A hev global energy management method considering engine start-stop and shift interval

Abstract

The application discloses a HEV global energy management method considering engine start-stop and gear shifting interval, comprising the following steps: selecting state variables and control variables, dividing vehicle driving conditions into N running stages, and discretizing power system state space grid; taking minimum fuel consumption as an objective function, and establishing a dynamic programming reverse solving equation; starting from the N-1 section, determining all possible decisions of each state space grid point and the state space grid point of the next stage corresponding to each decision, and screening out a first feasible decision from all possible decisions; screening out an optimal decision from the first feasible decision according to the objective function, updating the time sequence array of the current state space grid point, and repeating the process to complete the calculation of all state space grid points in the current stage; repeating the above steps to calculate the previous stage until the initial stage; starting from the state space grid point corresponding to the initial state, searching for an optimal decision sequence in a forward direction until the terminal stage.

Description

Technical Field

[0001] This invention belongs to the field of global optimization energy management technology for hybrid electric vehicles, and specifically relates to a global energy management method for HEVs that takes into account engine start-stop and shift intervals. Background Technology

[0002] Hybrid electric vehicles (HEVs) have two power sources: an engine and an electric motor. They can operate in different modes according to the power demands under different driving conditions, combining the advantages of both gasoline-powered and pure electric vehicles. By rationally adjusting the operating points of the engine and electric motor through energy management strategies, the efficiency of the powertrain components can be improved, thereby significantly improving the overall fuel economy of the vehicle. Therefore, energy management strategies are the core technology of HEVs.

[0003] Dynamic programming is a commonly used global optimization method that solves complex problems by transforming them into multi-stage subproblems. Therefore, dynamic programming-based energy management strategies have been widely used to solve for the optimal control sequence of hybrid electric vehicles. However, existing dynamic programming-based energy management strategies have some problems: 1. When solving dynamic programming problems, an objective function needs to be defined. Existing methods typically use fuel consumption as the objective function. Simultaneously, to ensure vehicle smoothness, frequent gear shifts need to be avoided. Existing methods usually define a penalty function for gear shifting operations as part of the objective function, thereby reducing gear shifting operations. However, this method only indirectly restricts gear shifting operations by minimizing fuel consumption and cannot strictly limit frequent gear shifts at the level of gear shift intervals. 2. When the gear is used as a state variable, the selection of the gear will be affected by vehicle speed due to the operating speed limitations of the powertrain components. Existing methods rarely filter gears based on vehicle speed, resulting in an excessive number of grid points in the state space, thus causing excessive computational load for DP solutions. 3. For hybrid vehicles, when the engine output power is 0, the engine will idle or shut off. Existing energy management strategies typically do not consider the engine's start-stop state. Furthermore, to ensure vehicle smoothness, it is also necessary to limit frequent engine start-stop operations. Existing methods reduce frequent engine start-stop by defining a penalty function for engine starting operations as part of the objective function. However, this method also indirectly limits engine start-stop operations by minimizing fuel consumption and cannot completely eliminate frequent engine start-stop at the level of engine start-stop intervals. Summary of the Invention

[0004] The purpose of this invention is to provide a global energy management method for HEVs that takes into account engine start-stop and shift intervals, which can effectively avoid frequent gear shifts of the transmission and frequent start-stop of the engine from the perspective of interval time.

[0005] The technical solution provided by this invention is as follows:

[0006] A global energy management method for HEVs that takes into account engine start-stop and shift intervals includes the following steps:

[0007] Battery SOC, transmission gear position, and engine start-stop state eng are taken as the state variables of the power system, and the change in battery SOC ΔSOC, transmission shift decision shift, and engine start-stop decision start are taken as the control variables of the power system. The vehicle driving conditions are divided into N operating stages, and the state space grid of the power system is discretized. State transition equations for battery SOC, transmission gear position, and engine start-stop state eng are established respectively.

[0008] Delete the state space grid points containing infeasible transmission gears at each operational stage;

[0009] With the goal of minimizing fuel consumption, a dynamic programming inverse solution equation is established. Starting from the (N-1)th stage, all possible decisions for each state space grid point and the state space grid point for the next stage corresponding to the possible decisions are determined, and the first feasible decision is selected from the possible decisions.

[0010] Among them, the state space grid points corresponding to the first feasible decision in the next stage simultaneously satisfy the transmission shift interval time constraint and the engine start-stop interval time constraint.

[0011] Based on the objective function, the optimal decision for transferring the current state space grid point to the next stage state space grid point is selected from the first feasible decisions;

[0012] The time series array of the current state space grid points is updated, and the next state space grid point of the current stage is calculated. This process is repeated until all state space grid points of the current stage are calculated. The calculation of the previous stage is repeated until the initial stage, and the optimal strategy for transferring all state space grid points of each running stage to the state space grid points of the next stage is obtained.

[0013] In the initial stage of vehicle operation, the battery SOC, transmission gear position, and engine start / stop status are obtained to determine the state space grid point corresponding to the current vehicle power system. Based on the result of the inverse solution of dynamic programming, the optimal decision sequence is searched in the forward direction from the state space grid point until the termination stage, and the vehicle's working state is controlled according to the optimal decision sequence.

[0014] Preferably, the HEV global energy management method that considers engine start-stop and shift intervals further includes:

[0015] In extreme driving conditions, when the first feasible decision does not exist, a second feasible decision is selected from the possible decisions, and the optimal strategy is obtained based on the second feasible decision.

[0016] Among them, the state space grid points corresponding to the second feasible decision in the next stage satisfy the engine start-stop interval time constraint and the transmission shift timing is minimized.

[0017] Preferably, the state transition equation for the battery's state of charge (SOC) is:

[0018] SOC(k+1) = SOC(k) + ΔSOC(k);

[0019] Where SOC(k+1) is the battery SOC in the (k+1)th operating stage, SOC(k) is the battery SOC in the kth operating stage, and ΔSOC(k) is the change in battery SOC in the kth operating stage, which is an integer multiple of the SOC calculation accuracy.

[0020] Preferably, the state transition equation for the gear is:

[0021]

[0022] Where gear(k+1) is the gear of the transmission in the (k+1)th operating stage, gear(k) is the gear of the transmission in the kth operating stage, and shift(k) is the transmission shift decision in the kth operating stage, including downshifting, holding, and upshifting, which are represented by -n, 0, and n respectively, where n is the number of gears changed for upshifting or downshifting, and n = 1, 2, 3, 4, 5.

[0023] Preferably, the state transition equation for the engine state is:

[0024] eng(k+1) = start(k);

[0025] Among them, the engine start-stop state eng includes off and on, represented by 0 and 1 respectively; eng(k+1) is the engine start-stop state in the (k+1)th operating stage; start is the engine start-stop decision, including off the engine and start the engine, represented by 0 and 1 respectively, and start(k) is the engine start-stop decision in the kth operating stage.

[0026] Preferably, before solving the equations in reverse dynamic programming, the process also includes: initializing the time series array of all state space grid points to (t 1-0 , t 2-0 ), t 1-0 =0,t 2-0 =0;

[0027] Among them, t 1-0t represents the initial value of the transmission shift timing. 2-0 This indicates the initial value of the engine start-stop sequence.

[0028] Preferably, the optimal decision is calculated using a dynamic programming algorithm based on the objective function;

[0029] The expression for the optimal decision is:

[0030] u * (j)=argmin[fuel(x(i),u(j))+J k+1 (x(j))];

[0031] Where fuel(x(i),u(j)) represents the fuel consumption generated when the state space grid point x(i) in the k-th operating stage is transferred to the (k+1)-th operating stage under the action of decision u(j); J k+1 (x(j)) represents the cumulative fuel consumption from the state space grid point x(j) in the (k+1)th running stage to the final stage.

[0032] Preferably, updating the timing array of the current state space grid points includes updating the transmission shift timing and the engine start-stop timing.

[0033] Preferably, based on the optimal decision u * (j) and its corresponding state space grid point x in the next stage * The time sequence array (t1, t2) of (j) updates the time sequence array of the current state space grid points; where: the update method for the transmission shift timing is:

[0034] When u * When the transmission shift decision in (j) is 0, if x * If the gear shift timing of (j) is not 0, then the gear shift timing of the current state space grid point is updated to t1-1; if x * If the gear shift timing of (j) is 0, then the gear shift timing of the current state space grid point is still 0;

[0035] When u * When the transmission shift decision in (j) is not 0, the transmission shift timing of the desired state space grid point is updated to S1-1;

[0036] Where t1 is x * (j) transmission shift timing, t2 is x * (j) is the engine start-stop sequence; S1 is the constraint value of the set transmission shift interval time.

[0037] Preferably, the method for updating the engine start-stop sequence is as follows:

[0038] When u * When the engine start-stop decision in (j) is 0, if x * If the engine start-stop timing of (j) is not 0, then the engine start-stop timing of the current state space grid point is updated to t2-1; if x * If the engine start-stop timing of (j) is 0, then the engine start-stop timing of the current state space grid point is still 0.

[0039] When u * When the engine start-stop decision in (j) is not 0, the engine start-stop timing of the current state space grid point is updated to S2-1;

[0040] S2 is the constraint value for the engine start-stop interval time.

[0041] The beneficial effects of this invention are:

[0042] This invention provides a global energy management method for HEVs that considers engine start-stop and gear shift intervals. The method treats engine start-stop state and transmission gear as state variables, and deletes some state space grid points according to driving conditions to reduce the computational load during solution. In the inverse solution stage of the dynamic programming algorithm, all possible decisions for each state space grid point in each stage are screened according to the preset transmission shift interval and engine start-stop interval, so that each state space grid point obtains an optimal decision that minimizes the objective function. At the same time, it effectively avoids frequent transmission shifts and frequent engine start-stops from the perspective of interval time. Taking into account the vehicle's power performance, smoothness and fuel economy, the method adjusts the transmission shift interval time constraint under extreme conditions while prioritizing the vehicle's power performance requirements. The resulting optimal decision sequence for the vehicle can avoid frequent transmission shifts as much as possible. Attached Figure Description

[0043] Figure 1 This is a flowchart of the HEV global energy management method that takes into account engine start-stop and shift intervals, as described in this invention.

[0044] Figure 2 This is a structural diagram of the powertrain system of the HEV to which this invention applies.

[0045] Figure 3 This is a state-space grid structure diagram of the powertrain system of the HEV described in this invention.

[0046] Figure 4 This is a schematic diagram illustrating the implementation of the first feasible decision screening method for state space grid points provided in an embodiment of the present invention.

[0047] Figure 5This is a schematic diagram illustrating the implementation of the second feasible decision-making screening method for state space grid points under extreme operating conditions provided in this embodiment of the invention.

[0048] Figure 6 This is a flowchart illustrating the reverse solution of the HEV global energy management method that considers engine start-stop and shift intervals as described in this invention.

[0049] Figure 7 This is a flowchart of the first feasible decision screening method for state space grid points according to the present invention.

[0050] Figure 8 This is a flowchart of the second feasible decision-making screening method for state-space grid points under extreme conditions as described in this invention.

[0051] Figure 9 This is a flowchart of the method for updating the temporal array of state space grid points according to the present invention.

[0052] Figure 10 This is a schematic diagram of the forward search process of the HEV global energy management method that takes into account engine start-stop and shift intervals as described in this invention. Detailed Implementation

[0053] The present invention will now be described in further detail with reference to the accompanying drawings, so that those skilled in the art can implement it based on the description.

[0054] like Figure 1 As shown, the present invention provides a global energy management method for HEV that takes into account engine start-stop and shift intervals, and the implementation process is as follows.

[0055] I. Establish HEV powertrain component models, including engine, motor, and battery models. Establish a longitudinal dynamics model of the entire vehicle and calculate the required power output. The engine model is established using steady-state fuel consumption map data; the engine fuel consumption rate is calculated by interpolating engine speed and torque.

[0056] be = f1(n) e ,T e )

[0057] Where be is the engine fuel consumption rate, and n e T is the engine speed. e This refers to the engine torque.

[0058] The motor model is built from static efficiency map data, and the motor efficiency is calculated by interpolating the motor speed and motor torque.

[0059] η m =f2(n m ,T m )

[0060] Where, η m n represents the motor efficiency. m T represents the motor speed. m T is the motor torque. m A value greater than 0 indicates that the motor is in drive mode; T m A value less than 0 indicates that the motor is in generator mode.

[0061] like Figure 2 As shown, based on the structural characteristics of hybrid electric vehicle powertrains, there is a power balance between the engine and the electric motor:

[0062]

[0063] Among them, P e P represents engine power. m P represents the motor power. re η represents the power required for the entire vehicle; η represents the efficiency of the transmission system.

[0064] The power battery model adopts the commonly used Rint model, which equates the battery to an equivalent circuit consisting of a voltage source and a resistor, and establishes a discrete model:

[0065]

[0066]

[0067] U oc =f3(SOC)

[0068] Where SOC is the battery charge state; I b Q represents the battery current. b Battery capacity; P b Battery power; U oc R is the battery open-circuit voltage, a function of the battery's state of charge (SOC); b η is the battery's internal resistance. c The inverter efficiency is represented by `sign`, where `T` is the sign function. m When >0, sign(T) m ) = 1; T m When <0, sign(T) m ) = -1;

[0069] Establish a longitudinal dynamics model for the entire vehicle and calculate the required power for the entire vehicle:

[0070]

[0071] Among them, P re (k) represents the vehicle's power requirement at time k; m represents the vehicle's curb weight; g represents gravitational acceleration; f represents the rolling resistance coefficient; Cd ρ is the air drag coefficient; A is the frontal area; ρ is the air density; θ(k) is the road slope angle at time k; δ is the vehicle rotational mass conversion factor; v(k) is the vehicle speed at time k. Let k be the acceleration at time k; r be the radius of the wheel.

[0072] 2. Divide the operating phases according to the driving conditions, and set the phase step size to 1 second. Select battery SOC, transmission gear position, and engine start / stop state as power system state variables. Select the change in battery SOC ΔSOC, transmission shift decision (shift), and engine start / stop decision (start) as power system control variables.

[0073] The state variable can be represented as: The control variable can be represented as:

[0074] Among them, the engine start / stop state eng includes two states: off and on, represented by 0 and 1 respectively; the transmission shift decision shift includes downshift, hold, and upshift, represented by -n, 0, and n respectively, where n = 1, 2, 3, 4, 5, and the highest gear of the transmission is 6th gear; the engine start / stop decision start includes turning off the engine and starting the engine, represented by 0 and 1 respectively.

[0075] Based on the accuracy of the battery SOC calculation, the battery SOC is discretized as: {SOC} min :Δsoc:SOC max}, where SOC min For the minimum allowed SOC, SOC max Δsoc represents the maximum allowable SOC, and Δsoc represents the accuracy of the battery SOC solution.

[0076] The gear positions of the transmission are discretized as: {1,2,3,4,5,6}, where the highest gear is 6.

[0077] The structure of the state-space grid of the discretized dynamical system is as follows: Figure 3 As shown.

[0078] Establish the state transition equations for battery SOC, transmission gear position, and engine state.

[0079] The state transition equation for battery SOC is:

[0080] SOC(k+1) = SOC(k) + ΔSOC(k)

[0081] Where SOC(k+1) is the battery SOC in the (k+1)th operating stage, SOC(k) is the battery SOC in the kth operating stage, and ΔSOC(k) is the change in battery SOC in the kth operating stage, which is an integer multiple of the solution accuracy Δsoc.

[0082] The state transition equation for the gearbox is:

[0083]

[0084] The state transition equation for the engine state is:

[0085] eng(k+1) = start(k)

[0086] That is, the start-stop state of the engine in stage k+1 is directly controlled by the engine start-stop decision in stage k.

[0087] Since the operating speed range of the engine and motor is limited, the selection of transmission gears will be affected by the vehicle speed. Therefore, it is necessary to calculate the selectable transmission gears of the hybrid vehicle at different vehicle speeds, delete the state space grid points where infeasible transmission gears are located at each stage, and reduce the computational load of DP solution.

[0088]

[0089]

[0090]

[0091] Among them, i g (k) represents the allowable range of the transmission ratio at time k; v(k) is the vehicle speed at time k and is not zero; n max n is the maximum speed of the transmission input shaft. min This refers to the minimum speed of the transmission input shaft; n m_max n is the maximum speed of the motor. e_max This refers to the engine's maximum speed; n m_min This is the minimum speed of the motor; n e_min This is the engine's lowest speed.

[0092] 3. Define a time sequence array T(t1,t2) and initialize the time sequence arrays of all state space grid points for all stages to (0,0). Set the transmission shift interval time constraint value to S1, indicating that the time interval between two consecutive transmission shift operations is not less than S1 seconds; set the engine start-stop interval time constraint value to S2, indicating that the time interval between two consecutive engine start-stop operations is not less than S2 seconds. In the time sequence array T(t1,t2), t1 represents the transmission shift sequence, reflecting the time interval between the stage where the state space grid point is located and the most recent transmission shift operation. Specifically, if the transmission shift sequence t1 of a certain state space grid point is not 0, it means that when starting the forward search for the optimal decision sequence from that state space grid point, the transmission shift will occur after (S1-1)-t1 seconds in the stage where the state space grid point is located. t2 represents the engine start-stop sequence, reflecting the time interval between the stage where the state space grid point is located and the most recent engine start-stop operation. Specifically, if the engine start-stop timing t2 at a certain state space grid point is not 0, it means that when the forward search for the optimal decision sequence starts from that state space grid point, the engine start-stop operation will be performed after (S2-1)-t2 seconds in the stage where that state space grid point is located.

[0093] IV. Defining fuel consumption as the objective function, the objective function of the state space grid point x(i) in stage K is the cumulative minimum fuel consumption from stage K to the final stage, expressed as:

[0094]

[0095] fuel(x(i),u(j))=f1(n e ,T e )+f4(x(i),u(j))

[0096]

[0097] in, Let x(i) represent the cumulative minimum fuel consumption from grid point x(i) in the state space of stage K to the final stage, and fuel(x(i),u(j)) represents the fuel consumption generated when the state space grid point x(i) in stage K is transferred to stage K+1 under the action of decision u(j); f4 represents the fuel consumption generated due to engine starting. When the engine changes from the off state to the on state under the action of the starting decision, the starting fuel consumption c will be generated; x(j) represents the state space grid point x(i) in stage K is transferred to stage K+1 under the action of decision u(j); This represents the cumulative minimum fuel consumption from grid point x(j) in the state space of stage K+1 to the final stage.

[0098] like Figure 6As shown, a DP inverse solution equation is established. Starting from stage K, all possible decisions [u(j1), u(j2), ... u(jk)] for the first state space grid point in the current stage are calculated. n [x(j1), x(j2), ... x(j)] and its state space grid points [x(j1), x(j2), ... x(j)] transitioning to stage K+1. n Each decision must satisfy the constraints of the power system components:

[0099] P e_min ≤P e (x(i),u(j))≤P e_max

[0100] P m_min ≤P m (x(i),u(j))≤P m_max

[0101] P b_min ≤P b (x(i),u(j))≤P b_max

[0102] Secondly, based on the state space grid point x(j) transferred to the next stage k The time series array T) k (t k1 ,t k2 ) for decision u(j k The filtering is performed where 1 ≤ k ≤ n. The implementation diagram and flowchart of the filtering method are shown below. Figure 4 and Figure 7 As shown, the specific process is as follows:

[0103] (1) When u(j k When both the transmission shift decision (shift) and the engine start-stop decision (start) in the given configuration are 0, the decision is reserved as the first feasible decision. Figure 4 The decision u(j1) of grid point x(i) in the state space.

[0104] (2) When u(j k When the shift decision in the transmission is not 0, if x(j) k ) transmission shift timing t k1 If it is not 0, then the decision is excluded, such as... Figure 4 The decision u(j3) of grid point x(i) in the state space; if x(j k ) transmission shift timing t k1 If the value is 0, then further determine x(j) k Engine start-stop timing t k2 Does it meet the engine start-stop interval time constraint?

[0105] (3) When u(j k When the engine start / stop decision start in x(j) is not 0, if x(j) k Engine start-stop timing t k2 If it is not 0, then the decision is excluded, such as... Figure 4 The decision u(j2) of grid point x(i) in the state space; if x(j k Engine start-stop timing t k2 If the value is 0, the decision is reserved as the first feasible decision, such as... Figure 4 The decision u(j4) of grid point x(i) in the state space.

[0106] Decision u(j) k After the filtering is completed, repeat the above steps to filter the next decision for the state space grid point until all possible decisions for the state space grid point have been filtered.

[0107] Fifth, under extreme driving conditions, the transmission gears will inevitably change frequently. To meet the transmission shift interval constraints, all possible decisions for the state space grid points will be eliminated, resulting in no first feasible decision. Figure 5 In the K-stage state space grid point x(i), the time sequence array of its transition to the K+1-stage state space grid point under all possible decisions cannot simultaneously satisfy the transmission shift interval time constraint and the engine start-stop interval time constraint. To ensure that the dynamic programming algorithm can solve the problem normally, it is necessary to consider all possible decisions [u(j1), u(j2), ... u(j...]] of the desired state space grid point. n The state space grid points that are re-selected under various possible decisions are still denoted as [x(j1), x(j2), ... x(jj)]. n Based on the transitioned state space grid point x(j) k The time series array T) k (t k1 ,t k2 ) for decision u(j k The filtering is performed where 1 ≤ k ≤ n. The implementation diagram and flowchart of the filtering method are shown below. Figure 5 and Figure 8 As shown, the specific process is as follows:

[0108] (1) When u(j k When the engine start / stop decision start in x(j) is not 0, if x(j) k Engine start-stop timing t k2 If it is not 0, then the decision is excluded, such as... Figure 5 The decisions u(j2) and u(j5) for grid point x(i) in the state space; if the engine start-stop timing tk2 If the value is 0, the decision is temporarily retained, and the next decision for the desired state space grid point is selected, such as... Figure 5 The decision u(j3) of grid point x(i) in the state space.

[0109] (2) After all possible decisions for the desired state space grid points have been filtered, the remaining feasible decisions are denoted as [u(j1), u(j2), ... u(j...]. m )],like Figure 5 The feasible decisions for state space grid point x(i) are [u(j1), u(j3), u(j4)]. The state space grid points that the desired state space grid point transitions to the next stage under the action of each feasible decision are denoted as [x(j1), x(j2), ... x(j4)]. m The corresponding time series array is denoted as [T1(t)]. 11 ,t 12 ),T2(t 21 ,t 22 ),...T m (t m1 ,t m2 )], where 1≤m≤n, such as Figure 5 The timing array [T1(t)] 11 ,t 12 ),T3(t 31 ,t 32 ),T4(t 41 ,t 42 )).

[0110] (3) Retain the timing array with the smallest transmission shift timing and its corresponding state space grid point, denoted as [T1(t)]. 11 ,t 12 ),T2(t 21 ,t 22 ),...T l (t l1 ,t l2 [x(j1),x(j2),...x(j)] and [x(j)] l )], each state space grid point [x(j1),x(j2),...x(j l The corresponding feasible decisions are [u(j1), u(j2), ... u(j)]. l [)] is the second feasible decision for the desired state space grid points, where 1≤l≤m, as shown in the example. Figure 5 The time series array T1(t) 11 ,t 12 ) and T4(t 41 ,t 42 ) transmission shift timing t 11 and t41 Since they are the smallest and equal, the decision [u(j1),u(j4)] is the second feasible decision for grid point x(i) in the state space.

[0111] VI. After selecting feasible decisions for the desired state-space grid points, the optimal decision u for the desired state-space grid points is calculated using a dynamic programming algorithm based on the objective function. * (j), the desired state space grid points are in u * Under the influence of (j), the state space grid point x of the next stage is transferred. * (j).

[0112] u * (j)=argmin[fuel(x(i),u(j))+J k+1 (x(j))]

[0113] VII. According to u * (j) and x * The time series array T(t1,t2) of (j) is the time series array T of the desired state space grid points. i (t i1 ,t i2 Update the flowchart as follows: Figure 9 As shown, the specific process is as follows:

[0114] (1) When u * When the transmission shift decision in (j) is 0, if x * If the gear shift timing of (j) is not 0, then the gear shift timing t of the required state space grid point is... i1 Updated to t1-1; if x * If the gear shift timing of (j) is also 0, then the gear shift timing t of the required state space grid point is... i1 Still 0.

[0115] That is, the transmission shift timing is updated to: max[t1-1,0]

[0116] (2) When u * When the transmission shift decision in (j) is not zero, the required transmission shift timing t at the state space grid point is obtained. i1 Updated to S1-1.

[0117] Engine start-stop timing t i2 The update method and transmission shift timing t i1 similar:

[0118] (1) When u * When the engine start-stop decision in (j) is 0, if x *If the engine start-stop timing of (j) is not 0, then the engine start-stop timing t of the required state space grid point is... i2 Updated to t2-1; if x * If the engine start-stop timing of (j) is 0, then the engine start-stop timing t of the required state space grid point is... i2 Still 0.

[0119] The engine start-stop timing sequence is updated to: max[t2-1,0]

[0120] (2) When u * When the engine start-stop decision in (j) is not 0, then the engine start-stop timing t of the required state space grid point is... i2 Updated to S2-1.

[0121] After the temporal array of the desired state space grid points is updated, repeat the above steps to calculate the next state space grid point for stage K. After all state space grid points for stage K are calculated, repeat the above steps to calculate each state space grid point for stage K-1, until the initial stage, where K = N-1, N-2...1.

[0122] 8. Based on the initial stage of the vehicle's battery SOC, transmission gear position, and engine start / stop state eng, determine the state space grid point x(1) corresponding to the current vehicle power system. Based on the result of the inverse solution of dynamic programming, start from this state space grid point and search for the optimal decision sequence U in the forward direction. * Until the termination stage, such as Figure 10 As shown, the vehicle's operating state is controlled according to the optimal decision sequence.

[0123] At each stage, the state space grid point x(k) is in its optimal decision u * Under the influence of (k), the state transition point is x(k+1) in the state space of the next stage. The state transition equation is:

[0124] x(k+1)=f(x(k),u * (k))

[0125] U * =[u * (1),u * (2),...,u * (N-1)]

[0126] Where x(k+1) is the state space grid point in stage k+1; x(k) is the state space grid point in stage k; u * x(k) is the optimal decision corresponding to x(k), where k = 1, 2, ..., N-1.

[0127] Although embodiments of the present invention have been disclosed above, they are not limited to the applications listed in the specification and embodiments. They can be applied to various fields suitable for the present invention. For those skilled in the art, other modifications can be easily made. Therefore, without departing from the general concept defined by the claims and their equivalents, the present invention is not limited to the specific details and illustrations shown and described herein.

Claims

1. A global energy management method for HEVs that considers engine start-stop and shift intervals, characterized in that, Includes the following steps: With batteries Gearbox and engine start / stop status As the state variables of the power system, with the battery Change Transmission shift decision Engine start-stop decision The control variables for the powertrain system are defined; the vehicle's operating conditions are divided into N operating stages, and the state space grid of the powertrain system is discretized; and battery states are established for each stage. Gearbox and engine start / stop status The state transition equation; Delete the state space grid points containing infeasible transmission gears at each operational stage; With the goal of minimizing fuel consumption, a dynamic programming inverse solution equation is established. Starting from the (N-1)th stage, all possible decisions for each state space grid point and the state space grid point for the next stage corresponding to the possible decisions are determined, and the first feasible decision is selected from the possible decisions. Among them, the state space grid points corresponding to the first feasible decision in the next stage simultaneously satisfy the transmission shift interval time constraint and the engine start-stop interval time constraint. Based on the objective function, the optimal decision for transferring the current state space grid point to the next stage state space grid point is selected from the first feasible decisions; The time series array of the current state space grid points is updated, and the next state space grid point of the current stage is calculated. This process is repeated until all state space grid points of the current stage are calculated. The calculation of the previous stage is repeated until the initial stage, and the optimal strategy for transferring all state space grid points of each running stage to the state space grid points of the next stage is obtained. Acquire the battery during the initial stage of vehicle operation. Gearbox and engine start / stop status The state space grid point corresponding to the current vehicle power system is determined. Based on the result of the inverse solution of dynamic programming, the optimal decision sequence is searched in the forward direction starting from the state space grid point until the termination stage. The working state of the vehicle is controlled according to the optimal decision sequence. In extreme driving conditions, when the first feasible decision does not exist, a second feasible decision is selected from the possible decisions, and the optimal strategy is obtained based on the second feasible decision. Among them, the state space grid points corresponding to the second feasible decision in the next stage satisfy the engine start-stop interval time constraint and the transmission shift timing is minimized; The battery The state transition equation is: ； in, For the first Battery during operation , For the first Battery during operation , For the first Battery during operation The change is The precision must be an integer multiple of the required precision. The gear of the transmission The state transition equation is: ； in, For the first Transmission gears during operation For the first The gears of the transmission during operation; For the first The transmission shifting decisions during operation include downshifting, holding, and upshifting, which are represented by -n, 0, and n, respectively, where n is the number of gears changed during upshifting or downshifting, n=1,2,3,4,5; The engine status State transition equation: ； Among them, engine start-stop state This includes being off and on, represented by 0 and 1 respectively; For the first Engine start-stop status during operation; Engine start-stop decisions, including shutting down the engine and starting the engine, are represented by 0 and 1 respectively. For the first Engine start-stop decision during operation.

2. The HEV global energy management method considering engine start-stop and shift intervals according to claim 1, characterized in that, Before solving the equations in reverse dynamic programming, the process also includes: initializing the time series arrays of all state space grid points to... ; in, This represents the initial value of the transmission shift timing. This indicates the initial value of the engine start-stop sequence.

3. The HEV global energy management method considering engine start-stop and shift intervals according to claim 2, characterized in that, Based on the objective function, the optimal decision is calculated using a dynamic programming algorithm; The expression for the optimal decision is: ； in, Indicates the first State space grid points during operation In decision-making Transferred to the first under the action Fuel consumption during operation; Indicates the first State space grid points during operation Cumulative fuel consumption up to the end of the phase.

4. The HEV global energy management method considering engine start-stop and shift intervals according to claim 3, characterized in that, Updating the timing array of the current state space grid points includes updating the transmission shift timing and the engine start-stop timing.

5. The HEV global energy management method considering engine start-stop and shift intervals according to claim 4, characterized in that, According to the optimal decision and its corresponding state space grid points in the next stage time series array Update the timing array of the current state space grid points; where: the update method for the transmission shift timing is as follows: when When the transmission shift decision is 0, if If the transmission shift timing is not 0, then the transmission shift timing of the current state space grid point is updated to... ;like If the gear shift timing of the transmission is 0, then the gear shift timing of the transmission at the current state space grid point is still 0. when When the transmission shift decision in the process is not zero, the transmission shift timing of the desired state space grid point is updated as follows: ; in, for The transmission shift timing, for Engine start-stop sequence; This is the constraint value for the set transmission shift interval time.

6. The HEV global energy management method considering engine start-stop and shift intervals according to claim 5, characterized in that, The method for updating the engine start-stop sequence is as follows: when When the engine start-stop decision is 0, if If the engine start-stop timing is not 0, then the engine start-stop timing of the current state space grid point is updated to... ;like If the engine start-stop timing is 0, then the engine start-stop timing of the current state space grid point is still 0; when If the engine start / stop decision value in the current state space grid is not zero, then the engine start / stop timing sequence of the current state space grid point is updated as follows: ; in, This is a constraint value for the engine start-stop interval.

Images