JPS593097B2 - Kikiijiyoukenchisouchi - Google Patents

Description

[Detailed description of the invention] The present invention relates to abnormality detection of equipment.

An object of the present invention is to provide an equipment abnormality detection device that can also identify spike-like abnormalities and drift-like abnormalities.

A feature of the present invention is that the average value, variance, and maximum amplitude of the output signal of a measuring instrument are determined, and abnormal conditions of the equipment are determined based on a pattern based on changes in the average value, variance, and maximum amplitude.

The present invention will be explained in detail below using examples.

FIG. 1 is an example in which the present invention is applied to power supply monitoring.

In a plant, a failure of a power source used in a control system has a significant impact on the operation of the entire plant, and may even result in a plant shutdown.

The present invention aims to detect signs of power failure as early as possible and display the situation to plant operators (hereinafter simply referred to as operators).

In FIG. 1, 1 is a power supply. In this example, three quantities, the average value, variance, and maximum amplitude of the output 1, are employed as the statistical quantities to be monitored, and an abnormal situation is determined from the change pattern of these three quantities.

Reference numerals 2, 3, and 4 in FIG. 1 are an average value calculation circuit, a dispersion calculation circuit, and a maximum amplitude calculation circuit, respectively.

Now, these operations are always performed using the latest N sample values.

At this time, the average value, variance, and maximum quasi-IN are determined by the following equations.

However, yi: signal sample value N: number of samples The outputs of the circuits 2 to 4 are converted into 0 and 1 signals, respectively, by the monitoring pattern creation circuit 5.

This conversion into 0 and 1 signals is carried out by a comparison circuit which sets preset values for the outputs of circuits 2 to 4 and sets them to 1 when these values are exceeded, and 0 when they do not.

When the power supply output has a normal distribution, each setting value can be determined as shown in Table 1.

In Table 1, α is the risk rate for determining abnormality, mol
σg are the average value and variance at normal times, respectively.

A combination of changes in the average value, variance, and maximum amplitude converted to 0 and 1 using the above method is defined as an equipment abnormality monitoring pattern, and is expressed by the following formula.

(11,12,13) ・・・・・・・・・
(4) The created monitoring pattern is classified by the determination circuit 6, and the display device 7 displays the monitoring pattern and the message.

For classification of monitoring patterns, a pattern list prepared by assigning monitoring patterns with the highest probability of occurrence to assumed abnormal situations is used.

Table 2 shows a list of patterns used to classify abnormal situations.

To classify abnormal situations from monitoring patterns, a discriminant function expressed by the following equation is used.

W2 22×11+2×12+13 ・・・・・・・・・
...(5) Table 3 shows the correspondence between the value of W and the classification of abnormal situations.

When an abnormality is detected in this device, an alarm is issued by an alarm device 8 to call the operator's attention.

In addition, the recorder 9 records the average value, variance, and maximum amplitude of the power output.

Figure 2 shows the situation when a contact failure occurs in the power supply circuit.
It shows the changes in the output voltage and the changes in the average value, variance, and maximum amplitude of the voltage at that time.

If only the maximum amplitude changes significantly, as shown in this figure, and the other average values and variance change small, this device will
It is determined to be a poor contact or spike-like noise.

10 is a reset circuit which provides a reset output when the determination of abnormality is completed.

When the present invention is applied to monitoring the operating status of a control valve of an operating section in a constant value automatic control system, the position signal is inputted to the abnormality detection devices 2 to 10 in FIG.

FIG. 3 shows changes in the average value, variance, and maximum amplitude of the valve position signal when the control valve opens and locks for some reason.

When locked, the fluctuation of the valve position signal becomes small, and the variance and maximum amplitude become close to O.

On the other hand, when the average value is at a locked position and greatly changes beyond the set value ε1, it is determined that the control valve is drifting.

Although two examples have been given as embodiments of the present invention, the present invention uses not only the three types of statistics to be monitored, such as the average value, variance, and maximum amplitude, but also higher-order moments or extreme value occurrence rates. can.

Furthermore, the present invention is not limited to the embodiments, and can be applied to various types of electrical and mechanical equipment.

Table 4 shows examples to which the present invention can be applied.

According to the present invention, spike-like abnormalities and drift-like abnormalities can also be easily identified.

Examples of devices to which the present invention is applicable 0 rotating equipment, 0 valves, 0 sliding bodies Mechanical system 0 filter, 0 mechanical setting device, 0 bearing etc.

[Brief explanation of drawings]

FIG. 1 is a diagram showing an embodiment of the present invention, FIG. 2 is an explanatory diagram of the operation of the embodiment shown in FIG. 1, and FIG. 3 is an explanatory diagram of the operation of another embodiment of the invention.

Claims (1)

[Claims] 1 A measuring device attached to a device, a calculating means for inputting the output signal of the measuring device and calculating the average value, variance, and maximum amplitude of the output signal, and a calculating means for calculating the average value, variance, and maximum amplitude of the output signal, and calculating the average value, variance, and maximum amplitude obtained by the calculating means. A device abnormality detection device comprising means for creating a pattern based on a change in amplitude, and means for determining an abnormality state of the device based on the pattern.