New energy vehicle digital twin modeling method and system based on multi-source data fusion

Abstract

The present invention relates to the field of data analysis, and in particular to a digital twin modeling method and system for new energy vehicles based on multi-source data fusion. The present invention collects and stores real-time battery stability characteristics corresponding to a target vehicle; obtains the operating condition data of the target vehicle to extract corresponding driving inertia characteristics, determines the energy consumption characterization value of the target vehicle, and marks the target vehicle; updates the twin model data set of the battery corresponding to the target vehicle in response to the marking result; analyzes the twin model data set according to a pre-built evaluation model to determine potential anomalies. The present invention generates a large amount of data based on vehicle driving. The present invention updates the twin model data set through the characteristics presented by driving operation behavior, so that during the evaluation process of the twin model data set, the accuracy of digital modeling is adjusted in combination with the matching of actual data and historical data, thereby improving the accuracy of battery anomaly evaluation.

Description

Technical Field

[0001] The present invention relates to the field of data analysis, and in particular to a new energy vehicle digital twin modeling method and system based on multi-source data fusion. Background Art

[0002] As a core component of new energy vehicles, the performance of batteries directly affects key indicators such as the vehicle's range, safety, and service life. Accurately evaluating and predicting battery performance is of great significance for optimizing the design of new energy vehicles, improving user experience, and reducing usage costs. For example, by accurately predicting battery life, it can help users reasonably arrange battery replacement time and reduce usage costs. At present, some car companies have adopted remote collection methods to monitor vehicle operating data, so as to analyze the battery operation status and issue timely warnings based on the monitoring data to ensure vehicle safety.

[0003] Chinese patent application publication number: CN114329760A, discloses a vehicle-mounted lithium-ion battery modeling and fault diagnosis method based on digital twins. The vehicle-mounted lithium-ion battery is modeled based on real-time data transmitted by the running vehicle, wireless communication, neural network algorithm and gradient descent optimization algorithm, and finally a twin model is obtained that can accurately analyze the battery operating status, predict the battery's health status throughout its life cycle and future performance. The twin model is used to determine whether the battery has an abnormality and provide real-time feedback information to the running vehicle, and further action or intervention is taken on the battery body. The method provided in this application establishes a dynamic connection between the vehicle-mounted battery body and the twin battery model throughout the battery life cycle; the operating status of the vehicle-mounted lithium-ion battery can be accurately monitored and analyzed in real time.

[0004] However, the prior art still has the following problems:

[0005] In reality, when conducting abnormal analysis of the battery performance of a large number of vehicles, the large amount of data generated during the vehicle driving process is not analyzed in a targeted manner, resulting in high computing power consumption. At the same time, only the characteristics presented by the battery itself are analyzed, and the abnormal impact of driving operation on the battery is not considered. As a result, the accuracy of battery abnormality assessment is reduced. Summary of the Invention

[0006] To this end, the present invention provides a new energy vehicle digital twin modeling method and system based on multi-source data fusion, which is used to overcome the problem in the prior art that, when performing abnormal analysis on the battery performance of a large number of vehicles, a large amount of data generated during the vehicle driving process is not analyzed in a targeted manner, resulting in high computing power consumption. At the same time, only the characteristics presented by the battery itself are analyzed, and the abnormal impact of driving operation on the battery is not considered, which in turn leads to reduced accuracy of battery abnormality assessment.

[0007] To achieve the above objectives, the present invention provides a new energy vehicle digital twin modeling method based on multi-source data fusion, which includes:

[0008] Collect and store the real-time battery stability characteristics corresponding to the target vehicle in real time;

[0009] Obtaining the operating condition data of the target vehicle to extract the corresponding driving inertia characteristics, determining the energy consumption characterization value of the target vehicle, and marking the target vehicle;

[0010] In response to the marking result, the twin model dataset of the battery corresponding to the target vehicle is updated, including:

[0011] Retrieving the motor system's sound frequency data to determine the abnormal sound time segment, obtaining driving operation characteristics within the abnormal sound time segment, and combining them with the auxiliary system's usage duration to evaluate battery consumption interference characterization parameters. Based on the consumption interference characterization parameters, the real-time battery stability characteristics within the abnormal sound time segment are filtered out and stored in the twin model dataset;

[0012] Analyze the twin model dataset based on the pre-built evaluation model to identify potential anomalies;

[0013] The evaluation model is used to compare the real-time battery stability characteristics in the twin model data set with the historical battery stability characteristics, determine the battery stability deviation value, and match the battery stability deviation value with the stability error range to determine whether the corresponding battery has potential abnormalities;

[0014] Among them, the driving inertia characteristics include the number of engine shutdowns within a predetermined driving distance and the average time interval from shutdown to restart; the driving operation characteristics include the frequency of accelerator pedal pressing and the average pedaling duration; the real-time battery stability characteristics include the duration of battery heating and the duration of maximum temperature maintenance.

[0015] Furthermore, the process of determining the energy consumption characterization value of the target vehicle includes:

[0016] The ratio of the number of stalls within the predetermined driving distance to the stall threshold is used as the first energy consumption characteristic;

[0017] taking a ratio of an average of time intervals from engine shutdown to engine restart within the predetermined driving distance to a threshold of an average of time intervals as a second energy consumption characteristic;

[0018] The sum of the first energy consumption characteristic and the second energy consumption characteristic is determined as the energy consumption characterization value.

[0019] Furthermore, marking the target vehicle includes:

[0020] If the energy consumption characterization value of the target vehicle is greater than or equal to the energy consumption characterization threshold, the target vehicle is marked as a high energy consumption vehicle.

[0021] Further, in response to the marking result, including:

[0022] If the target vehicle is marked as a high-energy consumption vehicle, the twin model dataset of the battery corresponding to the target vehicle is updated.

[0023] Furthermore, the process of determining the time domain segment of the abnormal sound includes:

[0024] Based on the sound frequency data, a time domain curve of the sound frequency of the target vehicle within a predetermined driving distance is constructed;

[0025] If there is any time domain segment whose corresponding slope is greater than or equal to the slope threshold, the time domain segment is determined as the abnormal sound time domain segment.

[0026] Furthermore, the process of evaluating the consumption interference characterization parameter for the battery includes,

[0027] The sum of the ratio of the accelerator pedal's stepping frequency to the stepping frequency threshold and the ratio of the average stepping duration to the average stepping duration threshold is used as the first consumption interference feature;

[0028] The ratio of the auxiliary system usage time to the usage time threshold is used as the second consumption interference feature;

[0029] The first consumption interference feature and the second consumption interference feature are weightedly summed to determine the consumption interference characterization parameter.

[0030] Furthermore, the process of screening out the real-time battery stability characteristics within the abnormal noise time domain based on the consumption interference characterization parameter includes:

[0031] Determine the consumption interference characterization parameters corresponding to each abnormal sound time domain segment;

[0032] If the consumption interference characterization parameter corresponding to the abnormal sound time domain segment is greater than or equal to a preset consumption interference characterization parameter threshold, the real-time battery stability feature corresponding to the abnormal sound time domain segment is screened out.

[0033] Furthermore, the process of determining the battery stability deviation value includes:

[0034] Recall historical battery stability characteristics, including the average duration of battery temperature rise and the average duration of maximum temperature maintenance;

[0035] Calculating a first deviation between a duration of battery heating up and an average of the durations of battery heating up;

[0036] Calculating a second deviation value between the maximum temperature maintenance time and the average of the maximum temperature maintenance time;

[0037] The sum of the first deviation value and the second deviation value is used as the battery stability deviation value.

[0038] Furthermore, the process of determining whether the corresponding battery has a potential abnormality includes:

[0039] If the battery stability deviation value is not within the stability error range, it is determined that the corresponding battery has a potential abnormality;

[0040] The stability error range is determined by obtaining a maximum value of the battery stability deviation and a minimum value of the battery stability deviation in the battery stability deviation value history data.

[0041] Furthermore, a system for applying a new energy vehicle digital twin modeling method based on multi-source data fusion is provided, including:

[0042] A battery collection module is used to collect and store the real-time battery stability characteristics corresponding to the target vehicle in real time;

[0043] An energy consumption analysis module is used to obtain the operating condition data of the target vehicle, extract the corresponding driving inertia characteristics, determine the energy consumption characterization value of the target vehicle, and mark the target vehicle;

[0044] A data update module is connected to the battery acquisition module and the energy consumption analysis module respectively, and updates the twin model data set of the battery corresponding to the target vehicle in response to the marking result, including:

[0045] Retrieving the motor system's sound frequency data to determine the abnormal sound time segment, obtaining driving operation characteristics within the abnormal sound time segment, and combining them with the auxiliary system's usage duration to evaluate battery consumption interference characterization parameters. Based on the consumption interference characterization parameters, the real-time battery stability characteristics within the abnormal sound time segment are filtered out and stored in the twin model dataset;

[0046] An analysis and evaluation module, which, together with the data update module, is used to analyze the twin model dataset based on a pre-built evaluation model to determine potential anomalies;

[0047] The evaluation model is used to compare the real-time battery stability characteristics in the twin model data set with the historical battery stability characteristics, determine the battery stability deviation value, and match the battery stability deviation value with the stability error range to determine whether the corresponding battery has potential abnormalities;

[0048] Among them, the driving inertia characteristics include the number of engine shutdowns within a predetermined driving distance and the average time interval from shutdown to restart; the driving operation characteristics include the frequency of accelerator pedal pressing and the average pedaling duration; the real-time battery stability characteristics include the duration of battery heating and the duration of maximum temperature maintenance.

[0049] Compared with the existing technology, the present invention collects and stores the real-time battery stability characteristics corresponding to the target vehicle in real time; obtains the operating condition data of the target vehicle to extract the corresponding driving inertia characteristics, determines the energy consumption characterization value of the target vehicle, and marks the target vehicle; in response to the marking results, updates the twin model dataset of the battery corresponding to the target vehicle; analyzes the twin model dataset according to the pre-built evaluation model to determine potential anomalies. The present invention updates the twin model dataset based on the characteristics presented by the driving operation behavior, so that during the evaluation process of the twin model dataset, the accuracy of the digital modeling is adjusted in combination with the matching of actual data and historical data, thereby improving the accuracy of battery anomaly evaluation.

[0050] In particular, the present invention takes into account improper driving operations during driving, among which frequent shutdowns may cause the battery to stop supplying power to the motor, and each time the battery is started, the battery needs to provide a large current to drive the motor, which will cause the battery to undergo large charge and discharge changes. It also means that the battery undergoes multiple charge and discharge cycles in a short period of time, which may cause the battery to be over-discharged or undercharged, affecting the battery's performance and life; the time interval from vehicle shutdown to startup reflects the stability of the vehicle's operating state, which in turn affects the battery's output power and battery stability. If the vehicle is shut down many times during driving and the time interval from shutdown to startup is short, the battery does not have enough time to dissipate heat, which will cause the battery temperature to continue to rise, further affecting the battery's performance and life; therefore, the present invention determines the energy consumption characterization value of the target vehicle through the driving inertia characteristics of the target vehicle to characterize the vehicle's energy consumption level, quantifies the abnormal risks brought to the battery by driving behavior, and provides data support for subsequent marking of the target vehicle. The present invention improves the accuracy of battery abnormality assessment.

[0051] In particular, the present invention updates the twin model data set corresponding to the target vehicle marked as high energy consumption, analyzes the characteristics of the motor system of the target vehicle during driving, and finds that the operating status of the battery is closely related to the sound of the motor system of the vehicle. By determining the abnormal sound time domain segment corresponding to the motor system during driving, that is, the time domain segment where the driving operation interferes more seriously with the battery consumption, in actual conditions, when the motor system operates at high frequency, the frequent stepping on the accelerator pedal will make the abnormal phenomenon more obvious, which can indicate the frequent acceleration and deceleration of the vehicle and has strong data characterization properties, and further reflects the degree of charge and discharge cycles experienced by the battery. Frequent charge and discharge will accelerate battery aging and shorten the effective service life of the battery. Based on the different driving habits of the driver, there is a situation where the accelerator pedal is stepped on for a long time during driving, which makes the When the vehicle is in a state of continuous acceleration or high-speed driving, the current output by the battery is large, and it is in a state of high-current discharge for a long time, which accelerates the aging of the battery. In addition, the high-current discharge will cause more heat to be generated inside the battery, and may also trigger the battery's thermal protection mechanism, limit the battery's charge and discharge power, and affect the vehicle's power performance; at the same time, the continuous use of the auxiliary system during driving will also affect the battery's energy burden and increase the battery's energy impact. Therefore, the present invention determines the consumption interference characterization parameters of the target vehicle through the driving operation characteristics of the target vehicle combined with the usage time of the auxiliary system to characterize the degree of loss of the battery's performance status affected by interference, quantify the degree of influence of the driver's driving habits on battery performance, and provide data support for the subsequent screening of real-time battery stability characteristics in the abnormal time domain. The present invention improves the accuracy of battery abnormality assessment.

[0052] In particular, specifically, the present invention analyzes the twin model data set according to the pre-built evaluation model to determine whether there is a potential abnormality in the battery. The duration of battery temperature rise and the corresponding maximum temperature maintenance time can reflect the heat dissipation performance of the battery thermal management system, which affects the performance and life of the battery. By matching the real-time data and historical conditions of the above two characteristics, the battery stability deviation value is determined to characterize the degree of deviation between the actual situation and the battery performance evaluated by the evaluation model, and then matched with the stability error range to determine whether the corresponding battery has a potential abnormality, thereby improving the accuracy of battery abnormality evaluation. BRIEF DESCRIPTION OF THE DRAWINGS

[0053] Figure 1 A schematic diagram of the steps of a new energy vehicle digital twin modeling method based on multi-source data fusion according to an embodiment of the invention;

[0054] Figure 2 A logic decision diagram for marking a target vehicle according to an embodiment of the invention;

[0055] Figure 3A logic decision diagram for determining the time domain segment of abnormal noise according to an embodiment of the present invention;

[0056] Figure 4 This is a logic decision diagram for determining whether a corresponding battery has a potential abnormality according to an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the objects and advantages of the present invention more clearly understood, the present invention is further described below in conjunction with embodiments; it should be understood that the specific embodiments described herein are merely used to explain the present invention and are not intended to limit the present invention.

[0058] The preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only used to explain the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0059] It should be noted that, in the description of the present invention, unless otherwise expressly specified or limited, the terms "mounted" and "connected" should be understood in a broad sense. For example, they may refer to fixed connections, detachable connections, or integral connections; they may refer to mechanical connections or electrical connections. Those skilled in the art will understand the specific meanings of the above terms in the present invention based on specific circumstances.

[0060] See also Figure 1 As shown, it is a schematic diagram of the steps of the new energy vehicle digital twin modeling method based on multi-source data fusion according to an embodiment of the present invention. The new energy vehicle digital twin modeling method based on multi-source data fusion according to an embodiment of the present invention includes:

[0061] Step S1: real-time battery stability characteristics corresponding to the target vehicle are collected and stored. During implementation, authorization is required when collecting relevant data of the target vehicle.

[0062] Step S2, obtaining the running condition data of the target vehicle to extract the corresponding driving inertia characteristics, determining the energy consumption characterization value of the target vehicle, and marking the target vehicle;

[0063] Step S3, in response to the marking result, updating the twin model dataset of the battery corresponding to the target vehicle, including:

[0064] Retrieving the motor system's sound frequency data to determine the abnormal sound time segment, obtaining driving operation characteristics within the abnormal sound time segment, and combining them with the auxiliary system's usage duration to evaluate battery consumption interference characterization parameters. Based on the consumption interference characterization parameters, the real-time battery stability characteristics within the abnormal sound time segment are filtered out and stored in the twin model dataset;

[0065] Step S4: Analyze the twin model dataset based on the pre-built evaluation model to determine potential anomalies;

[0066] The evaluation model is used to compare the real-time battery stability characteristics in the twin model data set with the historical battery stability characteristics, determine the battery stability deviation value, and match the battery stability deviation value with the stability error range to determine whether the corresponding battery has potential abnormalities;

[0067] Among them, the driving inertia characteristics include the number of engine shutdowns within a predetermined driving distance and the average time interval from shutdown to restart; the driving operation characteristics include the frequency of accelerator pedal pressing and the average pedaling duration; the real-time battery stability characteristics include the duration of battery heating and the duration of maximum temperature maintenance.

[0068] Specifically, the twin model dataset can be composed of a database to store the extracted relevant data.

[0069] Specifically, the functions of the evaluation model are all logical operations, and there is no limitation on the implementation form of the evaluation model. It can be a computer program that can perform numerical operations, extract real-time battery stability characteristics and compare them with historical battery stability characteristics, and determine the battery stability deviation value, and then match the battery stability deviation value with the stability error range.

[0070] Specifically, the vehicle operating condition data includes driving inertia characteristics, driving operation characteristics, usage time of the auxiliary system, sound frequency data of the motor system, real-time battery stability characteristics, and historical battery stability characteristics.

[0071] Specifically, the vehicle control system of the new energy vehicle can be used to obtain whether the accelerator pedal is stepped on, the time node of the shutdown and the time node of the start-up. This is a prior art. Furthermore, those skilled in the art can count the frequency and duration of the stepping on the accelerator pedal based on whether the accelerator pedal is stepped on, and determine the number of shutdowns within a predetermined driving distance and the time interval from shutdown to start-up based on the time node of the shutdown and the time node of the start-up. This will not be repeated here.

[0072] Specifically, the battery management system (BMS) equipped in new energy vehicles monitors the battery temperature in real time to determine how long the battery temperature rises and how long the maximum temperature is maintained. This will not be elaborated here.

[0073] Specifically, there is no specific limitation on the method for obtaining the sound frequency data of the motor system. The sound frequency of the motor system can be indirectly obtained by analyzing the electrical signal converted by the vibration sensor from the mechanical energy of the motor vibration using the vibration sensor.

[0074] Among them, the motor system refers to the core power unit of new energy vehicles, which is mainly composed of motors, motor controllers, transmission devices, etc. The battery is its energy source, and there is a close communication and collaborative working relationship between the two.

[0075] Specifically, the process of determining the energy consumption characterization value of the target vehicle includes:

[0076] The ratio of the number of stalls within the predetermined driving distance to the stall threshold is used as the first energy consumption characteristic;

[0077] taking a ratio of an average of time intervals from engine shutdown to engine restart within the predetermined driving distance to a threshold of an average of time intervals as a second energy consumption characteristic;

[0078] The sum of the first energy consumption characteristic and the second energy consumption characteristic is determined as the energy consumption characterization value.

[0079] In this embodiment, the purpose of setting the number of stalls threshold and the time interval mean threshold is to characterize the situation where the driver's driving behavior results in high energy consumption. The operating condition data of the target vehicle completing several historical trips of a predetermined driving distance is obtained, and the historical data of the number of stalls and the historical data of the mean time interval from stall to start are called. The average number of stalls and the average value of the mean time intervals are solved and used as the corresponding reference values ​​under normal conditions. Based on the purpose of setting the above two thresholds, the number of stalls threshold is determined as the product of the average number of stalls and the number deviation coefficient, and the time interval mean threshold is determined as the product of the average value of the mean time interval and the interval offset coefficient, wherein the number deviation coefficient is selected within the interval [1.2, 1.3], and the interval offset coefficient is selected within the interval [1.3, 1.35].

[0080] Specifically, with respect to the determination of the predetermined driving distance, in order to ensure that the collected relevant data is representative and can accurately reflect the driver's driving habits, in this embodiment, the predetermined driving distance is determined to be 25 kilometers.

[0081] Specifically, the present invention takes into account improper driving operations during driving, among which frequent shutdowns may cause the battery to stop supplying power to the motor, and each time the battery is started, the battery needs to provide a large current to drive the motor, which will cause the battery to undergo large charge and discharge changes. It also means that the battery undergoes multiple charge and discharge cycles in a short period of time, which may cause the battery to be over-discharged or undercharged, affecting the battery's performance and life; the time interval from vehicle shutdown to startup reflects the stability of the vehicle's operating state, which in turn affects the battery's output power and battery stability. If the vehicle is shut down many times during driving and the time interval from shutdown to startup is short, the battery does not have enough time to dissipate heat, which will cause the battery temperature to continue to rise, further affecting the battery's performance and life; therefore, the present invention determines the energy consumption characterization value of the target vehicle through the driving inertia characteristics of the target vehicle to characterize the vehicle's energy consumption level, quantifies the abnormal risks brought to the battery by driving behavior, and provides data support for subsequent marking of the target vehicle. The present invention improves the accuracy of battery abnormality assessment.

[0082] Specifically, see Figure 2 As shown, it is a logic decision diagram for marking a target vehicle according to an embodiment of the present invention. Marking the target vehicle includes:

[0083] If the energy consumption characterization value of the target vehicle is greater than or equal to the energy consumption characterization threshold, the target vehicle is marked as a high energy consumption vehicle;

[0084] If the energy consumption characterization value of the target vehicle is less than the energy consumption characterization threshold, there is no need to mark the target vehicle.

[0085] The energy consumption characterization threshold is selected within the interval [2.18, 2.24].

[0086] Specifically, in response to the marking result, including:

[0087] If the target vehicle is marked as a high-energy consumption vehicle, the twin model dataset of the battery corresponding to the target vehicle is updated.

[0088] Specifically, see Figure 3 As shown, it is a logic decision diagram for determining the abnormal sound time domain segment according to an embodiment of the present invention. The process of determining the abnormal sound time domain segment includes:

[0089] Based on the sound frequency data, a time domain curve of the sound frequency of the target vehicle within a predetermined driving distance is constructed;

[0090] If there is any time domain segment whose corresponding slope is greater than or equal to the slope threshold, the time domain segment is determined as the abnormal sound time domain segment.

[0091] In this embodiment, the time domain curve of the sound frequency of the target vehicle within the predetermined driving distance is constructed according to the following method, including:

[0092] Construct a rectangular coordinate system with time as the horizontal axis and sound frequency as the vertical axis;

[0093] Marking the coordinate points of the sound frequency at each moment in the rectangular coordinate system;

[0094] The coordinate points are connected with a smooth curve to obtain the sound frequency time domain curve.

[0095] Specifically, there is no limitation on the method of constructing the time domain curve of the sound frequency. For example, the time domain curve can be fitted by using MATLAB related fitting software, which will not be described in detail here.

[0096] Specifically, the purpose of setting the slope threshold is to characterize the situation where the sound frequency of the motor system is relatively abnormal. Therefore, the slope threshold is selected within the interval [0.55, 0.6].

[0097] Specifically, the process of evaluating the consumption interference characterization parameters for the battery includes:

[0098] The sum of the ratio of the accelerator pedal's stepping frequency to the stepping frequency threshold and the ratio of the average stepping duration to the average stepping duration threshold is used as the first consumption interference feature;

[0099] The ratio of the auxiliary system usage time to the usage time threshold is used as the second consumption interference feature;

[0100] The first consumption interference feature and the second consumption interference feature are weightedly summed to determine the consumption interference characterization parameter.

[0101] Specifically, during the actual driving of a new energy vehicle, the driver's driving operation can more intuitively reflect the degree of loss of vehicle and battery performance caused by driving behavior. Therefore, in implementation, priority is given to the driving operation characteristics, namely the average frequency and duration of the accelerator pedal's depression. Therefore, a slightly higher weight is assigned to the first consumption interference feature calculated based on the driving operation characteristics. Therefore, when performing weighted summation, the weight of the first consumption interference feature is set to 0.6, and the weight of the second consumption interference feature is set to 0.4.

[0102] In this embodiment, the purpose of setting the pedaling frequency threshold and the pedaling duration average threshold is to characterize the situation where the driving behavior has a more serious impact on the performance consumption of the battery, obtain the operating vehicle condition data of the target vehicle completing several historical trips of the predetermined driving distance, call the historical data of the pedaling frequency of the accelerator pedal in the corresponding abnormal noise time domain, and solve the average value of the pedaling frequency and the average value of the pedaling duration, and use them as the corresponding benchmark values ​​under normal circumstances. Based on the purpose of setting the above two thresholds, the pedaling frequency threshold is determined as the product of the average value of the pedaling frequency and the frequency deviation coefficient, and the pedaling duration average threshold is determined as the product of the average value of the pedaling duration and the duration deviation coefficient, wherein the frequency deviation coefficient is selected within the interval [1.3,1.5], and the duration deviation coefficient is selected within the interval [1.1,1.2].

[0103] The purpose of setting the usage time threshold is to characterize the situation where the auxiliary system places a greater energy burden on the battery during driving, obtain the operating condition data of the target vehicle completing several historical trips of a predetermined driving distance, call the historical usage time data of the auxiliary system, solve the average usage time, and use it as a benchmark value under normal circumstances. Based on the purpose of setting the usage time threshold, the usage time threshold is determined as the product of the average usage time and a time offset coefficient, wherein the time offset coefficient is selected within the interval [1.2, 1.25].

[0104] Specifically, the new energy vehicle auxiliary system refers to a series of devices designed to improve the functionality, comfort, safety and convenience of the vehicle, in addition to the vehicle's core systems such as the power system, transmission system, and braking system, including air conditioning, audio, headlights, electric power steering, etc.

[0105] Specifically, the present invention updates the twin model dataset corresponding to the target vehicle marked as high-energy consumption, analyzes the characteristics of the target vehicle's motor system during driving, and finds that the operating status of the battery is closely related to the noise of the vehicle's motor system. By determining the time domain segment corresponding to the abnormal noise of the motor system during driving, that is, the time domain segment where the driving operation interferes more seriously with the battery consumption, in actual situations, when the motor system operates at high frequency, frequent stepping on the accelerator pedal will make the abnormal phenomenon more obvious, which can indicate the frequent acceleration and deceleration of the vehicle and has strong data characterization properties, further reflecting the degree of charge and discharge cycles experienced by the battery. Frequent charging and discharging will accelerate battery aging and shorten the effective service life of the battery.

[0106] Due to the different driving habits of drivers, there are cases where the accelerator pedal is stepped on for a long time during driving, causing the vehicle to be in a state of continuous acceleration or high-speed driving. The current output by the battery is large, and it is in a high-current discharge state for a long time, which accelerates the aging of the battery. In addition, the high-current discharge will cause more heat to be generated inside the battery, and may also trigger the battery's thermal protection mechanism, limit the battery's charge and discharge power, and affect the vehicle's power performance; at the same time, the continuous use of the auxiliary system during driving will also affect the battery's energy burden and increase the battery's energy impact. Therefore, the present invention determines the consumption interference characterization parameters of the target vehicle through the driving operation characteristics of the target vehicle combined with the usage time of the auxiliary system to characterize the degree of loss of the battery's performance status affected by interference, quantify the degree of influence of the driver's driving habits on battery performance, and provide data support for the subsequent screening of real-time battery stability characteristics in the abnormal time domain segment. The present invention improves the accuracy of battery abnormality assessment.

[0107] Specifically, the process of screening out the real-time battery stability characteristics within the abnormal noise time domain based on the consumption interference characterization parameter includes:

[0108] Determine the consumption interference characterization parameters corresponding to each abnormal sound time domain segment;

[0109] If the consumption interference characterization parameter corresponding to the abnormal sound time domain segment is greater than or equal to a preset consumption interference characterization parameter threshold, the real-time battery stability feature corresponding to the abnormal sound time domain segment is screened out.

[0110] The consumption interference characterization parameter threshold is selected within the interval [1.68, 1.74].

[0111] Specifically, the process of determining the battery stability deviation value includes:

[0112] Recall historical battery stability characteristics, including the average duration of battery temperature rise and the average duration of maximum temperature maintenance;

[0113] Calculating a first deviation between a duration of battery heating up and an average of the durations of battery heating up;

[0114] Calculating a second deviation value between the maximum temperature maintenance time and the average of the maximum temperature maintenance time;

[0115] The sum of the first deviation value and the second deviation value is used as the battery stability deviation value.

[0116] In this embodiment, the first deviation value and the second deviation value are determined by the following method, including:

[0117] Calculating the absolute value of the first difference between the duration of battery heating up and the average duration of battery heating up;

[0118] taking a ratio of an absolute value of the first difference to an average duration of the battery temperature rise as a first deviation value;

[0119] Accordingly, the second absolute value of the difference between the maximum temperature maintenance time and the average maximum temperature maintenance time is calculated;

[0120] The ratio of the absolute value of the second difference to the average of the maximum temperature maintenance time is used as the second deviation value;

[0121] Specifically, by obtaining the historical operating condition data of the target vehicle completing several predetermined driving distances, calling the historical data of the battery heating duration and the historical data of the maximum temperature maintenance time, and then solving the average value to obtain the corresponding average battery heating duration and the average maximum temperature maintenance time;

[0122] Among them, for new energy vehicles, during the driving process, the corresponding suitable operating temperature range of the battery is [0℃, 40℃]. Therefore, in implementation, the battery temperature is raised from 40℃ to any temperature, and the time it takes for the temperature to remain unchanged per unit time is determined as the battery heating duration, and then the time it takes for the temperature to remain unchanged is determined as the maximum temperature maintenance duration.

[0123] Specifically, see Figure 4 As shown, it is a logic determination diagram for determining whether a corresponding battery has a potential abnormality according to an embodiment of the present invention. The process of determining whether a corresponding battery has a potential abnormality includes:

[0124] If the battery stability deviation value is not within the stability error range, it is determined that the corresponding battery has a potential abnormality;

[0125] The stability error range is determined by obtaining a maximum value of the battery stability deviation and a minimum value of the battery stability deviation in the battery stability deviation value history data.

[0126] In this embodiment, the purpose of setting the stability error range is to determine a boundary to divide the battery stability error allowance, which is a closed interval, the lower limit of the interval is the minimum value of the battery stability deviation, and the upper limit of the interval is the maximum value of the battery stability deviation.

[0127] Specifically, the present invention analyzes the twin model data set based on a pre-built evaluation model to determine whether the battery has potential abnormalities. The duration of battery temperature rise and the corresponding maximum temperature maintenance time can reflect the heat dissipation performance of the battery thermal management system, which affects the performance and life of the battery. By matching the real-time data and historical conditions of the above two characteristics, the battery stability deviation value is determined to characterize the degree of deviation between the actual situation and the battery performance evaluated by the evaluation model, and then matched with the stability error range to determine whether the corresponding battery has potential abnormalities, thereby improving the accuracy of battery abnormality evaluation.

[0128] Specifically, it also provides a system that applies a new energy vehicle digital twin modeling method based on multi-source data fusion, including:

[0129] A battery collection module is used to collect and store the real-time battery stability characteristics corresponding to the target vehicle in real time;

[0130] An energy consumption analysis module is used to obtain the operating condition data of the target vehicle, extract the corresponding driving inertia characteristics, determine the energy consumption characterization value of the target vehicle, and mark the target vehicle;

[0131] A data update module is connected to the battery acquisition module and the energy consumption analysis module respectively, and updates the twin model data set of the battery corresponding to the target vehicle in response to the marking result, including:

[0132] Retrieving the motor system's sound frequency data to determine the abnormal sound time segment, obtaining driving operation characteristics within the abnormal sound time segment, and combining them with the auxiliary system's usage duration to evaluate battery consumption interference characterization parameters. Based on the consumption interference characterization parameters, the real-time battery stability characteristics within the abnormal sound time segment are filtered out and stored in the twin model dataset;

[0133] An analysis and evaluation module, which, together with the data update module, is used to analyze the twin model dataset based on a pre-built evaluation model to determine potential anomalies;

[0134] The evaluation model is used to compare the real-time battery stability characteristics in the twin model data set with the historical battery stability characteristics, determine the battery stability deviation value, and match the battery stability deviation value with the stability error range to determine whether the corresponding battery has potential abnormalities;

[0135] Among them, the driving inertia characteristics include the number of engine shutdowns within a predetermined driving distance and the average time interval from shutdown to restart; the driving operation characteristics include the frequency of accelerator pedal pressing and the average pedaling duration; the real-time battery stability characteristics include the duration of battery heating and the duration of maximum temperature maintenance.

[0136] Specifically, there is no limitation on the specific structure of the battery acquisition module, energy consumption analysis module, data update module and analysis and evaluation module. They themselves or each unit therein can be composed of logic components or a combination of logic components, and the logic components include field programmable processors, computers or microprocessors in computers.

[0137] Thus far, the technical solutions of the present invention have been described in conjunction with the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art may make equivalent changes or substitutions to the relevant technical features, and the technical solutions after such changes or substitutions will fall within the scope of protection of the present invention.

Claims

1. A new energy vehicle digital twin modeling method based on multi-source data fusion, characterized in that: include: Collect and store the real-time battery stability characteristics corresponding to the target vehicle in real time; Obtaining the operating condition data of the target vehicle to extract the corresponding driving inertia characteristics, determining the energy consumption characterization value of the target vehicle, and marking the target vehicle; In response to the marking result, the twin model dataset of the battery corresponding to the target vehicle is updated. include, Retrieving the motor system's sound frequency data to determine the abnormal sound time segment, obtaining driving operation characteristics within the abnormal sound time segment, and combining them with the auxiliary system's usage duration to evaluate battery consumption interference characterization parameters. Based on the consumption interference characterization parameters, the real-time battery stability characteristics within the abnormal sound time segment are filtered out and stored in the twin model dataset; Analyze the twin model dataset based on the pre-built evaluation model to identify potential anomalies; The evaluation model is used to compare the real-time battery stability characteristics in the twin model data set with the historical battery stability characteristics, determine the battery stability deviation value, and match the battery stability deviation value with the stability error range to determine whether the corresponding battery has potential abnormalities; Among them, the driving inertia characteristics include the number of engine shutdowns within a predetermined driving distance and the average time interval from shutdown to restart; the driving operation characteristics include the frequency of accelerator pedal pressing and the average pedaling duration; the real-time battery stability characteristics include the duration of battery heating and the duration of maximum temperature maintenance.

2. The new energy vehicle digital twin modeling method based on multi-source data fusion according to claim 1 is characterized in that: The process of determining the energy consumption characterization value of the target vehicle includes: The ratio of the number of stalls within the predetermined driving distance to the stall threshold is used as a first energy consumption characteristic; taking a ratio of an average of time intervals from engine shutdown to engine restart within the predetermined driving distance to a threshold of an average of time intervals as a second energy consumption characteristic; The sum of the first energy consumption characteristic and the second energy consumption characteristic is determined as the energy consumption characterization value.

3. The new energy vehicle digital twin modeling method based on multi-source data fusion according to claim 2 is characterized in that: Marking the target vehicle, include, If the energy consumption characterization value of the target vehicle is greater than or equal to the energy consumption characterization threshold, the target vehicle is marked as a high energy consumption vehicle.

4. The new energy vehicle digital twin modeling method based on multi-source data fusion according to claim 3 is characterized in that: Responding to the marked results, including, If the target vehicle is marked as a high-energy consumption vehicle, the twin model dataset of the battery corresponding to the target vehicle is updated.

5. The new energy vehicle digital twin modeling method based on multi-source data fusion according to claim 1 is characterized in that: The process of determining the time domain segment of abnormal noise includes: Based on the sound frequency data, a time domain curve of the sound frequency of the target vehicle within a predetermined driving distance is constructed; If there is any time domain segment whose corresponding slope is greater than or equal to the slope threshold, the time domain segment is determined as the abnormal sound time domain segment.

6. The new energy vehicle digital twin modeling method based on multi-source data fusion according to claim 1 is characterized in that: The process of evaluating the consumption interference characterization parameters for the battery includes, The sum of the ratio of the accelerator pedal's stepping frequency to the stepping frequency threshold and the ratio of the average stepping duration to the average stepping duration threshold is used as the first consumption interference feature; The ratio of the auxiliary system usage time to the usage time threshold is used as the second consumption interference feature; The first consumption interference feature and the second consumption interference feature are weightedly summed to determine the consumption interference characterization parameter.

7. The new energy vehicle digital twin modeling method based on multi-source data fusion according to claim 6 is characterized in that: The process of screening out the real-time battery stability characteristics within the abnormal noise time domain based on the consumption interference characterization parameters includes: Determine the consumption interference characterization parameters corresponding to each abnormal sound time domain segment; If the consumption interference characterization parameter corresponding to the abnormal sound time domain segment is greater than or equal to a preset consumption interference characterization parameter threshold, the real-time battery stability feature corresponding to the abnormal sound time domain segment is screened out.

8. The new energy vehicle digital twin modeling method based on multi-source data fusion according to claim 1 is characterized in that: The process of determining the battery stability deviation value includes, Recall historical battery stability characteristics, including the average duration of battery temperature rise and the average duration of maximum temperature maintenance; Calculating a first deviation between a duration of battery heating up and an average of the durations of battery heating up; Calculating a second deviation value between the maximum temperature maintenance time and the average of the maximum temperature maintenance time; The sum of the first deviation value and the second deviation value is used as the battery stability deviation value.

9. The new energy vehicle digital twin modeling method based on multi-source data fusion according to claim 8 is characterized in that: The process of determining whether the corresponding battery has potential abnormalities includes: If the battery stability deviation value is not within the stability error range, it is determined that the corresponding battery has a potential abnormality; The stability error range is determined by obtaining a maximum value of the battery stability deviation and a minimum value of the battery stability deviation in the battery stability deviation value history data.

10. A system applying the new energy vehicle digital twin modeling method based on multi-source data fusion according to any one of claims 1 to 9, characterized in that: include: A battery collection module is used to collect and store the real-time battery stability characteristics corresponding to the target vehicle in real time; An energy consumption analysis module is used to obtain the operating condition data of the target vehicle, extract the corresponding driving inertia characteristics, determine the energy consumption characterization value of the target vehicle, and mark the target vehicle; A data update module is connected to the battery acquisition module and the energy consumption analysis module respectively, and updates the twin model data set of the battery corresponding to the target vehicle in response to the marking result, including: Retrieving the motor system's sound frequency data to determine the abnormal sound time segment, obtaining driving operation characteristics within the abnormal sound time segment, and combining them with the auxiliary system's usage duration to evaluate battery consumption interference characterization parameters. Based on the consumption interference characterization parameters, the real-time battery stability characteristics within the abnormal sound time segment are filtered out and stored in the twin model dataset; An analysis and evaluation module, which, together with the data update module, is used to analyze the twin model dataset based on a pre-built evaluation model to determine potential anomalies; The evaluation model is used to compare the real-time battery stability characteristics in the twin model data set with the historical battery stability characteristics, determine the battery stability deviation value, and match the battery stability deviation value with the stability error range to determine whether the corresponding battery has potential abnormalities; Among them, the driving inertia characteristics include the number of engine shutdowns within a predetermined driving distance and the average time interval from shutdown to restart; the driving operation characteristics include the frequency of accelerator pedal pressing and the average pedaling duration; the real-time battery stability characteristics include the duration of battery heating and the duration of maximum temperature maintenance.

Images