JP2575294Y2 - Vibration analyzer - Google Patents

Description

[Detailed description of the invention]

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a vibration analyzing apparatus, and more specifically, to an analyzing apparatus for determining whether a detected vibration is caused by an earthquake or other.

[0002]

2. Description of the Related Art A conventional vibration analyzer is disclosed in, for example, JP-A-4-38427. That is, an input wave such as a seismic wave is converted into a pulse wave, and the number of pulses generated within a fixed time of the converted pulse signal, the time during which the pulse is on, the time during which the pulse is off, and the time between these pulses Fuzzy inference is performed using the on-time ratio and the off-time ratio as features, and it is determined whether or not the input wave is a seismic wave.

[0003]

However, in the above-described conventional vibration analysis apparatus, a large number of types of feature amounts are required, and all the pulses generated within the predetermined time are extracted to extract the respective feature amounts. The number of occurrences, the on-time of each pulse, and the off-time between the preceding and succeeding pulses must be measured and compared with each other based on the measurement result to obtain a feature amount. The work becomes complicated. In addition, the number of fuzzy rules to be performed using the obtained feature amount also increases. As a result, the analysis processing becomes complicated and complicated, and the processing takes a long time.

The present invention has been made in view of the above background, and an object of the present invention is to provide a vibration analyzing apparatus capable of reducing the number of features to be used and analyzing a vibration waveform at high speed. To provide.

[0005]

In order to achieve the above-mentioned object, in a vibration analyzing apparatus according to the present invention, a signal generating circuit for converting a vibration waveform into a plurality of binary waveforms by a plurality of different threshold processing sections. And the time from the generation of the first pulse of the binary waveform converted by the low threshold to the generation of the first pulse of the binary waveform converted by the high threshold, and the high threshold A feature amount extracting unit that extracts an off time from a first pulse to a second pulse of the converted binary waveform as a feature amount, and a type of the vibration waveform using the feature amount extracted by the feature amount extracting unit as an input Determining means for determining the

[0006]

When a vibration is generated, an analog wave corresponding to the vibration is input to a signal generation circuit. Then, in this signal generation circuit, a plurality of binary waveforms are formed by different threshold values, and are sent to the next-stage feature value extracting means, where a predetermined feature value is extracted. Then, based on the extracted feature amount, the determination unit determines whether the vibration is caused by an earthquake. That is, in the case of an earthquake, first, there is a relatively small vibration (a first pulse of a binary waveform converted by a low threshold is generated), and after a certain period of time, a full-scale large shaking (a high threshold) The first pulse of the binary waveform converted by the value is generated). Therefore, if the period is long, the possibility of an earthquake is large. In addition, since such a large vibration is turned on and off at a predetermined interval, the possibility of an earthquake is large even when the off-time from the first pulse to the second pulse of the binary waveform converted by the high threshold is long. . On the other hand, in the case of an artificial impact, a large vibration is suddenly generated, and the vibration is generated at a relatively short cycle. Therefore, when both the period and the off time are short, the possibility of artificial vibration is large.

[0007]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Preferred embodiments of the vibration analyzing apparatus according to the present invention will be described below in detail with reference to the accompanying drawings. FIG. 1 shows an embodiment of the vibration analysis device according to the present invention. As shown in the figure, the vibration analysis device 10 of the present invention is connected to the output side of a vibration detector 11 such as an automatic horizontal seismic sensor. When the vibration detector 11 receives vibration, its output voltage fluctuates in proportion to the magnitude of the vibration, and its output waveform is an analog wave. In the device 10 of the present invention, the analog wave is used as input to determine whether the cause of the waveform is an earthquake or an impact caused by an artificial operation, and the analog wave is sent to the signal generation circuit 12. Is entered.

This signal generating circuit 12 is composed of two comparators 12a and 12b having different thresholds arranged in parallel, the threshold of the first comparator 12a is set to 4.0V, and , The threshold value of the second comparator 12b is 2.0
V is set. Thus, in each comparator, the analog wave is binarized and converted into an on / off pulse wave. That is, these comparators 12a and 12b become a threshold processing unit.

The outputs of the comparators 12a and 12b are input to the feature extraction unit 13 at the next stage. This feature value extraction unit 13 extracts a predetermined feature value based on the above two pulse data, and is specifically as follows. That is, when vibration occurs, as shown in FIG.
When the voltage is lower than V, a pulse is generated only in the pulse data 2 generated by the second comparator 12b, and when the voltage is 4 V or higher, the first comparator 1
A pulse is also generated in the pulse data 1 generated by 2a. Even when both pulses are generated at the same time, the generated pulse width (ON time) of the pulse data 2 is larger than that of the pulse data 2, and after the pulse of the pulse data 2 rises, the pulse data After the 1 pulse rises and the pulse of the pulse data 1 falls, the pulse of the pulse data falls.

Here, a calculation is performed to detect the vibration and to determine the characteristic amount of the pulse in both pulse data generated during a certain period after the first pulse (first pulse) P21 of the pulse data 2 is generated (rising). Target data,
Time (lag time) RT from the generation of the first pulse P21 to the generation of the first pulse P11 of the pulse data 1
And the OFF time (the time from the fall of the first pulse P11 to the rise of the second pulse P12) OFFT of the pulse data 1 until the first pulse P11 to the second pulse P12 are generated. Find the feature value.

The first pulse PZ21 of the pulse data 2
If the first pulse of the pulse data 1 does not occur even after a certain period of time has elapsed since the occurrence of, the above RT and OFFT are both set to zero.

The feature values RT and OF obtained as described above
The FT is input to the determination unit 14 at the next stage. The discriminating unit 14 performs fuzzy inference according to the membership function shown in FIG. 3 and the following rules with the given feature amount as an input, and determines whether the detected vibration waveform is due to an earthquake or not. To determine if it is artificial.

[0013]

[Table 1] <Rules> 1. IF OFFT IS BIG THEN DEC = EQ 2. IF RT IS BIG THEN DEC = EQ IF OFF IS IS SML AND IF RT IS SML THEN DEC = UEQ where DEC: determination result, EQ: earthquake, UEQ: impact (artificial vibration, etc.).

That is, when a normal earthquake occurs, first, vibration (initial tremor) associated with a longitudinal P wave having a high propagation speed but low energy (relatively small fluctuation) occurs, and thereafter the propagation speed is low but the energy is large. Oscillation is caused by the transverse wave (relatively large swing) S wave. And the vibration accompanying the transverse wave becomes like a relatively large swell. Therefore,
In the case of an earthquake, after a relatively small vibration (detected only by the second comparator 10b) first, a full-fledged large swing comes after a certain period of time. Therefore, during that period,
In other words, if the RT is large, it can be determined that an earthquake has occurred.
The above rule 1 is defined. Further, since such a large vibration is turned on and off at a predetermined interval, it can be determined that an earthquake has occurred even when the off time is long as in the rule 2.

On the other hand, for example, when an artificial shock is applied to the vibration detector 10, the vibration suddenly becomes large, the output voltage from the vibration detector 10 becomes large, and the vibration fluctuates at a relatively short cycle. Therefore, as in Rule 3, when both RT and OFFT are SML, it is determined that the vibration is artificial.

In this embodiment, after a certain period of time has elapsed after the generation of the first pulse of the pulse data 2, the apparatus is prepared for the next vibration analysis. That is, the pulse generated by the pulse data 2 after the elapse of the certain period becomes the first pulse P21.

Further, as is apparent from the above rule, the threshold value of the first comparator 12a is such that a pulse is not generated by the vibration associated with the P wave, but is generated by the vibration associated with the S wave. And the second comparator 12b
Is desirably set to a value such that a pulse is generated by the vibration accompanying the P-wave.

In the above-described embodiment, the judgment in the discriminating section is performed by using fuzzy inference. However, in the present invention, the comparator and the AND / OR are combined with the fact that the feature amount can be reduced. By appropriately combining the elements, it may be determined whether or not the earthquake is an earthquake based on whether each of the feature amounts is larger or smaller than a predetermined value.

[0019]

As described above, in the vibration analyzing apparatus according to the present invention, the number of feature values used for discriminating the vibration is small, and the judgment is performed using several pulses generated immediately after the start of the vibration. Therefore, the analysis of the vibration waveform can be performed at high speed.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a preferred embodiment of a vibration analysis device according to the present invention.

FIG. 2 is a diagram illustrating an operation of a feature amount extraction unit.

FIG. 3 is a diagram illustrating a membership function in a determination unit.

[Explanation of symbols]

 DESCRIPTION OF SYMBOLS 10 Vibration analyzer 12 Signal generation circuit 13 Feature extraction part 14 Discrimination part

Continuation of the front page (56) References JP-A-4-38427 (JP, A) JP-A-3-219535 (JP, A) JP-A-2-167435 (JP, A) JP-A-3-18787 (JP , A) JP-A-2-167436 (JP, A) (58) Fields investigated (Int. Cl. ⁶ , DB name) G01H 17/00 G01H 1/00 G01P 15/08 G01V 1/28-1/30

Claims (1)

(57) [Scope of request for utility model registration] A signal generation circuit for converting a vibration waveform into a plurality of binary waveforms by a plurality of threshold value processing units; The time until the first pulse of the binary waveform converted by the threshold value is generated and the off-time from the first pulse to the second pulse of the binary waveform converted by the high threshold value are defined as the feature amounts. A vibration analysis apparatus comprising: a feature amount extraction unit to be extracted; and a determination unit that determines a type of the vibration waveform by using the feature amount extracted by the feature amount extraction unit as an input.