JPH04358961A - Train-approach detector - Google Patents

Abstract

PURPOSE:To secure the safety for a job in a railway by detecting a train approach surely and outputting a signal of train-approach information at once. CONSTITUTION:This detector is provided with a sound detecting element 1, detecting a sound being generated with a train approaching and outputting an analog electric signal showing this sound, a filter 2 restricting a band of this electric signal, an analog-to-digital converter 3 for converting this electric signal, whose band is restricted by this filter 2, into a digital signal, and a computing element 4 which judges the presence of a time increment tendency and periodicity as to the digital signal, and outputting a signal of train-approach information only when this digital signal is so judged that it has the time increment tendency and the periodicity either.

Description

[Detailed description of the invention]

0001

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a train approach detection device used for safety purposes in railway work and the like.

0002

[Prior Art] Although safety confirmation is the most important matter when working in the dangerous area of railway tracks, conventionally, detection of approaching trains during construction on railway tracks has been done only by the worker's viewing, which is a safety concern. has become a big problem.

0003

SUMMARY OF THE INVENTION An object of the present invention is to eliminate the above-mentioned safety problems by providing means for reliably detecting the approach of a train and outputting train approach information.

0004

[Structure of the Invention] The train approach detection device according to the present invention is shown in FIG.
As shown in Figure 2, the sound obtained from the rails as the train approaches is detected and an analog electrical signal (
an acoustic detection element 1 that outputs an acoustic signal (hereinafter referred to as an "acoustic signal"), a filter 2 that limits the band of the acoustic signal, and an A/D converter that converts the acoustic signal whose band has been limited by the filter 2 into a digital signal. 3 and this A/D converter 3
The A/D converted digital signal is determined to have a temporal increasing tendency and to determine the presence or absence of periodicity, and it is determined that the digital signal has a temporal increasing tendency and periodicity. It is comprised of a computing unit 4 that outputs train approach information only when the train is approaching.

[0005]

Embodiments Hereinafter, embodiments of the present invention will be described with reference to the drawings.

In FIG. 1, a sound detection element 1 detects sound obtained from a rail and converts it into an acoustic signal (analog electrical signal). The ``sound obtained from the rails'' used here refers to the sound generated by collisions between wheels and rail joints while the train is running, in addition to the sounds caused by external disturbances. Normally, rail joints have a slight gap to accommodate expansion and contraction of the rail due to temperature changes, so collisions between wheels and rail joints generate sound, which propagates through the rail. Therefore, the sound detection element 1 is attached to the rail to detect the sound obtained from the rail, but when a train passes on the rail on which the sound detection element 1 is installed, the sound level changes as shown in FIG. 2. That is, as the train approaches the installation point of the sound detection element 1, the sound level increases (period A), and remains constant while passing over the installation point (
period B), and gradually descends after the train passes (period C). Note that the natural sound in FIG. 2 is sound caused by disturbance.

Furthermore, since the distance between the wheels of a train and the length of one rail are constant, the sound generated as the train runs has periodicity. FIG. 3 is an enlarged representation of the level fluctuation during the sound level rising period in FIG. 2, and the sound signal detected by the approach of the train has a period of time length T. Therefore, by limiting the band of the acoustic signal using the filter 2, the periodicity of the acoustic signal becomes more apparent.

The arithmetic unit 4 determines whether the digital signal from the A/D converter 3 contains information indicating the approach of a train as shown in FIGS. 2 and 3 as follows. (1) Determining whether there is a tendency for the sound level to increase over time (2)
Determining the presence or absence of periodicity

The determination operations (1) and (2) above will be explained below using the flowchart shown in FIG.

In the arithmetic unit 4, first in step 41, the digital signal output from the A/D converter 3 is input. Next, the process proceeds to step 42, where the digital signal (new signal) input in step 41 is compared with the past digital signal input last time, and it is determined whether or not there is a tendency for the sound level of the digital signal to increase over time. Do this. Here, only when the state in which the sound level of the new signal is higher than the past sound level continues for a certain period of time, the condition is satisfied is outputted, otherwise the process returns to step 41.

Next, in step 42, when a temporal increasing tendency in the sound level is recognized, the process proceeds to step 43, where it is determined whether or not the input digital signal has periodicity. This determination of the presence or absence of periodicity can be easily realized, for example, by using a method of correlating input digital signals. Then, the process proceeds to step 44 only when it is determined that there is periodicity, and otherwise returns to step 41. In step 44, when the condition is satisfied from step 43, train approach information is output.

[0012] Here, a method of correlating input digital signals when determining the presence or absence of periodicity will be explained. Taking a correlation means quantifying the similarity between two physical quantities, and the higher the similarity between the two, the larger the correlation value. Below, a description will be given of how to take the correlation between the acoustic signal widths T1 and T2. Here, in order to facilitate understanding, it is assumed that a simple acoustic signal is used, and that the only difference between the two signals is the level difference between 13 of the acoustic signal S1 and 23 of the acoustic signal S2 in FIG. However, both signals S1 and S2
It is assumed that both are digital series values consisting of 50 data. In addition, each data is 11-1 to 1 as shown in FIG.
1-10, 12-1 to 12-10, ...15-1 to 15-
Let's call it 10.

One of the two acoustic signals is fixed, the other is shifted by one data, multiplied, and the multiplication results are added. However, the portion that does not overlap with the fixed side (the dotted line portion in FIG. 7) is moved to the beginning for calculation. FIG. 7 shows a state in which the moving side is shifted by 30 data. The calculation method at this time is S30=11-1~23-1+11-2~23-2+1
1-3×23-3+…+13-10×25-10+14
-1×21-1+14-2×21-4+...+15-9×
22-9+15-10×22-10.

[0014] In this way, the total product sum S50 with the data shifted by 50 is determined from the total product sum S1 with the data shifted by 1, and these sums Sc are expressed by the following equation 1.
is the correlation value between the acoustic signal S1 and the acoustic signal S2.

[Math 1]

Of course, if the two physical quantities to be correlated are exactly the same, the correlation value will be maximum, and if the magnitude of the level or the sign is different, the correlation value will decrease according to the degree. To find the correlation value between the above two acoustic signals S1 and S2, the number of multiplications is 50 x 50 = 2500, and the number of additions is 50.
Although it is necessary to perform 2500 calculations (×50), such calculations are extremely easy if a microcomputer is used.

Note that in the flowchart of FIG. 4, the order of steps 42 and 43 may be reversed. Further, step 42,
Both processes in step 43 may be performed alternately in a short period of time, and when it is recognized that the conditions are met a certain number of times or more in both judgment results, the process may proceed to step 44 and output the train approach information.

As is clear from the above description, in this embodiment, by incorporating the increasing tendency and periodicity of the digital signal into the criteria for determining whether a train is approaching, it is possible to distinguish between sounds caused by disturbances and sounds caused by approaching trains. , it is possible to provide accurate train approach information and prevent malfunctions.

Furthermore, the train approach detection device according to the present invention has the following features:
It can be easily realized by using a microcomputer, DSP (digital signal processor), etc., and if it is made into an IC, it can be easily made smaller and lighter, so it is possible to construct a portable train approach detection device. It becomes possible.

Furthermore, by combining with a transmitter/receiver,
For example, it becomes possible to monitor train approach information at a point several kilometers away from the work site on the receiving side.

[0020]

[Effects of the Invention] According to the present invention, it is possible to early detect train approach information during track work, so
Not only does it significantly improve safety, it also contributes to labor savings. Furthermore, the device of the present invention can be easily integrated into an IC, and is lightweight and compact, so it has the advantage of being suitable for portability.

[Brief explanation of the drawing]

FIG. 1 is a block diagram showing the configuration of a train approach detection device according to the present invention.

FIG. 2 is a diagram showing changes in sound level as a train passes at one point on the rail.

FIG. 3 is an enlarged diagram illustrating level fluctuations during the sound level rising period in FIG. 2;

FIG. 4 is a flowchart showing the operation of the arithmetic unit in FIG. 1;

FIG. 5 is a diagram illustrating how to correlate two acoustic signals.

FIG. 6 is a diagram illustrating how to correlate two acoustic signals.

FIG. 7 is a diagram illustrating how to correlate two acoustic signals.

[Explanation of symbols]

1 Acoustic detection element 2 Filter 3 A/D converter 4 Arithmetic unit

Claims (1)

[Claims] 1. A sound detection element that detects sound generated as a train approaches and outputs an analog electrical signal representing the sound; a filter for limiting the band of the electrical signal; and a filter that limits the band by the filter. An A/D converter converts the electrical signal into a digital signal, and the digital signal is determined to determine whether or not there is an increasing tendency over time, and whether or not there is periodicity, and whether or not the digital signal has an increasing tendency over time. What is claimed is: 1. A train approach detection device comprising: a computing unit that outputs train approach information only when it is determined that the train approach information has periodicity.