JPH0592884A - Vibrating time risk inference device - Google Patents

Abstract

PURPOSE:To perform stop action at the time of earthquake occurrence or the like efficiently along the actual state of vibration by inputting the periodic data of vibration waveform and the vibrational acceleration data at the vibrating time to reason the risk caused by the vibration. CONSTITUTION:An inference device 1 is formed of a waveform detecting device 2 serving as a vibration detector for detecting acceleration waveform at the vibrating time, and a microcomputer 3. When an acceleration waveform signal is supplied from the waveform detecting device 2, whether its output level exceeds the set value level with the lapse of time is detected. The number of large peak value exceeding the set level and that of small peak value not exceeding the set level are respectively counted. Upon reaching the set time T, the fuzzy inference of risk is performed respectively by the count value on the time zone (periodic data of vibration waveform) and the count value on the peak value (vibrational acceleration data). In the antecedent part, the minimum operation is performed to compute compatibility, and the maximum operation is performed on the basis of the compatibility to compute the compatibility of the consequent part in the large-medium-small risks. A membership function is further added to draw inferences.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention is used, for example, in an automatic stop device attached to an elevator, a vehicle or the like for automatically stopping them in the event of an earthquake.

[0002]

2. Description of the Related Art Elevators and railway vehicles are equipped with an automatic stop device for detecting the vibration and automatically stopping the driving or traveling when strong vibration is applied during driving.

[0003]

However, such a conventional automatic stop device detects only the acceleration of vibration, and when the acceleration of vibration is large, it is automatically judged to be dangerous. Although the vehicle is stopped, there is a case in which even if the acceleration is the same, it may not be in a dangerous state, and there is a problem that it is not suitable for the actual situation of being used to stop driving or traveling in a dangerous situation.

The present invention has been made in view of the above problems, and it is an object of the present invention to provide an inference device that reliably infers the degree of danger when vibration occurs due to an earthquake or the like.

[0005]

In order to achieve the above object, according to the present invention, an inference device is provided with a cycle data creating means for creating cycle data within a set time of a vibration waveform based on data from a vibration detector. A vibration acceleration data creating means for creating vibration acceleration data within a set time based on the data from the vibration detector, and an inference means for inferring the risk of vibration by inputting the cycle data and the vibration acceleration data. It is configured as follows.

[0006]

According to the present invention, the risk of vibration is inferred by inputting the respective data of the period data of the vibration waveform and the vibration acceleration data.

[0007]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The present invention will be described below with reference to the drawings.

FIG. 1 is a block diagram showing the overall construction of an embodiment of a vibration risk inference apparatus of the present invention.

The inference apparatus 1 is composed of a waveform detecting device 2 as a vibration detector for detecting an acceleration waveform during vibration and a microcomputer 3, and the microcomputer 3 outputs its output in the acceleration waveform obtained from the waveform detecting device 2. A measuring means 4 for individually measuring each time by detecting whether or not the set value level is exceeded, a first counting means 5 for counting the times in the same time zone by the measurement, and an acceleration waveform with time. A peak value detecting means 6 for detecting each peak value, a second counting means 7 for counting those peak values having the same output level, and a setting time measuring means 8
Then, the operation as the inference means 10 for performing the fuzzy inference of the risk degree based on the count values obtained by the first and second counting means 5 and 7 within the set time is performed. The measuring means 4, the first counting means 5 and the time counting means 8 constitute a period data creating means, and the peak value detecting means 6, the second counting means 7 and the time counting means 8 constitute an acceleration data creating means.

Next, the inference operation of the inference apparatus 1 will be described with reference to the flowchart of FIG. 2 showing the processing operation of the microcomputer.

First, when an acceleration waveform signal as shown in FIG. 3 is given from the waveform detecting device 2, it is detected whether or not the output level thereof exceeds a set value level over time (step 1), and the setting is made. The lengths of time zones t1, t2, ... Separated by the value levels are measured (step 2). The times t1, t2, ... Have values from 0.1 sec to 1.0 sec. Next obtained peak value p
1, p2 ... Are detected over time (step 3), and it is detected whether or not each peak value exceeds a set value level (step 4).

.. are counted in the same time zone in each of the time zones t1, t2 ... (Step 5), and at the respective peak values p1, p2. The number of values is counted (step 6).

The count value (corresponding to the frequency) in the above time zones t1, t2, ... Is t in the short time zone (up to 0.2 sec).
[4] of 4, t5, t7, and t8, medium time zone (0.3 to
0.6 sec) is [4] of t1, t2, t3, and t9, and a long time period (0.7 sec-) is [1] of t6. In addition, the count value in the above peak value is p
[5] of 1, p2, p3, p8, p9, the small peak value is p
[4] of 4, p5, p6 and p7.

When the set time T is reached (step 7), fuzzy inference of the degree of danger is performed based on the count value (vibration waveform period data) related to the time zone and the count value (vibration acceleration data) related to the peak value (step 8). ). Then, the process returns to step 1 again, and the next inference is performed based on the data at the next set time T.

An example of fuzzy inference based on each of the above measured values will be described.

FIG. 4 is a membership function (A) of the antecedent part concerning the time zone and the peak value used in the fuzzy inference.
(B), (C), and (D) are respectively shown, and the membership function (A) is a membership function related to a short time zone, the horizontal axis is the short time zone count value, the vertical axis is the goodness of fit, and the large count value is It has three membership functions, medium and small. The membership function (B) is a membership function relating to a long-term band, and the horizontal axis represents the long-term band count value and the vertical axis represents the goodness of fit, and three membership functions of large, medium, and small count values are provided. The membership function (C) is a membership function related to a small peak value, and the horizontal axis represents the small peak value count value,
The vertical axis is the goodness of fit, and three membership functions of large, medium, and small count values are provided. The membership function (D) is a membership function related to the large peak count value, and the horizontal axis is the large peak count value and the vertical axis is the goodness of fit, and two membership functions of large count value and medium count value are provided.

FIG. 5 is a membership function relating to the degree of risk, where the horizontal axis is the degree of danger and the vertical axis is the degree of conformity.
It has three membership functions, medium and small. The definitions of high, medium, and small risk levels are, for example, “large emergency stop of target device”, medium “immediately stop target device.”, Small “target as it is. It is possible to operate the device. "

The fuzzy inference rule is composed of the following rule group consisting of an antecedent part regarding the count value of the time zone and the peak value, and an antecedent part of the risk degree as the inference result.

(1) If short time zone = medium and large peak value large then high risk level (2) If short time zone = medium and large peak value medium then high risk level (3) If short time zone = large and large Large peak value then the medium degree of risk (4) If long term = large and large peak value large then the high degree of risk (5) If short time zone = small and large large peak value then the low degree of risk (6) If short term zone = Small and small peak value large then medium risk (7) If short time zone = large and small peak value medium then low risk degree (8) If short time zone = small and small peak value small then low risk degree (9) If long time zone = large and small peak value large then high risk level (10) If long time zone = medium and small peak value large then medium risk level (11) If long time zone = small and large peak value small The degree of danger The above rules have been created in view of the characteristics of the period and acceleration of seismic waves and shock waves.

The fuzzy reasoning is such that, given the time zone and the count value of the peak value, the MIN in each antecedent part.
Calculation is performed to calculate the fitness of each, and MAX calculation is performed based on the fitness to calculate the suitability of the consequent part with high risk, medium risk, and low risk. , The degree of danger (0 to
100) is inferred.

When the inference apparatus 1 is installed in, for example, an automatic elevator stop device, when a high level of danger above a set level is inferred, the automatic stop device outputs a stop signal for the drive motor of the elevator. Elevators are stopped to avoid danger such as an earthquake.

[0022]

The present invention is configured as described above. Since the data of the period of the vibration waveform and the vibration acceleration data at the time of vibration are input to infer the risk due to vibration, the present invention is more accurate. It is now possible to infer the degree of danger associated with vibration, and by using this inference device, the stopping operation at the time of an earthquake such as an elevator can be efficiently performed according to the actual state of vibration.

[Brief description of drawings]

FIG. 1 is a block diagram of an inference device according to the present invention.

FIG. 2 is a flowchart for explaining the operation of the present invention.

FIG. 3 is an explanatory diagram of a vibration waveform.

FIG. 4 is an explanatory diagram of a membership function used on the input side of fuzzy inference.

FIG. 5 is an explanatory diagram of a membership function used on the output side of fuzzy inference.

[Explanation of symbols]

DESCRIPTION OF SYMBOLS 1 Inference device 2 Waveform detection device (vibration detector) 4 Measuring means (period data creating means) 5 First counting means (period data creating means) 6 Peak value detecting means (acceleration data creating means) 7 Second counting means (acceleration) Data creating means) 8 Clocking means (period data creating means, vibration acceleration data creating means)

Claims (1)

[Claims] 1. A cycle data creating means (4) (5) (8) for creating cycle data within a set time of a vibration waveform based on data from the vibration detector (2), and a vibration detector (2). Vibration acceleration data creating means (6) (7) (8) for creating vibration acceleration data within a set time based on the data from, and inferring the risk of vibration by inputting the cycle data and the vibration acceleration data. An inference device (10), and a risk inference device at the time of vibration, comprising: