CN115923764A - Hybrid electric vehicle energy control method and device based on running condition prediction - Google Patents
Hybrid electric vehicle energy control method and device based on running condition prediction Download PDFInfo
- Publication number
- CN115923764A CN115923764A CN202211441575.8A CN202211441575A CN115923764A CN 115923764 A CN115923764 A CN 115923764A CN 202211441575 A CN202211441575 A CN 202211441575A CN 115923764 A CN115923764 A CN 115923764A
- Authority
- CN
- China
- Prior art keywords
- condition
- vehicle
- power
- hybrid electric
- energy control
- 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.)
- Pending
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/60—Other road transportation technologies with climate change mitigation effect
- Y02T10/62—Hybrid vehicles
Landscapes
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The disclosure relates to a hybrid electric vehicle energy control method and device based on driving condition prediction. Wherein, the method comprises the following steps: analyzing the power characteristic of a hybrid electric vehicle to generate a power characteristic curve of the hybrid electric vehicle; based on a preset LSTM neural network model, taking the current speed and the current power of the hybrid electric vehicle as input, predicting the running condition, and generating the predicted speed and the predicted power of the hybrid electric vehicle; and generating a predicted vehicle condition based on the predicted vehicle speed, the predicted power and the power characteristic curve, and generating a vehicle control command based on the predicted vehicle condition and a preset energy control strategy to complete energy control on the hybrid electric vehicle. The method and the device realize the prediction of the running condition of the hybrid electric vehicle, perform energy flow control in advance according to the prediction result, reduce the frequency of starting the engine to enter a forced power supply mode due to electric quantity reduction, and improve the energy utilization efficiency.
Description
Technical Field
The disclosure relates to the field of new energy and electric automobiles, in particular to a hybrid electric vehicle energy control method and device based on running condition prediction.
Background
The power assembly of the plug-in hybrid electric vehicle comprises a fuel engine, a driving motor, a power battery and the like. The energy control of the hybrid electric vehicle is to reasonably coordinate and control the energy flow among all the powertrain components. The fuel-saving effect of the hybrid electric vehicle comes from energy recovery, engine idling working condition cancellation and benefits brought by running the engine in a high-efficiency area, but if the fuel-saving effect is improved to a higher level, a good energy control strategy cannot be left. Since each conversion of mechanical and electrical energy is lost efficiently, a good energy control strategy should minimize the number of conversions between mechanical and electrical energy for hybrid vehicle models in which the engine may participate in propulsion.
The setting of the energy control strategy of the existing hybrid electric vehicle needs to set the switching time of various energy flow modes according to the characteristic parameters of an engine, a driving motor and a power battery and a certain threshold value. Once the energy control strategy is set, it will not change. The energy control strategy has no high generalization and cannot be applied to all automobile running conditions. The energy control method of the single-motor plug-in hybrid electric vehicle distinguishes a driving mode in an electric quantity consumption stage, a driving mode in an electric quantity maintenance stage and a driving mode in an electric quantity supplement stage by setting a threshold value. The energy control strategy lacks prediction on the running condition of the automobile, and if the electric quantity is maintained at a lower level and a long-time high-power running demand follows, the automobile cannot quickly supplement the electric quantity, so that the automobile can continue running only if the automobile stops for standing and supplementing the electric quantity; in the energy control method based on the driving conditions, the driving road condition information of the automobile is calculated in advance by depending on a vehicle road cloud system, the corresponding working condition type is found by extracting the main characteristics and clustering the data, and a response energy control strategy is executed. However, the condition information cannot be defined accurately only by the extracted main characteristics (including running time, maximum speed, maximum acceleration, average acceleration and average deceleration), and the road condition of the automobile in the running process varies greatly.
Accordingly, there is a need for one or more methods to address the above-mentioned problems.
It is to be noted that the information disclosed in the above background section is only for enhancement of understanding of the background of the present disclosure, and thus may include information that does not constitute prior art known to those of ordinary skill in the art.
Disclosure of Invention
An object of the present disclosure is to provide a hybrid electric vehicle energy control method and apparatus based on driving condition prediction, thereby overcoming, at least to some extent, one or more of the problems due to the limitations and disadvantages of the related art.
According to one aspect of the present disclosure, there is provided a hybrid electric vehicle energy control method based on a driving condition prediction, including:
analyzing the power characteristic of the hybrid electric vehicle, and generating a power characteristic curve of the hybrid electric vehicle under a vehicle speed-power coordinate system;
based on a preset LSTM neural network model, taking the current speed and the current power of the hybrid electric vehicle as input, predicting the running condition of the hybrid electric vehicle, and generating the predicted speed and the predicted power of the hybrid electric vehicle;
and generating a predicted vehicle condition based on the predicted vehicle speed, the predicted power and the power characteristic curve, and generating a vehicle control command based on the predicted vehicle condition and a preset energy control strategy to complete energy control on the hybrid electric vehicle.
In an exemplary embodiment of the present disclosure, the method further comprises:
analyzing the series-parallel connection state of an engine and a driving motor of the hybrid electric vehicle and the power characteristic during combined operation, and generating a power characteristic curve of the hybrid electric vehicle and the charging and discharging state of the electric vehicle corresponding to the power characteristic curve under a vehicle speed-power coordinate system, wherein the power characteristic curve comprises a target torque curve, an external characteristic power curve of the driving motor, an external characteristic power curve of the engine, an engine economic working condition output power lower limit curve and a power curve when the engine and the driving motor work in parallel.
In an exemplary embodiment of the present disclosure, the method further includes training the LSTM neural network model based on historical operating parameters of the hybrid electric vehicle to generate the preset LSTM neural network model:
and based on the preset hyper-parameters of the LSTM neural network model, the historical speed and the historical power of the hybrid electric vehicle are used as input, and the training of the LSTM neural network model is completed according to loss function judgment.
In an exemplary embodiment of the present disclosure, the method further comprises:
establishing an LSTM neural network model population containing a preset number of LSTM neural network models, and setting a first random hyper-parameter for each LSTM neural network model in the LSTM neural network model population;
based on random super parameters set by each LSTM neural network model, taking the historical speed and the historical power of the hybrid electric vehicle as input, and judging according to a loss function, and performing a first round of training on the LSTM neural network models in the LSTM neural network model population to generate a first super parameter value;
calculating Euclidean distances of first hyper-parameter values of each LSTM neural network model, screening and generating second hyper-parameter values based on the Euclidean distances, and generating second random hyper-parameters based on the second hyper-parameter values and random scale factors;
setting a second random hyper-parameter based on each LSTM neural network model, taking the historical speed and the historical power of the hybrid electric vehicle as input, and performing a second round of training on the LSTM neural network models in the LSTM neural network model population according to loss function judgment.
In an exemplary embodiment of the present disclosure, the method further comprises:
and carrying out preset training on each LSTM neural network model in the LSTM neural network model population based on an iterative hyper-parameter method to generate a preset LSTM neural network model.
In an exemplary embodiment of the disclosure, the energy control strategy in the method comprises:
when the current vehicle condition is a pure electric working condition, the vehicle condition is predicted to be the pure electric working condition, and the energy control strategy is that when the electric quantity is lower than a threshold value, an engine is started to charge a power battery;
when the current vehicle condition is a pure electric condition, predicting the vehicle condition to enter a pure oil condition, and not processing the energy control strategy;
when the current vehicle condition is a pure electric condition, the vehicle condition is predicted to enter a parallel connection condition, and the energy control strategy is to start an engine in advance and charge a power battery;
when the current vehicle condition is a pure oil condition, predicting the vehicle condition to be the pure oil condition, and not processing the energy control strategy;
when the current vehicle condition is a pure oil condition, the vehicle condition is predicted to enter a parallel connection condition, and the energy control strategy is to preset partial power of the engine for charging and store electric energy in advance;
when the current vehicle condition is a pure oil condition, the vehicle condition is predicted to enter the pure oil condition, and the energy control strategy is to preset partial power of the engine for charging and store electric energy in advance;
when the current vehicle condition is a parallel working condition, the predicted vehicle condition is the parallel working condition, and the energy control strategy is to preset partial power of the engine for charging so as to prevent the reduction of the battery power caused by the long-time consumption of the power battery power;
when the current vehicle condition is a parallel working condition, predicting the vehicle condition to enter a series working condition, and not processing the energy control strategy;
when the current vehicle condition is a parallel working condition, the vehicle condition is predicted to enter a pure electric working condition, and the energy control strategy is to preset partial power of the engine for charging so as to prevent the reduction of the battery electric quantity caused by the long-time consumption of the electric quantity of the power battery;
when the current vehicle condition is a parallel working condition, predicting the vehicle condition to enter a pure oil working condition, and not processing the energy control strategy;
when the current vehicle condition is a series working condition, the predicted vehicle condition is the series working condition, the energy control strategy is to enter a power following mode, the power output by the engine follows the required power of the driving motor, and the power battery is not charged and does not need to be discharged outwards;
when the current vehicle condition is a series working condition, predicting the vehicle condition to enter a parallel working condition, and not processing the energy control strategy;
when the current vehicle condition is a series working condition, the vehicle condition is predicted to enter a pure electric working condition, and the energy control strategy is not processed.
In one aspect of the present disclosure, there is provided a hybrid electric vehicle energy control apparatus predicted based on a driving condition, including:
the power characteristic curve generating module is used for analyzing the power characteristic of the hybrid electric vehicle and generating the power characteristic curve of the hybrid electric vehicle under a vehicle speed-power coordinate system;
the vehicle condition prediction module is used for predicting the running condition of the hybrid electric vehicle by taking the current speed and the current power of the hybrid electric vehicle as input based on a preset LSTM neural network model, and generating the predicted speed and the predicted power of the hybrid electric vehicle;
and the energy control module is used for generating a predicted vehicle condition based on the predicted vehicle speed, the predicted power and the power characteristic curve, generating a vehicle control command based on the predicted vehicle condition and a preset energy control strategy and finishing energy control on the hybrid electric vehicle.
In an exemplary embodiment of the present disclosure, a hybrid electric vehicle energy control method based on a driving condition prediction, wherein the method comprises: analyzing the power characteristic of a hybrid electric vehicle to generate a power characteristic curve of the hybrid electric vehicle; based on a preset LSTM neural network model, taking the current speed and the current power of the hybrid electric vehicle as input, predicting the running condition, and generating the predicted speed and the predicted power of the hybrid electric vehicle; and generating a predicted vehicle condition based on the predicted vehicle speed, the predicted power and the power characteristic curve, and generating a vehicle control command based on the predicted vehicle condition and a preset energy control strategy to complete energy control on the hybrid electric vehicle. The method and the device realize the prediction of the running condition of the hybrid electric vehicle, perform energy flow control in advance according to the prediction result, reduce the frequency of starting the engine to enter a forced power supply mode due to the reduction of electric quantity, and improve the energy utilization efficiency.
It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the disclosure.
Drawings
The above and other features and advantages of the present disclosure will become more apparent by describing in detail exemplary embodiments thereof with reference to the attached drawings.
FIG. 1 illustrates a flow chart of a hybrid electric vehicle energy control method based on travel condition prediction according to an exemplary embodiment of the present disclosure;
FIG. 2 is a power characteristic diagram illustrating a hybrid electric vehicle energy control method based on driving condition prediction according to an exemplary embodiment of the present disclosure;
FIG. 3 is a schematic diagram illustrating predicted vehicle conditions for a hybrid electric vehicle energy control method based on driving condition prediction according to an exemplary embodiment of the present disclosure;
FIG. 4 shows a schematic block diagram of a hybrid electric vehicle energy control device based on a prediction of driving conditions according to an exemplary embodiment of the present disclosure.
Detailed Description
Example embodiments will now be described more fully with reference to the accompanying drawings. Example embodiments may, however, be embodied in many different forms and should not be construed as limited to the embodiments set forth herein; rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the concept of example embodiments to those skilled in the art. The same reference numerals denote the same or similar parts in the drawings, and thus, a repetitive description thereof will be omitted.
Furthermore, the described features, structures, or characteristics may be combined in any suitable manner in one or more embodiments. In the following description, numerous specific details are provided to give a thorough understanding of embodiments of the disclosure. One skilled in the relevant art will recognize, however, that the embodiments of the disclosure can be practiced without one or more of the specific details, or with other methods, components, materials, devices, steps, and so forth. In other instances, well-known structures, methods, devices, implementations, materials, or operations are not shown or described in detail to avoid obscuring aspects of the disclosure.
The block diagrams shown in the figures are functional entities only and do not necessarily correspond to physically separate entities. That is, these functional entities may be implemented in software, or in one or more software-hardened modules, or in different networks and/or processor devices and/or microcontroller devices.
In the exemplary embodiment, first, a hybrid electric vehicle energy control method based on the prediction of the running condition is provided; referring to fig. 1, the hybrid electric vehicle energy control method based on the driving condition prediction may include the steps of:
step S110, analyzing the power characteristic of the hybrid electric vehicle, and generating a power characteristic curve of the hybrid electric vehicle under a vehicle speed-power coordinate system;
step S120, based on a preset LSTM neural network model, taking the current speed and the current power of the hybrid electric vehicle as input, predicting the running condition of the hybrid electric vehicle, and generating the predicted speed and the predicted power of the hybrid electric vehicle;
and step S130, generating a predicted vehicle condition based on the predicted vehicle speed, the predicted power and the power characteristic curve, generating a vehicle control command based on the predicted vehicle condition and a preset energy control strategy, and finishing energy control on the hybrid electric vehicle.
In an exemplary embodiment of the present disclosure, a hybrid electric vehicle energy control method based on a driving condition prediction, wherein the method includes: analyzing the power characteristic of a hybrid electric vehicle to generate a power characteristic curve of the hybrid electric vehicle; based on a preset LSTM neural network model, taking the current speed and the current power of the hybrid electric vehicle as input, predicting the running condition, and generating the predicted speed and the predicted power of the hybrid electric vehicle; and generating a predicted vehicle condition based on the predicted vehicle speed, the predicted power and the power characteristic curve, and generating a vehicle control command based on the predicted vehicle condition and a preset energy control strategy to complete energy control on the hybrid electric vehicle. The method and the device realize the prediction of the running condition of the hybrid electric vehicle, perform energy flow control in advance according to the prediction result, reduce the frequency of starting the engine to enter a forced power supply mode due to electric quantity reduction, and improve the energy utilization efficiency.
Next, a hybrid electric vehicle energy control method based on the prediction of the running condition in this example embodiment will be further described.
In step S110, the power characteristics of the hybrid electric vehicle may be analyzed, and a power characteristic curve of the hybrid electric vehicle may be generated in a vehicle speed-power coordinate system.
In an embodiment of the present example, the method further comprises:
analyzing the series-parallel connection state of an engine and a driving motor of the hybrid electric vehicle and the power characteristic during combined operation, and generating a power characteristic curve of the hybrid electric vehicle and the charging and discharging state of the electric vehicle corresponding to the power characteristic curve under a vehicle speed-power coordinate system, wherein the power characteristic curve comprises a target torque curve, an external characteristic power curve of the driving motor, an external characteristic power curve of the engine, an engine economic working condition output power lower limit curve and a power curve when the engine and the driving motor work in parallel.
In the present exemplary embodiment, as shown in fig. 2, a power characteristic curve of the hybrid electric vehicle is shown, in which:
and 1.C represents the vehicle speed which can be driven by the engine, and I represents the maximum design vehicle speed of the whole vehicle.
2. The curve BHN represents the target torque, and when a driver steps on 100% of an accelerator pedal at a corresponding vehicle speed, the whole vehicle needs to provide corresponding driving power, and the driving power can be provided by driving an engine and a driving motor in series or in parallel;
3. the curve AFL represents the curve of the external characteristic power of the driving motor changing along with the vehicle speed;
4. the curve EK represents the curve of the change of the external characteristic power of the engine along with the change of the vehicle speed, under the hybrid structure of the embodiment of the invention, the engine can participate in the direct drive only after the vehicle speed exceeds 40km/h, and therefore, only the power characteristic of the vehicle speed exceeding 40km/h is intercepted;
5. the curve DJ represents the lower limit of the working output power of the engine in the high-efficiency area, and the specific oil consumption of the engine lower than the working output power is increased rapidly;
6. the curve GM represents the upper limit which can be reached by the sum of the power of the parallel driving of the engine and the driving motor due to the limit of the continuous discharge power of the power battery when the engine and the driving motor are driven in parallel;
7. in the area enclosed by all curves, the series mode, the parallel mode, the pure electric mode or the pure oil mode is defined. Wherein pure electric mode and parallel mode all can lead to the power battery electric quantity to descend, have consequently added "↓" sign, and under series mode and the pure oil mode, because the engine starts, can select to charge for power battery, also can select not to charge for power battery, have consequently added "↓" and accord with.
When the hybrid electric vehicle can always run in the power characteristic diagram, the part of electric quantity reduction can be balanced by the part of electric quantity supplement, and the engine does not need to be separately started for forced power supplement. Under some specific working conditions, the electric quantity is continuously reduced, so that the whole vehicle may be required to stand for a long time for power supplement.
The invention ensures that the hybrid electric vehicle always runs in the power characteristic diagram by predicting the running condition, reduces the frequency of starting the engine to enter a forced power supply mode due to insufficient electric quantity, and supplements the electric quantity of the power battery in advance in the working process of the engine.
In step S120, a driving condition of the hybrid electric vehicle may be predicted based on a preset LSTM neural network model and using the current speed and the current power of the hybrid electric vehicle as inputs, so as to generate a predicted speed and a predicted power of the hybrid electric vehicle.
In this exemplary embodiment, the method further includes training the LSTM neural network model based on historical operating parameters of the hybrid electric vehicle to generate the preset LSTM neural network model:
and based on the preset hyper-parameters of the LSTM neural network model, the historical speed and the historical power of the hybrid electric vehicle are used as input, and the training of the LSTM neural network model is completed according to the loss function judgment.
In an embodiment of the present example, the method further comprises:
establishing an LSTM neural network model population comprising a preset number of LSTM neural network models, and setting a first random hyper-parameter for each LSTM neural network model in the LSTM neural network model population;
based on random super parameters set by each LSTM neural network model, taking the historical speed and the historical power of the hybrid electric vehicle as input, and judging according to a loss function, and performing a first round of training on the LSTM neural network models in the LSTM neural network model population to generate a first super parameter value;
calculating Euclidean distances of first hyper-parameter values of all LSTM neural network models, screening and generating second hyper-parameter values based on the Euclidean distances, and generating second random hyper-parameters based on the second hyper-parameter values and random scale factors;
setting a second random hyper-parameter based on each LSTM neural network model, taking the historical speed and the historical power of the hybrid electric vehicle as input, and performing a second round of training on the LSTM neural network models in the LSTM neural network model population according to loss function judgment.
In an embodiment of the present example, the method further comprises:
and carrying out preset training on each LSTM neural network model in the LSTM neural network model population based on an iterative hyper-parameter method to generate a preset LSTM neural network model.
In the embodiment of the example, the hybrid electric vehicle is enabled to run in the power characteristic diagram all the time through the prediction of the running condition, the frequency of starting the engine to enter the forced power supplementing mode due to insufficient electric quantity is reduced, and the electric quantity of the power battery is supplemented in advance in the working process of the engine.
In the present exemplary embodiment, the C-WTVC operating condition is taken as an example, and defines a time-dependent vehicle speed curve. In fact, if the coefficient of sliding resistance of the vehicle is known, from the curve of the speed of the vehicle over time, a curve of the (speed, power) over time can be calculated, which is plotted as a three-dimensional space curve, which is a time series (variable over time). The LSTM can be used for learning and predicting the time series, and the LSTM can achieve the purpose of predicting the future time series through discarding learning to a certain degree according to the current time (vehicle speed and power) parameters and the past (vehicle speed and power) change trend.
In the present exemplary embodiment, as shown in the three-dimensional power characteristic diagram of fig. 3, a curve representing changes over time (vehicle speed, power) is shown, the three-dimensional curve of the left thick line is history information input to the LSTM network, and the three-dimensional curve of the right thin line is predicted travel condition information of the LSTM network. During this period, the vehicle is in pure oil condition.
To achieve accurate prediction, we need to train the LSTM neural network to learn the law of vehicle speed and power changes over time so that it can make accurate predictions.
Therefore, a group of LSTM neural networks are trained by adopting a population optimization method so as to find the most excellent LSTM neural network, in the training process, in order to improve the accuracy of LSTM prediction, n LSTM networks are established, and each network randomly sets an initial hyper-parameter. This is comparable to establishing a population and we perform our population optimization algorithm by the following scheme. The hyper-parameters are preset parameter values before neural network training, and comprise the number of hidden layers, training times, batch processing size, discarding probability, learning rate, gradient threshold value and the like. The difference of the super-parameter setting can result in the good and bad of the trained LSTM neural network.
Let the initial hyper-parameter setting of the ith LSTM network be
Each of which is->
The initial value is a random number. Defining Mean Square Error (MSE) as a loss function, observing and comparing the loss function of each LSTM network after each training is finished, and selecting the super parameter value ^ or the maximum value of the LSTM with the minimum loss function>
As the direction for the next round of training optimization. The hyper-parameters of all LSTM networks are updated before the next round of training begins. The updating method is to calculate the hyperparameter vector &'s in each LSTM network>
And the optimal hyperparameter vector->
The Euclidean distance between the training data and the training data is calculated according to a random scale factor eta and a Levy motion psi. The calculation formula is
The left side of the above formula represents the hyperparametric vector of the j +1 th iteration of the ith LSTM network, the 1 st item on the right side is the hyperparametric vector of the j th iteration of the ith LSTM network, the 2 nd item on the right side is the Euclidean distance between the hyperparametric vector of the j th iteration of the ith LSTM network and the hyperparametric vector of the optimal LSTM network of the j th iteration, and the 3 rd item on the right side is Levy motion, namely random motion is added, so that the neural network training process is prevented from falling into the local optimal solution.
After a plurality of times of iterative optimization, an LSTM network with the minimum loss function is found out and used as the reddest result of the population optimization algorithm.
The LSTM training input can be processed according to vehicle running condition data measured through tests, and can also be processed by a vehicle road cloud system according to running data recorded by sold vehicles. In a word, the data sources are various, and the requirement of neural network training on the data diversity can be met.
The trained LSTM neural network can predict the possible trend of the change of the vehicle speed and the time in a future period of time by recording the change of the real-time collected vehicle speed and power along with the time. If the electric quantity of the power battery is concentrated in an electric quantity reduction area of the power characteristic diagram in the future, the electric quantity of the power battery can be supplemented in an economic mode in advance to prevent sudden and urgent need of electricity, the engine cannot supplement electric energy slowly at an economic rotating speed, a high-power and rapid electricity supplementing strategy is inevitably adopted, and the fuel economy is reduced.
In step S130, a predicted vehicle condition may be generated based on the predicted vehicle speed, the predicted power, and the power characteristic curve, and a vehicle control command may be generated based on the predicted vehicle condition and a preset energy control strategy, so as to complete energy control of the hybrid electric vehicle.
In the embodiment of the present example, the energy control strategy in the method includes:
when the current vehicle condition is a pure electric working condition, the vehicle condition is predicted to be the pure electric working condition, and the energy control strategy is that when the electric quantity is lower than a threshold value, an engine is started to charge a power battery;
when the current vehicle condition is a pure electric condition, predicting the vehicle condition to enter a pure oil condition, and not processing the energy control strategy;
when the current vehicle condition is a pure electric condition, the vehicle condition is predicted to enter a parallel connection condition, and the energy control strategy is to start an engine in advance and charge a power battery;
when the current vehicle condition is a pure oil condition, predicting the vehicle condition to be the pure oil condition, and not processing the energy control strategy;
when the current vehicle condition is a pure oil condition, the vehicle condition is predicted to enter a parallel connection condition, and the energy control strategy is to preset partial power of the engine for charging and store electric energy in advance;
when the current vehicle condition is a pure oil condition, the vehicle condition is predicted to enter the pure oil condition, and the energy control strategy is to preset partial power of the engine for charging and store electric energy in advance;
when the current vehicle condition is a parallel working condition, the predicted vehicle condition is the parallel working condition, and the energy control strategy is to preset partial power of the engine for charging so as to prevent the reduction of the battery power caused by the long-time consumption of the power battery power;
when the current vehicle condition is a parallel working condition, predicting the vehicle condition to enter a series working condition, and not processing the energy control strategy;
when the current vehicle condition is a parallel working condition, the vehicle condition is predicted to enter a pure electric working condition, and the energy control strategy is to preset partial power of the engine for charging so as to prevent the reduction of the battery electric quantity caused by the long-time consumption of the electric quantity of the power battery;
when the current vehicle condition is a parallel working condition, predicting the vehicle condition to enter a pure oil working condition, and carrying out no treatment on the energy control strategy;
when the current vehicle condition is a series working condition, the predicted vehicle condition is the series working condition, the energy control strategy is to enter a power following mode, the power output by the engine follows the required power of the driving motor, and the power battery is not charged and does not need to be discharged outwards;
when the current vehicle condition is a series working condition, predicting the vehicle condition to enter a parallel working condition, and not processing the energy control strategy;
when the current vehicle condition is a series working condition, the vehicle condition is predicted to enter a pure electric working condition, and the energy control strategy is not processed.
In the exemplary embodiment, after the training of the LSTM network is completed, the time series of (vehicle speed, power) can be predicted. Training a mature LSTM network can predict which zone will be in for a future period of time (vehicle speed, power) based on historically collected time (vehicle speed, power) data. If it is predicted that the future driving condition will be in a region of the power characteristic diagram where the amount of electricity is reduced for a long time with a large power, the amount of electricity is intentionally supplemented in advance in the current driving condition. For example, when the vehicle runs at a position close to the upper bound of the OAFC area of the power characteristic diagram, the engine is started and enters a series mode, the engine runs under an economic working condition, and redundant power is used for charging the power battery; if the current vehicle runs at a position of a CDJI area of the power characteristic diagram close to an upper limit, the engine is started and enters a pure oil mode, the engine runs under an economic working condition, and redundant power is used for charging the power battery; if the current vehicle is operated in the DEKJ area of the power characteristic diagram, the output power of the engine is properly increased on the premise of not exceeding the external characteristic power of the engine, and the redundant power is used for charging the power battery. To counteract the negative effect of the continuous drop in charge during a future period of operation.
Specifically, the energy control strategy may be implemented according to the following table:
in the embodiment of the example, the energy flow control is performed in advance according to the prediction result by predicting the running condition of the hybrid electric vehicle, so that the frequency of starting the engine to enter the forced power supply mode due to the reduction of the electric quantity is reduced, and the energy utilization efficiency is improved. The trained LSTM network can predict the future running condition according to the current and historical running condition information, and the prediction can adapt to the complex change of the running condition of the automobile, so that the energy control strategy can adaptively change along with the running condition of the automobile, and the generalization of the control method is improved.
It should be noted that although the steps of the methods of the present disclosure are depicted in the drawings in a particular order, this does not require or imply that the steps must be performed in this particular order or that all of the depicted steps must be performed to achieve desirable results. Additionally or alternatively, certain steps may be omitted, multiple steps combined into one step execution, and/or one step broken down into multiple step executions, etc.
In addition, in the exemplary embodiment, a hybrid electric vehicle energy control apparatus based on the traveling condition prediction is also provided. Referring to fig. 4, the hybrid electric vehicle energy control apparatus 400 based on the prediction of the driving condition may include: a power profile generation module 410, a vehicle condition prediction module 420, and an energy control module 430. Wherein:
the power characteristic curve generating module 410 is used for analyzing the power characteristic of the hybrid electric vehicle and generating the power characteristic curve of the hybrid electric vehicle under a vehicle speed-power coordinate system;
the vehicle condition prediction module 420 is configured to predict a driving condition of the hybrid electric vehicle by using a current vehicle speed and a current power of the hybrid electric vehicle as inputs based on a preset LSTM neural network model, and generate a predicted vehicle speed and a predicted power of the hybrid electric vehicle;
and the energy control module 430 is configured to generate a predicted vehicle condition based on the predicted vehicle speed, the predicted power, and the power characteristic curve, and generate a vehicle control instruction based on the predicted vehicle condition and a preset energy control strategy to complete energy control on the hybrid electric vehicle.
The specific details of each of the above hybrid electric vehicle energy control device modules based on the driving condition prediction have been described in detail in a corresponding hybrid electric vehicle energy control method based on the driving condition prediction, and therefore, the details are not repeated herein.
It should be noted that although several modules or units of a hybrid electric vehicle energy control apparatus 400 based on travel condition prediction are mentioned in the above detailed description, such division is not mandatory. Indeed, the features and functions of two or more modules or units described above may be embodied in one module or unit, according to embodiments of the present disclosure. Conversely, the features and functions of one module or unit described above may be further divided into embodiments by a plurality of modules or units.
Furthermore, the above-described figures are merely schematic illustrations of processes involved in methods according to exemplary embodiments of the invention, and are not intended to be limiting. It will be readily understood that the processes shown in the above figures are not intended to indicate or limit the chronological order of the processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, e.g., in multiple modules.
Other embodiments of the disclosure will be apparent to those skilled in the art from consideration of the specification and practice of the disclosure disclosed herein. This application is intended to cover any variations, uses, or adaptations of the disclosure following, in general, the principles of the disclosure and including such departures from the present disclosure as come within known or customary practice within the art to which the disclosure pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the disclosure being indicated by the following claims.
It will be understood that the present disclosure is not limited to the precise arrangements described above and shown in the drawings and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is to be limited only by the terms of the appended claims.
Claims (7)
1. A hybrid electric vehicle energy control method based on driving condition prediction is characterized by comprising the following steps:
analyzing the power characteristic of the hybrid electric vehicle, and generating a power characteristic curve of the hybrid electric vehicle under a vehicle speed-power coordinate system;
based on a preset LSTM neural network model, taking the current speed and the current power of the hybrid electric vehicle as input, predicting the running condition of the hybrid electric vehicle, and generating the predicted speed and the predicted power of the hybrid electric vehicle;
and generating a predicted vehicle condition based on the predicted vehicle speed, the predicted power and the power characteristic curve, and generating a vehicle control command based on the predicted vehicle condition and a preset energy control strategy to complete energy control on the hybrid electric vehicle.
2. The hybrid electric vehicle energy control method based on traveling condition prediction according to claim 1, characterized by further comprising:
analyzing the series-parallel connection state of an engine and a driving motor of the hybrid electric vehicle and the power characteristic during combined operation, and generating a power characteristic curve of the hybrid electric vehicle and the charging and discharging state of the electric vehicle corresponding to the power characteristic curve under a vehicle speed-power coordinate system, wherein the power characteristic curve comprises a target torque curve, an external characteristic power curve of the driving motor, an external characteristic power curve of the engine, an engine economic working condition output power lower limit curve and a power curve when the engine and the driving motor work in parallel.
3. The method for hybrid electric vehicle energy control based on driving condition prediction according to claim 1, further comprising training the LSTM neural network model based on historical operating parameters of the hybrid electric vehicle to generate the preset LSTM neural network model:
and based on the preset hyper-parameters of the LSTM neural network model, the historical speed and the historical power of the hybrid electric vehicle are used as input, and the training of the LSTM neural network model is completed according to the loss function judgment.
4. The hybrid electric vehicle energy control method based on traveling condition prediction according to claim 3, characterized by further comprising:
establishing an LSTM neural network model population containing a preset number of LSTM neural network models, and setting a first random hyper-parameter for each LSTM neural network model in the LSTM neural network model population;
based on random super parameters set by each LSTM neural network model, taking the historical speed and the historical power of the hybrid electric vehicle as input, and judging according to a loss function, and performing a first round of training on the LSTM neural network models in the LSTM neural network model population to generate a first super parameter value;
calculating Euclidean distances of first hyper-parameter values of each LSTM neural network model, screening and generating second hyper-parameter values based on the Euclidean distances, and generating second random hyper-parameters based on the second hyper-parameter values and random scale factors;
setting a second random hyper-parameter based on each LSTM neural network model, taking the historical speed and the historical power of the hybrid electric vehicle as input, and performing a second round of training on the LSTM neural network models in the LSTM neural network model population according to loss function judgment.
5. The hybrid electric vehicle energy control method based on traveling condition prediction according to claim 4, characterized by further comprising:
and carrying out preset training on each LSTM neural network model in the LSTM neural network model population based on an iterative hyper-parameter method to generate a preset LSTM neural network model.
6. The method for hybrid electric vehicle energy control based on driving condition prediction according to claim 1, wherein the energy control strategy in the method comprises:
when the current vehicle condition is the pure electric working condition, predicting that the vehicle condition is the pure electric working condition, and starting an engine to charge a power battery when the electric quantity is lower than a threshold value according to an energy control strategy;
when the current vehicle condition is the pure electric condition, predicting the vehicle condition to enter the pure oil condition, and carrying out no treatment on the energy control strategy;
when the current vehicle condition is a pure electric condition, predicting the vehicle condition to be a parallel connection condition, and carrying out an energy control strategy to start an engine in advance and charge a power battery;
when the current vehicle condition is a pure oil condition, predicting the vehicle condition to be the pure oil condition, and carrying out no treatment on the energy control strategy;
when the current vehicle condition is a pure oil condition, predicting the vehicle condition to be a parallel connection condition, and presetting partial power for charging for an engine by an energy control strategy to store electric energy in advance;
when the current vehicle condition is a pure oil condition, predicting the vehicle condition to enter the pure oil condition, and presetting partial power for charging for an engine by an energy control strategy to store electric energy in advance;
when the current vehicle condition is the parallel working condition, predicting that the vehicle condition is the parallel working condition, and using part of power preset by the engine for charging according to an energy control strategy to prevent the reduction of the battery power caused by long-time consumption of the power battery power;
when the current vehicle condition is the parallel working condition, predicting the vehicle condition to enter the series working condition, and not processing the energy control strategy;
when the current vehicle condition is a parallel working condition, predicting the vehicle condition to enter a pure electric working condition, and presetting partial power for charging for an engine by an energy control strategy so as to prevent the reduction of the battery electric quantity caused by the long-time consumption of the electric quantity of a power battery;
when the current vehicle condition is the parallel working condition, predicting the vehicle condition to enter a pure oil working condition, and not processing the energy control strategy;
when the current vehicle condition is the series working condition, predicting that the vehicle condition is the series working condition, wherein the energy control strategy is to enter a power following mode, the power output by the engine follows the required power of the driving motor, and the power battery is not charged or does not need to be discharged outwards;
when the current vehicle condition is the series working condition, predicting the vehicle condition to enter the parallel working condition, and carrying out no treatment on the energy control strategy;
and when the current vehicle condition is the series working condition, predicting the vehicle condition to enter the pure electric working condition, and not processing the energy control strategy.
7. An energy control apparatus for a hybrid electric vehicle based on a prediction of a running condition, the apparatus comprising:
the power characteristic curve generating module is used for analyzing the power characteristic of the hybrid electric vehicle and generating the power characteristic curve of the hybrid electric vehicle under a vehicle speed-power coordinate system;
the vehicle condition prediction module is used for predicting the running condition of the hybrid electric vehicle by taking the current speed and the current power of the hybrid electric vehicle as input based on a preset LSTM neural network model, and generating the predicted speed and the predicted power of the hybrid electric vehicle;
and the energy control module is used for generating a predicted vehicle condition based on the predicted vehicle speed, the predicted power and the power characteristic curve, generating a vehicle control command based on the predicted vehicle condition and a preset energy control strategy and finishing energy control on the hybrid electric vehicle.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211441575.8A CN115923764A (en) | 2022-11-17 | 2022-11-17 | Hybrid electric vehicle energy control method and device based on running condition prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211441575.8A CN115923764A (en) | 2022-11-17 | 2022-11-17 | Hybrid electric vehicle energy control method and device based on running condition prediction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115923764A true CN115923764A (en) | 2023-04-07 |
Family
ID=86549695
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211441575.8A Pending CN115923764A (en) | 2022-11-17 | 2022-11-17 | Hybrid electric vehicle energy control method and device based on running condition prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115923764A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116749946A (en) * | 2023-08-21 | 2023-09-15 | 新誉集团有限公司 | Vehicle energy management method, device, equipment and readable storage medium |
-
2022
- 2022-11-17 CN CN202211441575.8A patent/CN115923764A/en active Pending
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116749946A (en) * | 2023-08-21 | 2023-09-15 | 新誉集团有限公司 | Vehicle energy management method, device, equipment and readable storage medium |
| CN116749946B (en) * | 2023-08-21 | 2023-10-20 | 新誉集团有限公司 | Vehicle energy management method, device, equipment and readable storage medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110775065B (en) | Hybrid electric vehicle battery life prediction method based on working condition recognition | |
| CN107688343B (en) | Energy control method of hybrid power vehicle | |
| CN110775043B (en) | Hybrid electric vehicle energy optimization method based on battery life attenuation pattern recognition | |
| Sezer et al. | A novel ECMS and combined cost map approach for high-efficiency series hybrid electric vehicles | |
| Gao et al. | Equivalent fuel consumption optimal control of a series hybrid electric vehicle | |
| CN110936949B (en) | Energy control method, equipment, storage medium and device based on driving condition | |
| CN108791270B (en) | Power component working point regulation and control method of hybrid power vehicle based on working condition prediction | |
| CN111619545B (en) | Hybrid electric vehicle energy management method based on traffic information | |
| CN110991757A (en) | Comprehensive prediction energy management method for hybrid electric vehicle | |
| CN111775925B (en) | Working mode decision method and device for power split hybrid electric vehicle | |
| CN109204300B (en) | Hybrid vehicle and method for controlling running mode thereof | |
| Ganji et al. | A study on look-ahead control and energy management strategies in hybrid electric vehicles | |
| CN112319462B (en) | Energy management method for plug-in hybrid electric vehicle | |
| CN108995642A (en) | A kind of vehicle distance increasing unit control method, system and vehicle | |
| Pham et al. | Analytical solution to energy management guaranteeing battery life for hybrid trucks | |
| CN106055830A (en) | PHEV (Plug-in Hybrid Electric Vehicle) control threshold parameter optimization method based on dynamic programming | |
| CN113815437B (en) | Predictive energy management method for fuel cell hybrid electric vehicle | |
| CN113479186B (en) | Energy management strategy optimization method for hybrid electric vehicle | |
| CN115923764A (en) | Hybrid electric vehicle energy control method and device based on running condition prediction | |
| CN104760591A (en) | Hybrid power comprehensive control system | |
| CN115230485A (en) | Short-term power smooth prediction-based fuel cell bus energy management method | |
| Ali et al. | Development and improvement of a situation-based power management method for multi-source electric vehicles | |
| Chen et al. | Driving cycle recognition based adaptive equivalent consumption minimization strategy for hybrid electric vehicles | |
| CN113246958B (en) | TD 3-based multi-target HEV energy management method and system | |
| Yao et al. | Control strategy for hybrid electric vehicle based on online driving pattern classification |
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 |