CN117698688B - Hybrid transmission system mode intelligent switching method based on short-time vehicle speed prediction - Google Patents
Hybrid transmission system mode intelligent switching method based on short-time vehicle speed prediction Download PDFInfo
- Publication number
- CN117698688B CN117698688B CN202410166740.6A CN202410166740A CN117698688B CN 117698688 B CN117698688 B CN 117698688B CN 202410166740 A CN202410166740 A CN 202410166740A CN 117698688 B CN117698688 B CN 117698688B
- Authority
- CN
- China
- Prior art keywords
- vehicle speed
- short
- time
- switching
- speed prediction
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 230000005540 biological transmission Effects 0.000 title claims abstract description 36
- 230000008569 process Effects 0.000 claims abstract description 19
- 230000002787 reinforcement Effects 0.000 claims abstract description 7
- 230000006870 function Effects 0.000 claims description 22
- 230000001133 acceleration Effects 0.000 claims description 14
- 239000011159 matrix material Substances 0.000 claims description 14
- 238000012549 training Methods 0.000 claims description 13
- 238000004364 calculation method Methods 0.000 claims description 10
- 238000012937 correction Methods 0.000 claims description 6
- 239000012634 fragment Substances 0.000 claims description 6
- 238000005259 measurement Methods 0.000 claims description 5
- 238000005457 optimization Methods 0.000 claims description 5
- 230000001052 transient effect Effects 0.000 claims description 5
- 238000012360 testing method Methods 0.000 claims description 4
- 230000002159 abnormal effect Effects 0.000 claims description 3
- 230000005611 electricity Effects 0.000 claims description 3
- 239000000446 fuel Substances 0.000 claims description 3
- 238000012545 processing Methods 0.000 claims description 3
- 230000001360 synchronised effect Effects 0.000 claims description 3
- 238000013528 artificial neural network Methods 0.000 abstract description 4
- 238000001914 filtration Methods 0.000 description 3
- 230000009471 action Effects 0.000 description 2
- 230000008901 benefit Effects 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 230000000694 effects Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000007547 defect Effects 0.000 description 1
- 238000011156 evaluation Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000002401 inhibitory effect Effects 0.000 description 1
- 230000003993 interaction Effects 0.000 description 1
- 238000011160 research Methods 0.000 description 1
- 230000007704 transition Effects 0.000 description 1
Landscapes
- Hybrid Electric Vehicles (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The invention belongs to the technical field of working mode switching control of hybrid vehicles, and particularly relates to an intelligent mode switching method of a hybrid transmission system based on short-time vehicle speed prediction. Firstly, collecting real vehicle driving data; secondly, analyzing the working mode switching performance of the hybrid system, and searching a working mode switching centralized area; then, a short-time vehicle speed prediction model is established based on a two-way long-and-short-term memory neural network and an extended Kalman filter, so that the prediction of the future vehicle speed is realized; and finally, establishing a hybrid transmission system mode intelligent switching algorithm based on reinforcement learning, realizing hybrid mode intelligent switching by utilizing vehicle state information and a predicted vehicle speed sequence in a working mode switching centralized area, reducing the mode switching times, optimizing a mode switching control process, and improving the running smoothness and the travel economy.
Description
Technical Field
The invention belongs to the technical field of working mode switching control of hybrid vehicles, and particularly relates to an intelligent mode switching method of a hybrid transmission system based on short-time vehicle speed prediction.
Background
The hybrid electric vehicle is provided with a hybrid power system which consists of a plurality of power sources such as an engine, a motor and the like, so that the whole vehicle has a plurality of working modes by the cooperative working modes of different power sources. For example, the working modes of the series-parallel hybrid electric vehicle comprise a pure electric mode, a range-extending mode, a hybrid electric mode and the like. The working mode decision of the hybrid electric vehicle is determined by the vehicle speed, the required torque, the state of a power system such as a battery SoC and the like, and different working modes are adopted under different state states.
At present, a mode decision method based on rules is mainly adopted for the working mode selection of the hybrid electric vehicle, and the method has the following two optimizable items, namely the problems that the mode is frequently switched when state variables such as vehicle speed and the like change at a mode switching boundary, so that driving smoothness is poor and fuel consumption, emission and the like are increased are caused, and the problem that the mode switching time cannot be predicted, and the driving experience is poor due to long mode switching transient process under the existing rules. Thus, if future vehicle speed information can be predicted at the mode switch boundary, an early decision can be made in this region based on the predicted information to obtain the intended operation mode. If the expected working mode is consistent with the current working mode, mode switching is not performed, and the effect of inhibiting frequent mode switching is achieved. If the expected working mode is inconsistent with the current working mode, the expected working mode can be sent to the mode switching control module, the power source speed regulation, clutch oil pre-filling and other works are performed in advance, the mode switching time is shortened, and the driving experience is improved. Therefore, it is necessary to research an accurate short-time vehicle speed prediction method and to conduct intelligent switching decision of the hybrid transmission system mode based on the short-time vehicle speed prediction method.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a hybrid transmission system mode intelligent switching method based on short-time vehicle speed prediction. Firstly, collecting real vehicle driving data; secondly, analyzing the working mode switching performance of the hybrid system, and searching a working mode switching centralized area; then, a short-time vehicle speed prediction model is established based on a two-way long-and-short-term memory neural network and an extended Kalman filter, so that the prediction of the future vehicle speed is realized; and finally, establishing a hybrid transmission system mode intelligent switching algorithm based on reinforcement learning, realizing hybrid mode intelligent switching by utilizing vehicle state information and a predicted vehicle speed sequence in a working mode switching centralized area, reducing the mode switching times, optimizing a mode switching control process, and improving the running smoothness and the travel economy.
The technical scheme of the invention is as follows:
a hybrid transmission system mode intelligent switching method based on short-time vehicle speed prediction comprises the following steps:
s1, acquiring real vehicle driving data, and acquiring a natural driving data set after synchronous processing;
s2, analyzing the working mode switching performance of the hybrid system based on the natural driving data set, and searching a region in which the working modes are switched;
s3, building a BiLSTM-EKF short-time vehicle speed prediction model and finishing training;
s4, constructing a hybrid transmission system mode intelligent switching model based on reinforcement learning and finishing training;
and S5, based on the BiLSTM-EKF short-time vehicle speed prediction model and the intelligent switching mode of the hybrid transmission system mode, the intelligent decision of the switching boundary of the hybrid transmission system mode is realized.
Preferably, the step S1 specifically includes:
s1-1, selecting a plurality of driving routes including high speed, suburbs and cities;
s1-2, selecting different drivers to drive vehicles, completing the multiple driving routes, and collecting driving data;
and S1-3, performing time synchronization on the acquired data of different acquisition channels in the driving data, and removing abnormal values to obtain a natural driving data set.
Preferably, the natural driving dataset includes vehicle speed, longitudinal acceleration, lateral acceleration, accelerator pedal opening, brake pedal opening, soC, operating mode, and wheel end demand torque.
Preferably, the step S2 specifically includes:
and S2-1, extracting process fragments of starting and stopping of the engine and switching of the working modes in batches from the natural driving data set, namely, fragments from the target working mode to the actual working mode after an instruction is sent out.
And step S2-2, finding out a vehicle speed section near the working mode switching boundary from the process segment, namely a working mode switching concentrated area.
Preferably, the BiLSTM-EKF short-time vehicle speed prediction model in the step S3 includes a BiLSTM reference short-time vehicle speed prediction model and an EKF short-time vehicle speed prediction correction model; the training is based on a natural driving dataset.
Preferably, the BiLSTM reference short-time vehicle speed prediction model takes a vehicle speed, a longitudinal acceleration, a lateral acceleration, an accelerator pedal opening and a brake pedal opening at a historical moment as an input sequence and takes a predicted vehicle speed sequence at a future moment as an output sequence.
Preferably, the forward LSTM calculation formula in the BiLSTM reference short-time vehicle speed prediction model is as follows:
wherein,is->The forward layer output of time of day,>is->The forward layer output of time of day,>is->The input sequence of the moments in time,fis an LSTM unit function;
the backward LSTM calculation formula is as follows:
wherein,is->Time backward layer output, < >>Is->Time backward layer output, < >>Is->The input sequence of the moments in time,fis an LSTM unit function;
the calculation formula of the predicted vehicle speed output sequence is as follows:
wherein,is a forward layer weight matrix, +.>Is a backward layer weight matrix,>for the output sequence, i.e. the predicted vehicle speed sequence.
Preferably, the EKF short-time vehicle speed prediction correction model includes a prediction update equation and a measurement update equation.
Preferably, in the step S4, the intelligent switching model of the hybrid transmission system mode adopts a deep Q learning algorithm.
Preferably, the hybrid transmission system mode intelligent switching model is used for acquiring the vehicle speed in real timeWheelEnd demand torque、SoCFor real-time data input, an optimal estimated vehicle speed sequence output by a BiLSTM-EKF short-time vehicle speed prediction model is adopted>And taking an expected working mode as output as prediction data input, wherein the reward function consists of oil consumption, electricity consumption, working mode switching frequency, working mode transient switching time optimization and rule punishment under the test working condition.
Compared with the prior art, the invention has the beneficial effects that:
1. the hybrid transmission system mode intelligent switching method based on short-time vehicle speed prediction adopts a short-time vehicle speed prediction method of a two-way long-short-term memory neural network and an extended Kalman filter, the extended Kalman filter is suitable for a nonlinear system, and the time sequence predicted by BiLSTM can be corrected, so that the accuracy of vehicle speed prediction is further improved.
2. According to the hybrid transmission system mode intelligent switching method based on short-time vehicle speed prediction, the working mode switching intelligent decision is made in the working mode switching centralized area by adopting deep Q learning, the deep Q learning rewarding function considers the working mode switching punishment, the working mode switching optimization rewarding and the like, and the trained model achieves multi-objective optimization promotion.
3. The intelligent mode switching method for the hybrid transmission system based on short-time vehicle speed prediction can calculate the expected working mode by utilizing deep Q learning according to the vehicle speed prediction sequence, can effectively solve the problem of frequent switching of the working mode switching boundary, and can optimize the transient control process of the working mode switching of the expected working mode, thereby improving driving experience.
4. According to the intelligent mode switching method for the hybrid transmission system based on short-time vehicle speed prediction, a working mode decision method combining short-time vehicle speed prediction and reinforcement learning is adopted in a working mode switching centralized area, and a mode switching strategy based on rules is adopted outside a working mode switching boundary, so that the stability of the working mode decision of the hybrid transmission system is guaranteed.
Drawings
So that the manner in which the above recited embodiments of the present invention and the manner in which the same are attained and can be understood in detail, a more particular description of the invention, briefly summarized above, may be had by reference to the embodiments thereof which are illustrated in the appended drawings, which drawings are intended to be illustrative, and which drawings, however, are not to be construed as limiting the invention in any way, and in which other drawings may be obtained by those skilled in the art without the benefit of the appended claims.
FIG. 1 is a flow chart of the hybrid powertrain mode intelligent switching method based on short-term vehicle speed prediction of the present invention.
FIG. 2 is a graph of statistical features of engine on segments in an embodiment of the invention.
FIG. 3 is a schematic diagram of a mixed mode decision boundary in this embodiment of the invention.
Fig. 4 is a short-time vehicle speed prediction result diagram in the embodiment of the invention.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, without conflict, the embodiments of the present invention and features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.
Example 1
As shown in fig. 1, the hybrid transmission mode intelligent switching method based on short-time vehicle speed prediction specifically includes:
step one, acquiring real vehicle driving data, and obtaining a natural driving data set after synchronous processing.
(1) And selecting a plurality of driving routes covering various working conditions such as high speed, suburbs, cities and the like, so that the driving data and the speed of the vehicle cover the full speed area.
(2) And selecting different drivers to drive the vehicles, naturally driving the vehicles in a plurality of driving routes, and collecting driving data.
(3) And carrying out time synchronization on the acquired data of different acquisition channels in the driving data, and removing abnormal values to obtain a natural driving data set. The outliers include outliers that deviate significantly from the actual values, etc.
Wherein the natural driving data set includes vehicle speed, longitudinal acceleration, lateral acceleration, accelerator pedal opening, brake pedal opening, soC, working mode, wheel end required torque, etc., as shown in Table 1
TABLE 1
And secondly, analyzing the working mode switching performance of the hybrid system based on the natural driving data set, and searching a region in which the working modes are switched.
(1) And extracting process fragments for starting and stopping the engine and switching the working modes in batches from the natural driving data set, namely, fragments for switching the target working mode to the actual working mode after an instruction is sent out.
(2) Looking for a region of concentrated operating mode switching, fig. 2 shows a region where the engine is frequently started and stopped in the driving data. And analyzing the distribution of variables such as the speed, the wheel end required torque, the SoC and the like of the engine starting and stopping and working mode switching events. Fig. 3 shows a working mode switching boundary of a certain two-gear series-parallel hybrid electric vehicle, line 1 and Line 2 are vehicle speed boundaries for switching the working modes, line 3 is a torque boundary for switching the working modes, and when the required torque of a wheel end is smaller than 0, the working modes exit from the first gear. Therefore, the working mode switching centralized area is a vehicle speed area near the working mode switching boundary, if the engine starting boundary is 55km/h, the working mode switching centralized area is 50-60km/h, and the following short-time vehicle speed prediction and intelligent mode switching decision-making action areas are all in the area.
And thirdly, establishing a short-time vehicle speed prediction model based on a two-way long-short-term memory neural network BiLSTM and an extended Kalman filter EKF, and completing training.
(1) In the natural driving dataset, the vehicle speed, longitudinal acceleration, lateral acceleration, accelerator pedal opening and brake pedal opening within 10 seconds of history are taken as input sequences.
(2) And constructing a BiLSTM reference short-time vehicle speed prediction model.
The invention adopts BiLSTM network to build a reference short-time speed prediction model, the input sequence in the step (1) is respectively connected into a forward LSTM and a backward LSTM, and then two hidden layers are connected together and are connected to an output layer together for prediction.
The forward LSTM calculation formula is as follows:
in the middle ofIs->The forward layer output of time of day,>is->The forward layer output of time of day,>is->Input sequence of moments,/->Is an LSTM unit function.
The backward LSTM calculation formula is as follows:
in the method, in the process of the invention,is->Time backward layer output, < >>Is->Time backward layer output, < >>Is->The input sequence of the moments in time,is an LSTM unit function.
The calculation formula of the predicted vehicle speed output sequence is as follows:
in the method, in the process of the invention,is a forward layer weight matrix, +.>Is a backward layer weight matrix,>a sequence is output for predicting vehicle speed.
(3) And training a BiLSTM reference short-time vehicle speed prediction model.
With a history of vehicle speed, longitudinal acceleration, and the like within 10 seconds,The lateral acceleration, the accelerator pedal opening and the brake pedal opening are used as inputs of a BiLSTM reference short-time vehicle speed prediction model, and a predicted vehicle speed output sequence of 10s in the future is predictedAnd training and testing a BiLSTM reference short-time vehicle speed prediction model by taking the root mean square error as an evaluation index, wherein the root mean square error has the following calculation formula:
in the method, in the process of the invention,is->Root mean square error in the time domain of individual time instants +.>To predict the time domain length, +.>Predicting the number of time points for the full-travel vehicle speed, +.>First->Time of day->Speed predicted value of seconds,/>Is->Time of dayFuture->Actual value of vehicle speed in seconds.
(4) And constructing an EKF short-time vehicle speed prediction correction model based on the BiLSTM standard short-time vehicle speed prediction model.
The extended Kalman filtering algorithm comprises a prediction update equation and a measurement update equation, wherein the prediction update equation is responsible for calculating the values of the prediction state and the prior estimation error covariance to construct the prior estimation of the next time state, and the measurement update equation is responsible for feedback to construct an improved posterior estimation. In the present invention, a vehicle speed sequence of 10 seconds in the future is estimated first, and then the first is calculatedA priori estimated covariance matrix for each moment +.>Completing a prediction update equation; subsequently calculate the Kalman gain->And based on the predicted vehicle speed sequence->Calculating optimal estimated vehicle speed sequence->As an output corrected by the EKF short-time vehicle speed prediction correction model.
The predictive update equation for extended kalman filtering is:
in the method, in the process of the invention,for estimating the vehicle speed sequence, the vehicle speed sequence is>Estimating a system state equation for vehicle speed,/->Is->Real speed of current point when predicting each moment, +.>For the input of the vehicle speed estimation system, < >>Is process noise->Is->Error covariance matrix of real vehicle speed and optimal estimated vehicle speed at moment +.>For state transition matrix>Is a state noise covariance matrix.
The measurement update equation for extended kalman filtering is:
in the method, in the process of the invention,for the covariance matrix of the observation noise, +.>Is->Jacobian matrix of function versus state>For observing noise +.>Is->Function (F)>Is->Covariance matrix between real value and optimal estimated value at each moment +.>Representing the identity matrix.
The vehicle speed prediction effect in the embodiment is shown in fig. 4, the vehicle speed after 10s is predicted by using a BiLSTM-EKF short-time vehicle speed prediction model, the vehicle speed prediction trend meets the use requirement, and the accuracy is improved compared with that of a single BiLSTM model and a single LSTM model.
And fourthly, constructing a hybrid transmission system mode intelligent switching model based on reinforcement learning and finishing training.
In the second step, a boundary of working mode switching is determined according to the natural driving data set, intelligent decision of an expected working mode is realized by short-time speed prediction in a working mode switching concentrated area, and specifically, a deep Q learning algorithm is adopted to realize a decision function. The method for building and training the intelligent switching model of the hybrid transmission system mode based on reinforcement learning comprises the following steps:
(1) Determining hybrid powertrain mode intelligent switching model inputs (state variables): with speed of vehicle collected in real timeWheel end demand torque->、/>For real-time data input, an optimal estimated vehicle speed sequence output by a BiLSTM-EKF short-time vehicle speed prediction model is adopted>As predictive data input.
Specifically, the state variablesThe method comprises the following steps:
(2) Determining the output (action variable) of a hybrid transmission mode intelligent switching model: the expected working modes, such as the two-gear series-parallel hybrid system, are a pure electric mode, a range extending mode, a first gear of a hybrid mode and a second gear of the hybrid mode.
(3) Determining a reward function of the hybrid transmission mode intelligent switching model: the reward function comprises five parts, including oil consumption, electricity consumption, working mode switching frequency, working mode transient switching time optimization and rule punishment in a prediction domain under a test working condition.
Specifically, the objective function is:
in the method, in the process of the invention,for the engine fuel consumption rate based on the engine speed and the engine torque +.>、/>、、/>And->Weight coefficients, which are all objective functions, +.>、/>、/>Engine speed, engine torque and engine power, respectively,/->Is the final value SoC, ">Is the start SoC->Increasing the times for switching the working mode under all working conditions, +.>For the average time increment of the operation mode switching control process in the prediction domain, +.>A penalty term for deviating from the operating mode switching region. Thereby realizing the only in the working mode switching concentrated areaAnd making a row decision, and making decisions in other areas according to the basic rules. The reward function is the inverse of the objective function.
(4) And when the average error of the Q value reaches an expected value or the training is finished after the maximum training cycle is reached, embedding the intelligent switching model of the hybrid transmission system mode into the whole vehicle controller, and realizing the intelligent decision of the switching boundary of the working mode.
In the present invention, unless explicitly specified and limited otherwise, the terms "mounted," "connected," "secured," and the like are to be construed broadly, and may be, for example, fixedly connected, detachably connected, or integrally formed; can be mechanically or electrically connected; can be directly connected or indirectly connected through an intermediate medium, and can be communicated with the inside of two elements or the interaction relationship of the two elements. The specific meaning of the above terms in the present invention can be understood by those of ordinary skill in the art according to the specific circumstances.
In the present invention, unless expressly stated or limited otherwise, a first feature "above" or "below" a second feature may include both the first and second features being in direct contact, as well as the first and second features not being in direct contact but being in contact with each other through additional features therebetween. Moreover, a first feature being "above," "over" and "on" a second feature includes the first feature being directly above and obliquely above the second feature, or simply indicating that the first feature is higher in level than the second feature. The first feature being "under", "below" and "beneath" the second feature includes the first feature being directly under and obliquely below the second feature, or simply means that the first feature is less level than the second feature.
In the present invention, the terms "first," "second," "third," "fourth" are used for descriptive purposes only and are not to be construed as indicating or implying relative importance. The term "plurality" refers to two or more, unless explicitly defined otherwise.
The above description is only of the preferred embodiments of the present invention and is not intended to limit the present invention, but various modifications and variations can be made to the present invention by those skilled in the art. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (8)
1. The intelligent mode switching method of the hybrid transmission system based on short-time vehicle speed prediction is characterized by comprising the following steps of:
s1, acquiring real vehicle driving data, and acquiring a natural driving data set after synchronous processing;
s2, analyzing the working mode switching performance of the hybrid system based on the natural driving data set, and searching a region in which the working modes are switched;
s3, building a BiLSTM-EKF short-time vehicle speed prediction model and finishing training;
s4, constructing a hybrid transmission system mode intelligent switching model based on reinforcement learning and finishing training;
s5, based on a BiLSTM-EKF short-time vehicle speed prediction model and a hybrid transmission system mode intelligent switching model, intelligent decision of a hybrid transmission system mode switching boundary is realized;
the natural driving data set comprises a vehicle speed, a longitudinal acceleration, a transverse acceleration, an accelerator pedal opening, a brake pedal opening, soC, a working mode and wheel end required torque;
the mode intelligent switching model of the hybrid transmission system is used for acquiring vehicle speed in real timeWheel end demand torque->、SoCFor real-time data input, an optimal estimated vehicle speed sequence output by a BiLSTM-EKF short-time vehicle speed prediction model is adopted>As predicted data input, the expected working mode is taken as output, and the reward function is composed of oil consumption, electricity consumption, working mode switching frequency and work efficiency under the test working conditionMode transient switching time optimization and rule punishment are performed;
the objective function is:
in the method, in the process of the invention,for the engine fuel consumption rate based on the engine speed and the engine torque +.>、/>、/>、And->Weight coefficients, which are all objective functions, +.>To predict the time domain length, +.>、/>、/>Engine speed, engine torque and engine power, respectively,/->Is the final value SoC，/>Is the start SoC->Increasing the times for switching the working mode under all working conditions, +.>For the average time increment of the operation mode switching control process in the prediction domain, +.>The penalty term deviating from the working mode switching region is used for realizing decision making only in the working mode switching centralized region, and decision making is carried out in other regions according to basic rules; the reward function is the inverse of the objective function.
2. The hybrid transmission mode intelligent switching method based on short-time vehicle speed prediction according to claim 1, wherein the step S1 specifically includes:
s1-1, selecting a plurality of driving routes including high speed, suburbs and cities;
s1-2, selecting different drivers to drive vehicles, completing the multiple driving routes, and collecting driving data;
and S1-3, performing time synchronization on the acquired data of different acquisition channels in the driving data, and removing abnormal values to obtain a natural driving data set.
3. The hybrid transmission mode intelligent switching method based on short-time vehicle speed prediction according to claim 1, wherein the step S2 specifically includes:
s2-1, extracting process fragments for starting and stopping an engine and switching the working modes in batches from a natural driving data set, namely, fragments for switching the target working mode to an actual working mode after an instruction is sent out;
and step S2-2, finding out a vehicle speed section near the working mode switching boundary from the process segment, namely a working mode switching concentrated area.
4. The hybrid transmission system mode intelligent switching method based on short-time vehicle speed prediction according to claim 1, wherein the BiLSTM-EKF short-time vehicle speed prediction model in step S3 includes a BiLSTM reference short-time vehicle speed prediction model and an EKF short-time vehicle speed prediction correction model; the training is based on a natural driving dataset.
5. The intelligent switching method of hybrid transmission modes based on short-time vehicle speed prediction according to claim 4, wherein the BiLSTM reference short-time vehicle speed prediction model takes a vehicle speed, longitudinal acceleration, lateral acceleration, accelerator pedal opening and brake pedal opening at historical moments as input sequences and a predicted vehicle speed sequence at future moments as output sequences.
6. The hybrid powertrain mode intelligent switching method based on short-time vehicle speed prediction according to claim 5, wherein a forward LSTM calculation formula in the BiLSTM reference short-time vehicle speed prediction model is as follows:
wherein,is->The forward layer output of time of day,>is->The forward layer output of time of day,>is->The input sequence of the moments in time,fis an LSTM unit function;
the backward LSTM calculation formula is as follows:
wherein,is->Time backward layer output, < >>Is->Time backward layer output, < >>Is->The input sequence of the moments in time,fis an LSTM unit function;
the calculation formula of the predicted vehicle speed output sequence is as follows:
wherein,is the forward layer output weight matrix, +.>Is the backward layer output weight matrix, +.>A sequence is output for predicting vehicle speed.
7. The hybrid powertrain mode intelligent switching method based on short-time vehicle speed prediction according to claim 6, wherein the EKF short-time vehicle speed prediction correction model includes a prediction update equation and a measurement update equation.
8. The hybrid transmission system mode intelligent switching method based on short-time vehicle speed prediction according to claim 1, wherein the hybrid transmission system mode intelligent switching model in step S4 adopts a deep Q learning algorithm.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410166740.6A CN117698688B (en) | 2024-02-06 | 2024-02-06 | Hybrid transmission system mode intelligent switching method based on short-time vehicle speed prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202410166740.6A CN117698688B (en) | 2024-02-06 | 2024-02-06 | Hybrid transmission system mode intelligent switching method based on short-time vehicle speed prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117698688A CN117698688A (en) | 2024-03-15 |
| CN117698688B true CN117698688B (en) | 2024-04-09 |
Family
ID=90146546
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202410166740.6A Active CN117698688B (en) | 2024-02-06 | 2024-02-06 | Hybrid transmission system mode intelligent switching method based on short-time vehicle speed prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117698688B (en) |
Citations (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20180070290A (en) * | 2016-12-16 | 2018-06-26 | 현대자동차주식회사 | Hybrid vehicle and method of controlling mode transition |
| CN113221449A (en) * | 2021-04-27 | 2021-08-06 | 中国科学院国家空间科学中心 | Ship track real-time prediction method and system based on optimal strategy learning |
| CN114048889A (en) * | 2021-10-08 | 2022-02-15 | 天津大学 | Aircraft trajectory prediction method based on long-term and short-term memory network |
| CN114620051A (en) * | 2022-03-15 | 2022-06-14 | 燕山大学 | Time-varying domain prediction energy management method and device |
| CN114802180A (en) * | 2022-05-19 | 2022-07-29 | 广西大学 | Mode prediction system and method for hybrid electric vehicle power system coordination control |
| CN115257691A (en) * | 2022-07-28 | 2022-11-01 | 同济大学 | Hybrid electric vehicle mode switching control method based on reinforcement learning |
| CN115496257A (en) * | 2022-07-23 | 2022-12-20 | 南京信息工程大学 | Short-term vehicle speed prediction based on space-time fusion |
| CN116070783A (en) * | 2023-03-07 | 2023-05-05 | 北京航空航天大学 | Learning type energy management method of hybrid transmission system under commute section |
| CN116946107A (en) * | 2023-09-21 | 2023-10-27 | 北京航空航天大学 | Hybrid system mode decision and power distribution method under energy track following |
-
2024
- 2024-02-06 CN CN202410166740.6A patent/CN117698688B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20180070290A (en) * | 2016-12-16 | 2018-06-26 | 현대자동차주식회사 | Hybrid vehicle and method of controlling mode transition |
| CN108216184A (en) * | 2016-12-16 | 2018-06-29 | 现代自动车株式会社 | Hybrid vehicle and the method for control model transformation |
| CN113221449A (en) * | 2021-04-27 | 2021-08-06 | 中国科学院国家空间科学中心 | Ship track real-time prediction method and system based on optimal strategy learning |
| CN114048889A (en) * | 2021-10-08 | 2022-02-15 | 天津大学 | Aircraft trajectory prediction method based on long-term and short-term memory network |
| CN114620051A (en) * | 2022-03-15 | 2022-06-14 | 燕山大学 | Time-varying domain prediction energy management method and device |
| CN114802180A (en) * | 2022-05-19 | 2022-07-29 | 广西大学 | Mode prediction system and method for hybrid electric vehicle power system coordination control |
| CN115496257A (en) * | 2022-07-23 | 2022-12-20 | 南京信息工程大学 | Short-term vehicle speed prediction based on space-time fusion |
| CN115257691A (en) * | 2022-07-28 | 2022-11-01 | 同济大学 | Hybrid electric vehicle mode switching control method based on reinforcement learning |
| CN116070783A (en) * | 2023-03-07 | 2023-05-05 | 北京航空航天大学 | Learning type energy management method of hybrid transmission system under commute section |
| CN116946107A (en) * | 2023-09-21 | 2023-10-27 | 北京航空航天大学 | Hybrid system mode decision and power distribution method under energy track following |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117698688A (en) | 2024-03-15 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108087541B (en) | Multi-performance comprehensive optimal gear decision system of automobile stepped automatic transmission | |
| CN110775065B (en) | Hybrid electric vehicle battery life prediction method based on working condition recognition | |
| CN110775043B (en) | Hybrid electric vehicle energy optimization method based on battery life attenuation pattern recognition | |
| WO2021114742A1 (en) | Comprehensive energy prediction and management method for hybrid electric vehicle | |
| CN110562239B (en) | Variable-domain optimal energy management control method and device based on demand power prediction | |
| CN108791270B (en) | Power component working point regulation and control method of hybrid power vehicle based on working condition prediction | |
| CN104842797B (en) | Electric automobile future average current drain is estimated, remaining continual mileage method of estimation and system | |
| CN108909702A (en) | A kind of plug-in hybrid-power automobile energy management method and system | |
| CN111619545B (en) | Hybrid electric vehicle energy management method based on traffic information | |
| CN110985651B (en) | Automatic transmission multi-parameter fusion gear shifting strategy based on prediction | |
| CN110341546B (en) | Method for estimating endurance mileage of pure electric vehicle | |
| CN111267827B (en) | Energy management method and system for hybrid electric vehicle | |
| CN109910866A (en) | Hybrid vehicle energy management method and system based on road condition predicting | |
| CN112429005B (en) | Pure electric vehicle personalized gear shifting rule optimization method considering transmission efficiency and application | |
| CN105035079A (en) | Power switching coordination control method of coaxial parallel hybrid electric vehicle with engine torque observer | |
| CN112721907B (en) | Efficient energy management method of network-connected hybrid electric vehicle considering road gradient | |
| CN112590760B (en) | Double-motor hybrid electric vehicle energy management system considering mode switching frequency | |
| CN112937547A (en) | Plug-in hybrid power bus energy management method based on global working conditions | |
| CN112319462B (en) | Energy management method for plug-in hybrid electric vehicle | |
| Chrenko et al. | Novel classification of control strategies for hybrid electric vehicles | |
| CN107839541A (en) | Electric vehicle control method | |
| CN113479186A (en) | Hybrid electric vehicle energy management strategy optimization method | |
| CN110509914A (en) | A kind of energy consumption optimization method of parallel hybrid electric vehicle | |
| von Parallelhybridfahrzeugen | On the power split control of parallel hybrid vehicles: from global optimization towards real-time control | |
| CN107323308A (en) | A kind of fuel cell car tractive torque computational methods with predictive ability |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |