Vehicle electrical performance analysis method based on big data

Abstract

The invention relates to the technical field of vehicle electrical system analysis, and particularly discloses a vehicle electrical performance analysis method based on big data, which comprises a data preparation stage, a data preprocessing stage and a data application stage, and is characterized in that based on vehicle big data information, data processing, data application innovation, data information screening and combination are carried out, and vehicle electrical performance analysis is carried out. According to the method, the use frequency of each electric appliance is obtained, optimization of electric balance calculation from estimated value calculation to refined calculation in each subdivision state is achieved, and on the basis that the performance of the whole vehicle is met, the matching problem caused by redundant design of a generator and a storage battery and estimation errors is avoided; through data information cleaning, integration and data graphical design, on-line monitoring, remote diagnosis, abnormity analysis and electric equipment use condition analysis of the vehicle are realized, improvement of product design quality is assisted, cost is saved, and combination of product design and product operation is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle electrical system analysis, and particularly relates to a method for analyzing vehicle electrical performance based on big data. Background Art

[0002] In the industry, the research on electrical balance mainly calculates the dynamic balance of electric energy generation and consumption among generators, storage batteries, and vehicle electrical loads. Its energy flow and equivalent circuit are as shown in Figure 1 、 Figure 2 ; Among them, the calculation of vehicle electrical loads is related to the use and working states of electrical equipment. The working state of electrical equipment is determined by complex situations such as seasons, environments, and traffic conditions. For the convenience of analysis and calculation, the concept of the use frequency coefficient of electrical equipment is currently introduced to roughly estimate the electricity consumption of the whole vehicle. The equivalent load of vehicle electrical appliances can be calculated by P 总 =Σ(P i ×K i ), where P i refers to the weighted power of the i-th electrical appliance; K i refers to the use frequency of the i-th electrical appliance.

[0003] The disadvantages of the above existing technologies are as follows:

[0004] Disadvantage 1: In the current electrical balance calculation, the calculation of vehicle electrical loads is based on the frequency coefficient. The use frequency of electrical loads lacks a reference basis and is roughly evaluated and adjusted by designers according to experience. The theoretical calculation data deviate greatly among different personnel.

[0005] Disadvantage 2: Currently, the equivalent load of vehicle electrical appliances is calculated by P 总 =Σ(P i ×K i ). For a certain electrical appliance, only the power consumption during equipment operation is considered, and the power consumption during equipment standby is not considered, resulting in deviation in calculation accuracy.

[0006] Disadvantage 3: Currently, for the usage conditions of electrical appliances, such as usage duration, usage frequency, etc., methods such as investigation, estimation, and local data analysis are mostly used, which have problems such as long design cycles, low data accuracy, and narrow coverage; moreover, the design of component life is evaluated by estimation, lacking data support, and having defects such as unreasonable design life and excessive redundancy.

[0007] Disadvantage 4: Compared with traditional fault troubleshooting, for troubleshooting bus-related, irregular, and sporadic electrical performance faults, post-event troubleshooting is more difficult, has a long troubleshooting cycle, requires higher personnel capabilities, has low problem handling efficiency, and there is no simple, easy-to-use, and easily recognizable troubleshooting method; for example, for the problem of battery power shortage, battery sensors can be used, but the cost is high, the accuracy is low, and the industry usage rate is low. For vehicles without battery sensors, battery problems are all analyzed retrospectively, and cannot be prevented or reminded in advance. There are few means for analyzing the reasons for actual vehicle power shortage, and the analysis cycle is long and the cost is high.

[0008] Therefore, it is necessary to design a vehicle electrical performance analysis method based on big data to solve the problems of low efficiency, high cost, or waste of long-cycle verification resources in the existing vehicle electrical performance analysis design verification and after-sales fault handling. Summary of the Invention

[0009] Aiming at the problems existing in the prior art, the purpose of the present invention is to provide a vehicle electrical performance analysis method based on big data. Based on vehicle big data information, through data processing, it realizes innovation in the method of data analysis and application, and proposes a method for accurately calculating electrical balance, vehicle online monitoring or remote diagnosis, and a method for analyzing the usage duration / frequency / life / demand of electrical equipment, which can assist in improving product design quality, saving costs, solving the problems of low efficiency, high cost, or waste of long-cycle verification resources in design verification and after-sales fault handling, and realizing the combination of product design and product operation.

[0010] The technical solution adopted by the present invention to solve its technical problems is: A vehicle electrical performance analysis method based on big data, including the following steps:

[0011] S1. Data preparation stage: Determine the vehicle model and signals to be collected, formulate the recognition scheme and feasibility analysis for each signal, and download the vehicle's original data;

[0012] S2. Data preprocessing stage: Formulate a single-day collection and statistics template, determine the screening conditions for signal name, date, season, start time, end time, operating area, day / night, and weather, and complete the summary of the required vehicle data;

[0013] S3. Data application stage: Based on big data information, according to the goals to be achieved, analyze, compare, merge, and calculate a large amount of data, and realize the value breakthrough of data application innovation by presenting the data in a graphical and statistical result manner;

[0014] S4. The applications in step S3 include but are not limited to the application of electrical balance calculation software, the application of the statistical service life of the battery, and vehicle anomaly monitoring.

[0015] Preferably, in the data preparation stage of step S1, it is necessary to clarify the vehicle models to be analyzed, count the number of vehicles, determine the time period for data collection, further determine the signals, configurations, and functional information to be analyzed, sort out the identified message information and the processing scheme for message signals, complete the feasibility analysis, and perform requirement input and download of the original vehicle data.

[0016] Preferably, in the data preprocessing stage of step S2, the data is primarily preprocessed according to the objectives to be achieved. The amount of vehicle data is extremely large and the data information is relatively scattered. To facilitate data analysis, it is necessary to first determine the screening conditions in combination with the signal requirements, formulate a daily collection template, determine the screening conditions for each vehicle to be collected, and complete the data aggregation.

[0017] Preferably, in the application of the electrical balance calculation software in step S4, the big data is combined to preprocess the data. The electrical balance calculation software screens different conditions, thereby calculating the usage frequency coefficients of various electrical appliances under different combined conditions during the day and night in spring and autumn, during the day and night in summer, on rainy summer nights, during the day and night in winter, and on snowy winter nights while driving. This frequency coefficient is the statistical result of the actual operation data, and is more accurate in data precision than the traditional method of estimating the frequency coefficient, providing data support for the accurate calculation of the electrical balance; for the convenient and quick implementation of loading the frequency coefficient, the front-end interface of the electrical balance calculation software is matched and associated with the database through keywords, and the frequency coefficient is automatically loaded into the electrical balance calculation software.

[0018] Preferably, the electrical balance calculation formula in the electrical balance calculation software is: I i =I S ×(1 - K i ) + I W ×K i ,

[0019] In the formula: i represents the average current of the i-th electrical appliance as I i , the standby current is I S , the working current is I W , the frequency coefficient corresponding to the electrical appliance is K i , and the vehicle electrical appliances are equivalent to

[0020] Preferably, in the application of the battery statistical service life in step S4, through the analysis of big data, various sub-conditions are obtained: the usage duration, usage times, open-circuit voltage of the battery, changes in the minimum starting SOC of the battery, and extreme values of the usage of the electrical load under different provinces, sub-markets, weather conditions, and seasons, which are used for scenarios such as product design reference, customer usage habits, and test case analysis; in cooperation with the after-sales system, the service life of components is statistically calculated.

[0021] Preferably, for the vehicle anomaly monitoring in step S4, by processing the original vehicle big data, signals of some corresponding relationships are packaged into modules: vehicle start module, lighting detection module, cruise module, battery charge and discharge module, which can remotely and online reproduce the vehicle operation status and are used in scenarios such as vehicle remote diagnosis, online monitoring, test tracking verification, and anomaly data analysis. At the same time, it can also monitor vehicle anomalies, make early predictions, and give reminders.

[0022] The present invention has the following beneficial effects:

[0023] The vehicle electrical performance analysis method based on big data designed by the present invention obtains the operation data of electrical equipment through big data, can calculate the frequency coefficient of electrical equipment under different conditions, realizes the optimization of the electrical balance calculation from manual estimation to precise calculation relying on data support, reduces the differences caused by human factors, and avoids redundant designs of generators and batteries and matching problems caused by estimation errors.

[0024] The vehicle electrical performance analysis method based on big data designed by the present invention can obtain data such as the usage duration / number of times of electrical equipment, differential configuration requirements, fan fault analysis, and battery power shortage analysis through the application innovation of vehicle big data, which is used for product design and selection to improve the product design quality and analyze customers' differential requirements.

[0025] The vehicle electrical performance analysis method based on big data designed by the present invention realizes the reproduction of the vehicle operation status in the form of charts through the application innovation of vehicle big data, which is used for remote diagnosis, reduces the difficulty of electrical system maintenance in the production and after-sales links, reduces the maintenance cost, improves the vehicle operation efficiency, and at the same time can also monitor vehicle anomalies, make early predictions and give reminders, improving the customer's vehicle use experience and vehicle attendance rate. Description of the Drawings

[0026] Figure 1 It is a schematic diagram of the energy flow direction of the vehicle in different states.

[0027] Figure 2 It is an equivalent circuit diagram among the generator, battery, and electrical equipment.

[0028] Figure 3 It is a flowchart of the vehicle original data download.

[0029] Figure 4 It is a flowchart of data preprocessing.

[0030] Figure 5 It is a bar chart for the analysis of the vehicle cruise usage in some provinces.

[0031] Figure 6 It is a statistical trend chart of the vehicle operation data. Detailed implementation manners

[0032] The following will further describe in detail the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts shall fall within the protection scope of the present invention.

[0033] Embodiment 1:

[0034] A vehicle electrical performance analysis method based on big data includes the following steps:

[0035] S1. Data preparation stage: Determine the vehicle model and signals to be collected, formulate the identification scheme and feasibility analysis for each signal, and download the original vehicle data;

[0036] S2. Data preprocessing stage: Formulate a single-day collection and statistics template, determine the screening conditions for signal name, date, season, start time, end time, operating area, day / night, weather, and complete the summary of the required vehicle data;

[0037] S3. Data application stage: Based on the big data information, according to the goals to be achieved, analyze, compare, merge, and calculate a large amount of data, and realize the value breakthrough of data application innovation in the form of graphing and statistical results;

[0038] S4. The applications in step S3 include but are not limited to the application of electrical balance calculation software, the application of the statistical service life of the battery, and vehicle abnormal monitoring.

[0039] Embodiment 2: Specific implementation technical solutions combining the method of Embodiment 1.

[0040] The present invention is mainly based on vehicle big data information, and through data processing, realizes data application innovation; meets multi-condition screening, calculates the usage frequency of electrical appliances under different conditions such as sub-markets, provinces, seasons, weather, etc., realizes the optimization of electrical balance calculation from estimated calculation to refined calculation under each sub-condition, and on the basis of meeting the vehicle performance, avoids redundant design of generators and batteries and matching problems caused by estimation errors; through cleaning, integrating big data information and data graphing design, constructs modules such as vehicle startup, light detection, cruise, battery charge and discharge, etc., and can realize the analysis of vehicle online monitoring, remote diagnosis, abnormal analysis, usage of electrical equipment, etc., and maximize the value application of big data.

[0041] The present invention can be divided into three major steps: data preparation, data preprocessing, and data application.

[0042] In the data preparation stage, such as Figure 3As shown, it is necessary to clarify the vehicle models to be analyzed, count the number of vehicles, and determine the time period for data collection. Further, it is necessary to determine the information on signals, configurations, and functions to be analyzed, sort out the identified message information and the processing scheme for message signals, complete the feasibility analysis, and perform requirement input and download of vehicle original data.

[0043] In the data preprocessing stage, as Figure 4 shown, the data is mainly preprocessed according to the objectives to be achieved, so as to improve the efficiency of data analysis. Taking the data of a certain vehicle model as an example, the data volume of a single vehicle in one year is 90G. The data volume of vehicles is very large and the data information is relatively scattered. In order to facilitate data analysis, it is necessary to first combine the signal requirements to formulate a daily collection template, clarify rules such as the operation area, day / night, and weather, and complete the statistical results of each selected vehicle every day. Since the daily data is still relatively scattered, it is necessary to formulate a unified summary template to divide the data according to screening conditions such as provinces, sub-markets, seasons, day, night, sunny days, rain and snow, etc., complete data summarization, and achieve data preprocessing.

[0044] In the data application stage, based on big data information, according to the objectives to be achieved, through the methods of analyzing, comparing, merging, and calculating a large amount of data, the data is presented in the form of graphics and statistical results to achieve a value breakthrough in data application innovation.

[0045] Application 1: Combine big data to preprocess data, screen different conditions through software, and according to the formula: Ki = T W / TON Calculate the usage frequency coefficients of various electrical appliances during driving under different combined conditions such as spring and autumn during the day, spring and autumn at night, summer during the day, summer at night, summer rainy night, winter during the day, winter at night, and winter snowy night. Among them, TW refers to the equipment working duration, and TON refers to the total vehicle power-on duration. The data selected for calculation is the effective data within the range of 2% - 98% of the selected duration to avoid abnormal data. This frequency coefficient is the statistical result of actual operation data, and the data accuracy is more accurate than the traditional method of estimating the frequency coefficient, providing data support for the accurate calculation of the electrical balance. To facilitate and quickly load the frequency coefficient, the front-end interface of the software is matched and associated with the database, and the frequency coefficient is automatically loaded into the electrical balance calculation software.

[0046] Technical solution 1 corresponding to Application 1: According to the actual vehicle test data, non-switching electrical equipment will also consume power in the standby state. The power consumption of a single electrical equipment has a small impact on the overall vehicle electrical balance. However, with the development of vehicle intelligence, there are more and more vehicle controllers and sensors on the vehicle, and their standby power consumption will have a prominent impact on the overall vehicle electrical balance. Therefore, the electrical balance calculation formula can be improved to: I i = I S ×(1 - K i ) + I W ×Ki (where \(i\) represents that the average current of the \(i\)-th electrical device is \(I\) i , the standby current is \(I\) S , the operating current is \(I\) W , and the frequency coefficient corresponding to the electrical device is \(K\) i ), the vehicle electrical appliances are equivalent to Regarding the frequency coefficient \(K\) in the formula i , the present invention analyzes the operation data of the vehicle through big data, obtains the usage frequencies of various electrical loads, provides a more practical and higher-precision frequency coefficient for the electric balance calculation, thereby improving the accuracy of the electric balance calculation and avoiding calculation differences caused by human factors.

[0047] Application 2: By analyzing big data, obtain the usage duration, usage times, open-circuit voltage of the battery, changes in the minimum starting SOC of the battery, and extreme values of the electrical load usage under various subdivision conditions (such as provinces, sub-markets, weather, seasons, etc.), for scenarios of product design reference, customer usage habits, and test case analysis; cooperate with the after-sales system to count the service life of parts, and assist in improving product design quality, saving costs, product promotion, and market positioning.

[0048] Technical solution 2 corresponding to Application 2: By cleaning big data information and combining software data processing, convert a large amount of data streams into easy-to-recognize statistical values or chart formats, and be able to directly obtain information such as cruise usage rate, single-day lighting-on duration, and minimum starting voltage, so that the data generates value. Taking the analysis of cruise usage in some provinces as an example, through cleaning the original data, screening out the daily cruise usage data of each vehicle, and forming summary information, use the pandas library in python to extract the three columns of data of "vehicle VIN", "province", and "cruise usage" that need to be used from the summary data, screen the cruise for each "province", remove duplicate values, increase the cleaning of abnormal data, obtain the total number of vehicles in each province and the number of vehicles using cruise through a for loop, thereby calculating the cruise usage proportion in some provinces, and use matplotlib to realize the charting of the data. The statistical example is shown in Figure 5 .

[0049] Application 3: By processing the original vehicle big data, pack some signals with corresponding relationships into modules, such as vehicle start module, lighting detection module, cruise module, battery charge and discharge module, etc., which can remotely and online or reproduce the vehicle operation state, and are used in scenarios such as vehicle remote diagnosis, online monitoring, test tracking verification, and abnormal data analysis, reduce the difficulty of electrical system maintenance in the production and after-sales links, greatly improve the maintenance efficiency, reduce the maintenance cost, and at the same time can also realize the abnormal monitoring of the vehicle, achieve early prediction and reminder, and improve the customer's vehicle usage experience and vehicle attendance rate.

[0050] Technical solution three corresponding to Application 3: The original vehicle data is processed by software and presented in the form of charts to display vehicle operation data for analyzing driving behaviors and analyzing / comparing abnormal vehicle data. The design solution is to determine the main observation signals and auxiliary observation signals required for the construction module, determine the analysis conditions for data input, such as data sources, data dates, etc., process the data using the pandas library in Python, extract and merge the required signals according to the input conditions, complete data cleaning, delete useless and duplicate data, sort various signals in sequence according to time, determine the null value filling rule according to the signal type, ensure that each observation signal uses the same time axis, complete data processing, further generate chart data, and the generated chart data needs to be plotted with a unified starting point of the time axis to avoid the chart time axes not being on the same line. Through a large amount of data chart information, filter out the data curves of normal vehicle states, which can be used to identify and analyze abnormal data; it can also be used for remote analysis of vehicle faults or online monitoring, without the need to record messages on-site, improving the development verification efficiency and vehicle repair timeliness rate. The statistical example is shown in Figure 6 .

[0051] Through the application innovation of vehicle big data, this invention charts the vehicle operation data, realizes functions such as vehicle remote diagnosis, online monitoring, test tracking verification, and data anomaly monitoring, innovates the existing traditional analysis means, improves the product design verification and after-sales service efficiency, and reduces the development and maintenance costs. Through the application innovation of vehicle big data, the usage frequency coefficients of electrical equipment under different working conditions and scenarios can be screened out, realizing the optimization of the electrical balance calculation from manual estimation to precise calculation relying on data support. Through the application innovation of vehicle big data, statistical data on the usage duration, usage frequency, and customer demand analysis of vehicle electrical equipment can be obtained to guide design selection and analysis of customer differentiated needs.

[0052] This invention is not limited to the above embodiments. Anyone should know that structural changes made under the inspiration of this invention, as long as they have the same or similar technical solutions as this invention, fall within the protection scope of this invention.

[0053] The technologies, shapes, and structures not detailedly described in this invention are all well-known technologies.

Claims

1. A method for analyzing the electrical performance of vehicles based on big data, characterized in that It includes the following steps: S1. Data preparation stage: Determine the vehicle model and signals to be collected, formulate the recognition scheme and feasibility analysis for each signal, and download the original vehicle data; S2. Data preprocessing stage: Formulate a single-day collection and statistics template, determine the screening conditions for signal name, date, season, start time, end time, operation area, day / night, and weather, and complete the summary of required vehicle data; S3. Data application stage: Based on the information of big data, according to the goals to be achieved, through the analysis, comparison, combination, and calculation of a large amount of data, realize the value breakthrough of data application innovation in the form of graphical and statistical results; S4. The applications in step S3 include but are not limited to the application of the electric balance calculation software, the application of the statistical service life of the battery, and vehicle anomaly monitoring.

2. The vehicle electrical performance analysis method based on big data according to claim 1, characterized in that In the data preparation stage of step S1, it is necessary to clarify the vehicle model to be analyzed, the number of vehicles to be counted, the time period for data collection, further determine the signals, configurations, and functional information to be analyzed, sort out the recognized message information and the processing scheme for the message signals, complete the feasibility analysis, and perform demand input and download of the original vehicle data.

3. The vehicle electrical performance analysis method based on big data according to claim 1, wherein In the data preprocessing stage of step S2, the data is mainly preprocessed according to the goals to be achieved. The amount of vehicle data is very large and the data information is relatively scattered. In order to facilitate data analysis, it is necessary to first determine the screening conditions in combination with the signal requirements, formulate a single-day collection template, determine the screening conditions required for each vehicle to be collected, and complete the data summary.

4. The vehicle electrical performance analysis method based on big data according to claim 1, characterized in that, For the application of the electric balance calculation software in step S4, the big data is combined to preprocess the data. The electric balance calculation software screens different conditions to calculate the usage frequency coefficients of various electrical appliances under different combined conditions during driving in spring and autumn during the day, spring and autumn at night, summer during the day, summer at night, summer rainy night, winter during the day, winter at night, and winter snowy night. This frequency coefficient is the statistical result of the actual operation data, and the data accuracy is more accurate than the traditional method of estimating the frequency coefficient, providing data support for the accurate calculation of the electric balance; for the convenient and fast loading of the frequency coefficient, the front-end interface of the electric balance calculation software is matched and associated with the database through keywords, and the frequency coefficient is automatically loaded into the electric balance calculation software.

5. The method for analyzing the electrical performance of a vehicle based on big data according to claim 4, wherein The electrical balance calculation formula in the electrical balance calculation software is: I i = I S × (1 - K i ) + I W × K i , Where: i represents the average current of the i-th electrical device as I i , the standby current is I S , the operating current is I W , the frequency coefficient corresponding to the electrical device is K i , the vehicle electrical appliances are equivalent to 6. The vehicle electrical performance analysis method based on big data according to claim 1, wherein For the application of the statistical service life of the battery in step S4, through the analysis of big data, obtain the usage duration, usage times, open-circuit voltage of the battery, changes in the minimum starting SOC of the battery, and extreme values of the usage of electrical loads under each sub-condition: province, sub-market, weather, and season, which are used for product design reference, customer usage habits, and test case analysis scenarios; Cooperate with the after-sales system to count the service life of parts.

7. The vehicle electrical performance analysis method based on big data according to claim 1, wherein The vehicle anomaly monitoring in step S4 processes the original vehicle big data and packages signals with corresponding relationships into modules: vehicle start-up module, lighting detection module, cruise module, and battery charge and discharge module. It can remotely monitor online or reproduce the vehicle operation status, and is used in scenarios such as vehicle remote diagnosis, online monitoring, test tracking verification, and anomaly data analysis. At the same time, it can also achieve vehicle anomaly monitoring, with early prediction and reminder.

Images