CN112319461A - Hybrid electric vehicle energy management method based on multi-source information fusion - Google Patents
Hybrid electric vehicle energy management method based on multi-source information fusion Download PDFInfo
- Publication number
- CN112319461A CN112319461A CN202011289194.3A CN202011289194A CN112319461A CN 112319461 A CN112319461 A CN 112319461A CN 202011289194 A CN202011289194 A CN 202011289194A CN 112319461 A CN112319461 A CN 112319461A
- Authority
- CN
- China
- Prior art keywords
- vehicle
- soc
- power battery
- speed
- vehicle speed
- 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.)
- Granted
Links
- 238000007726 management method Methods 0.000 title claims abstract description 40
- 230000004927 fusion Effects 0.000 title claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 52
- 230000007774 longterm Effects 0.000 claims abstract description 33
- 239000000446 fuel Substances 0.000 claims description 32
- 230000008569 process Effects 0.000 claims description 27
- 238000004422 calculation algorithm Methods 0.000 claims description 13
- 238000005457 optimization Methods 0.000 claims description 10
- 238000010276 construction Methods 0.000 claims description 9
- 230000033001 locomotion Effects 0.000 claims description 9
- 238000003062 neural network model Methods 0.000 claims description 9
- 230000005611 electricity Effects 0.000 claims description 6
- 230000007704 transition Effects 0.000 claims description 6
- 238000000342 Monte Carlo simulation Methods 0.000 claims description 3
- 238000012937 correction Methods 0.000 claims description 3
- 230000006870 function Effects 0.000 description 28
- 210000002569 neuron Anatomy 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 239000011159 matrix material Substances 0.000 description 2
- 238000012544 monitoring process Methods 0.000 description 2
- 238000012549 training Methods 0.000 description 2
- 238000007476 Maximum Likelihood Methods 0.000 description 1
- 238000013528 artificial neural network Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000005540 biological transmission Effects 0.000 description 1
- 238000004364 calculation method Methods 0.000 description 1
- 239000003990 capacitor Substances 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 238000005315 distribution function Methods 0.000 description 1
- 230000005284 excitation Effects 0.000 description 1
- 238000013507 mapping Methods 0.000 description 1
- 238000005309 stochastic process Methods 0.000 description 1
Images
Classifications
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W20/00—Control systems specially adapted for hybrid vehicles
- B60W20/10—Controlling the power contribution of each of the prime movers to meet required power demand
- B60W20/11—Controlling the power contribution of each of the prime movers to meet required power demand using model predictive control [MPC] strategies, i.e. control methods based on models predicting performance
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2510/00—Input parameters relating to a particular sub-units
- B60W2510/24—Energy storage means
- B60W2510/242—Energy storage means for electrical energy
- B60W2510/244—Charge state
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2520/00—Input parameters relating to overall vehicle dynamics
- B60W2520/10—Longitudinal speed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2554/00—Input parameters relating to objects
- B60W2554/80—Spatial relation or speed relative to objects
- B60W2554/804—Relative longitudinal speed
-
- B—PERFORMING OPERATIONS; TRANSPORTING
- B60—VEHICLES IN GENERAL
- B60W—CONJOINT CONTROL OF VEHICLE SUB-UNITS OF DIFFERENT TYPE OR DIFFERENT FUNCTION; CONTROL SYSTEMS SPECIALLY ADAPTED FOR HYBRID VEHICLES; ROAD VEHICLE DRIVE CONTROL SYSTEMS FOR PURPOSES NOT RELATED TO THE CONTROL OF A PARTICULAR SUB-UNIT
- B60W2555/00—Input parameters relating to exterior conditions, not covered by groups B60W2552/00, B60W2554/00
- B60W2555/60—Traffic rules, e.g. speed limits or right of way
Landscapes
- Engineering & Computer Science (AREA)
- Automation & Control Theory (AREA)
- Transportation (AREA)
- Mechanical Engineering (AREA)
- Electric Propulsion And Braking For Vehicles (AREA)
Abstract
The invention provides a hybrid electric vehicle energy management method based on multi-source information fusion, and belongs to the field of hybrid electric vehicle energy management. The method comprises the following steps: the method comprises the steps that a long-term running working condition of a vehicle in the future is built by combining traffic information of a current running road section of the vehicle, a vehicle speed-time relation curve of the whole travel of the vehicle is obtained through prediction, and a reference track of the power battery SOC of the whole travel of the vehicle is obtained through prediction of an objective function of the power battery SOC, which is solved by taking the vehicle speed-time relation curve as constraint; the method comprises the steps that the actual speed of a vehicle at the current moment, the speed of a front vehicle and the relative distance between the self vehicle and the front vehicle are combined, and the short-term speed data of the vehicle in the future are obtained through prediction; and predicting to obtain the future engine output torque and the future power battery output power of the vehicle by combining the actual SOC of the power battery, the reference track of the SOC of the power battery and the short-term vehicle speed data of the vehicle in the future. The invention comprehensively considers the long-term running working condition and the short-term running working condition of the vehicle, and the formulated vehicle energy management scheme is more reasonable.
Description
Technical Field
The invention relates to a hybrid electric vehicle energy management method based on multi-source information fusion, and belongs to the technical field of hybrid electric vehicle energy management.
Background
Compared with a conventional Vehicle or a pure electric Vehicle, a Plug-in hybrid electric Vehicle (PHEV) has at least two energy sources, which generally include an internal combustion engine and electric energy stored in a battery or a super capacitor. The biggest difference between the plug-in hybrid electric vehicle type and the traditional hybrid electric vehicle type is that the electric quantity of a battery is increased and the external charging under the condition of parking can be realized, so that the electric quantity in the power battery needs to be consumed as much as possible in a single driving task of the plug-in hybrid electric vehicle, the fuel consumption is further reduced on the basis of the traditional hybrid electric vehicle system, the reasonable consumption of the electric quantity of the battery is completely consumed when the driving task is finished, and the optimization of the fuel economy is realized.
For example, the invention patent document with the publication number CN105946857B discloses a parallel PHEV energy management method based on an intelligent transportation system, which calculates a predicted working condition vehicle speed-time history according to working condition characteristic parameters of a driving path, then generates a reference SOC by combining the predicted working condition vehicle speed-time history and the working condition characteristic parameters of the driving path, then calculates an engine torque threshold adjustment coefficient and a pure electric vehicle speed threshold adjustment coefficient by taking the reference SOC as a control target and taking an actual SOC as feedback, determines a pure electric-engine driving mode switching threshold, determines a driving mode of a current vehicle by combining the switching threshold, and determines an engine torque and a motor distribution torque according to the driving mode of the vehicle.
The current energy management method of the plug-in hybrid electric vehicle usually predicts the long-term driving condition of the vehicle in the future by using the road traffic information at a certain moment, then determines a reference SOC by combining the predicted long-term driving condition of the vehicle in the future, and performs energy management on the vehicle by combining the reference SOC and the actual SOC.
Disclosure of Invention
The invention aims to provide a hybrid electric vehicle energy management method based on multi-source information fusion, which is used for solving the problem that a formulated vehicle energy management scheme is not reasonable enough when the vehicle is subjected to energy management only according to a long-term running working condition established by road traffic information at a certain moment.
In order to achieve the aim, the invention provides a hybrid electric vehicle energy management method based on multi-source information fusion, which comprises the following steps:
acquiring traffic information of a current running road section of a vehicle, the actual speed of the vehicle and the actual SOC of a power battery in real time; the traffic information comprises preceding vehicle motion information, traffic flow information, traffic signal lamp information and road speed limit information, wherein the preceding vehicle motion information comprises the speed of a preceding vehicle and the relative distance between the current vehicle and the preceding vehicle;
establishing a future long-term running working condition of the vehicle based on a Markov chain by combining traffic information of a current running road section of the vehicle, predicting to obtain a vehicle speed-time relation curve of the whole travel of the vehicle, establishing an objective function of a power battery SOC (system on chip) by taking the running cost of the whole travel of the vehicle as the lowest principle, solving the objective function of the power battery SOC by using a dynamic programming algorithm by taking the vehicle speed-time relation curve as constraint, predicting to obtain a reference track of the power battery SOC of the whole travel of the vehicle, and updating the long-term running working condition and the reference track of the power battery SOC once at set time intervals by using the traffic information of the current running road section of the vehicle obtained in real time; the whole travel of the vehicle is the travel of the vehicle from the current position to the end position;
the method comprises the steps that the actual speed of a vehicle at the current moment, the speed of a front vehicle and the relative distance between the self vehicle and the front vehicle are combined, and the short-term speed data of the vehicle in the future are obtained through prediction;
and predicting to obtain the future engine output torque and the future power battery output power of the vehicle by combining the actual SOC of the power battery, the reference track of the SOC of the power battery and the short-term vehicle speed data of the vehicle in the future, so as to realize the energy management of the hybrid electric vehicle.
The invention has the beneficial effects that: the long-term running working condition is periodically updated based on the real-time updated multi-source traffic information, so that the constructed long-term running working condition is ensured to be consistent with the actual road condition, meanwhile, the energy management of the vehicle is also fusion of the multi-source information by combining the long-term running working condition and the short-term running working condition, the formulated vehicle energy management scheme can be further ensured to be consistent with the traffic information of the current running road section of the vehicle and the actual running state of the vehicle, and the formulated vehicle energy management scheme is more reasonable; and secondly, the reference track of the SOC of the power battery in the whole travel of the vehicle is updated regularly based on traffic information, so that electric energy can be better utilized, and the battery can be healthier to use.
Further, in the above method, the method further comprises the step of feedback correcting the vehicle speed data of the vehicle in a short term in the future by using the actual vehicle speed of the vehicle.
Further, in the above method, the process of predicting the future engine output torque and power battery output power of the vehicle includes: solving an energy management optimization objective function by using a simplified dynamic programming algorithm in combination with the actual SOC of the power battery, the reference track of the SOC of the power battery and the short-term vehicle speed data of the vehicle in the future, and predicting to obtain the future engine output torque and the future power output of the power battery of the vehicle; the energy management optimization objective function is as follows:
wherein,
wherein E is an energy management optimization objective function, L (x (t), u (t) ═ βfuel·Qfuel(t)+βele·Pbatt(t), u (t) is output torque of the engine, x (t) is SOC of the power battery, betafuelAs a price of fuel, Qfuel(t) is the fuel rate in L/s; beta is aeleAt a price of electricity, Pbatt(t) is the output power of the power battery, p is the prediction time domain, h (SOC (t +1)) is a penalty function for converging the SOC to a set value, beta is a penalty coefficient, and the SOC isr(t) is the SOC reference value at time t, and SOC (t) is the actual SOC value at time t.
Further, in the above method, the construction process of the future long-term driving condition of the vehicle includes: dividing the whole travel of the vehicle into M road sections, wherein M is more than or equal to 1, determining the average speed of the current road section according to the traffic information of the current road section, and obtaining a speed state cluster Q of the current road section according to the determined speed state cluster classification rule by combining the average speed of the current road section1Randomly selecting a cluster of vehicle speed states equal to Q from the vehicle speed segment candidate set1As a first road segment; determining a vehicle speed state cluster of the vehicle speed section according to the average vehicle speed of the vehicle speed section and the determined vehicle speed state cluster classification rule; determining a cluster of vehicle speed states Q for a next road segment based on a Markov chain2Selecting a cluster of vehicle speed states as Q from the vehicle speed segment candidate set2And the speed segment with the difference between the initial speed and the last speed value of the previous segment smaller than the set value is taken as a second segment; deleting the used vehicle speed sections from the vehicle speed section candidate set in the construction process, and repeating the steps until the total number of the constructed road sections is equal to the total number M of the road sections, so as to finish the construction of the long-term running working condition of the vehicle; the vehicle speed segment candidate set is obtained by dividing vehicle speed segments into historical vehicle speed data of a vehicle, the vehicle speed segment candidate set comprises a plurality of vehicle speed segments and a vehicle speed state cluster corresponding to each vehicle speed segment, and the division rule of the vehicle speed segments is as follows: the movement of the vehicle from one idle speed to the next idle speed is defined as a vehicle speed segment.
Further, in the above method, the determining a state cluster Q of the next link based on the markov chain2The process comprises the following steps: generating a random number r between (0,1) based on the Monte Carlo method, if r satisfies
Then the state cluster Q of the next road segment is determined2Is q; wherein, PijIn order to regard the driving process of the vehicle as the transition probability of the vehicle speed segment from the vehicle speed state cluster i to the vehicle speed state cluster j when the driving process of the vehicle is regarded as the Markov process, Q is the total number of the categories of the vehicle speed state cluster.
Further, in the method, a target function of the power battery SOC is solved by using a simplified dynamic programming algorithm under a constraint condition by combining a vehicle speed-time relation curve of the whole travel of the vehicle, and a reference track of the power battery SOC of the whole travel of the vehicle is obtained through prediction; the target function of the SOC of the power battery is as follows:
Pbatt(t)＝VI(t)-I(t)2R
j is an objective function of the SOC of the power battery, L (x (t), u (t)) is an instantaneous objective function, u (t) is the output torque of the engine, x (t) is the SOC of the power battery, and M is the number of segments dividing the running condition; beta is afuelAs a price of fuel, Qfuel(t) is the fuel rate in L/s; beta is aeleAt a price of electricity, Pbatt(t) is the output power of the power battery, V is the open-circuit voltage of the power battery, R is the internal resistance of the power battery, I (t) is the discharge current of the power battery, D is the electric quantity of the power battery, and SOCiniIs an initial SOC value, SOCr(t) is the SOC reference value at time t;
the constraint conditions are as follows:
in the formula, Te(t)、Tm(t)、Treq(T) engine torque, motor torque and vehicle demand torque, respectively, the vehicle demand torque being obtained from a vehicle speed-time relationship curve of the entire vehicle travel, Tbrake(t) is mechanical braking torque, ωe(t) and ωmAnd (t) the engine rotating speed and the motor rotating speed respectively, SOC (t) is the SOC of the power battery, min represents a lower limit value, and max represents an upper limit value.
Further, in the method, the actual vehicle speed of the vehicle at the current moment, the vehicle speed of the vehicle ahead and the relative distance between the vehicle and the vehicle ahead are input into a pre-trained BP neural network model to predict and obtain the vehicle speed data of the vehicle in a short term in the future.
Drawings
FIG. 1 is a flow chart of a hybrid electric vehicle energy management method based on multi-source information fusion in an embodiment of the method of the present invention;
FIG. 2 is a schematic diagram of the application of real-time traffic information to the construction of a long-term future driving condition of a vehicle in an embodiment of the method of the present invention;
FIG. 3 is a vehicle speed segment division schematic in an embodiment of the method of the present invention;
fig. 4 is a schematic diagram of a topological network structure of a BP neural network prediction model in the embodiment of the method of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is described in further detail below with reference to the accompanying drawings and embodiments.
The method comprises the following steps:
the method for managing energy of a hybrid electric vehicle based on multi-source information fusion (hereinafter referred to as the method of the present embodiment) of the present embodiment is shown in fig. 1, and the method of the present embodiment will be described in detail below by taking an example of applying the method of the present embodiment to a plug-in hybrid electric vehicle, but of course, the method of the present embodiment can also be applied to a common hybrid electric vehicle.
As shown in fig. 1, the method of this embodiment includes the following steps:
the traffic information of the current driving road section of the vehicle can be acquired through the vehicle-mounted radar and the vehicle-mounted vision system, and the traffic information comprises: the vehicle-mounted system comprises front vehicle motion information, traffic flow information, traffic signal lamp information and road speed limit information, wherein the front vehicle motion information comprises the speed of a front vehicle and the relative distance between the self vehicle and the front vehicle.
as shown in fig. 2, the process of constructing the long-term future driving condition of the vehicle using the real-time traffic information includes: the traffic monitoring platform acquires real-time traffic information of a current driving road section of a vehicle, the acquired multi-source traffic information is fused through a fusion strategy, a Markov chain is utilized to process and predict the fused traffic information to obtain a vehicle speed-time relation curve of the whole travel of the vehicle, the long-term driving working condition of the vehicle in the future is updated once every 300s by combining the traffic information of the current driving road section of the vehicle acquired in real time, and then the vehicle speed-time relation curve of the whole travel of the vehicle is updated once every 300 s.
The traffic information is acquired from different sensors, and meanwhile, the information acquired by the different sensors is different, so that the traffic information used for constructing the long-term running working condition is multi-source information, namely, the embodiment constructs the future long-term running working condition of the vehicle by fusing the multi-source traffic information; there are many fusion strategies in the prior art for fusing multi-source traffic information, and when the fusion strategy is applied, an existing fusion strategy is selected according to actual needs, for example, a D-S evidence theory is adopted to fuse the multi-source traffic information, and the basic principle is as follows: a plurality of sensors are used as data sources, respective judgment is made under the same identification frame according to a basic probability distribution function, and reasonable judgment is fused together through a combination rule.
The following details the principles and processes for constructing a vehicle's future long-term driving behavior using Markov chains:
firstly, a vehicle speed segment candidate set is constructed by dividing vehicle speed segments into historical vehicle speed data of a vehicle, and the vehicle speed segment candidate set comprises a plurality of vehicle speed segments and a vehicle speed state cluster corresponding to each vehicle speed segment. The construction process of the vehicle speed segment candidate set comprises the following steps: dividing the historical speed data of the vehicle into vehicle speed segments, as shown in fig. 3, the dividing rule of the vehicle speed segments is as follows: the movement of the vehicle from one idle speed to the next idle speed is defined as a vehicle speed segment: then, determining a vehicle speed state cluster of a corresponding vehicle speed segment according to the average speed of each vehicle speed segment to obtain a vehicle speed segment candidate set; the detailed classification rule of the vehicle speed state clusters is shown in table 1 (in the present embodiment, the total number of categories Q of the vehicle speed state clusters is taken as an example of 6), and for example, when the average speed of a certain vehicle speed segment falls within the vehicle speed section [0, 10 ], the state cluster of the vehicle speed segment is 1. It is easy to know that one vehicle speed segment corresponds to one vehicle speed state cluster, and one vehicle speed state cluster corresponds to a plurality of vehicle speed segments.
TABLE 1 vehicle speed State Cluster Classification rule Table
The markov process is typically a stochastic process whose future state changes independently of its past, depending only on its present state. The vehicle speed changes with the change of the traffic environment during the running process of the vehicle, and since there is a certain uncertainty in the influence of many factors on the traffic environment, the running process of the vehicle can be regarded as a markov process, and the markov characteristic of the running process of the vehicle can be expressed as follows:
P{X(n+1)＝j|X(n)＝i}＝pij(n)
wherein, { X (n); n ≧ 0} is Markov chain, pij(n) transition probability of the vehicle speed segment from the state cluster i to the state cluster j at n time, pijCan be obtained by maximum likelihood estimation, i.e.
NijThe number of times the vehicle speed segment transitions from state cluster i to state cluster j.
When a Markov chain has N states, its Markov chain's state transition matrix P is as follows:
generating a random number r between (0,1) based on the Monte Carlo method, if r satisfies
The state cluster of the next vehicle speed segment can be determined to be q.
The process of constructing the long-term running working condition of the vehicle according to the acquired real-time traffic information and the state transition matrix comprises the following steps:
first, the long-term driving condition should be initialized before the long-term driving condition is constructedgDividing a driving route of the whole travel of the vehicle into M road sections (for example, M is 14), acquiring real-time traffic information of each road section from a traffic monitoring platform, and sending the real-time traffic information to a target vehicle;
secondly, determining a vehicle speed state cluster Q1 of the current road section according to the average vehicle speed of the current road section, and randomly selecting a vehicle speed state cluster equal to Q from the vehicle speed segment candidate set1As a first road segment; by passing
Determining a status cluster Q for a next road segment2Selecting a cluster of vehicle speed states as Q from the vehicle speed segment candidate set2And the speed section with the difference between the initial speed and the last speed value of the previous section being less than a set value (for example, 1km/h) is taken as a second section; if the total number of the constructed road sections is less than the total number of the road sections, repeating the steps to construct the next road section if the total number of the constructed road sections is less than the total number of the road sections, completing construction of a long-term running working condition of the vehicle until the total number of the constructed road sections is equal to the total number of the road sections, and forming a vehicle speed-actual relation curve of the whole travel of the vehicle by the vehicle speed segments of all the constructed road sections; in addition, the used vehicle speed segment should be deleted from the vehicle speed segment candidate set, so as to avoid repeated use.
Finally, considering the limitation of information transmission in practical application, assuming that the obtained traffic information is updated once every 300s, reconstructing the long-term driving condition every 300s according to the real-time updated traffic data, wherein the reconstruction process is the same as the previous two steps.
In another embodiment, the total number Q of the categories of the vehicle speed state clusters and the vehicle speed interval value corresponding to each vehicle speed state cluster can be adjusted according to actual needs, and the number M of the road sections divided by the vehicle driving route can also be adjusted according to actual needs.
Step 3, solving an objective function of the SOC of the power battery by using a simplified dynamic programming algorithm under a constraint condition by combining a speed-time relation curve of the whole journey of the vehicle, and predicting to obtain a reference track of the SOC of the power battery of the whole journey of the vehicle; the vehicle speed-time relation curve of the whole travel of the vehicle is updated every 300s, and the vehicle speed-time relation curve of the whole travel of the vehicle is an important constraint condition for solving the reference track of the power battery SOC, so the reference track of the power battery SOC of the whole travel of the vehicle is also updated every 300 s.
The method comprises the following steps of constructing a target function of the SOC of the power battery on the basis of the lowest running cost of the whole travel of a vehicle, wherein the target function of the SOC of the power battery is as follows:
Pbatt(t)＝VI(t)-I(t)2R
j is an objective function of the SOC of the power battery, L (x (t), u (t)) is an instantaneous objective function, u (t) is the output torque of the engine, x (t) is the SOC of the power battery, and M is the number of segments dividing the running condition; beta is afuelAs a price of fuel, Qfuel(t) is the fuel rate in L/s; beta is aeleAt a price of electricity, Pbatt(t) is the output power of the power battery, V is the open-circuit voltage of the power battery, and R isThe internal resistance of the power battery, I (t) is the discharge current of the power battery, D is the electric quantity of the power battery, SOCiniIs an initial SOC value, SOCr(t) is the SOC reference value at time t;
the constraint conditions of the objective function J of the power battery SOC are as follows:
in the formula, Te(t)、Tm(t)、Treq(T) engine torque, motor torque and vehicle demand torque, respectively, the vehicle demand torque being obtained from a vehicle speed-time relationship curve of the entire vehicle travel, Tbrake(t) is mechanical braking torque, ωe(t) and ωmAnd (t) the engine rotating speed and the motor rotating speed respectively, SOC (t) is the SOC of the power battery, min represents a lower limit value, and max represents an upper limit value.
In the embodiment, the target function of the power battery SOC is solved by using a simplified dynamic programming algorithm, and the reference track of the power battery SOC is obtained. Wherein the control variable of the simplified DP algorithm is the torque T of the engineeThe state variable is the power battery SOC, and the solving process of the reference track of the power battery SOC is as follows: firstly, solving an objective function J of the SOC of the power battery under a constraint condition to obtain the output power P of the power batterybatt(t), then combining the formula: pbatt(t)＝VI(t)-I(t)2R and
solving to obtain the SOC reference track SOC of the power batteryr(t)。
In the embodiment, the simplified DP algorithm is used for calculating the SOC reference track of the power battery, compared with the traditional DP algorithm, the calculation time can be greatly reduced while certain solving precision is guaranteed, so that the SOC reference track of the power battery can be rapidly planned based on the traffic information of the current driving road section of the vehicle acquired in real time, the instantaneity of the SOC reference track of the power battery is guaranteed, and the dynamic traffic information is followed. Of course, as another embodiment, the reference trajectory of the power battery SOC of the vehicle for a long term in the future may also be obtained by solving an objective function of the power battery SOC using a conventional DP algorithm.
Step 4, predicting to obtain the vehicle speed data of the vehicle in the future short term by combining the actual vehicle speed of the vehicle at the current moment, the vehicle speed of the vehicle in front and the relative distance between the vehicle and the vehicle in front;
wherein the current time front vehicle speed v2(t) and the relative distance x (t) between the own vehicle and the front vehicle can be detected by using a vehicle-mounted millimeter wave radar.
In the present embodiment, the actual vehicle speed (i.e., the vehicle speed v) of the vehicle at the present time is determined1(t)), vehicle speed v of preceding vehicle2(t) inputting the relative distance x (t) between the own vehicle and the preceding vehicle into a pre-trained BP neural network model to predict and obtain the vehicle speed data of the vehicle in the short term in the future; the mapping relationship between the input quantity and the output quantity of the BP neural network model can be expressed as follows: v. of1(t+1)＝f(v1(t),v2(t), x (t)), FIG. 4, by v1(t)、v2(t) and x (t) to obtain v1(t +1) by v1(t+1)、v2(t +1) and x (t +1) give v1(t +2), and so on, finally obtaining the vehicle speed data of the vehicle in the short term in the future.
In this embodiment, the BP neural network model adopts a conventional 3-layer network structure, as shown in fig. 4, the numbers of neurons in the input layer and the output layer are respectively selected to be 3 and 1 according to the input and output of the network, the number of neurons in the hidden layer is determined to be 4 according to a trial and error method, a tansig function is selected as an excitation function of the neurons in the hidden layer, and the BP neural network model can be infinitely close to any function after multiple learning training. The training data when the BP neural network model is trained are historical speed data of the vehicle, historical speed data of a front vehicle corresponding to the historical speed data, and historical relative distance between the vehicle and the front vehicle.
As other embodiments, BP neural network models of other network structures can be selected according to actual needs.
Step 5, predicting to obtain the future engine output torque and the power battery output power of the vehicle by combining the actual SOC of the power battery, the reference track of the SOC of the power battery in the whole travel of the vehicle and the short-term vehicle speed data of the vehicle in the future, so as to realize the energy management of the hybrid electric vehicle; and optimizing the BP neural network model by combining the actual vehicle speed of the vehicle, thereby carrying out feedback correction on the short-term vehicle speed data of the vehicle in the future, wherein the purpose of the feedback correction is to make the predicted result more accord with the actual situation in the next time domain so as to make the predicted result more approximate to the actual vehicle running.
In the embodiment, a simplified dynamic programming algorithm is utilized to solve an energy management optimization objective function, and the future engine output torque and the power battery output power of the vehicle are obtained through prediction; wherein, the energy management optimization objective function is as follows:
wherein,
wherein E is an energy management optimization objective function, L (x (t), u (t) ═ βfuel·Qfuel(t)+βele·Pbatt(t), u (t) is engine output torque, x (t) is SOC of power battery, betafuelAs a price of fuel, Qfuel(t) is the fuel rate in L/s; beta is aeleFor the price of electricity, p is the prediction time domain, h (SOC (t +1)) is a penalty function for converging the SOC to a set value, β is a penalty coefficient, SOCr(t) is the SOC reference value at time t, and SOC (t) is the actual SOC value at time t.
Wherein h (SOC (t)) is a constraint function for the SOC reference track in the prediction time domain, the penalty coefficient beta is a very large positive number to ensure that the SOC reference track can be completely constrained, and when SOC (t) is in SOC (state of charge), (d) is a constraint function for the SOC reference track in the prediction time domain, the penalty coefficient beta is a very large positive numberr(t) lower h (SOC (t)) is a larger penalty, when SOC (t) is in SOCrThe upper time h (SOC (t)) is 0, so that the actual track of the SOC can be effectively restrainedAbove the SOC reference track, the actual SOC value is guaranteed to be always larger than the SOC reference value, and therefore the use of electric energy is guaranteed to be evenly distributed in the whole process.
As shown in fig. 1, the future engine output torque and the power battery output power of the vehicle predicted in step 5 are input into a plug-in hybrid system model (i.e., a PHEV model), so as to realize energy management of the hybrid vehicle.
In conclusion, the long-term driving working condition of the embodiment is periodically updated based on the real-time updated multi-source traffic information, so that the constructed long-term driving working condition is ensured to be consistent with the actual road condition, meanwhile, the embodiment realizes the energy management of the vehicle by combining the long-term driving working condition and the short-term driving working condition, is also the fusion of the multi-source information, can further ensure that the formulated vehicle energy management scheme is consistent with the traffic information of the current driving road section of the vehicle and the actual running state of the vehicle, and is more reasonable; and secondly, the reference track of the SOC of the power battery in the whole travel of the vehicle is updated regularly based on traffic information, so that electric energy can be better utilized, and the battery can be healthier to use.
Claims (7)
1. A hybrid electric vehicle energy management method based on multi-source information fusion is characterized by comprising the following steps:
acquiring traffic information of a current running road section of a vehicle, the actual speed of the vehicle and the actual SOC of a power battery in real time; the traffic information comprises preceding vehicle motion information, traffic flow information, traffic signal lamp information and road speed limit information, wherein the preceding vehicle motion information comprises the speed of a preceding vehicle and the relative distance between the current vehicle and the preceding vehicle;
establishing a future long-term running working condition of the vehicle based on a Markov chain by combining traffic information of a current running road section of the vehicle, predicting to obtain a vehicle speed-time relation curve of the whole travel of the vehicle, establishing an objective function of a power battery SOC (system on chip) by taking the running cost of the whole travel of the vehicle as the lowest principle, solving the objective function of the power battery SOC by using a dynamic programming algorithm by taking the vehicle speed-time relation curve as constraint, predicting to obtain a reference track of the power battery SOC of the whole travel of the vehicle, and updating the long-term running working condition and the reference track of the power battery SOC once at set time intervals by using the traffic information of the current running road section of the vehicle obtained in real time; the whole travel of the vehicle is the travel of the vehicle from the current position to the end position;
the method comprises the steps that the actual speed of a vehicle at the current moment, the speed of a front vehicle and the relative distance between the self vehicle and the front vehicle are combined, and the short-term speed data of the vehicle in the future are obtained through prediction;
and predicting to obtain the future engine output torque and the future power battery output power of the vehicle by combining the actual SOC of the power battery, the reference track of the SOC of the power battery and the short-term vehicle speed data of the vehicle in the future, so as to realize the energy management of the hybrid electric vehicle.
2. The hybrid electric vehicle energy management method based on multi-source information fusion of claim 1, characterized by further comprising the step of performing feedback correction on the vehicle speed data of the vehicle in the short term in the future by using the actual vehicle speed of the vehicle.
3. The hybrid electric vehicle energy management method based on multi-source information fusion according to claim 1 or 2, characterized in that the process of predicting the future engine output torque and power battery output power of the vehicle comprises: solving an energy management optimization objective function by using a simplified dynamic programming algorithm in combination with the actual SOC of the power battery, the reference track of the SOC of the power battery and the short-term vehicle speed data of the vehicle in the future, and predicting to obtain the future engine output torque and the future power output of the power battery of the vehicle; the energy management optimization objective function is as follows:
wherein,
wherein E is an energy management optimization objective function, L (x (t), u (t) ═ βfuel·Qfuel(t)+βele·Pbatt(t), u (t) is output torque of the engine, x (t) is SOC of the power battery, betafuelAs a price of fuel, Qfuel(t) is the fuel rate in L/s; beta is aeleAt a price of electricity, Pbatt(t) is the output power of the power battery, p is the prediction time domain, h (SOC (t +1)) is a penalty function for converging the SOC to a set value, beta is a penalty coefficient, and the SOC isr(t) is the SOC reference value at time t, and SOC (t) is the actual SOC value at time t.
4. The hybrid electric vehicle energy management method based on multi-source information fusion of claim 3, characterized in that the construction process of the future long-term running condition of the vehicle comprises the following steps: dividing the whole travel of the vehicle into M road sections, wherein M is more than or equal to 1, determining the average speed of the current road section according to the traffic information of the current road section, and obtaining a speed state cluster Q of the current road section according to the determined speed state cluster classification rule by combining the average speed of the current road section1Randomly selecting a cluster of vehicle speed states equal to Q from the vehicle speed segment candidate set1As a first road segment; determining a vehicle speed state cluster of the vehicle speed section according to the average vehicle speed of the vehicle speed section and the determined vehicle speed state cluster classification rule; determining a cluster of vehicle speed states Q for a next road segment based on a Markov chain2Selecting a cluster of vehicle speed states as Q from the vehicle speed segment candidate set2And the speed segment with the difference between the initial speed and the last speed value of the previous segment smaller than the set value is taken as a second segment; deleting the used vehicle speed sections from the vehicle speed section candidate set in the construction process, and repeating the steps until the total number of the constructed road sections is equal to the total number M of the road sections, so as to finish the construction of the long-term running working condition of the vehicle; the vehicle speed segment candidate set is obtained by dividing the historical vehicle speed data of the vehicle into vehicle speed segments, and the vehicle speed segment candidate set comprises a plurality of vehicle speed segments and each vehicle speed segmentCorresponding vehicle speed state clusters, and the dividing rule of the vehicle speed segment is as follows: the movement of the vehicle from one idle speed to the next idle speed is defined as a vehicle speed segment.
5. The hybrid vehicle energy management method based on multi-source information fusion of claim 4, characterized in that the Markov chain-based determination of the state cluster Q of the next road segment2The process comprises the following steps: generating a random number r between (0,1) based on the Monte Carlo method, if r satisfies
Then the state cluster Q of the next road segment is determined2Is q; wherein, PijIn order to regard the driving process of the vehicle as the transition probability of the vehicle speed segment from the vehicle speed state cluster i to the vehicle speed state cluster j when the driving process of the vehicle is regarded as the Markov process, Q is the total number of the categories of the vehicle speed state cluster.
6. The hybrid electric vehicle energy management method based on multi-source information fusion of claim 5, characterized in that an objective function of the power battery SOC is solved by using a simplified dynamic programming algorithm under a constraint condition in combination with a vehicle speed-time relation curve of the whole vehicle travel, and a reference trajectory of the power battery SOC of the whole vehicle travel is obtained by prediction; the target function of the SOC of the power battery is as follows:
Pbatt(t)＝VI(t)-I(t)2R
wherein J is an objective function of the SOC of the power battery, L (x (t), u (t)) is an instantaneous objective function, u (t) is the output torque of the engine, x (t) is the SOC of the power battery, and M is the driving working conditionThe number of divided segments; beta is afuelAs a price of fuel, Qfuel(t) is the fuel rate in L/s; beta is aeleAt a price of electricity, Pbatt(t) is the output power of the power battery, V is the open-circuit voltage of the power battery, R is the internal resistance of the power battery, I (t) is the discharge current of the power battery, D is the electric quantity of the power battery, and SOCiniIs an initial SOC value, SOCr(t) is the SOC reference value at time t;
the constraint conditions are as follows:
in the formula, Te(t)、Tm(t)、Treq(T) engine torque, motor torque and vehicle demand torque, respectively, the vehicle demand torque being obtained from a vehicle speed-time relationship curve of the entire vehicle travel, Tbrake(t) is mechanical braking torque, ωe(t) and ωmAnd (t) the engine rotating speed and the motor rotating speed respectively, SOC (t) is the SOC of the power battery, min represents a lower limit value, and max represents an upper limit value.
7. The hybrid electric vehicle energy management method based on multi-source information fusion of claim 1 or 2, characterized in that the actual vehicle speed, the vehicle speed in front and the relative distance between the vehicle and the vehicle in front at the current moment are input into a pre-trained BP neural network model to predict and obtain the vehicle speed data of the vehicle in the short term in the future.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011289194.3A CN112319461B (en) | 2020-11-17 | 2020-11-17 | Hybrid electric vehicle energy management method based on multi-source information fusion |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011289194.3A CN112319461B (en) | 2020-11-17 | 2020-11-17 | Hybrid electric vehicle energy management method based on multi-source information fusion |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN112319461A true CN112319461A (en) | 2021-02-05 |
| CN112319461B CN112319461B (en) | 2021-11-09 |
Family
ID=74322463
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011289194.3A Active CN112319461B (en) | 2020-11-17 | 2020-11-17 | Hybrid electric vehicle energy management method based on multi-source information fusion |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112319461B (en) |
Cited By (11)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113085666A (en) * | 2021-05-18 | 2021-07-09 | 北京理工大学 | Energy-saving driving method for layered fuel cell automobile |
| CN113240002A (en) * | 2021-05-11 | 2021-08-10 | 北京理工新源信息科技有限公司 | Internet of vehicles big data preprocessing system, device and method |
| CN113525344A (en) * | 2021-09-17 | 2021-10-22 | 北京航空航天大学 | Electric energy distribution method for P2 plug-in hybrid electric vehicle |
| CN113554337A (en) * | 2021-08-03 | 2021-10-26 | 青岛理工大学 | Plug-in hybrid electric vehicle energy management strategy construction method fusing traffic information |
| CN113581163A (en) * | 2021-07-26 | 2021-11-02 | 江苏大学 | Multi-mode PHEV mode switching optimization and energy management method based on LSTM |
| CN114506311A (en) * | 2022-02-22 | 2022-05-17 | 燕山大学 | Time-varying domain prediction energy management method and device, automobile and storage medium |
| CN116331184A (en) * | 2023-03-31 | 2023-06-27 | 长安大学 | Vehicle energy control method and system based on machine vision and radar information fusion |
| CN116901927A (en) * | 2023-08-08 | 2023-10-20 | 广州汽车集团股份有限公司 | Energy control method, device, equipment and storage medium |
| CN116985778A (en) * | 2023-09-27 | 2023-11-03 | 北京航空航天大学 | Hybrid vehicle global energy management optimization method under multitasking cross-core deployment |
| CN118034253A (en) * | 2024-03-11 | 2024-05-14 | 北京交通大学 | On-line optimization calibration method, system and equipment for control parameters of vehicle power system |
| CN118238797A (en) * | 2024-05-27 | 2024-06-25 | 比亚迪股份有限公司 | Intelligent management system, control method and related equipment for new energy vehicle energy |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120010767A1 (en) * | 2010-06-10 | 2012-01-12 | Massachusetts Institute Of Technology | Hybrid electric vehicle and method of control using path forecasting |
| CN103954913A (en) * | 2014-05-05 | 2014-07-30 | 哈尔滨工业大学深圳研究生院 | Predication method of electric vehicle power battery service life |
| CN108909702A (en) * | 2018-08-23 | 2018-11-30 | 北京理工大学 | A kind of plug-in hybrid-power automobile energy management method and system |
| WO2020005488A1 (en) * | 2018-06-26 | 2020-01-02 | Toyota Motor Engineering & Manufacturing North America, Inc. | Real time trajectory optimization for hybrid energy management utilizing connected information technologies |
| CN110696815A (en) * | 2019-11-21 | 2020-01-17 | 北京理工大学 | Prediction energy management method of network-connected hybrid electric vehicle |
| CN111267831A (en) * | 2020-02-28 | 2020-06-12 | 南京航空航天大学 | Hybrid vehicle intelligent time-domain-variable model prediction energy management method |
| CN111267827A (en) * | 2020-02-13 | 2020-06-12 | 山东中科先进技术研究院有限公司 | Energy management method and system for hybrid electric vehicle |
| US20210213933A1 (en) * | 2018-06-15 | 2021-07-15 | The Regents Of The University Of California | Systems, apparatus and methods to improve plug-in hybrid electric vehicle energy performance by using v2c connectivity |
-
2020
- 2020-11-17 CN CN202011289194.3A patent/CN112319461B/en active Active
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20120010767A1 (en) * | 2010-06-10 | 2012-01-12 | Massachusetts Institute Of Technology | Hybrid electric vehicle and method of control using path forecasting |
| CN103954913A (en) * | 2014-05-05 | 2014-07-30 | 哈尔滨工业大学深圳研究生院 | Predication method of electric vehicle power battery service life |
| US20210213933A1 (en) * | 2018-06-15 | 2021-07-15 | The Regents Of The University Of California | Systems, apparatus and methods to improve plug-in hybrid electric vehicle energy performance by using v2c connectivity |
| WO2020005488A1 (en) * | 2018-06-26 | 2020-01-02 | Toyota Motor Engineering & Manufacturing North America, Inc. | Real time trajectory optimization for hybrid energy management utilizing connected information technologies |
| CN108909702A (en) * | 2018-08-23 | 2018-11-30 | 北京理工大学 | A kind of plug-in hybrid-power automobile energy management method and system |
| CN110696815A (en) * | 2019-11-21 | 2020-01-17 | 北京理工大学 | Prediction energy management method of network-connected hybrid electric vehicle |
| CN111267827A (en) * | 2020-02-13 | 2020-06-12 | 山东中科先进技术研究院有限公司 | Energy management method and system for hybrid electric vehicle |
| CN111267831A (en) * | 2020-02-28 | 2020-06-12 | 南京航空航天大学 | Hybrid vehicle intelligent time-domain-variable model prediction energy management method |
Cited By (19)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113240002A (en) * | 2021-05-11 | 2021-08-10 | 北京理工新源信息科技有限公司 | Internet of vehicles big data preprocessing system, device and method |
| CN113240002B (en) * | 2021-05-11 | 2024-03-19 | 北京理工新源信息科技有限公司 | Internet of vehicles big data preprocessing system, device and method |
| WO2022241898A1 (en) * | 2021-05-18 | 2022-11-24 | 北京理工大学 | Hierarchical energy-saving driving method for fuel cell vehicle |
| CN113085666A (en) * | 2021-05-18 | 2021-07-09 | 北京理工大学 | Energy-saving driving method for layered fuel cell automobile |
| CN113581163B (en) * | 2021-07-26 | 2024-03-19 | 江苏大学 | Multimode PHEV mode switching optimization and energy management method based on LSTM |
| CN113581163A (en) * | 2021-07-26 | 2021-11-02 | 江苏大学 | Multi-mode PHEV mode switching optimization and energy management method based on LSTM |
| CN113554337A (en) * | 2021-08-03 | 2021-10-26 | 青岛理工大学 | Plug-in hybrid electric vehicle energy management strategy construction method fusing traffic information |
| CN113554337B (en) * | 2021-08-03 | 2023-08-08 | 青岛理工大学 | Plug-in hybrid electric vehicle energy management strategy construction method integrating traffic information |
| CN113525344A (en) * | 2021-09-17 | 2021-10-22 | 北京航空航天大学 | Electric energy distribution method for P2 plug-in hybrid electric vehicle |
| CN114506311A (en) * | 2022-02-22 | 2022-05-17 | 燕山大学 | Time-varying domain prediction energy management method and device, automobile and storage medium |
| CN114506311B (en) * | 2022-02-22 | 2023-06-20 | 燕山大学 | Variable time domain prediction energy management method and device, automobile and storage medium |
| CN116331184A (en) * | 2023-03-31 | 2023-06-27 | 长安大学 | Vehicle energy control method and system based on machine vision and radar information fusion |
| CN116331184B (en) * | 2023-03-31 | 2024-04-02 | 长安大学 | Vehicle energy control method and system based on machine vision and radar information fusion |
| CN116901927A (en) * | 2023-08-08 | 2023-10-20 | 广州汽车集团股份有限公司 | Energy control method, device, equipment and storage medium |
| CN116901927B (en) * | 2023-08-08 | 2024-05-14 | 广州汽车集团股份有限公司 | Energy control method, device, equipment and storage medium |
| CN116985778B (en) * | 2023-09-27 | 2023-12-08 | 北京航空航天大学 | Hybrid vehicle global energy management optimization method under multitasking cross-core deployment |
| CN116985778A (en) * | 2023-09-27 | 2023-11-03 | 北京航空航天大学 | Hybrid vehicle global energy management optimization method under multitasking cross-core deployment |
| CN118034253A (en) * | 2024-03-11 | 2024-05-14 | 北京交通大学 | On-line optimization calibration method, system and equipment for control parameters of vehicle power system |
| CN118238797A (en) * | 2024-05-27 | 2024-06-25 | 比亚迪股份有限公司 | Intelligent management system, control method and related equipment for new energy vehicle energy |
Also Published As
| Publication number | Publication date |
|---|---|
| CN112319461B (en) | 2021-11-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN112319461B (en) | Hybrid electric vehicle energy management method based on multi-source information fusion | |
| Wu et al. | Continuous reinforcement learning of energy management with deep Q network for a power split hybrid electric bus | |
| WO2021103625A1 (en) | Short-term vehicle speed condition real-time prediction method based on interaction between vehicle ahead and current vehicle | |
| Lei et al. | An adaptive equivalent consumption minimization strategy for plug-in hybrid electric vehicles based on traffic information | |
| Wu et al. | A predictive energy management strategy for multi-mode plug-in hybrid electric vehicles based on multi neural networks | |
| Chen et al. | Energy management of a power-split plug-in hybrid electric vehicle based on genetic algorithm and quadratic programming | |
| CN109733378B (en) | Offline optimized online predicted torque distribution method | |
| US8612077B2 (en) | Hybrid electric vehicle and method of path dependent receding horizon control | |
| Li et al. | Data-driven based eco-driving control for plug-in hybrid electric vehicles | |
| Liu et al. | Energy management for hybrid electric vehicles based on imitation reinforcement learning | |
| CN109910866B (en) | Hybrid electric vehicle energy management method and system based on road condition prediction | |
| CN111267831A (en) | Hybrid vehicle intelligent time-domain-variable model prediction energy management method | |
| Singh et al. | Fuzzy logic and Elman neural network tuned energy management strategies for a power-split HEVs | |
| Boyali et al. | Real-time controller design for a parallel hybrid electric vehicle using neuro-dynamic programming method | |
| Liu et al. | Prediction of vehicle driving conditions with incorporation of stochastic forecasting and machine learning and a case study in energy management of plug-in hybrid electric vehicles | |
| Zeng et al. | A two-level stochastic approach to optimize the energy management strategy for fixed-route hybrid electric vehicles | |
| Liu et al. | Cooperative optimization of velocity planning and energy management for connected plug-in hybrid electric vehicles | |
| Katsargyri et al. | Optimally controlling hybrid electric vehicles using path forecasting | |
| CN110406526A (en) | Parallel hybrid electric energy management method based on adaptive Dynamic Programming | |
| WO2022028257A1 (en) | Method for predicting energy consumption-recovery ratio of new energy vehicle, and energy saving control method and system | |
| CN112339756A (en) | New energy automobile traffic light intersection energy recovery optimization speed planning algorithm based on reinforcement learning | |
| Zhang et al. | Energy management strategy of a novel parallel electric-hydraulic hybrid electric vehicle based on deep reinforcement learning and entropy evaluation | |
| CN115805840A (en) | Energy consumption control method and system for range-extending type electric loader | |
| Pan et al. | Grey wolf fuzzy optimal energy management for electric vehicles based on driving condition prediction | |
| CN115214607B (en) | Energy management method for plug-in hybrid electric vehicle |
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 |