A new energy vehicle operation state evaluation method based on multi-source perception data

Abstract

The application is suitable for the technical field of new energy vehicle operation control and energy management, and provides a new energy vehicle operation state evaluation method based on multi-source sensing data, which comprises the following steps: when it is judged that the current slope comprehensive index exceeds the reference interval, the upper limit of the regenerative braking power currently adopted by the target vehicle is obtained, and a correction factor is generated according to the difference between the current slope comprehensive index and the slope comprehensive index in the reference interval. The application discloses the implicit coupling law among the slope comprehensive index, the recovery interruption rate and the battery SOC recovery efficiency, and can accurately identify the reference interval which enables the turning change of the energy recovery performance in the sample sequence. Based on the reference interval, the application performs feedforward correction on the upper limit of the regenerative braking power before the vehicle enters a specific road section, so that the vehicle can still maintain better energy recovery continuity and higher SOC recovery efficiency in the case that the ACC throttle cover behavior cannot be avoided.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of new energy vehicle operation control and energy management, and particularly relates to a new energy vehicle operation state evaluation method based on multi-source perception data. BACKGROUND

[0002] New energy vehicles are generally equipped with a regenerative braking system based on a power motor, which can convert part of kinetic energy into electric energy and charge the power battery when the vehicle is sliding or braking. In a typical long downhill road condition, since the regenerative braking is in a long-time and stable operating state, the vehicle can achieve a high energy recovery efficiency, so the existing vehicle control system usually sets a special regenerative braking power upper limit based on the downhill scene to improve the overall recovery effect. On the other hand, automatic driving assistance functions such as ACC constant speed cruise have been widely used in new energy vehicles, which can maintain the vehicle speed through active braking during downhill, thereby further improving the utilization rate of regenerative braking.

[0003] However, the existing technology generally designs energy recovery strategies based on the ideal scene of "typical continuous downhill", without fully considering the complex situation of "long downhill superimposed with multiple small positive slope changes" which is common in actual roads. In such road sections, ACC will repeatedly trigger slight throttle instructions to override regenerative braking in the small slope rising section in order to maintain the target speed, so that the regenerative braking is intermittently interrupted. The existing technology only considers such override braking as normal fluctuations and empirically handles it through simple deduction, threshold punishment and other methods, lacks in-depth analysis of the implicit correlation between the frequency, duration of override braking and the characteristics of the road section, and also fails to consider the aftereffect of such interruption behavior on the battery SOC recovery efficiency after the vehicle completes the road section.

[0004] Further research shows that when the number and amplitude of positive slope changes are low, the slight override braking caused by ACC is less and has limited impact on energy recovery; but when the slope disturbance intensity reaches a certain interval, the override braking will significantly increase, causing the recovery interruption rate to suddenly rise, and the SOC recovery efficiency of the vehicle after completing the road section will also shift from the original stable state to a significant decline. The existing technology lacks means to identify such turning rules from the dimensions of Internet of Vehicles big data, and also lacks a feedforward control strategy to actively adjust the regenerative braking power upper limit based on predictable road conditions before the vehicle enters a specific road section. Therefore, how to identify the possible adverse energy recovery trend before entering such road sections and adjust the controllable parameters in advance through reasonable means to improve the SOC state of the vehicle after passing through the road section has become a technical problem to be solved. SUMMARY

[0005] The present application aims to provide a new energy vehicle operation state evaluation method based on multi-source perception data, which aims to solve the problems raised in the background art.

[0006] The application is implemented by a new energy vehicle running state evaluation method based on multi-source perception data, the method comprises:

[0007] When it is predicted that the target vehicle is in an ACC cruise mode and is about to enter a specific road section with several slope changes, the Internet of Vehicles database is called;

[0008] The slope comprehensive index of the current specific road section is calculated based on the database, and several samples consistent with the vehicle type configuration, running state and specific road section background conditions of the target vehicle but different in slope comprehensive index are obtained from the database;

[0009] The recovery interruption rate of each sample in the specific road section and the battery SOC recovery efficiency after completing the specific road section are calculated;

[0010] In the several samples, whether the recovery interruption rate and the battery SOC recovery efficiency both change from a change amplitude lower than a preset threshold to a change amplitude higher than the preset threshold after the same slope comprehensive index interval as the slope comprehensive index increases is identified, and if so, the interval is determined as a reference interval;

[0011] When it is judged that the current slope comprehensive index exceeds the reference interval, the regenerative braking power upper limit currently adopted by the target vehicle is obtained, and a correction factor is generated according to the difference between the current slope comprehensive index and the slope comprehensive index in the reference interval to feed forward correct the regenerative braking power upper limit.

[0012] As a further limitation of the technical scheme of the embodiment of the application, the calculation process of the slope comprehensive index comprises: obtaining high-precision map information of the specific road section based on the Internet of Vehicles database, and extracting the slope positive change times, positive slope change amplitude and road section length of the specific road section from the high-precision map information to generate the slope comprehensive index.

[0013] As a further limitation of the technical scheme of the embodiment of the application, the consistency of the vehicle type configuration, running state and specific road section background conditions of the target vehicle means that the selected sample and the target vehicle are in a consistent state in terms of vehicle model, vehicle power system configuration, vehicle mass, current driving speed, battery temperature and environmental conditions and road characteristics of the specific road section.

[0014] As a further limitation of the technical scheme of the embodiment of the present application, the calculation process of the recovery interruption rate comprises: obtaining historical driving data of a sample vehicle in a specific road section based on a vehicle networking database, and analyzing the historical driving data to determine a time interval in which the regenerative braking of the vehicle is in a working state; detecting the number of occurrences of regenerative braking interruption events caused by ACC driving requirements and the duration of each interruption event in the time interval, and calculating the recovery interruption rate according to the proportion of the cumulative duration of the interruption events to the total duration of the regenerative braking working time interval.

[0015] As a further limitation of the technical scheme of the embodiment of the present application, the battery SOC recovery efficiency is the ratio between the SOC increment caused by regenerative braking energy recovery in the battery state parameter of the vehicle after completing a specific road section and the amount of energy recovery opportunity of the specific road section, which is used to reflect the energy recovery effect in the specific road section.

[0016] As a further limitation of the technical scheme of the embodiment of the present application, the step of identifying whether the recovery interruption rate and the battery SOC recovery efficiency both change from a change amplitude below a preset threshold to a change amplitude above a preset threshold after a certain slope comprehensive index interval in a plurality of samples, if yes, comprises:

[0017] arranging the plurality of samples in a small-to-large order according to the slope comprehensive index to obtain a sample sequence;

[0018] analyzing the change trend of the recovery interruption rate and the change trend of the battery SOC recovery efficiency corresponding to the sample sequence, and determining whether both of them respectively present rising change and falling change after a certain slope comprehensive index interval, and whether the absolute values of their average change rates both jump from below a preset threshold to above a preset threshold;

[0019] if the above conditions are met, the slope comprehensive index interval is determined as the reference interval.

[0020] As a further limitation of the technical scheme of the embodiment of the present application, when it is determined that the current slope comprehensive index exceeds the reference interval, the step of obtaining the upper limit of the regenerative braking power currently adopted by the target vehicle, and generating a correction factor according to the difference between the current slope comprehensive index and the slope comprehensive index in the reference interval to feed forward correct the upper limit of the regenerative braking power comprises:

[0021] determining whether the slope comprehensive index of the specific road section currently faced by the target vehicle exceeds the reference interval, if yes, obtaining the upper limit of the regenerative braking power currently adopted by the target vehicle;

[0022] An average value of the slope comprehensive index of each sample in the reference interval is calculated, and a correction factor is generated based on the difference between the current slope comprehensive index and the average value, the correction factor is combined with a preset control correction strength coefficient to feed forward correct the regenerative braking power upper limit to obtain a corrected regenerative braking power upper limit;

[0023] The corrected regenerative braking power upper limit is used to update the regenerative braking control strategy to improve the battery SOC recovery efficiency in a specific road section.

[0024] As a further limitation of the technical scheme of the embodiment of the application, when the regenerative braking power upper limit is feed forward corrected, a preset feed forward correction model is used, and the feed forward correction model is:

[0025] ;

[0026] Wherein, refers to the corrected regenerative braking power upper limit, refers to the target automobile currently adopted regenerative braking power upper limit, refers to the maximum value of the regenerative braking power upper limit, refers to the current slope comprehensive index, refers to the average value, refers to the correction factor, refers to the preset control correction strength coefficient.

[0027] As a further limitation of the technical scheme of the embodiment of the application, the preset control correction strength coefficient is updated by an adaptive learning method based on a vehicle networking database, and specifically includes:

[0028] Based on the vehicle networking database, a plurality of correction records of the same specific road section are obtained, and the battery SOC recovery efficiency corresponding to different correction strength coefficients of the vehicle in different correction records is calculated respectively;

[0029] The recovery efficiencies are compared to determine the correction strength coefficient that optimizes the recovery efficiency as the target correction strength coefficient;

[0030] When the target correction strength coefficient and the current correction strength coefficient differ, the current correction strength coefficient is adjusted to the target correction strength coefficient according to a preset approximation strategy to realize adaptive updating of the correction strength coefficient.

[0031] As a further limitation of the technical scheme of the embodiment of the application, the driving energy source of the target automobile and the sample vehicle at least includes a power battery, and has a regenerative braking energy recovery function based on a power motor.

[0032] Compared with the prior art, the present application has the following beneficial effects:

[0033] The application realizes the active optimization of the energy recovery performance of the regenerative braking system by constructing a running state evaluation and feedforward control method for long downhill and multiple slope positive change road sections. First, the application innovatively proposes a "slope comprehensive index" for unified quantification of the slope disturbance intensity of a specific road section, and proposes two core representation indexes of "recovery interruption rate" and "battery SOC recovery efficiency" based on big data analysis of the Internet of Vehicles. Further, the application discloses the implicit coupling law among the slope comprehensive index, the recovery interruption rate and the battery SOC recovery efficiency, which can accurately identify the reference interval that enables the turning change of the energy recovery performance in the sample sequence. Based on the reference interval, the application feeds forwardly corrects the upper limit of the regenerative braking power before the vehicle enters the specific road section, so that the vehicle can still maintain better energy recovery continuity and higher SOC recovery efficiency in the case of inevitable ACC throttle override behavior. At the same time, the control correction strength coefficient is adaptively learned through big data of the Internet of Vehicles, so that the control strategy is continuously optimized. The application can significantly improve the energy recovery performance of the vehicle in complex downhill road conditions, and has good engineering implementability and application prospect. BRIEF DESCRIPTION OF DRAWINGS

[0034] Figure 1 The flowchart of the method provided by the embodiment of the application is shown.

[0035] Figure 2 The flowchart of determining the reference interval in the method provided by the embodiment of the application is shown.

[0036] Figure 3 The flowchart of feedforward correction of the upper limit of the regenerative braking power in the method provided by the embodiment of the application is shown. DETAILED DESCRIPTION

[0037] In order to make the objectives, technical solutions and advantages of the application clearer, the application will be further described in detail below with reference to the drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the application and do not limit the application.

[0038] Figure 1 The flowchart of the method provided by the embodiment of the application is shown.

[0039] Specifically, a new energy vehicle running state evaluation method based on multi-source perception data, the method specifically includes the following steps:

[0040] Step S100, when it is predicted that the target vehicle is in the ACC cruise mode and is about to enter a specific road section with long downhill and several slope changes, the Internet of Vehicles database is called.

[0041] Step S200, based on the database to calculate the current specific road slope comprehensive index, and from the database to obtain the target vehicle with the same type configuration, running state and the specific road background conditions, but the slope comprehensive index of several samples different. The driving energy of the target vehicle and the sample vehicle includes at least power battery and power motor, and has the regenerative braking energy recovery function based on the power motor.

[0042] The calculation process of the slope comprehensive index includes: based on the vehicle networking database to obtain the high-precision map information of the specific road, and extract the slope positive change times, the positive slope change amplitude and the road length of the specific road from the road characteristic parameters, and fuse the road characteristic parameters to generate the slope comprehensive index.

[0043] The specific road background conditions consistent with the target vehicle type configuration, running state and the target vehicle type configuration, running state and the specific road background conditions, specifically, the selected sample and the target vehicle are in consistent state in terms of vehicle model, vehicle power system configuration, vehicle mass, current speed, battery temperature and the specific road environment conditions and road characteristics.

[0044] In the embodiment of the application, the system first needs to predict whether the target vehicle will enter the specific road which meets the research conditions of the application. The prediction is realized based on the map information contained in the vehicle networking database. The vehicle networking database usually stores high-precision map data collected and summarized by multi-source perception devices, including road geometry, longitudinal slope information, road length, historical traffic conditions, etc. The slope distribution curve of the road in front of the target vehicle can be analyzed to identify whether there is a long downhill section and whether the long downhill section contains several times of positive slope change characteristics. The so-called positive slope change refers to the local uphill section or the slope rising section in the downhill trend. When it is identified that the road in front meets the conditions of "long distance downhill" and "contains several times of positive slope change", the system determines that the target vehicle will enter the specific road described in the application.

[0045] The reason for identifying this specific premise scenario is that when a new energy vehicle with energy recovery function is in ACC cruise control mode, it will perform braking action through automatic control during downhill process to maintain the current set speed. This braking action will trigger the operation of the regenerative braking system, thereby realizing electric energy recovery, which is the energy recovery mode commonly adopted by existing vehicles. In a typical long downhill scenario, the regenerative braking system has a high continuous operation time and a large recoverable potential energy, so the vehicle usually sets a specific regenerative braking power upper limit for long downhill scenarios to improve the energy recovery effect of such road sections. It needs to be further explained that the regenerative braking power upper limit is not the larger the better. If the regenerative braking power is set too high, not only will it cause the regenerative braking force to be too strong and affect the smoothness and driving experience of the vehicle, but it may also cause safety hazards of vehicle acceleration and deceleration. At the same time, excessive regenerative braking power is easy to make the motor and battery pack work in a high load state for a long time, affecting their life stability; and when the regenerative power exceeds the instantaneous absorption capacity of the battery, the battery management system will trigger a forced reduction strategy, causing the regenerative power to be forced to decrease, thereby reducing the overall energy recovery effect.

[0046] However, research by those skilled in the art shows that when the long downhill road section contains a large number of slope positive changes, the ACC system will appear light accelerator pedal in some small uphill sections to cover the control behavior of regenerative braking in order to maintain speed, which will cause the energy recovery process to be frequently interrupted. However, further research shows that when the number and amplitude of slope positive changes are in a lower range, the accelerator pedal covering behavior of ACC to maintain speed has limited interference to regenerative braking, so the change of recovery interruption rate is relatively flat; but when the degree of slope positive change reaches a certain threshold interval, ACC will frequently trigger the accelerator pedal to maintain the set speed, and the triggering strength increases, thereby causing regenerative braking to be interrupted multiple times, causing the recovery interruption rate to show a rapid upward trend. More importantly, after the vehicle passes through such a specific road section, the battery SOC state parameter also shows a similar trend, i.e. the change is not obvious in a low intensity case, but the SOC recovery efficiency decreases significantly after exceeding a certain degree. The present application understands this phenomenon as an implicit coupling relationship between the continuity of regenerative braking and the behavior characteristics of the battery, and proposes to adjust the front feed control of the regenerative braking power upper limit when entering such a specific road section, so that the battery SOC state can be kept at a more optimal level after the vehicle passes through such a specific road section.

[0047] The vehicle networking database is a basic data source for realizing the above steps, can be jointly constituted by a vehicle body, road infrastructure and a cloud platform, and is supported by multi-source perception data as a core, and includes vehicle operation data, map and road environment data, power system state data and historical energy recovery behaviors, etc., to provide a comprehensive and reliable data basis for analysis modeling and feedforward control of the application. The database generally includes vehicle operation sampling data, regenerative braking working state information, longitudinal acceleration, brake pedal demand, power motor output, historical road section height and slope data, vehicle position trajectory, environmental temperature, road surface condition, average vehicle speed curve and other data types. These data all belong to information that can be collected, converged and analyzed in real time by the existing vehicle network system, and are mature in technology and widely applied, so it is feasible and reliable to build the data set required by the application based on the database.

[0048] In step S200, it is necessary to calculate the slope comprehensive index of the current specific road section based on the vehicle networking database. The index is used to quantitatively express the slope change characteristics in the specific road section, and provides a unified measurement method for the subsequent comparison basis of the sample judgment. Compared with directly using several single feature parameters, only using the number of positive slope changes or only using the slope change amplitude will lead to insufficient expression ability, because the energy recovery interruption behavior of the vehicle is often determined by multiple factors. The application extracts the number of positive slope changes, the positive slope change amplitude and the road section length and other road section characteristic parameters, and fuses these parameters to generate the slope comprehensive index, so as to completely describe the overall slope disturbance intensity of the specific road section. In this way, not only the quantization accuracy can be improved, but also a unified evaluation standard can be formed among different vehicles, different regions and different road sections, and the comparability in subsequent sample screening and rule analysis can be realized.

[0049] Further, since the slope comprehensive index can objectively quantify the longitudinal disturbance degree caused by the positive change of the slope in the road section, it can be regarded as a key index reflecting the potential strength of the ACC braking being covered by the throttle, and provides a clear basis for the subsequent analysis of the change trend of the recovery interruption rate.

[0050] The purpose of screening samples from the database is to provide reference data with similar background conditions as the target vehicle for subsequent coupling relationship identification. Although the screening conditions proposed in the present application are relatively strict, for example, the vehicle model, power system configuration, vehicle mass, driving speed, battery temperature, and environmental conditions and road characteristics of a specific road section all need to be consistent to ensure the comparability of sample characteristics, but due to the popularity of the current vehicle network system and the large scale of vehicles, a large amount of data can be easily formed through cloud aggregation, so strict screening conditions will not result in a lack of samples. At the same time, it should be noted that the screening of the present application does not necessarily require all conditions to be completely consistent, but requires consistency in key factors affecting energy recovery behavior to ensure the comparability of sample data. For example, the ambient temperature of the road area, the road surface friction condition, the speed limit interval, the lane curvature, and the traffic flow level all belong to the background conditions of a specific road section, which can be included in the screening range according to the actual situation to improve the reliability and stability of the analysis results.

[0051] Further, the new energy vehicle running state evaluation method based on multi-source perception data further comprises the following steps:

[0052] In step S300, the recovery interruption rate of each sample in a specific road section and the battery SOC recovery efficiency after completing the specific road section are calculated.

[0053] The calculation process of the recovery interruption rate includes: obtaining the historical driving data of the sample vehicle in the specific road section based on the vehicle network database, and analyzing the historical driving data to determine the time interval during which the regenerative braking of the vehicle is in the working state; the number of regenerative braking interruption events caused by ACC driving demand and the duration of each interruption event are detected in the time interval, and the recovery interruption rate is calculated according to the proportion of the cumulative duration of the interruption event to the total duration of the regenerative braking working time interval.

[0054] The battery SOC recovery efficiency refers to the ratio between the SOC increment caused by regenerative braking energy recovery in the battery state parameter of the vehicle after completing the specific road section and the energy recovery opportunity amount of the specific road section, which is used to reflect the energy recovery effect in the specific road section.

[0055] In the embodiments of the present application, the recovery interruption rate and the battery SOC recovery efficiency are both quantitative indexes proposed by the present application for the specific working condition of long downhill multi-section positive slope change under ACC operation. Traditional technologies usually only focus on the instantaneous power or total recovered electric quantity of regenerative braking, and cannot directly depict the continuity destruction of regenerative braking caused by ACC active throttle in long downhill road sections, and cannot capture the overall influence of such continuity destruction on the battery SOC behavior. Therefore, the present application proposes the above two types of indexes to solve the problems of the existing evaluation system.

[0056] Among them, the recovery interruption rate is used to quantify the degree of interruption of regenerative braking in the ACC speed maintaining process, and its core technical means includes: based on the vehicle networking database, the longitudinal dynamics data of the sample vehicle in a specific road section is analyzed, the time interval when the regenerative braking is in working state is identified, and the number of occurrence and the duration of the regenerative braking interruption event triggered by the ACC throttle command in the interval are detected. By calculating the ratio of the cumulative duration of all interruption events to the total duration of the regenerative braking working interval, the interruption rate index which objectively reflects the degree of destruction of regenerative braking continuity can be formed. The index is generated based on a large number of real historical data of vehicles, has a clear physical meaning and repeatability, and can scientifically describe the influence degree of ACC control on regenerative braking behavior.

[0057] Correspondingly, the battery SOC recovery efficiency is used to reflect the utilization degree of the effective SOC increment of the battery due to regenerative braking relative to the theoretical energy recovery opportunity provided by the road section itself after the vehicle passes through the specific road section. The index combines the SOC data recorded by the vehicle BMS, the energy data input by the regenerative braking, and the potential energy conditions determined by the slope characteristics of the road section, and can objectively evaluate the overall energy recovery effect of the vehicle in such a complex slope disturbance road section. Since the index takes the actual change of SOC as the result criterion, it can fully reflect the final influence of regenerative braking interruption on energy recovery utilization efficiency, thereby providing a scientific basis for subsequent feedforward correction strategy.

[0058] In summary, the present application creates two new quantitative indicators for describing the energy recovery behavior of ACC downhill working condition by using the big data of vehicle networking, namely the recovery interruption rate and the battery SOC recovery efficiency, so that the energy recovery process of the vehicle when passing through a specific road section with complex slope disturbance characteristics can be evaluated in a quantifiable, comparable and analyzable manner.

[0059] Further, the new energy vehicle running state evaluation method based on multi-source perception data further includes the following steps:

[0060] Step S400, identify whether the recovery interruption rate and the battery SOC recovery efficiency both appear a turning change from a change amplitude lower than a preset threshold to a change amplitude higher than the preset threshold after the same slope comprehensive index interval in the sample, if yes, the interval is determined as the reference interval.

[0061] Specifically, Figure 2 The flow chart for determining the reference interval is shown.

[0062] If yes, the interval is determined as the reference interval, specifically comprising the following steps:

[0063] In step S401, the plurality of samples are arranged according to the slope comprehensive index from small to large to obtain a sample sequence.

[0064] In step S402, the change trend of the recovery interruption rate and the change trend of the battery SOC recovery efficiency corresponding to the sample sequence are analyzed, and it is determined whether both of them show rising change and falling change respectively after a certain slope comprehensive index interval, and the absolute value of their average change rate jumps from below a preset threshold to above the preset threshold.

[0065] In step S403, if the above conditions are met, the slope comprehensive index interval is determined as the reference interval.

[0066] In the embodiment of the application, the purpose of step S400 is to verify whether the implicit coupling rule of "regenerative braking continuity destruction-battery SOC recovery efficiency decline" revealed by the foregoing analysis really exists in the historical samples, and to identify the corresponding slope comprehensive index interval as the basis for subsequent feedforward correction. By sorting the sample data, trend analysis and threshold detection, it can be objectively confirmed whether the recovery interruption rate and the battery SOC recovery efficiency will be simultaneously and significantly changed at a certain intensity of slope disturbance, so as to extract the interval with the most reference value for actual control.

[0067] In step S401, the plurality of samples are arranged according to the slope comprehensive index from small to large to obtain a sample sequence.

[0068] In step S402, the change trend of the recovery interruption rate and the change trend of the battery SOC recovery efficiency corresponding to the sorted sample sequence need to be analyzed. Specifically, since the interruption frequency and the duration of regenerative braking increase with the increase of the slope disturbance intensity, and the battery SOC recovery efficiency decreases with the increase of the slope disturbance intensity, the two indicators show opposite change directions of "increase" and "decrease". At the same time, by calculating the change rate between adjacent samples and taking the average value, it can be judged whether the change amplitude of each indicator increases significantly after a certain slope comprehensive index interval. The absolute value of the average change rate increases from below the preset threshold to above the threshold, indicating that the indicator enters the "sensitive change zone" from the "gentle change zone". The preset threshold can be determined based on statistical methods, for example, using the average value of the change rate of a large number of samples in the gentle change interval plus several times the standard deviation as the limit value, so that it has clear physical meaning and statistical constraints. This step is essentially a more calculable and verifiable mathematical expression of the "change amplitude" described above.

[0069] In step S403, when it is detected that the recovery interruption rate and the battery SOC recovery efficiency both show the "change amplitude jump" after the same slope comprehensive index interval, the slope comprehensive index interval is determined as the reference interval. The existence of the reference interval indicates that when the slope disturbance of the road section reaches this intensity, the regenerative braking interruption behavior will be significantly enhanced and the battery SOC recovery efficiency will be significantly reduced, which is a key road section feature point of the qualitative change of the vehicle energy recovery effect. It should be noted that the reference interval itself should have a relatively narrow span, otherwise it will not accurately represent the sensitivity of the critical change point, and it is not conducive to the pertinence and effectiveness of the subsequent feedforward correction strategy. Therefore, in specific applications, the present application will use the smallest index interval that can stably meet the above change conditions as the reference interval to ensure the accuracy and control significance of the interval.

[0070] Through the above steps, the present application can objectively identify the key slope disturbance range that causes the sharp change of energy recovery performance from a large amount of historical sample data, thereby providing a scientific, reliable and engineering meaningful basis for the subsequent regenerative braking power feedforward correction strategy.

[0071] Further, the new energy vehicle running state evaluation method based on multi-source perception data further includes the following steps:

[0072] In step S500, when it is judged that the current slope comprehensive index exceeds the reference interval, the upper limit of the regenerative braking power currently adopted by the target vehicle is obtained, and a correction factor is generated according to the difference between the current slope comprehensive index and the slope comprehensive index in the reference interval to feedforward correct the upper limit of the regenerative braking power.

[0073] Specifically, Figure 3A flowchart of feedforward correction of the regenerative braking power upper limit is shown.

[0074] When it is determined that the current slope comprehensive index exceeds the reference interval, the regenerative braking power upper limit currently adopted by the target vehicle is obtained, and a correction factor is generated according to the difference between the current slope comprehensive index and the slope comprehensive index in the reference interval to perform feedforward correction on the regenerative braking power upper limit, which specifically includes the following steps:

[0075] In step S501, it is determined whether the slope comprehensive index of the specific road section currently faced by the target vehicle exceeds the reference interval, and if so, the regenerative braking power upper limit currently adopted by the target vehicle is obtained.

[0076] In step S502, the average value of the slope comprehensive index of each sample in the reference interval is calculated, and a correction factor is generated based on the difference between the current slope comprehensive index and the average value. The correction factor is combined with a preset control correction strength coefficient to perform feedforward correction on the regenerative braking power upper limit, and a corrected regenerative braking power upper limit is obtained.

[0077] In step S503, the corrected regenerative braking power upper limit is used to update the regenerative braking control strategy to improve the battery SOC recovery efficiency in the specific road section.

[0078] In the feedforward correction of the regenerative braking power upper limit, a preset feedforward correction model is used, and the feedforward correction model is:

[0079]

[0080] Wherein, represents the corrected regenerative braking power upper limit, represents the regenerative braking power upper limit currently adopted by the target vehicle, represents the maximum acceptable value of the regenerative braking power upper limit, represents the current slope comprehensive index, represents the average value, represents the correction factor, represents the preset control correction strength coefficient.

[0081] The preset control correction strength coefficient is updated by an adaptive learning method based on a vehicle networking database, which specifically includes:

[0082] Based on the vehicle networking database, a number of correction records of the same type of specific road sections are obtained, and the battery SOC recovery efficiency corresponding to different correction strength coefficients of vehicles in different correction records is calculated.

[0083] The recovery efficiencies are compared to determine the correction strength coefficient that optimizes the recovery efficiency as the target correction strength coefficient. ​

[0084] When there is a difference between the target correction intensity coefficient and the current correction intensity coefficient, the current correction intensity coefficient is adjusted to the target correction intensity coefficient according to a preset approximation strategy, so as to realize adaptive updating of the correction intensity coefficient.

[0085] In the embodiment of the present application, the core purpose of step S500 is to perform feedforward control adjustment based on the identified reference interval before the vehicle enters the specific road section, so as to optimize the performance of the regenerative braking system in the future specific road section in advance, so that the vehicle can obtain more optimal battery SOC recovery efficiency after completing the specific road section. Compared with the traditional passive response mechanism relying only on the downhill control strategy, the present application identifies the road section characteristics that will cause the continuity of energy recovery to be destroyed in advance, and actively improves the upper limit of regenerative braking power before entering, so that the vehicle can improve the energy recovery amount as much as possible in the “recoverable part”, thereby compensating for the recovery interruption loss caused by slope disturbance, and realizing a more scientific and engineering feasible “feedforward correction”.

[0086] In step S501, it is necessary to first judge whether the slope comprehensive index of the specific road section that the current target vehicle faces exceeds the reference interval. When the slope comprehensive index is within the reference interval, the slope disturbance intensity faced by the vehicle is still in the “smooth change zone”, and the recovery interruption rate and the battery SOC recovery efficiency do not appear significant mutations, so no additional correction is needed; when the slope comprehensive index exceeds the reference interval, it means that the vehicle will face strong slope disturbance characteristics, and the ACC control will frequently trigger the throttle override braking, thereby causing the recovery interruption rate to increase sharply and ultimately leading to a significant decrease in the battery SOC recovery efficiency, so the current regenerative braking power upper limit needs to be obtained as a reference value for subsequent correction operations.

[0087] In step S502, the average value of the slope comprehensive index of each sample in the reference interval needs to be calculated, and a correction factor is generated based on the difference of the current slope comprehensive index compared with the average value. The difference can be used as an effective correction basis because the core theoretical basis of the present application is that the root cause of the sudden change of the recovery interruption rate and the decrease of the SOC recovery efficiency is the slope disturbance intensity itself, and the slope comprehensive index has objectively quantified this disturbance intensity. When the current slope comprehensive index is significantly greater than the average value of the reference interval, it means that the road has reached the risk interval of “recovery continuity destruction”. By combining the difference with the control correction intensity coefficient, a feedforward adjustment amount proportional to the slope disturbance degree can be formed, so that the upper limit of the regenerative braking power is accurately improved before entering the road section, and the pre-compensation of energy recovery loss is realized.

[0088] The mechanism essentially builds a closed-loop logical chain of "road section disturbance → recovery interruption → SOC loss → power upper limit compensation", forming a complete explainable causal relationship between road characteristics, battery behavior and control variables. Therefore, the correction factor used in this step not only has a clear engineering significance, but also has scientificity and rationality in physical performance.

[0089] In step S503, the corrected regenerative braking power upper limit is applied to the regenerative braking control strategy of the vehicle, so that it can recover electric energy at a higher power level in the subsequent descending working condition, thereby obtaining more effective energy, so that the vehicle can have a more optimal battery SOC state after completing the specific road section. This step ensures that the correction result falls to the vehicle control execution layer after feedforward prediction and correction, so that the entire control chain is closed.

[0090] The application also provides an adaptive learning mechanism for controlling the correction intensity coefficient. The correction intensity coefficient determines the strength of the feedforward adjustment of the application and is an important parameter of the entire feedforward correction model. In order to ensure that the parameter can continuously adapt to different vehicle models, different weather, different driving habits, and different battery attenuation degrees, the application proposes an adaptive updating mechanism based on a vehicle networking database. By analyzing the correction records of a plurality of historical specific road sections of the same type, the battery SOC recovery efficiency corresponding to different correction intensity coefficients is calculated, and the correction intensity coefficient that can obtain the best recovery efficiency is selected as the target value. The current coefficient is gradually adjusted to the target value according to the approximation strategy, so that the coefficient is continuously optimized in long-term use. This updating method can effectively improve the long-term adaptability and generalization ability of the model, so that the vehicle can maintain good recovery performance in different use scenarios.

[0091] The formula of the feedforward correction model generates a correction amount in a linear manner by combining the current regenerative braking power upper limit, the slope comprehensive index difference, and the correction intensity coefficient, and simultaneously limits the maximum available value of the regenerative braking power, so that the correction result has intuitive understandability and engineering safety boundary. In addition, the application is not limited to using a linear model, but can also use an exponential, polynomial, or regression model trained based on machine learning, as long as it can reasonably express the relationship between the slope disturbance degree and the power upper limit correction amount.

[0092] The following examples illustrate the effect of the entire implementation process:

[0093] Assumptions: the current slope comprehensive index is 0.45; the average index of the samples in the reference interval is 0.30; the current regenerative braking power upper limit is 20 kW; the correction intensity coefficient is initially set to 0.6; and the maximum regenerative braking power allowed value is 25 kW.

[0094] First, the difference is calculated as the current index minus the average index, resulting in 0.15. The difference is then divided by the average index and multiplied by the correction intensity coefficient to obtain a correction factor of 0.09. The correction factor is then increased by 1 and multiplied by the original power limit to obtain a corrected power limit of approximately 21.8 kilowatts. Since this value is lower than the maximum acceptable value, it is used directly.

[0095] In actual descent sections, due to the increased power limit, the vehicle's regenerative braking system was able to recover more energy during permissible periods to offset the energy loss caused by throttle coverage interruptions. Ultimately, the battery SOC recovery efficiency measured after the vehicle completed the section was approximately 5% to 8% higher than without correction, demonstrating the effectiveness of this feedforward correction mechanism.

[0096] The overall beneficial effects of this invention are reflected in the following aspects: This invention can accurately identify key road segment feature points that cause damage to the continuity of regenerative braking and establish a direct coupling relationship with the phenomenon of decreased battery SOC recovery efficiency; through a feedforward correction method, the vehicle can actively adjust its energy recovery capability before entering a specific road segment, effectively improving the battery SOC state after completing the road segment and compensating for the energy loss caused by the near-forced throttle coverage behavior; through an adaptive learning mechanism, this invention can continuously optimize control parameters to adapt to different vehicle and environmental changes; this invention can be implemented entirely based on existing vehicle network databases, map data, and vehicle control interfaces, with low implementation costs and high engineering feasibility.

[0097] This invention has broad application prospects. As the frequency of use of new energy vehicles on highways, mountain roads, and long-distance commuting scenarios continues to increase, the probability of vehicles encountering complex gradient road conditions has significantly increased. This invention can be directly deployed in the vehicle's infotainment system to predict upcoming specific road conditions in real time and proactively optimize energy recovery, which has significant value in improving vehicle range, battery life, and driving economy. Simultaneously, it can also provide key data support for intelligent cruise control systems, adaptive driving systems, and future cloud-based collaborative vehicle control, making it a highly promising foundational technology in the field of intelligent electric vehicle energy management.

[0098] It should be understood that although the steps in the flowcharts of the various embodiments of the present invention are shown sequentially according to the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the various embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0099] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. The program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, storage, databases, or other media used in the embodiments provided in this application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), dual data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link DRAM (SLDRAM), Rambus direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0100] The technical features of the above embodiments can be combined in any way. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0101] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

[0102] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for assessing the operating status of new energy vehicles based on multi-source sensing data, characterized in that, The method includes: When it is predicted that the target vehicle is in ACC cruise control mode and is about to enter a specific road section with a long downhill slope and several slope changes, the vehicle network database is retrieved. The slope comprehensive index is calculated based on the database for a specific road segment, and several samples with the same vehicle type, operating status, and specific road segment background conditions as the target vehicle are obtained from the database, but with different slope comprehensive indexes. The calculation process of the slope comprehensive index includes: obtaining high-precision map information of a specific road segment based on the vehicle network database, extracting road segment feature parameters such as the number of positive slope changes, the magnitude of positive slope changes, and the length of the road segment, and fusing the road segment feature parameters to generate the slope comprehensive index. Calculate the recycling interruption rate of each sample within a specific road segment and the battery SOC recovery efficiency after completing the specific road segment; In several samples, as the comprehensive slope index increases, whether the recycling interruption rate and battery SOC recovery efficiency both show a change in magnitude from below the preset threshold to above the preset threshold after the same comprehensive slope index range. If so, the range is determined as the reference range. When the current comprehensive slope index exceeds the reference range, the upper limit of the regenerative braking power currently used by the target vehicle is obtained, and a correction factor is generated based on the difference between the current comprehensive slope index and the comprehensive slope index in the reference range to perform feedforward correction on the upper limit of the regenerative braking power.

2. The method for assessing the operating status of new energy vehicles based on multi-source sensing data according to claim 1, characterized in that, The consistency with the target vehicle's vehicle model, operating status, and specific road conditions refers specifically to the fact that the selected sample is consistent with the target vehicle in terms of vehicle model, vehicle powertrain configuration, vehicle weight, current driving speed, battery temperature, and the environmental conditions and road characteristics of the specific road section.

3. The method for assessing the operating status of new energy vehicles based on multi-source sensing data according to claim 1, characterized in that, The calculation process of the regenerative braking interruption rate includes: obtaining historical driving data of sample vehicles in a specific road segment based on the vehicle network database, and parsing the historical driving data to determine the time interval in which the vehicle's regenerative braking is in operation; detecting the number of regenerative braking interruption events caused by ACC driving demand and the duration of each interruption event within the time interval, and calculating the regenerative braking interruption rate based on the proportion of the cumulative duration of the interruption events to the total duration of the time interval.

4. The method for assessing the operating status of new energy vehicles based on multi-source sensing data according to claim 1, characterized in that, The battery SOC recovery efficiency refers to the ratio between the SOC increment caused by regenerative braking energy recovery in the battery state parameters of a vehicle after completing a specific road segment and the energy recovery opportunity of that specific road segment, which is used to reflect the energy recovery effect within that specific road segment.

5. The method for assessing the operating status of new energy vehicles based on multi-source sensing data according to claim 1, characterized in that, The steps for identifying whether, in several samples, as the comprehensive slope index increases, the recycling interruption rate and battery SOC recovery efficiency both exhibit a transition from below a preset threshold to above a preset threshold after falling within the same comprehensive slope index range, and if so, determining this range as the reference range, include: A sample sequence is obtained by arranging several samples in ascending order of comprehensive slope index. Analyze the trend of recycling interruption rate and battery SOC recovery efficiency corresponding to the sample sequence, and determine whether both show an upward and downward change respectively after a certain slope comprehensive index range, and whether the absolute value of their average change rate jumps from below the preset threshold to above the preset threshold. If the above conditions are met, then the range of comprehensive slope indexes shall be determined as the reference range.

6. The method for assessing the operating status of new energy vehicles based on multi-source sensing data according to claim 5, characterized in that, When the current comprehensive slope index exceeds the reference range, the steps of obtaining the upper limit of regenerative braking power currently used by the target vehicle, and generating a correction factor based on the difference between the current comprehensive slope index and the comprehensive slope index within the reference range to perform feedforward correction on the upper limit of regenerative braking power include: Determine whether the comprehensive gradient index of the specific road segment currently faced by the target vehicle exceeds the reference range. If so, obtain the upper limit of the regenerative braking power currently used by the target vehicle. Calculate the average value of the comprehensive slope index of each sample in the reference interval, and generate a correction factor based on the difference between the current comprehensive slope index and the average value. Combine the correction factor with the preset control correction intensity coefficient to perform feedforward correction on the upper limit of regenerative braking power, and obtain the corrected upper limit of regenerative braking power. The revised regenerative braking power limit will be used to update the regenerative braking control strategy to improve battery SOC recovery efficiency in specific road sections.

7. The method for assessing the operating status of new energy vehicles based on multi-source sensing data according to claim 6, characterized in that, When performing feedforward correction on the upper limit of regenerative braking power, a preset feedforward correction model is adopted, which is as follows: ； in, This refers to the revised upper limit of regenerative braking power. This refers to the upper limit of regenerative braking power currently used in the target vehicle. This refers to the maximum possible value of the regenerative braking power. This refers to the current comprehensive slope index. This refers to the average value. This refers to the correction factor. This refers to the preset control correction strength coefficient.

8. The method for assessing the operating status of new energy vehicles based on multi-source sensing data according to claim 7, characterized in that, The preset control correction intensity coefficient is updated using an adaptive learning method based on the vehicle network database, specifically including: Based on the vehicle network database, several historical correction records of specific road sections of the same type were obtained, and the battery SOC recovery efficiency corresponding to different correction intensity coefficients of the vehicle in different correction records was calculated respectively. Compare the various recovery efficiencies and determine the correction intensity coefficient that optimizes the recovery efficiency as the target correction intensity coefficient; When there is a difference between the target correction intensity coefficient and the current correction intensity coefficient, the current correction intensity coefficient is adjusted to the target correction intensity coefficient according to a preset approximation strategy to achieve adaptive updating of the correction intensity coefficient.

9. The method for assessing the operating status of new energy vehicles based on multi-source sensing data according to any one of claims 1-8, characterized in that, The target vehicle and the sample vehicle are powered by at least a power battery and have a regenerative braking energy recovery function based on a power motor.

Images