Collaborative optimization method for wireless charging interval site selection of signalized intersection and ecological driving of intelligent networked electric vehicle

Abstract

The invention discloses a collaborative optimization method for signalized intersection wireless charging interval site selection and intelligent network connection electric vehicle ecological driving. The collaborative optimization method comprises the following steps: 1, determining a driving speed curve of an electric vehicle; 2, constructing an upper-layer planning model taking minimization of weighted wireless charging facility construction cost and motorcade driving cost as objective functions, and a lower-layer planning model for optimizing an ecological driving track after a motorcade enters an upstream road of the signalized intersection; and 3, solving the bilayer programming model by using a mixed algorithm based on a radial basis function (RBF) algorithm and a genetic algorithm (GA), and determining the position of the wireless charging interval. According to the invention, through collaborative optimization of the position of the wireless charging section and the ecological driving track of the motorcade, the time and energy consumption of the vehicle passing through the signalized intersection are fully reduced, and the traffic jam at the signalized intersection and the charging difficulty of the electric vehicle are relieved at the same time.

Description

Technical Field

[0001] The present invention belongs to the technical field of urban traffic control systems, and relates to the fields of electric vehicle ecological driving, road wireless charging location selection, and collaborative trajectory optimization and operations optimization algorithms during vehicle collisions. Specifically, it is a method for collaborative optimization of wireless charging section site selection at a signal intersection and ecological driving of intelligent networked electric vehicles. Background Art

[0002] In recent years, the rapid development of intelligent connected vehicle technology has accelerated the transformation of vehicles from traditional driver driving to intelligent connected vehicle collaborative ecological driving. Autonomous driving and vehicle networking technology have become important development directions in the future transportation field.

[0003] With the support of current technology, intelligent connected vehicles can be equipped with advanced on-board sensors, control systems, actuators, wireless communication systems and other devices to achieve information exchange and sharing between vehicles and X (people, vehicles, roads, etc.). Through sensors, cars can detect and perceive the environment around the car in real time, capture information such as road conditions, obstacles, other vehicles and pedestrians, and use advanced algorithms and models to make intelligent decisions. And through comparison, it is found that the advantages of equipping electric vehicles with intelligent connected vehicle technology will be more obvious. Compared with traditional cars, electric vehicles are easier to integrate advanced control systems and Internet systems, and it is also easier to control the vehicle's power system to achieve high-precision autonomous driving. However, the corresponding problems such as fleet ecological driving optimization, charging facility construction and layout optimization have also become more prominent.

[0004] First, during the process of optimizing the ecological driving trajectory of the vehicle, when the optimal trajectories of the front and rear vehicles conflict, only optimizing the rear vehicle will cause congestion in the rear traffic due to the obstruction of the front vehicle, which cannot be alleviated, and will eventually cause unnecessary time and energy consumption. Secondly, for the location of wireless charging facilities, there are very few related literatures that combine the actual situation of vehicles driving on the road section within a certain period of time to solve it. Setting the location of the wireless charging interval based on experience alone cannot make the road traffic conditions reach the optimal level. In addition, the traditional algorithms are inefficient in solving complex optimization problems and lack real-time response capabilities, resulting in delayed optimization results, which is not conducive to quickly obtaining the optimal trajectory of the fleet and easily affects the safety of vehicles on the road. Summary of the invention

[0005] The present invention aims to address the deficiencies of the above-mentioned prior art and proposes a collaborative optimization method for the site selection of wireless charging intervals at signalized intersections and the ecological driving of intelligent connected electric vehicles, aiming to determine the location of wireless charging interval facilities and optimize the ecological driving trajectory of vehicles to reduce the time and energy consumption of vehicles passing through intersections, while alleviating traffic congestion at intersections and the difficulty of charging electric vehicles.

[0006] In order to achieve the above-mentioned object of the invention, the present invention adopts the following technical scheme:

[0007] The collaborative optimization method for wireless charging section site selection at a signal intersection and ecological driving of an intelligent networked electric vehicle is characterized in that it is applied to a single lane with a length of D, and the starting point D of the single lane is start Located upstream of the signalized intersection, let the starting point D start The position is the coordinate origin of the single lane, at the starting point D start A vehicle detection device is provided at the intersection for obtaining the time when the electric vehicle enters the single lane, the end of the single lane being located at the exit position D of the signalized intersection. cross,end , the entry position of the signalized intersection is D cross,start , at the starting point D start and drive into position D cross,start There is a wireless charging zone between E,start ,D E,end ]; among them, D E,start Indicates the starting point of the interval, D E,end represents the end point of the interval; the collaborative optimization method is performed according to the following steps:

[0008] Step 1. The total number of electric vehicles entering the single lane in any time period T is obtained as I; and each electric vehicle travels on the single lane according to the speed curve, and the speed curve of the i-th electric vehicle in the process of traveling on the single lane includes five stages:

[0009] The first stage is the uniform speed driving stage. When the i-th electric vehicle enters the single lane, The time until the first stage of the i-th electric car ends During the time period, the i-th electric car enters the single lane at an initial speed From the starting point of the bicycle lane D start Drive at a constant speed to the position of the i-th electric car at the end of the first stage Department;

[0010] The second stage is the uniform deceleration stage. The time until the second stage of the i-th electric car ends During the time period, the i-th electric car decelerates from The electric car is driven at a uniform deceleration until the end of the second stage. At, the speed is Slow down to the minimum speed of the i-th electric vehicle on a single lane ;

[0011] The third stage is the uniform speed driving stage. The time until the third stage of the i-th electric vehicle ends During the time period, the i-th electric car from The i-th electric car is located at a constant speed at the end of the third stage. Department;

[0012] The fourth stage is the uniform acceleration stage. The time until the fourth stage of the i-th electric car ends During the time period, the i-th electric car accelerates from The i-th electric car is located at the end of the fourth stage. At, the speed is Speed ​​up to ;

[0013] The fifth stage is the uniform speed driving stage. The time until the fifth stage of the i-th electric car ends During the time period, the i-th electric car from Drive at a constant speed to the entry position D of the signalized intersection cross,start ;

[0014] Step 2. Construct a two-level planning model for the coordinated optimization of wireless charging station site selection and intelligent connected electric vehicle ecological driving:

[0015] Step 2.1. Construct an upper-level planning model for wireless charging interval site selection;

[0016] Step 2.2. Construct the lower-level planning model of the ecological driving of intelligent connected electric vehicles;

[0017] Step 3: Use the RBF-GA hybrid algorithm to solve the two-level programming model and obtain the wireless charging interval [D E,start ,D E,end ]The optimal evaluation position Z on a single lane best :

[0018] The collaborative optimization method of the present invention is also characterized in that step 2.1 comprises:

[0019] Step 2.1.1. Use formula (1) to construct the objective function F of the upper-level planning model up :

[0020] (1)

[0021] In formula (1), , , They are respectively the time cost coefficient, the power cost coefficient, and the number of electric vehicles passing through the wireless charging interval [D E,start ,D E,end ] construction cost coefficient, , They are the proportion of vehicle driving consumption and wireless charging area construction consumption, The exit position D of the i-th electric vehicle when it exits the single lane and reaches the signalized intersection cross,end Time, E i is the net energy consumed by the i-th electric vehicle passing through a single lane;

[0022] Step 2.1.2. Use formula (2) to construct the constraints of the upper-level planning model:

[0023] (2).

[0024] Furthermore, the step 2.2 includes:

[0025] Step 2.2.1: Use formula (3) to construct the objective function F of the lower-level planning model down :

[0026] (3)

[0027] Step 2.2.2: Use equations (4) to (24) to construct the constraints of the lower-level planning model:

[0028] (4)

[0029] (5)

[0030] (6)

[0031] (7)

[0032] (8)

[0033] (9)

[0034] (10)

[0035] (11)

[0036] (12)

[0037] (13)

[0038] (14)

[0039] (15)

[0040] (16)

[0041] (17)

[0042] (18)

[0043] (19)

[0044] (20)

[0045] (twenty one)

[0046] (twenty two)

[0047] (twenty three)

[0048] (twenty four)

[0049] In formula (4) to formula (24), , They respectively represent the maximum acceleration and minimum acceleration allowed for an electric vehicle when driving on a single lane; Indicates the maximum speed that an electric vehicle is allowed to travel on a single lane; , They represent the positions of the i-th electric car and the i+1-th electric car at time t, respectively. , They represent the minimum headway distance when an electric vehicle stops and the reaction time of the electric vehicle to receive signals and perform related operations. represents the speed of the i+1th electric car at time t; , They represent the i-th electric vehicle entering the wireless charging interval [D E,start ,D E,end ] and leave the wireless charging area [D E,start ,D E,end ] time; , , They represent the instantaneous power, traction power, and recovery power of the i-th electric vehicle at time t respectively; For the quality of electric vehicles, represents the acceleration of the ith electric car at time t, g is the acceleration due to gravity, 𝜃 is the inclination of the road, C r , c1, c2 are three rolling resistance parameters, ρ air is the air mass density, C D is the aerodynamic drag coefficient of the electric vehicle, A f is the front windshield area of ​​the electric vehicle, v i (t) represents the speed of the ith electric car at time t, Indicates v i (t) squared; , , They are the transmission system efficiency factor, motor efficiency factor, and battery efficiency factor. represents the recovery braking energy efficiency factor of the i-th electric vehicle at time t, 𝛼 is the parameter of the electric vehicle; , They represent the traction energy consumption and regeneration energy of the i-th electric vehicle when it is driving on a single lane, , They represent the total power consumption and total charging power of the i-th electric vehicle passing through a single lane, P charge It is the charging capacity per unit time.

[0050] Furthermore, the step 3 comprises:

[0051] Step 3.1. Define the maximum number of iterations as R max , the current number of iterations is n, and the maximum number of consecutive successes is suc max , the maximum number of consecutive failures is fai max , the initial step length is stp len , the size of the candidate point set is N size The upper and lower limits of the predicted score coefficient are μ max , μ min , initialize n=1;

[0052] Step 3.2. D E,start and D E,end As the decision variable of the upper-level planning model, and according to the number of decision variables of the upper-level planning model 2, in D start and D cross,start After randomly generating 2+1 initial evaluation positions, they are stored in the evaluation position set Z; each initial evaluation position is a combination of 2 decision variables;

[0053] Step 3.3. Bring any initial evaluation position z in Z into the lower-level planning model for solution to obtain the objective function value corresponding to z ;in, ;Will Bring it into the upper-level planning model for solution and obtain the objective function value corresponding to z After that, it is stored in the objective function set W;

[0054] Step 3.4. Let the minimum value in W be the optimal objective function value W best , W best The corresponding optimal evaluation position is recorded as Z best ;

[0055] Step 3.5. Based on the set Z and the set W, get the RBF parameter RBF after the nth update n,par ;

[0056] Step 3.6. According to Z best When constructing the nth iteration, a group of length N size The candidate evaluation position set P n ;

[0057] Step 3.7. RBF-based n,par , get the predicted value score corresponding to p , distance score , and thus calculate P n The comprehensive score of the candidate evaluation position p in ,in, is the weight coefficient for the nth iteration; and ,in, , are the minimum weight and maximum weight of the nth iteration respectively;

[0058] Step 3.8. According to P n The comprehensive score of each candidate evaluation position in is selected, and the candidate evaluation position corresponding to the minimum value is selected as the best evaluation position z at the nth iteration. n , and z n After depositing Z, follow the process in step 3.3 to get z n The corresponding upper-level planning model objective function is , and the updated W;

[0059] Step 3.9. Judgment Is it greater than W? best , if so, then W best and Z best Keep unchanged, let suc n =0, will fai n-1 +1 to fai n , otherwise, let Z best =z n , W best = ; will suc n-1+1 assigned to suc n , so fai n =0;

[0060] Step 3.10. When fai n >fai max When Assign to ,Will Assign to ,Will Assign to stp len After that, determine stp len Is it greater than D E,end If so, let stp len =D E,end ; Otherwise, directly execute step 3.11, where , and Represent 3 multiples respectively, and ; ;

[0061] When success n >suc max , when stp len / 2 assign value to stp len After that, if stp len <D E,start , then let stp len =D E,start ; Otherwise, go directly to step 3.11;

[0062] Step 3.11. If n=R max , then the final W is obtained best and Z best Otherwise, assign n+1 to n and return to step 3.5 to execute sequentially.

[0063] Furthermore, in step 3.3, the objective function value corresponding to z is obtained according to the following process: :

[0064] Step 3.3.1: Let the period of the traffic light at the intersection be t c , and the red light is used as the signal light color at the beginning of each cycle, where the green light is in cycle t c The percentage of the following is t G ; After the i-th electric vehicle enters the single lane, the number of vehicles that conflict with the i-th electric vehicle is recorded as , and initialize ;

[0065] Step 3.3.2: Calculate the time it takes for the i-th electric car to reach D cross,startThe cycle number of the signal light ;in, Indicates rounding up;

[0066] Step 3.3.3: According to The maximum number of electric vehicles that can pass within a green light period and the vehicle sequence j of the i-th electric vehicle passing through the signalized intersection in the h-th cycle i ,Will After assigning the value to h, execute step 3.3.4; where, Indicates rounding down;

[0067] Step 3.3.4: Calculate the time it takes for the i-th electric car to reach D in the h-th cycle. cross,start The time limit And the time limit ;in, Indicates that the i-1th electric car arrives at D cross,start time;

[0068] Step 3.3.5: Judgement Is it true? If so, the i-th electric car and the i-1-th electric car form a fleet. Assign to Then, execute step 3.3.6; otherwise, it means that the i-th electric car is the only vehicle in the fleet, and directly execute step 3.3.6;

[0069] Step 3.3.6: Take the electric car in front of the i-th electric car in the fleet The car is Electric vehicles, The end time of the five-stage speed curve of each electric vehicle is used as the decision variable, and the GA algorithm is used to solve the lower-level planning model to obtain the first The combination of five decision variables for an electric vehicle And store the serial number in the electric vehicle decision variable set L At the corresponding position, Substitute into equation (3)-(24) to determine The objective function value of an electric vehicle Store the sequence number in the electric vehicle lower layer objective function value set Y The corresponding position;

[0070] like , then directly execute step 3.3.8; otherwise, execute step 3.3.7;

[0071] Step 3.3.7: In the convoy, except for the first vehicle, The five decision variables of the ikth electric car besides the first electric car Store it in the position corresponding to the serial number ik in L, and Substitute into equations (3) to (24) to determine the objective function value of the ikth vehicle: And store it in the position corresponding to the serial number ik in Y, ;

[0072] If the time interval between the ik-1th electric vehicle and the ikth electric vehicle entering the single lane is When , the five decision variables of the ikth electric vehicle in the fleet are calculated using formula (26), including:

[0073] (26)

[0074] In formula (26), , , , , They represent the five decision variables of the ikth electric car, , , , , They represent the five decision variables of the ik-1th electric car;

[0075] If the time interval between the ik-1th electric vehicle and the ikth electric vehicle entering the single lane is greater than When , the five decision variables of the ikth electric vehicle in the fleet are calculated, including:

[0076] (27)

[0077] In formula (27), , , , They represent the time point when the ik-1th electric vehicle enters the single lane and starts to decelerate uniformly, and the time point when the ikth electric vehicle enters the single lane and starts to decelerate uniformly, respectively. , They respectively represent that if the time interval between the ik-1th electric vehicle and the ikth electric vehicle entering the single lane is The estimated time to enter the single lane and start deceleration;

[0078] like When , according to formula (27) we get , and use formula (28) to get the remaining four decision variables:

[0079] (28)

[0080] like When , five decision variables are obtained using formula (29):

[0081] (29)

[0082] Step 3.3.8: Determine the Electric car and Does the electric vehicle satisfy equation (13)? If so, If yes, execute step 3.3.10; otherwise, execute step 3.3.9:

[0083] Step 3.3.9: Determine the Is the electric car the first vehicle to pass through the signal intersection in the hth signal cycle? If so, The shared After the electric vehicles form a team, return to step 3.3.6 and execute in sequence; otherwise, Assign to , will The total number of electric vehicles from the first electric vehicle to the i-th electric vehicle After the electric vehicles form a fleet, the execution returns to 3.3.6 and executes in order;

[0084] Step 3.3.10: If i=I, then sum all the values ​​in Y to get Otherwise, assign i+1 to i and return to step 3.3.2 and execute in sequence.

[0085] Furthermore, in step 3.3.6, the GA algorithm is used to solve the lower-level planning model according to the following process;

[0086] Step 3.3.6.1: Define the maximum number of iterations as G max , the current number of iterations is g, the population size is pop, initialization g=1, and the initialization step size is stp;

[0087] The first car in the convoy is Serial number of the electric car Let it be u, then the optimal decision variable corresponding to the u-th electric car is , and initialize Empty; initialized The corresponding optimal objective function value is infinite;

[0088] Randomly generate a set of g-generation population P with size pop that meets the constraints of equations (4)-(24)g ; and P g Each individual in represents the five decision variables of the u-th electric car;

[0089] Step 3.3.6.2: P g Mutate to get the g-th generation offspring ;

[0090] Step a: Randomly select P g The bth individual p g,b , ; For the bth individual p g,b The corresponding five decision variables The decision variables are Mutate within and determine whether the mutated individuals meet the constraints of formula (4)-formula (24). If they meet the constraints, add the corresponding mutated individuals to Otherwise, the corresponding mutated individuals are discarded; Indicates the decision variables that need to be varied;

[0091] Step b: Judgement Whether the number of individuals in reaches pop, if so, it means that , otherwise, return to step a and execute sequentially;

[0092] Step 3.3.6.4: P g and Any individual in , and calculate according to the procedure in step 3.3.7 The decision variables of each electric vehicle in the fleet are substituted into the lower-level planning model for solution, and the obtained The objective function value of the team, so as to find P g and The minimum objective function value and its corresponding individuals ;

[0093] like < , then let Assign to , Assign to ;otherwise, and remain unchanged;

[0094] Step 3.3.6.5: If g = G max , then the output , otherwise, execute step 3.3.6.6;

[0095] Step 3.3.6.6: Join the g+1 generation population P g+1 ,like Belong to P g The individuals in From P g When deleting The individuals at the corresponding positions also If belong The individuals in from When deleting The individuals at the corresponding positions also start from P g Delete from;

[0096] Step 3.3.6.7: Compare P g and The objective function values ​​corresponding to the individuals in the remaining positions in P are selected, and the individuals corresponding to the positions with the minimum objective function values ​​are added to P g+1 , thus obtaining the g+1th generation population P containing pop individuals g+1 ,

[0097] Step 3.3.6.8: After assigning g+1 to g, return to step 3.3.6.2.

[0098] Further, the step 3.6 includes:

[0099] Step 3.6.1. At the nth iteration, from Z best Select the rth decision variable d from the two decision variables r , ;

[0100] Step 3.6.2. Use equation (25) to calculate d r After mutation, the mutated rth decision variable d is obtained r,1 :

[0101] (25)

[0102] Step 3.6.3. Z best Another decision variable d y With d r,1 Compare and select the larger value as D E,end , the smaller value is taken as D E,start , thus forming a candidate evaluation position p, ,and ;

[0103] Step 3.6.4. Determine whether p satisfies the constraint conditions shown in formula (2). If so, store it in P n, otherwise, discard p;

[0104] Step 3.6.5. Obtain N according to the process of steps 3.6.1 to 3.6.4. size candidate evaluation locations.

[0105] An electronic device of the present invention includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the collaborative optimization method, and the processor is configured to execute the program stored in the memory.

[0106] The present invention provides a computer-readable storage medium, wherein a computer program is stored on the computer-readable storage medium, and the computer program executes the steps of the collaborative optimization method when executed by a processor.

[0107] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0108] 1. In view of the fact that vehicles in a convoy often travel on the road at uneven intervals in real life. The present invention optimizes the convoy's driving energy consumption and driving time as dual objectives. First, each vehicle is optimized for its trajectory separately, and then the vehicles with conflicting trajectories are combined into a following convoy for overall collaborative optimization to reduce energy consumption and time costs. This can improve the situation where the rear vehicles are congested due to the optimal trajectory of the front vehicle, so that the vehicles in the convoy can drive ecologically with the least overall energy consumption and driving time, which is ultimately of great significance to improving road capacity and reducing energy loss.

[0109] 2. The present invention establishes a two-level programming model to solve the problem of collaborative optimization of the location of wireless charging intervals and the trajectory of the fleet. The actual driving conditions of the vehicle can be used to determine the location of the wireless charging interval, and the layout of these wireless charging intervals in turn affects the optimization of the driving trajectory of the electric vehicle. This two-way interactive relationship enables the two to promote each other and jointly improve the charging efficiency of electric vehicles and the rationality of the driving path. In addition, the model takes into account the instantaneous energy consumption of the vehicle during driving, and can more accurately calculate the vehicle driving cost, making it more in line with reality.

[0110] 3. The present invention adopts the RBF-GA hybrid algorithm to solve the problem of collaborative optimization of the wireless charging interval position and the vehicle ecological driving trajectory. Under complex constraints, the RBF-GA algorithm can quickly adapt to different traffic conditions and efficiently locate the optimal wireless charging interval position. It can also provide the optimal trajectory for each vehicle in a limited time according to the time when the vehicle enters the road and the different front and rear environments. This not only ensures the safe driving of the vehicle, but also significantly improves the overall energy efficiency and operating efficiency of the system, providing reliable technical support for collaborative optimization problems in complex traffic scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0111] Figure 1 It is the flow chart of RBF-GA hybrid algorithm;

[0112] Figure 2 It is a single lane road with charging facilities;

[0113] Figure 3 It is a five-stage speed control process diagram when the vehicle is driving;

[0114] Figure 4 This is a schematic diagram of a convoy with evenly distributed vehicles;

[0115] Figure 5 The distance traveled by the vehicle at a constant speed in the first stage is greater than Schematic diagram;

[0116] Figure 6 The distance traveled by the vehicle at a constant speed in the first stage is less than Schematic diagram. DETAILED DESCRIPTION

[0117] In this embodiment, Figure 1 As shown in FIG, a collaborative optimization method for wireless charging section site selection at a signalized intersection and ecological driving of intelligent connected electric vehicles is applied to a single lane with a length of D, such as Figure 2 As shown, the single lane includes a wireless charging area, a signalized intersection, and a vehicle detection device, and the starting point D of the single lane start Located upstream of the signalized intersection, let the starting point D start The position is the coordinate origin of the single lane, at the starting point D start A vehicle detection device is provided at the intersection for obtaining the time when the electric vehicle enters the single lane, the end of the single lane being located at the exit position D of the signalized intersection. cross,end , the entry position of the signalized intersection is D cross,start , at the starting point D start and drive into position D cross,start There is a wireless charging zone between E,start ,D E,end ]; among them, D E,start Indicates the starting point of the interval, D E,end represents the end point of the interval; the collaborative optimization method is performed according to the following steps:

[0118] Step 1. Obtain the total number of electric vehicles entering the single lane in any time period T as I; and each electric vehicle enters the single lane as follows Figure 3 The five-stage speed curve shown in the figure is used to drive, and the speed curve of the i-th electric vehicle during driving on a single lane includes five stages:

[0119] The first stage is the uniform speed driving stage. When the i-th electric vehicle enters the single lane, The time until the first stage of the i-th electric car ends During the time period, the i-th electric car enters the single lane at an initial speed From the starting point of the bicycle lane D start Drive at a constant speed to the position of the i-th electric car at the end of the first stage Department;

[0120] The second stage is the uniform deceleration driving stage. The time until the second stage of the i-th electric car ends During the time period, the i-th electric car decelerates from The electric car is driven at a uniform deceleration until the end of the second stage. At, the speed is Slow down to the minimum speed of the i-th electric vehicle on a single lane ;

[0121] The third stage is the uniform speed driving stage. The time until the third stage of the i-th electric vehicle ends During the time period, the i-th electric car from The i-th electric car is located at a constant speed at the end of the third stage. Department;

[0122] The fourth stage is the uniform acceleration stage. The time until the fourth stage of the i-th electric car ends During the time period, the i-th electric car accelerates from The i-th electric car is located at the end of the fourth stage. At, the speed is Speed ​​up to ;

[0123] The fifth stage is the uniform speed driving stage. The time until the fifth stage of the i-th electric car ends During the time period, the i-th electric car from Drive at a constant speed to the entry position D of the signalized intersection cross,start .

[0124] Step 2. Construct a two-level planning model for the coordinated optimization of wireless charging station site selection and intelligent connected electric vehicle ecological driving:

[0125] Step 2.1: Construct an upper-level planning model for wireless charging interval site selection;

[0126] Step 2.1.1: Use formula (1) to construct the objective function F of the upper-level planning model up :

[0127] (1)

[0128] In formula (1), , , They are respectively the time cost coefficient, the power cost coefficient, and the number of electric vehicles passing through the wireless charging interval [D E,start ,D E,end ] construction cost coefficient, , They are the proportion of vehicle driving consumption and wireless charging area construction consumption, The exit position D of the i-th electric vehicle when it exits the single lane and reaches the signalized intersection cross,end Time, E i is the net energy consumed by the i-th electric car passing through a single lane.

[0129] Step 2.1.2: Use formula (2) to construct the constraints of the upper-level planning model:

[0130] (2)

[0131] Step 2.2: Construct the lower-level planning model of ecological driving of intelligent connected electric vehicles;

[0132] Step 2.2.1: Use formula (3) to construct the objective function F of the lower-level planning model down :

[0133] (3)

[0134] Step 2.2.2: Use equations (4) to (24) to construct the constraints of the lower-level planning model:

[0135] (4)

[0136] (5)

[0137] (6)

[0138] (7)

[0139] (8)

[0140] (9)

[0141] (10)

[0142] (11)

[0143] (12)

[0144] (13)

[0145] (14)

[0146] (15)

[0147] (16)

[0148] (17)

[0149] (18)

[0150] (19)

[0151] (20)

[0152] (twenty one)

[0153] (twenty two)

[0154] (twenty three)

[0155] (twenty four)

[0156] In formula (4) to formula (24), , They respectively represent the maximum acceleration and minimum acceleration allowed for an electric vehicle when driving on a single lane; Indicates the maximum speed that an electric vehicle is allowed to travel on a single lane; , They represent the positions of the i-th electric car and the i+1-th electric car at time t, respectively. , They represent the minimum headway distance when an electric vehicle stops and the reaction time of the electric vehicle to receive signals and perform related operations. represents the speed of the i+1th electric car at time t; , They represent the i-th electric vehicle entering the wireless charging interval [D E,start ,D E,end ] and leave the wireless charging area [D E,start ,D E,end ] time; , , They represent the instantaneous power, traction power, and recovery power of the i-th electric vehicle at time t respectively; For the quality of electric vehicles, represents the acceleration of the ith electric car at time t, g is the acceleration due to gravity, 𝜃 is the inclination of the road, C r , c1, c2 are three rolling resistance parameters, ρ air is the air mass density, C D is the aerodynamic drag coefficient of the electric vehicle, A f is the front windshield area of ​​the electric vehicle, v i (t) represents the speed of the ith electric car at time t, Indicates v i (t) squared; , , They are the transmission system efficiency factor, motor efficiency factor, and battery efficiency factor. represents the recovery braking energy efficiency factor of the i-th electric vehicle at time t, 𝛼 is the parameter of the electric vehicle; , They represent the traction energy consumption and regeneration energy of the i-th electric vehicle when it is driving on a single lane, , They represent the total power consumption and total charging power of the i-th electric vehicle passing through a single lane, P charge It is the charging capacity per unit time.

[0157] Step 3: Use the RBF-GA hybrid algorithm to solve the two-level programming model and obtain the wireless charging interval [D E,start ,D E,end ]The optimal evaluation position Z on a single lane best :

[0158] Step 3.1: Define the maximum number of iterations as R max , the current number of iterations is n, and the maximum number of consecutive successes is suc max , the maximum number of consecutive failures is fai max , the initial step length is stp len , the size of the candidate point set is N size The upper and lower limits of the predicted score coefficient are μ max , μmin , initialize n=1;

[0159] Step 3.2: D E,start and D E,end As the decision variable of the upper-level planning model, and according to the number of decision variables of the upper-level planning model 2, in D start and D cross,start After randomly generating 2+1 initial evaluation positions, they are stored in the evaluation position set Z; each initial evaluation position is a combination of 2 decision variables.

[0160] Step 3.3: Bring any initial evaluation position z in Z into the lower-level planning model for solution to obtain the objective function value corresponding to z ;in, ;Will Bring it into the upper-level planning model for solution and obtain the objective function value corresponding to z After that, it is stored in the objective function set W;

[0161] Among them, in step 3.3, the objective function value corresponding to z is obtained according to the following process: :

[0162] Step 3.3.1: Let the period of the traffic light at the intersection be t c , and the red light is used as the signal light color at the beginning of each cycle, where the green light is in cycle t c The percentage of the following is t G ; After the i-th electric vehicle enters the single lane, the number of vehicles that conflict with the i-th electric vehicle is recorded as , and initialize ;

[0163] Step 3.3.2: Calculate the time it takes for the i-th electric car to reach D cross,start The cycle number of the signal light ;in, Indicates rounding up;

[0164] Step 3.3.3: According to The maximum number of electric vehicles that can pass within a green light period and the vehicle sequence j of the i-th electric vehicle passing through the signalized intersection in the h-th cycle i ,Will After assigning the value to h, execute step 3.3.4; where, Indicates rounding down;

[0165] Step 3.3.4: Calculate the time it takes for the i-th electric car to reach D in the h-th cycle. cross,start The time limit And the time limit ;in, Indicates that the i-1th electric car arrives at D cross,start time.

[0166] Step 3.3.5: Judgement Is it true? If so, the i-th electric car and the i-1-th electric car form a fleet. Assign to Then, execute step 3.3.6; otherwise, it means that the i-th electric car is the only vehicle in the fleet, and directly execute step 3.3.6;

[0167] Step 3.3.6: Take the electric car in front of the i-th electric car in the fleet The car is Electric vehicles, The end time of the five-stage speed curve of each electric vehicle is used as the decision variable, and the GA algorithm is used to solve the lower-level planning model to obtain the first The combination of five decision variables for an electric vehicle And store the serial number in the electric vehicle decision variable set L At the corresponding position, Substitute into equation (3)-(24) to determine The objective function value of an electric vehicle Store the sequence number in the electric vehicle lower layer objective function value set Y The corresponding position;

[0168] like , then directly execute step 3.3.8; otherwise, execute step 3.3.7.

[0169] Among them, in step 3.3.6, the GA algorithm is used to solve the lower-level planning model according to the following process;

[0170] Step 3.3.6.1: Define the maximum number of iterations as G max , the current number of iterations is g, the population size is pop, initialization g=1, and the initialization step size is stp;

[0171] The first car of the team Serial number of the electric car Let it be u, then the optimal decision variable corresponding to the u-th electric car is , and initialize Empty; initialized The corresponding optimal objective function value is infinite;

[0172] Randomly generate a set of g-generation population P with size pop that meets the constraints of equations (4)-(24)g ; and P g Each individual in represents the five decision variables of the u-th electric car;

[0173] Step 3.3.6.2: P g Mutate to get the g-th generation offspring ;

[0174] Step a: Randomly select P g The bth individual p g,b , ; For the bth individual p g,b The corresponding five decision variables The decision variables are Mutate within, obtain pop mutated individuals, and determine whether the mutated individuals meet the constraints of formula (4)-formula (24). If they meet the constraints, add the corresponding mutated individuals to Otherwise, the corresponding mutated individuals are discarded; Indicates the decision variables that need to be varied;

[0175] Step b: Judgement Whether the number of individuals in reaches pop, if so, it means that , otherwise, return to step a and execute sequentially.

[0176] Step 3.3.6.4: P g and Any individual in , and calculate according to the procedure in step 3.3.7 The decision variables of each electric vehicle in the fleet are substituted into the lower-level planning model for solution, and the obtained The objective function value of the team, so as to find P g and The minimum objective function value and its corresponding individuals ;

[0177] like < , then let Assign to , Assign to ;otherwise, and Remain unchanged.

[0178] Step 3.3.6.5: If g = G max , then the output , otherwise, execute step 3.3.6.6;

[0179] Step 3.3.6.6: Join the g+1 generation population P g+1 ,like Belong to P g The individuals in From P g When deleting The individuals at the corresponding positions also If belong The individuals in from When deleting The individuals at the corresponding positions also start from P g Delete from;

[0180] Step 3.3.6.7: Compare P g and The objective function values ​​corresponding to the individuals in the remaining positions in P are selected, and the individuals corresponding to the positions with the minimum objective function values ​​are added to P g+1 , thus obtaining the g+1th generation population P containing pop individuals g+1 ,

[0181] Step 3.3.6.8: After assigning g+1 to g, return to step 3.3.6.2.

[0182] Step 3.3.7: In the convoy, except for the first vehicle The five decision variables of the ikth electric car besides the first electric car Store it in the position corresponding to the serial number ik in L, and Substitute into equations (3) to (24) to determine the objective function value of the ikth vehicle: And store it in the position corresponding to the serial number ik in Y, ;

[0183] If the time interval between the ik-1th electric vehicle and the ikth electric vehicle entering the single lane is When Figure 4 The figure shows a convoy of electric vehicles entering a single lane. The time interval between adjacent electric vehicles in the convoy is When , the following trajectory of the electric vehicle, therefore, we can use formula (30) to calculate the five decision variables of the ikth electric vehicle in the fleet, including:

[0184] (30)

[0185] In formula (30), , , , , They represent the five decision variables of the ikth electric car, , , , , They represent the five decision variables of the ik-1th electric car;

[0186] If the time interval between the ik-1th electric vehicle and the ikth electric vehicle entering the single lane is greater than When Figure 5 , Figure 6 As shown, Figure 5 It shows that when the distance traveled at a constant speed in the first stage by the ik-1 electric car is greater than At this time, the trajectory of the ikth electric car is followed by the ik-1th electric car. Figure 6 It shows that when the distance traveled at a constant speed in the first stage by the ik-1 electric car is less than At this time, the ikth electric vehicle is catching up with the trajectory of the ik-1th electric vehicle. Figure 5 , Figure 6 The five decision variables that can be calculated for the ikth electric vehicle in the fleet include:

[0187] (31)

[0188] In formula (31), , , , They represent the time point when the ik-1th electric vehicle enters the single lane and starts to decelerate uniformly, and the time point when the ikth electric vehicle enters the single lane and starts to decelerate uniformly, respectively. , They respectively represent that if the time interval between the ik-1th electric vehicle and the ikth electric vehicle entering the single lane is The estimated time to enter the single lane and start deceleration;

[0189] like When , according to formula (31) we get , and use formula (32) to get the remaining four decision variables:

[0190] (32)

[0191] like When , five decision variables are obtained using formula (29):

[0192] (33)

[0193] Step 3.3.8: Determine the Electric car and Does the electric vehicle satisfy equation (13)? If so, , execute step 3.3.10; otherwise, go to step 3.3.9:

[0194] Step 3.3.9: Determine the Is the electric car the first vehicle to pass through the signal intersection in the hth signal cycle? If so, The shared After the electric vehicles form a team, return to step 3.3.6 and execute in sequence; otherwise, Assign to , will The total number of electric vehicles from the first electric vehicle to the i-th electric vehicle After the electric vehicles form a fleet, the execution returns to 3.3.6 and executes in order;

[0195] Step 3.3.10: If i=I, then sum all the values ​​in Y to get Otherwise, assign i+1 to i and return to step 3.3.2 and execute in sequence.

[0196] Step 3.4: Let the minimum value in W be the optimal objective function value W best , W best The corresponding optimal evaluation position is recorded as Z best ;

[0197] Step 3.5: According to the set Z and the set W, get the RBF parameter RBF after the nth update n,par ;

[0198] (25)

[0199] (26)

[0200] In formula (25)-formula (26), is a (n+2)×1 dimensional matrix composed of all function values ​​in the set W in the order in which they are stored in the set. zero1 and zero2 are 3×3 dimensional zero matrices and 3×1 dimensional zero matrices respectively. The two decision variables in z are transformed according to D E,start In front, D E,end The following sequence forms a 1×2 dimensional matrix , we can get n+2 1×2 dimensional matrices from Z , ; for each Number them, and then randomly select two numbers q1 and q2 to find the corresponding The Euclidean distance between The corresponding position of row q1 and column q2 of , we can finally get a (n+2)×(n+2) dimensional matrix ; U is a (n+2)×3 dimensional matrix, U T is the transpose of U, Is the first The transpose of It's the second The transpose of It is the (n+2)th The transpose of .

[0201] Step 3.6: According to Z best When constructing the nth iteration, a group of length N size The candidate evaluation position set P n :

[0202] Step 3.6.1: At the nth iteration, from Z best Select the rth decision variable d from the two decision variables r , ;

[0203] Step 3.6.2: Use equation (27) to calculate d r After mutation, the mutated rth decision variable d is obtained r,1 :

[0204] (27)

[0205] Step 3.6.3: Z best Another decision variable d y With d r,1 Compare and select the larger value as D E,end , the smaller value is taken as D E,start , thus forming a candidate evaluation position p, ,and ;

[0206] Step 3.6.4: Determine whether p satisfies the constraints shown in formula (2). If so, store it in P n , otherwise, discard p;

[0207] Step 3.6.5: Obtain N by following the process from Step 3.6.1 to Step 3.6.4 size candidate evaluation locations.

[0208] Step 3.7: RBF-based n,par , get the predicted value score corresponding to p , distance score , and thus calculate P n The comprehensive score of the candidate evaluation position p in ,in, is the weight coefficient for the nth iteration; and ,in, , are the minimum weight and maximum weight of the nth iteration respectively;

[0209] (28)

[0210] (29)

[0211] In formula (28)-(29), d j Represents RBF n,par The value corresponding to the jth row in Indicates the Euclidean distance, p o Represents RBF n,par The value of the third row from the bottom, Represents RBF n,par The last two rows form a 2×1 dimensional matrix, S n(p) is the predicted value of p based on the current sample point Z, D n(p) is the minimum Euclidean distance of p to the current sample point Z, , , , Respectively represent S n(p) and D n(p) The maximum and minimum values ​​of .

[0212] Step 3.8: According to P n The comprehensive score of each candidate evaluation position in is selected, and the candidate evaluation position corresponding to the minimum value is selected as the best evaluation position z at the nth iteration. n , and z n After depositing Z, follow the process in step 3.3 to get z n The corresponding upper-level planning model objective function is , and the updated W;

[0213] Step 3.9: Judgement Is it greater than W? best , if so, then W best and Z best Keep unchanged, let suc n =0, will fai n-1 +1 to fai n , otherwise, let Z best =z n , W best = ; will suc n-1 +1 assigned to suc n , so fai n =0.

[0214] Step 3.10: When the n >fai max When Assign to ,Will Assign to ,Will Assign to stp len After that, determine stp len Is it greater than D E,end If so, let stp len =D E,end ; Otherwise, directly execute step 3.11, where, , and Represent 3 multiples respectively, and ; ;

[0215] When success n >suc max , when stp len / 2 assign value to stp len After that, if stp len <D E,start , then let stp len =D E,start ; Otherwise, go directly to step 3.11;

[0216] Step 3.11: If n=R max , then the final W is obtained best and Z best Otherwise, assign n+1 to n and return to step 3.5 to execute sequentially.

[0217] In this embodiment, an electronic device includes a memory and a processor, wherein the memory is used to store a program that supports the processor to execute the above method, and the processor is configured to execute the program stored in the memory.

[0218] In this embodiment, a computer-readable storage medium stores a computer program on the computer-readable storage medium, and the computer program executes the steps of the above method when executed by a processor.

Claims

1. A collaborative optimization method for wireless charging section site selection at a signalized intersection and ecological driving of intelligent networked electric vehicles, characterized in that Applied to a single lane with a length of D, the starting point of the single lane is D start Located upstream of the signalized intersection, let the starting point D start The position is the coordinate origin of the single lane, at the starting point D start A vehicle detection device is provided at the intersection for obtaining the time when the electric vehicle enters the single lane, the end of the single lane being located at the exit position D of the signalized intersection. cross,end , the entry position of the signalized intersection is D cross,start , at the starting point D start and drive into position D cross,start There is a wireless charging zone between E,start ,D E,end ]; among them, D E,start Indicates the starting point of the interval, D E,end represents the end point of the interval; the collaborative optimization method is performed according to the following steps: Step 1. The total number of electric vehicles entering the single lane in any time period T is obtained as I; and each electric vehicle travels on the single lane according to the speed curve, and the speed curve of the i-th electric vehicle in the process of traveling on the single lane includes five stages: The first stage is the uniform speed driving stage. When the i-th electric vehicle enters the single lane, The time until the first stage of the i-th electric car ends During the time period, the i-th electric car enters the single lane at an initial speed From the starting point of the bicycle lane D start Drive at a constant speed to the position of the i-th electric car at the end of the first stage Department; The second stage is the uniform deceleration stage. The time until the second stage of the i-th electric car ends During the time period, the i-th electric car decelerates from The electric car is driven at a uniform deceleration until the end of the second stage. At, the speed is Slow down to the minimum speed of the i-th electric vehicle on a single lane ; The third stage is the uniform speed driving stage. The time until the third stage of the i-th electric car ends During the time period, the i-th electric car from The i-th electric car is located at a constant speed at the end of the third stage. Department; The fourth stage is the uniform acceleration stage. The time until the fourth stage of the i-th electric car ends During the time period, the i-th electric car accelerates from The i-th electric car is located at the end of the fourth stage. At, the speed is Speed ​​up to ; The fifth stage is the uniform speed driving stage. The time until the fifth stage of the i-th electric car ends During the time period, the i-th electric car from Drive at a constant speed to the entry position D of the signalized intersection cross,start ; Step 2. Construct a two-level planning model for the coordinated optimization of wireless charging station site selection and intelligent connected electric vehicle ecological driving: Step 2.

1. Construct an upper-level planning model for wireless charging interval site selection; Step 2.

2. Construct the lower-level planning model of ecological driving of intelligent connected electric vehicles; Step 3: Use the RBF-GA hybrid algorithm to solve the two-level programming model and obtain the wireless charging interval [D E,start ,D E,end ]The optimal evaluation position Z on a single lane best。 2. The collaborative optimization method according to claim 1, characterized in that: The step 2.1 comprises: Step 2.1.

1. Use formula (1) to construct the objective function F of the upper-level planning model up : (1) In formula (1), , , They are respectively the time cost coefficient, the power cost coefficient, and the number of electric vehicles passing through the wireless charging interval [D E,start ,D E,end ] construction cost coefficient, , They are the proportion of vehicle driving consumption and wireless charging area construction consumption, The exit position D of the i-th electric vehicle when it exits the single lane and reaches the signalized intersection cross,end Time, E i is the net energy consumed by the i-th electric vehicle passing through a single lane; Step 2.1.

2. Use formula (2) to construct the constraints of the upper-level planning model: (2)。 3. The collaborative optimization method according to claim 2, characterized in that: The step 2.2 comprises: Step 2.2.1: Use formula (3) to construct the objective function F of the lower-level planning model down : (3) Step 2.2.2: Use equations (4) to (24) to construct the constraints of the lower-level planning model: (4) (5) (6) (7) (8) (9) (10) (11) (12) (13) (14) (15) (16) (17) (18) (19) (20) (21) (22) (23) (24) In formula (4) to formula (24), , They respectively represent the maximum acceleration and minimum acceleration allowed for an electric vehicle when driving on a single lane; Indicates the maximum speed that an electric vehicle is allowed to travel on a single lane; , They represent the positions of the i-th electric car and the i+1-th electric car at time t, respectively. , They represent the minimum headway distance when an electric vehicle stops and the reaction time of the electric vehicle to receive signals and perform related operations. represents the speed of the i+1th electric car at time t; , They represent the i-th electric vehicle entering the wireless charging interval [D E,start ,D E,end ] and leave the wireless charging area [D E,start ,D E,end ] time; , , They represent the instantaneous power, traction power, and recovery power of the i-th electric vehicle at time t respectively; For the quality of electric vehicles, represents the acceleration of the ith electric car at time t, g is the acceleration due to gravity, 𝜃 is the inclination of the road, C r , c1, c2 are three rolling resistance parameters, ρ air is the air mass density, C D is the aerodynamic drag coefficient of the electric vehicle, A f is the front windshield area of ​​the electric vehicle, v i (t) represents the speed of the ith electric car at time t, Indicates v i (t) squared; , , They are the transmission system efficiency factor, motor efficiency factor, and battery efficiency factor. represents the recovery braking energy efficiency factor of the i-th electric vehicle at time t, 𝛼 is the parameter of the electric vehicle; , They represent the traction energy consumption and regeneration energy of the i-th electric vehicle when it is driving on a single lane, , They represent the total power consumption and total charging power of the i-th electric vehicle passing through a single lane, P charge It is the charging capacity per unit time.

4. The collaborative optimization method according to claim 1, characterized in that: The step 3 comprises: Step 3.

1. Define the maximum number of iterations as R max , the current number of iterations is n, and the maximum number of consecutive successes is suc max , the maximum number of consecutive failures is fai max , the initial step length is stp len , the size of the candidate point set is N size The upper and lower limits of the predicted score coefficient are μ max , μ min , initialize n=1; Step 3.

2. D E,start and D E,end As the decision variable of the upper-level planning model, and according to the number of decision variables of the upper-level planning model 2, in D start and D cross,start After randomly generating 2+1 initial evaluation positions, they are stored in the evaluation position set Z; each initial evaluation position is a combination of 2 decision variables; Step 3.

3. Bring any initial evaluation position z in Z into the lower-level planning model for solution to obtain the objective function value corresponding to z ;in, ;Will Bring it into the upper-level planning model for solution and obtain the objective function value corresponding to z After that, it is stored in the objective function set W; Step 3.

4. Let the minimum value in W be the optimal objective function value W best , W best The corresponding optimal evaluation position is recorded as Z best ; Step 3.

5. Based on the set Z and the set W, get the RBF parameter RBF after the nth update n,par ; Step 3.

6. According to Z best When constructing the nth iteration, a group of length N size The candidate evaluation position set P n ; Step 3.

7. RBF-based n,par , get the predicted value score corresponding to p , distance score , and thus calculate P n The comprehensive score of the candidate evaluation position p in ,in, is the weight coefficient for the nth iteration; and ,in, , are the minimum weight and maximum weight of the nth iteration respectively; Step 3.

8. According to P n The comprehensive score of each candidate evaluation position in is selected, and the candidate evaluation position corresponding to the minimum value is selected as the best evaluation position z at the nth iteration. n , and z n After depositing Z, follow the process in step 3.3 to get z n The corresponding upper-level planning model objective function is , and the updated W; Step 3.

9. Judgment Is it greater than W? best , if so, then W best and Z best Keep unchanged, let suc n =0, will fai n-1 +1 to fai n , otherwise, let Z best =z n , W best = ; will suc n-1 +1 assigned to suc n , so fai n =0; Step 3.

10. When fai n >fai max When Assign to ,Will Assign to ,Will Assign to stp len After that, determine stp len Is it greater than D E,end If so, let stp len =D E,end ; Otherwise, directly execute step 3.11, where , and Represent 3 multiples respectively, and ; ; When success n >suc max , when stp len / 2 assign value to stp len After that, if stp len <D E,start , then let stp len =D E,start ; Otherwise, go directly to step 3.11; Step 3.

11. If n=R max , then the final W is obtained best and Z best Otherwise, assign n+1 to n and return to step 3.5 to execute sequentially.

5. The collaborative optimization method according to claim 4, characterized in that: In step 3.3, the objective function value corresponding to z is obtained according to the following process. : Step 3.3.1: Let the period of the traffic light at the intersection be t c , and the red light is used as the signal light color at the beginning of each cycle, where the green light is in cycle t c The percentage of the following is t G ; After the i-th electric vehicle enters the single lane, the number of vehicles that conflict with the i-th electric vehicle is recorded as , and initialize ; Step 3.3.2: Calculate the time it takes for the i-th electric car to reach D cross,start The cycle number of the signal light ;in, Indicates rounding up; Step 3.3.3: According to The maximum number of electric vehicles that can pass within a green light period and the vehicle sequence j of the i-th electric vehicle passing through the signalized intersection in the h-th cycle i ,Will After assigning the value to h, execute step 3.3.4; where, Indicates rounding down; Step 3.3.4: Calculate the time it takes for the i-th electric car to reach D in the h-th cycle. cross,start The time limit And the time limit ;in, Indicates that the i-1th electric car arrives at D cross,start time; Step 3.3.5: Judgement Is it true? If so, the i-th electric car and the i-1-th electric car form a fleet. Assign to Then, execute step 3.3.6; otherwise, it means that the i-th electric car is the only vehicle in the fleet, and directly execute step 3.3.6; Step 3.3.6: Take the electric car in front of the i-th electric car in the fleet The car is Electric vehicles, The end time of the five-stage speed curve of each electric vehicle is used as the decision variable, and the GA algorithm is used to solve the lower-level planning model to obtain the first The combination of five decision variables for an electric vehicle And store the serial number in the electric vehicle decision variable set L At the corresponding position, Substitute into equation (3)-(24) to determine The objective function value of an electric vehicle Store the sequence number in the electric vehicle lower layer objective function value set Y The corresponding position; like , then directly execute step 3.3.8; otherwise, execute step 3.3.7; Step 3.3.7: In the convoy, except for the first vehicle, Five decision variables for the ikth electric car besides the first electric car Store it in the position corresponding to the serial number ik in L, and Substitute into equations (3) to (24) to determine the objective function value of the ikth vehicle: And store it in the position corresponding to the serial number ik in Y, ; If the time interval between the ik-1th electric vehicle and the ikth electric vehicle entering the single lane is When , the five decision variables of the ikth electric vehicle in the fleet are calculated using formula (26), including: (26) In formula (26), , , , , They represent the five decision variables of the ikth electric car, , , , , They represent the five decision variables of the ik-1th electric car; If the time interval between the ik-1th electric vehicle and the ikth electric vehicle entering the single lane is greater than When , the five decision variables of the ikth electric vehicle in the fleet are calculated, including: (27) In formula (27), , , , They represent the time point when the ik-1th electric vehicle enters the single lane and starts to decelerate uniformly, and the time point when the ikth electric vehicle enters the single lane and starts to decelerate uniformly, respectively. , They respectively represent that if the time interval between the ik-1th electric vehicle and the ikth electric vehicle entering the single lane is The estimated time to enter the single lane and start deceleration; like When , according to formula (27) we get , and use formula (28) to get the remaining four decision variables: (28) like When , five decision variables are obtained using formula (29): (29) Step 3.3.8: Determine the Electric car and Does the electric vehicle satisfy equation (13)? If so, If yes, execute step 3.3.10; otherwise, execute step 3.3.9: Step 3.3.9: Determine the Is the electric car the first vehicle to pass through the signal intersection in the hth signal cycle? If so, The shared After the electric vehicles form a fleet, return to step 3.3.6 and execute in sequence; otherwise, Assign to , will The total number of electric vehicles from the first electric vehicle to the i-th electric vehicle After the electric vehicles form a fleet, the execution returns to 3.3.6 and executes in order; Step 3.3.10: If i=I, then sum all the values ​​in Y to get Otherwise, assign i+1 to i and return to step 3.3.2 and execute in sequence.

6. The collaborative optimization method according to claim 5, characterized in that: In step 3.3.6, the GA algorithm is used to solve the lower-level planning model according to the following process; Step 3.3.6.1: Define the maximum number of iterations as G max , the current number of iterations is g, the population size is pop, initialization g=1, and the initialization step size is stp; The first car in the convoy is Serial number of the electric car Let it be u, then the optimal decision variable corresponding to the u-th electric car is , and initialize Empty; initialized The corresponding optimal objective function value is infinite; Randomly generate a set of g-generation population P with size pop that meets the constraints of equations (4)-(24) g ; and P g Each individual in represents the five decision variables of the u-th electric car; Step 3.3.6.2: P g Mutate to get the g-th generation offspring ; Step a: Randomly select P g The bth individual p g,b , ; For the bth individual p g,b The corresponding five decision variables The decision variables are Mutate within and determine whether the mutated individuals meet the constraints of formula (4)-formula (24). If they meet the constraints, add the corresponding mutated individuals to Otherwise, the corresponding mutated individuals are discarded; Indicates the decision variables that need to be varied; Step b: Judgement Whether the number of individuals in reaches pop, if so, it means that , otherwise, return to step a and execute sequentially; Step 3.3.6.4: P g and Any individual in , and calculate according to the procedure in step 3.3.7 The decision variables of each electric vehicle in the fleet are substituted into the lower-level planning model for solution, and the obtained The objective function value of the team, so as to find P g and The minimum objective function value and its corresponding individuals ; like < , then let Assign to , Assign to ;otherwise, and remain unchanged; Step 3.3.6.5: If g = G max , then the output , otherwise, execute step 3.3.6.6; Step 3.3.6.6: Join the g+1 generation population P g+1 ,like Belong to P g The individuals in From P g When deleting The individuals at the corresponding positions also If belong The individuals in from When deleting The individuals at the corresponding positions also start from P g Delete from; Step 3.3.6.7: Compare P g and The objective function values ​​corresponding to the individuals in the remaining positions in P are selected, and the individuals corresponding to the positions with the minimum objective function values ​​are added to P g+1 , thus obtaining the g+1th generation population P containing pop individuals g+1 , Step 3.3.6.8: After assigning g+1 to g, return to step 3.3.6.

2.

7. The collaborative optimization method according to claim 6, characterized in that: The step 3.6 includes: Step 3.6.

1. At the nth iteration, from Z best Select the rth decision variable d from the two decision variables r , ; Step 3.6.

2. Use equation (25) to calculate d r After mutation, the mutated rth decision variable d is obtained r,1 : (25) Step 3.6.

3. Z best Another decision variable d y With d r,1 Compare and select the larger value as D E,end , the smaller value is taken as D E,start , thus forming a candidate evaluation position p, ,and ; Step 3.6.

4. Determine whether p satisfies the constraint conditions shown in formula (2). If so, store it in P n , otherwise, discard p; Step 3.6.

5. Obtain N according to the process of steps 3.6.1 to 3.6.

4. size candidate evaluation locations.

8. An electronic device, comprising a memory and a processor, characterized in that: The memory is used to store a program that supports the processor to execute the collaborative optimization method described in any one of claims 1 to 7, and the processor is configured to execute the program stored in the memory.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the collaborative optimization method according to any one of claims 1 to 7 are executed.

Images