Gearbox gear transmission efficiency optimization method and system based on machine learning

Abstract

The invention discloses a gearbox gear transmission efficiency optimization method and system based on machine learning, and relates to the technical field of automobile gearboxes. Comprising the following steps: driving optimization: continuously learning and optimizing the transmission efficiency of a gearbox gear through a transmission optimization method based on target information and image information. According to the method, two methods for providing driving habits for the driver are provided through the set demand determination stage and the set matching output stage, whether the driver is replaced or not is judged, the corresponding driving habits are provided for the replaced driver, and the two methods supplement each other, so that the driving feeling of a new driver is improved, and the driving experience of the new driver is improved. Through a set transmission optimization method, driving habits are continuously learned and optimized according to target information and image information, and self-adaptive adjustment is performed on a gear shifting decision according to a vehicle following distance and traffic information, so that the transmission efficiency of a gearbox gear is improved, the fuel economy is improved, pause is reduced, and the driving feeling of a driver is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of automotive gearboxes, and specifically provides an optimization method and system for the transmission efficiency of gearbox gears based on machine learning. Background Technique

[0002] The gearbox is an important component in vehicles or mechanical equipment, mainly used to adjust the rotational speed and torque output by the engine to adapt to different driving or working conditions. There are various types of gearboxes, specifically including manual gearboxes, automatic gearboxes, continuously variable transmissions, and dual-clutch gearboxes. As the core component of the vehicle's power transmission system, its performance directly affects the driving experience and fuel economy of the vehicle.

[0003] The transmission efficiency optimization method and device with the patent publication number CN112380673A. The transmission efficiency optimization method includes: determining each component related to the drag torque of the transmission; classifying each component and obtaining the expression of the resistance torque of each component; converting and reducing the resistance torque of each component to the input end of the transmission through the relationship of the gear ratio; obtaining the theoretical value of the drag torque of the transmission, comparing the theoretical value with the drag torque of the transmission obtained through actual experimental tests to verify the theoretical value of the drag torque of the transmission; obtaining the main influencing factors of the drag torque of the transmission, and obtaining the improved drag torque of the transmission; predicting the transmission efficiency according to the improved drag torque of the transmission and comparing it with the measured value of the transmission efficiency to optimize the transmission efficiency, so that the deviation between the predicted value and the measured value of the transmission efficiency is less than the deviation threshold. This application can quickly decompose the drag torque of the transmission and optimize the transmission efficiency.

[0004] Optimizing the transmission efficiency of gearbox gears, that is, optimizing the shifting strategy of the gearbox. The transmission efficiency optimization methods based on the above and similar principles do not fully combine the actual driving situation during the optimization process. They pursue transmission efficiency more than driving feel, resulting in a rather rigid shifting logic and being unable to adaptively adjust according to the actual usage situation. As a result, the jerks during gear shifting of the vehicle may increase, affecting the riding comfort. Moreover, in the existing technology, the shifting strategies of gearboxes mostly have learning behaviors, which optimize the shifting strategy by learning the driving habits of drivers. However, when changing drivers or driving a new vehicle, they are not adapted to this learning method in a short time, which is likely to lead to a reduction in the driving experience. Therefore, the present invention is proposed. Summary of the Invention

[0005] The purpose of the present invention is to provide an optimization method and system for the transmission efficiency of gearbox gears based on machine learning to solve the problems raised in the above background technique.

[0006] To achieve the above purpose, the present invention provides the following technical solution: An optimization method for the transmission efficiency of gearbox gears based on machine learning, the method includes: Driving optimization: Based on the target information and in cooperation with the image information, continuously learn and optimize the transmission efficiency of the gearbox of the vehicle through the transmission optimization method; Information acquisition: Acquire the image information in front of the vehicle, the weight of the main driver, and the previous driving information; Requirement determination: Preset the standard driving preferences, judge the change in the weight of the main driver to obtain a judgment result. When the judgment result indicates that the weight of the main driver has changed, obtain the user requirements, and select the corresponding standard driving preferences based on the user requirements to obtain the target driving habits; Feature establishment: Establish the correlation between the optimized driving habits and the weight of the main driver, and establish a repository for storing the optimized driving habits and the weight of the main driver; Matching output: Monitor the real-time weight of the main driver. When the real-time weight of the main driver changes, select the optimized driving habit corresponding to the real-time weight of the main driver through the matching method based on the repository to obtain the matching driving habit, and push the matching driving habit to the driver; Driving recommendation: Based on the real-time position of the vehicle, recommend the driving habits of the repeated routes to the user through the recommendation method to obtain the recommended driving habits, and push the recommended driving habits to the driver; Information unification: The matching driving habit, the recommended driving habit, and the target driving habit are all target information; Information feedback: Based on the previous driving information, judge the driving habits of the user through the driving judgment method to obtain the user portrait and the improvement method, obtain the feedback address information, and feedback the user portrait and the improvement method to the user based on the feedback address information; Optimization of gear transmission efficiency: By optimizing the shifting operation and driving habits, reduce the slip and impact of the gears during the transmission process, improve the gear meshing efficiency, and thus indirectly achieve the improvement of the transmission efficiency; The transmission optimization method includes: obtaining the real-time position of the vehicle, monitoring whether the shifting decision is executed, obtaining the following distance based on the image information, obtaining the traffic information based on the real-time position of the vehicle, and optimizing the shifting decision through the deep optimization method based on the following distance and traffic information and replacing the shifting decision in the target information to obtain the optimized driving habits.

[0007] Furthermore, the deep optimization method includes: obtaining the shifting strategy based on the target information, presetting the distance threshold and the deviation value, judging whether there are passive shifting factors in the following distance and traffic information to obtain the result information. The passive shifting factors include that the following distance is less than the distance threshold and the red light prohibits passing. When the result information indicates that there are passive shifting factors, during the execution of the shifting decision, reduce the shifting speed based on the shifting strategy in cooperation with the deviation value. When the result information indicates that there are no passive shifting factors, during the execution of the shifting decision, increase the shifting speed based on the shifting strategy in cooperation with the deviation value, and continuously optimize the shifting speed through the exploration learning method and adjust the target information to obtain the optimized driving habits.

[0008] Furthermore, the inquiry learning method includes: presetting a safety distance and an added value. When the following distance is less than or equal to the safety distance, during the process of making a gearshift decision, based on the gearshift strategy and the deviation value, the gearshift speed is accelerated. The driver's feedback on the optimized gearshift decision is obtained to get feedback information. When the feedback information indicates dissatisfaction with the optimized gearshift decision, based on the feedback information and the added value, the deviation value is increased or decreased to obtain an optimized deviation value. The optimized deviation value is used to replace the pre-optimization deviation value to increase or decrease the gearshift speed. When the feedback information indicates satisfaction with the optimized gearshift decision, the gearshift strategy in the target information is correspondingly replaced with the optimized gearshift strategy to obtain an optimized driving habit.

[0009] Furthermore, the recommendation method includes: obtaining the user's route requirements, presetting a target weight, obtaining the vehicle driving route based on the user's route requirements, judging the weight of the vehicle driving route to obtain a specific weight. When the specific weight exceeds the target weight, the optimized gearshift decision during the vehicle driving route is recorded to obtain a target decision. Based on the target decision, the gearshift decision in the target information is replaced to obtain a recommended driving habit, and the recommended driving habit is pushed to the user.

[0010] Furthermore, the driving judgment method: preset a time period, intercept the past driving information based on the time period to obtain paragraph information, extract the number of times of hard braking and hard accelerating in the paragraph information to obtain corresponding numbers, preset a grading standard and a target number. The grading standard includes an aggressive type and an economy type. The aggressive type means that the number of times of hard braking and hard accelerating within the time period exceeds the target number, and the economy type means that the number of times of hard braking and hard accelerating within the time period does not exceed the target number. Based on the relationship between the corresponding number and the target number, a user profile is obtained. When the user profile is of the aggressive type, the corresponding number and the time period are extracted to obtain reference information, the method for improving hard braking and hard accelerating is obtained to get correction information, and the reference information and the correction information are integrated to obtain an improvement method.

[0011] Furthermore, the process of presetting the standard driving preference is: preset a standard gearshift speed. The standard driving preference includes an aggressive preference, a gentle preference, and an economy preference. The gentle preference is obtained based on the standard gearshift speed, the aggressive preference is obtained by increasing the deviation value based on the standard gearshift speed, and the economy preference is obtained by reducing the deviation value based on the standard gearshift speed.

[0012] Furthermore, the matching method includes: presetting a change value, obtaining the real-time weight of the main driver after the change to get a target weight, combining the change value with the target weight to obtain a target weight range, retrieving the main driver weight within the target weight range in the repository based on the target weight range to obtain a selected weight, and extracting the optimized driving habit corresponding to the selected weight in the repository based on the association relationship to obtain a matching driving habit.

[0013] A gear transmission efficiency optimization system based on machine learning uses the above-mentioned gear transmission efficiency optimization method based on machine learning.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: The gear transmission efficiency optimization method and system based on machine learning provide two methods for providing driving habits for drivers through the set demand determination stage and matching output stage. By judging whether the driver is replaced and providing corresponding driving habits for the replaced driver, and the two methods complement each other to facilitate improving the driving experience of new drivers. Through the set transmission optimization method, continuously learn and optimize driving habits according to target information and image information, and adaptively adjust the shift decision according to the following distance and traffic information to improve the transmission efficiency of the gearbox gears, thereby improving fuel economy and reducing jerks, and thus improving the driving experience of the driver.

[0015] At the same time, obtain the user portrait and improvement methods by judging the driving habits of the driver through the driving judgment method according to the past driving information. By feeding back the user portrait and improvement methods to the driver, it is convenient for the driver to understand their own driving habits, and it is also convenient for the driver to refine their driving habits, improve fuel economy. In the exploration of learning methods, set a safe distance. When the following distance is less than or equal to the safe distance, the shift speed is increased, that is, the downshift speed is increased to ensure the safety of the driver. By obtaining the driver's feelings about the optimized shift decision to further optimize the shift decision, it is convenient to find the driving habits suitable for the driver to improve fuel economy and the driving experience of the driver.

[0016] At the same time, by presetting the standard driving preferences, it is convenient to provide driving habits for new drivers who have not driven the vehicle according to their needs, avoiding new drivers adopting the driving habits of the previous driver when driving the vehicle, so as to improve the driving experience of the driver. At the same time, it will not interfere with the driving habits of the previous driver, which is beneficial to improving the driving experience of any driver and is convenient to use. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the main process structure of the present invention; Figure 2 It is a schematic diagram of the secondary process structure of the present invention; Figure 3 It is a schematic diagram of the deep optimization method structure of the present invention; Figure 4 It is a schematic diagram of the passive shift factor structure of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts belong to the scope of protection of the present invention.

[0019] Optimizing the gear transmission efficiency can reduce the loss of energy during transmission, thereby improving fuel economy, reducing fuel consumption. An efficient gear transmission can reduce the impact and vibration during meshing, lower the noise level, and improve ride comfort. Optimizing the gear transmission efficiency can reduce the wear of gears and bearings, reduce friction and heat generation, lower the operating temperature of the transmission, extend the service life of the lubricant and the transmission, and improving the transmission efficiency can reduce fuel consumption, thereby reducing exhaust emissions and meeting increasingly stringent environmental protection regulations.

[0020] As Figures 1 - 4 shown, the present invention provides a technical solution: a method for optimizing the gear transmission efficiency of a transmission based on machine learning. The method includes: Driving optimization: Continuously learning and optimizing the gear transmission efficiency of the transmission through a transmission optimization method based on target information and image information. Information acquisition: Acquire the image information in front of the vehicle, the weight of the main driver, and the previous driving information. Requirement determination: Preset standard driving preferences, judge the change in the weight of the main driver to obtain a judgment result. When the judgment result feedbacks that the weight of the main driver has changed, obtain the user's requirements, and select the corresponding standard driving preferences based on the user's requirements to obtain the target driving habit. Feature establishment: Establish the correlation between the optimized driving habit and the weight of the main driver, and establish a repository for storing the optimized driving habit and the weight of the main driver. Matching output: Monitor the real-time weight of the main driver. When the real-time weight of the main driver changes, select the optimized driving habit corresponding to the real-time weight of the main driver through a matching method based on the repository to obtain the matching driving habit, and push the matching driving habit to the driver. Driving recommendation: Based on the real-time position of the vehicle, recommend the driving habit of the repeated route to the user through a recommendation method to obtain the recommended driving habit, and push the recommended driving habit to the driver. Information unification: The matching driving habit, the recommended driving habit, and the target driving habit are all target information. Information feedback: Judge the user's driving habit through a driving judgment method based on the previous driving information to obtain the user portrait and improvement method, obtain the feedback address information, and feedback the user portrait and improvement method to the user based on the feedback address information. Optimization of gear transmission efficiency: By optimizing gear shifting operations and driving habits, reduce slippage and impact during gear transmission, improve gear meshing efficiency, and thereby indirectly achieve an increase in transmission efficiency; It should be noted that during the driving optimization stage, through the transmission optimization method, the transmission efficiency of the gearbox gears is continuously learned and optimized based on the target information and image information. By optimizing the driving habits, the driving habits are made more compatible with the transmission of the gearbox gears, and faster responses are achieved, such as decelerating in advance and downshifting appropriately, so that the gears mesh in a more ideal state, reducing impact. The slip and impact on the gear engagement surface will cause additional friction and energy loss. By reducing the above-mentioned impact, the transmission efficiency can be improved. Through the overall solution, the transmission of the gearbox gears is optimized according to the driving habits to complete the optimization of the transmission efficiency of the gearbox gears, enabling the gear combination to work continuously within the optimal load and speed range, and making the transmission system operate in a more efficient state to ensure the driver's driving experience and fuel economy. At the same time, the driving habits in the target information are optimized through the transmission optimization method to obtain optimized driving habits. In the information acquisition stage, the image information in front of the vehicle can be directly obtained through an in-vehicle camera or a subsequently installed driving recorder. The weight of the main driver is the weight of the driver sitting in the main driver's seat, and the past driving information is the previous driving information, which may specifically include the throttle usage frequency, fuel supply intensity, brake usage frequency, brake intensity, etc. In the demand determination stage, it is judged whether the driver is replaced by determining whether the weight of the main driver has changed. When the weight of the main driver changes, it is feedback that the driver is replaced. Specifically, a deviation range can be added to the weight of the main driver during the judgment to avoid errors. When it is judged that the driver is replaced, the driver's needs, that is, the user needs, are obtained. The specific user needs may be the driving habit needs of the driver, such as aggressive and gentle types, etc. There are driving habits in the standard driving preferences that are consistent with driving habits such as aggressive and gentle types. By extracting the features of the user needs and then matching the extracted features in the standard driving preferences to obtain the target driving habit. This process of feature extraction can be obtained by training the corresponding feature extraction model. The process of matching to obtain the target driving habit is to perform the same feature matching. It is also possible to extract the features in the standard driving preferences through a feature recognition model, and then compare the features of the standard driving preferences with the features of the user needs to judge whether the two are consistent to obtain the result. When the result reflects that the two are consistent, the features of the standard driving preference are extracted to obtain the target driving habit. The target driving habit is obtained by selecting the driving habit corresponding to the user needs from the standard driving preferences according to the user needs. When the driver is driving, the vehicle operates according to the target driving habit. In the feature establishment stage, by establishing the association relationship between the optimized driving habit and the weight of the main driver, it is convenient for subsequent retrieval. By establishing a repository, the optimized driving habit and the weight of the main driver can be stored for subsequent search. In the matching output stage, by monitoring the real-time weight of the main driver and judging whether the real-time weight of the main driver has changed. When it changes, the optimized driving habit corresponding to the real-time weight of the main driver is found in the repository through the matching method to match the corresponding driver by the driver's weight.By outputting driving habits based on the main driver's weight in the two stages of demand determination and matching output, they can complement each other. During use, when there is no sample in the repository, driving habits are output based on the main driver's weight in demand determination. Also, when neither of them has a driving habit that satisfies the driver, another method is used to generate a driving habit to meet the driver's needs. In the driving recommendation stage, the set recommendation method determines the route based on the vehicle's real-time position and provides corresponding driving habits for the driver according to the repeated route. In the information integration stage, the target information is one of the matching driving habit, recommended driving habit, and target driving habit. That is, when the transmission optimization method performs optimization processing, one of the driving habits in the target information is optimized. In the information feedback stage, based on past driving information, the driving habit of the driver is judged through a driving judgment method to obtain a user profile and improvement method. By feeding back the user profile and improvement method to the driver, it is convenient for the driver to understand their own driving habits and refine their driving habits to improve fuel economy. The feedback address information is the driver's address information, including but not limited to mobile phone numbers and email addresses. The specific process for the driver to feedback information based on their feelings about the gearshift decision can be to establish an independent information receiving and sending interface. Through the information receiving and sending interface as an intermediary, as the medium between the driver and the system, the driver sends their own feelings to the information receiving and sending interface, and after the information receiving and sending interface preprocesses the data of the information sent by the driver, it is fed back to the system. Conversely, the information receiving and sending interface sends the correction information generated by the system to the driver according to the driver's address information.,

[0021] As Figure 1 shown, the transmission optimization method includes: obtaining the vehicle's real-time position, monitoring whether a gearshift decision is executed, obtaining the following distance based on image information, obtaining traffic information based on the vehicle's real-time position, and optimizing the gearshift decision through a deep optimization method based on the following distance and traffic information and replacing the gearshift decision in the target information to obtain an optimized driving habit.

[0022] It should be noted that by monitoring whether a gearshift decision is executed, optimization preparation can be carried out. The gearshift decision is optimized through a deep optimization method combined with traffic information and following distance and the gearshift decision in the target information is replaced to obtain an optimized driving habit. By adaptively adjusting the gearshift decision according to the following distance and traffic information, the transmission efficiency of the gearbox gears can be improved, thereby improving fuel economy and reducing jerks, and thus improving the driver's driving experience.

[0023] As Figure 1As shown in the figure, the depth optimization method includes: obtaining a shifting strategy based on target information, presetting a spacing threshold and a deviation value, determining whether there are passive shifting factors in the following distance and traffic information to obtain result information. The passive shifting factors include that the following distance is less than the spacing threshold and red light prohibits passing. When the result information feedback indicates that there are passive shifting factors, during the process of executing the shifting decision, based on the shifting strategy and in cooperation with the deviation value, reduce the shifting speed. When the result information feedback indicates that there are no passive shifting factors, during the process of executing the shifting decision, based on the shifting strategy and in cooperation with the deviation value, increase the shifting speed. Continuously optimize the shifting speed through an exploration and learning method and adjust the target information to obtain an optimized driving habit.

[0024] It should be noted that the process of obtaining the shifting strategy based on target information, that is, extracting the shifting strategy from the corresponding driving habit. The shifting strategy refers to the shifting speed. The spacing threshold and the deviation value are preset according to actual usage requirements. The specific spacing threshold refers to the spacing value between the current vehicle and the vehicle in front. Determine whether there are passive shifting factors. When passive shifting factors occur, the shifting strategy is optimized in real time. When the result information feedback indicates that there are passive shifting factors, during the process of executing the shifting decision, based on the shifting strategy and in cooperation with the deviation value, reduce the shifting speed, that is, reduce the deviation value in the shifting strategy to lower the shifting speed. By reducing the shifting speed, improve fuel economy and reduce jerks. When the result information feedback indicates that there are no passive shifting factors, during the process of executing the shifting decision, based on the shifting strategy and in cooperation with the deviation value, increase the shifting speed, that is, increase the deviation value in the shifting strategy to increase the shifting speed, to improve fuel economy. The smaller the value of the deviation value, the better, so as to explore a driving habit suitable for the driver.

[0025] The exploration and learning method includes: presetting a safety spacing and an addition value. When the following distance is less than or equal to the safety spacing, during the process of executing the shifting decision, based on the shifting strategy and in cooperation with the deviation value, increase the shifting speed. Obtain the driver's feedback on the feeling of the optimized shifting decision to get feedback information. When the feedback information indicates dissatisfaction with the optimized shifting decision, based on the feedback information and in cooperation with the addition value, increase or decrease the deviation value to obtain an optimized deviation value. Replace the optimized deviation value with the pre-optimized deviation value to increase or decrease the shifting speed. When the feedback information indicates satisfaction with the optimized shifting decision, then replace the shifting strategy in the target information with the optimized shifting strategy to obtain an optimized driving habit.

[0026] It should be noted that the safety distance and the added value are set according to the actual usage. The safety distance is the safe distance maintained from the vehicle in front. When the following distance is less than or equal to the safety distance, the shifting speed is increased, that is, the downshifting speed is increased, to ensure the safety of the driver. By obtaining the driver's feelings about the optimized shifting decision, the shifting decision can be further optimized, which is convenient for finding the driving habits suitable for the driver, so as to improve fuel economy and the driver's driving experience. The further optimization process is to combine the added value with the deviation value to change the size of the deviation value, so as to find the driving habits suitable for the driver.

[0027] As Figure 1 shown, the recommended method includes: obtaining the user's route requirements, presetting the target weight, obtaining the vehicle driving route based on the user's route requirements, judging the weight of the vehicle driving route to obtain the specific weight. When the specific weight exceeds the target weight, record the optimized shifting decision on the vehicle driving route to obtain the target decision, replace the shifting decision in the target information based on the target decision to obtain the recommended driving habit, and push the recommended driving habit to the user.

[0028] It should be noted that the user's route requirements are obtained according to the navigation information input by the user, and the target weight can be set by oneself. The specific target weight can refer to the number of occurrences. The process of judging the weight of the vehicle driving route, that is, judging the number of occurrences of the vehicle driving route. The number of occurrences is the repeated driving times of the vehicle driving route. When the number of occurrences exceeds the target weight, record the optimized shifting decision on the vehicle driving route to obtain the target decision, and replace the shifting decision in the target information to obtain the recommended driving habit to improve the driver's driving experience. The process of pushing the recommended driving habit to the user is to push the recommended driving habit to the driver.

[0029] Driving judgment method: preset a time period, intercept the past driving information based on the time period to obtain the paragraph information, extract the number of times of hard braking and hard accelerating in the paragraph information to obtain the corresponding number of times, preset a grading standard and a target number of times. The grading standard includes aggressive and economical types. The aggressive type is when the number of times of hard braking and hard accelerating within the time period exceeds the target number of times, and the economical type is when the number of times of hard braking and hard accelerating within the time period does not exceed the target number of times. Obtain the user portrait based on the relationship between the corresponding number of times and the target number of times. When the user portrait is aggressive, extract the corresponding number of times and the time period to obtain the reference information, obtain the method to improve hard braking and hard accelerating to obtain the correction information, and integrate the reference information and the correction information to obtain the improvement method.

[0030] It should be noted that the time segment is formulated according to actual usage needs, specifically refers to a time span, and the past driving information is arbitrarily intercepted based on the time segment to obtain the segment information, and the segment information is used as sample information to analyze the user's driving habits. The classification standard and target number are formulated according to actual usage needs. When the number of heavy brakes and heavy accelerators exceeds the target number, the user is aggressive, and the rest are economical. The specific classification standard can be formulated according to actual usage. The process of obtaining methods to improve heavy brakes and heavy accelerators and obtain correction information can be retrieved on the retrieval platform. For example, practice smooth acceleration and deceleration, avoid slamming on the accelerator or brake, and try to predict the road conditions in advance while driving, so as to have an idea and adjust the accelerator and brake in time. By feeding back the correction information to the user, it is convenient for the user to adjust their own driving habits and help improve fuel economy.

[0031] like Figure 2 As shown, the process of presetting the standard driving preference is: presetting the standard shifting speed, the standard driving preference includes aggressive preference, mild preference and economic preference, the mild preference is obtained based on the standard shifting speed, the aggressive preference is obtained by increasing the deviation value based on the standard shifting speed, and the economic preference is obtained by reducing the deviation value based on the standard shifting speed.

[0032] It should be noted that by presetting standard driving preferences, it is possible to provide driving habits according to the needs of novice drivers who have not driven the vehicle before, so as to avoid the novice driver adopting the driving habits of the previous driver when driving the vehicle, so as to improve the driver's driving experience. At the same time, it will not interfere with the driving habits of the previous driver, which is conducive to improving the driving experience of any driver and is easy to use.

[0033] like Figure 1 As shown, the matching method includes: presetting a change value, obtaining the real-time weight of the main driver after the change to obtain a target weight, combining the change value with the target weight to obtain a target weight range, retrieving the main driver's weight within the target weight range in a repository based on the target weight range to obtain a selected weight, and extracting an optimized driving habit corresponding to the selected weight in the repository based on an association relationship to obtain a matched driving habit.

[0034] It should be noted that the size of the change value is formulated according to actual usage. For example, if the size of the change value is 2 kilograms, the change value is combined with the target weight to obtain the target weight range. Specifically, the target weight range is obtained by adding the change value to the target weight and reducing the change value. When the target weight is 65 kilograms, the target weight range is 63 kilograms to 67 kilograms. Based on the target weight range, the main driving weight within the target weight range is retrieved from the repository to obtain the selected weight. This process, that is, the weight within the target range is retrieved from the repository to obtain the selected weight. When the number of selected weights is multiple, the weight closest to the target weight is selected to obtain the selected weight, and the posture habits corresponding to the selected weight are extracted from the repository according to the association relationship to obtain the matching driving habits, so as to judge the driver and provide the driver with driving habits that conform to the driver's habits, which is convenient for use.

[0035] Embodiment: When the driver is driving a vehicle, the transmission gear transmission efficiency optimization method provided by the present invention is used. When commuting on urban traffic roads or when there is congestion during commuting, long-term rigid optimization of the gear transmission efficiency will lead to frequent jerking, which may cause discomfort to the driver and passengers in the vehicle. Therefore, it is necessary to set up a transmission gear transmission efficiency optimization method to reduce jerking and improve fuel economy and driving experience.

[0036] In the specific implementation process: first, it is necessary to obtain the front image information of the vehicle according to the on-board front camera or the driving recorder, and at the same time obtain the weight of the main driver, that is, the driver's weight and the driver's previous driving information. The user's driving habits are analyzed according to the previous driving information through the driving judgment method and an improvement method is provided, so that the user can understand and improve his own driving habits. When the driver gets on the vehicle for the first time, the user needs of the driver are obtained, and the standard driving preferences corresponding to the user needs are found in the preset standard driving preferences according to the user needs, so as to improve the driving experience of the first-time driver. When the driver is not getting on the vehicle for the first time, the real-time weight of the main driver is monitored, and the driving habits corresponding to the real-time weight are found in the storage library through the matching method, so as to store the driving habits of different drivers and output them to the corresponding driver, so as to improve the driver's driving experience and improve fuel economy. During the driving process of the vehicle, the gear shifting speed is adaptively adjusted in real time according to the image information in front of the vehicle through the transmission optimization method, so as to improve the transmission efficiency of the gearbox gear, thereby improving the driver's driving experience and improving fuel economy. By improving the transmission efficiency of the gearbox gear, noise and vibration can be reduced, the service life can be extended, the maintenance cost can be reduced, and environmental protection requirements can be met.

[0037] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is limited by the attached embodiments and their equivalents.

Claims

1. A method for optimizing transmission efficiency of a gearbox based on machine learning, characterized in that: The method comprises: Driving optimization: Based on target information and image information, the transmission optimization method is used to continuously learn and optimize the transmission efficiency of the gearbox gears; Information acquisition: obtain the image information in front of the vehicle, the main driver's weight and previous driving information; Demand determination: preset standard driving preferences, judge the change of the main driving weight to obtain the judgment result, and when the judgment result is fed back that the main driving weight has changed, obtain the user's needs, and select the corresponding standard driving preferences based on the user's needs to obtain the target driving habits; Feature establishment: establish the association between optimized driving habits and main driving weight, and establish a repository for storing optimized driving habits and main driving weight; Matching output: monitor the real-time weight of the main driver. When the real-time weight of the main driver changes, the optimized driving habits corresponding to the real-time weight of the main driver are selected through a matching method based on the repository to obtain matching driving habits, and the matching driving habits are pushed to the driver; Driving recommendation: Based on the real-time location of the vehicle, a recommended driving habit of a repeated route is recommended to the user through a recommendation method to obtain a recommended driving habit, and the recommended driving habit is pushed to the driver; Information unification: matching driving habits, recommended driving habits and target driving habits are all target information; Information feedback: Based on the previous driving information, the user's driving habits are judged by the driving judgment method to obtain the user portrait and improvement method, the feedback address information is obtained, and the user portrait and improvement method are fed back to the user based on the feedback address information; Gear transmission efficiency optimization: By optimizing the gear shifting operation and driving habits, the gear slip and impact during the transmission process are reduced, the gear meshing efficiency is improved, and thus the transmission efficiency is indirectly improved; The transmission optimization method includes: obtaining the real-time position of the vehicle, monitoring whether the gear shifting decision is executed, obtaining the following vehicle distance based on image information, obtaining traffic information based on the real-time position of the vehicle, optimizing the gear shifting decision through a deep optimization method based on the following vehicle distance and traffic information, and replacing the gear shifting decision in the target information to obtain optimized driving habits.

2. The method for optimizing the transmission efficiency of a gearbox based on machine learning according to claim 1, characterized in that: The deep optimization method includes: obtaining a shifting strategy based on target information, presetting a spacing threshold and a deviation value, judging whether there are passive shifting factors in the following vehicle spacing and traffic information to obtain result information, wherein the passive shifting factors include the following vehicle spacing being less than the spacing threshold and the red light prohibiting passage; when the result information feedback indicates that there are passive shifting factors, in the process of executing the shifting decision, the shifting speed is reduced based on the shifting strategy in combination with the deviation value; when the result information feedback indicates that there are no passive shifting factors, in the process of executing the shifting decision, the shifting speed is increased based on the shifting strategy in combination with the deviation value; and the shifting speed is continuously optimized by an exploratory learning method and the target information is adjusted to obtain an optimized driving habit.

3. The method for optimizing the transmission efficiency of a gearbox based on machine learning according to claim 2, characterized in that: The exploratory learning method includes: presetting a safety distance and an added value, and when the following distance is less than or equal to the safety distance, in the process of executing the gear shift decision, accelerating the gear shift speed based on the gear shift strategy and the deviation value, obtaining the driver's feedback on the feeling of the optimized gear shift decision to obtain feedback information, when the feedback information is that the optimized gear shift decision is not satisfactory, increasing or decreasing the deviation value based on the feedback information and the added value to obtain the optimized deviation value, replacing the deviation value before optimization with the optimized deviation value to increase or decrease the gear shift speed, and when the feedback information is that the optimized gear shift decision is satisfactory, the gear shift strategy in the target information is replaced with the optimized gear shift strategy to obtain the optimized driving habit.

4. The method for optimizing transmission efficiency of a gearbox based on machine learning according to claim 1, characterized in that: The recommendation method includes: obtaining user route requirements, presetting target weights, obtaining a vehicle driving route based on the user route requirements, determining the weight of the vehicle driving route to obtain a specific weight, when the specific weight exceeds the target weight, recording optimized shifting decisions on the vehicle driving route to obtain a target decision, replacing the shifting decisions in the target information based on the target decision to obtain a recommended driving habit, and pushing the recommended driving habit to the user.

5. The method for optimizing transmission efficiency of a gearbox based on machine learning according to claim 1, characterized in that: The driving judgment method includes: presetting a time segment, intercepting past driving information based on the time segment to obtain segment information, extracting the number of times the brake and the accelerator are stepped on hard in the segment information to obtain the corresponding number, presetting a grading standard and a target number, the grading standard including an aggressive type and an economical type, the aggressive type means that the number of times the brake and the accelerator are stepped on hard in the time segment exceeds the target number, the economical type means that the number of times the brake and the accelerator are stepped on hard in the time segment does not exceed the target number, obtaining a user portrait based on the relationship between the corresponding number and the target number, when the user portrait is an aggressive type, extracting the corresponding number and the time segment to obtain reference information, obtaining a method for improving the heavy braking and the heavy accelerator to obtain correction information, and integrating the reference information and the correction information to obtain an improved method.

6. The method for optimizing the transmission efficiency of a gearbox based on machine learning according to claim 5, characterized in that: The process of presetting the standard driving preference is: presetting the standard shifting speed, the standard driving preference includes aggressive preference, mild preference and economic preference, the mild preference is obtained based on the standard shifting speed, the aggressive preference is obtained by increasing the deviation value based on the standard shifting speed, and the economic preference is obtained by reducing the deviation value based on the standard shifting speed.

7. The method for optimizing transmission efficiency of a gearbox based on machine learning according to claim 1, characterized in that: The matching method includes: presetting a change value, obtaining the real-time weight of the main driver after the change to obtain a target weight, combining the change value with the target weight to obtain a target weight range, retrieving the main driver's weight within the target weight range in a repository based on the target weight range to obtain a selected weight, and extracting an optimized driving habit corresponding to the selected weight in the repository based on an association relationship to obtain a matched driving habit.

8. A gearbox gear transmission efficiency optimization system based on machine learning, characterized in that: A method for optimizing the transmission efficiency of a gearbox based on machine learning as described in any one of claims 1 to 7 is used.

Images