Dynamic operation method of urban rail transit train in abnormal weather

Abstract

The invention provides an urban rail transit train dynamic operation method under abnormal weather, and the method specifically comprises the steps: constructing plan nodes for stopping a train station, and setting the plan operation time between adjacent plan nodes; according to the weather data, obtaining a road condition abnormal subarea on the train running path; calculating predicted operation time between the adjacent plan nodes according to the road condition abnormal subareas; correcting the corresponding planned operation time based on the predicted operation time; and adjusting the train operation based on the corrected planned operation time. According to the method, the predicted running time of the train in the section is calculated by determining the abnormal road condition section influenced by the abnormal weather, so that the planned running time in the train plan is corrected, and the problem that the train running time is prolonged due to the deviation between the actual running time and the predicted running time of the train is solved; the running efficiency of the train is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of urban rail transit trains, and in particular to a method for dynamically operating urban rail transit trains under abnormal weather conditions. Background Art

[0002] In the urban rail transit signal system, the automatic train monitoring system ATS is responsible for adjusting train operations according to the plan diagram. When the train is running in the interval, ATS can calculate the optimal running speed of the train at different locations in the current area, and then calculate the running time and send it to the train, and the train will run accordingly. At the same time, ATS will send this plan to the ground passenger information system PIS, and passengers can check the train's expected arrival information based on PIS to arrange their travel plans reasonably.

[0003] Under normal circumstances, trains will run according to the train plan issued by ATS, and the actual running time of the train is consistent with the estimated running time in the train plan. However, in abnormal weather such as rain and snow, in order to ensure running safety, the running speed of the train will inevitably be reduced, which will cause the actual running time of the train to deviate from the estimated running time in the train plan, thereby extending the overall running time of the train, causing more serious delays and lower train running efficiency. Summary of the invention

[0004] The purpose of the present invention is to overcome the shortcomings of the prior art that when facing abnormal weather such as rain and snow, the actual running time of the train deviates from the estimated running time in the train plan, the overall running time of the train will be extended, the driving delay is serious, and the train operation efficiency is low. A dynamic operation method for urban rail transit trains under abnormal weather is provided, which takes into account the impact of abnormal weather on train operation, determines the abnormal road condition zones affected by the abnormal weather, and calculates the estimated running time of the train in the zone according to the abnormal road condition zones, thereby correcting the planned running time in the train plan, reducing the problem of extended train running time caused by the deviation between the actual train running time and the estimated running time, and improving the running efficiency of the train.

[0005] The purpose of the present invention is achieved through the following technical solutions:

[0006] The method for dynamic operation of urban rail transit trains under abnormal weather conditions includes:

[0007] Construct the planned nodes for train stops and set the planned running time between adjacent planned nodes;

[0008] Obtain abnormal road condition zones on the train running route based on weather data;

[0009] Calculate the estimated running time between adjacent planning nodes based on abnormal road conditions;

[0010] Correct the corresponding planned operation time based on the estimated operation time;

[0011] Adjust train operations based on the revised planned operating time.

[0012] Under the influence of abnormal weather, the running speed of the train will be reduced. Then, for the nodes where the train stops, the actual arrival time or actual departure time of the train will deviate from the train plan. If the operation of the train at the nodes where the train stops is still controlled according to the train plan, this deviation will continue to widen, greatly increasing the overall running time of the train and making the train delay more serious. Therefore, the abnormal road condition partition under the influence of abnormal weather is obtained to determine the estimated running time of the train after the running speed is reduced due to abnormal weather in the abnormal road condition partition, so as to correct the planned running time of the train plan according to the estimated running time, reduce the deviation between the actual arrival time or actual departure time and the train plan, so as to facilitate the adjustment of the train operation, and reduce the overall running time of the train and reduce driving delays under the premise of ensuring safety.

[0013] Furthermore, the step of obtaining abnormal road condition zones on the train running path according to weather data includes:

[0014] Obtain historical environmental change data and train operation data, and extract characteristic information including actual abnormal weather conditions, road conditions in the train operation path area, train operation speed, train stop protection distance and train type;

[0015] Construct a neural network model, take feature information as input, and actual road condition abnormality partitions under different weather conditions as output, and train the neural network;

[0016] The current weather data is input into the trained neural network to obtain the abnormal road condition partitions on the train running path.

[0017] Constructing a neural network model to use historical data to learn the train speed patterns under various circumstances means that the neural network model can provide more accurate predictions of abnormal road conditions by taking into account a variety of influencing factors, such as rainfall, snowfall, water and snow accumulation, early and late departures, train types, etc. The neural network model can also be updated and optimized based on the actual added data. It is based on real-time data for prediction and can quickly adapt to the current actual situation. In the face of emergencies, such as heavy snowfall, typhoon landing, etc., it can adjust the prediction results in time to ensure the accuracy of the prediction of abnormal road conditions, and thus ensure the accuracy of the calculation of the subsequent estimated running time.

[0018] Furthermore, the calculation of the estimated running time between adjacent planned nodes according to the abnormal road condition partitions includes:

[0019] Obtain all abnormal traffic partitions between adjacent planning nodes, and obtain the speed limit conditions of each abnormal traffic partition;

[0020] Calculate the estimated operation sub-time of each abnormal road condition partition based on the corresponding speed limit condition and the area length of each abnormal path partition;

[0021] The estimated running sub-times of all abnormal traffic partitions between adjacent planning nodes are summed to obtain the estimated running time between adjacent nodes.

[0022] The distance between adjacent planning nodes is generally not too long, and the weather conditions between adjacent planning nodes are similar. Once there is an abnormal road condition partition, it can be regarded as abnormal weather between adjacent planning nodes. However, there are also differences in the degree of abnormality of abnormal weather, such as rainfall, wind speed, etc. These factors have different degrees of influence on train operation. Therefore, the adjacent planning nodes can be divided into different abnormal road condition partitions according to the degree of weather abnormality. According to the speed limit conditions under the influence of the corresponding weather in different abnormal road condition partitions, the optimal running speed of the train in the corresponding abnormal road condition partition can be determined. Combined with the area length of the corresponding abnormal road condition partition, the estimated running sub-time of the train in the corresponding abnormal road condition partition can be calculated, and then the estimated running sub-time of all abnormal road condition partitions is summed to determine the total estimated running time of the train between adjacent planning nodes to ensure the accuracy of the estimated running time.

[0023] Furthermore, the step of obtaining all abnormal traffic zones between adjacent plan nodes and obtaining the speed limit condition of each abnormal traffic zone includes:

[0024] Determine the extent to which each abnormal weather type affects operating speed;

[0025] Obtain the corresponding abnormal weather type of each abnormal road condition partition, and classify each abnormal road condition partition according to the degree of influence of the corresponding abnormal weather type on the running speed;

[0026] The corresponding speed limit conditions are matched according to the divided levels.

[0027] Taking into account the real-time changes in weather conditions, in order to avoid the potential safety hazards caused by the excessive speed of trains when the abnormal weather becomes more severe, the train speed needs to be adjusted in time according to the abnormal weather conditions. Therefore, the degree of influence of the specific abnormal weather type on the running speed can be classified into levels according to the impact of the specific abnormal weather type on the running speed, so that the corresponding speed limit conditions can be quickly matched and the running speed of the train can be adjusted in time. This ensures the safe operation of the train, and at the same time avoids the time waste caused by the slow running speed of the train when the abnormal weather is reduced, thus ensuring the operational efficiency of the train.

[0028] Further, the correcting the corresponding planned operating time based on the estimated operating time includes:

[0029] Compare the estimated running time and the planned running time of adjacent planning nodes;

[0030] When the estimated running time exceeds the planned running time, the planned running time between the adjacent planned nodes is corrected to the estimated running time;

[0031] When the estimated running time does not exceed the planned running time, the planned running time between the adjacent plan nodes is maintained.

[0032] Furthermore, the step of constructing the planned nodes for train stops and setting the planned running time between adjacent planned nodes includes:

[0033] Obtain the train schedule and determine the scheduled arrival time and departure time of each node where the train stops;

[0034] Get the actual arrival time and departure time of the node where the train currently stops;

[0035] When the actual arrival time is inconsistent with the planned arrival time, or the actual departure time is inconsistent with the planned departure time, according to the train's schedule diagram, the node where the train stops is used as the initial schedule node, and the schedule node where the train stops is constructed, and the schedule running time between adjacent schedule nodes is set.

[0036] Since the correction of the planned running time is proposed based on the deviation between the actual running time and the planned running time of the train, when this deviation is detected, the construction of the plan node is started to reduce the amount of calculation.

[0037] Furthermore, the method of constructing a train stop plan node based on the train plan diagram, taking the current stop node as the initial plan node, and setting the planned running time between adjacent plan nodes includes:

[0038] When the actual arrival time is inconsistent with the planned arrival time, the estimated departure time of the initial planning node is set based on the actual arrival time and the station stop time;

[0039] When the actual departure time is inconsistent with the planned departure time, the estimated departure time of the initial planning node is set based on the actual departure time.

[0040] Furthermore, adjusting the train operation based on the revised planned operation time includes:

[0041] Obtain the estimated departure time of the previous plan node among the adjacent plan nodes, and calculate the estimated arrival time of the next plan node in combination with the revised planned running time between the adjacent plan nodes;

[0042] Calculate the time difference between the expected arrival time of the next planned node and the planned arrival time;

[0043] Update station stop time according to the time difference between early and late points;

[0044] Calculate the estimated departure time of the train at the next planning node based on the updated station stop time;

[0045] The train operation is controlled according to the calculated estimated departure time.

[0046] Further, the updating of the station stop time according to the time difference between the early and late points includes:

[0047] Get the train's stop time range and determine whether the time difference between the early and late trains is within the stop time range;

[0048] When the time difference between the morning and the evening is within the stop time range, the stop time shall be taken as the stop time;

[0049] When the time difference between the morning and late points is not within the stop time range, the maximum stop time or the minimum stop time shall be used as the stop time.

[0050] In order to ensure the safety of train operation, it is not possible to reduce the train running time by adjusting the speed of the train in the abnormal road condition zone, but the train stop time can be flexibly adjusted, which can be used to compensate for the time deviation between the actual running time and the planned running time under the influence of abnormal weather. However, considering the train boarding and alighting needs at the station stop node, when adjusting the station stop time, it is necessary to adjust it within the station stop time range to avoid affecting the passengers' travel needs.

[0051] Furthermore, it also includes:

[0052] Collect weather data in real time, and when weather changes are detected, update the abnormal road condition partition according to the corresponding weather data;

[0053] The estimated running time and planned running time between the remaining adjacent nodes are recalculated according to the updated traffic abnormality partition.

[0054] Taking into account the variability of weather, in order to ensure the operating efficiency of the train, the weather changes are detected in real time, and the corresponding abnormal road conditions are updated in time, so as to adjust the corresponding expected operating time and planned operating time to ensure the accuracy of subsequent train operation adjustments.

[0055] The beneficial effects of the present invention are:

[0056] (1) Obtain the abnormal road condition partition under the influence of abnormal weather, so as to determine the estimated running time of the train after the running speed is reduced due to the abnormal weather in the abnormal road condition partition, so as to correct the planned running time of the train according to the estimated running time, reduce the deviation between the actual arrival time or the actual departure time and the train plan, so as to facilitate the adjustment of the train operation, and reduce the overall running time of the train and reduce the travel delay under the premise of ensuring safety.

[0057] (2) According to the speed limit conditions in different road condition abnormality zones under the influence of corresponding weather conditions, the optimal operating speed of the train in the corresponding road condition abnormality zone can be determined. Combined with the area length of the corresponding road condition abnormality zone, the estimated operating sub-time of the train in the corresponding road condition abnormality zone can be calculated. Then, the estimated operating sub-times of all road condition abnormality zones are summed to determine the total estimated operating time of the train between adjacent planning nodes, thereby ensuring the accuracy of the estimated operating time.

[0058] (3) Furthermore, according to the degree of influence of specific abnormal weather types on the running speed, the degree of influence on the running speed of the train in each abnormal road condition zone is classified into different levels, so that the corresponding speed limit conditions can be quickly matched and the running speed of the train can be adjusted in time. This ensures the safe operation of the train and avoids the time waste caused by the train running too slowly when the abnormal weather is reduced, thus ensuring the operating efficiency of the train.

[0059] (4) Use the adjustment of station stop time to compensate for the deviation of train running time caused by abnormal weather. When adjusting the station stop time, make adjustments within the station stop time range to avoid affecting passengers' travel needs and ensure the normal operation of the train. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic diagram of a process of the present invention;

[0061] Figure 2 This is a schematic diagram of the operation of a train 1 between platforms ST0401-ST0301-ST0201-ST0101 in a snowy weather according to an embodiment of the present invention;

[0062] Figure 3 It is a schematic diagram of the actual running route of a train 1 between platforms ST0401-ST0301-ST0201-ST0101 according to an embodiment of the present invention. DETAILED DESCRIPTION

[0063] The present invention is further described below in conjunction with the accompanying drawings and embodiments.

[0064] Example:

[0065] Under the influence of abnormal weather such as rain and snow, in the actual operation process, in order to ensure the safety of train operation, the train's running speed will be reduced. Then the actual arrival time or actual departure time of the train recorded at the node where the train stops will deviate from the train plan. If the operation of the train at the node where the train stops is still controlled according to the train plan, this deviation will continue to widen, greatly increasing the overall running time of the train and making the train delay more serious.

[0066] In view of the above problems, this embodiment proposes a method for dynamic operation of urban rail transit trains under abnormal weather conditions, such as Figure 1 As shown, including:

[0067] Construct the planned nodes for train stops and set the planned running time between adjacent planned nodes;

[0068] Obtain abnormal road condition zones on the train running route based on weather data;

[0069] Calculate the estimated running time between adjacent planning nodes based on abnormal road conditions;

[0070] Correct the corresponding planned operation time based on the estimated operation time;

[0071] Adjust train operations based on the revised planned operating time.

[0072] The node where the train stops is equivalent to a platform. The train running time between each node can be displayed at the platform so that users can obtain the running status of the train. After the planned running time is corrected and the train operation is adjusted, the relevant information of the train running time displayed at the platform can be adjusted accordingly, so that passengers can timely obtain the updated running time and arrival time of the urban rail transit train, so as to change the corresponding travel plan and improve the passenger's sense of use.

[0073] Moreover, the method described in this embodiment is proposed to address the deviation between the actual running time and the planned running time of the train. Therefore, not all the nodes of the train station stops in the train plan are directly adopted. Instead, corresponding plan nodes are constructed according to actual conditions, and the objects that need to be adjusted are determined to implement targeted train operation adjustments and ensure the accuracy of train operation adjustments.

[0074] Taking into account the diversity of weather, the abnormal road condition partitions with abnormal weather conditions on the train running path are obtained based on weather data. In the abnormal path partitions, the train speed will be reduced and the operation time will be extended due to the abnormal weather. Therefore, based on the obtained abnormal road condition partitions, the estimated running time between adjacent plan nodes is determined, that is, the estimated running time of the train after the running speed is reduced due to abnormal weather in the abnormal road condition partitions, so as to correct the planned running time of the train plan according to the estimated running time, reduce the deviation between the actual arrival time or the actual departure time and the train plan, so as to facilitate the adjustment of the train operation.

[0075] The method of obtaining abnormal road condition partitions on the train running path according to weather data includes:

[0076] Obtain historical environmental change data and train operation data, and extract characteristic information including actual abnormal weather conditions, road conditions in the train operation path area, train operation speed, train stop protection distance and train type;

[0077] Construct a neural network model, take feature information as input, and actual road condition abnormality partitions under different weather conditions as output, and train the neural network;

[0078] The current weather data is input into the trained neural network to obtain the abnormal road condition partitions on the train running path.

[0079] For the prediction of abnormal road conditions, neural network models in deep learning, such as recurrent neural network models and long short-term memory network models, can be used to achieve it.

[0080] This embodiment takes the long short-term memory network model as an example. Historical environmental change data and train operation data can be collected first, and pre-processing operations such as cleaning, denoising, and difference can be performed to ensure data quality and continuity.

[0081] Then, feature information is extracted from the preprocessed data, and features related to abnormal weather and road conditions are proposed, such as actual abnormal weather conditions such as rainfall, snowfall, wind speed, track area range, track status, and other train operation path area road conditions, train operation speed, train parking protection distance, and train type.

[0082] The extracted feature information is then serialized by slicing it in chronological order, and a long short-term memory network model is constructed using a deep learning framework such as Keras or TensorFlow. The serialized data is divided into a training set and a validation set, the model is trained, and the model parameters are optimized using the back-propagation algorithm.

[0083] According to the current weather conditions, the corresponding abnormal road conditions can be accurately identified through the trained neural network model.

[0084] After determining the abnormal road condition zones, we can obtain the areas where the train's running speed will be affected accordingly during operation, and then calculate the estimated running time under the influence of abnormal weather through the abnormal road condition zones.

[0085] Specifically, the estimated running time between adjacent planning nodes is calculated based on the abnormal road condition partitions, including:

[0086] Obtain all abnormal traffic partitions between adjacent planning nodes, and obtain the speed limit conditions of each abnormal traffic partition;

[0087] Calculate the estimated operation sub-time of each abnormal road condition partition based on the corresponding speed limit condition and the area length of each abnormal path partition;

[0088] The estimated running sub-times of all abnormal traffic partitions between adjacent planning nodes are summed to obtain the estimated running time between adjacent nodes.

[0089] Considering the limited distance between adjacent planning nodes, although the weather conditions between adjacent planning nodes may be different, they are basically similar. Once there is an abnormal road condition partition between adjacent planning nodes, it can be regarded as abnormal weather between the adjacent planning nodes, and the adjacent planning nodes are composed of different path abnormality partitions.

[0090] Although the weather conditions are basically similar, the degree of abnormality of the abnormal weather faced in different road condition abnormal zones is not exactly the same. Taking rainy and snowy weather as an example, different road condition abnormal zones may face different rainfall amounts. The greater the rainfall, the higher the degree of abnormality of the abnormal weather. At this time, it is necessary to further reduce the running speed of the train to ensure the safety of the train operation.

[0091] Taking into account that the train will exhibit different speed reductions when facing different abnormal weather conditions, the speed limit conditions of each abnormal road condition zone under the influence of the corresponding abnormal weather are obtained respectively, and the optimal operating speed of the train in the corresponding abnormal road condition zone is determined. Then, combined with the area length of the corresponding abnormal road condition zone, the estimated running sub-time of the train in the corresponding abnormal road condition zone can be calculated. Then, the estimated running sub-times of all abnormal road condition zones are summed to determine the total estimated running time of the train between adjacent planned nodes to ensure the accuracy of the estimated running time.

[0092] The step of obtaining all abnormal traffic partitions between adjacent plan nodes and obtaining the speed limit condition of each abnormal traffic partition includes:

[0093] Determine the extent to which each abnormal weather type affects operating speed;

[0094] Obtain the corresponding abnormal weather type of each abnormal road condition partition, and classify each abnormal road condition partition according to the degree of influence of the corresponding abnormal weather type on the running speed;

[0095] The corresponding speed limit conditions are matched according to the divided levels.

[0096] The degree of influence of the abnormal weather type on the running speed can be obtained by analyzing the historical weather data and the historical train running data. The speed limit condition can be set based on the premise of train running safety, according to the information such as train type, train braking distance, and train braking influencing factors. The speed limit condition is the optimal train running speed under the premise of ensuring the safety of train running under slippery track conditions in the abnormal road condition zone.

[0097] In this embodiment, the speed limit levels of trains in abnormal road condition zones in different abnormal weather types are specifically divided into levels 0 to 10. Level 0 means that in the abnormal road condition zone, the running speed of the train is highly affected by the abnormal weather and cannot pass. Level 10 means that in the abnormal road condition zone, the running speed of the train is not affected by the abnormal weather.

[0098] Specifically, the function of grading abnormal road condition partitions can also be directly integrated into the above-mentioned neural network model by increasing the amount of information in the training set of the neural network model so that the neural network model can give the operating speed limit level of each predicted abnormal road condition partition when outputting the prediction results of the abnormal road condition partitions.

[0099] The correcting of the corresponding planned operating time based on the estimated operating time includes:

[0100] Compare the estimated running time and the planned running time of adjacent planning nodes;

[0101] When the estimated running time exceeds the planned running time, the planned running time between the adjacent planned nodes is corrected to the estimated running time;

[0102] When the estimated running time does not exceed the planned running time, the planned running time between the adjacent plan nodes is maintained.

[0103] When the estimated running time exceeds the planned running time, it proves that the train speed has been reduced due to abnormal weather between the adjacent planned nodes. At this time, there is a deviation between the planned running time recorded in the train plan and the actual operation of the train. At this time, it is necessary to correct the planned running time between the adjacent planned nodes to the estimated running time to reduce the deviation between the train plan and the actual operation of the train.

[0104] If the estimated running time does not exceed the planned running time, it proves that the train is in normal operating condition and can operate normally according to the train plan without any adjustment.

[0105] The step of constructing the planned nodes for train stops and setting the planned running time between adjacent planned nodes includes:

[0106] Obtain the train schedule and determine the scheduled arrival time and departure time of each node where the train stops;

[0107] Get the actual arrival time and departure time of the node where the train currently stops;

[0108] When the actual arrival time is inconsistent with the planned arrival time, or the actual departure time is inconsistent with the planned departure time, according to the train's schedule diagram, the node where the train stops is used as the initial schedule node, and the schedule node where the train stops is constructed, and the schedule running time between adjacent schedule nodes is set.

[0109] Since the dynamic operation method of this embodiment is proposed for the deviation between the train plan and the actual operation of the train, in order to reduce the amount of calculation and improve the efficiency of train operation control, the dynamic operation method proposed in this embodiment is executed only when a deviation in the running time is detected.

[0110] Specifically, this embodiment identifies the deviation between the train plan and the actual operation of the train by comparing the actual arrival time with the planned arrival time, and the actual departure time with the planned departure time.

[0111] Under the influence of abnormal weather, not only will the speed of the train be reduced during operation, but the train may also be unable to leave the platform normally in order to avoid extreme abnormal weather or be affected by delays of the train ahead. Therefore, it is not only necessary to compare the actual arrival time with the planned arrival time, but also to further compare the actual departure time with the planned departure time to ensure the accuracy of identifying the deviation between the train plan and the actual operation of the train, and to improve the efficiency of subsequent train operation adjustments.

[0112] The method of constructing a train stop plan node based on the train plan diagram, taking the current stop node as the initial plan node, and setting the planned running time between adjacent plan nodes includes:

[0113] When the actual arrival time is inconsistent with the planned arrival time, the estimated departure time of the initial planning node is set based on the actual arrival time and the station stop time;

[0114] When the actual departure time is inconsistent with the planned departure time, the estimated departure time of the initial planning node is set based on the actual departure time.

[0115] The estimated departure time of the initial planning node directly affects the calculation and correction of the subsequent estimated running time and planned running time. In order to ensure the accuracy of subsequent calculations, the estimated departure time of the corresponding initial planning node is set for the two deviation situations.

[0116] The adjusting the train operation based on the revised planned operation time includes:

[0117] Obtain the estimated departure time of the previous plan node among the adjacent plan nodes, and calculate the estimated arrival time of the next plan node in combination with the revised planned running time between the adjacent plan nodes;

[0118] Calculate the time difference between the expected arrival time of the next planned node and the planned arrival time;

[0119] Update station stop time according to the time difference between early and late points;

[0120] Calculate the estimated departure time of the train at the next planning node based on the updated station stop time;

[0121] The train operation is controlled according to the calculated estimated departure time.

[0122] In order to ensure the safe operation of the train, it is impossible to adjust the running time by adjusting the train running speed. However, in the actual operation process, the station stop time of the train can be flexibly adjusted. Therefore, this embodiment uses the update of the station stop time to compensate for the running time deviation of the train under the influence of abnormal weather, thereby improving the train operation efficiency.

[0123] The updating of the station stop time according to the time difference between the early and late points includes:

[0124] Get the train's stop time range and determine whether the time difference between the early and late trains is within the stop time range;

[0125] When the time difference between the morning and the evening is within the stop time range, the stop time shall be taken as the stop time;

[0126] When the time difference between the morning and late points is not within the stop time range, the maximum stop time or the minimum stop time shall be used as the stop time.

[0127] Taking into account that it takes a certain amount of time for passengers to get on and off the bus at stop nodes, when adjusting the stop time, it is necessary to make adjustments within the stop time range to avoid affecting the normal ride of passengers.

[0128] Changes in weather conditions will also lead to changes in road conditions. Therefore, weather data is collected in real time. When weather changes are detected, the abnormal road condition partition is updated according to the corresponding weather data.

[0129] The estimated running time and planned running time between the remaining adjacent nodes are recalculated according to the updated traffic abnormality partition.

[0130] Each time the planned running time is revised, the information displayed at the nodes where each train stops is updated accordingly to ensure that the actual running status of the train is transmitted in a timely manner.

[0131] Take the case of train 1 running between platforms ST0401-ST0301-ST0201-ST0101 in snowy weather as an example. Figure 2 As shown, the allowed stop time for each station is 20s to 50s. There are abnormal path partitions between ST0201-ST0101 and ST0301-ST0201.

[0132] Specifically, the plan for train 1 to run between the four platforms is: ST0401 (9:00:00-9:00:30)-ST0301 (9:04:30-9:05:00)-ST0201 (9:09:00-9:09:30)-ST0101 (9:13:30-9:14:00). However, due to the snowy weather, train 1 left ST0301 at 9:06:20, 80 seconds late.

[0133] At this time, ST0301 is used as the initial planning node Leg1, and the estimated departure time of planning node Leg1 is its actual departure time 9:06:20. According to the planning diagram, the remaining planning nodes are set in sequence, including planning nodes Leg2 and Leg3, which are ST0201 and ST0101 respectively.

[0134] According to the weather data between the planning nodes Leg1 and Leg2 and the planning nodes Leg2 and Leg3, the abnormal road condition partitions are divided and the speed limit levels are divided respectively. It is determined that there are two abnormal road condition partitions between the planning nodes Leg1 and Leg2, and their speed limit levels are level 6 and level 7 respectively. There are two abnormal road condition partitions between the planning nodes Leg2 and Leg3, and their speed limit levels are level 6 and level 5 respectively.

[0135] Based on the planning diagram, it can be seen that the planned running time of train 1 between the planning nodes Leg1 and Leg2 is 240s. Combined with the abnormal road condition partition, it is calculated that the estimated running time of train 1 between the planning nodes Leg1 and Leg2 should be 250s.

[0136] Furthermore, based on the abnormal road condition partition between the planned nodes Leg2 and Leg3, it is calculated that the estimated running time of Train 1 between the planned nodes Leg2 and Leg3 should be 260s.

[0137] Based on the calculation result of the expected running time, the planned running time is corrected, and then it can be calculated that the expected arrival time of train 1 at the planned node Leg2 = the expected departure time of Leg1 + the expected running time between Leg1 and Leg2 = 9:06:20 + 250s = 9:10:30.

[0138] From this, we can see that the time difference between the early and late points of the planned node Leg2 = 9:10:30-9:09:30 = 60s.

[0139] Therefore, the stop time is updated to the minimum stop time, which is 20s.

[0140] On this basis, the estimated departure time of the planned node Leg2 = the estimated arrival time of Leg2 + the stop time = 9:10:30 + 20s = 9:10:50.

[0141] Repeating the above calculation, we can obtain the estimated arrival time of the planned node Leg3 = 9:15:10, and the estimated departure time of the planned node Leg3 = 9:15:30.

[0142] The actual running route of train 1 between platforms ST0401-ST0301-ST0201-ST0101 is as follows: Figure 3 shown.

[0143] The above-mentioned estimated arrival time and estimated departure time are synchronized in real time to the ground passenger information system PIS of each planning node so that passengers can obtain the actual operation status of the train 1.

[0144] Subsequently, the current weather conditions are monitored in real time to update the abnormal road conditions partition according to the weather conditions, so as to recalculate the planned running time of train 1 to achieve the operation adjustment of train 1.

[0145] After the operation adjustment of train 1 is completed, the operation time deviations of other subsequent trains can be analyzed based on the adjusted operation time of train 1, and operation adjustments can be made according to the dynamic operation method proposed in this embodiment.

[0146] The above-described embodiment is only a preferred solution of the present invention and does not limit the present invention in any form. There are other variations and modifications without exceeding the technical solution described in the claims.

Claims

1. A method for dynamic operation of urban rail transit trains under abnormal weather conditions, characterized in that: include: Construct the planned nodes for train stops and set the planned running time between adjacent planned nodes; Obtain abnormal road condition zones on the train running route based on weather data; Calculate the estimated running time between adjacent planning nodes based on abnormal road conditions; Correct the corresponding planned operation time based on the estimated operation time; Adjust train operations based on the revised planned operating time.

2. The method for dynamic operation of urban rail transit trains under abnormal weather conditions according to claim 1, characterized in that: The method of obtaining abnormal road condition partitions on the train running path according to weather data includes: Obtain historical environmental change data and train operation data, and extract characteristic information including actual abnormal weather conditions, road conditions in the train operation path area, train operation speed, train stop protection distance and train type; Construct a neural network model, take feature information as input, and actual road condition abnormality partitions under different weather conditions as output, and train the neural network; The current weather data is input into the trained neural network to obtain the abnormal road condition partitions on the train running path.

3. The method for dynamic operation of urban rail transit trains under abnormal weather conditions according to claim 1, characterized in that: The abnormal traffic partitioning calculates the estimated running time between adjacent planning nodes, including: Obtain all abnormal traffic partitions between adjacent planning nodes, and obtain the speed limit conditions of each abnormal traffic partition; Calculate the estimated operation sub-time of each abnormal road condition partition based on the corresponding speed limit condition and the area length of each abnormal path partition; The estimated running sub-times of all abnormal traffic partitions between adjacent planning nodes are summed to obtain the estimated running time between adjacent nodes.

4. The method for dynamic operation of urban rail transit trains under abnormal weather conditions according to claim 3, characterized in that: The step of obtaining all abnormal traffic partitions between adjacent plan nodes and obtaining the speed limit condition of each abnormal traffic partition includes: Determine the extent to which each abnormal weather type affects operating speed; Obtain the corresponding abnormal weather type of each abnormal road condition partition, and classify each abnormal road condition partition according to the degree of influence of the corresponding abnormal weather type on the running speed; The corresponding speed limit conditions are matched according to the divided levels.

5. The method for dynamic operation of urban rail transit trains under abnormal weather conditions according to claim 1, characterized in that: The correcting of the corresponding planned operating time based on the estimated operating time includes: Compare the estimated running time and the planned running time of adjacent planning nodes; When the estimated running time exceeds the planned running time, the planned running time between the adjacent planned nodes is corrected to the estimated running time; When the estimated running time does not exceed the planned running time, the planned running time between the adjacent plan nodes is maintained.

6. The method for dynamic operation of urban rail transit trains under abnormal weather conditions according to claim 1, characterized in that: The step of constructing the planned nodes for train stops and setting the planned running time between adjacent planned nodes includes: Obtain the train schedule and determine the scheduled arrival time and departure time of each node where the train stops; Get the actual arrival time and departure time of the node where the train currently stops; When the actual arrival time is inconsistent with the planned arrival time, or the actual departure time is inconsistent with the planned departure time, according to the train's schedule, the node where the train stops is used as the initial schedule node, and the schedule node where the train stops is constructed, and the schedule running time between adjacent schedule nodes is set.

7. The method for dynamic operation of urban rail transit trains under abnormal weather conditions according to claim 6, characterized in that: The method of constructing a train stop plan node based on the train plan diagram, taking the current stop node as the initial plan node, and setting the planned running time between adjacent plan nodes includes: When the actual arrival time is inconsistent with the planned arrival time, the estimated departure time of the initial planning node is set based on the actual arrival time and the station stop time; When the actual departure time is inconsistent with the planned departure time, the estimated departure time of the initial planning node is set based on the actual departure time.

8. The method for dynamic operation of urban rail transit trains under abnormal weather conditions according to claim 7, characterized in that: The adjusting the train operation based on the revised planned operation time includes: Obtain the estimated departure time of the previous plan node among the adjacent plan nodes, and calculate the estimated arrival time of the next plan node in combination with the revised planned running time between the adjacent plan nodes; Calculate the time difference between the expected arrival time of the next planned node and the planned arrival time; Update station stop time according to the time difference between early and late points; Calculate the estimated departure time of the train at the next planning node based on the updated station stop time; The train operation is controlled according to the calculated estimated departure time.

9. The method for dynamic operation of urban rail transit trains under abnormal weather conditions according to claim 8, characterized in that: The updating of the station stop time according to the time difference between the early and late points includes: Get the train's stop time range and determine whether the time difference between the early and late trains is within the stop time range; When the time difference between the morning and the evening is within the stop time range, the stop time shall be taken as the stop time; When the time difference between the morning and late points is not within the stop time range, the maximum stop time or the minimum stop time shall be used as the stop time.

10. The method for dynamic operation of urban rail transit trains under abnormal weather conditions according to claim 1, characterized in that: Also includes: Collect weather data in real time, and when weather changes are detected, update the abnormal road condition partition according to the corresponding weather data; The estimated running time and planned running time between the remaining adjacent nodes are recalculated according to the updated traffic abnormality partition.

Images