JP5544418B2 - Plant diagnostic device, diagnostic method, and diagnostic program - Google Patents

Description

  The present invention relates to a plant diagnostic apparatus, diagnostic method, and diagnostic program.

  When an abnormal transient or accident occurs in the plant, the plant diagnosis device detects the occurrence of the abnormality or accident based on the measurement data from the plant.

  Patent Document 1 discloses a diagnostic apparatus that uses adaptive resonance theory (ART). A diagnostic apparatus using ART has a function of classifying multidimensional data into categories according to the similarity.

  In the technique of Patent Document 1, first, normal measurement data is classified into a plurality of categories (normal categories) using ART. Next, current measurement data is classified into categories by ART. When this measurement data cannot be classified into normal categories, a new category (new category) is generated. The occurrence of a new category means that the state of the plant has changed. Therefore, the occurrence of an abnormality is determined based on the occurrence of a new category, and an abnormality is diagnosed when the occurrence rate of the new category exceeds a threshold value.

JP 2005-165375 A

  In Patent Document 1, an abnormality is diagnosed on the premise that “the new category is generated when the current measurement data cannot be classified into the normal category, that is, when the state of the plant changes”.

  However, the state of the plant changes not only when an abnormality occurs but also when it deteriorates over time and when operating conditions change. At this time, a new category is generated. For this reason, it cannot be a new category or an immediate plant abnormality.

  Here, the operating conditions are not directly related to the equipment characteristics such as the environmental conditions (atmospheric temperature, humidity, etc.) of the place where the plant is installed and the amount determined by the operator's operation (power generation amount generated in the plant). It is an external environmental factor. Regardless of whether or not an abnormality has occurred, the values of various measurement data also change when the operating conditions change.

  Although the deterioration over time and the change in operating conditions are not abnormal, the method of Patent Document 1 diagnoses this as abnormal. In other words, the normal state is diagnosed as abnormal, and as a result, erroneous reporting occurs.

  An object of the present invention is to reduce the occurrence rate of false alarms by distinguishing aging deterioration, operating condition changes, and abnormalities that are factors of state changes.

  The plant diagnosis apparatus of the present invention detects an abnormality of a diagnosis target from a measurement signal of a diagnosis target plant, and notifies the abnormality. The measurement signal of the diagnosis target plant is classified into a category, stored, and classified. State change detection unit that determines that the state has changed when it does not belong to the category, output from the abnormality determination unit that determines that an abnormality has occurred in the diagnosis target plant using the signal from the state change detection unit as part of the input Alarm generating means for informing the outside, and a false alarm prevention unit for detecting an event other than an abnormality in the diagnosis target plant and blocking an output of the abnormality determination unit.

  In addition, the false alarm prevention unit obtains a weighting factor using the measurement signal of the diagnosis target plant, and when the amount of change in the weighting factor exceeds the maximum value, an aged deterioration detection unit that determines that the range of aging deterioration has been exceeded. In addition, the aging deterioration detection unit may prevent the output of the abnormality determination unit when it does not deviate from the aging deterioration range.

  The false alarm prevention unit includes an operation condition change detection unit that classifies and stores the operation condition data of the diagnosis target plant into categories, and determines that the operation conditions have not changed when a new category does not occur. The condition change detection unit may block the output of the abnormality determination unit when the operating condition is changing.

  In addition, the operating condition data that is an input of the operating condition change detection unit is preferably composed of external environmental factors that are not directly related to the characteristics of the devices that constitute the diagnosis target plant.

  The plant diagnosis apparatus of the present invention detects an abnormality of a diagnosis target from a measurement signal of a diagnosis target plant, and notifies the abnormality. The measurement signal of the diagnosis target plant is classified into a category, stored, and classified. State change detection unit that determines that the state has changed when it does not belong to the category, the weighting factor is obtained using the measurement signal of the diagnosis target plant, and the range of deterioration over time when the amount of change in the weighting factor exceeds the maximum value Aged deterioration detection unit that determines that the vehicle has deviated from the operation condition, and the operation condition data of the diagnosis target plant is classified and stored in a category, and the operation condition change detection unit that determines that the operation condition has not changed when a new category does not occur The state change detection unit determines that the state has changed, the aging deterioration detection unit determines that it has deviated from the range of aging deterioration, and the operating condition change detection unit changes the operating condition. Abnormality determining unit for determining that an abnormality has occurred in diagnostic object when it is determined that no, and a warning generating means allowed to informing the output of the abnormality determination unit to the outside.

  The plant diagnosis apparatus of the present invention includes a measurement signal database that stores measurement signals of a diagnosis target plant, a processing data extraction unit that extracts a diagnosis signal used for diagnosing the state of the diagnosis target plant from the measurement signal database, Reference signal database for storing diagnostic signals, classification means for classifying data stored in the reference signal database into categories, classification result database for storing categories as normal categories, and latest diagnosis extracted by processing data extraction means Diagnosis means for diagnosing whether the status of the plant to be diagnosed belongs to normal, abnormal, operating condition change, or aged deterioration using information on normal categories stored in the signal and classification result database, and diagnostic results of the diagnostic means Diagnostic result database to save and save to diagnostic result database The information is output to the image display device, and the diagnosis means includes an abnormality determination unit, a state change detection unit, an aging deterioration detection unit, and an operating condition change detection unit, and the state change detection unit Is equipped with a function to determine that the state has changed when the latest diagnostic signal extracted by the processing data extraction unit does not belong to the normal category stored in the classification result database. The function of deriving the relationship between the time variation width and the weight coefficient variation amount and determining that it has deviated from the range of aging deterioration when the maximum value of the weight coefficient variation amount in the normal state is exceeded during diagnosis. The change detection unit determines that the operating condition has not changed when a new category does not occur when operating condition data composed of external environmental factors not directly related to the device characteristics is classified into categories. In the abnormality determination unit, the state change detection unit determines that the state has changed, the aging deterioration detection unit determines that it has deviated from the range of aging deterioration, and the operating condition change detection unit changes the operating condition. When it is not determined that it is not, it is determined as abnormal.

  The plant diagnosis method of the present invention detects abnormality of a diagnosis target plant from the measurement signal of the diagnosis target plant, and notifies the abnormality, classifies the measurement signal of the diagnosis target plant into a category, stores it, and classifies it. It is determined that the condition has changed when the device does not belong to the selected category, and the weighting factor is obtained using the measurement signal to be diagnosed. Determine and store the operating condition data to be diagnosed into categories and store them. If no new category occurs, determine that the operating conditions have not changed and determine that an abnormality has occurred in the diagnosis target plant. Let me know.

  The plant diagnosis program of the present invention is a plant diagnosis program that detects an abnormality of a diagnosis target plant from a measurement signal of the diagnosis target plant and notifies the abnormality, and stores the measurement signal of the diagnosis target in a category. , A state change detection step for determining that the state has changed when it does not belong to the classified category, an abnormality determination step for determining that an abnormality has occurred in the diagnosis target as a part of the input from the signal from the state change detection step, an abnormality determination step The alarm generation step for informing the outside of the output to the outside, and the false alarm prevention step for detecting an event other than the abnormality to be diagnosed and blocking the output of the abnormality determination unit.

  The false alarm prevention step is a aging deterioration detection step in which a weighting factor is obtained using the measurement signal of the diagnosis target plant, and when the change amount of the weighting factor exceeds the maximum value, it is determined that the range of aging deterioration has been exceeded. In addition, the aged deterioration detection step preferably prevents the output of the abnormality determination step when it does not deviate from the aged deterioration range.

  Further, the false alarm prevention step includes an operation condition change detection step for storing the operation condition data of the plant to be diagnosed classified into categories and determining that the operation condition has not changed when a new category does not occur, In the condition change detection step, it is preferable to prevent the output of the abnormality determination step when the operating condition is changing.

  In addition, the driving condition data that is the input of the driving condition change detection step is preferably composed of external environmental factors that are not directly related to the characteristics of the devices that constitute the diagnosis target.

  The plant diagnosis program of the present invention detects an abnormality of the diagnosis target plant from the measurement signal of the diagnosis target plant, and notifies the abnormality, classifies the measurement signal of the diagnosis target plant into a category, stores it, and classifies it. The state change detection step that determines that the state has changed when it does not belong to the selected category, the weighting factor is obtained using the measurement signal of the diagnosis target plant, and if the amount of change in the weighting factor exceeds the maximum value, Aging detection step for determining that the range has been exceeded, operating condition data of the plant to be diagnosed is classified and stored in a category, and operating condition change detection for determining that the operating condition has not changed when a new category does not occur It is determined that the state has changed in the step and the state change detection step, and it is determined that it has deviated from the range of the aging deterioration in the aging deterioration detection step. And abnormality determining step of determining the operating condition in the operating conditions change detecting step abnormality occurs in the diagnostic object when it is determined that no change, and a warning generating step allowed to informing the output of the abnormality determining step to the outside.

  By distinguishing aging, operating condition changes, and abnormalities, the rate of false alarms can be reduced.

  Further, by providing the operator with information on whether the cause of the change in the data trend belongs to aging deterioration, change in operating conditions, or abnormality, it is useful for planning a maintenance plan for the plant.

It is a block diagram which shows the diagnostic apparatus of this invention. It is a figure which shows the flowchart in normal state learning mode. It is a figure which shows the flowchart in diagnostic mode. It is a figure which shows the diagnosis which operates 2 modes for every sampling period. It is a figure which shows the diagnosis which performs normal state learning mode for a predetermined period, and diagnostic mode for every period. It is a figure which shows the example performed at the timing which the operator set. It is a block diagram which shows a data pre-processing apparatus and an ART module. It is a block diagram which shows F0 layer of an ART module. It is a block diagram which shows F1 layer of an ART module. It is a graph which shows the result of carrying out the category classification | category by the Example of FIG. It is the graph which showed an example of the time-dependent change of the measurement signal at the time of abnormality occurrence. It is a screen for setting data items used for diagnosis. It is a block diagram explaining the classification | category means 400 of FIG. It is a figure of the screen which shows the aspect of the data preserve | saved at a measurement signal database. It is a figure of the screen which shows the aspect of the data preserve | saved at a reference signal database. It is a figure of the screen which shows the aspect of the data preserve | saved at a classification result database. It is a figure which shows the structure of the abnormality determination part 510. FIG. It is the figure which displayed the aspect of the data of a diagnostic result database on the screen. It is a screen display example that displays the diagnosis results in the diagnosis result database in time series It is a figure explaining operation | movement of an aged deterioration detection part. It is a figure explaining the criteria of aged deterioration range confirmation. It is a figure explaining operation | movement of the 1st Example of an operating condition change detection part. It is a figure explaining operation | movement of the 2nd Example of an operating condition change detection part. It is a figure explaining operation | movement of the 3rd Example of an operating condition change detection part. It is a block diagram which shows a gas turbine power plant. It is a figure explaining the example of an operation result when the diagnostic apparatus of this invention is applied to a gas turbine power plant. It is the figure which summarized the operating condition of each part of a diagnostic means.

  Hereinafter, the present invention will be described with reference to the drawings.

  FIG. 1 is a block diagram showing a diagnostic apparatus of the present invention. In this figure, the state of the plant 100 is diagnosed by the diagnostic device 200.

  The diagnosis device 200 includes a processing data extraction unit 300, a classification unit 400, a diagnosis unit 500, and an alarm generation unit 600 as arithmetic units. The diagnostic apparatus 200 includes a measurement signal database 230, a reference signal database 240, a classification result database 250, and a diagnosis result database 260 as databases. In this figure, the database is abbreviated as DB. The database here is a component of the diagnostic apparatus 200, but information recorded in each database is digitized and is usually called an electronic file (electronic data).

  The diagnostic apparatus 200 includes an external input interface 210 and an external output interface 220 as interfaces with the outside. Then, the measurement signal 1 including values obtained by measuring various state quantities of the plant 100 via the external input interface 210, and the operation of the external input device 910 including the keyboard 920 and the mouse 930 provided in the operation management room 900. The generated external input signal 2 is input to the diagnostic apparatus 200. Further, the image display information 14 is output from the diagnostic device 200 to the image display device 940 (image display unit) in the operation management room 900 via the external output interface 220.

  In this embodiment, the processing data extraction means 300, classification means 400, diagnosis means 500, alarm generation means 600, measurement signal database 230, reference signal database 240, classification result database 250, and diagnosis result database 260 are all diagnosed. Although inside the apparatus 200, some of these may be arranged outside the diagnostic apparatus 200 to communicate only data. In the present embodiment, there is one plant to be diagnosed, but a plurality of plants can be diagnosed by one diagnostic device 200.

  Hereinafter, the operation of the diagnostic apparatus 200 will be described. First, the measurement signal 3 input via the external input interface 210 is stored in the measurement signal database 230.

  In the processing data extraction unit 300, the diagnostic signal 6 used for diagnosis is extracted from the measurement signal 5 stored in the measurement signal database 230 and stored in the reference signal database 240. The reference signal database 240 stores measurement signals for a period determined by the operator as normal. Further, the processing data extraction means 300 extracts data items based on operator settings. Details of this will be described later with reference to FIG.

  The classifying unit 400 classifies the reference signal 7 into categories. The classification result 8 is stored in the classification result database 250. The processing content of the classification unit 400 will be described later with reference to FIG.

  The diagnosis unit 500 processes the latest diagnosis signal 6 and the classification result 9 stored in the classification result database 250 to diagnose the state of the plant. The diagnosis unit 500 includes an abnormality determination unit 510, a state change detection unit 520, an aged deterioration detection unit 530, and an operating condition change detection unit 540.

  The state change detection unit 520 determines whether the state of the plant 100 has changed. The state change determination result 15 output from the state change detection unit 520 is “1” when it is determined that the state has changed, and is “0” when the state has not changed.

  In the state change detection unit 520, the latest diagnostic signal 6 extracted by the processing data extraction unit 300 is compared with the classification result 9 stored in the classification result database 250, and when it belongs to the category included in the classification result 9 The diagnostic signal 6 is classified into the category. On the other hand, the latest diagnostic signal 6 extracted by the processing data extracting means 300 is compared with the classification result 9, and when it does not belong to the category included in the classification result 9, it is expressed as a new category (hereinafter referred to as a new category). ). Detailed processing contents of the state change detection unit 520 will be described later with reference to FIG.

  The aging deterioration detection unit 530 determines whether or not the state change of the plant is within the aging deterioration range. The aged deterioration determination result 16 output from the aged deterioration detection unit 530 is “1” when the change in the state of the plant is out of the aged deterioration range, and is “0” when it is within the aged deterioration range. . Detailed processing contents of the aging deterioration detection unit 530 will be described later with reference to FIG.

  The operating condition change detection unit 540 determines whether or not the operating condition of the plant has changed. The operating condition change determination result 17 output from the operating condition change detection unit 540 is “1” when the operating condition is not changed, and is “0” when the operating condition is changed. Detailed processing contents of the operating condition change detection unit 540 will be described later with reference to FIG.

  The abnormality determination unit 510 determines whether an abnormality has occurred in the plant using the state change determination result 15, the aging deterioration determination result 16, and the operating condition change determination result 17. Detailed processing contents of the abnormality determination unit 510 will be described later with reference to FIG.

  The diagnosis result 10 output from the diagnosis unit 500 includes the determination result in the abnormality diagnosis unit 510, the state change determination result 15, the aging deterioration determination result 16, and the operating condition change determination result 17. The diagnosis result 10 is stored in the diagnosis result database 260.

  The alarm generation means 600 determines whether or not to generate an alarm using the diagnosis result 11 stored in the diagnosis result database 260 and the measurement signal 4 at the latest time stored in the measurement signal database 230.

  The alarm generation means 600 has the following criteria (conditions 1 and 2) for generating the following two types of alarms, and determines whether to generate an alarm by arbitrarily combining them. Here, an arbitrary combination of alarms is generated, for example, an alarm is generated when both of the following condition 1 and the following condition 2 are satisfied, and when either of the following condition 1 or the following condition 2 is satisfied, an alarm is generated. For example.

  Condition 1: The measurement signal 4 at the latest time deviates from a predetermined range (threshold value).

  Condition 2: The ratio determined as abnormal by the abnormality determination unit 510 within a predetermined period exceeds a certain value (threshold value).

  Note that the threshold values in conditions 1 and 2 are values set by the operator.

  When the alarm generation unit 600 determines to generate an alarm, the alarm generation unit 600 transmits the alarm signal 13 to the external output interface 220. The alarm signal 13 is converted into image display information 14 by the external output interface 220 and displayed on the image display device 940.

  In this embodiment, an alarm is communicated to the operator using the image display device 940. However, the present invention is not limited to this. First, an alarm is generated by generating an alarm sound. You may make it notice. Further, after the alarm sound is generated or simultaneously with the generation of the alarm sound, the image display information 940 may be set to be displayed on the image display device 940.

  In addition, the diagnosis result 12 (the determination result in the abnormality diagnosis unit 510, the state change determination result 15, the aging deterioration determination result 16, the operating condition change determination result 17) stored in the diagnosis result database 260 is sent to the external output interface 220. It can also be displayed on the image display device 940.

  The diagnostic device information 50 stored in the measurement signal database 230, the reference signal database 240, the classification result database 250, and the diagnostic result database 260 can be displayed on the image display device 940. Further, these pieces of information can be corrected using the external input signal 2 by operating the external input device 910 as necessary.

  A feature of the present invention is that the diagnosis unit 500 includes an abnormality determination unit 510, an aging deterioration detection unit 530, and an operating condition change detection unit 540, and the aging deterioration detection and the operation condition change detection are determined to be abnormal. Not to be.

  That is, the state change detection unit 520 positively detects an abnormality of the plant, but the content of this detection includes deterioration due to aging that should not be abnormal in nature and changes in operating conditions. False alarms are prevented by detecting aging deterioration and changes in operating conditions and comprehensively determining them by the abnormality determination unit 510. In this sense, it can be said that the aging deterioration detection unit 530 and the operating condition change detection unit 540 constitute a false alarm prevention unit for the state change detection unit 520.

  In the state change detection unit 520, the new category is generated when the current measurement data cannot be classified into the normal category, that is, when the state of the plant has changed.

  However, since the state of the plant changes not only when an abnormality occurs, but also when aging and operating conditions change, a new category is generated when the state change detection unit 520 is operated. Here, the operating conditions are the environmental conditions (atmospheric temperature, humidity, etc.) of the place where the plant is installed, the power generation amount output from the plant, and the like. Regardless of whether an abnormality has occurred or not, the values of various measurement data also change when the operating conditions change.

  Aging and changes in operating conditions are not abnormal, but this is diagnosed as abnormal in the method of judging abnormality only by the occurrence of a new category. This causes a false alarm.

  In the present invention, the diagnosis unit 500 includes the abnormality determination unit 510, the aging deterioration detection unit 530, and the operation condition change detection unit 540, thereby distinguishing the aging deterioration, the operation condition change, and the abnormality that are the cause of the state change, The incidence of false alarms can be reduced.

  In addition, in order to distinguish aged deterioration, operating condition change, and abnormality using a diagnosis device that does not include the abnormality determining unit 510, the aging deterioration detecting unit 530, and the operating condition change detecting unit 540 of the present invention, a long-term Operation data is required. That is, when operating the normal state learning mode, all operating conditions and data capable of evaluating aging degradation are required. Therefore, the period from the start of plant operation to the start of diagnosis becomes longer. By using the diagnostic device of the present invention, it is possible to distinguish abnormalities, aging deterioration, and changes in operating conditions even when there are few operations that can be used for diagnosis, so the period required for introducing the diagnostic device can be shortened.

  FIG. 2 is a flowchart showing the basic operation of the diagnostic apparatus 200 of FIG. 2A shows the normal state learning mode, and FIG. 2B shows the diagnostic mode.

  In the following description, the components described in FIG. 1 are also used.

  The diagnostic apparatus 200 has two basic operations: a normal state learning mode for classifying normal data into categories based on information stored in the reference signal database 240, and a diagnostic mode for diagnosing the state of the plant 100.

  In FIG. 2A, the normal state learning mode is executed by sequentially performing steps 1000 and 1010.

  First, in step S1000, the processing data extraction unit 300 is operated to extract the diagnostic signal 6 from the measurement signal 5 in the measurement signal database 230. The diagnostic signal 6 is stored in the reference signal database 240. The data stored in the reference signal database 240 is data for a period when the operator (operator) determines that the operation state of the plant 100 is normal. Data items stored in the reference signal database 240 will be described later with reference to FIG.

  Next, in step S 1010, the classification unit 400 is operated to classify the reference signal 7 stored in the reference signal database 240 and store the classification result 8 in the classification result database 250.

  As described above, in the normal state learning mode of FIG. 2A, the processing data extraction means 300, the measurement signal database 230, the reference signal database 240, the classification means 400, the classification of the left column in the diagnostic apparatus 200 of FIG. Processing using the result database 250 is executed.

  In the diagnostic mode shown in FIG. 2 (b), steps S1100, S1110, and S1120 are executed in order.

  First, in step S1100, the measurement signal 1 from the plant 100 is taken into the diagnostic device 200 via the external input interface 210, and the measurement signal 3 is stored in the measurement signal database 230.

  Next, the processing data extraction unit 300 is operated, the measurement signal 5 is extracted from the measurement signal database 230, and the diagnostic signal 6 with the latest time is transmitted to the diagnostic unit 500.

  In step S1110, the diagnostic unit 500 is operated. In the diagnostic unit 500, the calculation unit is operated in the order of the state change detection unit 520, the aging deterioration detection unit 530, the operating condition change detection unit 540, and the abnormality determination unit 510. The diagnostic result 10 output from the diagnostic means 500 is transmitted to the diagnostic result database 260 and stored. The diagnostic result 12 output from the diagnostic result database 260 is converted into the image display information 14 by the external output interface 220 and output to the image display device 940.

  In step S1120, alarm generation means 600 is operated to determine whether an alarm can be generated. When an alarm is generated, the alarm signal 13 output from the alarm generation unit 600 is converted into image display information 14 by the external output interface 220 and output to the image display device 940. Thereby, an alarm is notified to the operator (operator) of the plant 100.

  Thus, in the diagnostic mode of FIG. 2B, the diagnostic means 500 (state change detection unit 520, aging deterioration detection unit 530, operating condition change detection unit 540, right column in the diagnostic apparatus 200 of FIG. The abnormality determination unit 510), the external output interface 220, the image display device 940, and the processing using the alarm generation means 600 are executed. Needless to say, the external input interface 210, the measurement signal database 230, and the processing data extraction means 300 are used as the premise.

  FIG. 3 is a diagram for explaining the timing for executing the normal state learning mode flowchart (FIG. 2A) and the diagnostic mode flowchart (FIG. 2B) of the diagnostic apparatus 200.

  The diagnostic apparatus 200 acquires the measurement signal 1 from the plant 100 for each sampling period.

  In FIG. 3A, diagnosis is performed by operating both the normal state learning mode and the diagnostic mode for each sampling period.

  Further, in FIG. 3B, the normal state learning mode is operated every predetermined set period, and only the diagnostic mode is operated every sampling period for diagnosis.

  Further, in FIG. 3C, the operator performs an operation of setting a learning period and a diagnosis period, and the normal state learning mode and the diagnosis mode are operated at this timing.

  In either method, a diagnosis mode is executed every sampling period, and the state of the plant can be diagnosed online.

  In the following, the processing contents in the normal state learning mode and the processing contents in the diagnostic mode will be described in order in this order, but as a premise, processing that classifies the input process amount in either mode is executed. doing. The classification unit 400 corresponds to the normal state learning mode, and the state change detection unit 520 corresponds to the diagnosis mode. Therefore, the concept of the category classification method as common recognition will be described with reference to the block diagram of FIG. 4 before explaining the operation in each mode.

  In the following, a case where adaptive resonance theory (ART) is applied to the classification unit 400 and the state change detection unit 520 will be described, but other clustering methods such as vector quantization may be used.

  As shown in FIG. 4A, the classification unit 400 and the state change detection unit 520 are configured by a data preprocessing device 710 and an ART module 720. Further, the ART module 720 includes an F0 layer 721, an F1 layer 722, an F2 layer 723, a memory 724, and a selection subsystem 725, which are coupled to each other. The F0 layer 721 and the F1 layer 722 are configured as shown in FIGS. 4B and 4C, for example.

  In FIG. 4A, first, the data preprocessing device 710 converts operation data into input data of the ART module 720. Specifically, the following formulas (1) and (2) are executed. Hereinafter, the procedure (process) will be described.

  First, the maximum value and the minimum value are calculated for each measurement item. Data is normalized using the calculated maximum and minimum values. Here, the normalization method will be described by taking the process amount xi of the plant as an example.

  The number of data of xi is N and the nth measurement value is xi (n). Further, when the maximum value and the minimum value in N pieces of data are Max_i and Min_i, respectively, normalized data Nxi (n) is expressed by the following equation (1).

  Here, it is a constant of α (0 ≦ α <0.5), and the data is normalized to the range of [α, 1−α] by the above equation (1).

Next, the complement of the normalized data is calculated and added to the input data. The complement CNxi (n) of the normalized data Nxi (n) is calculated by the following equation (2).

  In the data preprocessing device 710, the above equations (1) and (2) are executed for a plurality of input data, and the normalized data Nxi (n) obtained as a result and the complement CNxi (n) of the normalized data are obtained. ) Is input to the ART module 720 as input data Ii (n). The above procedure is included in the input data conversion processing of the operation data to the ART module 720 performed in the data preprocessing device 710.

  In the ART module 720, the input data Ii (n) is classified into a plurality of categories. For this purpose, the ART module 720 includes an F0 layer 721, an F1 layer 722, an F2 layer 723, a memory 724, and a selection subsystem 725, which are coupled to each other. The F1 layer 722 and the F2 layer 723 are coupled via a weighting factor. The weighting factor represents the prototype (prototype) of the category into which the input data is classified. Here, the prototype represents a representative value of the category.

  Next, the algorithm of the ART module 720 will be described.

  The outline of the algorithm when input data is input to the ART module 720 is as shown in the following processing 1 to processing 5.

  Process 1: The input vector is normalized by the F0 layer 721 whose processing content is shown in FIG. 4B, and noise is removed.

  Process 2: By comparing the input data input to the F1 layer 722 and the weighting coefficient, an appropriate category candidate is selected according to the processing content of FIG.

  Process 3: The validity of the category selected by the selection subsystem 725 is evaluated by the ratio with the parameter ρ. If it is determined to be valid, the input data is classified into the category, and the process proceeds to process 4. On the other hand, if it is not judged to be valid, the category is reset, and an appropriate category candidate is selected from the other categories (repeat processing 2). Increasing the value of parameter ρ makes the category classification finer, and decreasing the value of ρ makes the classification coarse. This parameter ρ is referred to as a vigilance parameter.

  Process 4: When all the existing categories are reset in Process 2, it is determined that the input data belongs to the new category, and a new weighting factor representing the prototype of the new category is generated.

  Process 5: When input data is classified into category J, weight coefficient WJ (new) corresponding to category J uses past weight coefficient WJ (old) and input data p (or data derived from input data). And updated by the following equation (3).

  Here, Kw is a learning rate parameter (0 <Kw <1), and is a value that determines the degree to which the input vector is reflected in the new weighting factor.

  The characteristic of the data classification algorithm of the ART module 720 is in the processing 4 described above.

  In the process 4, when input data different from the pattern recorded (saved) in the classification result database 250 in FIG. 1 is input, a new pattern can be recorded without changing the recorded pattern. Therefore, it is possible to record a new pattern while recording a pattern learned in the past.

  As described above, when the operation data given in advance is given as input data, the ART module 720 learns the given pattern. Therefore, when new input data is input to the learned ART module 720, it is possible to determine which pattern in the past is close by the above algorithm. If the pattern has never been experienced before, it is classified into a new category.

  Specifically, the above-described processing 1 to processing 5 are executed according to the following procedure.

  First, FIG. 4B is a block diagram showing the configuration of the F0 layer 721. The F0 layer 721 is composed of processing function blocks 71, 72, 73, and 74. In the processing function blocks 71, 72, 73, and 74, the following expressions (4), (5), (6), and (7) are executed, respectively. Then, a normalized input vector is obtained.

  In the series of processes in the F0 layer 721 in FIG. 4B, the input data Ii given to the processing function block 71 is normalized again at each time, and finally, from the processing function block 74, the F1 layer 721 and the selection subsystem are normalized. The normalized input vector Ui to be output to 725 is created. In the following equations (4), (5), (6), and (7), “i” represents the number of item data, and “0” represents the F1 layer.

  First, Wi0 is calculated from the input data Ii according to the equation (4). Here, a is a constant.

  Next, Xi0 obtained by normalizing Wi0 in equation (4) is calculated using equation (5). Here, W0 is obtained by normizing Wi0.

  Then, Vi0 obtained by removing noise from Xi0 is calculated using equation (6). However, θ is a constant for removing noise. Since the minute value becomes 0 by the calculation of the equation (6), noise of the input data is removed.

  Finally, a normalized input vector Ui0 is obtained using equation (7). However, V0 is obtained by normizing Vi0. The finally obtained Ui0 becomes the input of the F1 layer.

  FIG. 4C is a block diagram showing the configuration of the F1 layer 722. As shown in FIG. In the F1 layer 722, Ui0 obtained by the equation (7) is held as a short-term memory, and Pi that is finally input to the F2 layer 723 is calculated.

  The F2 layer is composed of processing function blocks 75, 76, 77, 78, 79, 80. In each processing function block 75, 76, 77, 78, 79, 80, the following (8) (9) (10) (11), (12), and (13) are executed. These calculation formulas are collectively shown in formulas (8) to (13). However, a and b are constants, f () is a function shown by the equation (6), and Tj is a fitness calculated by the F2 layer 722. Note that the denominators of equations (9), (11), and (12) are normized.

  FIG. 5 shows the result of categorizing the process quantity of the input plant using the algorithm of the ART module 720 of FIG. FIG. 5A is a graph showing an example of the classification result. This figure displays two items of measurement data as an example, and is represented by a two-dimensional graph. The vertical axis and the horizontal axis indicate standardized measurement data for each item.

  The measurement data is divided into a plurality of categories 750 (circles shown in FIG. 5A) by the ART module 720 in FIG.

In this figure, the two-dimensional measurement data is shown as a two-dimensional graph. However, the present invention is not limited to this, and a category may be created using multi-dimensional coordinates for three or more measurement data. Good.
The classification result of FIG. 5A is obtained by the classification means 400 in the normal state learning mode of FIG. 2A described above, and this result is stored in the classification result database 250. Further, in the diagnosis mode of FIG. 2 (b), it is obtained by the state change detection unit 520.

  FIG. 5B shows an example of a change over time when the measurement signal 1 acquired from the plant 100 changes due to the occurrence of an abnormality. The horizontal axis represents time, and the vertical axis represents measurement signals, category numbers, and the occurrence rate (generation frequency) of new categories. Data D1 and D2 correspond to items A and B, respectively.

  In this figure, items A and B are initially stable at a substantially constant value, but data D1 (item A) decreases immediately before time t1, and then data D2 (item B) immediately after time t1. increased. Thereafter, the data D2 (item B) decreases, and finally both the data D1 (item A) and the data D2 (item B) increase.

  Further, before reaching time t1, the classified category numbers are 1 to 4, which is a reference time category, that is, a normal category. On the other hand, after the time t1, the category numbers of the items A and B are 5 to 7, which is a new category indicating that the state of the plant has changed.

  Along with this, the occurrence rate (occurrence frequency) of the new category increases immediately after the time t1, and it has been diagnosed that the state has changed beyond the threshold. Here, the occurrence ratio of new categories is calculated using a moving average of the number of new categories that occur in a predetermined period.

  Such a state change is detected by the state change detection unit 520 in the diagnosis mode of FIG. The state change detection unit 520 diagnoses that the state has changed when the moving average of the number of new categories generated in a predetermined period exceeds the threshold, or diagnoses that the state has changed when the generated category is a new category. In the case of the normal category, one of the methods for diagnosing every sample can be selected if the state has not changed.

  FIG. 6A is a screen of the image display device 940 in FIG. 1 for setting data items to be extracted by the processing data extraction unit 300. Data items used for diagnosis are determined according to the purpose of diagnosis (contents of abnormality to be detected).

  This screen 940 can be scrolled vertically and horizontally, and has tabs (in the example shown, tabs of group 1 and group 2 are displayed). On the screen, process quantities (data items) A, B, C, D, etc. of the plant are displayed together with the process number PID. In addition, the maximum and minimum values of each process quantity (data item) are displayed, and the operator can divide into groups while judging the correlation between each process quantity or the purpose of diagnosis (content of abnormality to be detected). Select the combination of process quantities that you want to manage. The tab of group 1 in the figure indicates that the process quantities (data items) A, C, and D are selected and positioned as groups to be monitored in relation to each other.

  FIG. 6B is a block diagram showing the configuration of the classifying unit 400. For the process amount in the group set in FIG. 6A, the data preprocessing device and ART described in FIG. Configure to place modules. That is, the classification means 400 executes a series of processes in the data preprocessing device and the ART module as many as the number of groups.

  FIG. 7A, FIG. 7B, and FIG. 7C show modes of data stored in the measurement signal database 230, the reference signal database 240, and the classification result database 250 in FIG. 1, respectively. . These figures may be considered as display screens of the image display device 940 of FIG. Note that the screen 940 is vertically and horizontally scrolled, and tabs are displayed as necessary, as in FIG. 6A or other screens.

  As shown in FIG. 7A, the measurement signal database 230 stores values of a plurality of data items (items A, B, C, etc.) measured at the plant 100 for each sampling period (time on the vertical axis). Is done. On the display screen 940, the values of these data items (items A, B, C, etc.) are displayed in time series on the vertical axis, and a wide range of data can be obtained by using scroll boxes 56a and 56b that can be moved vertically and horizontally. Scroll display is possible.

  In the case of the reference signal database 240 in FIG. 7B, by selecting the tabs 57a and 57b indicating the data sheets of the references 1 and 2, only the items classified for each reference can be displayed together. .

  In the processing data extraction unit 300 in FIG. 1, a data group used for diagnosis of the plant 100 is extracted from the measurement signal database 230. For example, in FIG. 7B, there are two reference signal data groups ("reference 1" and "reference 2"), and when "reference 1" is selected, the data groups are item A, item C, and item. A state constituted by D is displayed. The data corresponding to the data item of group 1 set in FIG. 6A is reference 1, and the data corresponding to the data item of group 2 is reference 2.

  As described above, in the measurement signal database 230, the measurement values of all the data items are stored in time series as one data group as shown in FIG. 7B, the measured values of the data items extracted by the processing data extracting unit 300 are stored in a time series as a plurality of data groups.

  Further, FIG. 7C is a display screen showing the mode of data stored in the classification result database 250 of FIG. On the left side of FIG. 7C, the time and the data at that time are displayed in relation to the classified category number, and on the right side of FIG. 7C, the relationship between the category number and the weighting factor is displayed. Yes. As described above, the classification result database 250 stores the classification results for each data group stored in the reference signal database 240.

  FIG. 8A is a block diagram configuring the abnormality determination unit 510. The abnormality determination unit 510 uses the state change determination result 15 in the state change detection unit 520, the aging deterioration determination result 16 in the aging deterioration detection unit 530, and the operation condition change determination result 17 in the operation condition change detection unit 540. Determine whether an abnormality has occurred in the plant.

  As summarized in FIG. 13, the state change determination result 15 is “1” when it is determined that the state has changed, and is “0” when the state has not changed. The aged deterioration judgment result 16 is “1” when the state change of the plant is out of the aged deterioration range, and is “0” when it is within the aged deterioration range. The operating condition change determination result 17 is “1” when the operating condition is not changed, and is “0” when the operating condition is changed.

  The result of inputting these digital signals to the AND gate 514 in FIG. That is, when the state of the plant changes, the change in the state of the plant deviates from the range of deterioration over time, and the operating condition does not change, the abnormality determination result 10 is set to “1”, and it is determined that an abnormality has occurred. To do. Under other conditions, the abnormality determination result 10 is “0”.

  FIG. 8B is a diagram in which a mode of data stored in the diagnosis result database 260 is displayed on the screen 940. The state change determination result, the operating condition change determination result, the aging deterioration determination result, and the abnormality determination result are each stored in time series and displayed on the screen. Further, in the diagnosis result database 260, the relationship between the time shown in FIG. 7C obtained as a result of operating the state change detection unit 520 and the category number into which the data at that time is classified, and the category number and The relationship with the weighting factor is also saved.

  FIG. 8C is a screen display example when the diagnostic result 12 stored in the diagnostic result database 260 is displayed in time series on the image display device 940. The time-series results can be numerically displayed as shown in FIG. 8B, and can also be displayed as a graph as shown in FIG. 8C. In addition to displaying “0” and “1”, it is also possible to display a moving average value for a predetermined period. In FIG. 8C, the curve 551 is the moving average value of the state change determination result, the curve 552 is the moving average value of the operating condition change determination result, the curve 553 is the moving average value of the aged deterioration determination result, and the curve 554 is the abnormality determination result. The moving average values of are respectively shown.

  FIG. 9A is a diagram for explaining the operation of the aged deterioration detection unit 530. As shown in the graph at the upper left of FIG. 9A, the measurement value gradually changes due to aging. That is, when time is shown on the horizontal axis and the amount of process is shown on the vertical axis, the items A and B tend to change slowly in the case of aging. On the other hand, as shown in the upper right graph, the measured value may change outside the range of aging deterioration, and at this time, there are many cases where an abnormality has occurred or the operating condition has changed. In the upper right graph, the horizontal axis indicates time, and the vertical axis indicates the amount of process.

  Further, it is also possible to determine the identification accompanying the above time change by the change amount of the measured value and the change amount Δz of the category weighting coefficient. Since there is a correlation between these indices, Δz can be used to determine aging degradation. Each graph in the lower part of FIG. 9A shows the change amount Δz of the category weight coefficient between different times with the item A on the horizontal axis and the item B on the vertical axis. In the aging change on the left, the change amount Δz of the category weighting coefficient is small, but when the measured value changes outside the range of aging deterioration, the change amount Δz of the category weighting coefficient appears large.

  Even in a normal state, the tendency of data changes with the aging of the plant. The degree of change in the data trend at the normal state is recorded, and if it is within this range, it is determined that it is within the range of aging degradation.

  FIG. 9B is a diagram for explaining a criterion for determining whether the range of aged deterioration or the range of aged deterioration has been exceeded. This figure shows the time variation width and weighting factor actually measured, with the time variation width on the horizontal axis and the weighting factor variation amount on the vertical axis. Whether or not the relationship of the amount of change has deviated from the boundary determines whether it is within the range of secular change.

  In this embodiment, the determination is made based on the result of summarizing the relationship between the time change width and the weight coefficient change amount in the normal state. Here, the amount of change in the weighting coefficient with respect to the time change width is calculated using the following equation (14).

  In Expression (14), Δt is a time change width, i is a code for identifying a data item, 1 ≦ i ≦ n (n is the total number of data items), and Wi (t) is a data item i at time t. Is a weighting factor.

  In the diagnostic mode, when the maximum value of the change amount of the weighting coefficient in the normal state is exceeded, it is determined that the range of aging deterioration has been exceeded. In this embodiment, the maximum value of the change amount of the weighting coefficient in the normal state is determined as the threshold value. However, the average value of the change amount of the weighting coefficient in the normal state can be set as the threshold value.

  FIG. 10A is a diagram for explaining a first example of the operating condition change detection unit 540. In the present embodiment, the same ART processing (processing by the data preprocessing device 710c and the ART module 720d) of FIG. 4 as that performed by the classification unit 400 and the state change detection unit 520 is executed. However, the difference is that the input to the ART in this case is not the process amount of the plant but the operating condition data. That is, it is determined whether or not the operating condition has changed using the result of classifying the group data composed only of the operating condition data with the ART. Candidate operating conditions include external environmental factors that are not directly related to device characteristics, such as “generated power” and “temperature”.

  If a new category occurs when the operating condition data is classified by ART, it is highly likely that the operating condition has changed. Conversely, when a new category does not occur when the operating condition data is classified by ART and a new category occurs when the measurement data is classified, the cause of the state change is not a change in the operating condition. In this way, it is determined whether or not the operating conditions have changed.

  FIG. 10B is a diagram illustrating a second example of the operating condition change detection unit 540.

  Again, ART is used. In ART, as shown in FIG. 10B, items A and B are shown on the vertical axis and the horizontal axis, and data at the reference time is classified into several normal categories. In this operation, when a new category is generated, the similarity between the target data and the normal category is calculated, and the normal category that is most similar (similarity is maximized) is extracted.

  The similarity S is calculated using, for example, the following equation (15), and a normal category having the smallest S (similarity is minimized) is extracted.

  Here, i is a code for identifying a data item, and 1 ≦ i ≦ n (n is the total number of data items). J is a code for identifying a category, and 1 ≦ j ≦ m (m is the total number of normal categories). Further, Sj is the similarity, Di is the value of the data item i of the target data, and Wij is the weighting coefficient of the data item i in category j.

  Next, the contribution Ci of each data item is calculated using, for example, the following equation (16).

  A data item with a higher contribution is far from the normal category, and can be said to be a data item that causes a new category to occur. When this data item is operating condition data, it is determined that the cause of the change in the state is that the operating condition has changed.

  In the example shown in FIG. 10B, it is seen how much the new category coordinates change on the vertical axis and the horizontal axis with respect to the position of the normal category on the two-dimensional coordinates. In this figure, it can be seen that the change (contribution) on the horizontal axis is larger than that on the vertical axis, and the contribution of item A is large.

  FIG. 10C is a diagram for explaining a third embodiment of the operating condition change detection unit 540. In this embodiment, the functions described in FIGS. 10A and 10B are combined.

  The result 542 determined using FIG. 10A and the contribution processing result 543 determined using FIG. 10B are input to the operating condition change determination unit 544. The operating condition change determination unit 544 performs AND or OR digital processing and outputs an operating condition change determination result 545.

  Below, operation | movement at the time of applying the diagnostic apparatus 200 of this invention to a thermal power plant is demonstrated. FIG. 11 is a block diagram showing a thermal power plant.

  In this figure, the thermal power plant 100 includes a gas turbine generator 110, a control device 120, and a data transmission device 130. The gas turbine generator 110 includes a generator 111, a compressor 112, a combustor 113, and a turbine 114.

  At the time of power generation, the air sucked by the compressor 112 is compressed into compressed air, and this compressed air is sent to the combustor 113 and mixed with fuel and burned. The turbine 114 is rotated using the high-pressure gas generated by the combustion, and the generator 111 generates power.

  In the control device 120, the output of the gas turbine generator 110 is controlled according to the power demand. The control device 120 uses the operation data 102 measured by a sensor (not shown) installed in the gas turbine generator 110 as input data. The operation data 102 is state quantities such as intake air temperature, fuel input amount, turbine exhaust gas temperature, turbine rotation speed, generator power generation amount, turbine shaft vibration, and the like, and is measured at each sampling period. It also measures weather information such as atmospheric temperature.

  In the control device 120, the control signal 101 for controlling the gas turbine generator 110 is calculated using these operation data 102.

  The signal data transmission device 130 transmits the measurement signal 1 including the operation data 102 measured by the control device 120 and the control signal 101 calculated by the control device 120 to the diagnosis device 200.

  FIG. 12 is a diagram for explaining the results of diagnosis using the diagnostic device 200 for the thermal power plant described in FIG. 11.

  FIG. 12 shows the result when the generator output a and the atmospheric temperature b are selected as the data items of the operating condition data of the group 1, and the generator output a, the atmospheric temperature b and the fuel flow rate c are selected as the data items of the group 2. It is.

During time T ₀ -T ₁ , normal data is used to learn the normal state in the learning mode. Even in a normal state, the efficiency decreases due to secular change, and the fuel flow rate c necessary to obtain the same output a increases.

Diagnosing a start using the diagnostic device 200 from time _{T 1.} Although it is in a normal state between times T _{1 and} T ₂ , the fuel flow rate c increases due to the influence of secular change. As a result, when the state change detection unit 520 is operated due to a change in the value of the fuel flow rate c, a new category is generated. Therefore, the state change determination result d after time T ₁ is passed, it becomes close to 1. In addition, the aging deterioration detection unit 530 determines that the increase rate of the fuel flow rate c is within the range at the time of learning, and thus is within the aging deterioration range. From the above, the abnormality determination unit 510 determines that no abnormality has occurred.

Between times _T 2 -T ₃ is ambient temperature b rises to 1 temporary. As the atmospheric temperature b rises, the efficiency decreases, so the fuel flow rate c required to obtain the same output a increases. As a result, when the state change detection unit 520 is operated, a new category is generated. In addition, the aged deterioration detection unit 530 determines that the rate of increase in the fuel flow rate c deviates from the learning range. The operating condition change detection unit 540 determines that the operating condition f has changed since the atmospheric temperature b has changed. From the above, the abnormality determination unit 510 determines that no abnormality has occurred.

After the atmospheric temperature is lowered, and the abnormality is generated at time T _4, the fuel flow rate c is increased accordingly. Since the increasing rate of the fuel flow rate c deviates from the range of the aging deterioration e and the operating condition f has not changed, the abnormality determination unit 510 determines that an abnormality has occurred.

If using the result of the state change detecting section 520 only performed abnormality determination, the time T ₁ after the diagnosis that an abnormality has occurred. This is a false alarm. By using the aging deterioration detection unit 530 and the operating condition change detection unit 540 described in the present invention, it is possible to determine that the range of the times T _{1 to} T ₄ is normal, and to reduce the false alarm rate.

  Thus, by using the diagnostic apparatus 200 of the present invention, an effect of reducing the false alarm rate can be obtained. Further, by providing the operator with information on whether the cause of the change in the data trend belongs to aging deterioration, change in operating conditions, or abnormality, it is useful for planning a maintenance plan for the plant.

  Further, the diagnostic unit 500 of the present invention includes an abnormality determination unit 510, a state change detection unit 520, an aging deterioration detection unit 530, and an operating condition change detection unit 540. However, the aging deterioration detection unit 530 or the operating condition It is also possible to use only one of the change detection units 540 so as to detect either aged deterioration or a change in operating conditions.

  It should be noted that the items illustrated and described above can be configured as a diagnostic device, implemented as a diagnostic method using a computer, or a series of procedures can be implemented as a program.

  The present invention can be widely applied to various plants as a diagnostic apparatus, a diagnostic method, and a diagnostic program for a plant.

100: Plant 200: Diagnosis apparatus 210: External input interface 220: External output interface 230: Measurement signal database 240: Reference signal database 250: Classification result database 260: Diagnosis result database 300: Process data extraction means 400: Classification means 500: Diagnosis Means 510: Abnormality determination unit 520: State change detection unit 530: Aging deterioration detection unit 540: Operation condition change detection unit 600: Alarm generation unit 900: Operation management room 910: External input device 920: Keyboard 930: Mouse 940: Image display apparatus

Claims (10)

A diagnosis device for a plant that detects an abnormality of a diagnosis target plant from a measurement signal of the diagnosis target plant and notifies the abnormality, classifies the measurement signal of the diagnosis target plant into a category, stores it, and generates a new category A state change detection unit that determines that the state has changed when it does not belong to the classified category, an abnormality determination unit that determines that an abnormality has occurred in the diagnosis target plant as a part of the signal from the state change detection unit, An alarm generating means for informing the outside of the output of the abnormality determination unit, an event other than an abnormality of the plant to be diagnosed is detected, and an error report blocking unit for blocking the output of the abnormality determination unit is configured.
The false alarm prevention unit obtains a weighting factor corresponding to the category using the measurement signal of the diagnosis target plant, and deviates from the range of aging deterioration when the change amount of the weighting factor corresponding to the category exceeds the maximum value. An aging deterioration detection unit that determines that the aging deterioration detection unit prevents the output of the abnormality determination unit when it does not deviate from the range of aging deterioration. A diagnosis device for a plant that detects an abnormality of a diagnosis target plant from a measurement signal of the diagnosis target plant and notifies the abnormality, classifies the measurement signal of the diagnosis target plant into a category, stores it, and generates a new category A state change detection unit that determines that the state has changed when it does not belong to the classified category, an abnormality determination unit that determines that an abnormality has occurred in the diagnosis target plant as a part of the signal from the state change detection unit, An alarm generating means for informing the outside of the output of the abnormality determination unit, an event other than an abnormality of the plant to be diagnosed is detected, and an error report blocking unit for blocking the output of the abnormality determination unit is configured.
The diagnosis target plant is a power plant, and the false alarm prevention unit classifies and stores measurement data including at least one of a power generation output or an atmospheric temperature as a measurement signal of the power plant, and a new category is generated. without the the case belonging to the classified category comprises determining the operating condition change detecting unit and the operating conditions are not changed, the operation condition change detecting unit, when the operating condition is changed, the abnormality A plant diagnostic apparatus characterized in that the output of the determination unit is blocked. The plant diagnostic apparatus according to claim 2,
A plant diagnostic apparatus, wherein a measurement signal that is an input of an operating condition change detection unit is composed of external environmental factors that are not directly related to the characteristics of the equipment that constitutes the diagnosis target plant. A diagnosis device for a plant that detects an abnormality of a diagnosis target from a measurement signal of a diagnosis target plant and notifies the abnormality, classifies and stores the measurement signal of the diagnosis target plant into a category, and a new category is generated, State change detection unit that determines that the state has changed when it does not belong to the classified category, the weight coefficient corresponding to the category is obtained using the measurement signal of the diagnosis target plant, and the amount of change in the weight coefficient corresponding to the category is the maximum value When it exceeds the age limit, the aging deterioration detection unit that determines that it has deviated from the range of aging deterioration, the measurement signal of the diagnosis target plant is classified and stored in the category, and no new category is generated, and it belongs to the classified category The operating condition change detection unit that determines that the operating condition has not changed, the state change detection unit determines that the state has changed, and the aging degradation An abnormality determination unit that determines that an abnormality has occurred in the diagnosis target when it is determined that the departure part has deviated from the range of aging deterioration and the operation condition change detection unit has determined that the operation condition has not changed, the abnormality determination A diagnostic device comprising alarm generating means for informing the output of the unit to the outside. A measurement signal database for storing the measurement signal of the diagnosis target plant, a processing data extraction means for extracting a diagnosis signal used for diagnosing the state of the diagnosis target plant from the measurement signal database, and a reference for storing the diagnosis signal A signal database, a classification unit that classifies data stored in the reference signal database, a classification result database that stores the category as a normal category, and the latest diagnostic signal extracted by the processing data extraction unit; A diagnosis means for diagnosing whether the state of the diagnosis target plant belongs to normal, abnormal, operating condition change, or aged deterioration using information of the normal category stored in the classification result database; and Diagnostic result database for storing diagnostic results and the diagnostic results In the diagnostic apparatus of the plant for outputting information stored in database in the image display device,
The diagnosis unit includes an abnormality determination unit, a state change detection unit, an aging deterioration detection unit, and an operating condition change detection unit. The state change detection unit is the latest diagnosis extracted by the processing data extraction unit. A function of determining that the state has changed when the signal does not belong to the normal category stored in the classification result database, and the aging deterioration detection unit is a weight corresponding to a time change width and a category in a normal state A function of deriving a relationship between coefficient variation amounts, and having a function of determining that the range of aged deterioration has been exceeded when a maximum value of a weight coefficient variation amount corresponding to a category in a normal state is exceeded during diagnosis, the operating condition change detection unit, belonging to a new category is not generated, classified category when classified into categories formed measurement signal by external environmental factors not directly related to the equipment characteristics When provided with a function of determining the operating conditions are not changed, in the abnormality determining unit determines that the state has changed by the state change detecting section, and deviates from the range of aging at the aging detector The plant diagnosis apparatus is characterized in that it is determined as abnormal when the operating condition change detecting unit does not determine that the operating condition has not changed. A diagnosis method for detecting an abnormality of a diagnosis target plant from a measurement signal of a diagnosis target plant and notifying the abnormality, classifying and storing the measurement signal of the diagnosis target plant into a category, generating a new category, and classifying If it is determined that the status has changed when it does not belong to the selected category, the weighting factor corresponding to the category is obtained using the measurement signal of the diagnosis target plant, and the amount of change in the weighting factor corresponding to the category exceeds the maximum value , It is determined that it has deviated from the range of aging deterioration, the measurement signal to be diagnosed is classified and stored in a category, and if a new category does not occur and belongs to the classified category, it is determined that the operating condition has not changed. A plant diagnosis method that determines that an abnormality has occurred in the diagnosis target plant and notifies the outside of the abnormality. A diagnosis program for a plant that detects an abnormality of a diagnosis target plant from a measurement signal of the diagnosis target plant and notifies the abnormality, classifies and stores the measurement signal of the diagnosis target plant into a category, and generates a new category A state change detection step for determining that the state has changed when the device does not belong to the classified category, an abnormality determination step for determining that an abnormality has occurred in the diagnosis target as a part of the input from the signal from the state change detection step, An alarm generation step for notifying the output of the abnormality determination step to the outside, an event other than the abnormality of the diagnosis target is detected, and a false alarm prevention step for blocking the output of the abnormality determination step is configured.
In the false alarm prevention step, a weighting factor corresponding to the category is obtained using the measurement signal of the diagnosis target plant, and when the change amount of the weighting factor corresponding to the category exceeds a maximum value, it deviates from the range of aging deterioration. And a aging deterioration detecting step for determining that the aging deterioration detecting step prevents an output of the abnormality determining step when the aging deterioration does not deviate from the aging deterioration range. A diagnosis program for a plant that detects an abnormality of a diagnosis target plant from a measurement signal of the diagnosis target plant and notifies the abnormality, classifies and stores the measurement signal of the diagnosis target plant into a category, and generates a new category A state change detection step for determining that the state has changed when the device does not belong to the classified category, an abnormality determination step for determining that an abnormality has occurred in the diagnosis target as a part of the input from the signal from the state change detection step, An alarm generation step for notifying the output of the abnormality determination step to the outside, an event other than the abnormality of the diagnosis target is detected, and a false alarm prevention step for blocking the output of the abnormality determination step is configured.
The diagnosis target plant is a power plant, and the false alarm prevention step classifies and stores measurement data including at least one of a power generation output or an atmospheric temperature as a measurement signal of the power plant, and a new category is generated. Without changing the operation condition, the operation condition change detection step determines that the operation condition has changed. When the operation condition changes, the abnormality determination A plant diagnostic program characterized in that the output of a step is blocked. The plant diagnostic program according to claim 8,
A plant diagnostic program characterized in that a measurement signal that is an input of an operating condition change detection step is composed of external environmental factors that are not directly related to the characteristics of the equipment that constitutes the diagnostic object. A diagnosis program for a plant that detects an abnormality of a diagnosis target plant from a measurement signal of the diagnosis target plant and notifies the abnormality, classifies and stores the measurement signal of the diagnosis target plant into a category, and generates a new category The state change detection step that determines that the state has changed when it does not belong to the classified category, the weight coefficient corresponding to the category is obtained using the measurement signal of the diagnosis target plant, and the amount of change in the weight coefficient corresponding to the category is the maximum When the value exceeds the value, the aging deterioration detection step that determines that the range of aging deterioration has been deviated and the measurement signal of the diagnosis target plant are classified and stored in a category, and no new category is generated, and it belongs to the classified category In the operation condition change detection step for determining that the operation condition has not changed, the state change detection step Is detected, and it is determined in the aging deterioration detection step that it has deviated from the range of aging deterioration, and the operation condition change detection step determines that the operating condition has not changed. A plant diagnosis program comprising: an abnormality determination step for determining that the failure has occurred; and an alarm generation step for notifying the output of the abnormality determination step to the outside.

Images