Battery life prediction method and system based on charging pile historical data

Abstract

The present application relates to the technical field of electric vehicle battery life prediction, and particularly relates to a battery life prediction method and system based on charging pile historical data. The battery life prediction method based on charging pile historical data comprises the following steps: step 1: obtaining electric vehicle charging data recorded on a charging pile and uploading the electric vehicle charging data to a cloud server for storage; step 2: calculating the remaining life of the electric vehicle battery according to the obtained electric vehicle charging data; step 3: predicting the future remaining life of the electric vehicle battery based on historical charging data when the electric vehicle is charging; and step 4: providing the user with the remaining life of the electric vehicle battery at the current charging time and the predicted future remaining life of the electric vehicle battery. The remaining life of the electric vehicle battery is clearly presented, and accurate data support is provided for the measures of the vehicle owner to prolong the life.

Description

Technical Field

[0001] The present invention relates to the technical field of electric vehicle battery life prediction, and in particular to a battery life prediction method and system based on charging pile historical data. Background Art

[0002] The battery life of electric vehicles (electric bicycles, electric cars) is one of the important indicators for measuring electric vehicle performance. The power sources of electric vehicles are currently generally lithium batteries, lead-acid batteries, etc. Since battery replacement is very expensive, for example, the cost of electric vehicle batteries is usually calculated in tens of thousands of yuan, this is undoubtedly a considerable expense and burden for users.

[0003] How to extend the service life of batteries in electric vehicles is a problem that many electric vehicle owners hope to solve. However, existing electric vehicles have difficulty in clearly presenting battery life information and cannot provide owners with reference to vehicle life information during use. As a result, the measures taken by owners to extend the service life cannot be supported by accurate data.

[0004] Some intelligent electric vehicles are equipped with smart sensors to monitor battery life, providing users with a theoretical basis for extending the battery lifespan. However, installing smart sensors not only increases the purchase cost, but also, because battery life is a dynamic and difficult variable to monitor, existing technical solutions using smart sensors to monitor battery lifespan can only provide a rough estimate and are unable to quantify the battery lifespan of electric vehicles, providing users with a more accurate understanding. Summary of the Invention

[0005] In order to solve the above problems, the technical solution adopted by the present invention is:

[0006] The present invention provides a battery life prediction method based on charging pile historical data, comprising:

[0007] Step 1: Obtain the electric vehicle charging data recorded on the charging pile and upload the electric vehicle charging data to the cloud server for storage;

[0008] Step 2: Calculate the remaining life of the electric vehicle battery based on the acquired electric vehicle charging data;

[0009] Step 3: Predict the remaining life of the electric vehicle battery based on the historical charging data of the electric vehicle;

[0010] Step 4: Provide the user with the remaining life of the electric vehicle battery at the time of current charging and the predicted remaining life of the electric vehicle battery in the future.

[0011] Furthermore, in step 1, the electric vehicle charging data include:

[0012] (1)

[0013] in, The first The data of electric vehicle charging, T is the charging time, The serial number of the electric vehicle. is the battery capacity information of the electric vehicle, Indicates the single charging time of the electric vehicle. The remaining power when charging the battery of the electric vehicle, S is the charging mode.

[0014] Furthermore, in step 2, calculating the remaining life of the electric vehicle battery includes:

[0015] Calculate the battery life of the electric vehicle for the i-th time:

[0016] (2)

[0017] Where Le is the lifespan value; m is the single charge capacity (battery capacity after charging minus the remaining capacity of the electric vehicle battery when charging); β is the charging rate compensation value, and the value of β is determined according to the charging mode; γ is the compensation value; and They are used to describe the changing trend and degree of the nonlinear model of battery life value, Take 0.01, Take 0.1;

[0018] Substitute equation (2) to calculate the remaining life of the electric vehicle battery:

[0019] (3)

[0020] Wherein, Lt is the remaining life; Le 标称 Le is the life value of the battery at its nominal capacity; 失效 It is the life value of the battery when it is at failure capacity.

[0021] Furthermore, in step 3, the remaining life of the electric vehicle battery is predicted based on the general autoregressive model. :

[0022] (7)

[0023] in, is the Green function of the model, It is a constant item in the system settings.

[0024] Furthermore, the Green function is used to eliminate the deviation in the historical data:

[0025] In formula (7) is the Green function of the model:

[0026] (8)

[0027] Substitute equation (8) to solve for the future time t ， Lifespan value:

[0028] (11);

[0029] in, is a constant item in the system settings, Represents the product variable.

[0030] Furthermore, in step 3, based on the customer's demand for prediction results of electric vehicle lifespan at future moments, electric vehicle charging data at different time intervals are selected to predict the future remaining lifespan of electric vehicle batteries.

[0031] Furthermore, in step 1, obtaining the electric vehicle charging data also includes supplementing the charging data of the electric vehicle charged with mains electricity.

[0032] The present invention also provides a battery life prediction system based on charging pile historical data, which is used to implement the above-mentioned battery life prediction method based on charging pile historical data. The system includes:

[0033] Electric vehicle charging data acquisition module;

[0034] Cloud servers;

[0035] Display terminal;

[0036] The electric vehicle charging data acquisition module collects the electric vehicle charging data and uploads the collected electric vehicle data to the cloud server;

[0037] The cloud server includes a data receiving module, a data storage module, a remaining life calculation module, a future remaining life prediction module and a data output module;

[0038] Wherein, the data receiving module receives the electric vehicle charging data uploaded by the electric vehicle charging data acquisition module, and transmits the electric vehicle charging data to the data storage module for storage;

[0039] The remaining life calculation module retrieves the electric vehicle charging data from the data storage module and calculates the remaining life of the electric vehicle battery at the corresponding moment;

[0040] The future remaining life prediction module predicts the future remaining life of the electric vehicle battery at a future moment according to the result of the remaining life calculation module;

[0041] The data output module outputs the remaining life of the electric vehicle battery at the current charging moment and the predicted remaining life of the electric vehicle battery at a future moment to the image display terminal;

[0042] The display terminal is used to display the remaining life of the electric vehicle battery at the current charging moment and the predicted remaining life of the electric vehicle battery at a future moment to the user.

[0043] Furthermore, the electric vehicle charging data acquisition module includes a charging pile charging data acquisition unit and a mains charging data input unit;

[0044] Wherein, the charging pile charging data acquisition unit acquires the electric vehicle charging data recorded on the charging pile;

[0045] The mains charging data supplement unit collects the charging data of the electric vehicle during mains charging.

[0046] Furthermore, the future remaining life prediction module further includes a prediction time selection module, which selects electric vehicle charging data at different time intervals to predict the future remaining life of the electric vehicle battery according to the customer's selection of the future electric vehicle life prediction;

[0047] The predicted time selection module is in communication with the display terminal, and the user selects and inputs the future time through the display terminal.

[0048] The beneficial effects of the present invention are embodied in:

[0049] The present invention uses the historical data recorded when the electric vehicle is charging on the charging pile to calculate the remaining life of the electric vehicle battery at the current charging moment, and predict the remaining life of the electric vehicle battery at future moments. It can clearly present the remaining life of the electric vehicle battery and provide accurate data support for the owner's measures to extend the life.

[0050] At the same time, the historical data of electric vehicles recorded in the charging pile history is used to calculate the remaining life of the electric vehicle battery. This does not require the storage space of the electric vehicle and the computing power of the chip, reduces the storage and computing burden of the vehicle, reduces the installation of smart sensors and other hardware measures on the electric vehicle, and can also reduce the cost for users when purchasing electric vehicles. BRIEF DESCRIPTION OF THE DRAWINGS

[0051] Figure 1 A flowchart of a battery life prediction method based on charging pile historical data provided by the present invention;

[0052] Figure 2 This is a system block diagram of a battery life prediction system based on charging pile historical data provided by the present invention;

[0053] Figure 3 This is a system block diagram of a battery life prediction system based on charging pile historical data provided by the present invention, which has a prediction time selection module. DETAILED DESCRIPTION

[0054] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments of the present invention.

[0055] Reference Figure 1 The present invention provides a battery life prediction method based on charging pile historical data, comprising:

[0056] Step 1: Obtain the electric vehicle charging data recorded on the charging pile and upload the electric vehicle charging data to the cloud server for storage;

[0057] Step 2: Calculate the remaining life of the electric vehicle battery based on the acquired electric vehicle charging data;

[0058] Step 3: Predict the remaining life of the electric vehicle battery based on the historical charging data of the electric vehicle;

[0059] Step 4: Provide the user with the remaining life of the electric vehicle battery at the time of current charging and the predicted remaining life of the electric vehicle battery in the future.

[0060] Electric vehicles use electricity as energy to drive the vehicle, reducing carbon emissions from vehicle use. Currently, the ownership of electric vehicles is gradually increasing. In order to meet the demand for fast charging of electric vehicles, electric vehicle owners still use charging piles as the charging source for electric vehicles when conditions are met. The charging piles can collect information about each charging of the electric vehicle. By saving this information in the cloud server, it serves as basic data for calculating the remaining life of the electric vehicle battery.

[0061] The remaining life of the electric vehicle battery is calculated based on the historical charging data left by each charging station of the electric vehicle, quantifying the remaining life of the electric vehicle battery. The remaining life of the electric vehicle battery is then predicted based on the historical charging data recorded by the charging station when the electric vehicle is charging.

[0062] Finally, accurate data on the remaining life of the electric vehicle battery during current charging and the predicted remaining life of the electric vehicle battery in the future are provided to the user, providing accurate data support for the owner's measures to extend the life of the electric vehicle battery.

[0063] Furthermore, in step 1, the electric vehicle charging data include:

[0064] (1)

[0065] in, The first The data of electric vehicle charging is stored in the order of time T. The ID of the charging vehicle. A charging vehicle has only one ID, which is assigned by the system or filled in by the user. The battery capacity information of the electric vehicle. The change of battery capacity information is the main data for calculating the remaining life of the electric vehicle. Indicates the single charging time of the electric vehicle. The remaining power when charging the electric vehicle's battery. S is the charging mode. The charging pile has fast charging and slow charging options. Different charging modes correspond to different charging times.

[0066] Because data recording requires memory resources, sufficient data can truly reflect battery life data, while too little data can lead to inaccurate analysis. Therefore, according to actual testing, this solution can reduce memory hardware resource usage by 10% to 15% compared to traditional solutions. Furthermore, this solution based on the aforementioned subordinate data can also improve battery life prediction accuracy by 5% to 8% compared to traditional solutions.

[0067] Furthermore, in step 2, calculating the remaining life of the electric vehicle battery includes:

[0068] Calculate the battery life of the electric vehicle for the i-th time:

[0069] (2)

[0070] Wherein, Le is the life value; m is the single charge capacity; β is the charging rate compensation value, the value of β is determined by the charging mode, β is 1 in fast charge mode, and β is 2 in fast charge mode; γ is the compensation value, which is generally 10; and are used to control the change trend and degree of the nonlinear model used to describe the battery life value, Take 0.01, Take 0.1;

[0071] Substitute equation (2) to calculate the remaining life of the electric vehicle battery:

[0072] (3)

[0073] Wherein, Lt is the remaining life; Le标称 Le is the life value of the battery at its nominal capacity; 失效 It is the life value of the battery when it is at failure capacity.

[0074] In formula (3), Le is the life value calculated from the charging data of the i-th electric vehicle. When calculating the life value of the battery at the nominal capacity and failure capacity, the default is no charging, that is, the charging power and time are both zero. The failure capacity of the battery is generally taken as 90% of the nominal capacity of the battery.

[0075] A time series is a data sequence that is sorted in chronological order, changes over time, and is interrelated. It meets the solution requirements for lithium battery life prediction in the invention, and at the same time has the advantages of a relatively stable and accurate solution, and has a certain degree of robustness.

[0076] Furthermore, in step 3, the remaining life of the electric vehicle battery is predicted based on the general autoregressive model. :

[0077] According to formula (2) and formula (3), the change in its historical remaining life data can be calculated as follows:

[0078] (4)

[0079] The lifespan of electric vehicle batteries and , which has the following relationship:

[0080] (5)

[0081] make , then: , which is a first-order nonhomogeneous difference equation, then:

[0082]

[0083] (6)

[0084] By analogy with the above formula, we can get:

[0085] (7)

[0086] in, is the coefficient of variation, It is a constant item in the system settings.

[0087] Furthermore, the Green function is used to eliminate the deviation in the historical data:

[0088] In formula (7) is the Green function of the model:

[0089] (8)

[0090] Introducing product variables , we have the following formula:

[0091] (9)

[0092] Then the target formula (5) can be expressed as:

[0093] (10)

[0094] Substitute equation (8) to solve for the future time t ， Lifespan value:

[0095] (11)

[0096] in, is a constant item in the system settings, Represents the product variable.

[0097] For electric vehicles, battery life is a relatively ambiguous concept. To eliminate bias in historical data and analyze the impact of interference in historical data on battery life prediction, this solution introduces the Green function to describe the effect of such data interference on battery life. This approach treats the historical charging data of electric vehicles as a first-order dynamic sequence and uses this data to predict battery life.

[0098] Furthermore, in step 3, based on the customer's demand for prediction results of electric vehicle lifespan at future moments, electric vehicle charging data at different time intervals are selected to predict the future remaining lifespan of electric vehicle batteries.

[0099] For example, the user selects a time interval for charging intervals, where n is a positive integer. When n=1, the electric vehicle's historical charging data is sequentially selected from (t), (t-1), (t-2),...; when n=2, the electric vehicle's historical charging data is sequentially selected from (t), (t-2), (t-4),...; and when n=2, the electric vehicle's historical charging data is sequentially selected from (t), (t-5), (t-10). The user selects n based on the desired remaining battery life of the electric vehicle to obtain the predicted remaining battery life. Furthermore, in step 1, obtaining electric vehicle charging data also includes supplementing charging data for electric vehicles charged with mains power.

[0100] Due to the uneven match between the allocation of charging pile resources and the number of electric vehicles, not all users can use charging piles to charge their electric vehicles when driving them. When electric vehicles need to be temporarily charged in some places where there are no charging piles, they are generally charged with AC power. The charging piles cannot directly record the situation of AC power charging, which affects the prediction of the remaining life of the electric vehicle battery in the future.

[0101] The charging data of electric vehicles charged with mains electricity is fed into the cloud server to improve the accuracy of prediction of the remaining life of electric vehicle batteries in the future.

[0102] It should be noted that when the charging data of an electric vehicle charged with mains electricity is added, the value of the charging rate compensation value β corresponding to the charging mode is changed accordingly according to the difference in actual charging time.

[0103] Reference Figure 2 Furthermore, the present invention also provides a battery life prediction system based on charging pile historical data, which is used to implement the above-mentioned battery life prediction method based on charging pile historical data. The system includes:

[0104] Electric vehicle charging data acquisition module;

[0105] Cloud servers;

[0106] Display terminal;

[0107] The electric vehicle charging data acquisition module collects the electric vehicle charging data and uploads the collected electric vehicle data to the cloud server;

[0108] The cloud server includes a data receiving module, a data storage module, a remaining life calculation module, a future remaining life prediction module and a data output module;

[0109] Wherein, the data receiving module receives the electric vehicle charging data uploaded by the electric vehicle charging data acquisition module, and transmits the electric vehicle charging data to the data storage module for storage;

[0110] The remaining life calculation module retrieves the electric vehicle charging data from the data storage module and calculates the remaining life of the electric vehicle battery at the corresponding moment;

[0111] The future remaining life prediction module predicts the future remaining life of the electric vehicle battery at a future moment according to the result of the remaining life calculation module;

[0112] The data output module outputs the remaining life of the electric vehicle battery at the current charging moment and the predicted remaining life of the electric vehicle battery at a future moment to the image display terminal;

[0113] The display terminal is used to display the remaining life of the electric vehicle battery at the current charging moment and the predicted remaining life of the electric vehicle battery at a future moment to the user.

[0114] The display terminal may include the display screen of the charging station or the customer's mobile phone terminal to display the remaining life of the electric vehicle battery and interact with user selections.

[0115] Reference Figure 2 , further, the electric vehicle charging data acquisition module includes a charging pile charging data acquisition unit and a mains charging data supplement unit;

[0116] Wherein, the charging pile charging data acquisition unit acquires the electric vehicle charging data recorded on the charging pile;

[0117] The mains charging data supplement unit collects the charging data of the electric vehicle during mains charging.

[0118] Due to the uneven match between the allocation of charging pile resources and the number of electric vehicles, not all users can use charging piles to charge their electric vehicles when driving them. When electric vehicles need to be temporarily charged in some places where there are no charging piles, they are generally charged with AC power. The charging piles cannot directly record the situation of AC power charging, which affects the prediction of the remaining life of the electric vehicle battery in the future.

[0119] The mains charging data replenishment unit collects the charging data of electric vehicles charged by the mains, stores all the charging times of the electric vehicles, and improves the accuracy of the prediction of the remaining life of the electric vehicle battery in the future.

[0120] It should be noted that when the charging data of an electric vehicle charged with mains electricity is added, the value of the charging rate compensation value β corresponding to the charging mode is changed accordingly according to the difference in actual charging time.

[0121] Reference Figure 3 Furthermore, the future remaining life prediction module further includes a prediction time selection module, which selects electric vehicle charging data at different time intervals to predict the future remaining life of the electric vehicle battery according to the customer's selection of the electric vehicle life prediction at the future time;

[0122] For example, if the user selects a charging interval of n, where n is a positive integer, then when n=1, the electric vehicle's historical charging data time series is (t), (t-1), (t-2),...; when n=2, the electric vehicle's historical charging data time series is (t), (t-2), (t-4),...; and when n=2, the electric vehicle's historical charging data time series is (t), (t-5), (t-10). The user selects n based on the desired remaining battery life of the electric vehicle to obtain the predicted remaining battery life.

[0123] The predicted time selection module is in communication with the display terminal, and the user selects the future time through the display terminal. Human-computer interaction is performed at the display terminal for the customer to select the required information.

[0124] In describing the embodiments of the present invention, it should be understood that "-" and "~" represent a range between two values, and the range includes the endpoints. For example, "AB" represents a range greater than or equal to A and less than or equal to B. "A~B" represents a range greater than or equal to A and less than or equal to B.

[0125] In describing the embodiments of the present invention, the term "and / or" is used herein to describe a relationship between associated objects, indicating that three possible relationships exist. For example, "A and / or B" can represent three situations: A exists alone, A and B exist simultaneously, and B exists alone. Furthermore, the character " / " is generally used herein to indicate that the associated objects are in an "or" relationship.

[0126] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A battery life prediction method based on charging pile historical data, characterized in that: include: Step 1: Obtain the electric vehicle charging data recorded on the charging pile and upload the electric vehicle charging data to the cloud server for storage; Step 2: Calculate the remaining life of the electric vehicle battery based on the acquired electric vehicle charging data; Step 3: Predict the remaining life of the electric vehicle battery based on the historical charging data of the electric vehicle; Step 4: Provide the user with the remaining life of the electric vehicle battery at the time of current charging and the predicted remaining life of the electric vehicle battery in the future; Among them, in step 1, electric vehicle charging data include: （1） in, The first The data of electric vehicle charging, T is the charging time, The serial number of the electric vehicle. is the battery capacity information of the electric vehicle, Indicates the single charging time of the electric vehicle. The remaining power when charging the battery of the electric vehicle, S is the charging mode; In step 2, the remaining life of the electric vehicle battery is calculated including: Calculate the battery life of the electric vehicle for the i-th time: （2） Wherein, Le is the life value; m is the single charge capacity; β is the charging rate compensation value, and the value of β is determined according to the charging mode; γ is the compensation value; and They are used to describe the changing trend and degree of the nonlinear model of battery life value, Take 0.01, Take 0.1; Substitute equation (2) to calculate the remaining life of the electric vehicle battery: （3） Wherein, Lt is the remaining life; Le 标称 Le is the life value of the battery at its nominal capacity; 失效 It is the life value of the battery when it is at failure capacity.

2. The battery life prediction method based on charging pile historical data according to claim 1 is characterized in that: In step 3, the remaining life of the electric vehicle battery is predicted based on the general autoregressive model. : （7） in, is the Green function of the model, It is a constant item in the system settings.

3. The battery life prediction method based on charging pile historical data according to claim 2 is characterized in that: Eliminate the deviation in historical charging data through Green’s function: In formula (7) is the Green function of the model: （8） Substitute equation (8) to solve for the future time t ， Lifespan value: （11）； in, is a constant item in the system settings, Represents the product variable.

4. The battery life prediction method based on charging pile historical data according to claim 1 is characterized in that: In step 3, based on the customer's demand for prediction results of electric vehicle lifespan at future moments, electric vehicle charging data at different time intervals are selected to predict the future remaining lifespan of electric vehicle batteries.

5. The battery life prediction method based on charging pile historical data according to claim 1 is characterized in that: In step 1, obtaining the electric vehicle charging data also includes adding the charging data of the electric vehicle charged with mains electricity.

6. A battery life prediction system based on charging pile historical data, characterized in that: A system for implementing the battery life prediction method based on charging pile historical data according to any one of claims 1 to 5 includes: Electric vehicle charging data acquisition module; Cloud servers; Display terminal; The electric vehicle charging data acquisition module collects the electric vehicle charging data and uploads the collected electric vehicle data to the cloud server; The cloud server includes a data receiving module, a data storage module, a remaining life calculation module, a future remaining life prediction module and a data output module; Wherein, the data receiving module receives the electric vehicle charging data uploaded by the electric vehicle charging data acquisition module, and transmits the electric vehicle charging data to the data storage module for storage; The remaining life calculation module retrieves the electric vehicle charging data from the data storage module and calculates the remaining life of the electric vehicle battery at the corresponding moment; The future remaining life prediction module predicts the future remaining life of the electric vehicle battery at a future moment according to the result of the remaining life calculation module; The data output module outputs the remaining life of the electric vehicle battery at the current charging moment and the predicted remaining life of the electric vehicle battery at a future moment to the image display terminal; The display terminal is used to display the remaining life of the electric vehicle battery at the current charging moment and the predicted remaining life of the electric vehicle battery at a future moment to the user.

7. The battery life prediction system based on charging pile historical data according to claim 6, characterized in that: The electric vehicle charging data acquisition module includes a charging pile charging data acquisition unit and a mains charging data input unit; Wherein, the charging pile charging data acquisition unit acquires the electric vehicle charging data recorded on the charging pile; The mains charging data supplement unit collects the charging data of the electric vehicle during mains charging.

8. The battery life prediction system based on charging pile historical data according to claim 6 or claim 7, characterized in that: The future remaining life prediction module further includes a prediction time selection module, which selects electric vehicle charging data at different time intervals to predict the future remaining life of the electric vehicle battery according to the customer's selection of the electric vehicle life prediction at the future time; The predicted time selection module is in communication with the display terminal, and the user selects and inputs the future time through the display terminal.

Images