Charging scheduling method and device for new energy mine cards

Abstract

The invention provides a new energy mine card charging scheduling method and device, and belongs to the technical field of mine car operation charging scheduling. When a certain vehicle completes a certain operation task, the current vehicle waiting time t11 required to wait before the charging pile when the current vehicle completes the current operation task is calculated, and on the basis that the current vehicle is charged when the current operation task is completed, the current vehicle waiting time t11 is calculated by adopting a particle swarm optimization algorithm; the shortest time required for other vehicles to wait for charging in front of the charging pile is obtained and serves as the future optimal estimated waiting time t12 of other vehicles, and the weighted sum of t11 and t12 is calculated and serves as the global waiting time T1 for charging of the current vehicle when the current operation task is finished. And calculating the global waiting time T2 of charging when the next operation task of the current vehicle is finished. And T1 and T2 are compared, if T1 is smaller than T2, the current vehicle is charged when the operation task of the current trip is finished, and otherwise, charging is not carried out.

Description

Technical Field

[0001] The present invention relates to a charging scheduling method and device for a new energy mining truck, belonging to the technical field of charging scheduling for mining truck operations. Background Art

[0002] At present, in the production and transportation process of open-pit mines, new energy mining trucks are gradually being used in large quantities. However, during the continuous operation of mining trucks, as the power is consumed, the SOC of the mining trucks continues to decrease, and the mining trucks need to be charged from time to time. Due to the limited charging equipment and the randomness and uncertainty of human intervention in the scheduling plan, long queues for charging will occur when multiple vehicles work together, which seriously affects the operational efficiency of the full-process collaborative operation of multiple vehicles.

[0003] The Chinese patent document with publication number CN110675016A discloses a cloud intelligent scheduling system and charging method for an unmanned mine car transportation system based on an edge-cloud architecture. Through information interaction between intelligent driving terminals, intelligent road test equipment, an intelligent cloud platform and a wireless charging system, the intelligent cloud platform can schedule the charging of rechargeable autonomous driving mine cars in the mining area. Summary of the Invention

[0004] The purpose of the present invention is to provide a charging scheduling method and device for new energy mining trucks, which is used to solve the problem that the queuing time of new energy mining truck transport vehicles in open-pit mines is too long during charging scheduling, affecting the operating efficiency.

[0005] In order to achieve the above-mentioned object, the present invention provides a charging scheduling method for a new energy mining card, comprising the following steps:

[0006] S1. Obtain a dispatch schedule for a multi-car set during operation, including the start and end times of different cars in the multi-car set at different operations; also obtain the SOC of each car at the start of operation, the SOC consumed by each car during each operation, and the time required for each car to be fully charged at different SOCs;

[0007] S2. When each vehicle completes each operation task, obtain the usage status of the charging pile and the status of vehicles currently waiting for charging at that moment, and calculate the current vehicle waiting time t11 for charging at the end of the current operation task;

[0008] S3. Input the dispatch schedule, the SOC of each vehicle at the start of the operation, the SOC consumed by each vehicle during each operation, the time required for each vehicle to be fully charged at different SOCs, the vehicle ID, the current trip, the current usage of the charging pile, and the current status of vehicles waiting for charging into the particle swarm algorithm. Calculate the optimal future estimated waiting time t12 of other vehicles in the multi-vehicle formation if the current vehicle is charged at the end of the current trip. The optimal future estimated waiting time of other vehicles is the shortest time required for other vehicles to wait for charging at the charging pile.

[0009] S4. Calculate the weighted sum of t11 and t12 as the global waiting time T1 for the current vehicle to charge at the end of the current trip task;

[0010] S5. Calculate the current vehicle's waiting time t21 when the current vehicle is charged at the end of the next operation, and determine the optimal future estimated waiting time t22 of other vehicles in the multi-vehicle formation when the current vehicle is charged at the end of the next operation in accordance with step S3; calculate the weighted sum of t21 and t22 as the global waiting time T2 for the current vehicle to be charged at the end of the next operation;

[0011] S6. If T1 is less than T2, the current vehicle will be charged at the end of the current operation task; if T1 is not less than T2, the current vehicle will not be charged at the end of the current operation task, and will be re-judged when the next operation task is completed.

[0012] When scheduling charging for new energy mining vehicles in an open-pit mine, when a vehicle completes a particular trip, the waiting time t11 of the current vehicle that the vehicle needs to wait at the charging station to charge upon completing the current trip is calculated. Based on the current vehicle charging upon completion of the current trip, a particle swarm algorithm is used to design charging strategies for other vehicles in the multi-vehicle formation upon completion of different trips. The minimum waiting time required for other vehicles to wait at the charging station for charging is calculated as the future optimal estimated waiting time t12 for other vehicles. The weighted sum of t11 and t12 is calculated as the global waiting time T1 for the current vehicle to charge upon completion of the current trip. Using the same method, the global waiting time T2 for the current vehicle to charge upon completion of the next trip is calculated. T1 and T2 are compared. If T1 is less than T2, the current vehicle is charged upon completion of the current trip; otherwise, it is not charged and the decision is made again upon completion of the next trip. Each vehicle in a multi-vehicle formation uses this method to make judgments when each operation task is completed, so as to reasonably plan the charging timing of each vehicle in the multi-vehicle formation, avoiding the problem of vehicles waiting too long for charging at charging piles, resulting in reduced operation efficiency.

[0013] Furthermore, the fitness calculation formula used by the particle swarm algorithm in S3 is:

[0014]

[0015] Where i represents the current trip, m represents the evaluation trip, m = min (total trips minus i, N), j represents the vehicle index value, n represents the total number of vehicles currently participating in the operation, if the jth vehicle in the i-th trip chooses to charge, then I(i, j) = 1, otherwise I(i, j) = 0, γ is the attenuation factor, 0 < γ < 1, t ij It represents the waiting time for the jth vehicle in the i-th trip to choose charging, and N is the maximum number of trips.

[0016] Furthermore, in the above method, the future optimal estimated waiting time includes a penalty item for the situation where the vehicle's current SOC is insufficient to complete the next work task and no charging has been arranged; the vehicle's current SOC is calculated based on the vehicle's SOC when it starts working, the SOC consumed by the vehicle in each work task and the current trip.

[0017] If the current SOC of the vehicle is not enough to complete the next operation task, then the vehicle obviously needs to be charged. If the vehicle has not been arranged to be charged under such circumstances, it is not realistic. Therefore, a sufficiently large value is added as a penalty item to intervene in the future optimal estimated waiting time in this case to ensure that the determined charging scheduling plan meets actual needs.

[0018] Furthermore, in the above method, as the number of trips increases, the future optimal estimated waiting time is attenuated by a discount factor; the greater the number of trips, the smaller the discount factor.

[0019] As the number of trips increases, the uncertainty of vehicle scheduling may increase due to factors such as road conditions and weather. The greater the number of trips, the greater the uncertainty. Therefore, the future optimal estimated waiting time is attenuated through a discount factor.

[0020] Furthermore, in the above method, the relationship between the number of trips and the discount factor is represented by an exponential function, the discount factor is the dependent variable of the exponential function, the number of trips is the independent variable of the exponential function, and the base of the exponential function is greater than 0 and less than 1.

[0021] The relationship between the number of trips and the discount factor satisfies the exponential function form, and the future optimal estimated waiting time is decayed in an exponential decay manner.

[0022] Furthermore, the discount factor is:

[0023] C=soc_consume*1.5-soc

[0024] Among them, C is the discount factor, soc represents the soc of the vehicles involved in the operation, and soc_consume represents the soc consumed in one trip.

[0025] Furthermore, in the above method, when calculating the optimal estimated waiting time in the future, the future 10 trips are used as the maximum number of trips.

[0026] When calculating the optimal estimated waiting time in the future, the waiting time of at most the next 10 trips is calculated, which can reduce the amount of calculation.

[0027] Furthermore, in the above method, in S2, the current SOC of the current vehicle at the end of the current trip's work task is calculated based on the SOC of the vehicle when it starts working, the SOC consumed by the vehicle in each work task and the current trip. If the current SOC of the vehicle is lower than the set minimum threshold, the current vehicle directly chooses to charge at the end of the current trip; if the current SOC of the vehicle is greater than the set maximum threshold, the current vehicle directly chooses not to charge at the end of the current trip; the set minimum threshold is less than the set maximum threshold.

[0028] Taking into account the current SOC of the vehicle and the charging timing, the particle swarm algorithm is constrained as follows: if the current SOC of the vehicle is greater than the set maximum threshold, the vehicle does not need to be charged; if the current SOC of the vehicle is less than the set minimum threshold, the vehicle must be charged. This judgment can reduce the processing load of the particle swarm algorithm and improve the computational efficiency of the model.

[0029] The present invention also provides a charging scheduling device for a new energy mining card, comprising a memory and a processor, wherein the processor executes instructions in the memory to implement the above-mentioned charging scheduling method for the new energy mining card. BRIEF DESCRIPTION OF THE DRAWINGS

[0030] Figure 1 This is a flowchart of a charging scheduling method for a new energy mining card in an embodiment of the method of the present invention;

[0031] Figure 2 This is a flow chart of the particle swarm algorithm in an embodiment of the method of the present invention;

[0032] Figure 3 A schematic diagram of charging scheduling planning in an embodiment of the method of the present invention;

[0033] Figure 4 The figure is a structural diagram of a charging scheduling device for a new energy mining card in an embodiment of the present invention. DETAILED DESCRIPTION

[0034] In order to make the objectives, technical solutions and advantages of the present invention more clear, the present invention is further described in detail below with reference to the accompanying drawings and embodiments.

[0035] Method Example:

[0036] like Figure 1 As shown, a charging scheduling method for a new energy mining card of the present invention includes the following steps:

[0037] 1) Obtain the dispatch schedule for multi-vehicle formations. The dispatch schedule provides the start and end times of each vehicle for each mission. Each vehicle in the multi-vehicle formation performs its mission according to the start and end times of the corresponding mission in the dispatch schedule. The vehicles involved in the dispatch are free of abnormalities such as failure to charge or vehicle failure.

[0038] 2) Obtain the SOC of each vehicle before the first operation as the starting SOC of the corresponding vehicle.

[0039] 3) When a vehicle completes a mission, its vehicle ID and current mission number are obtained. Based on the vehicle ID and current mission number, the vehicle's end time for the mission can be retrieved from the schedule. Based on the vehicle's starting SOC, current mission number, and the SOC consumed by the vehicle during each mission, the vehicle's current SOC at the end of the mission can be obtained.

[0040] By collecting historical data and applying mathematical statistics, we can determine the SOC required for each vehicle during each mission. Through experimental testing, we can determine the time required for each vehicle to fully charge at different SOCs.

[0041] 4) When a vehicle completes a trip, it also obtains the current usage status of each charging station and the current SOC of vehicles currently waiting to charge. The current usage status of each charging station includes: whether each station is charging, the correspondence between the charging station and the charging vehicle, and the remaining charging time of the currently charging vehicle. The correspondence between charging stations and charging vehicles is as follows: charging station m is charging vehicle n, and there are k vehicles waiting to charge after vehicle n. The charging stations are not experiencing any anomalies such as failures or power outages.

[0042] If there are idle charging piles, that is, charging piles that are not charging any vehicles, the current vehicle waiting time t11 for charging at the end of the current operation task is 0. If all charging piles are charging vehicles, the remaining charging time of the current charging vehicle can be obtained based on the current SOC of the current charging vehicle and the time required for each vehicle to be fully charged at different SOCs. Based on the current SOC of the vehicle currently waiting to be charged, the time required to fully charge each waiting vehicle can be calculated. Based on the remaining charging time of the current charging vehicle and the time required to fully charge each waiting vehicle, the current vehicle waiting time t11 for charging at the end of the current operation task can be calculated.

[0043] 5) When a vehicle completes a mission, the scheduling schedule, the starting SOC of different vehicles, the SOC consumed by different vehicles in each mission, the time required for different vehicles to be fully charged at different SOCs, the vehicle ID, the current trip, the current charging pile usage, and the current SOC of the vehicle currently waiting for charging are input as follows: Figure 2 The particle swarm algorithm shown in the figure performs optimal charging strategy planning and calculates the shortest time required for other vehicles to wait at the charging pile for charging when the current vehicle is charged at the end of the current trip. This shortest time is used as the future optimal estimated waiting time t12 of other vehicles in the multi-vehicle formation.

[0044] Fitness calculation formula: i represents the current trip, m represents the evaluation trip, m = min (total trips minus i, 10), j represents the vehicle index value, n represents the total number of vehicles currently participating in the operation, excluding vehicles (waiting for charging, charging, and returning to the parking space after charging and waiting for operation). If the jth vehicle in the i-th trip chooses to charge, then I(i, j) = 1, otherwise I(i, j) = 0, γ is the attenuation factor, 0 < γ < 1, the default value is 0.8, t ij represents the waiting time for the jth vehicle in the i-th trip to choose charging;

[0045] Combination of fitness and particle swarm optimization

[0046] np.random.randint(i,m,size=(POP_SIZE,len(current_turn_trucks)))

[0047] i represents the current trip, m represents the evaluation trip, m = min (total trips minus i, N), pop_size represents the population size, current_turn_trucks represents the number of vehicles participating in the operation, N is the maximum number of trips, which can be 10;

[0048] Particle swarm velocity and position update formula:

[0049] v=w*v+c1*p1*(pbest-pcurr)+c2*p2*(gbest-pcurr);

[0050] pcurr=pcurr+v;

[0051] w represents the inertia factor, v represents the speed, c1 represents the individual learning factor, p1 and p2 represent the discount factors, a random number of (0, 1), pbest represents the individual optimal experience, pcurr represents the current position, c2 represents the social learning factor, and gbest represents the group optimal experience.

[0052] 6) Calculate the weighted sum of t11 and t12 as the global waiting time T1 for the current vehicle to charge at the end of the current trip task.

[0053] 7) Calculate the current vehicle waiting time t21, which corresponds to the situation where the current vehicle is charging at the end of the next operation. When the vehicle completes the next operation, the usage of each charging pile and the current SOC of the vehicle currently waiting for charging may change. In this case, the method for determining t11 in step 4 is used to determine t21. The method for determining t12 in step 5 is used to plan the optimal charging strategy for the vehicle when charging at the end of the next operation. Adjust the current operation and calculate the optimal estimated future waiting time t22 of other vehicles in the multi-vehicle formation when the vehicle is charging at the end of the next operation. The weighted sum of t21 and t22 is calculated as the global waiting time T2 for the current vehicle to charge at the end of the next operation.

[0054] 8) Using a greedy strategy, compare the global waiting time T1 and the global waiting time T2. If T1 is less than T2, the current vehicle will be charged at the end of the mission, otherwise it will not be charged at the end of the mission.

[0055] 9) If the vehicle is not charged at the end of the mission, then when the vehicle completes the next mission, repeat the above steps 1-8 to determine whether the vehicle is charged.

[0056] By adopting the present invention, the expected number of trips for vehicle charging can be estimated, and then a charging scheduling plan diagram for each vehicle can be established; when the vehicle completes the operation tasks of the expected number of trips, it is scheduled to charge. In this embodiment, the multi-vehicle formation includes 5 vehicles. As Figure 3 shown, the SOC of vehicle No. 2 at the start of operation is 88%, and the SOC consumed by vehicle No. 2 in each operation task is 10%. At the end of the 3rd operation task, the current SOC of vehicle No. 2 is 58%, and it is charged, and it takes 60 minutes to fully charge. Similarly, it can be seen from Figure 3 that vehicle No. 1 is charged at the end of the 6th operation task, vehicle No. 3 and vehicle No. 4 are charged at the end of the 2nd operation task, and vehicle No. 5 is charged at the end of the 5th operation task.

[0057] The global waiting time is equal to the weighted sum of the charging waiting time of the current vehicle and the future optimal estimated waiting time of other vehicles. Considering the situation that the current SOC of other vehicles in the multi-vehicle formation is not sufficient to complete the next operation task, a penalty term for the situation where the remaining SOC is not sufficient to complete a trip and has not been scheduled for charging is added to the future optimal estimated waiting time.

[0058] Considering that there are various uncertainties in the future, as the number of trips increases, the error in estimating the waiting time gradually increases. Therefore, a discount factor is introduced to attenuate the weight of the future optimal estimated waiting time in different trips. In this embodiment, the attenuation method is exponential attenuation, that is, the relationship between the number of trips and the discount factor is represented by an exponential function. The discount factor is the dependent variable of this exponential function, the number of trips is the independent variable of this exponential function, and the base of this exponential function is greater than 0 and less than 1.

[0059] Discount factor: if soc < soc_consume * 1.5, then C = soc_consume * 1.5 - soc

[0060] soc represents the soc of the vehicle participating in operation, soc_consume represents the soc consumed in one trip, and C represents the discount factor. If its SOC is not enough to run one trip when arranging charging for this vehicle, it means the arrangement is very unreasonable and needs to be punished, and the later the arrangement is, the greater the punishment. Multiplying by 1.5 means not waiting until the soc is not enough to run one trip to start punishing, but punishing in advance to avoid the situation where the soc is really not enough to run one operation trip.

[0061] When the value of the future optimal estimated waiting time is attenuated by the method of exponential reduction, in order to reduce the calculation amount, the waiting time for the future 10 trips is calculated at most.

[0062] In addition, according to the actual situation of the vehicle, charging is not required when the vehicle's current SOC is high, but charging is required when the vehicle's current SOC is low. Therefore, setting a maximum SOC threshold and a minimum SOC threshold, directly choosing not to charge when the SOC is above the maximum threshold, and directly choosing to charge when the SOC is below the minimum threshold, can improve the calculation efficiency of the model.

[0063] Device Example:

[0064] The present invention provides a charging scheduling device for a new energy mining card, such as Figure 4 As shown, the system includes memory, a processor, and an internal bus. The processor and memory exchange data and communicate with each other via the internal bus. The processor can be a microprocessor (MCU), a programmable logic device (FPGA), or other processing device. The memory can be any type of memory that stores information electrically, such as RAM and ROM; any type of memory that stores information magnetically, such as hard disks, floppy disks, magnetic tapes, magnetic core memory, bubble memory, and USB flash drives; any type of memory that stores information optically, such as CDs and DVDs; and other types of memory, such as quantum memory and graphene memory.

[0065] The processor executes the instructions in the memory to implement a charging scheduling method for a new energy mining card in the method embodiment. The implementation of this method has been clearly introduced in the method embodiment and will not be repeated here.

Claims

1. A charging scheduling method for a new energy mining card, characterized in that: The steps include: S1. Obtain a dispatch schedule for a multi-car set during operation, including the start and end times of different cars in the multi-car set at different operations; also obtain the SOC of each car at the start of operation, the SOC consumed by each car during each operation, and the time required for each car to be fully charged at different SOCs; S2. When each vehicle completes each operation task, obtain the usage status of the charging pile and the status of vehicles currently waiting for charging at that moment, and calculate the current vehicle waiting time t11 for charging at the end of the current operation task; S3. Input the dispatch schedule, the SOC of each vehicle at the start of the operation, the SOC consumed by each vehicle during each operation, the time required for each vehicle to be fully charged at different SOCs, the vehicle ID, the current trip, the current usage of the charging pile, and the current status of vehicles waiting for charging into the particle swarm algorithm. Calculate the optimal future estimated waiting time t12 of other vehicles in the multi-vehicle formation if the current vehicle is charged at the end of the current trip. The optimal future estimated waiting time of other vehicles is the shortest time required for other vehicles to wait for charging at the charging pile. S4. Calculate the weighted sum of t11 and t12 as the global waiting time T1 for the current vehicle to charge at the end of the current trip task; S5. Calculate the current vehicle's waiting time t21 when the current vehicle is charged at the end of the next operation, and determine the optimal future estimated waiting time t22 of other vehicles in the multi-vehicle formation when the current vehicle is charged at the end of the next operation in accordance with step S3; calculate the weighted sum of t21 and t22 as the global waiting time T2 for the current vehicle to be charged at the end of the next operation; S6. If T1 is less than T2, the current vehicle will be charged at the end of the current operation task; if T1 is not less than T2, the current vehicle will not be charged at the end of the current operation task, and will be re-judged when the next operation task is completed.

2. The charging scheduling method for new energy mining trucks according to claim 1, characterized in that: The fitness calculation formula used by the particle swarm algorithm in S3 is: Where i represents the current trip, m represents the evaluation trip, m = min (total trips minus i, N), j represents the vehicle index value, n represents the total number of vehicles currently participating in the operation, if the jth vehicle in the i-th trip chooses to charge, then I(i, j) = 1, otherwise I(i, j) = 0, γ is the attenuation factor, 0 < γ < 1, t ij It represents the waiting time for the jth vehicle in the i-th trip to choose charging, and N is the maximum number of trips.

3. The charging scheduling method for a new energy mining card according to claim 1 or 2, characterized in that: The optimal estimated waiting time in the future includes a penalty item for the situation where the vehicle's current SOC is insufficient to complete the next operation task and no charging has been arranged; the vehicle's current SOC is calculated based on the vehicle's SOC when it starts working, the SOC consumed by the vehicle in each operation task, and the current trip.

4. The charging scheduling method for the new energy mining card according to claim 3 is characterized in that: As the number of trips increases, the future optimal estimated waiting time is attenuated by a discount factor; the larger the number of trips, the smaller the discount factor.

5. The charging scheduling method for new energy mining trucks according to claim 4, characterized in that: The relationship between the number of trips and the discount factor is expressed by an exponential function, where the discount factor is the dependent variable of the exponential function, the number of trips is the independent variable of the exponential function, and the base of the exponential function is greater than 0 and less than 1.

6. The charging scheduling method for new energy mining trucks according to claim 4, characterized in that: The discount factor is: C=soc_consume*1.5-soc Among them, C is the discount factor, soc represents the soc of the vehicles involved in the operation, and soc_consume represents the soc consumed in one trip.

7. The charging scheduling method for new energy mining trucks according to claim 4, characterized in that: When calculating the optimal estimated waiting time in the future, the maximum number of trips is 10 in the future.

8. The charging scheduling method for new energy mining trucks according to claim 1, characterized in that: In S2, the current SOC of the vehicle at the end of the current trip is calculated based on the SOC of the vehicle when it starts working, the SOC consumed by the vehicle in each working task, and the current trip. If the current SOC of the vehicle is lower than the set minimum threshold, the current vehicle is directly charged at the end of the current trip. If the vehicle's current SOC is greater than the set maximum threshold, the current vehicle will directly choose not to charge at the end of the current trip; the set minimum threshold is less than the set maximum threshold.

9. A charging scheduling device for a new energy mining card, characterized in that: The method comprises a memory and a processor, wherein the processor executes instructions in the memory to implement the charging scheduling method for the new energy mining card according to any one of claims 1 to 8.

Images