JPS63100514A - Abnormality diagnosing device - Google Patents

Abstract

PURPOSE:To easily set a threshold of effective and accurate diagnosis even by an inexperienced operator, by providing the function which analyzes the characteristic of a signal as the diagnosis object. CONSTITUTION:The signal from a transmitter 1 or dual transmitters 2 is considered as a representative input signal. The input signal is diagnosed by a diagnosing part 3, and the warning or display processing is performed by a processing part 4. Meanwhile, the same input signal is inputted to a change rate analyzing part 5 and an analyzing part 7 of the deviation between dual signals to calculate an average value, a maximum value, and a standard deviation, and results are outputted to a CRT or a printer by an output part 6. In case of the standard deviation, there is 99.7% probability that it is within the normal range in the diagnosing part 3 as generally known, and the threshold is rationally set in accordance with the obtained characteristic of the input signal by these statistical means.

Description

DETAILED DESCRIPTION OF THE INVENTION [Field of Industrial Application] The present invention relates to a device having a function of diagnosing an abnormality in an input signal.

[Conventional technology]

A conventional abnormality diagnosis device has a function of diagnosing an abnormality in an input signal, and a function of recording and displaying the diagnosis results, as shown in FIG.

Among these, the following methods are generally used as representative methods for diagnosing input signals.

(1) Range over check It is detected that the signal is out of the normal use range (out of range) due to trouble on the signal output side, such as the transmitter, and determined to be abnormal.

(2) Change rate check (sudden change check) It is detected that the input signal moves at a change rate higher than the normal dynamic characteristics determined by the plant characteristics and detector characteristics, and is determined to be abnormal.

(3) Duplicated signal deviation check When a transmitter or the like is duplicated, the deviation between two input signals is monitored, and when the deviation becomes large, it is determined that there is an abnormality.

(4) For an input signal that has a correlation with an index value representing the state of the correlation check plant, it is detected that the signal deviates from the correlation and is determined to be abnormal.

In order to apply these methods to an actual plant, it is important to set them rationally; if the settings are too strict, there will be many malfunctions, and if the settings are too loose, abnormalities will not be detected.

[Problem that the invention seeks to solve]

Although the above-mentioned conventional technology has the fundamental function of discovering and processing an abnormality, the setting of a threshold value for determining an abnormality is determined based on conventional results and experience values. The signal characteristics vary from plant to plant and from signal to signal and are not the same, depending on the characteristics of the transmitter at position 2 of the detection pipe, the signal range, etc.

Therefore, there are problems in that it is difficult for an inexperienced person to make optimal settings, and it is difficult to make settings for signals to which conventional diagnosis has not been applied.

An object of the present invention is to enable threshold settings to be easily performed even without experience.

[Means for solving problems]

The above objective is achieved by understanding the characteristics of the input signal through analysis 2. That is, as shown in Figure 3, by providing a function to statistically analyze the characteristics of the input signal,
The characteristics of the signal can be easily understood, and the probability of malfunction can be quantitatively evaluated and set based on the maximum value of the rate of change, the standard deviation of the rate of change, etc.

First, regarding the rate of change check, in order to set the threshold, the input signal in the device is sampled at specified intervals, the rate of change is determined for each sampling, and the rate of change is determined from this and the past rate of change. Provide a function to calculate the maximum value and standard deviation.

In addition, regarding the deviation check of duplex B, in order to set the threshold, the signal input into the device is sampled at specified intervals, the deviation is obtained for each sampling, and from this and the past deviation, the deviation is calculated. Provide functions to calculate average value, maximum value, and standard deviation.

This makes it possible to statistically evaluate signal characteristics.

(Effect) The characteristics obtained through analysis are 1.111 times the statistical characteristics of the input signal. Taking the standard deviation as an example, it is well known that 3 falls within the normal range with a probability of 99.7% or more. By using these statistical methods, it is possible to make reasonable settings based on the characteristics of the input signal.

Therefore, the function of analyzing the characteristics of the input signal makes it possible to obtain data for making settings that are effective and do not cause malfunctions.

〔Example〕

An embodiment of the present invention will be described below with reference to FIG.

As a typical input signal, transmitter 1. Consider the signal from duplex oscillator 2. The input signal is diagnosed by the diagnosis section 3,
The processing unit 4 performs processing such as alarms and displays.

On the other hand, the same input signal is analyzed by the change rate analysis section 5. It is also input to the deviation analysis section 7 of the duplicated signal. Here, the average value and the maximum value 1vA standard deviation are calculated, and the output unit 6 outputs the results to a CRT or printer.

Here, flowcharts of the analysis sections 5 and 7 in FIG. 3 are shown in FIGS. 4 and 5.

FIG. 4 is a flowchart of the rate of change analysis section. In step 1, the rate of change R(+) is determined, in step 2, the maximum value RM^X of the rate of change is determined, in step 3, the standard deviation S is determined, and in step 4, the results are periodically output to a printer or CRT.

FIG. 5 is a flowchart of the duplex signal deviation analysis section, which is constructed based on the same concept as FIG. 4.

FIG. 6 shows the hardware configuration of this embodiment. Conventionally, an abnormality diagnosis device has an external cable 2 for transmitting an input signal.
7 is relayed by the external terminal section 23, passed through the input section 21 of the device, diagnosed and processed by the diagnosis/processing calculation section 20, and displayed on the lamp display section 29.
The processing calculation section 20 incorporates analysis calculations, and the analysis results are converted to C.
It is displayed on the RT 25 and printer 26.

According to this embodiment, since the analysis is performed using the diagnosis/processing calculation section 20, there is an effect that the increase in hardware is small.

A hardware configuration 28 of another embodiment of the present invention is shown in FIG.

According to this embodiment, since the analysis section is configured with independent hardware, only the analysis section can be made portable, and the analysis section can be used only until the diagnostic threshold is set. This has the effect of making it possible to analyze signals from multiple plants using one analysis unit.

Although the above configurations are all taken as an example of an abnormality diagnosis device as an independent device, the control device may have a diagnosis/processing function inside the control device, for example, like a boiler automatic control device in a thermal power plant.・In this case as well, Figure 6,
Similar to FIG. 7, an analysis function can be added.

〔Effect of the invention〕

According to the present invention, the characteristics of the input signal to be diagnosed are determined empirically,
Since it can be grasped quantitatively and statistically rather than intuitively, it has the advantage that effective and accurate threshold settings for diagnosis can be easily set even without experience.

[Brief explanation of the drawing]

Fig. 1 is a functional configuration diagram of an embodiment of the present invention, Fig. 2 is a conceptual explanatory diagram of a conventional abnormality diagnosis device, Fig. 3 is a conceptual explanatory diagram of the present invention, and Fig. 4 is an embodiment of the present invention. Of these, FIG. 5 is a flowchart of change rate analysis, FIG. 5 is a flowchart of duplicated signal deviation analysis, FIG. 6 is a hardware configuration diagram of one embodiment of the present invention, and FIG. 7 is a flowchart of another embodiment of the present invention. 6 which is a hardware configuration diagram 1... Transmitter, 2... Duplex transmitter, 3... Diagnosis section, 4... Processing section, 5... Rate of change analysis section, 6...
Output section, 7... Duplicated signal deviation analysis section, 8... Analysis section, 10... Abnormality diagnosis device, 11... Input signal, 2
0...Diagnosis/processing section, 21...Input section, 23...
External terminal section, 24... Analysis section, 25... CRT, 2
6...Printer, 27...External cable, 28...
・Connection cable, 29... Lamp display section. Confectionery + Mouth height 60 n Yu''''1 figure

Claims (1)

[Claims] 1. An abnormality diagnosis device having a function of diagnosing an abnormality in an input signal, characterized in that the device is provided with a function of analyzing characteristics of the signal to be diagnosed.