JP5461136B2 - Plant diagnostic method and diagnostic apparatus - Google Patents

Description

  The present invention relates to a plant diagnosis method and a diagnosis apparatus for diagnosing abnormal plant events based on plant operation data.

  2. Description of the Related Art Conventionally, various methods for remotely monitoring and diagnosing a plant based on various operation data of the plant have been proposed and put into practical use in plants such as power plants and mechanical equipment. This is because operation data is collected by sensors installed in each part of the plant, and based on this operation data, operators in the control room of the plant directly monitor various data of the plant, or from these short-term It is implemented by analyzing trend data showing long-term trends.

  FIG. 6 shows an example of a conventional plant diagnostic apparatus 100. The plant diagnosis apparatus 100 includes a controller / data collection apparatus 101 to which measurement values of various sensors 106 installed in each device of the plant are input. The controller / data collection device 101 collects the measurement values of the various sensors 106 and analyzes the operation data as necessary. The collected operation data or analysis data is stored in the data management device 103 and presented to the plant operator or maintenance manager 104 via the operation system 102. The plant operator or maintenance manager 104 uses the monitoring manual 105 to find out the prediction or occurrence of an abnormal plant event based on the operation data or analysis data, and controls various actuators 107 in the plant accordingly.

  Patent Document 1 (Japanese Patent Application Laid-Open No. 2003-122429) collects data related to chemical management of a plant in chemical management of a power plant, predicts chemical parameters that are expected to change due to a plant abnormal event, A chemical parameter change is detected based on a reference value for detecting fluctuations in the chemical parameter to create a predictive diagnosis matrix file, and an abnormal event in the plant is diagnosed according to a predetermined evaluation standard from the plant data fluctuation detection result file and the predictive diagnosis matrix file A plant abnormal event diagnosis apparatus configured as described above is disclosed.

JP 2003-122429 A

However, in the conventional plant diagnosis apparatus, the determination of abnormal events is largely dependent on the skill of the plant operator or maintenance manager, and the determination may differ depending on the skill level.
Judging signs of abnormality based on operation data as in Patent Document 1 is an effective method for abnormality diagnosis, but when operation data fluctuates greatly or the operation data deviates significantly from the normal operation range. Although it is easy to detect signs of abnormality, there is a possibility that the signs of abnormality may be overlooked if data values fluctuate gradually over a long period of time or if abnormal signs can be judged only with complex operating data. It was.

When the above problem becomes obvious, it is impossible to prevent the reduction in plant efficiency and the occurrence of troubles that could originally be prevented, resulting in an economic loss.
Therefore, in view of the problems of the prior art, an object of the present invention is to provide a plant diagnosis method and a diagnosis apparatus that can accurately grasp an abnormal event of a plant and prevent a trouble in advance.

In order to solve the above problems, a plant diagnosis method according to the present invention is a plant diagnosis method that diagnoses deterioration of the plant by a diagnosis device based on a plurality of operation data indicating the operation status of each part of the plant. In
In the diagnostic device, threshold values for the plurality of operation data are set, respectively.
The diagnostic device compares the plurality of operation data with the threshold values corresponding to these operation data, and determines that the operation data has an abnormality sign when the operation data exceeds the threshold value. An abnormal sign determination process;
The abnormality when the sign there been judged the operation data is present 2 or more, abnormal sites abnormality by the diagnosis device to specify site generated based on a combination of the operating data it is determined that there is the abnormality sign With a specific process,
Wherein the diagnostic device includes a monitoring type in which the two or more of the operational data of the plurality of operating data and parameters, and the threshold is more set respectively for a computed value calculated by the inspectional equation,
In the abnormality sign determination step, a comparison is made between the plurality of calculated values obtained by the monitoring formula and the threshold value corresponding to the calculated value, and the calculated value exceeds the threshold value. The value is judged to have an abnormal sign,
In the abnormal part specifying step, when there are two or more of the calculated values determined as having the abnormal sign, an abnormality has occurred by the diagnostic device based on the combination of the calculated values determined as having the abnormal sign And an abnormal part specifying step for specifying a part.

  In this way, by comparing a plurality of operation data with a threshold value, and when there are two or more operation data exceeding the threshold value, by specifying a site where an abnormality has occurred based on a combination of these operation data Thus, it is possible to quickly grasp the abnormality of each part that could not be determined only by a single driving data. Moreover, since abnormality can be determined quantitatively by comparing operation data with a threshold value and determining abnormality, it is possible to accurately detect abnormality regardless of the skill level of the engineer. .

In addition, the abnormality of each part includes deterioration and performance deterioration of the device, and further includes an abnormality sign indicating an abnormality. Each part includes each device or each facility included in the plant and each device included in each device or each facility.
A plant is a complex composed of a plurality of devices or facilities combined to exhibit a function of generating added value. Examples of the plant include a thermal power plant, a nuclear power plant, a petrochemical plant, an oil refining plant, and a waste treatment plant. The device or equipment includes equipment such as piping and an electric system and exhibits one or more functions.

  In the diagnostic device, a plurality of monitoring formulas using two or more of the plurality of operation data as parameters and a threshold value for a calculated value calculated by the monitoring formula are set. In the abnormality sign determination step, when the calculated value exceeds the threshold value by comparing the calculated value obtained by the monitoring formula with the threshold value corresponding to the calculated value, The calculated value is determined to have an abnormal sign, and when there are two or more calculated values determined to have the abnormal sign in the abnormal part specifying step, the combination of the calculated values determined to have the abnormal sign It is preferable to provide an abnormal site specifying step for specifying a site where an abnormality has occurred by the diagnostic device.

  In this way, when a calculated value obtained by a monitoring formula using two or more operation data as a parameter is compared with a threshold value, and there are two or more calculated values exceeding the threshold value, a combination of these calculated values. By specifying the site where the abnormality has occurred based on the above, it is possible to reliably detect the abnormality of the site that has been difficult to detect in the past. In particular, it is also possible to detect an abnormality of each device or each facility by detecting an abnormality by using the operation data of each device in combination.

Furthermore, it is preferable to output diagnostic information including a deterioration status of the specified part by an output means after specifying the part where the abnormality has occurred in the abnormal part specifying step.
Thus, by outputting diagnostic information, diagnostic information can be used as an index in maintenance, parts replacement, or the like.

In addition, the plant diagnostic apparatus according to the present invention is a plant diagnostic apparatus for diagnosing plant deterioration.
A plurality of measuring instruments for measuring operation data of each part of the plant;
Data collection means for collecting the operation data measured by the plurality of measuring instruments;
A diagnostic device to which a plurality of the operation data selected from the data collection means are input,
The diagnostic device comprises:
Threshold values for each of the plurality of operation data are set,
The plurality of operation data input from the data collection means and the threshold values corresponding to these operation data are respectively compared, and when the operation data exceeds the threshold value, the operation data is indicated as an abnormal sign. An abnormality sign determination means for determining presence;
When there are two or more of the operation data determined to have the abnormality sign, comprising an abnormal part specifying means for specifying a part where an abnormality has occurred based on the combination of the operation data determined to have the abnormality sign,
The diagnostic device comprises:
Wherein two or more of the operational data of the plurality of operating data inputted from the data acquisition means and parameters were monitoring expression, and the threshold for the calculated computed value by the inspectional equation is more set respectively And
The abnormality sign determination means compares the calculated value obtained by the monitoring formula with the threshold value corresponding to the calculated value, and the calculated value when the calculated value exceeds the threshold value. Is a means to determine that there is an abnormal sign,
The abnormal part specifying means is means for specifying a part where an abnormality has occurred based on a combination of the calculated values determined to have an abnormal sign when there are two or more of the calculated values determined to have the abnormal sign It is characterized by being.

In this way, by selecting only necessary operation data from the large amount of operation data collected by the data collection means and inputting it to the diagnosis device, the arithmetic processing by the diagnosis device can be performed efficiently and the processing can be shortened.
In addition, by comparing a plurality of operation data with a threshold value and when there are two or more operation data exceeding the threshold value, by specifying a portion where an abnormality has occurred based on a combination of these operation data, It becomes possible to grasp at an early stage the abnormality of each part that could not be determined by only one operation data. Moreover, since abnormality can be determined quantitatively by comparing operation data with a threshold value and determining abnormality, it is possible to accurately detect abnormality regardless of the skill level of the engineer. .

  In addition, the diagnostic device includes a monitoring formula using two or more of the operation data of the plurality of operation data input from the data collection unit as parameters, and a threshold for a calculated value calculated by the monitoring formula; Are set, and the abnormality sign determination means compares the plurality of calculated values obtained by the monitoring formula with the threshold corresponding to the calculated value, and the calculated value is the threshold value. A means for determining that there is an abnormal sign when the calculated value exceeds the value, and the abnormal part specifying means determines that there is an abnormal sign when there are two or more calculated values determined to have the abnormal sign Preferably, it is a means for identifying a site where an abnormality has occurred based on the combination of the calculated values.

  In this way, when a calculated value obtained by a monitoring formula using two or more operation data as a parameter is compared with a threshold value, and there are two or more calculated values exceeding the threshold value, a combination of these calculated values. By specifying the site where the abnormality has occurred based on the above, it is possible to reliably detect the abnormality of the site that has been difficult to detect in the past. In particular, it is also possible to detect an abnormality of each device or each facility by detecting an abnormality by using the operation data of each device in combination.

Furthermore, it is preferable to provide an output means for outputting diagnostic information including the deterioration status of the part characterized by the abnormal part specifying means.
Thus, by outputting diagnostic information, diagnostic information can be used as an index in maintenance, parts replacement, or the like.

  Furthermore, it is preferable that the operation data collected by the data collecting means is input to the diagnostic device via a communication line, whereby the deterioration status of the plant is detected by the diagnostic device installed at a remote place from the plant. It becomes possible to grasp. Further, in the present invention, only necessary operation data is selected from a large amount of operation data collected by the data collection means and input to the diagnosis device, so that only necessary data transmission / reception can be performed. The load of can be kept low.

  As described above, according to the present invention, when a plurality of operation data is compared with a threshold and there are two or more operation data exceeding the threshold, an abnormality occurs based on the combination of these operation data. By specifying the part that has been performed, it becomes possible to grasp at an early stage the abnormality of each part that could not be determined only by a single driving data. Moreover, since abnormality can be determined quantitatively by comparing operation data with a threshold value and determining abnormality, it is possible to accurately detect abnormality regardless of the skill level of the engineer. .

1 is an overall configuration diagram of a plant diagnostic apparatus according to an embodiment of the present invention. It is a figure explaining the threshold value with respect to driving | operation data. It is a table | surface for specifying an abnormal precursor. It is a table | surface for pinpointing an abnormal precursor different from FIG. It is a flowchart of the plant diagnostic method which concerns on embodiment of this invention. It is a whole block diagram of the conventional plant diagnostic apparatus.

  Hereinafter, exemplary embodiments of the present invention will be described in detail with reference to the drawings. However, the dimensions, materials, shapes, relative arrangements, and the like of the components described in this embodiment are not intended to limit the scope of the present invention unless otherwise specified, but are merely illustrative examples. Not too much.

FIG. 1 is an overall configuration diagram of a plant diagnostic apparatus according to an embodiment of the present invention.
A plant monitoring device 13 is installed at a plant site 10 where a plant to be diagnosed according to the present invention is installed. The plant monitoring device 13 includes a control controller / data collection unit 14 to which measurement values from various sensors 11 provided in each part of a plant (not shown) are input, and an input function and memory of the control controller / data collection unit 14. Operation system 15 having a management function, etc., data management means 16 for managing operation data collected by the controller / data collection means 15 or analysis data obtained by analyzing the operation data, and data collected by the controller / data collection means 15 A transmission means 17 for transmitting operation data is provided. Various actuators 12 are controlled based on the operation data collected by the controller / data collection means 14 or the analysis data obtained by analyzing the operation data.

A plant is a complex composed of a plurality of devices or facilities combined to exhibit a function of generating added value. Examples of the plant include a thermal power plant, a nuclear power plant, a petrochemical plant, an oil refining plant, and a waste treatment plant. The device or equipment includes equipment such as piping and an electric system and exhibits one or more functions.
The various sensors 11 are installed in a plant apparatus or facility and measure an operation state. For example, the opening degree of the damper, the flow rate or temperature of water flowing through the pipe, the pressure in the apparatus, the current of the motor, and the like are measured. The various sensors 11 and the control controller / data collection means 14 are connected by electrical wiring, and it is preferable that the measured values measured by the various sensors 11 are directly input to the control controller / data collection means 14.

The control controller / data collection means 14 includes a calculation unit composed of a CPU and a storage medium (not shown) in which a software program for realizing various calculation processes is recorded, and a program stored in the storage medium as necessary. Are read out by the calculation unit and this program is executed to realize various calculation processes.
The controller / data collection means 14 selects only the operation data required by the plant diagnostic device 21 described later from the operation data collected from the various sensors 11 and transmits it via the transmission means 17. .

  A plant diagnostic apparatus 21 is installed at a plant diagnostic base 20 provided outside the plant site 10. The plant diagnosis apparatus 21 mainly includes a calculation unit 22 that performs various calculation processes, a diagnosis database (diagnosis DB) 25 that stores diagnosis results calculated by the calculation unit 22, and diagnosis results calculated by the calculation unit 22. Output means 26 and receiving means 27 for receiving the operation data transmitted from the controller / data collecting means 14. The computing unit 22 includes a computing unit including a CPU, and a storage medium (not shown) in which a software program that realizes the functions of the abnormal sign determination unit 23 and the abnormal part specifying unit 24 shown below is recorded. Thus, each function is realized by the arithmetic unit reading the program stored in the storage medium and executing the program.

The receiving means 27 receives the operation data collected by the controller / data collecting means 14 via the communication line 30. That is, it is an operation data input means in the plant diagnosis device 21. The communication line 30 includes a public line such as the Internet or a dedicated line.
Here, the operation data is input to the plant diagnosis device 21 via the communication line 30, but the present invention is not limited to this, and may be input via a recording medium. Operation data may be input directly by an operator.

The calculation means 22 includes an abnormality sign determination means 23 and an abnormal part identification means 24.
Further, threshold values corresponding to a plurality of operation data are respectively set in the calculation means 22. For example, as shown in FIG. 2, a threshold value consisting of at least one of an upper limit and a lower limit is set for a reference value most suitable for driving. FIG. 2 shows a case where both the upper threshold and the lower threshold are set. Moreover, it is preferable that the threshold value is set corresponding to the driving load.

The abnormality sign determination means 23 compares the operation data received by the receiving device 27 with a threshold value corresponding to the operation data. If the operation data exceeds the threshold value, the operation data indicates that there is an abnormality sign. judge.
The abnormal part specifying unit 24 specifies a part where an abnormality has occurred based on a combination of operation data determined to have an abnormal sign when there are two or more operating data determined to have an abnormal sign by the abnormal sign determining unit 23 To do.

For example, the abnormal part specifying means 24 has a table as shown in FIG. In the table of FIG. 3, when an abnormality sign of the XX site is performed, it is performed by a combination of data 1, data 2, and data X. Similarly, when the abnormality sign of the xx part is performed, the data 2, the data X, and the data Y are used. In this way, a plurality of operation data necessary for performing an abnormality sign is set in advance in each part. Note that a horizontal line is drawn in the column of unnecessary data.
For each operation data, it is determined whether or not the threshold value is exceeded using the graph shown in FIG. 2. If the threshold value is exceeded, x is entered in the corresponding column. If the operation data does not exceed the threshold value and is within the allowable range, a circle is entered in the corresponding field.
In each part, when all the operation data necessary for performing an abnormality sign is filled with x, it is determined that this part has an abnormality sign.

  In this way, by comparing a plurality of operation data with a threshold value, and when there are two or more operation data exceeding the threshold value, by specifying a site where an abnormality has occurred based on a combination of these operation data Thus, it is possible to quickly grasp the abnormality of each part that could not be determined only by a single driving data. Moreover, since abnormality can be determined quantitatively by comparing operation data with a threshold value and determining abnormality, it is possible to accurately detect abnormality regardless of the skill level of the engineer. .

Further, as another method for specifying an abnormal site, an abnormal site is specified using a monitoring formula using two or more operation data as parameters and a threshold value for a calculated value calculated by the monitoring formula. Also good.
The monitoring formula is set as the following formula, for example.
Monitoring formula = [data 1] × ([data X] − [data Y]) / 2
A threshold is set for the calculated value of this monitoring formula. For example, assuming that the threshold value of the calculated value is 2.0, it is determined that the calculated value is 0 to 2.0 and is normal, and when the calculated value exceeds 2.0, it is determined that there is an abnormal sign.

In the table shown in FIG. 4, “X” is input when the calculated value of the monitoring formula 1 exceeds the threshold value, and “◯” is input when it is within the allowable range. As in FIG. 3, a plurality of monitoring formulas necessary for performing an abnormality sign are set in advance in each part. When all the monitoring formulas necessary for performing an abnormality sign are filled with x, it is determined that there is an abnormality sign for this part.
In this way, when a calculated value obtained by a monitoring formula using two or more operation data as a parameter is compared with a threshold value, and there are two or more calculated values exceeding the threshold value, a combination of these calculated values. By specifying the site where the abnormality has occurred based on the above, it is possible to reliably detect the abnormality of the site that has been difficult to detect in the past. In particular, it is also possible to detect an abnormality of each device or each facility by detecting an abnormality by using the operation data of each device in combination.

Next, referring to FIG. 5, a plant diagnosis method according to an embodiment of the present invention will be described.
First, in the plant site 10, the operation data measured by the various sensors 11 is collected by the controller / data collection means 14 (S1). As an example, a boiler main body, boiler auxiliary equipment such as an air heater, a ventilation fan, a suit blower, a raw material crushing mill, a gas turbine main body, a steam turbine main body, a water system device such as a feed water pump, a circulating water pump, and a condenser In addition, when diagnosing abnormalities and long-term deterioration of a thermal power plant having a dust collection device, a desulfurization device, a denitration device or the like, the operation data measured by the sensor 11 installed in each device is used. Data is collected by the controller / data collecting means 14.
The controller / data collection means 14 stores these operation data in the data management means 16 and presents them to the driver via the operation system 15. Further, the controller / data collection means 14 selects a plurality of operation data required for the abnormality diagnosis of the plant by the plant diagnosis device 21 and transmits it by the transmission means 17.

  In the plant diagnosis base 20, the reception means 27 of the plant diagnosis apparatus 21 receives a plurality of operation data transmitted by the transmission means 17 via the communication line 30, and inputs them to the calculation means 22 (S2). The abnormality sign determination means 23 of the calculation means 22 compares a plurality of operation data with threshold values corresponding to these operation data as shown in FIG. 2 (S3), and the operation data sets the threshold value. It is determined whether it exceeds (S4), and the operation data exceeding the threshold value is determined to have an abnormality sign.

When there is no operation data determined to have an abnormality sign or when the operation data is less than 2, diagnosis information without an abnormality sign is stored in the diagnosis DB 25 (S5), and the diagnosis information is output by the output means 26. Output (S6).
When there are two or more operation data determined as having an abnormal sign, as shown in FIG. 3 or FIG. 4, an abnormal part is specified based on a combination of data determined to have an abnormal sign (S7). Then, diagnostic information including the abnormal part and its situation is stored in the diagnostic DB 25 (S8), and the diagnostic information is output by the output means 26 (S9). For example, in the case of the above-mentioned thermal power plant, the operation data measured by the raw material crushing mill includes the damper opening, the primary air flow rate, the primary air temperature, the primary air pressure, the lower chamber pressure, the upper chamber pressure, the outer temperature, the main motor. Current, rotary separator speed, etc. If it is determined that there are abnormal signs in the four items of primary air pressure, outer temperature, main motor current, and rotary separator speed among these, the abnormal part is the raw material grinding mill The diagnosis is made at a specific site, and this diagnosis information is output. The operation data used for the diagnosis may be operation data obtained from two or more different devices or facilities (for example, a boiler body and a water system device) arranged in the plant. When a thermal power plant is targeted for diagnosis, performance such as power generation efficiency, boiler efficiency, turbine efficiency, or flue gas processing efficiency of the entire plant can be diagnosed by performing diagnosis using a plurality of operation data.

  As described above, according to this embodiment, when a plurality of operation data is compared with a threshold value and there are two or more operation data exceeding the threshold value, an abnormality has occurred based on a combination of these operation data. By specifying a part, it becomes possible to grasp at an early stage an abnormality of each part that could not be determined only by a single driving data. Moreover, since abnormality can be determined quantitatively by comparing operation data with a threshold value and determining abnormality, it is possible to accurately detect abnormality regardless of the skill level of the engineer. .

DESCRIPTION OF SYMBOLS 10 Plant site 11 Various sensors 12 Various actuators 13 Plant monitoring apparatus 14 Controller / data collection means 15 Data management means 16 Operation system 17 Receiving means 20 Plant diagnosis base 21 Plant diagnosis apparatus 22 Calculation means 23 Abnormal sign determination means 24 Abnormal part identification Means 25 Diagnostic Database (Diagnostic DB)
26 output means 27 receiving means

Claims (5)

In a plant diagnosis method for diagnosing deterioration of the plant by a diagnostic device based on a plurality of operation data indicating the operation status of each part of the plant,
In the diagnostic device, threshold values for the plurality of operation data are set, respectively.
The diagnostic device compares the plurality of operation data with the threshold values corresponding to these operation data, and determines that the operation data has an abnormality sign when the operation data exceeds the threshold value. An abnormal sign determination process;
When the operating data it is determined that there is the abnormality indication is present 2 or more, abnormal abnormality by the diagnosis device to specify site generated based on the combination of the operating data it is determined that there is the abnormality sign A site identification process,
Wherein the diagnostic device includes a monitoring type in which the two or more of the operational data of the plurality of operating data and parameters, and the threshold is more set respectively for a computed value calculated by the inspectional equation,
In the abnormality sign determination step, a comparison is made between the plurality of calculated values obtained by the monitoring formula and the threshold value corresponding to the calculated value, and the calculated value exceeds the threshold value. The value is judged to have an abnormal sign,
In the abnormal part specifying step, when there are two or more of the calculated values determined as having the abnormal sign, an abnormality has occurred by the diagnostic device based on the combination of the calculated values determined as having the abnormal sign A plant diagnostic method comprising: an abnormal site specifying step of specifying a site.   2. The plant diagnosis method according to claim 1, wherein after the part where an abnormality has occurred is specified in the abnormal part specifying step, diagnostic information including a deterioration state of the specified part is output by an output unit. In a plant diagnostic device for diagnosing plant degradation,
A plurality of measuring instruments for measuring operation data of each part of the plant;
Data collection means for collecting the operation data measured by the plurality of measuring instruments;
A diagnostic device to which a plurality of the operation data selected from the data collection means are input,
The diagnostic device comprises:
Threshold values for each of the plurality of operation data are set,
The plurality of operation data input from the data collection means and the threshold values corresponding to these operation data are respectively compared, and when the operation data exceeds the threshold value, the operation data is indicated as an abnormal sign. An abnormality sign determination means for determining presence;
When there are two or more of the operation data determined to have the abnormality sign, comprising an abnormal part specifying means for specifying a part where an abnormality has occurred based on the combination of the operation data determined to have the abnormality sign,
The diagnostic device comprises:
Wherein two or more of the operational data of the plurality of operating data inputted from the data acquisition means and parameters were monitoring expression, and the threshold for the calculated computed value by the inspectional equation is more set respectively And
The abnormality sign determination means compares the calculated value obtained by the monitoring formula with the threshold value corresponding to the calculated value, and the calculated value when the calculated value exceeds the threshold value. Is a means to determine that there is an abnormal sign,
The abnormal part specifying means is means for specifying a part where an abnormality has occurred based on a combination of the calculated values determined to have an abnormal sign when there are two or more of the calculated values determined to have the abnormal sign A diagnostic device for a plant, characterized in that   The plant diagnosis apparatus according to claim 3, further comprising an output unit configured to output diagnosis information including a deterioration state of the part characterized by the abnormal part specifying unit.   The plant diagnosis apparatus according to claim 3, wherein the operation data collected by the data collection means is input to the diagnosis apparatus via a communication line.

Images