Urban rail train energy-saving operation strategy online optimization method based on reinforcement learning

Abstract

The invention discloses an urban rail train energy-saving operation strategy online optimization method based on reinforcement learning, and relates to the technical field of urban rail train control, and the method comprises the steps: obtaining the historical operation data of an urban rail train and the operation route information of the urban rail train; and analyzing and processing according to the historical operation data of the urban rail train and the related position information of the operation route, and determining a ramp acceleration position and a curve deceleration position. The method comprises the following steps: analyzing speed data of historical operation of an urban rail train to determine station position information, ramp position information and curve position information of an operation route; the historical running speed data of the urban rail train is analyzed and processed through the ramp position information of the running route and the curve position information of the running route, the ramp acceleration position and the curve deceleration position are determined, and the problem that energy is wasted due to the fact that ramp acceleration and curve deceleration are too fast is avoided. Energy-saving control of the urban rail train in the running process is achieved, and the running cost is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of urban rail train control, and in particular to an online optimization method for an urban rail train energy-saving operation strategy based on reinforcement learning. Background Art

[0002] Urban rail trains, also known as inter-city rail services, refer to passenger trains that travel between adjacent important cities or city clusters. Generally, the total running distance is short, the ride time is short, the number of cities passed is small, and there are no sleeper cars. Intercity trains refer to medium- and short-distance passenger trains that only serve a few specific cities. They are mainly used to strengthen the connection between nearby cities and facilitate cross-city travel and cultural exchanges between surrounding areas. Although intercity trains are not as widely used as long-distance trains, with the large-scale rise of many city clusters in my country and the large-scale construction of a large number of intercity railways, the status of intercity trains in railway transportation is increasing.

[0003] Urban rail trains will pass through slopes and curves during operation. When the urban rail train passes through a slope, if it accelerates too fast, it will cause energy waste. When the urban rail train passes through a curve, if it decelerates too fast, it will also cause energy waste when it speeds up again to return to its normal operating speed. Therefore, in order to achieve energy saving of urban rail trains, an online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning is proposed. Summary of the invention

[0004] In order to solve the above technical problems, an online optimization method for the energy-saving operation strategy of urban rail trains based on reinforcement learning is provided. This technical solution solves the problem proposed in the above background technology that the urban rail train will pass through ramps and curves during operation. When the urban rail train passes through a ramp, if it accelerates too fast, it will cause energy waste. When the urban rail train passes through a curve, it decelerates too fast, and when the urban rail train re-accelerates to restore to the normal operating speed, it will also cause energy waste.

[0005] In order to achieve the above purpose, the technical solution adopted by the present invention is: An online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning, comprising: Obtain historical operation data of urban rail trains and operation route information of urban rail trains; Analyze and process the running route information of the urban rail train to obtain relevant position information of the running route, wherein the relevant position information of the running route specifically includes station position information of the running route, ramp position information of the running route, and curve position information of the running route; Analyze and process the historical operation data of urban rail trains and the relevant position information of the operation route to determine the acceleration position on the slope and the deceleration position on the curve; According to the station location information, ramp acceleration position and curve deceleration position of the running route, the speed adjustment instruction is issued to the urban rail train to change the running speed of the urban rail train.

[0006] Preferably, the step of analyzing and processing the running route information of the urban rail train to obtain the relevant position information of the running route specifically includes the following steps: Perform data extraction and processing based on the historical operation data of the urban rail train to obtain the historical operation speed data of the urban rail train; Compare and judge the historical speed data of urban rail trains to determine the station location information of the running route; Extract and process the historical running speed data of the urban rail train to obtain the speed data of a single completed running of the urban rail train; Perform difference calculation on the speed data of a single completed run of the urban rail train to obtain the speed change of the urban rail train; The speed change of the urban rail train is compared and judged to determine the slope position information and the curve position information of the running route.

[0007] Preferably, the comparing and judging the historical running speed data of the urban rail train to determine the station location information of the running route specifically comprises the following steps: Data extraction and processing are performed based on the historical running speed data of the urban rail train to obtain the zero speed data of the urban rail train; The running route information of the urban rail train is positioned and processed according to the zero speed data of the urban rail train to determine the stop position of the urban rail train; Counting and processing the historical running speed data of the urban rail train according to the stop position of the urban rail train and the zero speed data of the urban rail train, and obtaining the stop number information corresponding to the stop position of the urban rail train; Determine and process the stop number information corresponding to the stop position of the urban rail train and the set stop threshold; If the number of stops is greater than the set stop threshold, the stop location of the urban rail train is the station location information of the running route; If the number of stops is less than or equal to the set stop threshold, the stop position of the urban rail train is a temporary stop position.

[0008] Preferably, the comparing and judging process of the speed change of the urban rail train to determine the ramp position information and the curve position information of the running route specifically includes the following steps: The speed change of the urban rail train and the first speed setting threshold and the second speed setting threshold are judged and processed respectively; If the speed change of the urban rail train is greater than or equal to the first speed setting threshold, positioning the running route information of the urban rail train is performed according to the speed change of the urban rail train to determine the slope position information of the running route; If the speed change of the urban rail train is less than the first speed setting threshold, and the speed change of the urban rail train is greater than the second speed setting threshold, the urban rail train accelerates normally; If the speed change of the urban rail train is less than the second speed setting threshold, and the speed change of the previous urban rail train is not greater than the first speed setting threshold, determining the curve position information of the running route; If the speed change of the urban rail train is less than the second speed setting threshold, and the speed change of the previous urban rail train is greater than the first speed setting threshold, the ramp position information of the running route is determined.

[0009] Preferably, the analyzing and processing of the historical running data of the urban rail train and the relevant position information of the running route to determine the ramp acceleration position and the curve deceleration position specifically comprises the following steps: Classify and process the historical running speed data of urban rail trains using the number of completed runs of urban rail trains as a feature, and obtain the single completed running speed data of several groups of urban rail trains; Analyze and process the speed data of several groups of urban rail trains that have completed a single run to determine the ramp acceleration position; The speed data of several groups of urban rail trains that have completed a single run are analyzed and processed to determine the deceleration position on the curve.

[0010] Preferably, the analyzing and processing of the speed data of a single completed operation of a plurality of groups of urban rail trains to determine the ramp acceleration position specifically comprises the following steps: Extracting and processing speed data of a plurality of groups of urban rail trains that have completed a single run according to the ramp position information of the running route to determine a plurality of groups of ramp speed data; Sorting and processing are performed according to several groups of ramp speed data to obtain the minimum ramp speed; Extracting and processing the speed data of a single completed run of a plurality of groups of urban rail trains according to the minimum ramp speed, and obtaining the speed data of a single completed run of the urban rail trains corresponding to the minimum ramp speed; The acceleration position is positioned according to the speed data of a single completed run of the urban rail train corresponding to the minimum ramp speed to determine the ramp acceleration position.

[0011] Preferably, the specific calculation formula for obtaining the minimum ramp speed is:

[0012] In the formula, is the minimum ramp speed; are several groups of ramp speed data; n is the specific number of several groups of ramp speed data; min() is the minimum function.

[0013] Preferably, the analyzing and processing of the speed data of a single completed operation of a plurality of groups of urban rail trains to determine the deceleration position of the curve specifically comprises the following steps: Extracting and processing the speed data of several groups of urban rail trains that have completed a single run according to the curve position information of the running route, and obtaining several groups of curve speed data; Sorting and processing are performed according to a plurality of groups of curve speed data to obtain a maximum curve speed; Extracting and processing the speed data of a single completed run of a plurality of groups of urban rail trains according to the maximum curve speed, and obtaining the speed data of a single completed run of the urban rail train corresponding to the maximum curve speed; The deceleration position is positioned according to the speed data of a single completed run of the urban rail train corresponding to the maximum curve speed to determine the curve deceleration position.

[0014] Preferably, the specific calculation formula for obtaining the maximum curve speed is:

[0015] In the formula, is the maximum cornering speed; are several groups of curve speed data; m is the specific number of several groups of curve speed data; max() is the maximum value function.

[0016] Preferably, the issuing of a speed adjustment instruction to the urban rail train according to the station location information, the ramp acceleration position and the curve deceleration position of the running route to change the running speed of the urban rail train specifically comprises the following steps: Get the real-time position and speed of the urban rail train; The real-time position of the urban rail train is calculated and processed with the station position information, ramp acceleration position, and curve deceleration position of the running route to obtain the distance to the station, the distance to the ramp acceleration position, and the distance to the curve deceleration position; The real-time speed of the urban rail train is calculated and processed with the arrival station distance, the arrival ramp acceleration position distance, and the arrival curve deceleration position distance, and the arrival station deceleration information, ramp acceleration information, and curve deceleration information are obtained; According to the deceleration information at the arrival station, the acceleration information on the slope and the deceleration information on the curve, the speed adjustment instruction is issued to the urban rail train to change the running speed of the urban rail train.

[0017] Compared with the prior art, the present invention provides an online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning, which has the following beneficial effects: The present invention first analyzes the historical running speed data of the urban rail train to determine the station location information of the running route, the ramp location information of the running route and the curve location information of the running route. Then, the historical running speed data of the urban rail train is analyzed and processed according to the ramp location information of the running route and the curve location information of the running route to determine the ramp acceleration position and the curve deceleration position, thereby avoiding the problem of wasting energy due to excessive ramp acceleration and curve deceleration, realizing energy-saving control of the urban rail train during operation and reducing the operation cost. BRIEF DESCRIPTION OF THE DRAWINGS

[0018] Figure 1 A schematic flow chart of steps S100-S400 in an online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning proposed by the present invention; Figure 2 A schematic flow chart of steps S201-S205 in an online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning proposed by the present invention; Figure 3 A flowchart of steps S2021-S2026 in an online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning proposed by the present invention; Figure 4 A schematic flow chart of steps S2051-S2055 in an online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning proposed by the present invention; Figure 5 A schematic flow chart of steps S301-S303 in an online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning proposed by the present invention; Figure 6 A schematic flow chart of steps S3021-S3024 in an online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning proposed by the present invention; Figure 7 A flow chart of steps S3031-S3034 in an online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning proposed by the present invention; Figure 8 This is a flow chart of steps S401-S404 in an online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning proposed by the present invention. DETAILED DESCRIPTION

[0019] The following description is used to disclose the present invention so that those skilled in the art can implement the present invention. The preferred embodiments described below are only examples, and those skilled in the art may think of other obvious variations.

[0020] Reference Figure 1As shown, an online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning includes: S100, obtaining historical operation data of urban rail trains and operation route information of urban rail trains; S200, analyzing and processing the running route information of the urban rail train to obtain relevant position information of the running route, wherein the relevant position information of the running route specifically includes station position information of the running route, ramp position information of the running route, and curve position information of the running route; S300, analyzing and processing the historical operation data of the urban rail train and the relevant position information of the operation route to determine the ramp acceleration position and the curve deceleration position; S400, issuing a speed adjustment instruction to the urban rail train according to the station location information, the ramp acceleration position and the curve deceleration position of the running route, so as to change the running speed of the urban rail train; Those skilled in the art can understand that, during the operation, the urban rail train will pass through ramps and curves. When the urban rail train passes through a ramp, it needs to speed up, and a large amount of energy will be consumed in the process of speeding up the urban rail train, because the urban rail train will travel on a flat road after passing the ramp, so it needs to slow down. Therefore, if the urban rail train speeds up too quickly after passing the ramp and then slows down, it will cause a waste of energy. In addition, the urban rail train needs to slow down when passing through a curve. When the urban rail train turns from a curve to a flat road, it needs to speed up. If the urban rail train slows down too quickly at the curve and then speeds up again, it will also cause a waste of energy. Therefore, the historical operation data of the urban rail train is analyzed to determine the optimal acceleration position on the ramp and the optimal deceleration position on the curve to reduce energy loss.

[0021] Reference Figure 2 As shown, analyzing and processing the running route information of the urban rail train to obtain the relevant location information of the running route specifically includes the following steps: S201, extracting and processing data according to historical operation data of urban rail trains to obtain historical operation speed data of urban rail trains; S202, comparing and judging the historical running speed data of the urban rail train to determine the station location information of the running route; S203, extracting and processing the historical running speed data of the urban rail train to obtain the speed data of a single completed running of the urban rail train; S204, performing difference calculation on the speed data of a single completed run of the urban rail train to obtain a speed change of the urban rail train; S205: Compare and judge the speed change of the urban rail train to determine the slope position information and the curve position information of the running route.

[0022] Reference Figure 3As shown, the historical running speed data of the urban rail train is compared and judged to determine the station location information of the running route, which specifically includes the following steps: S2021. Perform data extraction and processing based on historical running speed data of urban rail trains to obtain zero speed data of urban rail trains; S2022, positioning the running route information of the urban rail train according to the zero speed data of the urban rail train, and determining the stop position of the urban rail train; S2023, counting and processing the historical running speed data of the urban rail train according to the stop position of the urban rail train and the zero speed data of the urban rail train, and obtaining the stop number information corresponding to the stop position of the urban rail train; S2024, determining and processing the stop number information corresponding to the stop position of the urban rail train and the set stop threshold; S2025. If the number of stops is greater than the set stop threshold, the stop position of the urban rail train is the station position information of the running route; S2026: If the number of stops is less than or equal to the set stop threshold, the stop position of the urban rail train is a temporary stop position; In this embodiment, when the speed of the urban rail train is zero, it is the location where the urban rail train stops, but the location where the urban rail train stops is not necessarily the station location. Therefore, the number of times the speed of the urban rail train is zero at the same location, that is, the number of stops, is judged to determine the station location. Because the speed of the urban rail train is zero, it may be that the road is damaged and the urban rail train cannot run, or the urban rail train has an abnormal situation and stops urgently. Therefore, in order to exclude the above situation, the number of stops is judged.

[0023] Reference Figure 4 As shown, the comparison and judgment processing of the speed change of the urban rail train to determine the slope position information and the curve position information of the running route specifically includes the following steps: S2051, respectively judging and processing the speed change of the urban rail train and the first speed setting threshold and the second speed setting threshold; S2052: If the speed change of the urban rail train is greater than or equal to the first speed setting threshold, positioning the running route information of the urban rail train according to the speed change of the urban rail train to determine the ramp position information of the running route; S2053: If the speed change of the urban rail train is less than the first speed setting threshold, and the speed change of the urban rail train is greater than the second speed setting threshold, the urban rail train accelerates normally; S2054: If the speed change of the urban rail train is less than the second speed setting threshold, and the speed change of the previous urban rail train is not greater than the first speed setting threshold, determine the curve position information of the running route; S2055: If the speed change of the urban rail train is less than the second speed setting threshold, and the speed change of the previous urban rail train is greater than the first speed setting threshold, determine the ramp position information of the running route; In this embodiment, when the urban rail train passes through a slope, it needs to speed up, and when passing through a curve, it needs to slow down. Therefore, it can be judged by the speed change of the urban rail train whether the urban rail train passes through a curve or a slope. It is worth noting that when the urban rail train speeds up and passes through a slope, it needs to slow down. Therefore, through secondary judgment, it is determined whether the urban rail train passes through a curve or a slope to avoid misjudgment.

[0024] Reference Figure 5 As shown, according to the historical operation data of the urban rail train and the relevant position information of the operation route, the following steps are specifically included to determine the ramp acceleration position and the curve deceleration position: S301, classifying and processing the historical running speed data of urban rail trains based on the number of completed runs of urban rail trains as a feature, and obtaining the single completed running speed data of several groups of urban rail trains; S302, analyzing and processing the speed data of a single completed operation of a plurality of groups of urban rail trains to determine the ramp acceleration position; S303: Analyze and process the speed data of a single completed operation of a plurality of groups of urban rail trains to determine the deceleration position of the curve.

[0025] Reference Figure 6 As shown, analyzing and processing the speed data of a single completed operation of several groups of urban rail trains to determine the ramp acceleration position specifically includes the following steps: S3021, extracting speed data of several groups of speed data of a single completed run of urban rail trains according to the ramp position information of the running route, and determining several groups of ramp speed data; S3022, performing sorting processing according to a plurality of groups of ramp speed data to obtain a minimum ramp speed; S3023, extracting and processing the speed data of a single completed run of a plurality of groups of urban rail trains according to the minimum ramp speed, and obtaining the speed data of a single completed run of the urban rail trains corresponding to the minimum ramp speed; S3024, performing acceleration position positioning processing according to the speed data of a single completed operation of the urban rail train corresponding to the minimum ramp speed, and determining the ramp acceleration position; Among them, the specific calculation formula for obtaining the minimum ramp speed is:

[0026] In the formula, is the minimum ramp speed; are several groups of ramp speed data; n is the specific number of several groups of ramp speed data; min() is the minimum value function; In this embodiment, the historical running speed data of the urban rail train is screened according to the ramp position information of the running route to determine the speed of the urban rail train when passing the ramp, and then the speed passing the ramp is screened to select the minimum ramp speed. This is because the urban rail train needs to slow down when entering a flat road after passing the ramp. In order to avoid excessive deceleration, the historical running speed data of the urban rail train is screened to determine the minimum ramp speed, which indirectly reduces energy loss and also reduces the running cost of the urban rail train.

[0027] Reference Figure 7 As shown, analyzing and processing the speed data of a single completed operation of several groups of urban rail trains to determine the deceleration position of the curve specifically includes the following steps: S3031, extracting speed data of several groups of speed data of a single completed run of urban rail trains according to the curve position information of the running route, and obtaining several groups of curve speed data; S3032, performing sorting processing according to a plurality of groups of curve speed data to obtain a maximum curve speed; S3033, extracting and processing the speed data of a single completed run of a plurality of groups of urban rail trains according to the maximum curve speed, and obtaining the speed data of a single completed run of the urban rail train corresponding to the maximum curve speed; S3034, performing deceleration position positioning processing according to the speed data of a single completed run of the urban rail train corresponding to the maximum curve speed, and determining the curve deceleration position; Among them, the specific calculation formula for obtaining the maximum cornering speed is:

[0028] In the formula, is the maximum cornering speed; are several groups of curve speed data; m is the specific number of several groups of curve speed data; max() is the maximum value function; In this embodiment, the historical speed data of the urban rail train is screened according to the curve position information of the running route to determine the speed of the urban rail train when passing through the curve, and then the speed through the curve is screened to select the maximum curve speed. This is because the urban rail train needs to slow down when entering the curve, and needs to speed up when entering the flat road after passing the curve. In order to avoid excessive speed increase, the historical speed data of the urban rail train is screened to determine the maximum curve speed, which indirectly reduces energy loss and also reduces the operating cost of the urban rail train.

[0029] Reference Figure 8As shown, according to the station location information, ramp acceleration position and curve deceleration position of the running route, a speed adjustment instruction is issued to the urban rail train, and changing the running speed of the urban rail train specifically includes the following steps: S401, obtaining the real-time position and real-time speed of the urban rail train; S402, the real-time position of the urban rail train is respectively calculated and processed with the station position information, the ramp acceleration position, and the curve deceleration position of the running route to obtain the distance to the station, the distance to the ramp acceleration position, and the distance to the curve deceleration position; S403, the real-time speed of the urban rail train is calculated and processed with the arrival station distance, the arrival ramp acceleration position distance, and the arrival curve deceleration position distance, to obtain the arrival station deceleration information, ramp acceleration information, and curve deceleration information; S404, issuing a speed adjustment instruction to the urban rail train according to the deceleration information at the arrival station, the acceleration information on the ramp, and the deceleration information on the curve, so as to change the running speed of the urban rail train; In this embodiment, in order to make the urban rail train stop at the station accurately, the real-time speed and real-time position of the station and the urban rail train are calculated and analyzed, and the deceleration information of the urban rail train is determined, so that the speed of the urban rail train is exactly zero when it arrives at the station. In addition, when the urban rail train needs to pass through a ramp or a curve, in order to avoid excessive waste of energy, the real-time speed, real-time position, ramp acceleration position, and curve deceleration position of the urban rail train are calculated to determine how long it will take for the urban rail train to speed up or decelerate, so that the urban rail train passes the ramp or curve at the minimum ramp speed or the maximum curve speed, thereby reducing energy loss and reducing the operating cost of the urban rail train.

[0030] The above shows and describes the basic principles, main features and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The above embodiments and descriptions only describe the principles of the present invention. The present invention may be subject to various changes and improvements without departing from the spirit and scope of the present invention. These changes and improvements fall within the scope of the present invention. The scope of protection claimed by the present invention is defined by the attached claims and their equivalents.

Claims

1. An online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning, characterized in that: include: Obtain historical operation data of urban rail trains and operation route information of urban rail trains; Analyze and process the running route information of the urban rail train to obtain relevant position information of the running route, wherein the relevant position information of the running route specifically includes station position information of the running route, ramp position information of the running route, and curve position information of the running route; Analyze and process the historical operation data of urban rail trains and the relevant position information of the operation route to determine the acceleration position on the slope and the deceleration position on the curve; According to the station location information, ramp acceleration position and curve deceleration position of the running route, the speed adjustment instruction is issued to the urban rail train to change the running speed of the urban rail train.

2. According to claim 1, a reinforcement learning-based online optimization method for energy-saving operation strategy of urban rail trains is characterized in that: The step of analyzing and processing the running route information of the urban rail train to obtain the relevant position information of the running route specifically includes the following steps: Perform data extraction and processing based on the historical operation data of the urban rail train to obtain the historical operation speed data of the urban rail train; Compare and judge the historical speed data of urban rail trains to determine the station location information of the running route; Extract and process the historical running speed data of the urban rail train to obtain the speed data of a single completed running of the urban rail train; Perform difference calculation on the speed data of a single completed run of the urban rail train to obtain the speed change of the urban rail train; The speed change of the urban rail train is compared and judged to determine the slope position information and the curve position information of the running route.

3. According to claim 2, a reinforcement learning-based online optimization method for energy-saving operation strategy of urban rail trains is characterized in that: The process of comparing and judging the historical running speed data of the urban rail train to determine the station location information of the running route specifically includes the following steps: Data extraction and processing are performed based on the historical running speed data of the urban rail train to obtain the zero speed data of the urban rail train; The running route information of the urban rail train is positioned and processed according to the zero speed data of the urban rail train to determine the stop position of the urban rail train; Counting and processing the historical running speed data of the urban rail train according to the stop position of the urban rail train and the zero speed data of the urban rail train, and obtaining the stop number information corresponding to the stop position of the urban rail train; Determine and process the stop number information corresponding to the stop position of the urban rail train and the set stop threshold; If the number of stops is greater than the set stop threshold, the stop location of the urban rail train is the station location information of the running route; If the number of stops is less than or equal to the set stop threshold, the stop position of the urban rail train is a temporary stop position.

4. According to claim 2, a reinforcement learning-based online optimization method for energy-saving operation strategy of urban rail trains is characterized in that: The process of comparing and judging the speed change of the urban rail train to determine the ramp position information and the curve position information of the running route specifically includes the following steps: The speed change of the urban rail train and the first speed setting threshold and the second speed setting threshold are judged and processed respectively; If the speed change of the urban rail train is greater than or equal to the first speed setting threshold, positioning the running route information of the urban rail train is processed according to the speed change of the urban rail train to determine the slope position information of the running route; If the speed change of the urban rail train is less than the first speed setting threshold, and the speed change of the urban rail train is greater than the second speed setting threshold, the urban rail train accelerates normally; If the speed change of the urban rail train is less than the second speed setting threshold, and the speed change of the previous urban rail train is not greater than the first speed setting threshold, determining the curve position information of the running route; If the speed change of the urban rail train is less than the second speed setting threshold, and the speed change of the previous urban rail train is greater than the first speed setting threshold, the ramp position information of the running route is determined.

5. According to claim 1, a reinforcement learning-based online optimization method for energy-saving operation strategy of urban rail trains is characterized in that: The analysis and processing of the historical operation data of the urban rail train and the relevant position information of the operation route to determine the ramp acceleration position and the curve deceleration position specifically includes the following steps: Classify and process the historical running speed data of urban rail trains using the number of completed runs of urban rail trains as a feature, and obtain the single completed running speed data of several groups of urban rail trains; Analyze and process the speed data of several groups of urban rail trains that have completed a single run to determine the ramp acceleration position; The speed data of several groups of urban rail trains that have completed a single run are analyzed and processed to determine the deceleration position on the curve.

6. The online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning according to claim 5 is characterized in that: The analyzing and processing of the speed data of a single completed operation of a plurality of groups of urban rail trains to determine the ramp acceleration position specifically comprises the following steps: Extracting and processing speed data of a plurality of groups of urban rail trains that have completed a single run according to the ramp position information of the running route to determine a plurality of groups of ramp speed data; Sorting and processing are performed according to several groups of ramp speed data to obtain the minimum ramp speed; Extracting and processing the speed data of a single completed run of a plurality of groups of urban rail trains according to the minimum ramp speed, and obtaining the speed data of a single completed run of the urban rail trains corresponding to the minimum ramp speed; The acceleration position is positioned according to the speed data of a single completed run of the urban rail train corresponding to the minimum ramp speed to determine the ramp acceleration position.

7. The online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning according to claim 6 is characterized in that: The specific calculation formula for obtaining the minimum ramp speed is: In the formula, is the minimum ramp speed; are several groups of ramp speed data; n is the specific number of several groups of ramp speed data; min() is the minimum function.

8. The online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning according to claim 5 is characterized in that: The analyzing and processing of the speed data of a single completed operation of a plurality of groups of urban rail trains to determine the deceleration position of the curve specifically comprises the following steps: Extracting and processing the speed data of several groups of urban rail trains that have completed a single run according to the curve position information of the running route, and obtaining several groups of curve speed data; Sorting and processing are performed according to a plurality of groups of curve speed data to obtain a maximum curve speed; Extracting and processing the speed data of a single completed run of a plurality of groups of urban rail trains according to the maximum curve speed, and obtaining the speed data of a single completed run of the urban rail train corresponding to the maximum curve speed; The deceleration position is positioned according to the speed data of a single completed run of the urban rail train corresponding to the maximum curve speed to determine the curve deceleration position.

9. The online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning according to claim 8 is characterized in that: The specific calculation formula for obtaining the maximum curve speed is: In the formula, is the maximum cornering speed; are several groups of curve speed data; m is the specific number of several groups of curve speed data; max() is the maximum value function.

10. The online optimization method for energy-saving operation strategy of urban rail trains based on reinforcement learning according to claim 1 is characterized in that: The method of issuing a speed adjustment instruction to the urban rail train according to the station location information, the ramp acceleration position and the curve deceleration position of the running route to change the running speed of the urban rail train specifically includes the following steps: Get the real-time position and speed of the urban rail train; The real-time position of the urban rail train is calculated and processed with the station position information, ramp acceleration position, and curve deceleration position of the running route to obtain the distance to the station, the distance to the ramp acceleration position, and the distance to the curve deceleration position; The real-time speed of the urban rail train is calculated and processed with the arrival station distance, the arrival ramp acceleration position distance, and the arrival curve deceleration position distance, and the arrival station deceleration information, ramp acceleration information, and curve deceleration information are obtained; According to the deceleration information at the arrival station, the acceleration information on the slope and the deceleration information on the curve, the speed adjustment instruction is issued to the urban rail train to change the running speed of the urban rail train.

Images