Simulation Analysis and Evaluation Method for the Scheduling Algorithm of Hydrogen Fuel Cell Bus

Abstract

The present invention provides a simulation analysis and evaluation method for the scheduling algorithm of hydrogen fuel cell bus, which realizes the simulation operation analysis of the intelligent algorithm for hydrogen fuel cell bus based on vehicle operation big data. It comprehensively considers the actual operation environment and operation characteristics of hydrogen fuel cell bus, classifies and processes categories such as vehicle driving, operation timeliness, and hydrogen refueling behavior, and calculates the standard operation reference index values. By building a simulation environment and various disturbance simulation events in combination with the scheduling events during the operation process, it can effectively verify the scheduling adjustment ability of the intelligent algorithm in the face of different actual situations. The method of the present application overcomes the disadvantages of long time consumption and low efficiency in the verification of existing bus vehicle scheduling strategies, has extremely high applicability to different intelligent scheduling algorithms, and is beneficial to assisting relevant operation units to make timely scheduling decisions for the overall hydrogen fuel buses and hydrogen energy systems.

Description

Technical Field

[0001] The present invention belongs to the technical field of operation management of hydrogen fuel cell bus, and particularly relates to a method for simulating and analyzing the effects of relevant operation algorithms by using big data of hydrogen fuel cell buses. Background Art

[0002] For hydrogen fuel cell buses, their actual operation process involves various related factors such as bus depots, vehicles, drivers, routes, and traffic. Existing technologies rarely involve the research on the dispatching of this new type of new energy vehicle. Especially for the intelligent dispatching of hydrogen fuel cell buses, there is still a large technical gap at present. Before a mature intelligent dispatching method or algorithm for hydrogen fuel cell buses is put into practical application, comprehensive and long-term verification work is needed to test its effectiveness. However, the verification work has the problem of high cost, which will inevitably cause a certain obstacle to the popularization of bus intelligent dispatching based on the hydrogen energy system. At the same time, when using the traditional manual operation mode to evaluate the dispatching efficiency of the algorithm, it is also difficult to comprehensively measure the advantages and disadvantages of the algorithm in coordinating and managing multiple resources. Therefore, there is an urgent need in this field for a scientific analysis method for the operation algorithm of hydrogen fuel cell buses. On the one hand, it can reduce the trial threshold of the algorithm intelligent dispatching applied by the operation company as much as possible. On the other hand, it can comprehensively simulate the algorithm application of the hydrogen energy bus dispatching strategy and further verify the effectiveness of the algorithm. Summary of the Invention

[0003] In view of this, aiming at the above technical problems existing in this field, the present invention provides a simulation analysis and evaluation method for the dispatching algorithm of hydrogen fuel cell buses, which specifically includes the following steps:

[0004] Step 1: Comprehensively collect the static information of the hydrogen fuel cell buses actually put into operation currently, the static data and scheduling data of the actual bus routes, and the vehicle working data, and use these collected data as the basic data;

[0005] Step 2: Based on the data collected in Step 1, form the standard operation reference index values of hydrogen fuel cell buses, including performing the following operations on the hydrogen fuel cell buses actually put into operation currently: statistically obtaining the driving time and driving duration reference index values on different intervals included in the corresponding routes every day; matching the vehicle driving trajectories with the corresponding bus routes, and obtaining the shifts of the routes and their departure and arrival times; calculating the daily minimum hydrogen fuel pressure and average hydrogen consumption reference index values according to the hydrogen fuel pressure changes during vehicle driving; obtaining the daily alarm probability reference index values according to the alarm items occurring in the vehicle within a certain period;

[0006] Step 3: Build a simulation environment for operation scheduling and simulate operation disturbances, including performing the following operations on the hydrogen fuel cell buses actually put into operation currently: statistically analyze the passenger flow within a certain period to obtain the passenger flow distribution at different stations on each line for a single day, which is used for simulating passenger flow distribution for trips; set different levels of delays and corresponding scheduling execution rules according to the planned arrival time and actual arrival time of different trips on each line, which is used for simulating the occurrence of delay events; set the rules for stopping hydrogen refueling during vehicle operation, which is used for simulating hydrogen refueling behavior planning; set the rules for fault occurrence and vehicle stopping instructions according to the probability reference index value, which is used for simulating vehicle fault occurrence.

[0007] Step 4: Input the static information of the buses used for simulation and the static data of bus lines into the intelligent algorithm to be verified. The intelligent algorithm calculates the real-time scheduling plan. Perform intelligent operation scheduling according to the scheduling plan in the simulation environment. At the same time, based on the passenger flow distribution simulation for trips, delay event occurrence simulation, hydrogen refueling behavior planning simulation, and vehicle fault occurrence simulation obtained in Step 3, adjust the vehicle usage status and the arrival time at each station in the real-time scheduling plan, and let the intelligent algorithm recalculate and output the scheduling data corresponding to the adjusted real-time scheduling plan.

[0008] Step 5: Compare the scheduling plan and scheduling data obtained in Step 4 based on each reference index value in Step 2 to evaluate the scheduling effect of the intelligent algorithm; and compare the scheduling plan and scheduling data corresponding to the actual operation scheduling strategy of the bus operation company with the scheduling plan and scheduling data obtained by the intelligent algorithm to reflect the advantages and disadvantages of the two.

[0009] Further, the static information of the hydrogen fuel cell buses collected in Step 1 specifically includes: license plate number, internal vehicle number, frame number, name of the assigned driver, and number of assigned drivers.

[0010] The static data of the actual bus lines specifically includes: line name, line outbound path, line inbound path, round-trip / single-trip line, line station sequence, line station number, line station location, name of the line stop yard, line stop yard location. The scheduling data specifically includes: departure trips, planned departure time, actual departure time at each station, planned arrival time at each station, actual arrival time at each station, planned vehicle license plate number, actual vehicle license plate number, planned driver, actual driver, operating line name, departure yard, arrival yard, number of people getting on at each station, number of people getting off at each station.

[0011] The vehicle working data specifically includes: vehicle positioning, hydrogen fuel pressure value, fault alarm occurrence time, alarm content, fault end time, and alarm level collected based on the national standard 32960 new energy vehicle communication protocol.

[0012] Further, the driving time and driving duration reference index values of each day on the corresponding line in step 2 specifically include: defining the distance between every two stations on each line as an interval, counting the driving time of the vehicle in each interval through the driving records of each vehicle, and calculating the average driving duration within every 30-minute range of each interval based on the data of the previous 7 days as the driving duration reference index value;

[0013] Obtaining the shifts of the line and their departure and arrival times is specifically based on comparing the daily driving trajectories of the vehicles on the line with the line path. With a matching range within 100 meters, the number of times the vehicle conforms to the trajectory each time is found, which is the number of shifts of the vehicle operating on the line on that day; according to the starting trajectory time and the ending trajectory time of entering different intervals, it is judged as the departure time and arrival time of the shift;

[0014] The minimum hydrogen fuel pressure is specifically obtained by calculating the median of the hydrogen pressure values when the hydrogen fuel cell bus enters the hydrogen refueling station within the previous 7 days; the average hydrogen consumption reference index value is obtained by calculating the average hydrogen consumption of the vehicle based on the driving mileage and the change in hydrogen pressure of the same vehicle entering the hydrogen refueling station twice within the previous 7 days;

[0015] The probability reference index value is specifically calculated by dividing the alarm values of all vehicles on the selected line in the previous 7 days by 7 to obtain the average value.

[0016] Further, in step 3, the passenger flow distribution is specifically based on collecting the passenger transport data of the operating shifts of the selected line in the previous 7 days to obtain the number of passengers getting on and off at each station on the line within 7 days; setting the simulated passenger flow rule that the passenger flow is inversely proportional to the departure interval, and the arrival time of passengers at the station follows a negative exponential distribution (Poisson distribution), and setting the constraint condition that the number of passengers carried by the vehicle cannot exceed the limit number, thereby obtaining the corresponding distributed passenger flow distribution travel simulation;

[0017] A late arrival event is defined as occurring when the planned arrival time of each shift is more than 10 minutes different from the actual arrival time. Specifically, based on the planned arrival time and the actual arrival time of different shifts on each line, a single shift vehicle being late is defined as a minor late arrival, and more than one shift vehicle being late is defined as a large interval late arrival; according to the actual late arrival situation of the shifts, adjust the algorithm scheduling execution situation in the simulation environment, that is, extend the actual arrival time of the shifts in the same time period to simulate the occurrence of late arrival events;

[0018] The determination of the hydrogen refueling suspension rule is specifically based on calculating the downward trend of the hydrogen fuel pressure based on the mileage. When the pressure drops to or below the minimum hydrogen fuel pressure reference index value, the vehicle is arranged to be suspended for hydrogen refueling;

[0019] The vehicle suspension instruction rule specifically randomly generates a fault event in the operation dispatching according to the probability reference index value, and generates the corresponding vehicle suspension instruction based on the fault.

[0020] Furthermore, in step 4, the intelligent algorithm calculates the real-time scheduling plan and outputs the corresponding departure schedules, planned departure times for each station, planned arrival times for each station, planned vehicle license plate numbers, planned drivers, operating routes, departure stations, and arrival stations. Based on the passenger flow distribution travel simulation, late event occurrence simulation, hydrogen refueling behavior planning simulation, and vehicle failure occurrence simulation obtained in step 3, the real-time scheduling plan is adjusted and the following data is output for comparative analysis: planned schedules, actual schedules, planned departure times for each schedule, actual departure times for each schedule, planned arrival times for each schedule, actual arrival times for each schedule, planned vehicle license plate numbers, actual vehicle license plate numbers, planned drivers, actual drivers, and actual numbers of passengers getting on and off at each station.

[0021] Furthermore, in step 5, the scheduling effect of the intelligent algorithm is evaluated specifically based on the number of scheduling shifts for a certain route, the number of line vehicles, the average full load rate, the average passenger waiting time, and the average driver working hours index, and the passenger satisfaction index and the line vehicle carrying rate index are set to adjust the execution process of the intelligent algorithm.

[0022] The above-mentioned simulation analysis and evaluation method for the hydrogen fuel cell bus scheduling algorithm provided by the present invention realizes the operation simulation analysis of the intelligent algorithm for hydrogen fuel cell buses based on vehicle operation big data. It comprehensively considers the actual operation environment and operation characteristics of hydrogen fuel cell buses, classifies and processes categories such as vehicle driving, operation timeliness, and hydrogen refueling behavior, and calculates the standard operation reference index values. By building a simulation environment and various disturbance simulation events in combination with the scheduling events in the operation process, it can effectively verify the scheduling adjustment ability of the intelligent algorithm in the face of different actual situations. The method of this application overcomes the disadvantages of long time consumption and low efficiency in verifying the existing bus vehicle scheduling strategies, has extremely high applicability to different intelligent scheduling algorithms, and is conducive to assisting relevant operation units to make timely scheduling decisions for the entire hydrogen fuel buses and hydrogen energy system. BRIEF DESCRIPTION OF THE DRAWINGS

[0023] Figure 1 It is a flowchart of the method provided by the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0024] The technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art without creative efforts based on the embodiments of the present invention fall within the protection scope of the present invention.

[0025] The simulation analysis and evaluation method for the hydrogen fuel cell bus scheduling algorithm provided by the present invention, as Figure 1 shown, specifically includes the following steps:

[0026] Step 1: Comprehensively collect the static information of the hydrogen fuel cell bus currently in actual operation, the static data and scheduling data of the actual bus routes, and the vehicle operation data, and use these collected data as the basic data.

[0027] Step 2: Based on the data collected in Step 1, form the standard operation reference index values of the hydrogen fuel cell bus, including performing the following on the hydrogen fuel cell bus currently in actual operation: statistically obtaining the driving time and driving duration reference index values on different intervals included in the corresponding route every day; matching the corresponding bus route according to the vehicle driving trajectory, and obtaining each shift of the route and its departure and arrival times; calculating the daily minimum hydrogen fuel pressure and average hydrogen consumption reference index values according to the change of hydrogen fuel pressure during vehicle driving; obtaining the daily alarm probability reference index value according to the alarm items that occurred within a certain period of time for the vehicle.

[0028] Step 3: Build a simulation environment for operation scheduling and simulate operation disturbances, including performing the following on the hydrogen fuel cell bus currently in actual operation: statistically obtaining the passenger flow situation within a certain period of time to get the passenger flow distribution at different stations on each route for passenger flow distribution travel simulation; setting different late arrival levels and corresponding scheduling execution rules according to the planned arrival time and actual arrival time of different shifts on each route for late arrival event occurrence simulation; setting the refueling stop rules during vehicle driving for refueling behavior planning simulation; setting the fault occurrence and vehicle stop instruction rules according to the probability reference index value for vehicle fault occurrence simulation.

[0029] Step 4: Input the static information of the bus used for simulation and the static data of the bus route into the intelligent algorithm to be verified, and the intelligent algorithm calculates to obtain the real-time scheduling plan; perform intelligent operation scheduling according to the scheduling plan in the simulation environment, and at the same time, based on the passenger flow distribution travel simulation, late arrival event occurrence simulation, refueling behavior planning simulation, and vehicle fault occurrence simulation obtained in Step 3, adjust the vehicle usage status and the time to reach each station in the real-time scheduling plan, and the intelligent algorithm recalculates and outputs the scheduling data corresponding to the adjusted real-time scheduling plan.

[0030] Step 5: Compare the scheduling plan and scheduling data obtained in Step 4 based on each reference index value in Step 2 to evaluate the scheduling effect of the intelligent algorithm; and compare the scheduling plan and scheduling data corresponding to the actual operation scheduling strategy of the bus operation company with the scheduling plan and scheduling data obtained by the intelligent algorithm to reflect the advantages and disadvantages of the two.

[0031] In a preferred embodiment of the present invention, the static information of the hydrogen fuel cell bus collected in step 1 specifically includes: license plate number, internal vehicle number, vehicle identification number, configured driver name, and the number of configured drivers;

[0032] The static data of the actual bus line specifically includes: line name, outbound path of the line, inbound path of the line, round-trip / single-trip line, sequence of line stops, line stop numbers, line stop positioning, name of the line stop yard, line stop yard positioning. The scheduling data specifically includes: departure frequency, planned departure time, actual departure time at each stop, planned arrival time at each stop, actual arrival time at each stop, planned vehicle license plate number, actual vehicle license plate number, planned driver, actual driver, name of the operating line, departure yard, arrival yard, number of people getting on at each stop, number of people getting off at each stop;

[0033] The vehicle working data specifically includes: vehicle positioning collected based on the national standard 32960 new energy vehicle communication protocol, hydrogen fuel pressure value, fault alarm occurrence time, alarm content, fault end time, and alarm level.

[0034] In a preferred embodiment of the present invention, the reference index values of the driving time and driving duration on different intervals included in the corresponding line every day in step 2 specifically include: defining the distance between every two stops on each line as an interval, statistically calculating the driving time of the vehicle in each interval through the driving records of each vehicle, and calculating the average driving duration within every 30 minutes range of each interval based on the data of the previous 7 days as the reference index value of the driving duration;

[0035] Obtaining the shifts of the line and their departure and arrival times is specifically based on the comparison between the daily driving trajectories of the vehicles on the line and the line path. With a matching range within 100 meters, the number of times the vehicle conforms to the trajectory each time is found, which is the number of shifts of the vehicle operating this line on the current day; based on the starting trajectory time and the ending trajectory time of entering different intervals, it is judged as the departure time and arrival time of the shift;

[0036] The lowest hydrogen fuel pressure is specifically obtained by calculating the median of the hydrogen pressure values when the hydrogen fuel cell bus enters the hydrogen refueling station within the previous 7 days; the reference index value of the average hydrogen consumption is obtained by calculating the average hydrogen consumption of the vehicle based on the driving mileage and the change in hydrogen pressure of the same vehicle entering the hydrogen refueling station twice within the previous 7 days;

[0037] The probability reference index value is specifically calculated by dividing the alarm values of all vehicles on the selected line in the previous 7 days by 7 days to obtain the average value.

[0038] In a preferred embodiment of the present invention, in step 3, the passenger flow distribution is specifically collected based on the passenger transport data of the operating shifts in the previous 7 days of the selected route, and the number of passengers getting on and off at each station of the route within 7 days is obtained; the simulated passenger flow rule is set that the passenger flow is inversely proportional to the departure interval, and the arrival time of passengers at the station follows a negative exponential distribution (Poisson distribution), and the constraint condition is set that the number of passengers carried by the vehicle cannot exceed the limit number, and thus the corresponding distributed passenger flow distribution travel simulation is obtained.

[0039] The late arrival event is defined as occurring when the planned arrival time of each shift is more than 10 minutes different from the actual arrival time. Specifically, based on the planned arrival time and the actual arrival time of different shifts on each route, a single shift vehicle being late is defined as a minor late arrival, and more than one shift vehicle being late is defined as a large interval late arrival; according to the actual late arrival situation of the shifts, the algorithm scheduling execution situation in the simulation environment is adjusted, that is, the actual arrival time of the shifts in the same time period is extended to simulate the occurrence of the late arrival event.

[0040] The determination of the hydrogen addition rule for vehicle deactivation is specifically based on calculating the downward trend of the hydrogen fuel pressure based on the mileage. When the pressure drops to or below the minimum hydrogen fuel pressure reference index value, the vehicle is arranged to be deactivated for hydrogen addition.

[0041] The vehicle deactivation instruction rule specifically randomly generates a fault event in the operation scheduling according to the probability reference index value, and generates the corresponding vehicle deactivation instruction according to the fault.

[0042] In a preferred embodiment of the present invention, in step 4, the intelligent algorithm calculates to obtain the real-time scheduling plan, and outputs the corresponding departure shifts, the planned departure time of each station, the planned arrival time of each station, the planned vehicle license plate number, the planned driver, the operation route, the departure station, and the arrival station; based on the passenger flow distribution travel simulation, the simulation of the occurrence of the late arrival event, the simulation of the hydrogen addition behavior planning, and the simulation of the occurrence of vehicle faults obtained in step 3, after adjusting the real-time scheduling plan, the following data are output for comparative analysis: the planned shifts, the actual shifts, the planned departure time of each shift, the actual departure time of each shift, the planned arrival time of each shift, the actual arrival time of each shift, the planned vehicle license plate number, the actual vehicle license plate number, the planned driver, the actual driver, and the actual number of passengers getting on and off at each station.

[0043] In a preferred embodiment of the present invention, in step 5, the scheduling effect of the intelligent algorithm is specifically evaluated based on the number of scheduling shifts of a certain route, the number of route vehicles, the average full load rate, the average waiting time of passengers, and the average working hours of drivers, and the passenger satisfaction index and the vehicle carrying rate index of the route are set to adjust the execution process of the intelligent algorithm.

[0044] It should be understood that the sequence numbers of the steps in the embodiments of the present invention do not indicate the order of execution, and the execution order of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present invention.

[0045] Although the embodiments of the present invention have been shown and described, those of ordinary skill in the art can understand that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention. The scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. A simulation analysis and evaluation method for a hydrogen fuel cell bus dispatching algorithm, characterized by: The specific steps include: Step 1: Comprehensively collect static information of hydrogen fuel cell buses currently in operation, static data and scheduling data of actual bus routes, and vehicle operation data, and use these collected data as basic data; Step 2: Based on the data collected in step 1, standard operating reference index values for hydrogen fuel cell buses are formed, including the following steps for hydrogen fuel cell buses currently in operation: statistically analyzing the driving time and driving duration reference index values for different sections of the corresponding routes each day; matching the corresponding bus routes based on the vehicle's driving trajectory, and obtaining the frequency of each bus and its departure and arrival times; calculating the daily minimum hydrogen fuel pressure and average hydrogen consumption reference index values based on the changes in hydrogen fuel pressure during vehicle driving; and obtaining the daily probability reference index value of alarm occurrence based on the alarm items that occur on the vehicle within a certain period; Step 3: Build a simulation environment for operation scheduling and operation disturbance simulation, including executing the following for hydrogen fuel cell buses currently in operation: Count the passenger flow within a certain period to obtain the passenger flow distribution at different stations on each route on a single day, which is used for passenger flow distribution travel simulation; Set different delay levels and corresponding scheduling execution rules based on the planned arrival time and actual arrival time of different flights on each route, which is used for delay event simulation; Set the deactivation and hydrogenation rules during vehicle driving, which is used for hydrogenation behavior planning simulation; Set the fault occurrence and vehicle deactivation instruction rules based on the probability reference index value, which is used for vehicle fault occurrence simulation; Step 4: Input the static information of the bus used for simulation and the static data of the bus route into the intelligent algorithm to be verified, and calculate the real-time scheduling plan by the intelligent algorithm; perform intelligent operation scheduling according to the scheduling plan in the simulation environment, and adjust the vehicle usage status and arrival time of each station in the real-time scheduling plan based on the passenger flow distribution travel simulation, delay event simulation, hydrogen refueling behavior planning simulation, and vehicle failure simulation obtained in Step 3. The intelligent algorithm recalculates and outputs the scheduling data corresponding to the adjusted real-time scheduling plan; Step 5: Compare the scheduling plan and scheduling data obtained in step 4 based on the reference index values in step 2 to evaluate the scheduling effect of the intelligent algorithm; The scheduling plans and scheduling data corresponding to the actual operation scheduling strategy of the bus operating company are compared with the scheduling plans and scheduling data obtained by the intelligent algorithm to reflect the advantages and disadvantages of the two.

2. The method according to claim 1, wherein: The static information of the hydrogen fuel cell bus collected in step 1 specifically includes: license plate number, internal vehicle number, frame number, assigned driver name, and assigned driver number; The static data of actual bus routes specifically include: route name, route outbound path, route return path, round-trip / one-way route, route station sequence, route station number, route station location, route stop station name, route stop station location; scheduling data specifically include: departure frequency, planned departure time, actual departure time of each stop, planned arrival time of each stop, actual arrival time of each stop, planned vehicle license plate number, actual vehicle license plate number, planned driver, actual driver, route name, departure stop, arrival stop, number of passengers boarding at each stop, number of passengers getting off at each stop; The vehicle working data specifically includes: vehicle positioning, hydrogen fuel pressure value, fault alarm occurrence time, alarm content, fault end time, and alarm level collected based on the national standard 32960 new energy vehicle communication protocol.

3. The method according to claim 1, wherein: The calculation of the driving time and driving duration reference index values for each day in different sections included in the corresponding route in step 2 specifically includes: defining the distance between each two stations on each route as a section, calculating the driving time of each vehicle in each section based on the driving records of each vehicle, and calculating the average driving time of each section within a 30-minute range based on the data of the previous 7 days as the driving duration reference index value; The routes and their departure and arrival times are obtained by comparing the daily driving trajectories of the vehicles on the route with the route path. The number of times the vehicle matches the trajectory within a 100-meter range is found, which is the number of times the vehicle runs on the route on that day. The departure and arrival times of the routes are determined based on the time of entering the starting trajectory of different sections and the time of entering the ending trajectory. The minimum hydrogen fuel pressure is calculated based on the median hydrogen pressure of hydrogen fuel cell buses entering the hydrogen refueling station within the previous 7 days. The average hydrogen consumption reference index value is obtained by calculating the average hydrogen consumption of the vehicle based on the mileage and hydrogen pressure changes of the same vehicle entering the hydrogen refueling station twice within the previous 7 days. The probability reference index value is calculated based on the average value of the various alarm values of all vehicles on the selected route in the previous 7 days, divided by 7 days.

4. The method according to claim 1, wherein: In step 3, the passenger flow distribution is specifically based on the passenger data collected from the previous 7 days of the selected route, and the number of passengers getting on and off at each station on the route within 7 days is obtained; the simulation passenger flow rule is set as the passenger flow is inversely proportional to the departure interval, and the passenger arrival time follows the negative exponential Poisson distribution, and the constraint condition is set as the number of passengers on a vehicle cannot exceed the limit, thereby obtaining the corresponding distributed passenger flow distribution travel simulation; A delay event is defined as a difference of more than 10 minutes between the planned arrival time and the actual arrival time of each shift. Based on the planned arrival time and actual arrival time of different shifts on each line, a delay of a single shift is defined as a minor delay, while delays of more than one shift are defined as a large delay. Based on the actual delays of shifts, the algorithm scheduling execution in the simulation environment is adjusted, that is, the actual arrival time of shifts in the same time period is extended to simulate the occurrence of delay events. The determination of the deactivation rule for hydrogen refueling is specifically based on the downward trend of hydrogen fuel pressure calculated by mileage. When the pressure drops to or below the minimum hydrogen fuel pressure reference index value, the vehicle is arranged to be deactivated for hydrogen refueling. The vehicle deactivation instruction rule specifically randomly generates fault events in operation scheduling based on the probability reference index value, and generates corresponding vehicle deactivation instructions based on the fault.

5. The method according to claim 1, wherein: In step 4, the intelligent algorithm calculates the real-time scheduling plan and outputs the corresponding departure schedule, the planned departure time of each station, the planned arrival time of each station, the planned vehicle license plate number, the planned driver, the operation route, the departure station, and the arrival station; Based on the passenger flow distribution travel simulation, delay event simulation, hydrogen refueling behavior planning simulation, and vehicle failure simulation obtained in step 3, adjust the real-time scheduling plan and output the following data for comparative analysis: planned shifts, actual shifts, planned departure time of each shift, actual departure time of each shift, planned arrival time of each shift, actual arrival time of each shift, planned vehicle license plate number, actual vehicle license plate number, planned driver, actual driver, and the actual number of passengers getting on and off at each station.

6. The method according to claim 1, wherein: In step 5, the scheduling effect of the intelligent algorithm is evaluated based on the number of scheduled shifts, number of line vehicles, average full load rate, average passenger waiting time, and average driver working time indicators of a certain line, and the passenger satisfaction index and line vehicle load rate index are set to adjust the execution process of the intelligent algorithm.

Images