A backup battery rental management system

Abstract

The present invention discloses a spare battery rental management system, which relates to the technical field of battery rental management. The present invention deploys a sensor network in a vehicle to monitor multi-dimensional health data such as voltage, current, temperature, etc. of the battery in real time, and transmits the data to a data processing center through an Internet of Things device. A battery health status prediction model is constructed based on a support vector machine (SVM). The health status and remaining service life (RUL) of the battery are predicted in real time through the trained SVM model. The dynamic prediction method based on actual data greatly improves the prediction accuracy, and can more accurately judge the maintenance needs of the battery, reducing excessive maintenance and insufficient maintenance. The maintenance cycle is dynamically adjusted according to the health status prediction results, and a personalized maintenance plan is formulated. The system will remind the customer on the vehicle center console when to go to a battery swap station for inspection, and formulate detailed maintenance content and measures according to the specific situation. The battery service life is extended and the maintenance efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of battery rental management, and in particular to a backup battery rental management system. Background Art

[0002] The electric vehicle battery leasing model is an innovative business model that aims to reduce the high cost of car purchases for consumers. By leasing batteries, consumers do not need to pay the full price of the battery, but instead pay the leasing fee on a monthly or per-use basis. However, this model requires the construction and maintenance of a large number of battery swap stations in order to provide timely battery replacement services for vehicles that lease batteries.

[0003] For example, the patent publication number CN114387728A discloses a battery rental method and a battery rental system, the battery rental method comprising: step S10: for identifying customer needs; step S20: for analyzing and processing information related to customer needs, and forming a battery rental customer model according to customer needs; step S30: for predicting customer trajectories according to customer needs and the battery rental customer model, and arranging battery rental equipment according to the predicted customer trajectory; step S40: the user initiates a battery rental request through a handheld terminal APP or IC card according to demand, and the background server controls the battery rental equipment according to the request information; step S50: the system is initialized, and the system detection module detects whether the battery rental module is in normal working condition. If it is in normal working condition, the system initialization is completed, and the battery rental equipment allows the user to take out the battery when the user initiates a rental request; according to the battery rental method of the present invention, user satisfaction can be improved and the intelligence level of battery rental can be improved.

[0004] As mentioned above, including the current battery swapping methods, battery swapping stations are indispensable. The construction and maintenance of battery swapping stations require a lot of money and resources. The current penetration rate of battery swapping stations is much lower than that of charging stations, which leads to the inconvenience of battery swapping for consumers under the leasing model. The main reasons for this problem are: a large amount of money needs to be invested in the initial construction, the payback period is long, and the company faces great financial pressure. At the same time, the repair and maintenance of the battery during the lease period are all borne by the company. The cost of warranty services will increase with the increase of leasing time, further increasing the burden on the company. The uncertainty of battery life and frequent maintenance needs increase operating costs and reduce the overall profit margin.

[0005] It can be seen that on the user side, due to the insufficient number of battery swap stations, consumers are unable to replace batteries in time when needed, resulting in travel inconvenience and affecting user satisfaction; on the business side, high infrastructure and maintenance costs make it difficult for companies to make profits, especially in the initial market promotion stage; insufficient coverage of battery swap stations and high cost investment limit the promotion and popularization of the battery leasing model, affecting the expansion of market share; therefore, there is an urgent need for a backup battery rental management system to solve such problems. Summary of the invention

[0006] In view of the deficiencies of the prior art, the present invention provides a backup battery rental management system to solve the problems of high cost and low coverage in the battery rental model in the prior art.

[0007] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0008] The present invention provides a backup battery rental management system, comprising:

[0009] The data acquisition module is deployed in the vehicle to monitor the battery's voltage, current, and temperature in real time.

[0010] The data acquisition module includes:

[0011] A sensor network is provided on the battery mounting frame, and a sensor interface is provided for connecting sensors. A plurality of sensors are provided on the vehicle and connected to the sensor interface to form a sensor network.

[0012] The data acquisition unit collects multi-dimensional health data of the battery in real time and transmits it to the data processing center, which is deployed in the vehicle's center console;

[0013] The health status prediction module builds a battery health status prediction model based on the support vector machine (SVM) machine learning algorithm and predicts the health status of the battery in the data processing center;

[0014] The health status prediction module includes:

[0015] The state prediction unit uses the trained SVM model to predict the health status of the battery in real time and outputs it on the vehicle's center console;

[0016] Personalized maintenance plan module, which develops personalized maintenance plans based on health status prediction results and reminds customers on the vehicle center console when to go to the battery swap station for inspection;

[0017] The personalized maintenance plan module includes:

[0018] The prediction result analysis unit analyzes the battery health status prediction results and determines when the battery needs maintenance or replacement;

[0019] A maintenance plan generation unit generates a personalized maintenance plan based on the prediction results;

[0020] The user reminder unit reminds customers on the vehicle center console when they need to go to the battery swap station for inspection;

[0021] In the battery swap station management module, battery swap station staff refer to the health status prediction results to perform battery maintenance and replacement in advance, reduce the probability of sudden failures, extend battery life, and reduce maintenance costs;

[0022] The battery swap station management module includes:

[0023] A maintenance plan receiving unit receives and processes the maintenance plan and health status prediction results from the vehicle;

[0024] The battery maintenance unit performs battery maintenance and testing according to the maintenance plan;

[0025] Energy replenishment module, which sets up additional functions at gas stations, battery swap stations, and charging stations, provides energy replenishment methods that combine battery swapping and charging, and improves the utilization rate of facilities;

[0026] Energy replenishment modules include:

[0027] Multifunctional stations: additional charging piles and battery swapping functions are deployed at gas stations, additional charging piles and refueling functions are deployed at battery swapping stations, and additional battery swapping functions are deployed at charging stations;

[0028] IoT data collection module: deploy IoT devices in gas stations, battery swap stations, and charging stations with battery swap functions to collect battery status and inventory data in real time;

[0029] The IoT data collection module includes:

[0030] The inventory management unit uses linear programming to optimize battery distribution and inventory management, and analyzes battery inventory data collected by IoT devices to predict future battery demand;

[0031] The distribution route optimization unit uses linear programming algorithms to optimize battery distribution and inventory management to ensure that users can obtain batteries immediately and guarantee the timeliness of battery supply.

[0032] The present invention is further configured such that the maintenance plan generating unit generates a personalized maintenance plan based on the prediction result in the following manner:

[0033] RUL

[0034] Obtain the predicted remaining service life RUL of the battery from the health status prediction module, expressed as yc, that is, the predicted remaining service life of the battery, expressed in the number of charge and discharge cycles;

[0035] Set the maintenance cycle threshold. The specific maintenance cycle is determined according to the health status of the battery. It is expressed as τ, that is, the maintenance cycle threshold, which indicates the number of charge and discharge cycles after which maintenance is required.

[0036] Dynamically adjust the maintenance cycle based on the battery health status prediction results:

[0037] Maintenance cycle mode: Among them, τ d is the short maintenance period when the health status is poor, τ c is the long maintenance period when the health status is good, τ lj To define critical thresholds for differentiating good from bad health status;

[0038] The present invention is further configured such that the maintenance plan generating unit generates a personalized maintenance plan based on the prediction result, and the step further comprises:

[0039] According to the battery health status and usage, formulate specific maintenance content, including inspection items and maintenance measures;

[0040] Inspection items include: voltage, current, temperature, internal resistance, and charge and discharge times;

[0041] Maintenance measures include: battery balancing, cleaning contact points, and changing coolant;

[0042] Mathematical model for setting maintenance content: maintenance content: M = {m1, m2, ...m i}, where M represents the maintenance project set, m i is the i-th maintenance item;

[0043] The maintenance items are weighted and selected according to the health status prediction results. The maintenance items are selected as follows: Among them, w i represents the item weight, θ i It represents the item selection threshold, 1 means the item is needed, 0 means it is not needed;

[0044] According to the maintenance content, estimate the maintenance cost, maintenance cost: where c i represents the cost of the ith maintenance item, m i Indicates whether to select the item;

[0045] Generate personalized maintenance plan, including maintenance cycle, specific content and estimated cost;

[0046] The present invention is further configured such that the inventory management unit is optimized using a linear programming model, specifically:

[0047] Collect inventory data, including:

[0048] Current inventory I(t), that is, the current battery inventory of each battery swap station;

[0049] Historical demand D(t), that is, the battery demand of each battery swap station in the past period of time;

[0050] Replenishment quantity P(t), which is the number of batteries at each replenishment;

[0051] Remove outliers and missing values, use the time series model ARIMA to forecast demand, and use historical demand data to predict future battery demand; in, is the predicted demand in k days, f is the time series prediction function, D(t), D(t-1), ..., D(tn) are the historical demand data;

[0052] The present invention is further configured that the inventory management unit adopts a linear programming model to optimize the method further comprising:

[0053] Construct a linear programming model and define the objectives and constraints of the linear programming model:

[0054] Objective function:

[0055] Minimize total cost: Among them, C ij is the delivery cost from warehouse i to battery swap station j, Q ij is the delivery volume;

[0056] Constraints:

[0057] Inventory balance constraint: Among them I j (t+1) is the inventory of battery swap station j at time (t+1), Q ij is the delivery volume from warehouse i to battery swap station j, is the predicted demand of battery swap station j in the next k days;

[0058] Inventory non-negative constraint: Among them I j (t+1) means that the inventory of battery swap station j at time t+1 must be non-negative;

[0059] Maximum inventory capacity constraint: Among them I j,max is the maximum inventory capacity of battery swap station j;

[0060] Use linear programming algorithm to solve the model and get the distribution plan;

[0061] The present invention also discloses a backup battery rental management method, using the above backup battery rental management system, comprising the following steps:

[0062] Step 1. Deploy a sensor network in the vehicle to monitor battery performance parameters, including multi-dimensional health data of battery voltage, current, and temperature;

[0063] Step 2. Build a health status prediction model based on support vector machine (SVM) to predict the health status of the battery;

[0064] Step 3. Based on the health status prediction results, a personalized maintenance plan is developed, and the customer is reminded on the vehicle center console of the time before which he or she needs to go to the battery swap station for inspection;

[0065] Step 4. The staff of the battery swap station refer to the health status prediction results to perform battery maintenance and battery replacement in advance, reduce the probability of sudden failures, extend battery life, and reduce maintenance costs;

[0066] Step 5. Additional functions are set up at gas stations, battery swap stations, and charging stations. That is, additional charging piles and battery swap functions are deployed at gas stations, charging piles and refueling functions are deployed at battery swap stations, and battery swap functions are deployed at charging stations; energy replenishment methods combining battery swapping and charging are provided to improve the utilization rate of charging facilities;

[0067] Step 6. Deploy IoT devices at gas stations, battery swap stations, and charging stations with battery swap functions to collect battery status and inventory data in real time;

[0068] Step 7. Use linear programming to optimize battery distribution and inventory management;

[0069] The present invention is further configured such that, in step 2, a battery health status prediction model is constructed based on a support vector machine (SVM) in the following manner:

[0070] Collect battery performance data, including the battery voltage V during each charge and discharge, the current I during each charge and discharge, the battery temperature T under different working conditions, the number of charge and discharge cycles c, and the internal resistance R of the battery over time;

[0071] Collect tag data, including remaining useful life, expressed in terms of number of cycles, of the battery's remaining life;

[0072] Remove missing values ​​and outliers, normalize each feature, and standardize data of different dimensions. Among them, x is the original data, and x′ is the normalized data;

[0073] Extract key features from raw data:

[0074] Voltage change rate: Temperature change rate: Here Δt represents the time interval;

[0075] Divide the data set and use RBF kernel radial basis function for nonlinear data mapping. Where γ is the parameter of the kernel function; x i and x jRepresents the feature vector of the i-th and j-th samples, each x i is a feature vector, including the normalized voltage, current, temperature, number of charge and discharge cycles, internal resistance, voltage change rate and temperature change rate characteristics, x i =[V′ i ,I′ i ,T′ i ,c' i ,R′ i ,ΔV′ i ,ΔT′ i ];

[0076] The present invention is further configured such that, in step 2, the method of constructing a battery health status prediction model based on a support vector machine SVM further includes:

[0077] Perform model training and use the cross-validation method to select the optimal hyperparameters C and γ, where C is the penalty parameter and γ controls the width of the RBF kernel function.

[0078] Use the training set data to train the SVM model. Where i,j are sample retrievals, α i and α j is the Lagrange multiplier, y i is the label, y j is the remaining service life of the battery, K(x i ,x j ) is the kernel function;

[0079] Use the test set to evaluate the performance of the model, analyze the confusion matrix of the prediction results, and evaluate the classification effect of the model;

[0080] Optimize the model's hyperparameters based on the evaluation results; select important features and remove redundant features based on the model performance;

[0081] The present invention is further configured such that in step 2, the health state of the battery is predicted based on the health state prediction model:

[0082] Collect the current battery voltage V, current I, temperature T, charge and discharge cycle number c, and internal resistance R, and normalize the data collected in real time;

[0083] Calculate the voltage change rate ΔV and the temperature change rate ΔT, and construct the feature vector: X = [V', I', T', c', R', ΔV', ΔT'];

[0084] Use the SVM model to make predictions, and input the feature vector X into the SVM model for prediction;

[0085] Interpret the prediction results and inform the user of the battery health status and remaining lifespan.

[0086] Compared with the prior art, the present invention has the following beneficial effects:

[0087] The present invention deploys a sensor network in the vehicle to monitor the multi-dimensional health data of the battery, such as voltage, current, temperature, etc., in real time, and transmits the data to the data processing center through the Internet of Things device. A battery health status prediction model is constructed based on the support vector machine (SVM). The health status and remaining service life (RUL) of the battery are predicted in real time through the trained SVM model. The dynamic prediction method based on actual data greatly improves the prediction accuracy, can more accurately judge the maintenance needs of the battery, and reduce the situation of over-maintenance and under-maintenance.

[0088] The present invention dynamically adjusts the maintenance cycle and formulates a personalized maintenance plan based on the health status prediction results; the system will remind customers on the vehicle center console when they need to go to the battery swap station for inspection, and formulate detailed maintenance content and measures according to the specific situation; the service life of the battery is extended and the maintenance efficiency is improved;

[0089] The present invention analyzes the battery inventory data collected by IoT devices, uses a time series model to predict future battery demand, and optimizes battery distribution and inventory management through linear programming, optimizes inventory and distribution strategies, and reduces operating costs;

[0090] The present invention deploys multifunctional facilities in gas stations, battery swap stations and charging stations to provide an energy supplement method combining battery swapping and charging; improves the utilization rate of facilities and the flexibility of services;

[0091] The present invention can provide real-time battery status information and maintenance suggestions. Users can directly obtain this information on the vehicle center console without having to make their own judgments. At the same time, through optimized inventory and delivery management, it ensures that users can replace batteries in time when needed, significantly improving user convenience and satisfaction;

[0092] The problem of high cost and low coverage in the battery leasing model in the prior art is solved. BRIEF DESCRIPTION OF THE DRAWINGS

[0093] Figure 1 This is a framework diagram of a backup battery rental management system of the present invention. DETAILED DESCRIPTION

[0094] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0095] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0096] The present invention is further described in detail below in conjunction with the accompanying drawings:

[0097] Example 1

[0098] This solution is aimed at the separate car purchase method, that is, the car and the battery are separated when the user purchases the car, and the battery is leased for use. Only the bare car without the battery is purchased, which can greatly reduce the cost of car purchase. However, the battery leasing market in this context has the problems described above, namely the high cost and low coverage problem in the battery leasing model.

[0099] See also Figure 1 The present invention provides a backup battery rental management system, comprising:

[0100] The data acquisition module is deployed in the vehicle to monitor the battery's voltage, current, and temperature in real time.

[0101] The data acquisition module includes:

[0102] A sensor network is provided on the battery mounting frame, and a sensor interface is provided for connecting sensors. A plurality of sensors are provided on the vehicle and connected to the sensor interface to form a sensor network.

[0103] The data acquisition unit collects multi-dimensional health data of the battery in real time and transmits it to the data processing center, which is deployed in the vehicle's center console;

[0104] The health status prediction module builds a battery health status prediction model based on the support vector machine (SVM) machine learning algorithm and predicts the health status of the battery in the data processing center;

[0105] The health status prediction module includes:

[0106] The state prediction unit uses the trained SVM model to predict the health status of the battery in real time and outputs it on the vehicle's center console;

[0107] Personalized maintenance plan module, which develops personalized maintenance plans based on health status prediction results and reminds customers on the vehicle center console when to go to the battery swap station for inspection;

[0108] The personalized maintenance plan module includes:

[0109] The prediction result analysis unit analyzes the battery health status prediction results and determines when the battery needs maintenance or replacement;

[0110] A maintenance plan generation unit generates a personalized maintenance plan based on the prediction results;

[0111] The user reminder unit reminds customers on the vehicle center console when they need to go to the battery swap station for inspection;

[0112] The maintenance plan generation unit generates a personalized maintenance plan based on the prediction results in the following way:

[0113] RUL

[0114] Obtain the predicted remaining service life RUL of the battery from the health status prediction module, expressed as yc, that is, the predicted remaining service life of the battery, expressed in the number of charge and discharge cycles;

[0115] Set the maintenance cycle threshold. The specific maintenance cycle is determined according to the health status of the battery. It is expressed as τ, that is, the maintenance cycle threshold, which indicates the number of charge and discharge cycles after which maintenance is required.

[0116] Dynamically adjust the maintenance cycle based on the battery health status prediction results:

[0117] Maintenance cycle mode: Among them, τ d is the short maintenance period when the health status is poor, τ c is the long maintenance period when the health status is good, τ lj To define critical thresholds for differentiating good from bad health status;

[0118] The maintenance plan generating unit generates a personalized maintenance plan based on the prediction result, and the step further includes:

[0119] According to the battery health status and usage, formulate specific maintenance content, including inspection items and maintenance measures;

[0120] Inspection items include: voltage, current, temperature, internal resistance, and charge and discharge times;

[0121] Maintenance measures include: battery balancing, cleaning contact points, and changing coolant;

[0122] Mathematical model for setting maintenance content: maintenance content: M = {m1, m2, ...m i}, where M represents the maintenance project set, m i is the i-th maintenance item;

[0123] The maintenance items are weighted and selected according to the health status prediction results. The maintenance items are selected as follows: Among them, w i represents the item weight, θ i It represents the item selection threshold, 1 means the item is needed, 0 means it is not needed;

[0124] According to the maintenance content, estimate the maintenance cost, maintenance cost: where c i represents the cost of the ith maintenance item, m i Indicates whether to select the item;

[0125] Generate personalized maintenance plan, including maintenance cycle, specific content and estimated cost;

[0126] In the battery swap station management module, battery swap station staff refer to the health status prediction results to perform battery maintenance and replacement in advance, reduce the probability of sudden failures, extend battery life, and reduce maintenance costs;

[0127] The battery swap station management module includes:

[0128] A maintenance plan receiving unit receives and processes the maintenance plan and health status prediction results from the vehicle;

[0129] The battery maintenance unit performs battery maintenance and testing according to the maintenance plan;

[0130] Energy replenishment module, which sets up additional functions at gas stations, battery swap stations, and charging stations, provides energy replenishment methods that combine battery swapping and charging, and improves the utilization rate of facilities;

[0131] Energy replenishment modules include:

[0132] Multifunctional stations: additional charging piles and battery swapping functions are deployed at gas stations, additional charging piles and refueling functions are deployed at battery swapping stations, and additional battery swapping functions are deployed at charging stations;

[0133] IoT data collection module: deploy IoT devices in gas stations, battery swap stations, and charging stations with battery swap functions to collect battery status and inventory data in real time;

[0134] The IoT data collection module includes:

[0135] The inventory management unit uses linear programming to optimize battery distribution and inventory management, and analyzes battery inventory data collected by IoT devices to predict future battery demand;

[0136] The distribution route optimization unit uses linear programming algorithms to optimize battery distribution and inventory management to ensure that users can obtain batteries immediately and ensure the timeliness of battery supply;

[0137] The inventory management unit is optimized using a linear programming model, specifically:

[0138] Collect inventory data, including:

[0139] Current inventory I(t), that is, the current battery inventory of each battery swap station;

[0140] Historical demand D(t), that is, the battery demand of each battery swap station in the past period of time;

[0141] Replenishment quantity P(t), which is the number of batteries at each replenishment;

[0142] Remove outliers and missing values, use the time series model ARIMA to forecast demand, and use historical demand data to predict future battery demand; in, is the predicted demand in k days, f is the time series prediction function, D(t), D(t-1), ..., D(tn) are the historical demand data;

[0143] Construct a linear programming model and define the objectives and constraints of the linear programming model:

[0144] Objective function:

[0145] Minimize total cost: Among them, C ij is the delivery cost from warehouse i to battery swap station j, Q ij is the delivery volume;

[0146] Constraints:

[0147] Inventory balance constraint: Among them I j (t+1) is the inventory of battery swap station j at time (t+1), Q ij is the delivery volume from warehouse i to battery swap station j, is the predicted demand of battery swap station j in the next k days;

[0148] Inventory non-negative constraint: Among them I j (t+1) means that the inventory of battery swap station j at time t+1 must be non-negative;

[0149] Maximum inventory capacity constraint: Among them I j,max is the maximum inventory capacity of battery swap station j;

[0150] Use linear programming algorithm to solve the model and get the distribution plan;

[0151] The present invention also discloses a backup battery rental management method, comprising:

[0152] Step 1. Deploy a sensor network in the vehicle to monitor battery performance parameters, including multi-dimensional health data of battery voltage, current, and temperature;

[0153] Step 2. Build a health status prediction model based on support vector machine (SVM) to predict the health status of the battery;

[0154] The method of building a battery health status prediction model based on support vector machine SVM is as follows:

[0155] Collect battery performance data, including the battery voltage V during each charge and discharge, the current I during each charge and discharge, the battery temperature T under different working conditions, the number of charge and discharge cycles c, and the internal resistance R of the battery over time;

[0156] Collect tag data, including remaining useful life, expressed in terms of number of cycles, of the battery's remaining life;

[0157] Remove missing values ​​and outliers, normalize each feature, and standardize data of different dimensions. Among them, x is the original data, and x is the normalized data;

[0158] Extract key features from raw data:

[0159] Voltage change rate: Temperature change rate: Here Δt represents the time interval;

[0160] Divide the data set and use RBF kernel radial basis function for nonlinear data mapping, K(x i ,x j ) = exp(-γ||x i -x j || 2 ), where γ is the parameter of the kernel function; x i and xj Represents the feature vector of the i-th and j-th samples, each x i is a feature vector, including the normalized voltage, current, temperature, number of charge and discharge cycles, internal resistance, voltage change rate and temperature change rate characteristics, x i =[V′ i ,I′ i ,T′ i ,c' i ,R′ i ,ΔV′ i ,ΔT′ i ];

[0161] Perform model training and use the cross-validation method to select the optimal hyperparameters C and γ, where C is the penalty parameter and γ controls the width of the RBF kernel function.

[0162] Use the training set data to train the SVM model. Where i,j are sample retrievals, α i and α j is the Lagrange multiplier, y i is the label, y j is the remaining service life of the battery, K(x i ,x j ) is the kernel function;

[0163] Use the test set to evaluate the performance of the model, analyze the confusion matrix of the prediction results, and evaluate the classification effect of the model;

[0164] Optimize the model's hyperparameters based on the evaluation results; select important features and remove redundant features based on the model performance;

[0165] Collect the current battery voltage V, current I, temperature T, charge and discharge cycle number c, and internal resistance R, and normalize the data collected in real time;

[0166] Calculate the voltage change rate ΔV and the temperature change rate ΔT, and construct the feature vector: X = [V', I', T', c', R', ΔV', ΔT'];

[0167] Use the SVM model to make predictions, and input the feature vector X into the SVM model for prediction;

[0168] Interpret the prediction results and inform the user about the battery health status and remaining life;

[0169] Step 3. Based on the health status prediction results, a personalized maintenance plan is developed, and the customer is reminded on the vehicle center console of the time before which he or she needs to go to the battery swap station for inspection;

[0170] Step 4. The staff of the battery swap station refer to the health status prediction results to perform battery maintenance and battery replacement in advance, reduce the probability of sudden failures, extend battery life, and reduce maintenance costs;

[0171] Step 5. Additional functions are set up at gas stations, battery swap stations, and charging stations. That is, additional charging piles and battery swap functions are deployed at gas stations, charging piles and refueling functions are deployed at battery swap stations, and battery swap functions are deployed at charging stations; energy replenishment methods combining battery swapping and charging are provided to improve the utilization rate of charging facilities;

[0172] Step 6. Deploy IoT devices at gas stations, battery swap stations, and charging stations with battery swap functions to collect battery status and inventory data in real time;

[0173] Step 7. Use linear programming to optimize battery distribution and inventory management.

[0174] The present invention aims to solve the problems of high cost, low coverage and poor user experience in traditional systems. A sensor network is deployed in the vehicle to monitor the multi-dimensional health data of the battery, such as voltage, current, temperature, etc. in real time. The sensor forms a network through the interface installed on the battery and transmits the data to the data processing center located in the vehicle center console. A battery health status prediction model is constructed based on the support vector machine (SVM) machine learning algorithm to predict the health status of the battery in real time. Through the trained SVM model, the status prediction unit performs real-time prediction in the data processing center and outputs the prediction result on the center console.

[0175] Based on the health status prediction results, the personalized maintenance plan module generates a maintenance plan and reminds the user; first, the prediction result analysis unit determines the maintenance or replacement time of the battery, and then the maintenance plan generation unit dynamically adjusts the maintenance cycle according to the predicted remaining service life RUL of the battery, and formulates specific maintenance content, including inspection items such as voltage, current, temperature, internal resistance, and number of charge and discharge times, as well as maintenance measures such as battery balancing, cleaning contact points, and replacing coolant; finally, the user reminder unit reminds the customer on the vehicle center console when to go to the battery swap station for inspection;

[0176] The battery swap station management module receives and processes maintenance plans and health status prediction results to perform battery maintenance and replacement in advance, reduce the probability of sudden failures, extend battery life, and reduce maintenance costs; the energy replenishment module deploys additional functions in gas stations, battery swap stations, and charging stations to provide energy replenishment methods that combine battery swapping and charging to improve facility utilization; IoT devices are deployed at the site to collect battery status and inventory data in real time; the inventory management unit analyzes data and uses linear programming to optimize battery distribution and inventory management;

[0177] Specifically, the inventory management unit collects current inventory, historical demand, and replenishment data, removes outliers and missing values, and uses a time series model (ARIMA) to predict future battery demand. Based on the prediction results, a linear programming model is constructed to minimize distribution costs, and constraints such as inventory balance, non-negative inventory, and maximum inventory capacity are set. By solving the linear programming model, battery distribution and inventory management are optimized to ensure that users can obtain batteries in a timely manner.

[0178] The above contents are only for explaining the technical idea of ​​the present invention and cannot be used to limit the protection scope of the present invention. Any changes made on the basis of the technical solution in accordance with the technical idea proposed by the present invention shall fall within the protection scope of the claims of the present invention.

Claims

1. A backup battery rental management system, characterized in that: include: The data acquisition module is deployed in the vehicle to monitor the battery's voltage, current, and temperature in real time. The data acquisition module includes: A sensor network is provided on the battery mounting frame, and a sensor interface is provided for connecting sensors. A plurality of sensors are provided on the vehicle and connected to the sensor interface to form a sensor network. The data acquisition unit collects multi-dimensional health data of the battery in real time and transmits it to the data processing center, which is deployed in the vehicle's center console; The health status prediction module builds a battery health status prediction model based on the support vector machine (SVM) machine learning algorithm and predicts the health status of the battery in the data processing center; The health status prediction module includes: The state prediction unit uses the trained SVM model to predict the health status of the battery in real time and outputs it on the vehicle's center console; Personalized maintenance plan module, which develops personalized maintenance plans based on health status prediction results and reminds customers on the vehicle center console when to go to the battery swap station for inspection; The personalized maintenance plan module includes: The prediction result analysis unit analyzes the battery health status prediction results and determines when the battery needs maintenance or replacement; A maintenance plan generation unit generates a personalized maintenance plan based on the prediction results; The user reminder unit reminds customers on the vehicle center console when they need to go to the battery swap station for inspection; The battery swap station management module allows station staff to refer to the health status prediction results and perform battery maintenance and replacement in advance; The battery swap station management module includes: A maintenance plan receiving unit receives and processes the maintenance plan and health status prediction results from the vehicle; The battery maintenance unit performs battery maintenance and testing according to the maintenance plan; Energy replenishment module, which sets up additional functions at gas stations, battery swap stations, and charging stations to provide energy replenishment methods that combine battery swapping and charging; Energy replenishment modules include: Multifunctional stations: additional charging piles and battery swapping functions are deployed at gas stations, additional charging piles and refueling functions are deployed at battery swapping stations, and additional battery swapping functions are deployed at charging stations; IoT data collection module: deploy IoT devices in gas stations, battery swap stations, and charging stations with battery swap functions to collect battery status and inventory data in real time; The IoT data collection module includes: The inventory management unit uses linear programming to optimize battery distribution and inventory management, and analyzes battery inventory data collected by IoT devices to predict future battery demand; The distribution route optimization unit uses linear programming algorithms to optimize battery distribution and inventory management.

2. A backup battery rental management system according to claim 1, characterized in that: The maintenance plan generation unit generates a personalized maintenance plan based on the prediction results in the following way: Get the predicted remaining service life RUL of the battery from the health status prediction module, expressed as RUL yc , which is the predicted remaining battery life, expressed in terms of the number of charge and discharge cycles; Set the maintenance cycle threshold. The specific maintenance cycle is determined according to the health status of the battery. It is expressed as τ, that is, the maintenance cycle threshold, which indicates the number of charge and discharge cycles after which maintenance is required. Dynamically adjust the maintenance cycle based on the battery health status prediction results: Maintenance cycle mode: Among them, τ d is the short maintenance period when the health status is poor, τ c is the long maintenance period when the health status is good, τ lj The critical threshold for distinguishing good from bad health status.

3. A backup battery rental management system according to claim 2, characterized in that: The maintenance plan generating unit generates a personalized maintenance plan based on the prediction result, and the step further includes: Formulate specific maintenance contents, including inspection items and maintenance measures, based on the battery health status and usage; Inspection items include: voltage, current, temperature, internal resistance, and charge and discharge times; Maintenance measures include: battery balancing, cleaning contact points, and changing coolant; Mathematical model for setting maintenance content: maintenance content: M = {m1, m2, ...m i }, where M represents the maintenance project set, m i is the i-th maintenance item; The maintenance items are weighted and selected according to the health status prediction results. The maintenance items are selected as follows: Among them, w i represents the item weight, θ i It represents the item selection threshold, 1 means the item is needed, 0 means it is not needed; According to the maintenance content, estimate the maintenance cost, maintenance cost: where c i represents the cost of the ith maintenance item, m i Indicates whether to select the item; Generate a personalized maintenance plan including maintenance cycles, specific content and estimated costs.

4. A backup battery rental management system according to claim 3, characterized in that: The inventory management unit is optimized using a linear programming model, specifically: Collect inventory data, including: Current inventory I(t), that is, the current battery inventory of each battery swap station; Historical demand D(t), that is, the battery demand of each battery swap station in the past period of time; Replenishment quantity P(t), which is the number of batteries at each replenishment; Remove outliers and missing values, use the time series model ARIMA to forecast demand, and use historical demand data to predict future battery demand; in, is the predicted demand in k days’ time, f is the time series prediction function, and D(t), D(t-1), ..., D(tn) are the historical demand data.

5. A backup battery rental management system according to claim 4, characterized in that: The optimization methods of the inventory management unit using the linear programming model also include: Construct a linear programming model and define the objectives and constraints of the linear programming model: Objective function: Minimize total cost: Among them, C ij is the delivery cost from warehouse i to battery swap station j, Q ij is the delivery volume; Constraints: Inventory balance constraint: Among them I j (t+1) is the inventory of battery swap station j at time (t+1), Q ij is the delivery volume from warehouse i to battery swap station j, is the predicted demand of battery swap station j in the next k days; Inventory non-negative constraint: Among them I j (t+1) Indicates that the inventory of battery swap station j at time t+1 must be non-negative; Maximum inventory capacity constraint: Among them I j,max is the maximum inventory capacity of battery swap station j; The linear programming algorithm is used to solve the model and obtain the distribution plan.

6. A backup battery rental management method, characterized in that: A backup battery rental management system according to any one of claims 1 to 5 is used, comprising the following steps: Step 1. Deploy a sensor network in the vehicle to monitor battery performance parameters, including multi-dimensional health data of battery voltage, current, and temperature; Step 2. Build a health status prediction model based on support vector machine (SVM) to predict the health status of the battery; Step 3. Based on the health status prediction results, a personalized maintenance plan is developed, and the customer is reminded on the vehicle center console of the time before which he or she needs to go to the battery swap station for inspection; Step 4. The staff of the battery swap station refers to the health status prediction results to perform battery maintenance and battery replacement in advance; Step 5. Additional functions are set up at gas stations, battery swap stations, and charging stations. That is, additional charging piles and battery swap functions are deployed at gas stations, additional charging piles and refueling functions are deployed at battery swap stations, and additional battery swap functions are deployed at charging stations; Step 6. Deploy IoT devices at gas stations, battery swap stations, and charging stations with battery swap functions to collect battery status and inventory data in real time; Step 7. Use linear programming to optimize battery distribution and inventory management.

7. A backup battery rental management method according to claim 6, characterized in that: In step 2, the battery health status prediction model is constructed based on the support vector machine SVM as follows: Collect battery performance data, including the battery voltage V during each charge and discharge, the current I during each charge and discharge, the battery temperature T under different working conditions, the number of charge and discharge cycles c, and the internal resistance R of the battery over time; Collect tag data, including remaining useful life, expressed in terms of number of cycles, of the battery's remaining life; Remove missing values ​​and outliers, normalize each feature, and standardize data of different dimensions. Among them, x is the original data, and x' is the normalized data; Extract key features from raw data: Voltage change rate: Temperature change rate: Here Δt represents the time interval; Divide the data set and use RBF kernel radial basis function for nonlinear data mapping, K(x i ,x j ) = exp(-γ||x i -x j || 2 ), where γ is the parameter of the kernel function; x i and x j Represents the feature vector of the i-th and j-th samples, each x i is a feature vector, including the normalized voltage, current, temperature, number of charge and discharge cycles, internal resistance, voltage change rate and temperature change rate characteristics, x i =[V' i ,I' i ,T' i ,c' i ,R' i ,ΔV' i ,ΔT' i ].

8. A backup battery rental management method according to claim 7, characterized in that: In step 2, the method of building a battery health status prediction model based on a support vector machine (SVM) also includes: Perform model training and use the cross-validation method to select the optimal hyperparameters C and γ, where C is the penalty parameter and γ controls the width of the RBF kernel function. Use the training set data to train the SVM model. Where i,j are sample retrievals, α i and α j is the Lagrange multiplier, y i is the label, y j is the remaining service life of the battery, K(x i ,x j ) is the kernel function; Use the test set to evaluate the performance of the model, analyze the confusion matrix of the prediction results, and evaluate the classification effect of the model; Optimize the model's hyperparameters based on the evaluation results; select important features and remove redundant features based on the model performance.

9. A backup battery rental management method according to claim 8, characterized in that: In step 2, the health status of the battery is predicted based on the health status prediction model: Collect the current battery voltage V, current I, temperature T, charge and discharge cycle number c, and internal resistance R, and normalize the data collected in real time; Calculate the voltage change rate ΔV and the temperature change rate ΔT, and construct the feature vector: X = [V', I', T', c', R', ΔV', ΔT']; Use the SVM model to make predictions, and input the feature vector X into the SVM model for prediction; Interpret the prediction results and inform the user of the battery health status and remaining lifespan.

Images