Vehicle adaptive cruise method based on vehicle speed planning

Abstract

The invention discloses a vehicle adaptive cruise method based on vehicle speed planning, which comprises the following steps that: a vehicle receives a satellite signal to determine a current position, and obtains the altitude and speed limit data of a front road; an n-order Bezier function is constructed, and a vehicle speed curve of the vehicle speed relative to the road position is generated through the Bezier function; an evaluation function of the vehicle speed curve is constructed, and constraints are established; a wild horse optimization algorithm is improved, and Gaussian distribution is introduced to generate an initial population and an individual position updating strategy based on Levy flight so as to improve the global search capability and the convergence speed; inputting the Bezier function and the evaluation function into an improved wild horse optimization algorithm, and finally outputting an optimal cruise speed curve; based on the optimal cruise speed curve, the conversion relation between fuel energy and electric energy is determined by using an ECMS global optimization strategy and introducing an equivalent factor, a global optimal control strategy can be obtained by minimizing a Hamilton value, then an optimal control method of an engine and a motor at each moment is obtained, and the engine is controlled.

Description

Technical Field

[0001] The invention relates to a vehicle adaptive cruising method based on vehicle speed planning. Background Art

[0002] In modern driving experience, adaptive cruise control has become an important technology to relieve driving fatigue. It can automatically control the vehicle to drive at the speed preset by the driver, significantly reducing fatigue during long-distance travel. However, in the face of complex speed limits on actual roads, such as the changing maximum speed limit and interval speed limit, the driver needs to constantly cancel and reset the cruise speed, which not only reduces the driving pleasure, but also may affect the smoothness and safety of driving. In this context, the importance of speed planning technology has become increasingly prominent. It is the key to realizing advanced adaptive cruise function and is directly related to driving comfort and energy efficiency. Through accurate speed planning, not only can the driving trajectory be optimized while ensuring safety, unnecessary acceleration and deceleration can be reduced, but also fuel economy can be effectively improved, environmental protection and energy saving can be achieved. Therefore, researching and developing a speed planning method based on road information is of great significance to improving adaptive cruise technology and achieving a smarter and more efficient assisted driving experience. Summary of the invention

[0003] The present invention proposes a vehicle adaptive cruise control method based on vehicle speed planning, which achieves energy-saving and efficient adaptive cruise control by acquiring detailed road information, updating slope data, establishing an energy-saving driving cost function, an optimal vehicle speed curve, and adjusting the acceleration by a controller, thereby improving driving comfort and fuel economy. The present invention adopts the following technical solutions:

[0004] A vehicle adaptive cruise control method based on vehicle speed planning comprises the following steps:

[0005] S1: The driver selects the vehicle navigation itinerary according to the destination;

[0006] S2: The vehicle receives satellite signals to determine the current location and obtains the altitude and speed limit data of the road ahead;

[0007] S3: constructing an n-order Bessel function, and generating a vehicle speed curve with respect to the road position through the Bessel function;

[0008] S4: construct the evaluation function of the vehicle speed curve and establish constraints;

[0009] S5: Improve the Mustang optimization algorithm by introducing Gaussian distribution to generate the initial population and the individual position update strategy based on Levy flight to improve its global search capability and convergence speed; then input the Bessel function and evaluation function into the improved Mustang optimization algorithm, and finally output the optimal cruising speed curve;

[0010] S6: Based on the optimal cruising speed curve, the ECMS global optimization strategy is used to introduce an equivalent factor to determine the conversion relationship between fuel energy and electrical energy. The global optimal control strategy can be obtained by minimizing the Hamiltonian value, and then the optimal control method of the engine and motor at each moment is obtained, and the engine is controlled.

[0011] Furthermore, in S2, the road ahead is discretized at equal intervals, and the altitude data at each point is obtained using linear interpolation; for the road segments, the road slope on each segment is solved using the starting point height difference and the starting point interval.

[0012] Furthermore, in S3, the Bessel function is used to describe the relationship between the vehicle speed and the road distance, specifically:

[0013] Define a set of control points (x i ,y i ), where x i Represents the road distance, y i represents the vehicle speed at this distance. Assuming there are n+1 control points (P0, P1, ..., P n ), construct an n-order Bezier curve, where n ranges from [10,30], and the generated speed curve has the following parametric equation:

[0014]

[0015] Among them, t is a parameter, ranging between [0,1]; is the binomial coefficient, calculated as P i =(x i ,y i ) is the i-th control point.

[0016] Furthermore, in S4, the evaluation function of the vehicle speed curve is:

[0017] L=Cost power +Cost Mobility +Cost comfort

[0018] Cost power is the energy consumption required for vehicle driving, Cost Mobility Cost is the penalty for a vehicle speed that deviates from the road speed limit. comfort It is driving comfort;

[0019] Cost power =w1·P req (t)

[0020] Cost Mobility =w2(x2(t)-v l) 2

[0021] Cost comfort =w3·u1(t) 2

[0022] Among them, w1, w2 and w3 are weight factors to ensure the balance between energy economy, power and driving comfort; v l is the speed limit on the test road;

[0023]

[0024] Where m is the mass of the vehicle, g is the acceleration of gravity, f is the rolling resistance coefficient, C D is the air resistance coefficient, ρ is the air density, A f is the frontal area; when P req When (t) is positive, it indicates the total power spent on vehicle state changes. req When (t) is negative, it describes the power charged into the battery.

[0025] Furthermore, the evaluation function has the following constraints:

[0026] Acceleration constraint: The maximum acceleration is set to u1(t),

[0027]

[0028] Among them, F t is the maximum driving force of the vehicle's powertrain. The acceleration range only applies to acceleration and deceleration during speed planning. The constraint is expressed as a set of allowable accelerations:

[0029]

[0030] Among them, u min is the minimum acceleration, u1(t) is the vehicle acceleration at time t, u max is the maximum acceleration, t0 is the initial time, and T is the total driving time of the vehicle;

[0031] Speed ​​limit:

[0032] Among them, v min is the minimum vehicle speed, v1(t) is the vehicle speed at time t, u max is the maximum vehicle speed, t0 is the initial time, and T is the total vehicle travel time.

[0033] Furthermore, in S5, the improvement of the Mustang optimization algorithm is as follows:

[0034] (51) Use Gaussian distribution to generate the initial population;

[0035] The population size is N, and each candidate solution is a vector x i , where i = 1, 2, ..., N. The candidate solution here represents the coordinates of the control points of the Bezier curve mentioned above. The initial solution is generated by setting the mean μ and standard deviation σ as the center of the search space. The standard deviation can be adjusted according to the range of the search space. The candidate solution is expressed as:

[0036] x i =μ+σ×randn()

[0037] Where μ is the center of the search space, calculated as σ is the standard deviation, set to Assuming that 99.7% of the data falls within the range, randn() is a function that generates standard normal distributed random numbers;

[0038] Since the Gaussian distribution can generate values ​​that are out of bounds, these values ​​need to be processed to ensure that they are within the search space:

[0039] X i =max(X min ,min(X i ,X max ))

[0040] (52) Determine the leader of the initial population based on fitness and decide whether to engage in grazing and mating behaviors;

[0041] When grazing, individuals tend to forage near the leader, and their positions are updated as follows:

[0042]

[0043] in, is the position of individual i in the jth dimension after update; Stallion j is the position of the stallion in the jth dimension; x ij is the position of individual i in the jth dimension; Z is the modulation factor; R is a random number whose value range is [-2,2];

[0044] Z=R2⊙IDX+R3(~IDX)

[0045] Among them, ⊙ represents the dot product operation; ~ represents binary inversion; R2 and R3 are random numbers in the range [0,1]; IDX is a binary mask that determines whether R2 or R3 is used;

[0046] IDX=(P==0)

[0047] Among them, P is a Boolean variable, indicating whether the random condition is met;

[0048] P=R1<TDR

[0049] Among them, R1 is a random number in the range of [0,1]; TDR is a threshold that determines whether a certain update strategy is used in the current iteration;

[0050]

[0051] Among them, ite is the current iteration number; ite max is the maximum number of iterations;

[0052] During mating, individuals search for mates and their positions are updated as follows:

[0053]

[0054] in, represents the average of two individual positions; represents the position of foal k in population p; represents the position of foal i in population q; represents the position of foal j in population z;

[0055] Individuals that do not mate or graze adopt the Lévy flight mode to wander to enhance global search capabilities. The position update formula of Lévy flight is:

[0056] x′ i (t) = x i (t)+l·Levy(λ)

[0057] Where: x i (t) represents the i-th solution of the t-th generation, l is the weight controlling the step length, defined as l = 0.01×(x i (t)-x b ), where x b is the current optimal solution; Levy(λ) represents the path that obeys the Levy distribution and satisfies Levy~u=t -λ , where 1<λ≤3;

[0058] (53) Calculate the fitness of the new population and determine the leader based on the fitness. The leader leads the population to the new habitat and competes for the owned land. If it fails, it will move to another habitat. The leader's position is updated as follows:

[0059]

[0060] Among them, WH is the location of the habitat, that is, the location of the global optimal individual; is the current position of the leader of population i; Stallioni is a random number in the range [0,1];

[0061] (54) Repeat steps (52) to (53) until the maximum number of iterations is reached, end the loop, output the optimal fitness and its corresponding control point value, and substitute the optimized control point into the parameter equation of the Bessel function to obtain the optimal cruising speed curve of the vehicle.

[0062] Further, S6 includes the following steps:

[0063] (61) The engine fuel consumption rate, torque external characteristics, engine speed range, motor consumption rate, torque external characteristics, and motor speed range are respectively expressed as b e =f1(n e ,T e ), T e_max =f2(n e ), n e_min <n e <n e_max , η m =f3(n m ,T m ), T m_min =f4(n m ), n e_min <n e <n e_max ;

[0064] (62) Determine the feasible region of motor torque [T m_lim_min ,T m_lim_max ] and divided into N equal parts [T m_lim_min ,…,T m_i ,…T m_lim_max ], then calculate the possible values ​​of the corresponding engine torque T e_i =T d -T m_i ;

[0065] (63) Calculate the corresponding engine power P ei 、Motor power P mi and engine fuel consumption m f_i =(b i ·P ei 0.001) / 3600;

[0066] (64) Calculate the corresponding equivalent fuel consumption m fequ_i =m f_i +s(t)·P mi / 42500, and solve for the minimum equivalent fuel consumption, and then solve for the minimum equivalent fuel consumption m that meets the SOC limit fequ_min =min{mfequ_1 ,m fequ_2 ,…,m fequ_i}, and obtain the corresponding optimized motor torque T m_opt , and then through T e_opt =T d -T m_opt Finding the optimized engine torque;

[0067] (65) The minimum engine shutdown time and the minimum continuous start time are set to reduce the problem of frequent engine starts and stops in a very short time when only the instantaneous equivalent fuel consumption is minimized.

[0068] The present invention has the following beneficial effects:

[0069] (1) The present invention improves the driving experience by automatically adjusting the vehicle speed to comply with the road speed limit, avoiding the risk of speeding and reducing the user's frequent adjustment of the cruise speed. The significant feature of this method is that its optimal vehicle speed is only related to the position and is decoupled from time, so it can effectively avoid vehicle speed tracking errors caused by traffic jams.

[0070] (2) The Mustang optimization algorithm is significantly improved by introducing Gaussian distribution to generate the initial population and combining it with the Levy flight strategy to optimize the global search capability of the Mustang algorithm. Gaussian distribution ensures the diversity of the initial population, and Levy flight enhances the algorithm's exploration ability in the search space and improves the algorithm's ability to jump out of the local optimum, making the overall optimization process more efficient. The improved Mustang algorithm performs well in terms of convergence speed. Experimental results show that compared with the traditional Mustang algorithm, the optimization algorithm of the present invention has a significant improvement in convergence speed and can reach the optimal fitness more quickly. Combined with the evaluation function of the present invention, the vehicle speed curve can be optimized more accurately to ensure the best balance between energy economy, power and driving comfort.

[0071] (3) By using the ECMS global optimization strategy and combining it with the optimized vehicle speed curve, this method provides the best control solution for the engine and motor at each moment, significantly reducing the engine's idling fuel consumption and frequent start-stop problems. The experimental results show that while maintaining a similar battery SOC, the strategy of the present invention reduces fuel consumption to 6.56L, compared to 7.17L of traditional cruise control, with an energy saving optimization of 8.51%. This significant energy-saving effect shows that this method has extremely high application value in the adaptive cruise system of hybrid vehicles, and can effectively improve the vehicle's economy and environmental performance.

[0072] (4) The present invention not only improves the intelligence level of the adaptive cruise control system of hybrid vehicles and enhances the driving experience, but also ensures the real-time response capability of the system through efficient calculation methods, providing important technical support for the development of future intelligent transportation. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] Figure 1 It is a flow chart of the present invention.

[0074] Figure 2 The vehicle speed curve generated by the multidimensional Bessel function.

[0075] Figure 3 Flowchart of improved Mustang optimization algorithm for integrating multiple strategies.

[0076] Figure 4 Comparison chart of the wild-horse algorithm and the wild-horse algorithm optimized for mixed strategies.

[0077] Figure 5 Optimized comparison chart for engine start-stop.

[0078] Figure 6 The figure is a comparison chart of SOC curves of different strategies.

[0079] Figure 7 The following is a comparison chart of fuel consumption for different strategies. DETAILED DESCRIPTION

[0080] The present invention will be further described below in conjunction with the accompanying drawings.

[0081] The present invention is aimed at a hybrid vehicle model, combined with Figure 1 , the present invention comprises the following steps:

[0082] S1: The driver selects a vehicle itinerary based on the target longitude and latitude coordinates.

[0083] S2: The vehicle receives satellite signals through a GPS receiver to obtain the longitude and latitude coordinates of its current location, determines the current location, and then obtains the altitude and speed limit data of the road ahead through a high-precision map;

[0084] S3: Improve the Mustang optimization algorithm and apply it to obtain the optimal cruising speed curve.

[0085] S4: Based on the optimal cruising speed curve, the ECMS global optimization strategy is used, combined with the engine start-stop control strategy, to obtain the optimal control method for the engine and motor at each moment;

[0086] The process of improving the Mustang optimization algorithm and applying it to obtain the optimal cruising speed curve is as follows:

[0087] 1) Construct an n-order Bessel function and generate a speed curve of the vehicle speed with respect to the road position through the Bessel function, such as Figure 2 .

[0088] 2) Construct an evaluation function for the vehicle speed curve, covering factors such as energy consumption, speed deviation penalty, and driving comfort, and establish constraints.

[0089] 3) The Mustang optimization algorithm is improved by introducing Gaussian distribution to generate the initial population and an individual position update strategy based on Levy flight to improve its global search capability and convergence speed.

[0090] 4) The Bessel function and the evaluation function are input into the improved Mustang optimization algorithm, and the optimal cruising speed curve is finally output.

[0091] In S2, the road ahead is discretized at equal intervals, and the elevation data at each point is obtained using linear interpolation; for road segments, the road slope α on each segment is solved using the starting point height difference and the starting point interval.

[0092] The principle algorithm of the improved wild horse optimization algorithm proposed by S3 that integrates multiple strategies includes the following:

[0093] Use Bezier curves to describe the relationship between vehicle speed and road distance.

[0094] Define a set of control points (x i ,y i ), where x i Represents the road distance, y i represents the vehicle speed at this distance. Assume there are n+1 control points (P0, P1, ..., P n ), then we can construct an n-order Bezier curve, where n ranges from [10,30], and the generated speed curve is as follows Figure 2 As shown, the parametric equation is:

[0095]

[0096] Among them, t is a parameter, ranging between [0,1], is the binomial coefficient, calculated as P i =(x i ,y i ) is the i-th control point.

[0097] The evaluation function of the speed curve is established as:

[0098] L=Cost power +Cost Mobility +Cost comfort

[0099] Cost power is the energy consumption required for vehicle driving, Cost MobilityDescribes the penalty when the vehicle speed deviates from the road speed limit, which is the squared difference between the desired speed profile and the speed limit. comfort Describes driving comfort. This part makes the vehicle accelerate and decelerate more smoothly, increasing driving comfort.

[0100] Cost power =w1·P req (t)

[0101] Cost Mobility =w2(x2(t)-v l ) 2

[0102] Cost comfort =w3·u1(t) 2

[0103] Among them, w1, w2 and w3 are weight factors to ensure the balance between energy economy, power and driving comfort; v l is the speed limit on the test road.

[0104]

[0105] Where m is the mass of the vehicle, g is the acceleration of gravity, f is the rolling resistance coefficient, C D is the air resistance coefficient, ρ is the air density, A f is the frontal area; when P req When (t) is positive, it indicates the total power spent on vehicle state changes. req When (t) is negative, it describes the power charged into the battery.

[0106] Establishing constraints and initial conditions, when solving the optimal vehicle speed curve, the above evaluation function has the following constraints.

[0107] Acceleration constraint: To ensure that all acceleration solutions are feasible while providing maximum driving power and braking conditions, the maximum acceleration is set to u1(t).

[0108]

[0109] Among them, F t is the maximum driving force of the vehicle's powertrain. This acceleration range is only applicable to acceleration and deceleration during vehicle speed planning. This constraint can be expressed as a set of allowable accelerations:

[0110]

[0111] Among them, u min is the minimum acceleration, u1(t) is the vehicle acceleration at time t, u maxis the maximum acceleration, t0 is the initial time, and T is the total driving time of the vehicle.

[0112] Speed ​​Constraint: The speed is designed to vary within a specified range, taking into account the feasibility and necessity of speed adjustment in different traffic situations. This constraint can be specified as:

[0113]

[0114] Among them, v min is the minimum vehicle speed, v1(t) is the vehicle speed at time t, u max is the maximum vehicle speed, t0 is the initial time, and T is the total vehicle travel time.

[0115] like Figure 3 ,The optimal vehicle speed solution based on the hybrid strategy optimization wild horse algorithm includes the following steps:

[0116] (1) Use Gaussian distribution to generate the initial population.

[0117] The population size is N, and each candidate solution is a vector x i , where i = 1, 2, ..., N. The candidate solution here represents the coordinates of the control points of the Bezier curve mentioned above. The initial solution is generated by setting the mean μ and standard deviation σ as the center of the search space. The standard deviation can be adjusted according to the range of the search space. The candidate solution is expressed as:

[0118] x i =μ+σ×randn()

[0119] Where μ is the center of the search space and can be calculated as σ is the standard deviation, set to randn() is a function that generates standard normally distributed random numbers, assuming that 99.7% of the data fall within the range.

[0120] Since the Gaussian distribution may generate values ​​that are out of bounds, these values ​​need to be processed to ensure that they are within the search space:

[0121] x i =max(x min ,min(x i ,x max ))

[0122] (2) Determine the leader of the initial population based on fitness and decide whether to engage in grazing and mating behaviors.

[0123] When grazing, individuals tend to forage near the leader. Their position is updated as follows:

[0124]

[0125] in, is the position of individual i in the jth dimension after update; Stallion j is the position of the stallion in the jth dimension; x ij is the position of individual i in the jth dimension; Z is a modulation factor used to control the update amplitude and direction; R is a random number whose value range is [-2,2].

[0126] Z=R2⊙IDX+R3⊙(~IDX)

[0127] Among them, ⊙ represents the dot product operation; ~ represents binary inversion; R2 and R3 are random numbers in the range [0,1]; IDX is a binary mask that determines whether R2 or R3 is used.

[0128] IDX=(P==0)

[0129] Where P is a Boolean variable indicating whether the random condition is met.

[0130] P=R1<TDR

[0131] Among them, R1 is a random number in the range of [0,1]; TDR is a threshold that determines whether a certain update strategy is used in the current iteration.

[0132]

[0133] Among them, ite is the current iteration number; ite max is the maximum number of iterations.

[0134] During mating, individuals search for mates and their positions are updated as follows:

[0135]

[0136] in, represents the average of two individual positions; represents the position of foal k in population p; represents the position of foal i in population q; represents the position of foal j in population z.

[0137] Individuals that do not mate or graze adopt the Lévy flight mode to wander to enhance global search capabilities. The position update formula of Lévy flight is:

[0138] x i ′(t)=x i (t)+l·Levy(λ)

[0139] Where: xi (t) represents the i-th solution of the t-th generation, l is the weight controlling the step length, usually defined as l = 0.01×(x i (t)-x b ), where x b is the current optimal solution, Levy(λ) represents the path that obeys the Levy distribution, satisfying Levy~u=t -λ , where 1<λ≤3.

[0140] (3) Calculate the fitness of the new population and determine the leader based on the fitness. The leader leads the population to the new habitat and competes for the owned land. If it fails, it will move to another habitat. The leader's position is updated as follows:

[0141]

[0142] Among them, WH is the location of the habitat, that is, the location of the global optimal individual; is the current position of the leader of population i; Stallion i is a random number in the range [0,1].

[0143] (4) Repeat steps (2) to (3) until the maximum number of iterations is reached, then end the loop and output the optimal fitness and its corresponding control point value.

[0144] By substituting the optimized control points into the parameter equation of the Bessel function mentioned above, the optimal cruising speed curve of the vehicle can be obtained.

[0145] The present invention uses the ECMS global optimization strategy to provide the best control solution for the engine and motor at each moment, including the following steps:

[0146] (S41) The engine fuel consumption rate, torque external characteristics, and engine speed range are known. The motor fuel consumption rate, torque external characteristics, and motor speed range can be expressed as b e =f1(n e ,T e ), T e_max =f2(n e ), n e_min <n e <n e_max , η m =f3(n m ,T m ), T m_min =f4(n m ), n e_min <n e <n e_max ;

[0147] (S42) Determine the feasible region of the motor torque [T m_lim_min ,T m_lim_max ] and divided into N equal parts [T m_lim_min ,…,T m_i ,…T m_lim_max ], then calculate the possible values ​​of the corresponding engine torque T e_i =T d -T m_i ;

[0148] (S43) Calculate the corresponding engine power P ei 、Motor power P mi and engine fuel consumption m f_i =(b i ·P ei 0.001) / 3600;

[0149] (S44) Calculate the corresponding equivalent fuel consumption m fequ_i =m f_i +s(t)·P mi / 42500 and solve for the minimum equivalent fuel consumption, and then solve for the minimum equivalent fuel consumption m that meets the SOC limit fequ_min =min{m fequ_1 ,m fequ_2 ,…,m fequ_i}, and obtain the corresponding optimized motor torque T m_opt , and then through T e_opt =T d -T m_opt Find the optimal engine torque.

[0150] (S45) The minimum shutdown time and the minimum continuous start time of the engine are set to reduce the problem of frequent start and stop of the engine in a very short time when only minimizing the instantaneous equivalent fuel consumption is considered.

[0151] To verify the advantages of the present invention, the results of the conventional hybrid vehicle cruise control method and the hybrid vehicle adaptive cruise control based on speed planning of the present invention are compared, and the convergence speed of the hybrid strategy optimized Mustang algorithm of the present invention and the conventional Mustang algorithm in this method are compared.

[0152] Figure 4 The comparison between the hybrid strategy optimization of the present invention and the wild horse algorithm is shown, showing that compared with the original wild horse algorithm, the optimization algorithm proposed by the present invention converges faster in this optimization and the obtained result has better fitness.

[0153] Figure 5 The comparison of the engine start-stop optimization of the present invention is shown. After the optimization, the working state of the engine is more centralized, and the idling fuel consumption of the engine is reduced.

[0154] Figure 6 The SOC curve of the strategy of the present invention is compared with the SOC curves of other strategies.

[0155] Figure 7 The fuel consumption of the strategy of the present invention is compared with that of other strategies. Figure 6 By comparing the SOCs of the hybrid vehicles, it can be found that while maintaining a similar SOC of the battery, the fuel consumption of the hybrid vehicle adaptive cruise control based on vehicle speed planning of the present invention is 6.56L, compared with the fuel consumption of 7.17L of the cruise control, and the optimization range of the energy-saving optimization performance has reached 8.51%.

[0156] The above description is only a preferred embodiment of the present invention. It should be pointed out that a person skilled in the art can make several improvements without departing from the principle of the present invention, and these improvements should also be regarded as within the protection scope of the present invention.

Claims

1. A vehicle adaptive cruise control method based on vehicle speed planning, characterized in that: The following steps are involved: S1: The driver selects the vehicle navigation itinerary according to the destination; S2: The vehicle receives satellite signals to determine the current location and obtains the altitude and speed limit data of the road ahead; S3: constructing an n-order Bessel function, and generating a vehicle speed curve with respect to the road position through the Bessel function; S4: construct the evaluation function of the vehicle speed curve and establish constraints; S5: Improve the Mustang optimization algorithm by introducing Gaussian distribution to generate the initial population and the individual position update strategy based on Levy flight to improve its global search capability and convergence speed; then input the Bessel function and evaluation function into the improved Mustang optimization algorithm, and finally output the optimal cruising speed curve; S6: Based on the optimal cruising speed curve, the ECMS global optimization strategy is used to introduce an equivalent factor to determine the conversion relationship between fuel energy and electrical energy. The global optimal control strategy can be obtained by minimizing the Hamiltonian value, and then the optimal control method of the engine and motor at each moment is obtained, and the engine is controlled.

2. The vehicle adaptive cruise control method based on vehicle speed planning according to claim 1, characterized in that: In S2, the road ahead is discretized at equal intervals, and the elevation data at each point is obtained using linear interpolation; for road segments, the road slope on each segment is solved using the height difference of the starting point and the starting point interval.

3. The vehicle adaptive cruise control method based on vehicle speed planning according to claim 1, characterized in that: In S3, the Bessel function is used to describe the relationship between vehicle speed and road distance, specifically: Define a set of control points (x i ,y i ), where x i Represents the road distance, y i represents the vehicle speed at this distance. Assuming there are n+1 control points (P0, P1, ..., P n ), construct an n-order Bezier curve, where n ranges from [10,30], and the generated speed curve has the following parametric equation: Among them, t is a parameter, ranging between [0,1]; is the binomial coefficient, calculated as P i =(x i ,y i ) is the i-th control point.

4. The vehicle adaptive cruise control method based on vehicle speed planning according to claim 1, characterized in that: In S4, the evaluation function of the vehicle speed curve is: L=Cost power +Cost Mobility +Cost comfort Cost power is the energy consumption required for vehicle driving, Cost Mobility Cost is the penalty for a vehicle speed that deviates from the road speed limit. comfort It is driving comfort; Cost power =w1·P req (t) Cost Mobility =w2(x2(t)-v l ) 2 Cost comfort =w3·u1(t) 2 Among them, w1, w2 and w3 are weight factors to ensure the balance between energy economy, power and driving comfort; v l is the speed limit on the test road; Where m is the mass of the vehicle, g is the acceleration of gravity, f is the rolling resistance coefficient, C D is the air resistance coefficient, ρ is the air density, A f is the frontal area; when P req When (t) is positive, it indicates the total power spent on vehicle state changes. req When (t) is negative, it describes the power charged into the battery.

5. The vehicle adaptive cruise control method based on vehicle speed planning as claimed in claim 4, characterized in that: The evaluation function has the following constraints: Acceleration constraint: The maximum acceleration is set to u1(t), Among them, F t is the maximum driving force of the vehicle's powertrain. The acceleration range only applies to acceleration and deceleration during speed planning. The constraint is expressed as a set of allowable accelerations: Among them, u min is the minimum acceleration, u1(t) is the vehicle acceleration at time t, u max is the maximum acceleration, t0 is the initial time, and T is the total driving time of the vehicle; Speed ​​limit: Among them, v min is the minimum vehicle speed, v1(t) is the vehicle speed at time t, u max is the maximum vehicle speed, t0 is the initial time, and T is the total vehicle travel time.

6. The vehicle adaptive cruise control method based on vehicle speed planning according to claim 1, characterized in that: In S5, the improvements to the Mustang optimization algorithm are as follows: (51) Use Gaussian distribution to generate the initial population; The population size is N, and each candidate solution is a vector x i , where i = 1, 2, ..., N. The candidate solution here represents the coordinates of the control points of the Bezier curve mentioned above. The initial solution is generated by setting the mean μ and standard deviation σ as the center of the search space. The standard deviation can be adjusted according to the range of the search space. The candidate solution is expressed as: x i =μ+σ×randn() Where μ is the center of the search space, calculated as σ is the standard deviation, set to Assuming that 99.7% of the data falls within the range, randn() is a function that generates standard normal distributed random numbers; Since the Gaussian distribution can generate values ​​that are out of bounds, these values ​​need to be processed to ensure that they are within the search space: X i =max(X min ,min(X i ,X max )) (52) Determine the leader of the initial population based on fitness and decide whether to engage in grazing and mating behaviors; When grazing, individuals tend to forage near the leader, and their positions are updated as follows: in, is the position of individual i in the jth dimension after update; Stallion j is the position of the stallion in the jth dimension; x ij is the position of individual i in the jth dimension; Z is the modulation factor; R is a random number whose value range is [-2,2]; Z=R2⊙IDX+R3⊙(~IDX) Among them, ⊙ represents the dot product operation; ~ represents binary inversion; R2 and R3 are random numbers in the range [0,1]; IDX is a binary mask that determines whether R2 or R3 is used; IDX=(P==0) Among them, P is a Boolean variable, indicating whether the random condition is met; P=R1<TDR Among them, R1 is a random number in the range of [0,1]; TDR is a threshold that determines whether a certain update strategy is used in the current iteration; Among them, ite is the current iteration number; ite max is the maximum number of iterations; During mating, individuals search for mates and their positions are updated as follows: in, represents the average of two individual positions; represents the position of foal k in population p; represents the position of foal i in population q; represents the position of foal j in population z; Individuals that do not mate or graze adopt the Lévy flight mode to wander to enhance global search capabilities. The position update formula of Lévy flight is: x i ′(t)=x i (t)+l·Levy(λ) Where: x i (t) represents the i-th solution of the t-th generation, l is the weight controlling the step length, defined as l = 0.01×(x i (t)-x b ), where x b is the current optimal solution; Levy(λ) represents the path that obeys the Levy distribution and satisfies Levy~u=t -λ , where 1<λ≤3; (53) Calculate the fitness of the new population and determine the leader based on the fitness. The leader leads the population to the new habitat and competes for the owned land. If it fails, it will move to another habitat. The leader's position is updated as follows: Among them, WH is the location of the habitat, that is, the location of the global optimal individual; is the current position of the leader of population i; Stallion i is a random number in the range [0,1]; (54) Repeat steps (52) to (53) until the maximum number of iterations is reached, end the loop, output the optimal fitness and its corresponding control point value, and substitute the optimized control point into the parameter equation of the Bessel function to obtain the optimal cruising speed curve of the vehicle.

7. The vehicle adaptive cruise control method based on vehicle speed planning according to claim 1, characterized in that: S6 includes the following steps: (61) The engine fuel consumption rate, torque external characteristics, engine speed range, motor consumption rate, torque external characteristics, and motor speed range are respectively expressed as b e =f1(n e ,T e ), T e_max =f2(n e ), n e_min <n e <n e_max , η m =f3(n m ,T m ), T m_min =f4(n m ), n e_min <n e <n e_max ; (62) Determine the feasible region of motor torque [T m_lim_min ,T m_lim_max ] and divided into N equal parts [T m_lim_min ,…,T m_i ,…T m_lim_max ], then calculate the possible values ​​of the corresponding engine torque T e_i =T d -T m_i ; (63) Calculate the corresponding engine power P ei 、Motor power P mi and engine fuel consumption m f_i =(b i ·P ei 0.001) / 3600; (64) Calculate the corresponding equivalent fuel consumption m fequ_i =m f_i +s(t)·P mi / 42500, and solve for the minimum equivalent fuel consumption, and then solve for the minimum equivalent fuel consumption m that meets the SOC limit fequ_min =min{m fequ_1 ,m fequ_2 ,…,m fequ_i }, and obtain the corresponding optimized motor torque T m_opt , and then through T e_opt =T d -T m_opt Find the optimized engine torque; (65) The minimum engine shutdown time and the minimum continuous start time are set to reduce the problem of frequent engine starts and stops in a very short time when only the instantaneous equivalent fuel consumption is minimized.

Images