Method and system for predicting the remaining life of a vehicle's starting battery

Abstract

This invention relates to a method and system for predicting the remaining lifespan of a vehicle's starter battery. The method includes: constructing a starter battery simulation model; for an actual vehicle with a complete starter battery lifespan and a target vehicle whose starter battery remaining lifespan is to be predicted, calculating historical battery current data from the starter battery simulation model based on historical data of engine speed and battery voltage; constructing a prediction model using machine learning based on historical data of the actual vehicle's engine start-stop count, cumulative battery operating time, battery voltage, and battery current, as well as the actual battery lifespan; and inputting the historical data of the target vehicle's engine start-stop count, cumulative battery operating time, battery voltage, and battery current into the prediction model to predict the remaining lifespan of the target vehicle's starter battery. This invention enables low-cost, high-accuracy, and convenient prediction of starter battery remaining lifespan, thereby achieving stable, safe, and efficient operation of engineering machinery.

Description

Technical Field

[0001] This invention relates to the field of vehicle starter batteries, and more particularly to a method and system for predicting the remaining life of a starter battery. Background Technology

[0002] Many construction machines, such as loaders or excavators (and for simplicity, these machines are conventionally referred to as "vehicles" in this article), are equipped with starting batteries, such as lead-acid batteries (or simply "batteries"), which are crucial components for starting the vehicle's engine. The starting battery connects to the alternator driven by the engine to receive charging, and also connects to the vehicle's starter motor to supply power for starting the engine and vehicle. Furthermore, the starting battery can also power in-vehicle electrical equipment such as lights, air conditioning, and controllers when the engine / alternator is not operating. A healthy and stable starting battery ensures that the vehicle can be started quickly and reliably, and maintains the stability and safety of vehicle operation and use.

[0003] However, starter batteries have a limited design lifespan, and different vehicle operating conditions can cause variations in their actual lifespan. If a starter battery suddenly fails due to a malfunction or the end of its lifespan, and timely repair or replacement is difficult, it can lead to temporary downtime and affect the overall vehicle operation. Monitoring the condition of the starter battery and predicting its remaining lifespan in advance would be crucial for the stable, safe, and efficient operation of the vehicle. Existing starter battery lifespan prediction methods are mostly based on battery operating parameters such as voltage, current, and internal resistance, but these parameters require specialized sensors installed on the vehicle for measurement or estimation. Therefore, how to predict starter battery lifespan in a low-cost, simple, and accurate manner remains a key technical challenge. Summary of the Invention

[0004] The purpose of this invention is to solve the above-mentioned problems and / or other defects existing in the prior art.

[0005] According to a first aspect of the present invention, a method for predicting the remaining life of a starting battery of a vehicle, the starting battery being charged by a generator driven by the vehicle's engine, the method comprising the steps of: constructing a starting battery simulation model capable of calculating battery current based on engine speed and battery voltage; for at least one actual vehicle having a complete starting battery lifespan, calculating historical data of its battery current from the starting battery simulation model based on historical data of its engine speed and battery voltage; constructing a prediction model for predicting the remaining life of the starting battery by means of machine learning, based on historical data of the actual vehicle's engine start-stop count, cumulative battery operating time, battery voltage, and the calculated historical data of its battery current, as well as the actual battery life; for a target vehicle, calculating historical data of its battery current from the starting battery simulation model based on its engine speed and battery voltage; and inputting the historical data of the target vehicle's engine start-stop count, cumulative battery operating time, battery voltage, and the calculated historical data of its battery current into the prediction model to predict the remaining life of the target vehicle's starting battery.

[0006] According to an exemplary embodiment, constructing the startup battery simulation model includes the following sub-steps: (a) establishing an initial simulation model involving at least a startup battery and a generator; (b) conducting an experimental run of the startup battery connected to the generator and acquiring experimental data of the battery current using sensors; and (c) correcting model parameters by comparing the difference between the battery current of the startup battery during the experimental run, which is calculated by the initial simulation model, and the corresponding experimental data of the acquired battery current, in order to reduce the difference to below a predetermined threshold.

[0007] According to one exemplary implementation, the sub-step (c) is performed by repeatedly training and tuning the initial simulation model in a machine learning manner.

[0008] According to one exemplary implementation, the prediction model is a machine learning model based on a time series prediction algorithm.

[0009] According to one exemplary implementation, the time series prediction algorithm includes one or more of the following: Long Short-Term Memory (LSTM) network algorithm, Random Forest, Support Vector Machine, and Backpropagation Neural Network.

[0010] A second aspect of the invention provides a computer-readable storage medium having a computer program stored thereon, the computer program including executable instructions that, when executed by a processor, implement a method according to any of the configurations described above.

[0011] A third aspect of the present invention provides an electronic device comprising: a processor; and a memory for storing executable instructions of the processor; wherein the processor is configured to execute the executable instructions to implement a method according to any of the above-described configurations.

[0012] A fourth aspect of the invention provides a system for predicting the remaining life of a vehicle's starter battery, the starter battery being charged by a generator driven by the vehicle's engine, the system comprising: a voltage sensor capable of acquiring battery voltage; an engine speed sensor capable of acquiring engine speed; a first device capable of acquiring the number of engine start-stop cycles; a second device capable of acquiring the cumulative battery operating time; and an electronic device as described above, the electronic device being directly or indirectly communicatively connected to the voltage sensor, the engine speed sensor, the first device, and the second device.

[0013] In the method and system for predicting the remaining life of a starter battery according to the present invention, a starter battery simulation model simulating the actual operating conditions of the starter battery is first constructed based on the basic physical principles of starter battery operation. This simulation model allows the battery current to be calculated from engine speed and battery voltage, which are normal background monitoring parameters of the vehicle. This starter battery simulation model can be considered a virtual sensor, enabling the acquisition of the battery current—a key parameter for predicting the remaining battery life—without the need to install a real current sensor on the vehicle, significantly saving costs. Furthermore, based on historical data of engine start-stop cycles, cumulative battery operating time, and battery voltage from at least one actual vehicle with a complete starter battery lifespan, along with historical data of the battery current calculated from the starter battery simulation model and the actual battery life, a prediction model is constructed using machine learning. This prediction model can then predict the remaining life of the starter battery of the target vehicle. Therefore, real-time monitoring of the operating status of the starter battery of construction machinery can be achieved in a low-cost, convenient, and rapid manner, estimating the remaining life of the starter battery and maintenance / replacement time, thereby reducing unnecessary downtime of construction machinery and improving production efficiency. In predicting the remaining lifespan of starter batteries, in addition to battery voltage and current, parameters such as engine start-stop cycles and cumulative battery operating time, which have a significant impact on battery life, are also considered, greatly improving the accuracy of lifespan prediction. For construction machinery manufacturers, the predicted remaining lifespan of starter batteries allows them to promptly understand their usage status and make advance arrangements for spare parts management and after-sales service, thereby improving the quality and efficiency of after-sales service. Attached Figure Description

[0014] The features and advantages of the present invention will now be described in detail with reference to the accompanying drawings and by way of non-limiting embodiments, in which:

[0015] Figure 1This is a flowchart of a method for predicting the remaining life of a startup battery according to an embodiment of the present invention;

[0016] Figure 2 This is a schematic structural diagram of a system for predicting the remaining life of a startup battery according to an embodiment of the present invention. Detailed Implementation

[0017] Exemplary embodiments of the present invention will now be described in detail with reference to the accompanying drawings. Numerous specific details are set forth in the following description to enable those skilled in the art to fully understand the invention. However, it will be apparent to those skilled in the art that implementations of the invention may not include some of these specific details. Furthermore, it should be understood that the invention is not limited to the specific embodiments described. Rather, the invention can be conceived to be practiced with any combination of the features and elements described herein, regardless of whether they relate to different embodiments. Therefore, the following aspects, features, embodiments, and advantages are illustrative only and should not be construed as elements or limitations of the claims unless expressly set forth in the claims.

[0018] Figure 1 A method for predicting the remaining lifespan of a vehicle's starter battery according to an embodiment of the present invention is shown. The method may include the following steps: S110: constructing a starter battery simulation model; S120: obtaining historical data of the battery current of an actual vehicle with a complete starter battery lifespan from the starter battery simulation model; S130: constructing a prediction model using machine learning based on historical data of the actual vehicle's engine start-stop count, cumulative battery operating time, battery voltage, and battery current, as well as the actual battery lifespan; S140: obtaining historical data of the target vehicle's battery current from the starter battery simulation model; and S150: inputting the historical data of the target vehicle's engine start-stop count, cumulative battery operating time, battery voltage, and battery current into the prediction model to predict the remaining lifespan of the target vehicle's starter battery.

[0019] Step S110 is used to construct a starting battery simulation model. A "starting battery simulation model" refers to a computer model that abstracts the real starting battery system into a mathematical model based on the fundamental physical principles of starting battery operation, and uses computer algorithms to implement this mathematical model to simulate the actual operating conditions of the starting battery, thereby enabling prediction and analysis. In a real engineering machinery starting battery system, the starting battery is connected to a generator driven by the vehicle's engine to be charged by the electricity generated by the generator, and also connected to battery loads on the vehicle, such as the starter motor and other electrical equipment, to supply them with power. Since the generator can charge the starting battery, the rotational speed of the generator shaft (referred to as "generator speed") affects the starting battery's state parameters, such as the battery current. Because the generator is driven by the engine, there is a pre-designed proportional relationship between the engine speed and the generator speed; therefore, it can also be said that the engine speed affects the battery current. Furthermore, the battery voltage also affects the battery current. Thus, with all the parameters of the starting battery system determined, changes in engine speed (or the corresponding generator speed) and battery voltage will cause specific changes in the battery current. Therefore, for a starter battery simulation model that simulates the actual operation of a starter battery, using engine speed and battery voltage as input parameters (independent variables), the battery current can be calculated as its output parameter (dependent variable). For construction machinery, engine speed and battery voltage are normally monitored in real-time / periodically, and can be obtained from the vehicle manufacturer's or user's database. Therefore, the starter battery simulation model can be considered a "virtual sensor" that can calculate the battery current from the known values ​​of the monitored parameters, thus eliminating the need for costly physical current sensors to be installed on the vehicle's starter battery. It should be noted that the parameters of the battery load powered by the starter battery (such as internal resistance) may also affect the battery current. However, given the complexity and variability of battery loads, and the fact that the starter battery supplies power to the battery load for a much shorter time than the time the starter battery is charged by the alternator, the impact of changes in battery load on the battery current is ignored here.

[0020] Building a startup battery simulation model can be achieved by programming various mathematical models (similar to functional relationships) that reflect the operating principles and laws of a startup battery system in a computer. Several commercially available software programs are available for building such simulation models of battery systems, typically MATLAB, or other common power system simulation software.

[0021] When constructing the aforementioned starter battery simulation model, model parameters need to be set to simulate the actual operating conditions of the starter battery. Model parameters can be represented as weights or coefficients preceding the input parameters (engine speed and battery voltage) in the functional relationship describing the battery model, characterizing the respective influence weights of engine speed and battery voltage on the battery current. In different starter battery systems, due to differences in the inherent physical parameters of different starter batteries and / or generators, these weights or coefficients will also differ between different starter battery systems. In this case, a rough initial system simulation model (at least involving the starter battery and generator) can be established using the estimated model parameters. Since the estimated model parameters may not be accurate, the battery current calculated from this initial model based on engine speed and battery voltage may also be inaccurate. Therefore, a real starting battery system can be built in a laboratory, for example, and the starting battery connected to the generator can be tested to simulate various operating conditions of the starting battery. This allows for the acquisition of experimental data P_0, V_0, and I_0 regarding engine speed (which corresponds to generator speed), battery voltage, and battery current. This experimental data can be measured by various physical sensors (such as speed sensors, voltage sensors, and current sensors) installed in the laboratory system, reflecting the actual operating status of the starting battery system. It should be noted that the above-mentioned experimental operation can also be conducted on a real engineering vehicle equipped with a battery current sensor. In this case, the experimental data P_0 of engine speed and V_0 of battery voltage can be obtained from a background database, while the experimental data I_0 of battery current can be measured by the installed battery current sensor. Finally, for the starting battery during experimental operation, the battery current I obtained from the initial simulation model (obtained by inputting the measured data of the engine speed P_0 and battery voltage V_0 into the initial simulation model) is compared with the corresponding measured data I_0 of the battery current obtained above. The difference between them is obtained, and the model parameters of the initial simulation model are corrected or adjusted according to the difference to reduce the difference. This correction or adjustment is repeated until the difference is reduced to below a predetermined threshold. At this point, the model parameters can be considered to be adjusted in place, and the initial simulation model naturally becomes a well-constructed starting battery simulation model with sufficient accuracy.

[0022] The aforementioned iterative corrections to the initial simulation model can be performed, for example, by iteratively training and tuning the initial simulation model using machine learning. The differences can be, for example, expressed as Mean Absolute Error (MAE), a commonly used error metric for evaluating model performance in regression tasks within machine learning. In one embodiment, when the MAE exceeds a predetermined threshold, for example, 0.1, it indicates that the model accuracy is still insufficient; therefore, the model parameters are corrected and training continues until the MAE is less than 0.1, at which point the model accuracy is considered to have met the requirements.

[0023] After the starter battery simulation model is constructed in step S110, the historical data of the battery current of an actual vehicle with a complete starter battery lifecycle can be obtained using the simulation model in step S120. In this application, "actual vehicle" refers to a vehicle that has experienced the entire process of at least one starter battery from its start-up to the end of its lifecycle and whose time history has been recorded in the background. In step S120, for at least one actual vehicle, the historical data of its engine speed and battery voltage are retrieved from the background database. In this application, "historical data" refers to the data with time information from the start-up of the starter battery until the end of its lifecycle, for the complete lifecycle of each starter battery. For example, engine speed, as one of the background monitoring parameters of the vehicle, is periodically (e.g., every few minutes) sent from the engine speed sensor on the vehicle to the vehicle's background database. Each sent engine speed is accompanied by time information, indicating the engine speed at which moment; and the historical data of engine speed is the engine speed at each moment in the complete lifecycle of each starter battery. Accordingly, the battery voltage and battery current at each moment in the complete lifecycle of each starter battery constitute the historical data of battery voltage and battery current, respectively. It should be noted that the historical data does not need to be data from all moments throughout the entire lifespan of the starter battery, but only data from at least a portion of those moments. In reality, for various reasons, the background database may not be able to record data from all moments, or some of the recorded data may be discarded due to being unqualified. In step S120, the historical data of engine speed and battery voltage are input into the constructed starter battery simulation model to calculate the historical data of battery current.

[0024] Next, in step S130, historical data on engine start-stop counts, cumulative battery operating time, battery voltage, and actual battery life can be retrieved from the background database of each actual vehicle. This data, combined with the historical battery current data obtained in step S120, is used to continuously perform machine learning on these data for each actual vehicle. This establishes the relationship between the four parameters—engine start-stop counts, cumulative battery operating time, battery voltage, and battery current—and battery life, thereby constructing a predictive model that can predict the remaining lifespan of the starter battery based on these four parameters. It should be noted that each engine start-stop cycle means the starter battery needs to be charged and discharged once; therefore, the number of engine start-stop cycles corresponds to the number of charge-discharge cycles of the starter battery, which is a crucial factor affecting the remaining lifespan of the starter battery. Similarly, the cumulative battery operating time represents the total time the starter battery has been charged or discharged since it was first put into use, and it also significantly impacts the remaining battery lifespan. Therefore, this invention considers engine start-stop counts and cumulative battery operating time, in addition to battery voltage and battery current, as influencing factors in predicting the remaining battery lifespan, which greatly improves the accuracy of lifespan prediction. Among the parameters mentioned above, the data on engine start-stop counts, cumulative battery operating time, battery voltage, and battery current all contain time information (indicating when the data is from), and the actual battery lifespan also indicates when the battery lifespan ends. Therefore, the remaining battery lifespan corresponding to the time specified in the data for each parameter can be calculated. Thus, a predictive model for predicting the remaining battery lifespan can be constructed by performing machine learning on the historical data of each parameter and the corresponding remaining battery lifespan data. In one embodiment, the historical data obtained from the background database can be retrieved based on each working cycle (charge-discharge cycle) of the starter battery's complete lifespan. That is, the number of engine start-stop counts (which should be the same as the number of working cycles), cumulative battery operating time, and battery voltage (wherein the battery voltage can be, for example, the average battery voltage at several moments within each working cycle) corresponding to each working cycle of the starter battery can be retrieved.

[0025] The specific construction process of the aforementioned predictive model may include conventional machine learning steps such as data reading / preparation, data processing, model selection, model training, model evaluation / validation, and model tuning. Here, "building" a predictive model refers to the process of encompassing these machine learning steps to ultimately generate a machine learning model with the required generalization ability that can be directly deployed and applied. It can be understood that the more actual vehicles or historical data learned, the more accurate the trained machine learning model will be. Specifically, the data processing step may include prioritizing the handling of null and singular values ​​in the model building data, for example, obtained from a backend database (e.g., using other values ​​near null or singular values ​​to fill in null values ​​or correct singular values) to ensure the completeness of all data used to build the model. Optionally, the model building data can be categorized to describe the battery's operating state as important reference information. For example, by analyzing the acquired model building data, it can be determined whether the starting battery is in a high-current discharge, low-current charging, or low-current discharge state at a given time. Furthermore, the model building data can be normalized to avoid affecting the accuracy of the overall model prediction results due to differences in data units. The model training step can employ the common 80 / 20 classification method to randomly split the dataset, with 80% of the data used as the training set and 20% as the validation set. The model validation and tuning steps can involve inputting validation set data and optimizing the model parameters to select the model with the highest accuracy.

[0026] In step S130, various suitable machine learning algorithms can be used to train the prediction model. Given the nature of the prediction model, time series forecasting algorithms capable of predicting future trends and patterns based on historical data are particularly suitable, such as the Long Short-Term Memory (LSTM) network algorithm. Of course, the invention is not limited to this; other possible algorithms include, but are not limited to, random forests, support vector machines, backpropagation neural networks, etc. The appropriate model algorithm can be selected by considering factors such as data characteristics, model algorithm complexity, computational resources, and prediction accuracy. Depending on the specific circumstances, different algorithms or a combination of multiple algorithms can be tried to improve the accuracy and reliability of the prediction results.

[0027] In step S140, for any target vehicle whose remaining lifespan of the starter battery is to be predicted, the historical data of its battery current can be calculated from the starter battery simulation model constructed in step S110, based on the historical data of its engine speed and battery voltage, as in step S120. It should be noted that step S140 is not necessarily as described in step S120. Figure 1 The procedure is not performed after step S130, but can be performed at any time after step S110 and before step S150 below, for example, before or between steps S120 or S130.

[0028] Next, in step S150, the historical data of the target vehicle's engine start-stop count, cumulative battery operating time, battery voltage (which can be obtained from its background database), and the historical data of its battery current obtained in step S140 are input into the prediction model to predict the remaining life of the target vehicle's starting battery.

[0029] Advantageously, the prediction results of the remaining life of the starting battery of the target vehicle output from the prediction model can be transmitted to the vehicle's display device for display to the user (e.g., different colors can be used to display different lengths of the remaining life of the starting battery to warn or remind the user) or transmitted to the manufacturer's or user's back-end database system, etc., so that the remaining life of the starting battery can be remotely monitored at any time, so as to make spare parts management and after-sales service arrangements in advance.

[0030] Figure 2 A schematic structural diagram of an exemplary configuration of a system 200 for predicting the remaining life of a vehicle's starter battery according to the present invention is shown. The system may include: a voltage sensor 10 mounted on the vehicle capable of acquiring battery voltage; an engine speed sensor 15 mounted on the vehicle capable of acquiring engine speed; a first device 20 mounted on the vehicle capable of acquiring the number of engine start-stop cycles; a second device 25 mounted on the vehicle capable of acquiring the cumulative battery operating time; and an electronic device 400. The electronic device 400 may include a processor 410 and a memory 420 for storing executable instructions of the processor 410. The processor 410 is configured to perform various steps of the method for predicting the remaining life of a starter battery according to any of the above embodiments by executing the executable instructions. The electronic device 400 may be in the form of a general-purpose computing device (such as a remote server or cloud platform, or an onboard high-performance computer or controller for construction machinery).

[0031] Voltage sensor 10, engine speed sensor 15, first device 20, and second device 25 can communicate with a remote backend database system 50, for example, via an in-vehicle remote communication terminal (e.g., a smart box on the vehicle, such as Product Link or T-box), to transmit real-time / periodic data (which may be directly measured or indirectly obtained through calculation or conversion) of battery voltage, engine speed, engine start-stop count, and cumulative battery operating time. Electronic device 400 can communicate with voltage sensor 10, engine speed sensor 15, first device 20, second device 25, and / or backend database system 50 to directly or indirectly acquire relevant data from these sensors or devices 10, 15, 20, 25, and to output or transmit predicted results of the remaining life of the starting battery to the backend database system 50. Electronic device 400 can also communicate with in-vehicle display device 30 to output the predicted results.

[0032] The present invention also provides a computer-readable storage medium having a computer program stored thereon, the program including executable instructions that, when executed by, for example, a processor, can implement the steps of the method for predicting the remaining lifespan of a startup battery according to any of the above embodiments. In some possible embodiments, aspects of the invention can also be implemented as a program product including program code that, when run on a terminal device, causes the terminal device to perform the steps of various exemplary embodiments of the method for predicting the remaining lifespan of a startup battery according to the present invention.

[0033] The program product for implementing the above method according to embodiments of the present invention can employ a portable compact disc read-only memory (CD-ROM) and include program code, and can run on a terminal device such as a remote server or an in-vehicle computer. However, the program product of the present invention is not limited thereto. In this document, a readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0034] The program product may employ any combination of one or more readable media. A readable medium may be a readable signal medium or a readable storage medium. A readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof.

[0035] The computer-readable storage medium may include data signals propagated in baseband or as part of a carrier wave, carrying readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. The readable storage medium may also be any readable medium other than a readable storage medium, capable of transmitting, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the readable storage medium may be transmitted using any suitable medium, including but not limited to wireless, wired, optical fiber, RF, etc., or any suitable combination thereof.

[0036] Program code for performing the operations of this application can be written in any combination of one or more programming languages, including object-oriented programming languages ​​such as Java and C++, and conventional procedural programming languages ​​such as C or similar languages. The program code can execute entirely on the user's computing device, partially on the user's device, as a standalone software package, partially on the user's computing device and partially on a remote computing device, or entirely on a remote computing device or server. In cases involving remote computing devices, the remote computing device can be connected to the user's computing device via any type of network, including a local area network (LAN) or a wide area network (WAN), or it can be connected to an external computing device (e.g., via the Internet using an Internet service provider).

[0037] Through the above description of the embodiments, those skilled in the art will readily understand that the exemplary embodiments described herein can be implemented by software or by combining software with necessary hardware. Therefore, the technical solutions according to embodiments of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) or on a network, including several instructions to cause a computing device (such as a personal computer, server, or network device, etc.) to execute the method for predicting the remaining lifespan of a startup battery according to embodiments of the present invention.

[0038] Industrial applicability

[0039] The method and system for predicting the remaining life of a starter battery according to the present invention can be applied to various engineering machinery or vehicles equipped with a starter battery, a generator and an engine. It can be any type of engineering machinery (e.g., loaders, excavators, etc.) that performs operations related to a specific industry (e.g., mining, construction, agriculture, transportation, etc.) and travels and operates in various working environments (e.g., mines, construction sites, farms and roads, etc.).

[0040] In implementing this invention, a starting battery simulation model that can calculate the battery current based on engine speed and battery voltage can be constructed first. Then, for both actual vehicles with a complete starting battery lifespan and target vehicles whose remaining starting battery lifespan needs to be predicted, historical data of battery current can be calculated from the starting battery simulation model based on historical data of engine speed and battery voltage. After constructing a prediction model through machine learning based on, for example, historical data of engine start-stop times, battery cumulative operating time, battery voltage and battery current, and actual battery lifespan of multiple actual vehicles, the remaining lifespan of the target vehicle's starting battery can be predicted by inputting the historical data of engine start-stop times, battery cumulative operating time, battery voltage and battery current of the target vehicle into the prediction model.

[0041] The method and system of this invention enable low-cost, convenient, and rapid real-time monitoring of the operating status of starter batteries in construction machinery, predicting the remaining lifespan of the starter batteries and maintenance / replacement time, thereby reducing unnecessary downtime and improving production efficiency. In predicting the remaining lifespan of the starter batteries, in addition to battery voltage and current, parameters that significantly impact battery lifespan, such as engine start-stop cycles and cumulative battery operating time, are considered, greatly improving the accuracy of lifespan prediction. Furthermore, for construction machinery manufacturers, the predicted remaining lifespan of the starter batteries allows for timely understanding of their usage status, enabling them to proactively manage spare parts and arrange after-sales service, thus improving the quality and efficiency of after-sales service.

[0042] Other embodiments of the invention will readily occur to those skilled in the art upon consideration of the specification and practice of the disclosure herein. This disclosure is intended to cover any variations, uses, or adaptations of the invention. The specification and examples are to be considered exemplary only, and the true scope and spirit of the invention are indicated by the appended claims.

Claims

1. A method for predicting the remaining life of a vehicle's starter battery, the starter battery being charged by a generator driven by the vehicle's engine, the method comprising the steps of: Construct a starting battery simulation model that can calculate the battery current based on engine speed and battery voltage; For at least one actual vehicle with a complete life cycle of the starter battery, the historical data of its battery current are obtained from the starter battery simulation model based on its historical engine speed and battery voltage data. Based on the actual vehicle's engine start-stop count, cumulative battery operating time, historical data of battery voltage, historical data of battery current obtained from calculation, and actual battery life, a predictive model for predicting the remaining life of the starting battery is constructed using machine learning. For the target vehicle, based on its historical engine speed and battery voltage data, the historical battery current data is obtained from the starting battery simulation model. as well as The target vehicle's engine start-stop count, cumulative battery operating time, historical data of battery voltage, and historical data of its battery current are input into the prediction model to predict the remaining life of the target vehicle's starting battery.

2. The method according to claim 1, characterized in that, The construction of the startup battery simulation model includes the following sub-steps: (a) establishing an initial simulation model involving at least the startup battery and the generator; (b) conducting an experimental run of the startup battery connected to the generator and acquiring experimental data of the battery current using sensors; and (c) correcting the model parameters by comparing the difference between the battery current of the startup battery during the experimental run, which is calculated by the initial simulation model, and the corresponding experimental data of the acquired battery current, so as to reduce the difference to below a predetermined threshold.

3. The method according to claim 2, characterized in that, The sub-step (c) is performed by repeatedly training and optimizing the initial simulation model using machine learning.

4. The method according to any one of claims 1 to 3, characterized in that, The prediction model is a machine learning model based on time series prediction algorithms.

5. The method according to claim 4, characterized in that, The time series prediction algorithm includes one or more of the following: Long Short-Term Memory (LSTM) network algorithm, Random Forest, Support Vector Machine, and Backpropagation Neural Network.

6. A computer-readable storage medium having a computer program stored thereon, the computer program including executable instructions that, when executed by a processor, implement the method according to any one of claims 1 to 5.

7. An electronic device (400), comprising: Processor (410); as well as Memory (420) for storing executable instructions of the processor; The processor (410) is configured to execute the executable instructions to implement the method according to any one of claims 1 to 5.

8. A system for predicting the remaining life of a vehicle's starter battery, the starter battery being charged by a generator driven by the vehicle's engine, the system comprising: A voltage sensor (10) capable of acquiring battery voltage; An engine speed sensor (15) capable of acquiring engine speed; A first device (20) capable of acquiring the number of engine start-stop cycles; A second device (25) capable of acquiring the cumulative battery operating time; as well as The electronic device (400) according to claim 7 is directly or indirectly communicatively connected to the voltage sensor, the engine speed sensor, the first device, and the second device.

Images