WO2024066702A1 - 混合动力车辆及其能量管理方法、装置及介质、电子设备 - Google Patents

混合动力车辆及其能量管理方法、装置及介质、电子设备 Download PDF

Info

Publication number
WO2024066702A1
WO2024066702A1 PCT/CN2023/108978 CN2023108978W WO2024066702A1 WO 2024066702 A1 WO2024066702 A1 WO 2024066702A1 CN 2023108978 W CN2023108978 W CN 2023108978W WO 2024066702 A1 WO2024066702 A1 WO 2024066702A1
Authority
WO
WIPO (PCT)
Prior art keywords
operating condition
hybrid vehicle
charge
sequence
energy management
Prior art date
Application number
PCT/CN2023/108978
Other languages
English (en)
French (fr)
Inventor
杨冬生
朱福堂
王春生
沈涛
武金龙
Original Assignee
比亚迪股份有限公司
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by 比亚迪股份有限公司 filed Critical 比亚迪股份有限公司
Publication of WO2024066702A1 publication Critical patent/WO2024066702A1/zh

Links

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
    • B60W10/00—Conjoint control of vehicle sub-units of different type or different function
    • B60W10/04—Conjoint control of vehicle sub-units of different type or different function including control of propulsion units
    • B60W10/06—Conjoint control of vehicle sub-units of different type or different function including control of propulsion units including control of combustion engines
    • 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
    • B60W10/00—Conjoint control of vehicle sub-units of different type or different function
    • B60W10/04—Conjoint control of vehicle sub-units of different type or different function including control of propulsion units
    • B60W10/08—Conjoint control of vehicle sub-units of different type or different function including control of propulsion units including control of electric propulsion units, e.g. motors or generators
    • 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
    • B60W40/00—Estimation or calculation of non-directly measurable driving parameters for road vehicle drive control systems not related to the control of a particular sub unit, e.g. by using mathematical models

Definitions

  • the present disclosure relates to the field of vehicle technology, and in particular to a hybrid vehicle and an energy management method, device and medium, and electronic equipment thereof.
  • the energy management strategy of hybrid vehicles is based on a dynamic programming algorithm, which adopts an equivalent fuel consumption minimum strategy (ECMS), and its equivalent factor is a fixed value.
  • ECMS equivalent fuel consumption minimum strategy
  • this strategy can only achieve optimal control under specific working conditions. When the driving conditions change, this strategy cannot guarantee the vehicle's fuel economy.
  • the present disclosure aims to solve at least one of the technical problems in the related art to a certain extent.
  • the purpose of the present disclosure is to propose a hybrid vehicle and its energy management method, device and medium, and electronic equipment to achieve online control of the hybrid vehicle and global optimal control of the hybrid vehicle to ensure fuel economy.
  • an embodiment of the first aspect of the present disclosure proposes an energy management method for a hybrid vehicle, the method comprising: obtaining an operating condition category sequence of a current navigation route of the hybrid vehicle and the mileage of each operating condition in the current navigation route, the operating condition category sequence and the mileage of each operating condition being obtained based on road characteristic parameters of the current navigation route; obtaining a target state of charge sequence based on the operating condition category sequence and the mileage of each operating condition; obtaining an equivalent factor sequence based on the operating condition category sequence and the target state of charge sequence; obtaining the instantaneous output power of the power battery of the hybrid vehicle at each moment based on the equivalent factor sequence; and controlling the hybrid vehicle based on the instantaneous output power of the power battery.
  • the energy management method for a hybrid vehicle of the above embodiment of the present disclosure may also have the following additional technical features:
  • the step of obtaining a target state of charge sequence according to the operating condition category sequence and the mileage of each operating condition includes: obtaining an actual state of charge of the power battery at the starting point of the current navigation route; determining the target state of charge of the hybrid vehicle at the starting point of the current navigation route according to the actual state of charge, the operating condition category sequence and the mileage of each operating condition.
  • the charge state variation range at the end of each operating condition is obtained; and the target charge state sequence is obtained according to the charge state variation range at the end of each operating condition.
  • the charge state variation range at the end of the first operating condition of the current navigation route is obtained based on the actual charge state, the road characteristic data of the first operating condition and the mileage of the first operating condition; the charge state variation range at the end of a non-first operating condition of the current navigation route is obtained based on the charge state variation range at the end of the operating condition before the non-first operating condition, the road characteristic data of the non-first operating condition and the mileage of the non-first operating condition.
  • the road characteristic data includes slope data and speed limit data.
  • the upper limit value of the charge state variation range of the operating condition is the charge state of the hybrid vehicle at the end of operating the operating condition in the power generation mode
  • the lower limit value of the charge state variation range of the operating condition is the charge state of the hybrid vehicle at the end of operating the operating condition in the pure electric mode
  • the target state of charge sequence is obtained by selecting a target state of charge in the range of state of charge variation corresponding to each operating condition, and the target state of charge sequence is obtained according to the selected target states of charge.
  • the operating condition category sequence and the mileage of each operating condition are obtained using a trained neural network model based on road characteristic parameters of the current navigation route
  • the training process of the neural network model includes: obtaining historical driving parameters of the hybrid vehicle on the current navigation route, and determining historical road characteristic parameters based on the historical driving parameters, and clustering the historical road characteristic parameters to obtain multiple operating condition categories; constructing a training data set based on the historical road characteristic parameters and the operating condition categories; constructing a neural network model, and training the neural network model using the training data set.
  • the operating condition categories include: ordinary urban roads, lightly congested urban roads, moderately congested urban roads, severely congested urban roads, expressways, highways, suburban roads, and township roads.
  • the road characteristic parameters include at least one of the following: average vehicle speed, maximum vehicle speed, speed standard deviation, average acceleration, maximum acceleration, minimum acceleration, acceleration standard deviation, acceleration time ratio, deceleration time ratio, uniform speed time ratio, idling time ratio, and accumulated mileage.
  • the instantaneous output power of the battery of the hybrid vehicle under the working condition corresponding to the equivalent factor is calculated according to the following formula:
  • H(u, SOC(t), t) is the Hamiltonian function established according to the equivalent fuel consumption minimum strategy
  • arg H(u, SOC(t), t) is the instantaneous output power of the power battery at time t
  • s(t) is the equivalent factor at time t
  • SOC(t) is the state of charge of the power battery at time t
  • u is the fuel consumption.
  • the controlling of the hybrid vehicle according to the instantaneous output power of the power battery includes: obtaining the instantaneous power demand of the hybrid vehicle at time t; subtracting the instantaneous output power of the power battery at time t from the instantaneous power demand at time t to obtain the instantaneous power demand of the engine of the hybrid vehicle at time t; and controlling the power battery and the engine according to the instantaneous output power of the power battery at time t and the instantaneous output power of the engine at time t.
  • a second aspect of the present disclosure provides a computer-readable storage medium having a computer program stored thereon.
  • the computer program is executed by a processor, the above hybrid vehicle energy management method is implemented.
  • the third aspect of the present disclosure proposes an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor.
  • the computer program is executed by the processor, the above hybrid vehicle energy management method is implemented.
  • the fourth aspect of the present disclosure proposes an energy management device for a hybrid vehicle, the device comprising: an acquisition module, used to acquire the operating condition category sequence of the current navigation route of the hybrid vehicle and the mileage of each operating condition in the current navigation route, the operating condition category sequence and the mileage of each operating condition are obtained according to the road characteristic parameters of the current navigation route; a target state of charge sequence is obtained according to the operating condition category sequence and the mileage of each operating condition; an equivalent factor sequence is obtained according to the operating condition category sequence and the target state of charge sequence; the instantaneous output power of the power battery of the hybrid vehicle at each moment is obtained according to the equivalent factor sequence; an energy management module, used to control the hybrid vehicle according to the instantaneous output power of the power battery.
  • a fifth aspect of the present disclosure proposes a hybrid vehicle, including the energy management device of the hybrid vehicle mentioned above.
  • the hybrid vehicle and its energy management method, device, medium, and electronic device of the disclosed embodiment can realize online control of the hybrid vehicle by obtaining the road characteristic parameters of the current navigation route of the hybrid vehicle, and using the pre-trained neural network model to obtain the working condition category sequence and the mileage of each working condition of the current navigation route according to the road characteristic parameters, and further, after obtaining the working condition category sequence and the mileage of each working condition, it is also necessary to obtain the target state of charge sequence according to the working condition category sequence and the mileage of each working condition, so as to obtain the equivalent factor sequence according to the working condition category sequence and the target state of charge sequence, so as to realize that the equivalent factor changes according to the working condition category of the road section, and ensure fuel economy.
  • the equivalent factor sequence is obtained based on the current navigation route, and the instantaneous output power of the power battery of the hybrid vehicle at each moment is obtained according to the equivalent factor sequence, the hybrid vehicle is controlled according to the instantaneous output power of the power battery, and energy management is realized based on the equivalent factor sequence on the current navigation route, so that the energy management is not affected by the frequency of the navigation map data output, the required calculation amount is small, and the fuel economy can be better ensured under the condition of changing working conditions.
  • FIG1 is a flow chart of an energy management method for a hybrid vehicle according to an embodiment of the present disclosure
  • FIG2 is a schematic diagram of a charge state change path of an example of the present disclosure
  • FIG3 is a schematic diagram of a structure of a neural network model of an example of the present disclosure.
  • FIG4 is a structural block diagram of an energy management device for a hybrid vehicle according to an embodiment of the present disclosure
  • FIG. 5 is a structural block diagram of a hybrid vehicle according to an embodiment of the present disclosure.
  • FIG. 1 is a flow chart of an energy management method for a hybrid vehicle according to an embodiment of the present disclosure.
  • the energy management method of a hybrid vehicle includes:
  • a whole vehicle simulation model is built in advance, and the energy consumption and other related performance of the hybrid vehicle under various working conditions are calculated offline through the whole vehicle simulation model, and the working conditions of the hybrid vehicle are divided into multiple working condition categories based on the performance calculation results.
  • the neural network model is trained in advance so that the neural network model can output a sequence of working condition categories after inputting the road feature parameters of the navigation map.
  • the road characteristic parameters of the hybrid vehicle's current navigation route are obtained according to the navigation map, and the road characteristic parameters obtained through the navigation map are input into the pre-trained neural network model to obtain the operating condition category and mileage of each section in the current navigation route.
  • the operating condition categories are sorted to obtain an operating condition category sequence. According to the operating condition category sequence and the mileage of each operating condition, the target state of charge sequence corresponding to the current navigation route is obtained, so as to obtain the equivalent factor sequence using the equivalent fuel consumption minimum strategy, and the energy management of the hybrid vehicle is performed according to the equivalent factor sequence.
  • the pre-trained neural network obtains the working condition category sequence and the mileage of each working condition of the current navigation route according to the road characteristic parameters, so as to realize the online control of the hybrid vehicle. Moreover, after obtaining the working condition category sequence and the mileage of each working condition, it is also necessary to obtain the target state of charge sequence according to the working condition category sequence and the mileage of each working condition, so as to obtain the equivalent factor sequence according to the working condition category sequence, the mileage of each working condition and the target state of charge sequence, so as to realize the change of the equivalent factor according to the working condition category of the road section to ensure fuel economy.
  • the equivalent factor sequence is obtained based on the current navigation route, and energy management is performed on the current navigation route based on the equivalent factor sequence, so that the energy management is not affected by the frequency of the navigation map data output, the required calculation amount is small, and the fuel economy can be better guaranteed under the condition of changing working conditions.
  • the target state of charge sequence is obtained according to the operating condition category sequence and the mileage of each operating condition, including: obtaining the current actual state of charge of the power battery; determining the range of charge change of the hybrid vehicle at the end of each operating condition according to the actual state of charge, the operating condition category sequence and the mileage of each operating condition; and obtaining the target state of charge sequence according to the range of charge change at the end of each operating condition.
  • the operating condition category sequence and the mileage of each operating condition can be sent to a third-party platform to obtain a target state of charge sequence, or the target state of charge sequence can be directly calculated locally.
  • the first state of charge and the second state of charge at the end of the first operating condition can be determined according to the current actual state of charge of the hybrid vehicle, the road characteristic data corresponding to the first operating condition in the operating condition category sequence, and the mileage of the first operating condition, and the charge state change range at the end of the first operating condition is obtained according to the first state of charge and the second state of charge, wherein the road characteristic data includes slope data and speed limit data, the first state of charge is the charge state of the hybrid vehicle at the end of the first operating condition in the power generation mode, and the second state of charge is the charge state of the hybrid vehicle at the end of the first operating condition in the pure electric mode; for each operating condition except the first operating condition in the operating condition category sequence, the charge state change range of the hybrid vehicle at the end of the operating condition is determined according to the charge state change range at the end of the operating condition corresponding to the previous operating condition, the road characteristic data corresponding to the operating condition, and the mileage of the operating condition.
  • a target state of charge may be selected from the range of change of the state of charge corresponding to each operating condition; and a target state of charge sequence may be obtained according to each target state of charge.
  • the first operating condition is operating condition 1 corresponding to segment AB
  • the actual state of charge of the hybrid vehicle at point A is F
  • the hybrid vehicle adopts the power generation mode, that is, the hybrid vehicle uses all fuel in operating condition 1 from point A to point B
  • the battery is in a charging state
  • the hybrid vehicle adopts the pure electric mode, that is, the hybrid vehicle uses all electricity in operating condition 1 from point A to point B
  • the battery is in a discharging state
  • the second state of charge at point B to be I that is, the lower limit of the state of charge of operating condition 1 is I
  • the range of charge change of the hybrid vehicle under operating condition 2 is determined.
  • the upper limit value of the charge of operating condition 1 is G as the actual charge of operating condition 2
  • the hybrid vehicle adopts the power generation mode, namely, the hybrid vehicle uses all fuel under operating condition 2, and the battery is in a charging state, to determine that the third charge of point C is J, that is, the upper limit value of the charge of operating condition 2 is J.
  • the lower limit value of the charge of operating condition 1 is I as the actual charge of operating condition 2, and the hybrid vehicle adopts the pure electric mode, namely, the hybrid vehicle uses all electricity under operating condition 2 from point B to point C, and the battery is in a discharging state, to determine that the fourth charge of point C is L, that is, the lower limit value of the charge of operating condition 2 is L, thereby determining that the range of charge change under operating condition 2 is [L, J].
  • the actual state of charge at point A is F.
  • point H is selected as the target state of charge value within the state of charge variation range [I, G]
  • F-H is a state of charge variation path for operating condition 1.
  • the target state of charge value of AB segment condition 1 also includes the state of charge upper limit value G and the state of charge lower limit value I.
  • AB segment condition 1 can form three state of charge change paths of F-G, F-H, and F-I.
  • the actual state of charge of BC segment condition 2 can be the target state of charge value of AB segment, that is, including G, H, and I.
  • the preset reference state of charge of BC segment includes J, K, and L. At this time, BC segment can form nine state of charge change paths of G-J, G-K, G-L, H-J, H-K, H-L, I-J, I-K, and I-L.
  • CD segment condition 3 can form fifteen state of charge change paths
  • the state-of-charge change path with the lowest fuel consumption and power consumption is selected as the optimal state-of-charge change path by comparison, and the optimal state-of-charge change path is used as the target state-of-charge change path of the working condition to control the hybrid vehicle for energy management.
  • the fuel consumption and power consumption of the state-of-charge change paths F-G, F-H and F-I are obtained respectively, and the fuel consumption and power consumption of the state-of-charge change path F-I are calculated to be the lowest, so the state-of-charge change path F-I is used as the target state-of-charge change path of the AB segment, and I is the optimal state of charge at the end of working condition 1.
  • the target state-of-charge change path of the BC segment is determined to be I-J
  • the target state-of-charge change path of the CD segment is determined to be J-P
  • the target state-of-charge change path of the DE segment is determined to be P-U
  • the target state-of-charge change path under the pre-driving road AE is F-I-J-P-U.
  • this process can be carried out on a third-party platform, which can reduce the amount of calculation for the vehicle, reduce the calculation time of the vehicle, and ensure fuel economy when the operating conditions change.
  • the training process of the neural network model includes: obtaining the historical driving parameters of the hybrid vehicle on the current navigation route, determining the historical road characteristic parameters based on the historical driving parameters, and clustering the historical road characteristic parameters to obtain multiple operating condition categories; constructing a training data set based on the historical road characteristic parameters and the operating condition categories; constructing a neural network model, and training the neural network model using the training data set.
  • the neural network model includes an input layer, a hidden layer, and an output layer, the input layer is used to input the road characteristic parameters after dimensionality reduction processing, and the output layer is used to output the operating condition category.
  • a neural network model as shown in FIG3 is constructed, including an input layer, a hidden layer and an output, etc.
  • the input layer inputs x1 , x2 , ... xn are road characteristic parameters, and in the specific example shown in FIG3, n is 4, and the neural network model supports road characteristic parameters with a dimension of 4.
  • the output layer outputs y as the working condition category.
  • the historical driving parameters of the hybrid vehicle on the current navigation route are obtained, and the historical road characteristic parameters are determined according to the historical driving parameters.
  • a cluster analysis algorithm is used to classify working conditions, that is, the historical road characteristic parameters are divided into different working condition categories using the Euclidean distance method, so that the neural network model is trained according to the historical road characteristic parameters and the corresponding working condition categories.
  • the historical road characteristic parameters can be reduced in dimension using the principal component analysis method, so as to train the neural network model using the historical road characteristic parameters after the dimension reduction process.
  • the neural network model also obtains the working condition category sequence and the mileage of each working condition of the current navigation route based on the road characteristic parameters after the dimension reduction process.
  • the operating condition categories include: ordinary urban roads, lightly congested urban roads, moderately congested urban roads, severely congested urban roads, expressways, highways, suburban roads, and township roads.
  • the road characteristic parameters include at least one of the following: average vehicle speed, maximum vehicle speed, speed standard deviation, average acceleration, maximum acceleration, minimum acceleration, acceleration standard deviation, acceleration time ratio, deceleration time ratio, uniform speed time ratio, idling time ratio, and accumulated mileage.
  • the instantaneous output power of the power battery of the hybrid vehicle at each moment is obtained according to the equivalent factor sequence, including: for each equivalent factor in the equivalent factor sequence, the instantaneous output power of the battery of the hybrid vehicle corresponding to the equivalent factor is calculated according to the following formula:
  • H(u, SOC(t), t) is the Hamiltonian function established according to the equivalent fuel consumption minimum strategy
  • arg H(u, SOC(t), t) is the instantaneous output power of the power battery of the hybrid vehicle at time t
  • s(t) is the equivalent factor at time t
  • SOC(t) is the state of charge of the power battery at time t
  • u is the fuel consumption.
  • a Hamiltonian function is established, and the principle of the minimum equivalent fuel consumption strategy is used to calculate each classified working condition and construct the relationship between the working condition category, the target state of charge and the equivalent factor.
  • Solving the Hamiltonian function can establish the relationship between the battery target state of charge and the equivalent factor under the current working condition category, thereby determining the optimal equivalent factor under each target state of charge of the current working condition category.
  • the instantaneous output power of the battery of the hybrid vehicle corresponding to the optimal equivalent factor can be obtained, so that the hybrid vehicle can be controlled according to the instantaneous output power of the power battery.
  • the instantaneous required power of the hybrid vehicle at time t can be obtained, and the instantaneous output power of the power battery at time t is subtracted from the instantaneous required power at time t to obtain the instantaneous required power of the engine of the hybrid vehicle at time t.
  • the power battery and the engine are controlled.
  • the output power can be obtained based on the equivalent factor of real-time optimization, realizing the optimal energy management in the entire time domain of the entire navigation route.
  • the energy management method for a hybrid vehicle of the disclosed embodiment can realize online control of the hybrid vehicle by obtaining the road characteristic parameters of the current navigation route of the hybrid vehicle, and using the pre-trained neural network model to obtain the operating condition category sequence and the mileage of each operating condition of the current navigation route according to the road characteristic parameters. Moreover, after obtaining the operating condition category sequence and the mileage of each operating condition, it is also necessary to obtain the target state of charge sequence according to the operating condition category sequence and the mileage of each operating condition, so as to obtain the equivalent factor sequence according to the operating condition category sequence, the mileage of each operating condition and the target state of charge sequence, so as to realize that the equivalent factor changes according to the operating condition category of the road section, and ensure fuel economy.
  • the equivalent factor sequence is obtained based on the current navigation route, and energy management is performed on the current navigation route based on the equivalent factor sequence, so that the energy management is not affected by the frequency of the navigation map data output, the required calculation amount is small, and the fuel economy can be better ensured under the condition of changing operating conditions.
  • the present disclosure proposes a computer-readable storage medium.
  • a computer program is stored on the processor, and when the computer program is executed by the processor, the energy management method of the hybrid vehicle described above is implemented.
  • the computer-readable storage medium of the disclosed embodiment can realize the above-mentioned hybrid vehicle energy management method by acquiring the road characteristic parameters of the current navigation route of the hybrid vehicle, and using the pre-trained neural network model to obtain the operating condition category sequence and the mileage of each operating condition of the current navigation route according to the road characteristic parameters, thereby realizing online control of the hybrid vehicle. Moreover, after obtaining the operating condition category sequence and the mileage of each operating condition, it is also necessary to obtain the target state of charge sequence according to the operating condition category sequence and the mileage of each operating condition, thereby obtaining the equivalent factor sequence according to the operating condition category sequence, the mileage of each operating condition and the target state of charge sequence, thereby realizing that the equivalent factor changes according to the operating condition category of the road section, and ensuring fuel economy.
  • the equivalent factor sequence is obtained based on the current navigation route, and energy management is performed on the current navigation route based on the equivalent factor sequence, so that energy management is not affected by the operating condition category sequence.
  • the frequency of the map data output is affected by the small amount of calculation required, which can better ensure fuel economy under changing operating conditions.
  • the present disclosure proposes an electronic device.
  • the electronic device includes a memory, a processor, and a computer program stored in the memory and executable on the processor.
  • the computer program is executed by the processor, the above-mentioned hybrid vehicle energy management method is implemented.
  • the electronic device of the disclosed embodiment by implementing the above-mentioned hybrid vehicle energy management method, can obtain the road characteristic parameters of the current navigation route of the hybrid vehicle, and use the pre-trained neural network model to obtain the operating condition category sequence and the mileage of each operating condition of the current navigation route according to the road characteristic parameters, so as to realize online control of the hybrid vehicle. Moreover, after obtaining the operating condition category sequence and the mileage of each operating condition, it is also necessary to obtain the target state of charge sequence according to the operating condition category sequence and the mileage of each operating condition, so as to obtain the equivalent factor sequence according to the operating condition category sequence, the mileage of each operating condition and the target state of charge sequence, so as to realize that the equivalent factor changes according to the operating condition category of the road section, and ensure fuel economy.
  • the equivalent factor sequence is obtained based on the current navigation route, and energy management is performed on the current navigation route based on the equivalent factor sequence, so that the energy management is not affected by the frequency of the navigation map data output, the required calculation amount is small, and the fuel economy can be better ensured under the condition of changing operating conditions.
  • FIG. 4 is a structural block diagram of an energy management device for a hybrid vehicle according to an embodiment of the present disclosure.
  • the energy management device 100 of a hybrid vehicle includes: an acquisition module 101 and an energy management module 102 .
  • the acquisition module 101 is used to obtain the road characteristic parameters of the current navigation route of the hybrid vehicle, and to use a pre-trained neural network model to obtain the operating condition category sequence and the mileage of each operating condition of the current navigation route based on the road characteristic parameters, and to obtain the target state of charge sequence based on the operating condition category sequence and the mileage of each operating condition, and to use the equivalent fuel consumption minimum strategy to obtain the equivalent factor sequence based on the operating condition category sequence, the mileage of each operating condition and the target state of charge sequence, and to obtain the instantaneous output power of the power battery of the hybrid vehicle at each moment based on the equivalent factor sequence; the energy management module 102 is used to control the hybrid vehicle according to the instantaneous output power of the power battery.
  • the energy management device for a hybrid vehicle of the disclosed embodiment can achieve online control of the hybrid vehicle by acquiring the road characteristic parameters of the current navigation route of the hybrid vehicle and using a pre-trained neural network model to obtain the operating condition category sequence and the mileage of each operating condition of the current navigation route based on the road characteristic parameters. Moreover, after obtaining the operating condition category sequence and the mileage of each operating condition, it is also necessary to obtain the target state of charge sequence based on the operating condition category sequence and the mileage of each operating condition, thereby obtaining an equivalent factor sequence based on the operating condition category sequence, the mileage of each operating condition and the target state of charge sequence, thereby achieving the change of the equivalent factor according to the operating condition category of the road section to ensure fuel economy.
  • the equivalent factor sequence is obtained based on the current navigation route, and the equivalent factor sequence is obtained based on the current navigation route.
  • Energy management is performed based on the efficiency factor sequence, so that energy management is not affected by the frequency of navigation map data output, the required calculation amount is small, and fuel economy can be better guaranteed under changing working conditions.
  • the present disclosure proposes a hybrid vehicle.
  • FIG. 5 is a structural block diagram of a hybrid vehicle according to an embodiment of the present disclosure.
  • a hybrid vehicle 10 includes the above-described hybrid vehicle energy management device 100 .
  • the hybrid vehicle of the disclosed embodiment can obtain the road characteristic parameters of the current navigation route of the hybrid vehicle through the energy management device of the hybrid vehicle, and obtain the operating condition category sequence and the mileage of each operating condition of the current navigation route according to the road characteristic parameters using the pre-trained neural network model, so as to realize online control of the hybrid vehicle. Moreover, after obtaining the operating condition category sequence and the mileage of each operating condition, it is also necessary to obtain the target state of charge sequence according to the operating condition category sequence and the mileage of each operating condition, so as to obtain the equivalent factor sequence according to the operating condition category sequence, the mileage of each operating condition and the target state of charge sequence, so as to realize that the equivalent factor changes according to the operating condition category of the road section, and ensure fuel economy.
  • the equivalent factor sequence is obtained based on the current navigation route, and energy management is performed on the current navigation route based on the equivalent factor sequence, so that the energy management is not affected by the frequency of the navigation map data output, the required calculation amount is small, and the fuel economy can be better ensured under the condition of changing operating conditions.
  • computer-readable media include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM).
  • the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or processing in other suitable ways if necessary, and then stored in a computer memory.
  • first and second are used for descriptive purposes only and should not be understood as indicating or implying relative importance or implicitly indicating the number of technical features indicated.
  • a feature defined as “first” or “second” may explicitly or implicitly include at least one of the features.
  • “plurality” means at least two, such as two, three, etc., unless otherwise clearly and specifically defined.
  • the terms “installed”, “connected”, “connected”, “fixed” and the like should be understood in a broad sense, for example, it can be a fixed connection, a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined.
  • installed can be a fixed connection, a detachable connection, or an integral connection
  • it can be a mechanical connection or an electrical connection
  • it can be a direct connection or an indirect connection through an intermediate medium, it can be the internal connection of two elements or the interaction relationship between two elements, unless otherwise clearly defined.
  • the specific meanings of the above terms in this disclosure can be understood according to specific circumstances.
  • a first feature being “above” or “below” a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium.
  • a first feature being “above”, “above” or “above” a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature.
  • a first feature being “below”, “below” or “below” a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

Landscapes

  • Engineering & Computer Science (AREA)
  • Chemical & Material Sciences (AREA)
  • Combustion & Propulsion (AREA)
  • Transportation (AREA)
  • Mechanical Engineering (AREA)
  • Automation & Control Theory (AREA)
  • Physics & Mathematics (AREA)
  • Mathematical Physics (AREA)
  • Electric Propulsion And Braking For Vehicles (AREA)
  • Hybrid Electric Vehicles (AREA)

Abstract

一种混合动力车辆(10)及其能量管理方法、装置(100)及介质、电子设备，涉及车辆技术领域。其中，方法包括：S12、获取混合动力车辆(10)的当前导航路线的工况类别序列和当前导航路线中各工况的里程，工况类别序列和各工况的里程是根据当前导航路线的道路特征参数得到的；S13、根据工况类别序列和各工况的里程获取目标荷电状态序列；根据工况类别序列和目标荷电状态序列得到等效因子序列；S14、根据等效因子序列得到混合动力车辆(10)的动力电池在各时刻的瞬时输出功率；根据动力电池的瞬时输出功率对混合动力车辆(10)进行控制。以实现混合动力车辆(10)的在线控制，以及实现混合动力车辆(10)的全局最优控制，保证燃油经济性。

Description

混合动力车辆及其能量管理方法、装置及介质、电子设备
相关申请的交叉引用
本公开要求于2022年09月29日提交的申请号为202211203935.0、名称为“混合动力车辆及其能量管理方法、装置及介质、电子设备”的中国专利申请的优先权，其全部内容通过引用结合在本公开中。
技术领域
本公开涉及车辆技术领域，尤其涉及一种混合动力车辆及其能量管理方法、装置及介质、电子设备。
背景技术
相关技术中，混合动力车辆的能量管理策略基于动态规划算法，该算法采用等效燃油消耗最小策略(Equivalent Consumption Minimum Strategy,ECMS)，其等效因子是一个定值。然而，该策略只能在特定的工况下实现最优控制，当行驶工况发生变化后，这种策略无法保证车辆燃油经济性。
发明内容
本公开旨在至少在一定程度上解决相关技术中的技术问题之一。为此，本公开的目的在于提出一种混合动力车辆及其能量管理方法、装置及介质、电子设备，以实现混合动力车辆的在线控制，以及实现混合动力车辆的全局最优控制，保证燃油经济性。
为达到上述目的，本公开第一方面实施例提出了一种混合动力车辆的能量管理方法，方法包括：获取混合动力车辆的当前导航路线的工况类别序列和当前导航路线中各工况的里程，工况类别序列和各工况的里程是根据当前导航路线的道路特征参数得到的；根据工况类别序列和各工况的里程获取目标荷电状态序列；根据工况类别序列和目标荷电状态序列得到等效因子序列；根据等效因子序列得到混合动力车辆的动力电池在各时刻的瞬时输出功率；根据动力电池的瞬时输出功率对混合动力车辆进行控制。
另外，本公开上述实施例的混合动力车辆的能量管理方法还可以具有如下附加的技术特征：
根据本公开的一个实施例，所述根据所述工况类别序列和所述各工况的里程获取目标荷电状态序列，包括：获取所述动力电池在所述当前导航路线起点的实际荷电状态；根据所述实际荷电状态、所述工况类别序列和所述各工况的里程，确定所述混合动力车辆在所 述各工况结束时的荷电状态变化范围；根据所述各工况结束时的荷电状态变化范围，得到所述目标荷电状态序列。
根据本公开的一个实施例，所述当前导航路线的首个工况结束时的荷电状态变化范围是根据所述实际荷电状态、所述首个工况的道路特征数据和所述首个工况的里程得到的；所述当前导航路线的非首个工况结束时的荷电状态变化范围是根据所述非首个工况的前一个工况结束时的荷电状态变化范围、所述非首个工况的道路特征数据和所述非首个工况的里程得到的。
根据本公开的一个实施例，所述道路特征数据包括坡度数据和限速数据。
根据本公开的一个实施例，所述工况的荷电状态变化范围的上限值是所述混合动力车辆以发电模式在所述工况运行结束时的荷电状态，所述工况的荷电状态变化范围的下限值是所述混合动力车辆以纯电模式在所述工况运行结束时的荷电状态。
根据本公开的一个实施例，所述目标荷电状态序列是分别在每个工况对应的荷电状态变化范围中选取一个目标荷电状态，根据选取的各个目标荷电状态得到的。
根据本公开的一个实施例，所述工况类别序列和所述各工况的里程是利用训练好的神经网络模型，根据所述当前导航路线的道路特征参数得到的，所述神经网络模型的训练过程包括：获取所述混合动力车辆在当前导航路线上的历史行驶参数，并根据所述历史行驶参数确定历史道路特征参数，以及对历史道路特征参数进行聚类处理，得到多个工况类别；基于所述历史道路特征参数和所述工况类别，构建训练数据集；构建神经网络模型，并利用所述训练数据集对所述神经网络模型进行训练。
根据本公开的一个实施例，所述工况类别包括：普通城市路、城市轻度拥堵路、城市中度拥堵路、城市严重拥堵路、快速路、高速路、市郊路、乡镇路。
根据本公开的一个实施例，所述道路特征参数包括以下至少一个：平均车速、最高车速、速度标准差、平均加速度、最大加速度、最小加速度、加速度标准差、加速时间比例、减速时间比例、匀速时间比例、怠速时间比例、累计行驶里程。
根据本公开的一个实施例，针对所述等效因子序列中的每个等效因子，根据下式计算得到与该等效因子对应的工况下所述混合动力车辆的电池瞬时输出功率：
其中，H(u，SOC(t)，t)为根据等效燃油消耗最小策略建立得到的哈密顿函数，arg H(u,SOC(t),t)为所述动力电池在t时刻的瞬时输出功率，为所述混合动力车辆的发动机燃油消耗率，s(t)为t时刻的等效因子，SOC(t)为所述动力电池在t时刻的荷电状态，为荷电状态变化率，u为油耗。
根据本公开的一个实施例，所述根据所述动力电池的瞬时输出功率对所述混合动力车辆进行控制，包括：获取所述混合动力车辆在t时刻的瞬时需求功率；将所述t时刻的瞬时需求功率减去所述动力电池在t时刻的瞬时输出功率，得到所述混合动力车辆的发动机在t时刻的瞬时需求功率；根据所述动力电池在t时刻的瞬时输出功率和所述发动机在t时刻的瞬时输出功率，对所述动力电池和所述发动机进行控制。
为达到上述目的，本公开第二方面实施例提出了一种计算机可读存储介质，其上存储有计算机程序，计算机程序被处理器执行时，实现上述的混合动力车辆能量管理方法。
为达到上述目的，本公开第三方面实施例提出了一种电子设备，包括存储器、处理器和存储在存储器上并可在处理器上运行的计算机程序，计算机程序被处理器执行时，实现上述的混合动力车辆能量管理方法。
为达到上述目的，本公开第四方面实施例提出了一种混合动力车辆的能量管理装置，装置包括：获取模块，用于获取混合动力车辆的当前导航路线的工况类别序列和当前导航路线中各工况的里程，工况类别序列和各工况的里程是根据当前导航路线的道路特征参数得到的根据工况类别序列和各工况的里程获取目标荷电状态序列；根据工况类别序列和目标荷电状态序列得到等效因子序列；根据等效因子序列得到混合动力车辆的动力电池在各时刻的瞬时输出功率；能量管理模块，用于根据动力电池的瞬时输出功率对混合动力车辆进行控制。
为达到上述目的，本公开第五方面实施例提出了一种混合动力车辆，包括上述的混合动力车辆的能量管理装置。
本公开实施例的混合动力车辆及其能量管理方法、装置及介质、电子设备，通过获取混合动力车辆当前导航路线的道路特征参数，利用预先训练好的神经网络模型根据道路特征参数得到当前导航路线的工况类别序列和各工况的里程，从而可以实现对混合动力车辆的在线控制，而且，在获取工况类别序列和各工况的里程之后，还需要根据工况类别序列和各工况的里程获取目标荷电状态序列，从而根据工况类别序列和目标荷电状态序列得到等效因子序列，从而实现等效因子根据路段的工况类别发生变化，保证燃油经济性。而且，由于是首先获取当前导航路线，基于当前导航路线得到等效因子序列，根据等效因子序列得到混合动力车辆的动力电池在各时刻的瞬时输出功率，根据动力电池的瞬时输出功率对混合动力车辆进行控制，实现在当前导航路线上基于该等效因子序列进行能量管理，使得能量管理不受导航地图数据输出的频率影响，所需计算量小，能够在工况变化的情况下更好地保证燃油经济性。
本公开附加的方面和优点将在下面的描述中部分给出，部分将从下面的描述中变得明显，或通过本公开的实践了解到。
附图说明
图1是本公开一个实施例的混合动力车辆的能量管理方法的流程图；
图2是本公开一个示例的荷电状态变化路径示意图；
图3是本公开一个示例的神经网络模型的结构示意图；
图4是本公开实施例的混合动力车辆的能量管理装置的结构框图；
图5是本公开实施例的混合动力车辆的结构框图。
具体实施方式
下面参考附图描述本公开实施例的混合动力车辆及其能量管理方法、装置及介质、电子设备，其中自始至终相同或类似的标号表示相同或类似的元件或具有相同或类似功能的元件。参考附图描述的实施例是示例性的，不能理解为对本公开的限制。
图1是本公开一个实施例的混合动力车辆的能量管理方法的流程图。
如图1所示，混合动力车辆的能量管理方法，包括：
S11，获取混合动力车辆当前导航路线的道路特征参数。
S12，利用预先训练好的神经网络模型根据道路特征参数得到当前导航路线的工况类别序列和各工况的里程。
S13，根据工况类别序列和各工况的里程获取目标荷电状态序列，并利用等效燃油消耗最小策略根据工况类别序列、各工况的里程和目标荷电状态序列得到等效因子序列。
S14，根据等效因子序列得到混合动力车辆的动力电池在各时刻的瞬时输出功率，根据动力电池的瞬时输出功率对混合动力车辆进行控制。
具体地，预先搭建整车仿真模型，通过整车仿真模型对混合动力车辆在各工况下进行能量消耗等相关性能离线计算，基于性能计算结果将混合动力车辆的工况划分为多个工况类别。而且，还预先对神经网络模型进行训练，使得神经网络模型可以在输入导航地图的道路特征参数之后，输出工况类别序列。
在混合动力车辆需要行驶时，根据导航地图获取混合动力车辆当前导航路线的道路特征参数，并将通过导航地图获取的道路特征参数输入预先训练好的神经网络模型，得到当前导航路线中各路段的工况类别和里程，同时，还对工况类别进行排序，得到工况类别序列。根据工况类别序列和各工况的里程得到与当前导航路线对应的目标荷电状态序列，从而利用等效燃油消耗最小策略得到等效因子序列，根据等效因子序列对混合动力车辆进行能量管理。
由此，通过获取混合动力车辆当前导航路线的道路特征参数，利用预先训练好的神经 网络模型根据道路特征参数得到当前导航路线的工况类别序列和各工况的里程，从而可以实现对混合动力车辆的在线控制，而且，在获取工况类别序列和各工况的里程之后，还需要根据工况类别序列和各工况的里程获取目标荷电状态序列，从而根据工况类别序列、各工况的里程和目标荷电状态序列得到等效因子序列，从而实现等效因子根据路段的工况类别发生变化，保证燃油经济性。而且，由于是首先获取当前导航路线，基于当前导航路线得到等效因子序列，在当前导航路线上基于该等效因子序列进行能量管理，使得能量管理不受导航地图数据输出的频率影响，所需计算量小，能够在工况变化的情况下更好地保证燃油经济性。
在本公开的一些实施例中，上述根据工况类别序列和各工况的里程获取目标荷电状态序列，包括：获取动力电池当前的实际荷电状态；根据实际荷电状态、工况类别序列和各工况的里程，确定混合动力车辆在每个工况结束时的荷电状态变化范围；根据每个工况结束时的荷电状态变化范围，得到目标荷电状态序列。
具体地，在接收到工况类别序列和各工况的里程后，可以将工况类别序列和各工况的里程发送至第三方平台得到目标荷电状态序列，也可直接本地计算得到目标荷电状态序列。
为了计算得到目标荷电状态序列，可以根据混合动力车辆当前的实际荷电状态、工况类别序列中首个工况对应的道路特征数据和首个工况的里程，确定首个工况结束时的第一荷电状态和第二荷电状态，根据第一荷电状态和第二荷电状态得到首个工况结束时的荷电状态变化范围，其中，道路特征数据包括坡度数据和限速数据，第一荷电状态为混合动力车辆以发电模式在首个工况结束时的荷电状态，第二荷电状态为混合动力车辆以纯电模式在首个工况结束时的荷电状态；针对工况类别序列中除首个工况之外的每个工况，根据工况的前一个工况对应的工况结束时荷电状态变化范围、工况对应的道路特征数据和工况的里程，确定混合动力车辆在工况结束时的荷电状态变化范围。
在该实施例中，可分别在每个工况对应的荷电状态变化范围中选取一个目标荷电状态；根据各个目标荷电状态得到目标荷电状态序列。
在一些示例中，参见图2，首个工况为AB段对应的工况1，混合动力车辆处于A点的实际荷电状态为F，以混合动力车辆采用发电模式即在A点至B点的工况1下混合动力车辆全部使用燃料，电池处于充电状态，来确定B点的第一荷电状态为G，即工况1的荷电状态上限值为G，以混合动力车辆采用纯电模式即在A点至B点的工况1下混合动力车辆全部电，电池处于放电状态，来确定B点的第二荷电状态为I，即工况1的荷电状态下限值为I，由此，可以确定工况1下的荷电状态变化范围为[I,G]。假设工况1下的实际荷电状态的F为70％，B点的荷电状态上限值G为75％，荷电状态下限值I为65％，则工况1下 荷电状态变化范围为[65％,75％]。
然后，根据工况2对应的工况数据和工况2的前一个工况即工况1对应的荷电状态变化范围，来确定混合动力车辆在工况2运行下的荷电状态变化范围。首先，以工况1的荷电状态上限值为G作为工况2的实际荷电状态，混合动力车辆采用发电模式即工况2下混合动力车辆全部使用燃料，电池处于充电状态，来确定C点的第三荷电状态为J，即工况2的荷电状态上限值为J。然后，以工况1的荷电状态下限值为I作为工况2的实际荷电状态，以混合动力车辆采用纯电模式即在B点至C点的工况2下混合动力车辆全部电，电池处于放电状态，来确定C点的第四荷电状态为L，即工况2的荷电状态下限值为L，由此，确定工况2下荷电状态变化范围为[L，J]。
继续参照如图2所示，A点的实际荷电状态为F，假设在荷电状态变化范围[I,G]内选取点H作为目标荷电状态值，则F-H为一条工况1的荷电状态变化路径。
AB段工况1的目标荷电状态值除上述处于荷电状态变化范围内的H点外，还包括荷电状态上限值G和荷电状态下限值I，此时，AB段工况1可形成F-G、F-H、F-I三个荷电状态变化路径。BC段工况2的实际荷电状态可为AB段的目标荷电状态值，即包括G、H、I，BC段的预设参考荷电状态包括J、K、L，此时，BC段可形成G-J、G-K、G-L、H-J、H-K、H-L、I-J、I-K和I-L九个荷电状态变化路径。以此类推，CD段工况3可形成十五个荷电状态变化路径，DE段工况4可形成三十个荷电状态变化路径。基于上述内容可知，混合动力车辆当前行驶道路AE工况下可形成3×9×15×30＝12150组荷电状态变化路径。
需要说明的是，每个工况的目标荷电状态值的数量越多，所产生的荷电状态变化路径的数量也就越多，则工况下的能耗最小的荷电状态变化路径的确定精度也就越高，同时所达到的混合动力车辆的能量管理效果也就越好。
通过计算得到不同荷电状态变化路径的油耗和电耗，对比选取油耗和电耗最小的荷电状态变化路径作为最优荷电状态变化路径，并将最优荷电状态变化路径作为该工况的目标荷电状态变化路径，以控制混合动力车辆进行能量管理。以AB段的工况1为例，分别获取荷电状态变化路径F-G、F-H和F-I的油耗和电耗，并计算得到荷电状态变化路径F-I的油耗和电耗最小，因此以荷电状态变化路径F-I作为AB段的目标荷电状态变化路径，I为工况1结束时的最优荷电状态。以此类推，确定BC段的目标荷电状态变化路径为I-J，CD段的目标荷电状态变化路径为J-P，DE段的目标荷电状态变化路径为P-U，则预行驶道路AE下的目标荷电状态变化路径为F-I-J-P-U。
由此，可以实现根据工况类别序列和各工况的里程获取目标荷电状态序列，而且，该过程可在第三方平台中进行，可以减少车辆的计算量，降低车辆的计算时间，能够在工况变化的情况下保证燃油经济性。
在本公开的一些实施例中，上述神经网络模型的训练过程包括：获取混合动力车辆在当前导航路线上的历史行驶参数，并根据历史行驶参数确定历史道路特征参数，以及对历史道路特征参数进行聚类处理，得到多个工况类别；基于历史道路特征参数和工况类别，构建训练数据集；构建神经网络模型，并利用训练数据集对神经网络模型进行训练。其中，神经网络模型包括输入层、隐藏层和输出层，输入层用于输入降维处理后的道路特征参数，输出层用于输出工况类别。
具体地，首先构建如图3所示的神经网络模型，包括输入层、隐藏层和输出等，输入层输入的x1、x2、…xn为道路特征参数，而在图3所示的具体示例中，n为4，该神经网络模型支持维度为4的道路特征参数，输出层输出的y为工况类别。
为了对神经网络模型进行训练，首先获取混合动力车辆在当前导航路线上的历史行驶参数，根据历史行驶参数确定历史道路特征参数。
而且，还采用聚类分析算法进行工况分类，即采用欧式距离法将历史道路特征参数分为不同的工况类别，从而根据历史道路特征参数和对应的工况类别对神经网络模型进行训练。
可选地，在对神经网络模型进行训练之前，还可利用主成分分析法对历史道路特征参数进行降维处理，以利用降维处理后的历史道路特征参数对神经网络模型进行训练。而且，神经网络模型还根据降维处理后的道路特征参数得到当前导航路线的工况类别序列和各工况的里程。
在本公开的一些实施例中，工况类别包括：普通城市路、城市轻度拥堵路、城市中度拥堵路、城市严重拥堵路、快速路、高速路、市郊路、乡镇路。
在本公开的一些实施例中，道路特征参数包括以下至少一者：平均车速、最高车速、速度标准差、平均加速度、最大加速度、最小加速度、加速度标准差、加速时间比例、减速时间比例、匀速时间比例、怠速时间比例、累计行驶里程。
在本公开的一些实施例中，根据等效因子序列得到混合动力车辆的动力电池在各时刻的瞬时输出功率，包括：针对等效因子序列中的每个等效因子，根据下式计算得到与该等效因子对应的混合动力车辆的电池瞬时输出功率：
其中，H(u，SOC(t)，t)为根据等效燃油消耗最小策略建立得到的哈密顿函数，arg H(u,SOC(t),t)为混合动力车辆的动力电池在t时刻的瞬时输出功率，为混合动力车辆的发动机燃油消耗率，s(t)为t时刻的等效因子，SOC(t)为动力电池在t时刻的荷电状态，为荷电状态变化率，u为油耗。
具体地，建立哈密顿函数，利用等效燃油消耗最小策略原理，计算各分类工况并构建工况类别、目标荷电状态和等效因子的关系。求解哈密顿函数可建立当前工况类别下的电池目标荷电状态与等效因子的关系，从而确定当前工况类别各目标荷电状态下的最优等效因子。在求得最优等效因子后，即可得到与该最优等效因子对应的混合动力车辆的电池瞬时输出功率，从而根据动力电池的瞬时输出功率对混合动力车辆进行控制，比如说，在获取动力电池在t时刻的瞬时输出功率之后，可以获取混合动力车辆在t时刻的瞬时需求功率，将t时刻的瞬时需求功率减去动力电池在t时刻的瞬时输出功率，得到混合动力车辆的发动机在t时刻的瞬时需求功率，根据动力电池在t时刻的瞬时输出功率和发动机在t时刻的瞬时输出功率，对动力电池和发动机进行控制。
由此，可以实现基于实时优化的等效因子得到输出功率，实现了全导航路线的全时域最优能量管理。
综上，本公开实施例的混合动力车辆的能量管理方法，通过获取混合动力车辆当前导航路线的道路特征参数，利用预先训练好的神经网络模型根据道路特征参数得到当前导航路线的工况类别序列和各工况的里程，从而可以实现对混合动力车辆的在线控制，而且，在获取工况类别序列和各工况的里程之后，还需要根据工况类别序列和各工况的里程获取目标荷电状态序列，从而根据工况类别序列、各工况的里程和目标荷电状态序列得到等效因子序列，从而实现等效因子根据路段的工况类别发生变化，保证燃油经济性。而且，由于是首先获取当前导航路线，基于当前导航路线得到等效因子序列，在当前导航路线上基于该等效因子序列进行能量管理，使得能量管理不受导航地图数据输出的频率影响，所需计算量小，能够在工况变化的情况下更好地保证燃油经济性。
基于上述实施例的混合动力车辆的能量管理方法，本公开提出了一种计算机可读存储介质。
在本公开实施例中，上存储有计算机程序，计算机程序被处理器执行时实现上述的混合动力车辆的能量管理方法。
本公开实施例的计算机可读存储介质，通过实现上述的混合动力车辆能量管理方法，可以通过获取混合动力车辆当前导航路线的道路特征参数，利用预先训练好的神经网络模型根据道路特征参数得到当前导航路线的工况类别序列和各工况的里程，从而可以实现对混合动力车辆的在线控制，而且，在获取工况类别序列和各工况的里程之后，还需要根据工况类别序列和各工况的里程获取目标荷电状态序列，从而根据工况类别序列、各工况的里程和目标荷电状态序列得到等效因子序列，从而实现等效因子根据路段的工况类别发生变化，保证燃油经济性。而且，由于是首先获取当前导航路线，基于当前导航路线得到等效因子序列，在当前导航路线上基于该等效因子序列进行能量管理，使得能量管理不受导 航地图数据输出的频率影响，所需计算量小，能够在工况变化的情况下更好地保证燃油经济性。
基于上述实施例的混合动力车辆的能量管理方法，本公开提出了一种电子设备。
在本公开实施例中，电子设备包括存储器、处理器和存储在存储器上并可在处理器上运行的计算机程序，计算机程序被处理器执行时，实现上述的混合动力车辆能量管理方法。
本公开实施例的电子设备，通过实现上述的混合动力车辆能量管理方法，可以通过获取混合动力车辆当前导航路线的道路特征参数，利用预先训练好的神经网络模型根据道路特征参数得到当前导航路线的工况类别序列和各工况的里程，从而可以实现对混合动力车辆的在线控制，而且，在获取工况类别序列和各工况的里程之后，还需要根据工况类别序列和各工况的里程获取目标荷电状态序列，从而根据工况类别序列、各工况的里程和目标荷电状态序列得到等效因子序列，从而实现等效因子根据路段的工况类别发生变化，保证燃油经济性。而且，由于是首先获取当前导航路线，基于当前导航路线得到等效因子序列，在当前导航路线上基于该等效因子序列进行能量管理，使得能量管理不受导航地图数据输出的频率影响，所需计算量小，能够在工况变化的情况下更好地保证燃油经济性。
图4是本公开实施例的混合动力车辆的能量管理装置的结构框图。
如图4所示，混合动力车辆的能量管理装置100包括：获取模块101、能量管理模块102。
具体地，获取模块101，用于获取混合动力车辆当前导航路线的道路特征参数，以及，利用预先训练好的神经网络模型根据道路特征参数得到当前导航路线的工况类别序列和各工况的里程，以及，根据工况类别序列和各工况的里程获取目标荷电状态序列，并利用等效燃油消耗最小策略根据工况类别序列、各工况的里程和目标荷电状态序列得到等效因子序列，以及，根据等效因子序列得到混合动力车辆的动力电池在各时刻的瞬时输出功率；能量管理模块102，用于根据动力电池的瞬时输出功率对混合动力车辆进行控制。
需要说明的是，本公开实施例的混合动力车辆的能量管理装置的其他具体实施方式，可以参见上述的混合动力车辆的能量管理方法。
本公开实施例的混合动力车辆的能量管理装置，通过获取混合动力车辆当前导航路线的道路特征参数，利用预先训练好的神经网络模型根据道路特征参数得到当前导航路线的工况类别序列和各工况的里程，从而可以实现对混合动力车辆的在线控制，而且，在获取工况类别序列和各工况的里程之后，还需要根据工况类别序列和各工况的里程获取目标荷电状态序列，从而根据工况类别序列、各工况的里程和目标荷电状态序列得到等效因子序列，从而实现等效因子根据路段的工况类别发生变化，保证燃油经济性。而且，由于是首先获取当前导航路线，基于当前导航路线得到等效因子序列，在当前导航路线上基于该等 效因子序列进行能量管理，使得能量管理不受导航地图数据输出的频率影响，所需计算量小，能够在工况变化的情况下更好地保证燃油经济性。
基于上述实施例的混合动力车辆的能量管理方法装置，本公开提出了一种混合动力车辆。
图5是本公开实施例的混合动力车辆的结构框图。
如图5所示，混合动力车辆10包括上述混合动力车辆的能量管理装置100。
本公开实施例的混合动力车辆，通过上述混合动力车辆的能量管理装置，可以通过获取混合动力车辆当前导航路线的道路特征参数，利用预先训练好的神经网络模型根据道路特征参数得到当前导航路线的工况类别序列和各工况的里程，从而可以实现对混合动力车辆的在线控制，而且，在获取工况类别序列和各工况的里程之后，还需要根据工况类别序列和各工况的里程获取目标荷电状态序列，从而根据工况类别序列、各工况的里程和目标荷电状态序列得到等效因子序列，从而实现等效因子根据路段的工况类别发生变化，保证燃油经济性。而且，由于是首先获取当前导航路线，基于当前导航路线得到等效因子序列，在当前导航路线上基于该等效因子序列进行能量管理，使得能量管理不受导航地图数据输出的频率影响，所需计算量小，能够在工况变化的情况下更好地保证燃油经济性。
需要说明的是，在流程图中表示或在此以其他方式描述的逻辑和/或步骤，可以被认为是用于实现逻辑功能的可执行指令的定序列表，可以具体实现在任何计算机可读介质中，以供指令执行***、装置或设备(如基于计算机的***、包括处理器的***或其他可以从指令执行***、装置或设备取指令并执行指令的***)使用，或结合这些指令执行***、装置或设备而使用。就本说明书而言，“计算机可读介质”可以是任何可以包含、存储、通信、传播或传输程序以供指令执行***、装置或设备或结合这些指令执行***、装置或设备而使用的装置。计算机可读介质的更具体的示例(非穷尽性列表)包括以下：具有一个或多个布线的电连接部(电子装置)，便携式计算机盘盒(磁装置)，随机存取存储器(RAM)，只读存储器(ROM)，可擦除可编辑只读存储器(EPROM或闪速存储器)，光纤装置，以及便携式光盘只读存储器(CDROM)。另外，计算机可读介质甚至可以是可在其上打印所述程序的纸或其他合适的介质，因为可以例如通过对纸或其他介质进行光学扫描，接着进行编辑、解译或必要时以其他合适方式进行处理来以电子方式获得所述程序，然后将其存储在计算机存储器中。
应当理解，本公开的各部分可以用硬件、软件、固件或它们的组合来实现。在上述实施方式中，多个步骤或方法可以用存储在存储器中且由合适的指令执行***执行的软件或固件来实现。如果用硬件来实现，和在另一实施方式中一样，可用本领域公知的下列技术中的任一项或他们的组合来实现：具有用于对数据信号实现逻辑功能的逻辑门电路的离散 逻辑电路，具有合适的组合逻辑门电路的专用集成电路，可编程门阵列(PGA)，现场可编程门阵列(FPGA)等。
在本说明书的描述中，参考术语“一个实施例”、“一些实施例”、“示例”、“具体示例”、或“一些示例”等的描述意指结合该实施例或示例描述的具体特征、结构、材料或者特点包含于本公开的至少一个实施例或示例中。在本说明书中，对上述术语的示意性表述不一定指的是相同的实施例或示例。而且，描述的具体特征、结构、材料或者特点可以在任何的一个或多个实施例或示例中以合适的方式结合。
在本说明书的描述中，术语“中心”、“纵向”、“横向”、“长度”、“宽度”、“厚度”、“上”、“下”、“前”、“后”、“左”、“右”、“竖直”、“水平”、“顶”、“底”“内”、“外”、“顺时针”、“逆时针”、“轴向”、“径向”、“周向”等指示的方位或位置关系为基于附图所示的方位或位置关系，不是指示或暗示所指的装置或元件必须具有特定的方位、以特定的方位构造和操作，不能理解为对本公开的限制。
此外，术语“第一”、“第二”仅用于描述目的，而不能理解为指示或暗示相对重要性或者隐含指明所指示的技术特征的数量。由此，限定有“第一”、“第二”的特征可以明示或者隐含地包括至少一个该特征。在本公开的描述中，“多个”的含义是至少两个，例如两个，三个等，除非另有明确具体的限定。
在本说明书的描述中，除非另有说明，术语“安装”、“相连”、“连接”、“固定”等术语应做广义理解，例如，可以是固定连接，也可以是可拆卸连接，或成一体；可以是机械连接，也可以是电连接；可以是直接相连，也可以通过中间媒介间接相连，可以是两个元件内部的连通或两个元件的相互作用关系，除非另有明确的限定。对于本领域的普通技术人员而言，可以根据具体情况理解上述术语在本公开中的具体含义。
在本公开中，除非另有明确的规定和限定，第一特征在第二特征“上”或“下”可以是第一和第二特征直接接触，或第一和第二特征通过中间媒介间接接触。而且，第一特征在第二特征“之上”、“上方”和“上面”可是第一特征在第二特征正上方或斜上方，或仅仅表示第一特征水平高度高于第二特征。第一特征在第二特征“之下”、“下方”和“下面”可以是第一特征在第二特征正下方或斜下方，或仅仅表示第一特征水平高度小于第二特征。
尽管上面已经示出和描述了本公开的实施例，可以理解的是，上述实施例是示例性的，不能理解为对本公开的限制，本领域的普通技术人员在本公开的范围内可以对上述实施例进行变化、修改、替换和变型。

Claims (15)

  1. 一种混合动力车辆的能量管理方法，其中，所述方法包括：
    获取所述混合动力车辆的当前导航路线的工况类别序列和所述当前导航路线中各工况的里程，所述工况类别序列和所述各工况的里程是根据所述当前导航路线的道路特征参数得到的；
    根据所述工况类别序列和所述各工况的里程获取目标荷电状态序列；
    根据所述工况类别序列和所述目标荷电状态序列得到等效因子序列；
    根据所述等效因子序列得到所述混合动力车辆的动力电池在各时刻的瞬时输出功率；
    根据所述动力电池的瞬时输出功率对所述混合动力车辆进行控制。
  2. 根据权利要求1所述的混合动力车辆的能量管理方法，其中，所述根据所述工况类别序列和所述各工况的里程获取目标荷电状态序列，包括：
    获取所述动力电池在所述当前导航路线起点的实际荷电状态；
    根据所述实际荷电状态、所述工况类别序列和所述各工况的里程，确定所述混合动力车辆在所述各工况结束时的荷电状态变化范围；
    根据所述各工况结束时的荷电状态变化范围，得到所述目标荷电状态序列。
  3. 根据权利要求2所述的混合动力车辆的能量管理方法，其中，所述当前导航路线的首个工况结束时的荷电状态变化范围是根据所述实际荷电状态、所述首个工况的道路特征数据和所述首个工况的里程得到的；所述当前导航路线的非首个工况结束时的荷电状态变化范围是根据所述非首个工况的前一个工况结束时的荷电状态变化范围、所述非首个工况的道路特征数据和所述非首个工况的里程得到的。
  4. 根据权利要求3所述的混合动力车辆的能量管理方法，其中，所述道路特征数据包括坡度数据和限速数据。
  5. 根据权利要求3或4所述的混合动力车辆的能量管理方法，其中，所述工况的荷电状态变化范围的上限值是所述混合动力车辆以发电模式在所述工况运行结束时的荷电状态，所述工况的荷电状态变化范围的下限值是所述混合动力车辆以纯电模式在所述工况运行结束时的荷电状态。
  6. 根据权利要求3-5中任一项所述的混合动力车辆的能量管理方法，其中，所述目标荷电状态序列是分别在每个工况对应的荷电状态变化范围中选取一个目标荷电状态，根据选取的各个目标荷电状态得到的。
  7. 根据权利要求1-6中任一项所述的混合动力车辆的能量管理方法，其中，所述工况类别序列和所述各工况的里程是利用训练好的神经网络模型，根据所述当前导航路线的道 路特征参数得到的，所述神经网络模型的训练过程包括：
    获取所述混合动力车辆在当前导航路线上的历史行驶参数，并根据所述历史行驶参数确定历史道路特征参数，以及对历史道路特征参数进行聚类处理，得到多个工况类别；
    基于所述历史道路特征参数和所述工况类别，构建训练数据集；
    构建神经网络模型，并利用所述训练数据集对所述神经网络模型进行训练。
  8. 根据权利要求7所述的混合动力车辆的能量管理方法，其中，所述工况类别包括：普通城市路、城市轻度拥堵路、城市中度拥堵路、城市严重拥堵路、快速路、高速路、市郊路、乡镇路。
  9. 根据权利要求1-8中任一项所述的混合动力车辆的能量管理方法，其中，所述道路特征参数包括以下至少一个：平均车速、最高车速、速度标准差、平均加速度、最大加速度、最小加速度、加速度标准差、加速时间比例、减速时间比例、匀速时间比例、怠速时间比例、累计行驶里程。
  10. 根据权利要求1-9中任一项所述的混合动力车辆的能量管理方法，其中，
    针对所述等效因子序列中的每个等效因子，根据下式计算得到与该等效因子对应的工况下所述混合动力车辆的电池瞬时输出功率：
    其中，H(u，SOC(t)，t)为根据等效燃油消耗最小策略建立得到的哈密顿函数，arg H(u,SOC(t),t)为所述动力电池在t时刻的瞬时输出功率，为所述混合动力车辆的发动机燃油消耗率，s(t)为t时刻的等效因子，SOC(t)为所述动力电池在t时刻的荷电状态，为荷电状态变化率，u为油耗。
  11. 根据权利要求1-10中任一项所述的混合动力车辆的能量管理方法，其中，所述根据所述动力电池的瞬时输出功率对所述混合动力车辆进行控制，包括：
    获取所述混合动力车辆在t时刻的瞬时需求功率；
    将所述t时刻的瞬时需求功率减去所述动力电池在t时刻的瞬时输出功率，得到所述混合动力车辆的发动机在t时刻的瞬时需求功率；
    根据所述动力电池在t时刻的瞬时输出功率和所述发动机在t时刻的瞬时输出功率，对所述动力电池和所述发动机进行控制。
  12. 一种计算机可读存储介质，其中，其上存储有计算机程序，所述计算机程序被处理器执行时实现如权利要求1-11中任一项所述的混合动力车辆的能量管理方法。
  13. 一种电子设备，其中，包括存储器、处理器和存储在存储器上并可在处理器上运行的计算机程序，所述计算机程序被所述处理器执行时，实现如权利要求1-11中任一项所 述的混合动力车辆的能量管理方法。
  14. 一种混合动力车辆的能量管理装置，其中，所述装置包括：
    获取模块，用于获取所述混合动力车辆的当前导航路线的工况类别序列和所述当前导航路线中各工况的里程，所述工况类别序列和所述各工况的里程是根据所述当前导航路线的道路特征参数得到的根据所述工况类别序列和所述各工况的里程获取目标荷电状态序列；根据所述工况类别序列和所述目标荷电状态序列得到等效因子序列；根据所述等效因子序列得到所述混合动力车辆的动力电池在各时刻的瞬时输出功率；
    能量管理模块，用于根据所述动力电池的瞬时输出功率对所述混合动力车辆进行控制。
  15. 一种混合动力车辆，其中，包括如权利要求14所述的混合动力车辆的能量管理装置。
PCT/CN2023/108978 2022-09-29 2023-07-24 混合动力车辆及其能量管理方法、装置及介质、电子设备 WO2024066702A1 (zh)

Applications Claiming Priority (2)

Application Number Priority Date Filing Date Title
CN202211203935 2022-09-29
CN202211203935.0 2022-09-29

Publications (1)

Publication Number Publication Date
WO2024066702A1 true WO2024066702A1 (zh) 2024-04-04

Family

ID=90384059

Family Applications (1)

Application Number Title Priority Date Filing Date
PCT/CN2023/108978 WO2024066702A1 (zh) 2022-09-29 2023-07-24 混合动力车辆及其能量管理方法、装置及介质、电子设备

Country Status (2)

Country Link
CN (1) CN117774934A (zh)
WO (1) WO2024066702A1 (zh)

Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN109895760A (zh) * 2019-04-03 2019-06-18 大连理工大学 基于soc参考轨迹的插电式混合动力汽车能量管理方法
US20190202436A1 (en) * 2017-12-31 2019-07-04 Hyliion Inc. Electric drive controller adaptation to through-the-road (ttr) coupled primary engine and/or operating conditions
WO2019133957A1 (en) * 2017-12-31 2019-07-04 Hyliion Inc. Supplemental electric drive with primary engine recognition for electric drive controller adaptation and crowdsourced hybrid system customization
CN110929920A (zh) * 2019-11-05 2020-03-27 中车戚墅堰机车有限公司 一种基于工况识别的混合动力列车能量管理方法
CN110936949A (zh) * 2019-12-12 2020-03-31 湖北文理学院 基于行驶工况的能量控制方法、设备、存储介质及装置
CN111591279A (zh) * 2020-04-20 2020-08-28 湖南大学 一种插电式混合动力汽车电池电量轨迹规划方法及***
CN112009455A (zh) * 2019-05-28 2020-12-01 北汽福田汽车股份有限公司 混合动力车辆的能量管理方法、装置及车辆
CN113650601A (zh) * 2021-09-23 2021-11-16 上海交通大学 车辆控制方法及***、服务器、车辆
CN114179777A (zh) * 2021-12-24 2022-03-15 河南科技大学 一种基于工况识别的变等效因子混合动力汽车能量管理方法

Patent Citations (9)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20190202436A1 (en) * 2017-12-31 2019-07-04 Hyliion Inc. Electric drive controller adaptation to through-the-road (ttr) coupled primary engine and/or operating conditions
WO2019133957A1 (en) * 2017-12-31 2019-07-04 Hyliion Inc. Supplemental electric drive with primary engine recognition for electric drive controller adaptation and crowdsourced hybrid system customization
CN109895760A (zh) * 2019-04-03 2019-06-18 大连理工大学 基于soc参考轨迹的插电式混合动力汽车能量管理方法
CN112009455A (zh) * 2019-05-28 2020-12-01 北汽福田汽车股份有限公司 混合动力车辆的能量管理方法、装置及车辆
CN110929920A (zh) * 2019-11-05 2020-03-27 中车戚墅堰机车有限公司 一种基于工况识别的混合动力列车能量管理方法
CN110936949A (zh) * 2019-12-12 2020-03-31 湖北文理学院 基于行驶工况的能量控制方法、设备、存储介质及装置
CN111591279A (zh) * 2020-04-20 2020-08-28 湖南大学 一种插电式混合动力汽车电池电量轨迹规划方法及***
CN113650601A (zh) * 2021-09-23 2021-11-16 上海交通大学 车辆控制方法及***、服务器、车辆
CN114179777A (zh) * 2021-12-24 2022-03-15 河南科技大学 一种基于工况识别的变等效因子混合动力汽车能量管理方法

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
"Master Thesis", 15 August 2020, JILIN UNIVERSITY, CN, article LI, JINKE: "Research on Travel Condition Prediction and Adaptive Control Strategy of Plug-in Hybrid Electric Vehicles", pages: 1 - 127, XP009553428 *

Also Published As

Publication number Publication date
CN117774934A (zh) 2024-03-29

Similar Documents

Publication Publication Date Title
CN110936949B (zh) 基于行驶工况的能量控制方法、设备、存储介质及装置
Tran et al. Thorough state-of-the-art analysis of electric and hybrid vehicle powertrains: Topologies and integrated energy management strategies
Zhou et al. A novel energy management strategy of hybrid electric vehicle via an improved TD3 deep reinforcement learning
CN110775065B (zh) 一种基于工况识别的混合动力汽车电池寿命预测方法
Zhang et al. Role of terrain preview in energy management of hybrid electric vehicles
Tian et al. Data-driven hierarchical control for online energy management of plug-in hybrid electric city bus
CN108528436B (zh) 一种内外层嵌套的ecms多目标双层优化方法
Wu et al. An optimized real-time energy management strategy for the power-split hybrid electric vehicles
US10668824B2 (en) Method for calculating a setpoint for managing the fuel and electricity consumption of a hybrid motor vehicle
Xu et al. Real-time reinforcement learning optimized energy management for a 48V mild hybrid electric vehicle
CN105882648B (zh) 一种基于模糊逻辑算法的混合动力***能量管理方法
Taherzadeh et al. A comprehensive study of the parameters impacting the fuel economy of plug-in hybrid electric vehicles
CN112668799A (zh) 基于行驶大数据的phev的智能能量管理方法和存储介质
Zheng et al. A deep reinforcement learning-based energy management strategy for fuel cell hybrid buses
CN113479186A (zh) 一种混合动力汽车能量管理策略优化方法
Fries et al. Operational strategy of hybrid heavy-duty trucks by utilizing a genetic algorithm to optimize the fuel economy multiobjective criteria
KR20180055054A (ko) 하이브리드 차량의 최적 운전점 결정 방법
Peng et al. Ecological Driving Framework of Hybrid Electric Vehicle Based on Heterogeneous Multi-Agent Deep Reinforcement Learning
Chen et al. Data-driven transferred energy management strategy for hybrid electric vehicles via deep reinforcement learning
JP2022500307A (ja) ハイブリッド自動車による燃料および電流の消費に対する管理設定値を計算する方法
Halima et al. Energy management of parallel hybrid electric vehicle based on fuzzy logic control strategies
WO2024066702A1 (zh) 混合动力车辆及其能量管理方法、装置及介质、电子设备
Gong et al. Energy Management for a DM-i Plug-in Hybrid Electric Vehicle via Continuous-Discrete Reinforcement Learning
Li et al. Predictive multi-objective operation strategy considering battery cycle aging for hybrid electric vehicles
Feng et al. Fuzzy energy management strategy for hybrid electric vehicles on battery state-of-charge estimation by particle filter

Legal Events

Date Code Title Description
121 Ep: the epo has been informed by wipo that ep was designated in this application

Ref document number: 23869948

Country of ref document: EP

Kind code of ref document: A1