JP2023082385A - Train operation management system, parameter production system and parameter production method - Google Patents

Abstract

To provide a parameter production system which can output a probable prediction diagram.SOLUTION: A train operation management system comprises an arithmetic unit to perform specified processes and a storage device to which the arithmetic unit is accessible. The arithmetic unit has: an input unit to input a plan diagram, an evaluation scenario of the plan diagram, an evaluation index, facility data and a beginning value input of the parameter; a prediction unit for predicting a diagram according to one or more evaluation scenarios and producing a prediction diagram; and a parameter narrowing down unit for comparing the plan diagram with the prediction diagram to narrow down the parameter value satisfying the evaluation index. The operation management system predicts the train diagram using the narrowed down parameter and outputs signals to control track facilities according to the predicted train diagram.SELECTED DRAWING: Figure 1A

Description

The present invention relates to a system for creating parameters necessary for predicting train operations in a train operation management system.

The train operation management business is a business that controls the operation time and running route (hereinafter referred to as a route) of the train based on the train operation plan, and supports transportation by train. In order to improve the efficiency of this train operation management work, a train operation management system having a train operation prediction function has been realized.

As background arts in this technical field, there are the following prior arts. In Patent Document 1 (Japanese Patent Laid-Open No. 2016-30542), in a train operation management system, an environment change detection unit that detects a location where the environment related to the train has changed, and a condition for storing influence range information corresponding to the situation of the train a storage unit, a location where the environment has changed, and an effect range specifying unit that specifies a range of impact of the environmental change based on the information on the range of impact stored in the condition storage unit; and a prediction unit for predicting train operation based on the determined constraint.

Further, in Patent Document 2 (Japanese Patent Application Laid-Open No. 2019-93906), an operation management support device of an embodiment includes a configuration information storage unit, an input unit, and an identification unit. The configuration information storage unit stores configuration information indicating each configuration provided in an operation management system that manages operation of trains. The input unit contains actual timetable data that shows the actual operation results when the trains managed by the operation management system that manages the operation of the trains are operated according to the route, and the actual timetable data that shows the operation results when the train operates on the route on time. Predetermined timetable data indicating the operation status are input. The specifying unit performs operation management, which is subject to correction so as to operate the train according to the predetermined timetable data, based on the difference between the actual timetable data and the predetermined timetable data and the configuration information stored in the configuration information storage unit. Identify the configurations provided in the system.

JP 2016-30542 A JP 2019-93906 A JP 2012-245801 A JP 2009-96221 A

The train operation prediction function, which is one of the functions of the train operation management system, expresses the operation of trains by diagrams showing the arrival and departure times of stations on the train operation route, and predicts future timetables. In Patent Document 1 (Japanese Patent Application Laid-Open No. 2016-30542), parameters necessary for timetable prediction are modeled in the form of standard operation time, crossing interval, and the like. To predict timetables plausibly, we must model signaling equipment, rolling stock, etc., and input appropriate parameters into the model. However, the correspondence between the control data of signaling equipment and vehicles and the parameters is not clear, and the number of parameters is large. In addition, trial and error was required to create parameters in situations where workers were not familiar with the train operation management system, or when it was difficult to obtain detailed control system data (e.g., operation curves) for new routes. .

For this reason, unless you are familiar with the behavior of the train operation control system, you cannot create plausible parameters, and even if you are familiar with the train operation control system, it takes time to create parameters because trial and error is required. . In addition, with on-board signalization of ground signaling equipment, control data can be updated by updating software, and it is expected that the frequency of creating parameters to follow the control data will increase, and efficient parameter creation is required.

As a method for supporting parameter creation, in the method described in Patent Document 2 (Japanese Patent Laid-Open No. 2019-93906), correction points in the configuration of the train operation management system are specified from the difference between the actual timetable and the predetermined timetable. This makes it possible to quickly find out which devices need modification. However, Patent Document 2 does not refer to a method for correcting the specified location. In addition, Patent Literature 2 cannot be applied when the actual timetable for the new route is not available.

An object of the present invention is to create parameters that enable output of a plausible predicted timetable from parameter value candidates with a degree of freedom.

A representative example of the invention disclosed in the present application is as follows. That is, a train operation management system comprising an arithmetic device for executing predetermined processing and a storage device accessible by the arithmetic device, the arithmetic device including a planned timetable, an evaluation scenario for the planned timetable, and an evaluation index , an input unit that receives input of initial values of equipment data and parameters, a prediction unit that predicts a timetable according to one or more of the evaluation scenarios and creates a predicted timetable, and a prediction unit that generates a predicted timetable, a parameter narrowing unit that compares the timetable with the predicted timetable and narrows down parameter values that satisfy the evaluation index, and the operation management system predicts the train timetable using the narrowed down parameters, It is characterized by outputting a signal for controlling the track equipment according to the train schedule.

According to one aspect of the present invention, plausible parameters can be created without trial and error. Problems, configurations, and effects other than those described above will be clarified by the following description of the embodiments.

1 is an overall configuration diagram of a parameter creation system of an embodiment; FIG. It is a figure which shows the physical structure of the parameter preparation system of a present Example. It is a figure which shows the track wiring represented by the installation data of a present Example. It is a figure which shows the table|surface which is a structural example of the planned timetable of a present Example. FIG. 10 is a diagram showing a table that is a configuration example of temporary parameters according to the embodiment; FIG. 10 is a diagram showing a table that is a configuration example of temporary parameters according to the embodiment; FIG. 10 is a diagram showing a table that is a configuration example of a robustness evaluation scenario of the embodiment; FIG. 10 is a diagram showing a table that is a configuration example of parameter application priority information according to the embodiment; It is a figure which shows the table|surface which is a structural example of the improvement log|history of a present Example. 4 is a sequence diagram of processing executed by the parameter creation system of the embodiment; FIG. 4 is a sequence diagram of processing executed by the parameter creation system of the embodiment; FIG. 4 is a flow chart showing the overall flow of processing by the parameter creation system of the embodiment; FIG. 10 is a flowchart of processing for calculating a constraint expression for each parameter in the case of the robustness evaluation scenario of the embodiment; FIG. 4 is a flow chart of a process (S400) for calculating a representative value from a set range of parameters summarizing a plurality of constraint expressions according to the present embodiment; It is a figure which shows the example of the input screen from the worker of a present Example. It is a figure which shows the example of the output screen shown to the worker of a present Example. It is a figure which shows the forecast timetable produced by the present Example. 10 is a graph illustrating adjustment of travel time between BC station and stop time at C station in evaluation scenario 2 in the present embodiment. FIG. 10 is a diagram showing setting ranges of a plurality of parameters calculated using a steady scenario and evaluation scenario 2 of the present embodiment;

Embodiments of the present invention will be described below with reference to the drawings.

FIG. 1A is an overall configuration diagram of the parameter creation system 100. As shown in FIG.

The parameter creation system 100 is a device that creates data for operating the train operation management system 99 . The train operation management system 99 collects information on trains from track facilities and transmits information necessary for train operation to the track facilities and trains. The parameter creation system 100 is positioned as a subsystem of the train operation management system 99, but may be a system separate from the train operation management system 99. For example, when the parameter creation system 100 and the train operation management system 99 are installed in different computer environments, data may be transferred through a network or a storage medium. In this case, the data that the worker 98 receives from the outside may always receive the latest data online, and the created data may be reflected in the train operation management system 99 . Also, if actual travel data can be used on an existing route, the actual timetable and the predicted timetable may be compared instead of the planned timetable. If the improvement history 105 is sufficiently stored, adjustment by the operator 98 may be unnecessary. With this configuration, even if control data needs to be periodically updated due to on-board signaling and automatic driving, the data that follows the control is automatically updated, and the burden of creating data can be reduced.

Parameter creation system 100 includes input unit 101 , prediction unit 102 , parameter narrowing unit 103 , result presentation unit 104 , and improvement history 105 . The input section 101 is a section that receives data input from the operator 98 . The prediction unit 102 is a part that predicts diamonds from parameters, and can be realized by the technology described in Japanese Patent Application Laid-Open No. 2012-245801, for example. A parameter narrowing-down unit 103 is a part that narrows down plausible values from among parameters with a degree of freedom. The parameter narrowing unit 103 includes a parameter value candidate creation unit 103A that creates parameter value candidates, and a constraint expression calculation unit 103B that determines whether the timetable prediction result satisfies the evaluation index 109B and calculates a constraint expression that the parameter value satisfies. and a parameter value selection unit 103C that selects one parameter value that satisfies all the constraint expressions for one or more evaluation scenarios.

The result presentation unit 104 is a part that presents the created parameters to the operator 98 . Improvement history 105 records feedback from workers 98 when presenting results. The parameter creation system 100 receives the planned timetable 107, the provisional parameters 108, the robustness evaluation scenario 109A, the evaluation index 109B, and the parameter application priority information 110 from the outside. These data 107-110 will be described later with reference to FIGS.

FIG. 1B is a diagram showing the physical configuration of the parameter creation system 100. As shown in FIG.

The parameter creation system 100 is composed of a computer having a processor (CPU) 1 , memory 2 , auxiliary storage device 3 and communication interface 4 . The parameter generation system 100 may have an input interface 5 and an output interface 6 .

Processor 1 is an arithmetic device that executes a program stored in memory 2 . By the processor 1 executing various programs, the functions of each functional unit (for example, the input unit 101, the prediction unit 102, the parameter narrowing unit 103, the result presentation unit 104, etc.) of the parameter creation system 100 are realized. Note that part of the processing performed by the processor 1 by executing the program may be performed by another arithmetic device (for example, hardware such as ASIC and FPGA).

The memory 2 includes ROM, which is a non-volatile storage element, and RAM, which is a volatile storage element. The ROM stores immutable programs (eg, BIOS) and the like. RAM is a high-speed and volatile storage element such as DRAM (Dynamic Random Access Memory), and temporarily stores programs executed by the processor 1 and data used when the programs are executed.

The auxiliary storage device 3 is, for example, a large-capacity, non-volatile storage device such as a magnetic storage device (HDD) or flash memory (SSD). The auxiliary storage device 3 also stores data used by the processor 1 when executing programs (for example, improvement history 105, etc.) and programs executed by the processor 1 . That is, the program is read from the auxiliary storage device 3, loaded into the memory 2, and executed by the processor 1, thereby realizing each function of the parameter creation system 100. FIG.

The communication interface 4 is a network interface device that controls communication with other devices (for example, train operation management system 99) according to a predetermined protocol.

The input interface 5 is an interface to which input devices such as a keyboard 7 and a mouse 8 are connected and receives input from an operator. The output interface 6 is an interface to which output devices such as a display device 9 and a printer (not shown) are connected, and which outputs results of program execution in a format that can be visually recognized by the operator. A user terminal connected to the parameter creation system 100 via a network may provide the input device and the output device. In this case, the parameter creation system 100 may have a web server function, and the user terminal may access the parameter creation system 100 using a predetermined protocol (eg, http).

Programs executed by the processor 1 are provided to the parameter creation system 100 via removable media (CD-ROM, flash memory, etc.) or a network, and stored in the non-volatile auxiliary storage device 3, which is a non-temporary storage medium. . Therefore, the parameter creation system 100 preferably has an interface for reading data from removable media.

The parameter creation system 100 is a computer system configured on one physical computer or on a plurality of logically or physically configured computers, and is constructed on a plurality of physical computer resources. It may operate on a virtual machine. For example, the input unit 101, the prediction unit 102, the parameter narrowing unit 103, and the result presentation unit 104 may operate on separate physical or logical computers, or may be combined into one physical or logical computer. It may work on the

FIG. 2 is a diagram showing the line wiring represented by the facility data 106. As shown in FIG.

The facility data 106 includes stations, tracks, traffic lights, and alignments. In the figure, there are stations A, B, and C, tracks #1 and #2, and signals 8111, 8121, 8211, 8212, 8311, and 8321. FIG. 2 shows an alignment with a dead end at station C and a turnaround operation.

FIG. 3 is a diagram showing a table T01 that is a configuration example of the planned timetable 107. As shown in FIG. 3 to 6 show each data in a tabular form, they may be configured in a form other than the tabular form. The planned timetable 107 (table T01) records the arrival/departure track numbers and arrival/departure times for each station relating to train operation, and includes a serial number T01a that is row record identification information and train identification information (train number). It includes a train number T01b, a station T01c, a track number T01d, an arrival time T01e, and a departure time T01f. The planned timetable 107 preferably includes the step size of the planned timetable (minimum time interval set when creating the planned timetable) in addition to the data shown in Table T01.

4A and 4B are diagrams showing a table T02, which is a configuration example of the temporary parameters 108. FIG. Temporary parameters 108 include parameters set between the departure station T02c and destination station T02d shown in FIG. minute, continuation interval, crossing failure interval). The tables shown in FIGS. 4A and 4B may be configured separately as shown or combined. These temporary parameters may be created for each train type or vehicle performance. The provisional parameter 108 (table T02) includes a serial number T02a, a parameter type T02b, a parameter section (departure station T02c, arrival station T02d, station T02e), and a provisional value T02f.

If the train operation management system 99 has already been operated and there is a basic parameter, the provisional parameter 108 is set to its value as an initial value. If there is no underlying parameter, set the default value determined by the parameter's specification. Examples of these parameters include stop time, turn-around time, continuation time interval, and crossing obstacle time interval described in a known document (Japanese Patent Laid-Open No. 2009-96221). Of these, the stop time and turnaround time are set for each station, and the continuation time interval and crossing obstacle time interval are set between stations. For example, on the route shown in FIG. 2, the continuation time interval is the time interval between the preceding train at B station and the continuation train, and the crossing obstacle time interval is the time when passing through a branch installed at A station or C station. This is the time interval that must be secured by the preceding and following trains. The parameter value may be a continuous value such as hour and minute, or a discrete value such as whether or not it is necessary to secure these time intervals.

FIG. 5 is a diagram showing a table T03 that is a configuration example of the robustness evaluation scenario 109A. The robustness evaluation scenario 109A expresses the robustness that is taken into consideration when preparing the planned timetable by the allowable amount of delay from each point. Robustness evaluation scenario 109A (Table T03) includes serial number T03a, station T03b where the assumed delay occurs, assumed delay time T03c, and the assumed delay entered into the station where the delay occurs (predicted timetable and planned timetable) and a station T03d where the departure times match).

FIG. 6 is a diagram showing a table T04 that is a configuration example of the parameter application priority information 110. As shown in FIG. The parameter application priority information 110 (table T04) is used when apportioning parameter adjustment amounts among a plurality of parameters. The amount of parameter adjustment may be determined by a method of quantifying a ratio including spare time peculiar to the parameter type. Parameter application priority information 110 (table T04) includes serial number T04a, parameter type T04, and priority T04c.

FIG. 7 is a diagram showing a table T05 that is a configuration example of the improvement history 105. As shown in FIG. The improvement history 105 is used for the next generation of parameters for accumulating feedback from the worker 98 when presenting results. The improvement history 105 (table T05) includes a serial number T05a, a parameter type T05b, a departure station T05c and a destination station T05d in the section of the parameter, a proposed setting width T05e, a proposed representative value T05f, an improvement type T05g, and an improvement instruction T05h. include. The setting width T05e indicates a settable range of parameters with a degree of freedom by narrowing down the parameters. The representative value is a set value selected from the set range. A method of selecting the setting value will be described later. The improvement type T05g is approved when a representative value is adopted as a parameter, selected when a value other than the representative value is selected from the set range, and is presented when a value outside the set range is specified or presented. If a parameter type other than the specified parameter type is specified, new is recorded. The improvement instruction T05h records the selected or specified value when the improvement type T05g is selected and new, and is provided to the train operation management system 99. FIG.

8A and 8B are sequence diagrams of the processing executed by the parameter creation system 100. FIG. 8A shows the overall processing, and FIG. 8B shows the predicted diagram generation processing based on the steady scenario and the predicted diagram generation based on the robustness evaluation scenario 109A. Indicates processing.

The worker 98 inputs input data including the facility data 106, the planned diagram 107, the temporary parameters 108, and the robustness evaluation scenario 109A into the input unit 101. The parameter application priority information 110 is created and incorporated by the developer of the parameter creating system 100 . The input unit 101 registers input data in predetermined databases 106-110. When the data registration is completed, the input unit 101 notifies the parameter narrowing unit 103 of the address where the registration data is stored.

The parameter narrowing section 103 sends the planned timetable 107 and the provisional parameters 108 to the prediction section 102 . The prediction unit 102 uses the data sent from the parameter narrowing unit 103 to predict the diamond. When the timetable prediction based on the provisional parameters 108 is completed, the parameter narrowing unit 103 creates the adjusted parameters 111 based on the planned timetable.

The parameter narrowing unit 103 is a prediction diagram created by the prediction unit 102 using the planned diagram 107, the parameters 111 adjusted by the planned diagram, and the robustness evaluation scenario 109A (predicted from the current time calculated by the parameters diamond). The parameter narrowing unit 103 uses the predicted timetable, the planned timetable 107 , the improvement history 105 , and the parameter application priority information 110 to calculate the parameter setting ranges and representative values, and sends them to the result presenting unit 104 .

The result presenting unit 104 presents the result of comparison between the planned timetable 107 and the predicted timetable, the result of comparison between the provisional parameters 108 and the set ranges and representative values of the parameters, and the applied robustness evaluation scenario 109A to the operator 98 .

The operator 98 instructs an adjustment method for the presented results from the output screen. The result presentation unit 104 stores instructions from the worker 98 in the improvement history 105 , creates adjusted parameters 111 based on the instructions, and presents them to the worker 98 .

FIG. 9 is a flow chart showing the overall flow of processing by the parameter creation system 100 of this embodiment.

First, input data is registered in the facility data 106, the planned timetable 107, the provisional parameters 108, the robustness evaluation scenario 109A, and the parameter application priority information 110 (S201).

Next, the diamond is predicted using the provisional parameters 108 (S202). After that, with respect to the arrival/departure times of each station, trains whose estimated timetable created in step S202 is later than the timetable of the planned timetable are extracted (S203). Since early departure is prohibited on railways, delayed trains are extracted. For example, the step size of the planned timetable is 5 seconds, the departure time of a certain train from station A on the planned timetable is 12:00:00, the arrival time at B station is 12:03:00, and the departure time from station A on the predicted timetable is 12:00:00. When the arrival time at B station is 00:00 and 12:03:05, this train is the train to be extracted. Also, when trains depart from Platforms 1 and 2 of Station A toward the same track at the same time, the departure time of one of the trains is delayed, so this train is a train to be extracted. Since the difference less than the step size of the planned timetable is included in the range of rounding error, the number of predictions can be reduced by the process of extracting trains that are slower than the step size of the planned timetable.

Then, for all the extracted trains (S204a-S204b), using a steady scenario (a scenario in which trains operate without delay according to the planned timetable), parameters related to locations where differences occur are changed, and timetables are predicted (S300). . Details of the process S300 will be described with reference to FIG. Further, processing for calculating a constraint expression indicating a value range for each parameter for each scenario will be described with reference to FIG. 10 .

Constraint expressions for each parameter created for each train among a plurality of robustness evaluation scenarios are put together to create a representative value (S400). Details of the process S400 will be described with reference to FIG. The representative values of the parameters that match the planned timetable are saved as the adjusted parameters 111 according to the planned timetable. In addition, by recording the parameter type and location of the change in the predicted timetable in the timetable prediction, the process S300 using the robustness evaluation scenario 109A is efficiently executed. The iterative processing of steps S204a and S204b may be performed in parallel for each train to shorten the time required for timetable prediction.

After that, the departure time of a certain train is delayed by the value described in the robustness evaluation scenario 109A (S205). By inserting a delay, it is possible to omit the spare time included in the planned timetable and obtain a predicted timetable calculated by combining parameters. Then, for each evaluation scenario of the robustness evaluation scenario 109A (S206a-S206b), timetable prediction in the delay state is performed to calculate a constraint expression representing the possible range of parameters (S300). The calculated constraint expressions for each evaluation scenario are put together to create setting widths and representative values (S400). The iterative processing of processing S206 may be performed in parallel for each evaluation scenario to shorten the time required for timetable prediction.

The created set width and representative value are presented to the operator 98 on the output screen displayed together with the provisional parameters 108 (S207). The operator 98 enters feedback on the improvement method with respect to the presented parameters, and the entered feedback is recorded into the improvement method. By enriching the improvement history 105, the amount of data used as the basis for determining the representative value increases, less confirmation of parameter values is required, and the accuracy of automatically adjusted parameters can be improved.

FIG. 10 is a flowchart of processing for calculating a constraint expression for each parameter in the case of the robustness evaluation scenario 109A, and shows processing S300 in detail.

First, diamonds are predicted with given parameters (S301). FIG. 14 shows a predicted timetable created using each of the planned timetable of FIG. 3 and the scenario of FIG. The diagram shown in FIG. 14 has the arrival time T06e and departure time T06f of each station for each train using the steady scenario, the arrival time T06g and departure time T06h for each train for each station using the evaluation scenario 1, and the evaluation scenario 2. It includes arrival time T06i and departure time T06j of each train at each station. In the evaluation scenario 1, prediction is made under the condition that the departure time at Station A is delayed by 90 seconds in #2. In the evaluation scenario 2, prediction is made under the condition that the departure time at B station is delayed by 30 seconds at #4.

After that, the planned timetable and the predicted timetable are compared to determine the parameters to be adjusted (S302). If the departure time and arrival time of each station do not match between the planned timetable and the predicted timetable, it is determined that there is a difference, and the parameters are changed. For the parameters to be changed, parameters are selected that define the section between the station where the difference first occurs on the route of the train (hereinafter referred to as the difference origin station) and the stations that depart from or arrive at the difference origin station. For example, when a difference occurs in the schedule from the time of arrival at B station in the alignment of FIG. Then, #1 set between A and B stations, #12 set at the branch between A and B stations, and #9 and #10 set at the interval between trains before and after between A and B stations) are Applicable. In the case of the steady-state scenario, the parameters of the location where the difference occurs are comprehensively selected. In the case of the robustness evaluation scenario 109A, continuous values that were effective in processing in the stationary scenario may be extracted. In this case, it is possible to efficiently discover parameters that cause changes in the predicted timetable, and reduce the number of timetable predictions.

Using the extracted parameters, repeated processing is executed for each parameter and for each set of parameters (S303a-S303b). The set of parameters is all possible combinations of the extracted parameters, and the iterative process is performed for all these combinations. For example, if it is necessary to secure the continuation interval, a combination of continuation intervals can occur, but if it is unnecessary to secure the interval, no combination of continuation intervals can occur. Iterations may be performed in parallel to reduce computation time. The reason why this is performed for all combinations is that there are parameter types that have a relationship in which changes are not reflected in the predicted timetable unless a certain parameter type is changed. For example, if the continuation interval, which is the interval between the departure of the preceding train at a station and the arrival of the own train, determines the predicted arrival time of the following train, Even if the hour and minute are shortened, the predicted arrival time at the station does not change.

Next, the amount of parameter change is determined. Refer to the equipment data 106 of the section of the applied robustness evaluation scenario 109A (the section from the station of T03b to the station of T03d), and determine whether there is a line that becomes a bottleneck in operation such as an intersection or a single track section. (S304). Whether or not there is a bottleneck can be determined by whether or not there is a bottleneck such as a single track or crossing on the track of the train. In the example of evaluation scenario 2, since there is a single track section at B station, the determination is YES. If the determination is YES, the first set value is set for the step size for adjusting the parameter (S305). If the determination is NO, a second set value is set for the step size for adjusting the parameter (S306). Note that the first setting value is predetermined to be smaller than the second setting value. For example, the first set value may be set to the minimum value of the parameter specification, and the second set value may be set to the step size of the planned timetable. At intersections, multiple trains intersect and become a bottleneck in operation management. Therefore, by finely adjusting the parameters, the possibility of creating parameters that can output a plausible forecast timetable increases. However, if the minimum value of the specification is used at all places, the amount of calculation becomes large, so it is better to adjust the parameters in large increments except at the bottleneck point. As a result, it is possible to reduce the time required to calculate parameters while maintaining the plausibility of the predicted timetable.

In the case of the robustness evaluation scenario 109A, subtract the determined step width of the parameter from the given parameter (S307), and if the parameter is 0 or more (S308), predict the diamond (S309). If the parameter is less than 0, change the parameter to be iterated. This process is repeated until the predicted timetable recovers at the assumed point of the robustness evaluation scenario 109A (S310). The reason for subtraction is that, in the case of parameters whose data is designed in units of time, if the prediction is made under the condition that the departure time is delayed, the prediction is not made in the direction of increasing the total required time.

In the case of a steady scenario, the processing in the loop of S303a-S303b is the processing that uses the interval between the preceding and following trains for the purpose of changing the train order, and the processing that uses the planned timetable increment width for the purpose of matching with the planned timetable. may be executed in two steps. As a result, the time required for timetable prediction can be shortened, and parameters can be adjusted efficiently.

For each parameter and for each set of parameters, parameter values and their predicted timetables for outputting predicted timetables for recovering train delays according to the robustness evaluation scenario 109A are extracted (S311). Based on the extracted parameter values, a constraint expression indicating the value range for each parameter and for each set of parameters is obtained (S312). If no parameter values are found that can recover the train delay according to the robustness evaluation scenario 109A, the temporary parameter values are left as they are. A flag should be set to indicate that no parameter values were found that could recover the train delay.

Here, FIG. 15 will be described as an example. A graph 700 shows the adjustment of travel time between BC station and stop time at C station in evaluation scenario 2 .

In evaluation scenario 2, a delay of up to 30 seconds at B station is acceptable. In other words, the planned timetable includes a 30-second allowance from the time of departure from B station to the time of departure from C station, which is the recovery point. If all 30 seconds of spare time is used up only by the running time between B and C stations, the running time between B and C stations will be shortened by 30 seconds from the adjusted 120 seconds (702) in the planned timetable. A certain range of 90 seconds or less is a possible range. Similarly, the stop time at Station C is also within a possible range of 30 seconds or less, which is 30 seconds shorter than the adjusted 60 seconds (703) in the planned timetable. A line segment 701 connecting the points 702 and 703 can be drawn. Therefore, with the parameters alone, the travel time between B and C stations will be 120 seconds or less, and the stop time at C station will be 60 seconds or less, so that they will not be larger than the parameters already adjusted in the planned timetable. The sum of the running time between BC station and the stopping time at C station is a constraint expression that calculates 150 seconds or less.

FIG. 11 is a flow chart of a process (S400) for calculating a representative value from the set range of parameters that summarize a plurality of constraint expressions.

First, a set width is calculated by summarizing constraint expressions for a plurality of parameters (S401). FIG. 16 shows an example of when the steady scenario and the evaluation scenario 2 are used. The possible range of values is narrowed down from the relationship between parameters in multiple scenarios. An example of the part already described in T07i is the applicability range of the steady-state scenario and the robustness evaluation scenario 109A in FIG.

Next, the process branches at the steady scenario and the robustness evaluation scenario 109A. If only the steady scenario is used, the median value of each parameter is set as the representative value (S408), and the process S400 ends.

In the case of the robustness evaluation scenario 109A, the improvement history 105 is referred to and it is determined whether or not there is a location where the planned timetable is the same, a combination of set widths, or a location where the range is very similar (S402). If such an improvement history 105 exists, the spare time apportionment ratio to each parameter is created from the improvement history 105 (S403). The spare hourly apportionment ratio is a ratio for apportioning the spare hourly portion in the constraint formula calculated by each scenario. If such an improvement history 105 does not exist, the parameter application priority information 110 is referenced to determine whether the parameter exists (S404). If such parameter application priority information 110 exists, the spare hourly apportionment ratio to each parameter is created from the parameter application priority information 110 (S405). When the margin time division proportional division ratio is created by the improvement history 105 or the parameter, each parameter value is subtracted by the proportional division margin time from the maximum value of the setting width of the combination of parameters that determined the proportional division ratio. (S406). With such a proportional division ratio of the spare time, it is possible to suppress the bias to the parameter with the spare time, and increase the possibility that the constraint due to the control that cannot be grasped in detail is reflected in the parameter.

If there is no target location in either the improvement history 105 or the parameter application priority information 110, the value that minimizes the difference from the adjusted parameters in the planned timetable for all the parameters that require adjustment is adopted. For example, it is assumed that the constraint expression between the evaluation scenarios is obtained by the constraint expression shown in T07i in FIG. 16 in process S401. A graph shown in FIG. 15 is obtained by summarizing the constraint equations. In the process of S402, it is determined whether the parameter exists in T05 of FIG. This time, #201, #202, and #207 of T07 are targeted, and since the combination of parameters and departure/arrival stations is not found in T05, none of them apply. Therefore, it is determined as NO in S402, and the process proceeds to S404. In processing S404, it is determined whether the parameter exists in T04 of FIG. Since the parameters of stop time and running time are included in T04, it is determined as YES in S404. In the process S405, 4:5 is created as the spare time apportionment ratio of the stop time and the running time. In step S406, a parameter value of 4:5 on the line segment between points 702 and 703 in FIG. 15 is used as a representative value. This processing is executed in a multi-dimensional space with a plurality of parameters as axes, and the marginal time division apportionment ratio among the plurality of parameters is determined.

FIG. 12 is a diagram showing an example of an input screen 500 from the worker 98. As shown in FIG.

The input screen 500 includes a planned timetable input section 510 , a parameter input section 520 , a robustness evaluation scenario input section 530 and an OK button 501 . The planned timetable input unit 510 includes a timetable streak display unit 511 , a planned timetable timetable input unit 512 , and a planned timetable interval input unit 513 . The data input to the planned timetable input section 512 is recorded in the planned timetable 107 . After inputting the timetable of the planned timetable, the planned timetable is drawn on the streak display portion 511 . Parameter input section 520 includes an inter-station parameter input section 521 and a station parameter input section 522 . Data input to the parameter input unit 520 is recorded in the temporary parameters 108 . Robustness evaluation scenario input section 530 includes individual setting 531 and batch setting 532 . The robustness evaluation scenario 109A is entered in individual settings 531 or collective settings 532 . The data input in the collective setting 532 is expanded for each station and converted into the format of T03 in FIG. 5 when recorded in the robustness evaluation scenario 109A. When the confirmation button 550 is operated in a state in which all the data have been input, a data registration event occurs. If the input data is insufficient, an error indicating the missing part is presented to the operator 98 to urge correction.

FIG. 13 is a diagram showing an example of an output screen 600 presented to the operator 98. As shown in FIG.

The output screen 600 includes a predicted timetable line display portion 610 , a parameter display portion 620 , a robustness evaluation scenario selection portion 630 , a redraw button 640 , and a confirm button 650 . Parameter display section 620 displays improvement history 105 (table T05 shown in FIG. 7). The predicted timetable streak display section 610 draws the data selected by the robustness evaluation scenario selection section 630 . For example, when the second evaluation scenario is selected by the robustness evaluation scenario selection unit 630, a delay is drawn like a line 611 during a stop at B station #1. In the predicted diamond line display portion 610, the predicted diamond outputting the result selected in the width drawing of the parameter display portion 620 is drawn. For example, when the second parameter is selected in the parameter display section 620, detailed data 613 of travel time between stations BC is popped up. This allows the operator 98 to know the basis for proposing the parameters.

As described above, the train operation management system 99 of the embodiment of the present invention includes the planned timetable 107, the robustness evaluation scenario 109A, the evaluation index 109B, the equipment data 106, and the input unit 101 that receives the input of the initial values of the parameters. , a prediction unit 102 that predicts a diagram according to one or more evaluation scenarios and creates a predicted diagram, and a parameter narrowing unit 103 that compares the planned diagram 107 and the predicted diagram and narrows down parameter values that satisfy the evaluation index. Then, a train schedule is predicted using the narrowed down parameters, and a signal for controlling the track equipment is output according to the predicted train schedule, so the operator 98 is not familiar with the detailed operation of the train operation control system 99. Even if there is no detailed data on the situation or control system, plausible parameters can be created from parameter value candidates with a degree of freedom. As a result, the early release of the train operation management system 99 and the changes in the operation mode and facilities can be quickly followed, and the operation management work can be facilitated.

In addition, the robustness evaluation scenario 109A includes the delay occurrence point and the delay time, and the evaluation index 109B includes at least one of the delay recovery point and the total delay time. can be created to reduce the

Further, the parameter narrowing-down unit 103 narrows down the parameter values as the first evaluation index in which the total delay time is 0 in the first evaluation scenario in which no trains are delayed, and creates the first parameters. And, in one or more second evaluation scenarios in which any delay train, any delay occurrence point, and any delay time are determined, the delay recovery point is used as one or more second evaluation indicators, and the parameter values are narrowed down. , and the means for creating the second parameters are executed in order, so that it is possible to efficiently narrow down the parameters by guessing which parameters to change, and to shorten the time required for proposing parameters.

The parameter narrowing unit 103 includes a parameter value candidate creation unit 103A, a constraint expression calculation unit 103B, and a parameter value selection unit 103C. For each evaluation scenario, the parameter value candidate creation unit 103A creates parameter value candidates is created, the prediction unit 103 creates a predicted timetable based on the planned timetable 107, the robustness evaluation scenario 109A, and parameter value candidates, and the constraint expression calculation unit 103B determines that the predicted timetable satisfies the evaluation index Then, the parameter value selection unit 103C selects one parameter value that satisfies all the constraint expressions for one or more robustness evaluation scenarios. Parameter values independent of the person 98 can be calculated.

In addition, the parameter value selection unit 103C proportionally divides the delay time by the adjustment amount of the required time from the delay occurrence point of the delayed train specified in the evaluation scenario to the delay recovery point specified in the evaluation index. parameters, thus reducing the chances of violating control constraints.

Further, the parameter value selection unit 103C calculates the ratio of proportional division based on the parameter improvement history 105 generated by the execution result, so that the accuracy of similar parameters can be improved by execution.

In addition, since the parameter value selection unit 103C calculates the proportional division ratio based on the weight of proportional division of the surplus time set for each parameter type, the parameter can improve the accuracy of

In addition, the parameter value candidate creation unit 103C creates parameter value candidates by setting small increments for changing parameters at locations where the alignment becomes a bottleneck in operation. By changing the precision of the parameter values in the portion where the calculation is performed, the amount of computer resources can be suppressed, and the parameter values can be efficiently narrowed down.

It should be noted that the present invention is not limited to the embodiments described above, but includes various modifications and equivalent configurations within the scope of the appended claims. For example, the above-described embodiments have been described in detail for easy understanding of the present invention, and the present invention is not necessarily limited to those having all the described configurations. Also, part of the configuration of one embodiment may be replaced with the configuration of another embodiment. Moreover, the configuration of another embodiment may be added to the configuration of one embodiment. Further, additions, deletions, and replacements of other configurations may be made for a part of the configuration of each embodiment.

In addition, each configuration, function, processing unit, processing means, etc. described above may be realized by hardware, for example, by designing a part or all of them with an integrated circuit, and the processor realizes each function. It may be realized by software by interpreting and executing a program to execute.

Information such as programs, tables, and files that implement each function can be stored in storage devices such as memories, hard disks, SSDs (Solid State Drives), or recording media such as IC cards, SD cards, and DVDs.

In addition, the control lines and information lines indicate those considered necessary for explanation, and do not necessarily indicate all the control lines and information lines necessary for mounting. In practice, it can be considered that almost all configurations are interconnected.

98 operator 99 train operation management system 100 parameter creation system 101 input unit 102 prediction unit 103 parameter narrowing unit 104 result presentation unit 105 improvement history 106 facility data 107 plan diagram 108 provisional parameter 109A robustness evaluation scenario 109B evaluation index 110 parameter application priority Speed information 500 Input screen 600 Output screen 700 Graph of constraint equation calculation image for travel time between BC station and stop time at C station

Claims (10)

A train operation management system,
A computing device that executes a predetermined process, and a storage device that can be accessed by the computing device,
an input unit in which the arithmetic device receives input of a planned timetable, an evaluation scenario of the planned timetable, an evaluation index, equipment data, and initial values of parameters;
a prediction unit in which the computing device predicts a diagram according to one or more of the evaluation scenarios and creates a predicted diagram;
The arithmetic device has a parameter narrowing unit that compares the planned timetable and the predicted timetable and narrows down parameter values that satisfy the evaluation index,
A train operation management system, wherein the train operation management system predicts a train schedule using the narrowed-down parameters, and outputs a signal for controlling track facilities according to the predicted train schedule. The train operation management system according to claim 1,
The evaluation scenario includes a delay occurrence point and a delay time,
A train operation management system, wherein the evaluation index includes at least one of a delay recovery point and a total delay time. The train operation management system according to claim 1,
The parameter narrowing unit
In a first evaluation scenario in which no delayed trains are determined, as a first evaluation index in which the total delay time is 0, the parameter values are narrowed down to create a first parameter,
In one or more second evaluation scenarios in which an arbitrary delayed train, an arbitrary delay occurrence point, and an arbitrary delay time are determined, the parameter value is narrowed down with the delay recovery point as one or more second evaluation indicators. , a train operation control system characterized by creating a second parameter. The train operation management system according to claim 1,
The parameter narrowing unit has a parameter value candidate creation unit, a constraint expression calculation unit, and a parameter value selection unit,
For each evaluation scenario,
The parameter value candidate creation unit creates parameter value candidates,
The prediction unit creates a predicted timetable based on the planned timetable, the evaluation scenario, and the parameter value candidates,
The constraint expression calculation unit determines whether the predicted timetable satisfies the evaluation index, calculates the constraint expression satisfied by the parameter value,
The train operation management system, wherein the parameter value selection unit selects one parameter value that satisfies all the constraint expressions for one or more of the evaluation scenarios. The train operation management system according to claim 4,
The parameter value selection unit proportionally divides the delay time by the adjustment amount of the required time from the delay occurrence point of the delayed train specified in the evaluation scenario to the delay recovery point specified by the evaluation index, and obtains a second A train operation control system characterized by creating parameters. The train operation management system according to claim 5,
The train operation management system, wherein the parameter value selection unit calculates the proportional division ratio based on a parameter improvement history generated by an execution result. The train operation management system according to claim 5,
The train operation management system, wherein the parameter value selection unit calculates the ratio of proportional division based on the weight of the proportional division of surplus time set for each type of parameter. The train operation management system according to claim 4,
The train operation management system, wherein the parameter value candidate creation unit creates parameter value candidates by setting a parameter change step width small at a location where an alignment becomes a bottleneck in operation. A parameterization system comprising:
A computing device that executes a predetermined process, and a storage device that can be accessed by the computing device,
an input unit in which the arithmetic device receives input of a planned timetable, an evaluation scenario of the planned timetable, an evaluation index, equipment data, and initial values of parameters;
a prediction unit in which the computing device predicts a diagram according to one or more of the evaluation scenarios and creates a predicted diagram;
The parameter creation system, wherein the computing device has a parameter narrowing unit that compares the planned timetable and the predicted timetable and narrows down parameter values that satisfy the evaluation index. A parameter creation method executed by a parameter creation system, comprising:
The parameter creation system has an arithmetic device that executes a predetermined process and a storage device that can be accessed by the arithmetic device,
The parameter creation method is
an input procedure in which the arithmetic device receives input of a planned timetable, an evaluation scenario for the planned timetable, an evaluation index, equipment data, and initial values of parameters;
a prediction procedure in which the computing device predicts a diagram according to one or more of the evaluation scenarios and creates a predicted diagram;
A parameter creation method, wherein the arithmetic unit compares the planned timetable with the predicted timetable and narrows down parameter values that satisfy the evaluation index.

Images