demand delay times are determined.

[0037] Figure 2 is a flowchart of key load characteristics of the battery swap station and key quantity characteristics of the related road according to some embodiments of the present specification, as shown in Figure 2 As shown, specifically, the key load characteristics of the battery swap station and the key quantity characteristics of the related road can be extracted according to the following flow: S11, a plurality of sets of sample data are obtained, wherein the sample data includes the battery swap load of the battery swap station, the quantity of battery swap demand vehicles and the demand delay time of the related road in a plurality of historical time periods of a sample area; S12, a plurality of load modal component characteristic factors (for example, time domain characteristic factors (for example, mean, variance, peak value, kurtosis, energy, etc.), frequency domain characteristics (for example, center frequency, bandwidth, spectral energy, etc.), time-frequency domain characteristics (for example, instantaneous frequency, instantaneous amplitude, time-frequency energy distribution, etc.), modal correlation characteristics (for example, energy ratio, correlation coefficient of adjacent modal components, etc.), etc.) and a plurality of quantity modal component factors (for example, time domain characteristic factors (for example, mean, variance, peak value, kurtosis, energy, etc.), frequency domain characteristics (for example, center frequency, bandwidth, spectral energy, etc.), time-frequency domain characteristics (for example, instantaneous frequency, instantaneous amplitude, time-frequency energy distribution, etc.), modal correlation characteristics (for example, energy ratio, correlation coefficient of adjacent modal components, etc.), etc.) are determined; S13, for each set of sample data, the battery swap load of the battery swap station in a plurality of historical time periods of the sample area is subjected to variational modal decomposition to obtain a plurality of load modal components, and the quantity of battery swap demand vehicles of the related road in a plurality of historical time periods of the sample area is subjected to variational modal decomposition to obtain a plurality of quantity modal components; S14, a first fitness function is constructed, wherein the independent variables of the first fitness function include the correlation coefficient of the set of load modal component characteristic factors and the demand delay time, and the greater the correlation coefficient of the set of load modal component characteristic factors and the demand delay time, the greater the function value of the first fitness function; S15, sample at least two load modal component feature factors from the plurality of load modal component feature factors as a load modal component feature factor set, generate a plurality of load modal component feature factor sets through multiple samplings, for each load modal component feature factor set, extract the key load features corresponding to the load modal component feature factor set from the plurality of load modal components of the plurality of sample data groups, calculate the cosine similarity of the key load features of any two sample data groups, and calculate the absolute value of the time difference of the demand delay time of any two sample data groups, take the cosine similarity of the key load features of any two sample data groups and the absolute value of the time difference of the demand delay time of any two sample data groups as two variables, and substitute them into the correlation coefficient (for example, Spearman rank correlation coefficient, Kendall rank correlation coefficient, etc.) calculation formula to calculate the correlation coefficient of the load modal component feature factor set and the demand delay time; S16, select the load modal component feature factor set with the maximum function value of the first fitness function, and extract the key load features of the battery swap station from the variational modal decomposition result; S17, construct a second fitness function, wherein the independent variables of the second fitness function include the correlation coefficient of the quantity modal component feature factor set and the demand delay time, and the greater the correlation coefficient of the quantity modal component feature factor set and the demand delay time, the greater the function value of the second fitness function; S18, sample at least two quantity modal component feature factors from the plurality of quantity modal component feature factors as a quantity modal component feature factor set, generate a plurality of quantity modal component feature factor sets through multiple samplings, for each quantity modal component feature factor set, extract the key quantity features corresponding to the quantity modal component feature factor set from the plurality of quantity modal components of the plurality of sample data groups, calculate the cosine similarity of the key quantity features of any two sample data groups, and calculate the absolute value of the time difference of the demand delay time of any two sample data groups, take the cosine similarity of the key quantity features of any two sample data groups and the absolute value of the time difference of the demand delay time of any two sample data groups as two variables, and substitute them into the correlation coefficient (for example, Spearman rank correlation coefficient, Kendall rank correlation coefficient, etc.) calculation formula to calculate the correlation coefficient of the quantity modal component feature factor set and the demand delay time; S19, select the quantity modal component feature factor set with the maximum function value of the second fitness function, and extract the key quantity features of the battery swap station from the variational modal decomposition result.

[0038] Calculate the cosine similarity of the key load features of the battery swap station and the key quantity features of the related road and the key load features of the plurality of sample data and the key quantity features of the related road, take the sample data with a cosine similarity greater than a cosine similarity threshold (for example, 0.5) as similar sample data, and take the demand delay time of the similar sample data as a plurality of candidate demand delay times.

[0039] It can be understood that by decomposing the battery swap load and the number of demand vehicles into multiple modal components through variational modal decomposition (VMD), the limitations of traditional time domain or frequency domain analysis are broken through, and the hidden patterns (such as periodic fluctuations, instantaneous impacts, etc.) in nonlinear and non-stationary signals are effectively captured, providing a richer information base for feature extraction. A complex feature factor library covering time domain (mean, variance, peak), frequency domain (center frequency, bandwidth), time-frequency domain (instantaneous frequency, time-frequency energy), and modal correlation (energy ratio, adjacent modal correlation coefficient) is constructed to comprehensively characterize the dynamic characteristics of the load and the number of demands, avoiding information loss caused by a single feature. By constructing the first and second fitness functions, the correlation coefficient (such as the Spearman rank correlation coefficient) between the feature factor set and the demand delay time is used as the optimization target to automatically select the feature combination with the strongest explanation power for the delay time. This data-driven approach reduces the subjectivity of manual feature selection and improves the correlation strength between features and target variables. By calculating the cosine similarity between the current key features and the historical sample features, sample data with a similarity higher than a threshold value is selected, and its demand delay time is directly referenced as a candidate value. This method uses the distribution rules of historical data to avoid the blindness of subjectively setting the delay time, and the robustness of the cosine similarity is enhanced due to its insensitivity to feature scale. Based on the data of multiple historical time periods, the model is trained to adapt to different time periods (such as peak / flat) and different regions (such as urban / rural) of battery swap demand mode differences, ensuring the applicability of the candidate delay time in different scenarios.

[0040] Figure 3 is a flowchart of determining the optimization scheme of the battery swap station of the management area according to some embodiments of the present specification, as shown in Figure 3 In some embodiments, based on the battery swap load of multiple battery swap stations, the number of battery swap demand vehicles of multiple roads, and the demand delay time corresponding to each battery swap station and the related road in multiple historical time periods, the optimization scheme of the battery swap station of the management area is determined, including: establishing a constraint condition, wherein the constraint condition at least includes the maximum number of battery swap positions of each battery swap station and the location constraint of the newly added battery swap station, for example, due to the limitation of site area and investment cost, a certain battery swap station can support a maximum of 20 battery swap positions, and the location constraint of the newly added battery swap station can include multiple locations that meet the construction conditions of the battery swap station after screening; The plurality of candidate battery swap station optimization schemes are generated, wherein the candidate battery swap station optimization scheme includes the battery swap capacity of each battery swap station, the position and battery swap capacity of each newly added battery swap station, for example, the candidate schemes are generated by enumeration or random sampling, for each existing battery swap station, a plurality of battery swap capacities are randomly allocated according to the maximum number of battery swap capacities of the battery swap station, at least one position is sampled from a plurality of positions as the position of the newly added battery swap station, and a plurality of battery swap capacities are randomly allocated according to the maximum number of battery swap capacities of the newly added battery swap station, and the newly added battery swap station is merged with the nearest existing battery swap station into one battery swap station. Based on the battery swap load of the plurality of battery swap stations in a plurality of historical time periods, the number of battery swap demand vehicles of the plurality of roads, and the demand delay time corresponding to each battery swap station and the related road, a plurality of battery swap demand scenarios are established. According to the plurality of candidate battery swap station optimization schemes and the plurality of battery swap demand scenarios, the battery swap station optimization scheme of the management area is determined.

[0041] Specifically, the battery swap load of the plurality of battery swap stations in a plurality of historical time periods and the number of battery swap demand vehicles of the plurality of roads can be sampled to determine a plurality of battery swap demand scenarios, wherein the battery swap demand scenario can include the real-time battery swap load of the plurality of battery swap stations and the number of battery swap demand vehicles of the related road of the battery swap station after being forward shifted according to the demand delay time corresponding to the battery swap station and the related road.

[0042] For each battery swap demand scenario, the demand battery swap capacity of each battery swap station corresponding to the battery swap demand scenario is calculated according to the number of battery swap demand vehicles of the related road of the battery swap station, the real-time battery swap load of the battery swap station, and the number of completed battery swaps corresponding to the demand delay time, wherein the number of completed battery swaps corresponding to the demand delay time can be determined according to the time required to complete one battery swap, the demand delay time, and the battery swap capacity of the battery swap station.

[0043] For example, the demand battery swap capacity of the battery swap station corresponding to the battery swap demand scenario can be calculated according to the following formula: wherein, is the demand battery swap capacity of the i-th battery swap station corresponding to the battery swap demand scenario, is the real-time battery swap load of the i-th battery swap station corresponding to the battery swap demand scenario, is the number of battery swap demand vehicles of the j-th related road of the i-th battery swap station, is the total number of related roads of the i-th battery swap station, is the maximum value of the demand delay time corresponding to the i-th battery swap station and all related roads, is the time required to complete one battery swap, is the battery swap capacity of the i-th battery swap station, is the rounding operation.

[0044] For each candidate battery swap station optimization scheme, the absolute value of the difference between the battery swap capacity of each battery swap station and the required battery swap capacity is calculated, the absolute values of the differences between the battery swap capacities of each battery swap station and the required battery swap capacities are summed to obtain a required difference sum, and the optimization value of the candidate battery swap station optimization scheme is calculated. The larger the required difference sum, the smaller the optimization value of the candidate battery swap station optimization scheme.

[0045] The candidate battery swap station optimization scheme with the largest optimization value is taken as the battery swap station optimization scheme of the management area.

[0046] It can be understood that by integrating historical battery swap load, road battery swap demand vehicle quantity and demand delay time, a dynamic model covering "time-space-demand" three dimensions is constructed, accurately capturing battery swap demand fluctuations in different time periods and different areas, and avoiding resource mismatch caused by traditional static planning. The road battery swap demand vehicle quantity is shifted forward according to the demand delay time to simulate the time distribution of vehicles actually arriving at the battery swap station, solving the spatio-temporal misalignment problem of "demand occurrence" and "demand satisfaction". The maximum battery swap capacity (such as 20) and the site location screening conditions (such as site area, investment cost, construction compliance) of the new station are determined to ensure that the scheme meets the physical limitations and policy requirements, reducing the risk of later implementation. With the goal of minimizing the required difference sum, the battery swap capacity configuration and construction cost are indirectly balanced to avoid resource idling caused by excessive investment and improve the return on investment.

[0047] S140, based on the battery swap station optimization scheme of the management area, optimizing multiple battery swap stations in the management area.

[0048] Specifically, the battery swap capacity of the existing battery swap station in the management area can be optimized according to the battery swap station optimization scheme of the management area, and / or a new battery swap station can be added.

[0049] S150, based on the relevant roads of each battery swap station, obtaining real-time battery swap influence data of the management area.

[0050] The real-time battery swap influence data of the management area includes the real-time quantity of battery swap demand vehicles of multiple roads.

[0051] S160, the cloud platform predicts the future battery swap load of multiple battery swap stations based on the real-time battery swap load data of multiple battery swap stations and the real-time battery swap influence data of the management area, and delivers to the edge computing device.

[0052] Specifically, it includes: For each battery swap station, based on the real-time battery swap load data of the battery swap station, the real-time quantity of battery swap demand vehicles of the relevant road, and the demand delay time corresponding to the battery swap station and the relevant road, the future battery swap load of the battery swap station is predicted.

[0053] In some embodiments, based on the real-time battery swap load data of the battery swap station, the real-time number of battery swap demand vehicles of the related road, and the demand delay time corresponding to the battery swap station and the related road, the future battery swap load of the battery swap station is predicted, including: Based on the battery swap load of the battery swap station in multiple historical time periods, the number of battery swap demand vehicles of each related road, and the demand delay time corresponding to the battery swap station and each related road, a load prediction model corresponding to the battery swap station is established; Based on the real-time battery swap load data of the battery swap station, the real-time number of battery swap demand vehicles of the related road of the battery swap station, and the demand delay time corresponding to the battery swap station and the related road, the future battery swap load of the battery swap station is predicted through the load prediction model corresponding to the battery swap station.

[0054] Specifically, the load prediction model can be a long short-term memory network model, which captures the long-term dependence relationship between the load and the road demand. The input of the load prediction model can be the real-time battery swap load data of the battery swap station, the real-time number of battery swap demand vehicles of the related road of the battery swap station, and the demand delay time corresponding to the battery swap station and the related road. The output of the load prediction model can be the future battery swap load. Based on the battery swap load of the battery swap station in multiple historical time periods, the number of battery swap demand vehicles of each related road, and the demand delay time corresponding to the battery swap station and each related road, multiple training samples are generated. The training samples are divided into a training set (such as the first 80% of historical data), a validation set (the middle 10%), and a test set (the last 10%). The mean absolute error or root mean square error is used to optimize the load prediction model.

[0055] The load prediction model can use the following rolling prediction mechanism: Single-step prediction: taking the current time t as the starting point, the load at t+Δt is predicted; Multi-step rolling: taking the prediction result at t+Δt as input, recursively predicting t+2Δt, t+3Δt,..., forming a load curve for the next K steps.

[0056] S170, the cloud platform receives a battery swap request initiated by a user and allocates an edge computing device.

[0057] For example, an edge computing device with the smallest real-time computing power load can be determined, and the battery swap request initiated by the user can be allocated to the edge computing device.

[0058] S180, the edge computing device generates an optimal battery swap strategy corresponding to the battery swap request based on the future battery swap load of multiple battery swap stations.

[0059] Specifically, it includes: A multi-index strategy evaluation function is established, wherein the multi-index strategy evaluation function is related to at least a battery swap station load balancing index and a battery swap time cost index; The optimal battery swap strategy corresponding to the battery swap request is generated based on future battery swap loads of the plurality of battery swap stations and a multi-index strategy evaluation function.

[0060] Specifically, for each battery swap station, a variance of the plurality of battery swap loads after adding 1 to the load of the battery swap station is calculated, a score of a battery swap station load balancing index is calculated according to the variance of the plurality of battery swap loads, the smaller the variance of the plurality of battery swap loads, the higher the score of the battery swap station load balancing index, and a driving time required for a shortest driving time path between the vehicle corresponding to the battery swap request and the battery swap station is determined, a score of a battery swap time cost index is calculated, wherein the longer the driving time, the smaller the score of the battery swap time cost index, and a weighted sum of the score of the battery swap station load balancing index and the score of the battery swap time cost index is obtained to obtain a multi-index strategy evaluation function value corresponding to the battery swap station.

[0061] The battery swap station with the maximum multi-index strategy evaluation function value is taken as a target battery swap station, and the optimal battery swap strategy corresponding to the battery swap request is generated, wherein the optimal battery swap strategy corresponding to the battery swap request includes the target battery swap station and the shortest driving time path between the vehicle corresponding to the battery swap request and the battery swap station.

[0062] Figure 3 is a schematic diagram of a vehicle battery swap management system applying cloud-edge collaboration technology according to some embodiments of the present specification, as Figure 3 The vehicle battery swap management system applying cloud-edge collaboration technology can include a data acquisition module, a correlation analysis module, a battery swap station optimization module, and a cloud-edge collaboration module.

[0063] The data acquisition module is configured to acquire historical battery swap demand data of a management area. The correlation analysis module is configured to determine a relevant road for each battery swap station based on the historical battery swap demand data of the management area. The battery swap station optimization module is configured to determine a battery swap station optimization scheme for the management area based on the historical battery swap demand data of the management area and the relevant road for each battery swap station, and optimize a plurality of battery swap stations in the management area based on the battery swap station optimization scheme for the management area. The cloud-edge collaboration module is configured to acquire real-time battery swap influence data of the management area based on the relevant road for each battery swap station, predict future battery swap loads of the plurality of battery swap stations based on real-time battery swap load data of the plurality of battery swap stations and the real-time battery swap influence data of the management area, and deliver to an edge computing device, receive a battery swap request initiated by a user, and allocate the edge computing device, and the edge computing device generates an optimal battery swap strategy corresponding to the battery swap request based on the future battery swap loads of the plurality of battery swap stations.

[0064] The vehicle battery swap management system applying cloud-edge collaboration technology can be used to execute the vehicle battery swap management method applying cloud-edge collaboration technology, which will not be described here.

[0065] Finally, it should be understood that the embodiments described herein are only given by way of example and that other modifications can occur to persons skilled in the art. Therefore, the scope of the present description is not intended to be limited to the embodiments described herein but is only limited by the claims.

Claims

1. A vehicle battery swap management method using cloud edge collaboration technology, characterized in that, The method comprises the following steps: obtaining historical battery swap demand data of a management area; determining the relevant roads of each battery swap station based on the historical battery swap demand data of the management area; determining a battery swap station optimization scheme of the management area based on the historical battery swap demand data of the management area and the relevant roads of each battery swap station; optimizing a plurality of battery swap stations in the management area based on the battery swap station optimization scheme of the management area; obtaining real-time battery swap influence data of the management area based on the relevant roads of each battery swap station; the cloud platform predicts the future battery swap load of the plurality of battery swap stations based on the real-time battery swap load data of the plurality of battery swap stations and the real-time battery swap influence data of the management area, and delivers the prediction result to the edge computing device; the cloud platform receives a battery swap request initiated by a user and allocates the edge computing device; the edge computing device generates an optimal battery swap strategy corresponding to the battery swap request based on the future battery swap load of the plurality of battery swap stations. 2.The vehicle battery swapping management method using cloud edge collaboration technology according to claim 1, wherein, The historical battery swap demand data of the management area comprises the battery swap load of a plurality of battery swap stations and the number of battery swap demand vehicles of a plurality of roads in a plurality of historical time periods; determining the relevant roads of each battery swap station based on the historical battery swap demand data of the management area comprises: for each battery swap station and each road, calculating a first correlation coefficient between the battery swap station and the road based on the battery swap load of the battery swap station and the number of battery swap demand vehicles of the road in a plurality of historical time periods; determining the relevant roads of each battery swap station based on the first correlation coefficient between each battery swap station and each road. 3.The vehicle battery swapping management method using cloud-edge collaboration technology according to claim 2, wherein, determining a battery swap station optimization scheme of the management area based on the historical battery swap demand data of the management area and the relevant roads of each battery swap station comprises: for each relevant road of the battery swap station, determining a demand delay time corresponding to the battery swap station and the relevant road based on the battery swap load of the battery swap station and the number of battery swap demand vehicles of the relevant road in a plurality of historical time periods; determining a battery swap station optimization scheme of the management area based on the battery swap load of a plurality of battery swap stations, the number of battery swap demand vehicles of a plurality of roads, and the demand delay time corresponding to each battery swap station and the relevant road in a plurality of historical time periods. 4.The vehicle battery swapping management method using cloud-edge collaboration technology according to claim 3, wherein, determining a demand delay time corresponding to the battery swap station and the relevant road based on the battery swap load of the battery swap station and the number of battery swap demand vehicles of the relevant road in a plurality of historical time periods comprises: determining a plurality of candidate demand delay times based on the battery swap load of the battery swap station and the number of battery swap demand vehicles of the relevant road in a plurality of historical time periods; for each candidate demand delay time, determining a number sequence of battery swap demand vehicles corresponding to the candidate demand delay time based on the candidate demand delay time and the number of battery swap demand vehicles of the relevant road in a plurality of historical time periods, calculating a second correlation coefficient between the battery swap station and the relevant road corresponding to the candidate demand delay time based on the battery swap load of the battery swap station and the number sequence of battery swap demand vehicles corresponding to the candidate demand delay time of the relevant road in a plurality of historical time periods; determining a demand delay time corresponding to the battery swap station and the relevant road based on the second correlation coefficient between the battery swap station and the relevant road corresponding to each candidate demand delay time. 5.The vehicle battery swapping management method using cloud-edge collaboration technology according to claim 4, wherein, determining a plurality of candidate demand delay times based on the battery swap load of the battery swap station and the number of battery swap demand vehicles of the relevant road in a plurality of historical time periods comprises: The battery swap load of the battery swap station in multiple historical time periods is decomposed by variational mode decomposition, and key load characteristics of the battery swap station are extracted from the variational mode decomposition result; The number of battery swap demand vehicles of the related road in multiple historical time periods is decomposed by variational mode decomposition, and key number characteristics of the related road are extracted from the variational mode decomposition result; Based on the key load characteristics of the battery swap station and the key number characteristics of the related road, a plurality of candidate demand delay times are determined. 6.The vehicle battery swapping management method using cloud-edge collaboration technology according to claim 3, wherein, Based on the battery swap load of the plurality of battery swap stations, the number of battery swap demand vehicles of the plurality of roads, and the demand delay time corresponding to each battery swap station and the related road in multiple historical time periods, a battery swap station optimization scheme of the management area is determined, including: establishing constraint conditions, wherein the constraint conditions at least include the maximum battery swap position number of each battery swap station and the position constraint of the newly added battery swap station; generating a plurality of candidate battery swap station optimization schemes, wherein the candidate battery swap station optimization scheme includes the battery swap position of each battery swap station, the position and battery swap position of each newly added battery swap station; Based on the battery swap load of the plurality of battery swap stations, the number of battery swap demand vehicles of the plurality of roads, and the demand delay time corresponding to each battery swap station and the related road in multiple historical time periods, a plurality of battery swap demand scenarios are established; According to the plurality of candidate battery swap station optimization schemes and the plurality of battery swap demand scenarios, the battery swap station optimization scheme of the management area is determined.

7. The vehicle battery swapping management method using cloud-edge collaboration technology according to any one of claims 3-6, characterized in that, The real-time battery swap influence data of the management area includes the real-time number of battery swap demand vehicles of the plurality of roads; The cloud platform predicts the future battery swap load of the plurality of battery swap stations based on the real-time battery swap load data of the plurality of battery swap stations and the real-time battery swap influence data of the management area, including: For each battery swap station, the future battery swap load of the battery swap station is predicted based on the real-time battery swap load data of the battery swap station, the real-time number of battery swap demand vehicles of the related road, and the demand delay time corresponding to the battery swap station and the related road. 8.The vehicle battery swapping management method using cloud-edge collaboration technology according to claim 7, wherein, Based on the real-time battery swap load data of the battery swap station, the real-time number of battery swap demand vehicles of the related road, and the demand delay time corresponding to the battery swap station and the related road, the future battery swap load of the battery swap station is predicted, including: Based on the battery swap load of the battery swap station, the number of battery swap demand vehicles of each related road, and the demand delay time corresponding to the battery swap station and each related road in multiple historical time periods, a load prediction model corresponding to the battery swap station is established; Through the load prediction model corresponding to the battery swap station, the future battery swap load of the battery swap station is predicted based on the real-time battery swap load data of the battery swap station, the real-time number of battery swap demand vehicles of the related road of the battery swap station, and the demand delay time corresponding to the battery swap station and the related road. 9.The vehicle battery swapping management method using cloud-edge collaboration technology according to any one of claims 1-6, wherein, The edge computing device generates an optimal battery swap strategy corresponding to the battery swap request based on the future battery swap load of the plurality of battery swap stations, including: establishing a multi-index strategy evaluation function, wherein the multi-index strategy evaluation function is at least related to a battery swap station load balancing index and a battery swap time cost index; Based on the future battery swap load of the plurality of battery swap stations and the multi-index strategy evaluation function, an optimal battery swap strategy corresponding to the battery swap request is generated. 10.A vehicle battery swap management system using cloud edge collaboration technology, characterized by The vehicle battery swap management method using the cloud-edge collaborative technology according to any one of claims 1-9, including: The data acquisition module is configured to acquire historical battery swap demand data of the management area; The correlation analysis module is configured to determine the relevant roads of each battery swap station based on the historical battery swap demand data of the management area; The battery swap station optimization module is configured to determine a battery swap station optimization scheme of the management area based on the historical battery swap demand data of the management area and the relevant roads of each battery swap station, and to optimize the plurality of battery swap stations of the management area based on the battery swap station optimization scheme of the management area; The cloud-edge collaboration module is configured to acquire real-time battery swap influence data of the management area based on the relevant roads of each battery swap station, to predict future battery swap loads of the plurality of battery swap stations based on real-time battery swap load data of the plurality of battery swap stations and the real-time battery swap influence data of the management area, and to deliver the prediction to an edge computing device, to receive a battery swap request initiated by a user and to allocate the edge computing device, and to generate an optimal battery swap strategy corresponding to the battery swap request based on the future battery swap loads of the plurality of battery swap stations.

Images