Online gear decision-making method and system for electric automobile

Abstract

The invention relates to the technical field of vehicle gear decision making, in particular to an electric vehicle online gear decision making method and system, and the method comprises the steps: obtaining the driving intention quantification of a target vehicle and the vehicle state and driving manipulation data related to vehicle speed prediction; constructing a driving intention quantification model, and calculating a performance preference expected value according to the vehicle state and driving manipulation data related to driving intention quantification; constructing a vehicle speed prediction model, and calculating a vehicle speed prediction result according to the vehicle speed prediction related data; based on a preset off-line gear shifting rule database, selecting a gear shifting rule table according to the performance preference expected value, and extracting a gear decision result by using the driving manipulation data and the vehicle speed prediction result; and based on a dynamic programming algorithm, an optimal gear decision sequence is extracted with the purpose of minimizing the gear shifting cost. The objective of the invention is to solve the technical problem of insufficient adaptability of gear decision working conditions under complex working conditions of an automobile, and improve the comprehensive performance and intelligent level of gear decision of a multi-gear electric driving system of the electric automobile.

Description

Technical Field

[0001] The present invention relates to the technical field of vehicle gear decision-making, and in particular to an online gear decision-making method and system for an electric vehicle. Background Art

[0002] In the field of electric vehicles, gear decision-making, as a key core technology of the step-type automatic transmission control system, plays a decisive role in the performance of the entire vehicle. Its core task is to accurately select the appropriate gear according to the real-time driving status of the vehicle. The scientific determination of the shift point is directly related to the comprehensive performance of the vehicle, such as power, economy, and driving comfort. With the increasing demand for the intelligent and personalized development of electric vehicles, how to effectively integrate the driver's diverse driving intentions into the gear decision-making mechanism so that it can flexibly adapt to complex and changeable driving conditions has become a key breakthrough in improving the intelligence and personalization level of electric vehicles.

[0003] At present, many existing technologies have been extensively studied in the field of electric vehicle gear decision-making. For example, the invention patent with publication number CN115264048A provides an automatic transmission intelligent gear decision design method based on data mining. By collecting driving data of excellent drivers of manual transmission vehicles in multiple road environments, after preprocessing and cleaning, a gear decision strategy is constructed for different driving conditions, and the shift boundary points are extracted using a data mining algorithm and the shift rule surface is fitted; the invention patent with publication number CN107131296A provides a pure electric vehicle two-speed transmission system control strategy for energy consumption. Based on driving intention and accelerator pedal information, a torque acquisition method in an economical driving mode is formulated, and a shift rule optimization model is established in combination with the motor efficiency MAP diagram and power requirements, and the cross particle swarm algorithm is used to solve it. However, these existing technologies have certain limitations. Existing technologies generally use offline methods to obtain gear shifting rules. In actual application scenarios, faced with real-time changing driving conditions, the extracted gear shifting rules are difficult to achieve real-time update and online dynamic adjustment, resulting in insufficient adaptability to working conditions and poor gear decision accuracy. They cannot fully meet the needs of efficient and intelligent operation of electric vehicles in complex real-world scenarios. Summary of the invention

[0004] In order to solve the technical problem of insufficient adaptability of gear decision-making under complex working conditions of automobiles and improve the comprehensive performance and intelligence level of gear decision-making of multi-gear electric drive systems of electric vehicles, the present invention provides an online gear decision-making method and system for electric vehicles, and the technical solutions adopted are as follows: The technical solution of the first aspect of the present invention provides an online gear decision method for an electric vehicle, the method comprising: Obtain vehicle state and driving maneuver data related to the target vehicle's driving intention quantification and speed prediction; Construct a driving intention quantification model and calculate the performance preference expectation value based on the vehicle status and driving manipulation data related to the driving intention quantification; Construct a vehicle speed prediction model and calculate the vehicle speed prediction result based on the vehicle speed prediction related data; Based on the preset offline shift schedule database, the shift schedule table is selected according to the expected performance preference value, and the gear decision result is extracted using the driving operation data and vehicle speed prediction results; Based on the dynamic programming algorithm, the optimal gear decision sequence is extracted with the goal of minimizing the gear shift cost.

[0005] Furthermore, the vehicle state and driving manipulation data related to the target vehicle's driving intention quantification and vehicle speed prediction are obtained, including: Obtaining the speed, accelerator pedal strength and accelerator pedal strength change rate of the target vehicle; The target vehicle's V2X communication module is used to establish communication with the front vehicle's V2X communication module to obtain the front vehicle's speed data.

[0006] Furthermore, a driving intention quantification model is constructed, and the performance preference expected value is calculated based on the vehicle state and driving manipulation data related to the driving intention quantification, including: Calculating a first performance preference expected value according to the speed, accelerator pedal strength, and accelerator pedal strength change rate of the target vehicle; A second performance preference expected value is calculated based on the first performance preference expected value.

[0007] Furthermore, a vehicle speed prediction model is constructed, and a vehicle speed prediction result is calculated based on vehicle speed prediction related data, including: Construct a vehicle speed prediction model based on regression neural network; Construct a traffic flow speed information matrix based on the speed data of the target vehicle and the preceding vehicle; The vehicle speed prediction model is optimized by using the traffic speed information data matrix to calculate the vehicle speed prediction results.

[0008] Furthermore, based on the preset offline shift schedule database, a shift schedule table is selected according to the expected value of performance preference, and the gear decision result is extracted using the driving operation data and the vehicle speed prediction result, including: In the startup phase, a shift schedule table matching the expected performance preference is selected from a preset offline shift schedule database; The target gear is determined from the gear shift schedule table based on the accelerator pedal strength and vehicle speed prediction results.

[0009] Furthermore, based on the dynamic programming algorithm, the optimal gear decision sequence is extracted with the goal of minimizing the gear shift cost, including: Initialize the gear index value, shift command index value and solution phase index value; Determine the gear position of the current stage according to the gear position index value of the current stage; Determine the shift instruction set of the current stage according to the shift instruction index value of the current stage; Calculate the gear set for the next stage according to the gear and shift instruction set for the current stage; Constructing a cost function based on the expected value of performance preference and a preset gear shift penalty function; Based on the cost function, with the goal of minimizing the gear shift cost, the objective function value of each stage and the optimal gear shift instruction that minimizes the objective function value are solved inversely from the last stage.

[0010] Furthermore, based on the dynamic programming algorithm, the optimal gear decision sequence is extracted with the goal of minimizing the gear shift cost, and further includes: According to the inverse solution results, the optimal gear shift instructions for each stage are extracted, and starting from the current gear, the target gears of the subsequent stages are solved forward.

[0011] Furthermore, the cost function is expressed as:

[0012] In the formula, Indicates The cost function value of the stage; Indicates The gear position of the stage; Indicates The shift command of the stage; represents the first performance preference expected value; represents the normalized first performance objective function value; represents the expected value of the second performance preference; represents the normalized second performance objective function value; Represents the preset gear shift penalty function.

[0013] Furthermore, the objective function is expressed as:

[0014] In the formula, Indicates Stage objective function value; Indicates the time domain length of vehicle speed prediction; Indicates The objective function value of the stage.

[0015] The technical solution of the second aspect of the present invention provides an online gear decision system for an electric vehicle, which adopts the online gear decision method for an electric vehicle described in the technical solution of the first aspect of the present invention, and the system comprises: a data acquisition module configured to acquire vehicle state and driving maneuver data related to driving intention quantification and vehicle speed prediction of the target vehicle; A driving intention quantification module configured to construct a driving intention quantification model and calculate a performance preference expectation value based on vehicle states and driving manipulation data related to the driving intention quantification; A vehicle speed prediction module is configured to construct a vehicle speed prediction model and calculate a vehicle speed prediction result based on vehicle speed prediction related data; A first gear decision module is configured to select a gear shifting schedule table based on a preset offline gear shifting schedule database according to a performance preference expectation value, and extract a gear decision result using driving control data and a vehicle speed prediction result; The second gear decision module is configured to extract an optimal gear decision sequence based on a dynamic programming algorithm with the goal of minimizing the gear shifting cost.

[0016] The present invention has the following beneficial effects: The electric vehicle online gear decision method provided by the present invention combines the driver's personalized needs with the vehicle's driving status by constructing a driving intention quantification model and a vehicle speed prediction model, and integrates the dynamic programming algorithm to minimize the gear shifting cost and perform rolling solution; in the startup phase, based on the performance preference expected value, driving operation data and vehicle speed prediction results, a gear decision result that matches it is selected from a preset offline gear shifting law database to preliminarily match the working condition requirements. In the rolling optimization phase, based on the dynamic programming algorithm and with the goal of minimizing the gear shifting cost, reverse solution is combined with forward optimization to calculate the optimal gear decision sequence, thereby achieving accurate gear decision under complex working conditions, effectively improving the decision accuracy, personalization and intelligence level, and ensuring the optimal overall performance of the vehicle. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions and advantages in the embodiments of the present invention or the prior art, the drawings required for use in the embodiments or the prior art descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0018] Figure 1 A method flow chart of an online gear decision method for an electric vehicle provided by an embodiment of the present invention; Figure 2 A schematic diagram of solving an online gear decision method for an electric vehicle provided by an embodiment of the present invention; Figure 3 A schematic diagram of an implementation method of an online gear decision method for an electric vehicle provided by an embodiment of the present invention Figure 1 ; Figure 4A schematic diagram of an implementation method of an online gear decision method for an electric vehicle provided by an embodiment of the present invention Figure 2 ; Figure 5 A schematic diagram of an implementation method of an online gear decision method for an electric vehicle provided by an embodiment of the present invention Figure 3 ; Figure 6 A schematic diagram of vehicle speed during the time period of 8:00-8:40 provided by an embodiment of the present invention; Figure 7 A schematic diagram of a first performance preference expected value for the time period 8:00-8:40 provided by an embodiment of the present invention; Figure 8 A schematic diagram of a gear decision result in the time period of 8:00-8:40 provided by an embodiment of the present invention; Fig. 9 A schematic diagram of vehicle speed during the time period of 13:40-14:20 provided by an embodiment of the present invention; Fig.10 A schematic diagram of a first performance preference expected value for the time period 13:40-14:20 provided by an embodiment of the present invention; Fig.11 A schematic diagram of a gear decision result in the time period of 13:40-14:20 provided by an embodiment of the present invention; Fig.12 A schematic diagram of vehicle speed during the time period of 17:20-18:00 provided by an embodiment of the present invention; Fig.13 A schematic diagram of a first performance preference expected value for the time period of 17:20-18:00 provided by an embodiment of the present invention; Fig.14 A schematic diagram of a gear decision result in the time period of 17:20-18:00 provided by an embodiment of the present invention; Fig.15 A schematic diagram of the structure of an online gear decision system for an electric vehicle provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0019] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the following is a detailed description of an electric vehicle online gear decision method and system proposed by the present invention, its specific implementation, structure, features and effects, in combination with the accompanying drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" does not necessarily refer to the same embodiment. In addition, specific features, structures or characteristics in one or more embodiments may be combined in any suitable form.

[0020] Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs.

[0021] The specific scheme of an online gear decision method and system for an electric vehicle provided by the present invention is described in detail below with reference to the accompanying drawings.

[0022] See also Figure 1 and Figure 2 As shown, it shows a method flow chart and a solution schematic diagram of an electric vehicle online gear decision method provided by an embodiment of the present invention, the method comprising: Step S100: Acquire vehicle state and driving manipulation data related to driving intention quantification and vehicle speed prediction of the target vehicle; Step S100 specifically includes: Step S110: Obtain the speed, accelerator pedal strength and accelerator pedal strength change rate of the target vehicle; specifically, a sensor can be set at the accelerator pedal to accurately measure the current position of the accelerator pedal through the sensor to obtain the accelerator pedal strength, and the change in the accelerator pedal position per unit time can be recorded to obtain the accelerator pedal strength change rate.

[0023] Step S120: Use the V2X communication module of the target vehicle to establish communication with the V2X communication module of the preceding vehicle to obtain the speed data of the preceding vehicle; specifically, the target vehicle communicates with the V2X communication module of the preceding vehicle through the V2X communication module to obtain the speed of the vehicle in front of the target vehicle; V2X stands for Vehicle to Everything, V stands for vehicle, and X stands for roads, people, vehicles, equipment and all other connectable devices; V2X communication module is a key component in the Internet of Vehicles, and V2X is a technology that enables vehicles to communicate with other entities such as other vehicles, infrastructure, and networks.

[0024] Step S200: constructing a driving intention quantification model, and calculating a performance preference expected value according to the vehicle state and driving manipulation data related to the driving intention quantification; Step S200 specifically includes: Step S210: calculating a first performance preference expected value according to the speed of the target vehicle, the accelerator pedal strength and the accelerator pedal strength change rate; Step S220: Calculate the second performance preference expected value according to the first performance preference expected value; Specifically, the present embodiment is based on the driver's power and economy expected quantification method of the fuzzy neural network and exponential smoothing-Markov model, wherein the power expected value is the first performance preference expected value, which can be expressed as , the expected value of economy, that is, the expected value of the second performance preference, can be expressed as , ; The specific technical principle is to establish an offline driver dynamics expectation quantitative model and a driving control characteristic parameter and vehicle operation status parameter prediction model based on the collection and processing of various manipulation test data of representative drivers. During driving, the throttle opening (i.e., accelerator pedal strength) and vehicle speed are collected in real time, and then the throttle opening, throttle opening change rate and vehicle speed at the next moment are predicted through the prediction model. The predicted parameters are input into the driver's dynamics expectation quantitative model to obtain the driver's current dynamics expectation value and economic expectation value. For the specific calculation method, please refer to the Chinese invention patent with publication number CN111027618A. The present invention does not further elaborate on the calculation of the first performance preference expectation value and the second performance preference value.

[0025] Step S300: constructing a vehicle speed prediction model, and calculating a vehicle speed prediction result according to vehicle speed prediction related data; Step S300 specifically includes: Step S310: constructing a vehicle speed prediction model based on a regression neural network; specifically, the vehicle speed prediction module in this embodiment is preferably a generalized regression neural network, including an input layer, a pattern layer, a summation layer and an output layer; the input vector of the vehicle speed prediction model is the historical vehicle speed data of the target vehicle and the current actual vehicle speeds of multiple preceding vehicles, and the output vector is the predicted vehicle speed of the target vehicle in the future; a training sample of the vehicle speed prediction model is composed of a training sample input vector and a training sample output vector, and the training sample input vector and output vector are extracted from the historical vehicle speed according to the model input and output vector according to the sliding time window, and a number of training samples form a training sample matrix; Step S320: construct a traffic flow speed information matrix based on the speed data of the target vehicle and the preceding vehicle; specifically, obtain the historical speed data of the target vehicle, and use the target vehicle V2X communication module to establish communication with the preceding vehicle V2X communication module to obtain the current speed data of the preceding vehicle, and construct the traffic flow speed information matrix; specifically, define the time length of the historical speed data of the target vehicle as , the predicted future speed time length is ; Use the target vehicle's V2X communication module to obtain the current moment in front The actual speed data of the vehicle is compared with the previous The traffic speed information at each moment is combined, and the traffic speed information matrix can be expressed as:

[0026] In the formula, represents the second traffic speed information matrix; express The actual speed of the host vehicle at the moment; Indicates the training time domain length, that is, the data time span used to train the GRNN model; Indicates the prediction time domain length, that is, how long in the future the main vehicle speed is predicted. It is the dimension of the output vector, indicating how many time points in the future the model will provide predictions for. The traffic flow speed information matrix includes the main vehicle and the front vehicle speed information matrix. The actual speed information of the vehicle at the target time; Step S330: using the traffic speed information data matrix to train and optimize the vehicle speed prediction model and calculate the vehicle speed prediction result; Step S330 specifically includes: Step S331: extracting the training sample input vector and the training sample output vector based on the traffic flow data information matrix, and updating the training sample matrix of the vehicle speed prediction model. The training sample matrix can be expressed as:

[0027] In the formula, Indicates from Time has come The target vehicle speed data at the moment and The speed data of the preceding vehicle at the moment The training sample input vector of the GRNN model;

[0028] In the formula, Represents the training sample output vector, as The expected output vector of the GRNN model provides the model with the target sequence to be predicted.

[0029] In the formula, Represents the training sample input matrix, Placed in The last column of can be used to update the training sample matrix to ensure that the matrix contains the latest historical and current speed information;

[0030] In the formula, Represents the training sample output matrix, Add to The last column of updates the output sample matrix and provides the model with a series of target vehicle speeds at time points as the expected output to evaluate the model's ability to predict the future speed of the target vehicle. The number of training samples is determined by Control, when When and The first column and ,This control mechanism ensures that the training sample matrix does not grow indefinitely, keeping its size within a reasonable range, while ensuring that the data in the matrix is ​​always the latest and most relevant to adapt to the ,continuously updated data and avoid overfitting or excessive ,computational burden caused by too much historical data.

[0031]

[0032] In the formula, Represents the predicted sample input vector.

[0033] Step S332: Based on the updated training sample matrix, extract the optimal vehicle speed prediction model with the goal of minimizing the root mean square error; use the updated and right A GRNN model is trained; the new training sample input vector is input into the trained model to obtain the predicted output; then, the performance of the model is evaluated by calculating the root mean square error, and the model with the smallest root mean square error is selected as the optimal model; Step S333: Input the prediction sample input vector into the optimal vehicle speed prediction model and output the vehicle speed prediction data; Input the prediction sample input vector The input is then fed into the optimal GRNN model, and the model is used to predict the future of the target vehicle. Predictions are made at speeds of seconds; Contains the latest speed information to predict future vehicle speed.

[0034] In this embodiment, by constructing a vehicle speed prediction model based on a generalized regression neural network, combining the historical vehicle speed data of the target vehicle and the real-time vehicle speed data of the preceding vehicle, a vehicle flow speed information matrix is ​​constructed and the training sample matrix is ​​continuously updated, and the model is trained and optimized with the goal of minimizing the root mean square error, and finally the accurate prediction of the future vehicle speed of the target vehicle is achieved. The vehicle speed prediction model provided by the method can be updated and learned online in real time according to the vehicle condition, and can adjust its own parameters and structure in time according to the changing traffic conditions, adapt to various traffic scenes, avoid prediction deviations caused by changes in vehicle conditions, and thus effectively improve the accuracy of vehicle speed prediction; the vehicle speed prediction model provides key speed prediction information for the online gear decision of electric vehicles, so that the gear decision can fully consider the future driving state of the vehicle, thereby improving the foresight and accuracy of the gear decision, better adapting to complex and changeable traffic conditions, helping to improve the vehicle's power performance, economy and driving comfort, and promoting the development of intelligent gear decision technology for electric vehicles.

[0035] Step S400: Based on a preset offline shifting schedule database, a shifting schedule table is selected according to the expected performance preference value, and the gear decision result is extracted using the driving operation data and the vehicle speed prediction result; specifically, the offline shifting database is obtained in the vehicle development stage through a large number of experiments and simulation calculations, for different combinations of the first expected performance preference value and the second expected performance preference value, corresponding shifting schedules are obtained, and these shifting schedules are stored in the transmission control unit TCU to form a database.

[0036] Step S400 specifically includes: Step S410: In the startup phase, a shifting schedule table matching the performance preference expected value is selected from a preset offline shifting schedule database based on the performance preference expected value; specifically, according to the first performance preference expected value and the second performance preference expected value, since the offline shifting database calculates the shifting schedule offline for different combinations of the first performance preference expected value and the second performance preference expected value, a shifting schedule table matching the performance preference expected value can be accurately selected from the preset offline shifting schedule database; Step S420: Determine the target gear from the gear shifting rule table according to the accelerator pedal strength and vehicle speed prediction results; specifically, use the selected gear shifting rule table, take the accelerator pedal strength and vehicle speed prediction results as input, and substitute them into the gear shifting rule table selected in step S410. The gear shifting rule table is essentially a table containing the correspondence between different vehicle speeds, accelerator pedal strengths and recommended gears. According to the current vehicle speed prediction value and accelerator pedal strength, the corresponding recommended gear is searched in the gear shifting rule table to determine the gear decision result.

[0037] When in use, the transmission control unit (TCU) first collects vehicle speed and accelerator pedal strength signals in real time, and then calculates the first performance preference expected value and the second performance preference expected value based on the collected vehicle speed and accelerator pedal strength signals, and then selects the shifting law that matches the current first performance preference expected value and the second performance preference expected value from the shifting law data stored in the TCU, and then determines the target gear according to the selected shifting law based on the predicted vehicle speed and accelerator pedal strength; step S400 selects the gear decision result that matches it from the preset offline shifting law database by combining the performance preference prediction value and the vehicle speed prediction result, thereby providing an efficient and accurate method for the gear decision of the electric vehicle.

[0038] Step S500: Based on the dynamic programming algorithm, the optimal gear decision sequence is extracted with the goal of minimizing the gear shifting cost; specifically, in dynamic programming, the entire gear decision process will be divided into a series of discrete decision stages, each stage corresponding to a specific time point or a discrete moment of the vehicle's driving state.

[0039] Step S500 specifically includes: Step S510: Initializing the gear index value, the shift instruction index value and the solution phase index value; Step S520: determining the gear position of the current stage according to the gear position index value of the current stage; determining the gear shift instruction set of the current stage according to the gear shift instruction index value of the current stage; calculating the gear position set of the next stage according to the gear position and the gear shift instruction set of the current stage; Specifically, this embodiment uses the gear position at the current stage as a state variable, which can be expressed as: , Indicates The state variables of the phase; Indicates gears, is the gear index value; the gear shift command index value is used as the control variable, which can be expressed as:

[0040] In the formula, Indicates Control variables of the stage; Indicates A gear shift control command, is the shift command index value, -1 means downshift, 0 means no shift, and 1 means upshift; The state transfer equation for calculating the gear position in the next stage is expressed as:

[0041] In the formula, express The state variable of the next stage, that is, the next gear obtained by the gear shifting operation.

[0042] Step S530: constructing a cost function based on the performance preference expected value and the preset shift penalty function; the expression of the cost function is:

[0043] In the formula, Indicates The cost function value of the stage; Indicates The gear position of the stage; Indicates The shift command of the stage; represents the first performance preference expected value; represents the normalized first performance objective function value; represents the expected value of the second performance preference; represents the normalized second performance objective function value; Represents the preset gear shift penalty function, which is used to punish unnecessary gear shift operations. Its value is proportional to the absolute value of the gear shift control instruction, and the proportional coefficient is the gear shift penalty coefficient : , Indicates A gear shift control command.

[0044] Specifically, this implementation is based on the first performance objective function value and the second performance objective function value Normalized calculation and , the expression is:

[0045] In the formula, shows the normalized value, Represents the original data, y max , y min are the maximum and minimum values ​​of the sample data respectively.

[0046] In this embodiment, the first performance sub-objective function value is calculated according to the backup power parameter of the motor; the expression of the first performance sub-objective function value is:

[0047] In the formula, represents the value of the first performance objective function; represents the motor backup power. The first performance objective function value is obtained by taking the motor backup power The opposite value of the backup power is used as the indicator to measure the value of the first performance objective function. The larger the backup power, the greater the value after taking the opposite value. The smaller it is, the more adjustments need to be made; The expression of motor backup power is:

[0048]

[0049] In the formula, Indicates the backup torque of the motor; Indicates the speed of the motor; 9550 is a conversion constant used to convert the units of torque and speed into the unit of power kW; Indicates the motor torque; Indicates the corresponding The maximum torque of the motor that can be provided under Indicates the mapping relationship between the accelerator pedal strength, motor speed and motor maximum torque; Indicates the maximum accelerator pedal strength; Indicates that the reserve torque is the maximum torque that the motor can provide at the corresponding speed The actual motor torque The difference between and reflects how much torque reserve the motor has; Indicates the maximum torque The accelerator pedal strength and motor speed Through the mapping relationship The maximum torque that the motor can output is determined by the maximum accelerator pedal strength and the current speed. This embodiment calculates the reserve power of the motor based on the actual operating parameters (torque, speed) of the motor, the accelerator pedal strength, and the maximum accelerator pedal strength, and further obtains the first performance sub-objective function value, which provides an effective indicator for quantifying the driver's target vehicle power performance. This indicator can reflect the power reserve of the motor under the current driving state, and then reflect the driver's potential demand for power. It helps to fully consider the driver's demand for power performance in the subsequent gear decision process, make the gear decision more in line with the driver's actual needs, and optimize the vehicle's operating performance under different working conditions.

[0050] In this embodiment, the second performance sub-objective function value is calculated according to the motor output power and the motor efficiency, and its expression is:

[0051] In the formula, represents the value of the second performance objective function; Indicates the motor output power (kW); Indicates motor efficiency; Indicates the mapping relationship between motor speed, motor torque and motor efficiency; represents a discrete time interval; Indicates that at discrete time intervals The energy output by the internal motor; at the same time The motor efficiency is the motor speed and motor torque The function of the motor speed and torque is that the motor efficiency is different at different speeds and torques, which in turn affects the economy. Specifically, a large number of motor performance tests are conducted in advance to obtain the performance of different motor speeds. and torque The motor efficiency data under the combination are fitted to obtain the mapping relationship between motor speed, motor torque and motor efficiency. , and stored in the electronic control unit ECU in the form of a table or function, and the efficiency of the current motor is determined by querying the stored mapping relationship according to the real-time motor speed and torque collected; this embodiment collects the motor speed, torque and other operating parameters, combines the predetermined discrete time interval, calculates the motor output power and efficiency, and then obtains the second performance sub-objective function value. This value reflects the relationship between the motor output energy and the motor efficiency in a specific discrete time interval, and provides a quantitative indicator for the vehicle economy.

[0052] Step S540: Based on the cost function, with the goal of minimizing the shifting cost, reversely solve the objective function value of each stage from the last stage and the optimal shifting instruction that minimizes the objective function value; the last stage refers to the stage corresponding to the last discrete moment in the entire decision time range, for example, for the future The gear position decision is made within the time step, so the last stage is the The stage corresponding to the time step is the end point of the entire decision sequence.

[0053] The expression of the objective function is:

[0054] In the formula, Indicates Stage objective function value; Indicates the index value of the solution stage; Indicates the time domain length of vehicle speed prediction; Indicates The cost function value of the stage; Indicates The objective function value of the stage; this embodiment uses a piecewise function, when When it is in the last stage, the comprehensive objective function value is 0. This is because there is no subsequent decision to be considered after the last stage, so the comprehensive objective function value of this stage is set to 0 as the boundary condition for reverse deduction; To ensure that the motor speed, torque, power and gear are within a reasonable range, the constraint condition expression is as follows:

[0055] In the formula, Indicates the peak speed of the motor; Indicates the maximum torque of the motor; Indicates the peak power of the motor; Step S550: Extract the optimal shifting instruction for each stage according to the reverse solution result, and start from the current gear and solve the target gear of each subsequent stage in a forward direction; specifically, after the reverse solution is completed, for each stage (i.e., each gear), the control variable that minimizes the objective function value has been determined. This embodiment starts from the current actual gear and uses it as the initial state variable. According to the corresponding optimal control variable obtained by the reverse solution, the target gear of the next stage is calculated using the state transfer equation. Repeat the above steps, as the vehicle travels, at each new moment, the target gear calculated at the previous moment is used as the current gear, and then the new target gear is calculated according to the corresponding control variable obtained by the reverse solution. In this way, the cycle is repeated, and the gear that the vehicle should switch to at each subsequent moment is gradually determined according to the current state and the determined optimal shifting strategy, so as to achieve the dynamic and real-time optimization decision of the vehicle gear, so as to minimize the shifting cost and improve the comprehensive performance of the vehicle, and obtain the optimal gear decision result of the entire decision sequence. Through forward solving, the theoretical optimal gear shifting strategy obtained by reverse solving is applied to the actual vehicle driving process. According to the real-time status of the vehicle, the target gear at each moment is gradually determined to complete the entire online gear decision-making process based on minimizing the gear shifting cost.

[0056] This embodiment is based on a dynamic programming algorithm, and extracts the optimal gear decision sequence with the goal of minimizing the gear shifting cost, taking into account multiple factors such as power, economy, and gear shifting penalty. By combining reverse solving and forward solving, under the premise of satisfying the constraints of various motor parameters, it can make full use of the performance preference expectation value and the gear shifting rule, accurately find the optimal gear shifting instruction for each stage, and thus obtain the optimal gear decision result for the entire decision sequence. It can not only effectively reduce unnecessary gear shifting operations and reduce gear shifting costs, but also dynamically adjust the gear according to the driver's expectations for power and economy, improve the vehicle's power performance and fuel economy, and enhance driving comfort and stability, making the gear decision of electric vehicles more intelligent, personalized and efficient, and adapting to complex and changeable driving conditions.

[0057] See also Figure 3-Figure 5 As shown, it illustrates how a pure electric vehicle can realize the functions of this invention through a power domain controller or a vehicle controller during actual driving. Figure 3-Figure 5 in The time before is the startup phase. The moment and after it is the rolling optimization stage, and the current moment is . In the startup phase, the power domain controller or vehicle controller obtains the speed of the preceding vehicle and the vehicle speed, obtains the vehicle speed in the predicted time domain through the online vehicle speed prediction module, and obtains the first performance preference expected value and the second performance preference expected value through the driver performance expectation quantification module based on the obtained vehicle speed, accelerator pedal strength and accelerator pedal strength change rate, and then selects the shifting law that matches the driver performance expectation quantification value in the offline calculated shifting law database, and then queries the shifting law table based on the accelerator pedal strength and predicted vehicle speed to obtain the target gear at the next moment. In the rolling optimization phase, at each moment, the power domain controller or vehicle controller uses the same algorithm as the startup phase to obtain the vehicle speed in the predicted time domain, as well as the first performance preference expected value and the second performance preference expected value, and then uses the DP algorithm to solve the personalized comprehensive performance optimal target gear at the next moment; when the car actually runs to the next moment, the rolling prediction of the vehicle speed and the quantified driving intention are used to solve the target gear using the DP algorithm, and then the target gear rolling decision is made in sequence until the controller exits the gear decision.

[0058] See also Figure 2 As shown, Figure 2 A specific embodiment of the present invention is provided, in which the gear decision is divided into two stages: a startup stage and a rolling optimization stage; is the current time (discrete interval is 1 second), The time when the startup phase ends and the rolling optimization phase begins. The length of the time domain for vehicle speed prediction, is the number of gears, is the number of shift commands, This is the startup phase, during which the target gear is obtained by looking up the offline shift schedule table; This is the rolling optimization stage, in which the target gear position is obtained by online rolling solution using dynamic programming; S1: Gear decision during the start-up phase: S11: Predict the vehicle speed using the vehicle speed prediction model.

[0059] S12: Driving intention quantification. The vehicle speed, accelerator pedal strength (i.e., accelerator pedal opening), and accelerator pedal strength change rate (i.e., accelerator pedal opening change rate) are collected in real time, and the driving intention is quantified through a driving intention quantification model to obtain a first performance preference expected value and a second performance preference expected value; S13: Determine the target gear position according to the offline gear shifting rule table. Based on the preset offline gear shifting rule database, select a gear shifting rule table that matches the first performance preference expected value and the second performance preference expected value; based on the selected gear shifting rule table, determine the target gear position according to the accelerator pedal strength and the vehicle speed prediction value; S14: Delay 1 second, .

[0060] S15: Determine whether to stop the gear decision, if so, stop the gear decision, otherwise continue to the next step.

[0061] S16: Judgment Is it less than If yes, then return to S11, otherwise enter the rolling optimization phase S2: Gear decision in rolling optimization phase: S21: predicting the vehicle speed using a vehicle speed prediction model; S22: Driving intention quantification, collecting vehicle speed, accelerator pedal strength (i.e., accelerator pedal opening), and accelerator pedal strength change rate (i.e., accelerator pedal opening change rate) in real time, and quantifying driving intention through a driving intention quantification model to obtain a first performance preference expected value and a second performance preference expected value; S23: DP solves the optimal gear, including two sub-steps: reverse solution and forward optimization: S231: Initialization parameters: gear index value , shift command index value and the solution phase index value ; S232: Reverse solution, the specific steps are as follows: S2321: Determine state variables , i.e. gear position ; S2322: Determine the control variables , i.e., the shift control command ; S2323: Calculate the state variables of the next stage according to the state transfer equation; S2324: Calculate the cost function value; S2325: Calculate the comprehensive objective function value; S2326: Determine whether the constraint conditions are met: If not, then And return to S2322, otherwise go to the next step; S2327: Judgment Is it less than :If so, then And return to S2322, otherwise go to the next step; S2328: Find and save the minimum objective function value and the corresponding control variable in the current state; S2329: Judgment Is it less than :If so, then And return to S2321, otherwise go to the next step; S23210: Judgment Is it equal to 1: If yes, proceed to the next step, otherwise And return to S2321; S233: Forward optimization, the specific steps are as follows: S2331: Set state variables is the current gear position; S2332: According to the current status The corresponding control variable , calculate the state variable of the next stage, i.e., the target gear position; S24: Delay 1 second, ; S25: Determine whether to stop the gear decision, if so, stop the gear decision, otherwise return to S21.

[0062] See also Figures 6 to 14 As shown, through the above method, this embodiment selects three time periods of 8:00-8:40, 13:40-14:20 and 17:20-18:00 on a certain road section (with a total length of 19.2 kilometers and 22 traffic lights), and adopts the technical solution of the present invention to make online gear decision for electric vehicles. The gear decision results of the three time periods are as follows: Figure 8 , Fig.11 and Fig.14 shown; according to Figure 8 , Fig.11 and Fig.14 It can be understood that the electric vehicle online gear decision method provided by the present invention has a high gear decision accuracy in the three time periods; the electric vehicle online gear decision method provided by the present invention combines the driver's personalized needs with the vehicle's driving status by constructing a driving intention quantification model and a vehicle speed prediction model, and integrates the dynamic programming algorithm to minimize the gear shifting cost rolling solution; based on the performance preference expected value and the vehicle speed prediction result, the gear decision result matching it is selected from the preset offline gear shifting law database to preliminarily match the working condition requirements. Based on the dynamic programming algorithm, and with the goal of minimizing the gear shifting cost, the reverse solution is combined with the forward optimization to extract the optimal gear decision sequence, so as to achieve accurate gear decision under complex working conditions, effectively improve the decision accuracy, personalization and intelligence level, and ensure the optimal overall performance of the whole vehicle.

[0063] See also Fig.15 , which shows a schematic diagram of the structure of an online gear decision system for an electric vehicle provided by an embodiment of the present invention, the system comprising: a data acquisition module configured to acquire vehicle state and driving maneuver data related to driving intention quantification and vehicle speed prediction of the target vehicle; A driving intention quantification module configured to construct a driving intention quantification model and calculate a performance preference expectation value based on vehicle states and driving manipulation data related to the driving intention quantification; A vehicle speed prediction module is configured to construct a vehicle speed prediction model and calculate a vehicle speed prediction result based on vehicle speed prediction related data; A first gear decision module is configured to select a gear shifting schedule table based on a preset offline gear shifting schedule database according to a performance preference expectation value, and extract a gear decision result using driving control data and a vehicle speed prediction result; The second gear decision module is configured to extract an optimal gear decision sequence based on a dynamic programming algorithm with the goal of minimizing the gear shifting cost.

[0064] It should be noted that the sequence of the above embodiments of the present invention is only for description and does not represent the advantages and disadvantages of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0065] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referenced to each other, and each embodiment focuses on the differences from other embodiments.

Claims

1. An online gear decision method for electric vehicles, characterized in that: The method comprises: Obtain vehicle state and driving maneuver data related to the target vehicle's driving intention quantification and speed prediction; Construct a driving intention quantification model and calculate the performance preference expectation value based on the vehicle status and driving manipulation data related to the driving intention quantification; Construct a vehicle speed prediction model and calculate the vehicle speed prediction result based on the vehicle speed prediction related data; Based on the preset offline shift schedule database, the shift schedule table is selected according to the expected performance preference value, and the gear decision result is extracted using the driving operation data and vehicle speed prediction results; Based on the dynamic programming algorithm, the optimal gear decision sequence is extracted with the goal of minimizing the gear shift cost.

2. The electric vehicle online gear decision method according to claim 1, characterized in that: Obtain vehicle state and driving maneuver data related to the target vehicle’s driving intention quantification and speed prediction, including: Obtaining the speed, accelerator pedal strength and accelerator pedal strength change rate of the target vehicle; The target vehicle's V2X communication module is used to establish communication with the front vehicle's V2X communication module to obtain the front vehicle's speed data.

3. The electric vehicle online gear decision method according to claim 1, characterized in that: Construct a driving intention quantification model and calculate the performance preference expectation value based on the vehicle status and driving manipulation data related to driving intention quantification, including: Calculating a first performance preference expected value according to the speed, accelerator pedal strength, and accelerator pedal strength change rate of the target vehicle; A second performance preference expected value is calculated based on the first performance preference expected value.

4. The electric vehicle online gear decision method according to claim 1, characterized in that: Construct a vehicle speed prediction model and calculate the vehicle speed prediction results based on the vehicle speed prediction related data, including: Construct a vehicle speed prediction model based on regression neural network; Construct a traffic flow speed information matrix based on the speed data of the target vehicle and the preceding vehicle; The vehicle speed prediction model is optimized by using the traffic speed information data matrix to calculate the vehicle speed prediction results.

5. The electric vehicle online gear decision method according to claim 1, characterized in that: Based on the preset offline shift schedule database, the shift schedule table is selected according to the expected performance preference value, and the gear decision results are extracted using the driving operation data and vehicle speed prediction results, including: In the startup phase, a shift schedule table matching the expected performance preference is selected from a preset offline shift schedule database; The target gear is determined from the gear shift schedule table based on the accelerator pedal strength and vehicle speed prediction results.

6. The electric vehicle online gear decision method according to any one of claims 1 to 5, characterized in that: Based on the dynamic programming algorithm, the optimal gear decision sequence is extracted with the goal of minimizing the gear shift cost, including: Initialize the gear index value, shift command index value and solution phase index value; Determine the gear position of the current stage according to the gear position index value of the current stage; Determine the shift instruction set of the current stage according to the shift instruction index value of the current stage; Calculate the gear set for the next stage according to the gear and shift instruction set for the current stage; Constructing a cost function based on the expected value of performance preference and a preset gear shift penalty function; Based on the cost function, with the goal of minimizing the gear shift cost, the objective function value of each stage and the optimal gear shift instruction that minimizes the objective function value are solved inversely from the last stage.

7. The electric vehicle online gear decision method according to claim 6, characterized in that: Based on the dynamic programming algorithm, the optimal gear decision sequence is extracted with the goal of minimizing the gear shift cost, which also includes: According to the inverse solution results, the optimal gear shift instructions for each stage are extracted, and starting from the current gear, the target gears of the subsequent stages are solved forward.

8. The electric vehicle online gear decision method according to claim 6, characterized in that: The expression of the cost function is: In the formula, Indicates The cost function value of the stage; Indicates The gear position of the stage; Indicates The shift command of the stage; represents the first performance preference expected value; represents the normalized first performance objective function value; represents the expected value of the second performance preference; represents the normalized second performance objective function value; Represents the preset gear shift penalty function.

9. The electric vehicle online gear decision method according to claim 8, characterized in that: The expression of the objective function is: In the formula, Indicates Stage objective function value; Indicates the time domain length of vehicle speed prediction; Indicates The objective function value of the stage.

10. The electric vehicle online gear decision system is characterized by: The electric vehicle online gear decision method according to any one of claims 1 to 9 is adopted, wherein the system comprises: a data acquisition module configured to acquire vehicle state and driving maneuver data related to driving intention quantification and vehicle speed prediction of the target vehicle; A driving intention quantification module configured to construct a driving intention quantification model and calculate a performance preference expectation value based on vehicle states and driving manipulation data related to the driving intention quantification; A vehicle speed prediction module is configured to construct a vehicle speed prediction model and calculate a vehicle speed prediction result based on vehicle speed prediction related data; A first gear decision module is configured to select a gear shifting schedule table based on a preset offline gear shifting schedule database according to a performance preference expectation value, and extract a gear decision result using driving control data and a vehicle speed prediction result; The second gear decision module is configured to extract an optimal gear decision sequence based on a dynamic programming algorithm with the goal of minimizing the gear shifting cost.

Images