JPH08220278A - Plant monitor device and monitor method - Google Patents

Abstract

PURPOSE: To provide a highly reliable plant monitor device and monitor method by automating a plant monitor and increasing a monitoring frequency for detecting the significant change of plant behavior and the error of an instrument drift, and shortening a detection time. CONSTITUTION: This device has a detector 1 to input measurement signals for the operation modes of various equipment in a plant, a data collection device 3 to record the output signals, a long-term trend analysis device 4 to analyze data sent from the device 3 by the preset time unit via the statistical analysis method, database 5 incorporating a monitor condition such as a threshold value for a long-term trend characteristic, an error judgement device 6 to make judgement about an error on the basis of a monitor condition such as a threshold value for the database 5, using the result of calculation with the device 4, and an alarm device 7 to generate an alarm on the basis of an error judgement signal.

Description

Detailed Description of the Invention

[0001]

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to a plant monitoring system for a nuclear power plant or the like, and more particularly to a plant monitoring system and a monitoring method for supporting operator's decision work.

[0002]

2. Description of the Related Art Conventionally, with respect to various components installed in, for example, a nuclear power plant, an abnormality in these components is detected, and the cause of the abnormality is analyzed and diagnosed to monitor the state of the plant. In addition, there is a plant monitoring device that supports the operator's decision work.

As a method of monitoring a plant, an average value, a maximum value, and a minimum value of various signals relating to the components in the plant collected by the data collecting device are sequentially calculated in real time by a long-term trend analysis device and recorded. Is being done. However, at this time, the abnormality detection by monitoring the long-term trend data was artificially performed after plotting the recorded data from the data collection device.

[0004]

In the method of detecting an abnormality by artificially monitoring long-term trend data, it is unavoidable that a time delay will occur in order to obtain the detection result, and the operator will be burdened, sometimes. There was a possibility of overlooking the abnormality, and there was a problem that reliability was low.

Further, the rate of change of long-term trend data captured in units of one hour or one day is successively calculated and monitored to detect anomalies, or the rate of change of various signals is monitored and preset. Although there is a demand for a plant monitoring device that automatically detects an abnormality and issues an alarm when the threshold value is exceeded, no device has realized this.

The object of the present invention is to automate the plant monitoring to increase the monitoring frequency to detect significant changes in plant behavior and abnormalities such as instrument drift, and shorten the detection time to improve reliability. It is to provide a high plant monitoring apparatus and monitoring method.

[0007]

In order to achieve the above object, a plant monitoring apparatus according to the invention of claim 1 is a detector for inputting measurement signals of operating states of various equipment in a plant, and an output of this detector. A data collection device that records signals, a long-term trend analysis device that analyzes the data sent from this data collection device in a predetermined time unit by a statistical analysis method, and a database with monitoring conditions such as threshold values of long-term trend characteristics. And an abnormality determination device that performs an abnormality determination based on a monitoring condition such as a threshold value of the database from the result calculated by the long-term trend analysis device, and an alarm device that issues an alarm by an abnormality determination signal. .

A plant monitoring method according to a second aspect of the present invention calculates a long-term tendency characteristic from a measurement signal of operating states of various equipments constituting a plant, and at the same time, calculates an average value and a standard value which are statistical amounts in a predetermined time unit. Deviation, maximum value and minimum value, compared with the threshold value set based on each statistic and rate of change value which is a long-term tendency characteristic during normal operation in advance for these change rates, and in the long-term tendency characteristic It is characterized in that it is determined that the plant is abnormal when a certain statistic exceeds the threshold value.

In the plant monitoring method according to the third aspect of the present invention, regarding the signal of the change rate which is the long-term trend characteristic, the change rate in a certain section does not exceed the set threshold value N.
The feature is that it is determined that the plant is abnormal when the number of times increases or decreases continuously.

In the plant monitoring method according to the fourth aspect of the present invention, the average value, the maximum value, the minimum value, and the standard deviation of the signals of the long-term trend characteristics change suddenly within the preset upper and lower threshold values. At this time, this sudden change is counted at the rate of change, and if a certain state continues after that, it is judged that the plant is abnormal.

A plant monitoring method according to a fifth aspect of the present invention is characterized in that preset threshold values and monitoring conditions such as abnormality determination are arbitrarily set based on long-term tendency characteristics during normal operation.

[0012]

According to the invention described in claim 1, the detector inputs a signal for measuring the operating state of the plant equipment, and the output signal is recorded in the data collecting device, and the recorded data is calculated by the long-term trend analyzing device. Then, the average value and standard deviation, the maximum value, the minimum value, and the rate of change, which are the statistics, are obtained.

This long-term tendency characteristic is compared with a monitoring condition such as a threshold value set in advance in the database based on each statistic which is the long-term tendency characteristic during normal operation in the abnormality judging device, When it is determined that the plant is abnormal beyond the normal operating range of the monitoring condition, the alarm device issues an alarm by the abnormality determination signal.

According to a second aspect of the present invention, the long-term tendency characteristics are calculated from the signals of the operating states of various equipments constituting the plant, and the average value and standard deviation, the maximum value and the minimum value, which are the statistics in a predetermined time unit, are calculated. By comparing these values with the preset threshold values, it is determined that the plant is abnormal when the various statistics, which are the long-term tendency characteristics, exceed the threshold values.

According to a third aspect of the present invention, in the case of a change rate signal having a long-term tendency characteristic, when the change rate in a certain section continuously increases or decreases N times without exceeding a set threshold value. Determined as plant abnormality.

According to a fourth aspect of the present invention, when the average value, the maximum value, the minimum value and the standard deviation of the signals of the long-term tendency characteristic change suddenly within the preset upper and lower threshold values, this sudden change occurs. Is counted as a rate of change, and if a certain state continues after that, it is determined that the plant is abnormal.

In a fifth aspect of the present invention, preset threshold values and monitoring conditions such as abnormality determination are arbitrarily set based on long-term tendency characteristics during normal operation.

[0018]

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS An embodiment of the present invention will be described with reference to the drawings. For the sake of convenience of description, a signal input from various devices in the plant and a detector for detecting the signal are shown as one unit. In the first embodiment, as shown in the block diagram of FIG. 1, a monitoring device is a detector 1 for inputting measurement signals of operating states of various equipment in a plant, and the detector 1.
A bandpass filter 2 that removes unnecessary noise through the signal from and a data collection device 3 that records the output signal.

Further, the data of the signals recorded in the data collecting device 3 are analyzed to calculate a long-term tendency characteristic x, and a long-term tendency analyzing device 4 and a statistical amount obtained by the calculation in the long-term tendency analyzing device 4. A database 5 that sets monitoring conditions such as threshold values and discrimination, which are set in advance for each of the average value, standard deviation, maximum value, minimum value, and rate of change, based on each statistic that is a long-term tendency characteristic during normal operation. An abnormality determination device 6 having the following determination functions (1) to (3) is provided.

(B) The statistic of the long-term tendency characteristic x calculated by the calculation of the long-term tendency analysis device 4 exceeds the threshold value set in advance, as compared with the threshold value. It is judged that the plant is abnormal. (B) When the rate of change calculated for the signal to be monitored at the time of monitoring the rate of change increases or decreases continuously N times without exceeding the threshold value, it is determined to be abnormal. (C) If the average value, standard deviation, maximum value, and minimum value of the signal to be monitored suddenly changes within the threshold range and then remains in a certain state, it is determined to be abnormal.

Further, when the abnormality determining device 6 determines that the abnormality is present, an alarm device 7 for issuing an alarm by the abnormality determining signal and a display by a CRT or the like for displaying the calculated long-term tendency characteristic x data and the determination result. It is composed of a device 8 and a plotting device 9 for plotting and outputting the device 8 as required.

Next, the operation of the above configuration will be described. Generally, the statistics calculated for long-term trend analysis are the following four items.

(A) Average value (xave ... sequentially updated).
(B) Standard deviation (xσ ... sequentially updated). (C) Maximum value (x
max ... maximum value in each storage cycle section). (D) Minimum value (x
min ... minimum value in each storage cycle section).

The present invention monitors the plant abnormality from the equipment abnormality detection by displaying the long-term tendency characteristic data consisting of these statistics in time series. As shown in FIG. 1, in various equipment of a plant, measurement signals of respective operating states are detected by a detector 1, and this output signal is recorded in a data collection device 3 via a bandpass filter 2. The data recorded in the data collection device 3 is
It is sent to the long-term trend analysis device 4 and the long-term trend characteristics x of each device
Is calculated.

The long-term tendency characteristic x calculated here is the average value, the standard deviation, the maximum value, the minimum value, and the change rate thereof, and the change rate Δx is as shown in the characteristic diagram of FIG.
It is calculated by the following formula (1). In addition, x is a long-term tendency characteristic,
Further, f represents a function, and calculates each average value (xave), standard deviation (xσ), maximum value (xmax), and minimum value (xmin).

Δx = f (x, dt) (1)

The long-term tendency characteristic x calculated by the long-term tendency analysis device 4 is input to the abnormality determination device 6 and stored in the database 5
Is compared with a threshold value set in advance based on the average value, standard deviation, maximum value, minimum value, and rate of change of each statistic that is the long-term tendency characteristic during normal operation.

When the abnormality determination device 6 determines that there is an abnormality, the alarm device 7 issues an alarm by this abnormality determination signal to notify the operator. The long-term tendency characteristic x and the judgment result at this time are displayed on the display device 8 and can be plotted and output by the plotting device 9. It should be noted that the determination of abnormality or normality at this time is made under the determination conditions as shown in the following second to fifth embodiments.

In the second embodiment, x _i (average value, standard deviation, maximum value, minimum value) of the long-term trend analysis result calculated by the long-term trend analysis device 4 is as shown in the characteristic diagram of FIG. It is judged to be abnormal when the upper limit A ₁ or the lower limit A ₂ of the respective preset threshold values is exceeded.
This is expressed by the following equation (2).

Abnormality = A ₁ ≦ x _i , or A ₂ ≧ x _i (2)

If plant operation or instrument surveillance is performed at the same time as this inspection, a sudden change in process signal or the like occurs, but the average value, the maximum value, the minimum value, and the standard deviation are affected. The rate of change you gave is
When it is determined that there is an abrupt change (according to the determination conditions shown in the third embodiment below), it is clearly not a change due to a plant abnormality, so conditions that are not determined to be abnormality can be set arbitrarily.

As shown in FIG. 3, in the long-term tendency characteristic of a signal, for example, the average value gradually increases for some reason, and the x _i data at time t _i is the upper limit A of the threshold value.
If it exceeds ₁ , the rate of change at this time has not changed suddenly, so it is determined that the plant is abnormal, not the change due to the above-described operation or instrument surveillance.

The third embodiment relates to a condition for determining whether or not the change rate Δx of the calculation result of the long-term trend analysis shown in FIG. 2 calculated by the long-term trend analysis device 4 is a sudden change.
As shown in the setting diagram of FIG. 4, the change rate Δx of the calculation result is
Within the shaded range consisting of the preset threshold values B _{0 to} B ₁ ,
Alternatively, when it is within the range determined to be abnormal within the shaded area consisting of the threshold values B _{2 to} B ₃ , the change in the rate of change Δx is not rapid, and the result is that the plant equipment to be inspected is abnormal. If so, it is determined that the plant has an abnormality.

Regarding this, the range determined to be abnormal in the case where the change rate increases in the following equation (3) and the range judged to be abnormal in the case where the change rate decreases in the equation (4) are shown. The threshold value B ₁
Since there is little change between B _{2 and} B _2, it is determined to be normal.

B ₁ <Δx <B ₀ (3) B ₂ >Δx> B ₃ (4)

Further, for example, a rapid change due to a plant operation or instrument surveillance, the change rate Δx
Becomes Δx> B ₀ or Δx <B ₃ , and it is determined that the change is abrupt and not abnormal. As shown in the transition characteristic diagram of FIG. 5, this is the long-term tendency characteristic x of a certain signal. For example, the average value is large due to the operation of the plant being performed from time t _{i- 6} to time t _i-5. It descended and returned again at time t _i-4 .

Here, the change rates Δx _i-5 and Δx _i-4 both suddenly change beyond the threshold values B ₃ and B ₀ , but since they return to the original values again, they are abnormal. Is not judged. However, after that, the change rates Δx _i-3 , Δx _i-2 , Δ
The x _i-1 values are all within the normal range of the threshold values B _{1 to} B ₂ . However, at t _i , the rate of change Δx _i rises, and since this rate of change exceeds the threshold value B _{1 in} FIG. 4, it is determined to be abnormal here because it is within the abnormal range.

In the fourth embodiment, the rate of change Δx of the long-term tendency characteristic x, which is the calculation result of the long-term tendency analysis in the long-term tendency analysis device 4, falls within the normal range according to the determination condition of the first embodiment or the second embodiment. In this state, if it increases or decreases continuously N times, it is determined to be abnormal.

For example, when N = 5 is set, as shown in the transition characteristic diagram of FIG. 6, the displacement of the average value, which is the long-term tendency characteristic x of a certain signal, gradually decreases for some reason. It is a very gradual change and the current time t _i
Then, the lower limit of the threshold value A ₂ has not been reached.

Therefore, in this state, it is not judged to be abnormal under the judgment conditions according to the first and second embodiments, but in the fourth embodiment, the change rate and the upper and lower limits of the threshold value are not exceeded. Since it has decreased 5 times in a row, it is determined to be abnormal. Accordingly, if the decreasing tendency continues as it is, it is possible to detect in advance an abnormal sign such as a drift of the instrument which may reach the lower limit A ₂ of the threshold value. The number N of times of determination can be arbitrarily set in the case of increase and the case of decrease.

In the fifth embodiment, the long-term trend characteristic x, which is the calculation result of the long-term trend analysis in the long-term trend analysis device 4, has a threshold upper limit A ₁ and a lower limit A as shown in the transition characteristic diagram of FIG. Within the normal range between the _two , a sudden change (Δ in FIG.
After occurrence of x _i > B ₀ , Δx _i <B ₃ ) an odd number of times (as a general rule, when the level does not return to the original level), a normal determination (according to the determination conditions of the first embodiment and the second embodiment) is given. Is N
If it continues, it is determined to be abnormal.

When N = 5 is set here, as shown in FIG. 7, the long-term tendency characteristic x of a certain signal, for example, the displacement of the average value suddenly changes once by the instrument surveillance, and then does not return to the original level. Since points 1 to 5 remain constant, even if they are regarded as normal in the discrimination conditions of the first and second examples, they are still unchanged five times. From the setting of N = 5, it is determined to be abnormal at time t _i . This makes it possible to easily detect an unrestored state of instrument surveillance.

The number of times N for the determination can be set arbitrarily. The thresholds A and B can be set arbitrarily for each statistic as shown in FIGS. 2 and 3, and a combination of these monitoring conditions can be selected for each signal. Therefore, it is possible to monitor the drift and the significant change from the fluctuations specific to each statistic of each signal.

Further, when it is determined that the plant equipment is abnormal, at the same time, an alarm is issued from the alarm device 7 to notify the operator and the contents of the long-term tendency characteristic x determined to be abnormal are displayed. It can be displayed on, for example, a CRT of the device 8. In addition, plotting can be output from the plotting device 9 as needed.

Since the abnormality of the plant is automatically detected from the change of the characteristic obtained by such long-term trend analysis, it is possible to reduce the burden on the plant operator by supporting the decision work and improve the reliability. .

[0046]

As described above, according to the present invention, it is possible to automate the monitoring of changes in long-term trend characteristics from the average value, standard deviation, and rate of change of statistics obtained by long-term trend analysis from the measurement signals of plant component equipment. To detect abnormalities in various devices,
This has the effects of increasing the frequency of monitoring and shortening the detection time of abnormalities, supporting the judgment work of operators who perform plant diagnosis by cause analysis, and improving the reliability of plant operation.

[Brief description of drawings]

FIG. 1 is a block configuration diagram of a plant monitoring apparatus according to a first embodiment of the present invention.

FIG. 2 is a characteristic diagram of long-term tendency characteristics (rate of change) of the first embodiment according to the present invention.

FIG. 3 is a characteristic diagram of a long-term tendency characteristic and a threshold value according to the second embodiment of the present invention.

FIG. 4 is a threshold value setting diagram of a change rate according to a third embodiment of the present invention.

FIG. 5 is a transition characteristic diagram of long-term tendency characteristics of the third embodiment according to the present invention.

FIG. 6 is a transition characteristic diagram of long-term tendency characteristics according to the fourth embodiment of the present invention.

FIG. 7 is a transition characteristic diagram of the long-term tendency characteristic of the fifth embodiment according to the present invention.

[Explanation of symbols]

1 ... Detector, 2 ... Band pass filter, 3 ... Data collection device, 4 ... Long-term trend analysis device, x ... Long-term trend characteristic, 5 ...
Database of monitoring conditions such as thresholds for long-term trend characteristics,
6 ... Abnormality determination device, 7 ... Alarm device, 8 ... Display device, 9 ...
Plot device, A ... Statistical amount threshold value, B ... Change rate threshold value, N ... Determination setting number of times.

Claims (5)

[Claims] 1. A detector for inputting measurement signals of operating conditions of various equipment in a plant, a data collecting device for recording output signals of the detector, and data sent from the data collecting device by a statistical analysis method. A long-term trend analysis device for analyzing in a predetermined time unit, a database with monitoring conditions such as threshold values of long-term trend characteristics, and monitoring conditions such as threshold values of the database from the results calculated by the long-term trend analysis device A plant monitoring device comprising: an abnormality determination device for determining an abnormality by means of an abnormality determination device; and an alarm device for issuing an alarm in response to an abnormality determination signal. 2. A long-term tendency characteristic is calculated from measurement signals of operating states of various equipments constituting a plant, and an average value and a standard deviation, a maximum value and a minimum value, which are statistics in a predetermined time unit, and these Compared with the threshold value set based on the change rate value and each statistic that is the long-term tendency characteristic in advance during normal operation, the various statistic that is the long-term tendency characteristic exceeds the threshold value. Therefore, the plant monitoring method is characterized by determining that the plant is abnormal. 3. A plant abnormality is determined when the rate of change signal, which is the long-term tendency characteristic, increases or decreases N times continuously without exceeding a set threshold value in a certain section. The plant monitoring method according to claim 2, wherein: 4. When the average value, the maximum value, the minimum value, and the standard deviation of the signals of the long-term tendency characteristic change suddenly within a preset upper and lower threshold value range, this rapid change is counted at a rate of change. The plant monitoring method according to claim 2, wherein the plant is determined to be abnormal when a certain state continues thereafter. 5. A plant monitoring method, wherein the preset threshold and monitoring conditions such as abnormality determination are arbitrarily set based on long-term tendency characteristics during normal operation.