JPH05112243A - Train automatic operation system - Google Patents

Abstract

PURPOSE:To provide a train automatic operation system which facilitates setting, such as scheduled time running, achieves a plurality of control targets, such as a target speed, stop precision, and a riding sensation, with a high balance, and performs excellent operation like a skillful operator. CONSTITUTION:Operation of a train is executed such that a notch command to a train drive part and a train 3 is delivered from an operation control part 2 so that an actual running speed follows a target speed in a target pattern memory 1. A fuzzy control part 4 changes the control parameter of the operation control part 2 at a real time through fuzzy inference according to a target speed pattern and a present running state. In this way, robust properties, such as a riding sensation and stop precision, can be improved white the advantages of a target speed pattern follow-up type control, such as scheduled time running, are favorably utilized.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to an automatic train operation system for automatically operating a train.

[0002]

2. Description of the Related Art In recent years, an automatic train operation system has been developed which automates train operation by a computer and enables one-man operation or even unmanned operation.

In such an automatic train operation system, a speed pulse generated by a tacho generator, an ATC (Au
tomatic Train Control) It has a function to output a control command (notch command) to the drive system and braking system by inputting the speed limit signal from the on-board system, the ground child detection signal, and the traveling information from the ground operation management system. ing.

In the conventional automatic train operation system, proportional control or the like is performed so that the train speed follows a target speed pattern determined for the position. However, in such a train automatic operation system, it is only the speed deviation between the target speed pattern and the actual speed of the train that is evaluated during running of the train that balances other control targets such as stop accuracy and ride comfort. There was a problem that it was difficult to achieve well.

For example, when stopping a train, it is necessary to accurately stop the train at a predetermined position because of the boarding position and the like set on the platform. However, a conventional train that performs proportional control based on a target speed pattern, etc. In an automatic driving system, it is necessary to increase the proportional gain in order to obtain sufficient stopping accuracy. In this case, if the notch operation is performed frequently, the riding comfort will be impaired because the actual train speed is adjusted to the target speed pattern. There was a problem such as.

On the other hand, an automatic train operation system has been developed which uses rule-based control without using the target speed pattern as described above. However, in the conventional automatic train operation system using such rule-based control, it is not known in advance how the train speed should be changed, so it is necessary to drive at a fixed time that the destination is reached by a predetermined time. There is a problem that it is difficult to set such as.

[0007]

As described above, in the conventional train automatic operation system, the system that uses the method of controlling tracking to the target speed pattern balances other control targets such as stop accuracy and riding comfort. There was a problem that it was difficult to achieve well. Further, in the system using the rule-based control, there is a problem that it is difficult to set the constant time running or the like. The present invention has been made in response to such a conventional situation, and can easily set a fixed-time running or the like and achieve a plurality of control targets such as target speed, stopping accuracy, and riding comfort in a well-balanced manner. You can
It is an object of the present invention to provide an automatic train operation system capable of performing good operation similar to that of a skilled operator.

[0008]

That is, an automatic train operation system of the present invention is an operation for sending a command signal based on a predetermined control parameter according to a deviation between a preset target speed pattern and an actual traveling speed. In a train automatic operation system including a control unit and a train drive unit that executes the operation of a train based on the command signal from the operation control unit, the actual traveling speed and the actual traveling position are sequentially input, and these From the traveling speed and traveling position of the target speed pattern, based on a preset fuzzy rule, the value of the control parameter of the operation control unit is determined,
It is characterized by having a fuzzy control unit for changing.

[0009]

In the automatic train operation system of the present invention having the above-described structure, the fuzzy control unit determines the optimum value of the control parameter of the operation control unit such as the proportional gain according to the target speed pattern and the current traveling condition. , The control parameter is always adjusted to the optimum value according to the driving situation. As a result, it is possible to improve, for example, the riding comfort and the stopping accuracy while taking advantage of the target speed pattern tracking type control.

[0010]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS The details of an automatic train operation system of the present invention will be described below with reference to the drawings for one embodiment.

FIG. 1 shows a main configuration of an automatic train operation system according to an embodiment of the present invention. In the figure, 1 is a target pattern memory table, 2 is an operation control unit, 3 is a train drive unit and a train. Reference numeral 4 is a fuzzy control unit.

The target pattern memory table 1 holds a target speed pattern predetermined for a position based on an operation plan or the like. This target speed pattern is given so as to be uniquely determined for the position of the train, as shown in FIG. In addition, in FIG. 2, A is an acceleration section, B is a coasting section,
C is a brake section.

The operation control unit 2 also inputs information about the current position of the train from the train drive unit and the train 3, and sets a corresponding target speed on the basis of the information about the current position of the train. Read from. Then, a notch command (command for adjusting the current value of the drive motor and the brake strength) is output according to the deviation between the target speed and the current speed from the train drive unit and the train 3.

In this embodiment, the operation control unit 2 is composed of a proportional control unit 2a and a notch fixing unit 2b. Then, in the proportional control unit 2a, a predetermined proportional gain k
Based on, the notch command is calculated from the deviation between the target speed and the current speed. Further, the notch fixing portion 2b is for suppressing the change of the notch for improving the riding comfort, and prohibits the notch change within the notch fixing time ft by the notch change suppressing algorithm described below. fe is the limit of deviation of acceleration / deceleration.

If t <ft and derr <fe then Maintain the current notch: Notch change suppression algorithm t: Notch maintenance time derr: Deviation between the currently output acceleration / deceleration value and the acceleration / deceleration instruction value by proportional control. The control unit 4 includes a target pattern feature amount extraction unit 4a, a fuzzy inference input value calculation unit 4b, a fuzzy inference unit 4c, and a fuzzy rule base 4d.

The target pattern feature amount extraction unit 4a reads the target pattern from the target pattern memory table 1 and extracts the feature amount. The characteristic amount of the target pattern includes the overall shape of the target traveling pattern, the acceleration section, the coasting section, the arrangement of the brake sections, and the size thereof.

The fuzzy inference input value computing unit 4b receives the feature amount extracted by the feature amount extracting unit 4a of the target pattern, and information about the current position and speed of the train from the train driving unit and the train 3. It Then, the fuzzy inference input value calculation unit 4b calculates the current traveling situation as a numerical value from these pieces of information. The calculation result and the current speed of the train are input to the fuzzy inference unit 4c as the current traveling status.

The following information can be given as specific examples of the current driving situation. Here, the reference acceleration / deceleration is the acceleration / deceleration of the target pattern, and the traveling mode is classified into acceleration, coasting, and deceleration according to the value of the reference acceleration / deceleration.

(1) Extra distance to the stop point
(2) Current reference acceleration (running mode) (3) Extra distance to the next running mode change point (4) Next reference acceleration (running mode) (5) Extra speed up to ATC speed limit (6)
Delay time from target arrival time (7) Speed deviation
(8) In the current speed fuzzy inference unit 4c,
The control parameter of the operation control unit 2 is adjusted by these information and the fuzzy rule in the fuzzy rule base 4d. In the present embodiment, the control parameters to be adjusted include various control parameters as shown below.

(1) Proportional gain (2) Dead band width of proportional control (3) Notch fixing time In this embodiment, the notch fixing time ft of the notch fixing portion 2b and the notch fixing time ft A case where the proportional gain k of the proportional control unit 2a is adjusted will be described.

In this case, the characteristic amount of the target speed pattern is the position where the traveling mode (acceleration, coasting, deceleration) is changed, and the following three amounts are used as inputs to the fuzzy inference unit 4c.

(1) Distance dx to the travel mode change point
(Distance from the current position to the closer one of the next drive mode change position and the current drive mode start position) (2) Reference acceleration / deceleration Ra of the current target speed pattern (can be calculated from the target speed pattern) ( 3) Current speed V Further, for example, the following fuzzy rules are used.

If dx = small and v = small and Ra = negative then ft = medium and k = large if dx = small and v = medium and Ra = negative then ft = medium and k = medium if dx = small and v = Large and Ra ＝ Negative then ft = Small and k = Medium if dx ＝ Medium and Ra ＝ Negative then ft = Medium and k = Medium if dx ＝ Large and Ra ＝ Negative then ft = Medium and k = Medium if v ＝ Small and Ra ＝ zero then ft = medium and k = large if dx ＝ small and Ra ＝ zero then ft = medium and k = small if Ra ＝ zero then ft = somewhat large and k = small if dx ＝ small and Ra ＝ positive then ft = Medium and k = Small if v = Small and Ra = Positive then ft = Large and k = Zero if Ra = Positive then ft = Somewhat large and k = Small The above fuzzy rules are previously fuzzy rule base 4d
The fuzzy inference unit 4c makes an inference according to this fuzzy rule.

In the automatic train operation system having the above structure, the proportional control unit 2a of the operation control unit 2 controls the actual speed of the train to match the target speed pattern in the target pattern memory table 1. That is, these deviations are multiplied by the proportional gain k so that the deviation between the target speed and the actual speed becomes zero, and the notch command corresponding to this value is sent out. At this time, if the notch operation is repeated frequently, the riding comfort is impaired. Therefore, in the notch fixing portion 2b, the notch change within the notch fixing time ft is prohibited by the change suppressing algorithm described above.

At the same time, the fuzzy reasoning unit 4c
According to the above-mentioned fuzzy rule, the values of the proportional gain k and the notch fixing time ft that are appropriate for the driving situation are obtained, and the values of the proportional gain k of the proportional control unit 2a and the notch fixing time ft of the notch fixing unit 2b are sequentially set to this appropriate value. Change the value.

If the above-mentioned fuzzy rule is used, the proportional gain k is small at the time of acceleration and coasting, the notch fixing time ft is set to be long, and frequent notch operations are eliminated to improve riding comfort. it can. Further, when the vehicle is stopped, the proportional gain is increased, and by shortening the notch fixing time before and after the travel mode change position, it is possible to improve the followability and the stopping accuracy.

As described above, according to the automatic train operation system of the present embodiment, the operation control unit 2 controls so that the actual traveling speed follows the target speed pattern, so that it is easy to set a fixed time traveling or the like. It can be carried out. Further, by changing the control parameter of the operation control unit 2 in real time by fuzzy reasoning according to the target speed pattern and the current traveling situation, while taking advantage of the target speed pattern following type control such as scheduled traveling performance, It is possible to improve riding comfort and robustness of stopping accuracy.

[0028]

As described above, according to the train automatic operation system of the present invention, it is possible to easily set a fixed time running and the like, and to control a plurality of target speeds, stopping accuracy, riding comfort and the like. The goal can be achieved in a well-balanced manner, and the same good operation as a skilled operator can be performed.

[Brief description of drawings]

FIG. 1 is a diagram showing a schematic configuration of an automatic train operation system according to an embodiment of the present invention.

FIG. 2 is a diagram showing an example of a driving target pattern.

[Explanation of symbols]

 1 Target pattern memory 2 Operation control unit 2a Proportional control unit 2b Notch fixing unit 3 Train drive unit and train 4 Fuzzy control unit 4a Target pattern feature extraction unit 4b Fuzzy inference input value operation unit 4c Fuzzy inference unit 4d Fuzzy rule base

Claims (2)

[Claims] 1. An operation control unit for transmitting a command signal based on a predetermined control parameter according to a deviation between a preset target speed pattern and an actual traveling speed, and the command signal from the operation control unit. In a train automatic operation system including a train drive unit that executes the operation of a train based on the actual traveling speed and the actual traveling position are sequentially input, from these traveling speed and traveling position and the target speed pattern, An automatic train operation system having a fuzzy control unit that determines and changes the value of the control parameter of the operation control unit based on a preset fuzzy rule. 2. An operation control unit for transmitting a command signal based on a predetermined proportional gain according to a deviation between a preset target speed pattern and an actual traveling speed, and the command signal from the operation control unit. In a train automatic operation system including a train drive unit that executes the operation of a train based on the actual traveling speed and the actual traveling position are sequentially input, from these traveling speed and traveling position and the target speed pattern, An automatic train operation system having a fuzzy control unit that determines and changes the value of the proportional gain of the operation control unit based on a preset fuzzy rule.