Complete plant monitoring device, complete plant monitoring method and program product

Abstract

The complete plant monitoring device (20) comprises: a detection value acquisition unit (211) for acquiring a detection value bundle; a first Mahalanobis distance calculation unit (212) for calculating the first Mahalanobis distance based on a unit space created by a past detection value bundle; a first SN ratio calculation unit (214) for calculating the first SN ratio of each of the plurality of evaluation items; a second Mahalanobis distance calculation unit (215) for calculating the second Mahalanobis distance by increasing or decreasing each value of the plurality of detection values; a second SN ratio acquisition unit (216) for converting the first SN ratio of each of the evaluation items into a second SN ratio based on the first Mahalanobis distance and the second Mahalanobis distance, and acquiring the second SN; and an addition unit (217) for calculating the sum of the plurality of second SN ratios acquired within a specified period for each of the plurality of evaluation items.

Description

Technical Field

[0001] The present invention relates to a complete set of equipment monitoring device, a complete set of equipment monitoring method and a program.

[0002] This application claims priority based on Japanese Patent Application No. 2019-236772 filed in Japan on December 26, 2019, and the contents are incorporated herein by reference. Background Art

[0003] In various plants such as gas turbine power plants, nuclear power plants, chemical plants, and their remote monitoring systems, a system that displays important sensor values ​​is used to monitor the long-term trend of the plants. In such a system, it is necessary to closely monitor the sensor values ​​manually to grasp the long-term trend, which has become one of the reasons for delaying mechanization and AI. Therefore, for example, Patent Document 1 describes a system that monitors the long-term trend of a plant by finding a regression formula that represents the long-term trend of an index that takes into account the margin and deviation up to the limit value of each of a plurality of sensor values. In this system, the period when the index of each sensor exceeds a specified threshold, that is, the period when an abnormality such as a failure is likely to occur, is predicted based on the regression formula.

[0004] Previous technical literature

[0005] Patent Literature

[0006] Patent Document 1: Japanese Patent Application Publication No. 2011-60012 Summary of the invention

[0007] Technical issues to be solved by the invention

[0008] However, in recent years, there has been a demand for a system that is more versatile and universal and can easily grasp long-term trends for various plants with different specifications, operating conditions, sensor types, etc.

[0009] The present invention has been made in view of such a problem, and provides a plant monitoring device, a plant monitoring method, and a program that can easily grasp long-term trends while improving versatility and generality.

[0010] Means for solving technical problems

[0011] According to one embodiment of the present invention, a complete plant monitoring device comprises: a detection value acquisition unit that acquires a set of detection values ​​for each of a plurality of evaluation items, namely, a detection value bundle; a first Mahalanobis distance calculation unit that calculates the first Mahalanobis distance of the detection value bundle based on a unit space created by a past detection value bundle; a first SN ratio calculation unit that calculates the first SN ratio of each of the plurality of evaluation items; a second Mahalanobis distance calculation unit that increases or decreases each value of the plurality of detection values ​​to calculate a second Mahalanobis distance corresponding to the increased or decreased detection values, respectively; a second SN ratio acquisition unit that converts the first SN ratio of each of the evaluation items into a second SN ratio based on the first Mahalanobis distance and the second Mahalanobis distance, and acquires the second SN ratio; and an addition unit that calculates the sum of the plurality of second SN ratios acquired within a prescribed period for each of the plurality of evaluation items.

[0012] According to one embodiment of the present invention, a complete set of equipment monitoring method includes: a step of obtaining a set of detection values, i.e., a detection value bundle, for each of a plurality of evaluation items; a step of calculating the first Mahalanobis distance of the detection value bundle based on a unit space created by a past detection value bundle; a step of calculating the first SN ratio of each of the plurality of evaluation items; a step of increasing or decreasing each value of the plurality of detection values ​​to calculate the second Mahalanobis distance corresponding to the increased or decreased detection values, respectively; a step of converting the first SN ratio of each of the evaluation items into a second SN ratio based on the first Mahalanobis distance and the second Mahalanobis distance, and obtaining the second SN ratio; and a step of calculating the sum of the plurality of second SN ratios obtained within a specified period for each of the plurality of evaluation items.

[0013] According to one embodiment of the present invention, a program causes a computer of a complete set of equipment monitoring devices to execute the following steps: a step of obtaining a set of detection values, i.e., a detection value bundle, for each of a plurality of evaluation items; a step of calculating a first Mahalanobis distance of the detection value bundle based on a unit space created by a past detection value bundle; a step of calculating a first SN ratio for each of a plurality of the evaluation items; a step of increasing or decreasing each value of a plurality of the detection values ​​to calculate a second Mahalanobis distance corresponding to the increased or decreased detection values, respectively; a step of converting the first SN ratio of each of the evaluation items into a second SN ratio based on the first Mahalanobis distance and the second Mahalanobis distance and obtaining the second SN ratio; and a step of calculating the sum of a plurality of the second SN ratios obtained within a specified period for each of the plurality of the evaluation items.

[0014] Effects of the Invention

[0015] According to the plant monitoring device, plant monitoring method and program of the present invention, versatility and generality are improved while long-term trends can be easily grasped. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 This is a diagram for explaining the outline of the plant monitoring device according to the first embodiment of the present invention.

[0017] Figure 2 It is a diagram showing the functional configuration of the plant monitoring device according to the first embodiment of the present invention.

[0018] Figure 3 This is a first flowchart showing an example of processing of the plant monitoring device according to the first embodiment of the present invention.

[0019] Figure 4 This is a diagram showing an example of detection values ​​according to the first embodiment of the present invention.

[0020] Figure 5 This is a diagram showing an example of changes in the added value of the second SN ratio according to the first embodiment of the present invention.

[0021] Figure 6 This is a second flowchart showing an example of processing of the plant monitoring device according to the first embodiment of the present invention.

[0022] Figure 7 This is a flowchart showing an example of processing of the plant monitoring device according to the second embodiment of the present invention.

[0023] Figure 8 This is a diagram showing an example of the hardware configuration of a plant monitoring device according to at least one embodiment of the present invention.

[0024] Fig. 9 This is a flowchart showing an example of processing of the plant monitoring device according to the first modified example of the present invention.

[0025] Fig.10 This is a flowchart showing an example of processing of the plant monitoring device according to the second modified example of the present invention. DETAILED DESCRIPTION

[0026] <First embodiment>

[0027] Below, reference Figures 1 to 6 A plant monitoring device 20 according to a first embodiment of the present invention will be described.

[0028] (Overall structure)

[0029] Figure 1 This is a diagram for explaining the outline of the plant monitoring device according to the first embodiment of the present invention.

[0030] like Figure 1 As shown, the plant monitoring device 20 according to the present embodiment is a device for monitoring the operating state of a plant 1 having a plurality of evaluation items. The plant monitoring device 20 acquires a detection value representing a state quantity of each evaluation item from sensors provided in each part of the plant 1. Then, the plant monitoring device 20 uses the Mahalanobis-Taguchi method (hereinafter referred to as the MT method) to determine whether the operating state of the plant 1 is normal or abnormal based on the acquired detection value.

[0031] The plant 1 involved in this embodiment is a gas turbine combined power generation plant, which includes a gas turbine 10, a gas turbine generator 11, a waste heat recovery boiler 12, a steam turbine 13, a steam turbine generator 14, and a control device 40. In addition, in another embodiment, the plant 1 can be a gas turbine power generation plant, a nuclear power generation plant, or a chemical plant.

[0032] The gas turbine 10 includes a compressor 101 , a combustor 102 , and a turbine 103 .

[0033] The compressor 101 compresses the air sucked from the air inlet. The compressor 101 is provided with temperature sensors 101A and 101B as sensors for detecting the temperature in the machine room of the compressor 101, which is one of the evaluation items. For example, the temperature sensor 101A detects the temperature of the machine room inlet (inlet air temperature) of the compressor 101, and the temperature sensor 101B detects the temperature of the machine room outlet (outlet air temperature). In addition, the compressor 101 may also be provided with a pressure sensor, a flow sensor, etc.

[0034] The combustor 102 mixes fuel F with compressed air introduced from the compressor 101 and burns the mixture to generate combustion gas. The combustor 102 is provided with a pressure sensor 102A as a sensor for detecting the pressure of the fuel F, which is one of the evaluation items.

[0035] The turbine 103 is driven to rotate by the combustion gas supplied from the combustor 102. The turbine 103 is provided with temperature sensors 103A and 103B as sensors for detecting the temperature in the engine room, which is one of the evaluation items. For example, the temperature sensor 103A can detect the temperature of the engine room inlet of the turbine 103 (inlet combustion gas temperature), and the temperature sensor 103B can detect the temperature of the engine room outlet (outlet combustion gas temperature).

[0036] The gas turbine generator 11 is connected to the rotor 104 of the turbine 103 via the compressor 101, and generates electricity by the rotation of the rotor 104. The gas turbine generator 11 is provided with a thermometer 11A as a sensor for detecting the temperature of lubricating oil, which is one of the evaluation items.

[0037] The waste heat recovery boiler 12 generates steam by heating water using combustion gas (exhaust gas) discharged from the turbine 103. The waste heat recovery boiler 12 is provided with a liquid level gauge 12A as a sensor for detecting the water level of a steam drum, which is one of the evaluation items.

[0038] The steam turbine 13 is driven by steam from the waste heat recovery boiler 12. The steam turbine 13 is provided with a temperature sensor 13A as a sensor for detecting the temperature in the engine room, which is one of the evaluation items. The steam discharged from the steam turbine 13 is converted back into water by the condenser 132 and is delivered to the waste heat recovery boiler 12 via a water supply pump.

[0039] The steam turbine generator 14 is connected to the rotor 131 of the steam turbine 13, and generates electricity by the rotation of the rotor 131. The steam turbine generator 14 is provided with a thermometer 14A as a sensor for detecting the temperature of the lubricating oil which is one of the evaluation items.

[0040] In addition, the above evaluation items are only examples and are not limited thereto. Other evaluation items of the plant 1 may include, for example, the output of the gas turbine generator 11, the pressure in the engine room of the turbine 103, the rotation speed and vibration of the rotor 104, etc. In this case, although not shown in the figure, it is considered that each part of the plant 1 is provided with a sensor for detecting the state quantity of these evaluation items.

[0041] The control device 40 is a device for controlling the operation of the plant 1 . When the plant monitoring device 20 determines that the operating state of the plant 1 is abnormal, the control device 40 controls the operation of each part of the plant 1 based on a control signal from the plant monitoring device 20 .

[0042] (Functional structure of plant monitoring device)

[0043] Figure 2 It is a diagram showing the functional configuration of the plant monitoring device according to the first embodiment of the present invention.

[0044] like Figure 2 As shown, the plant monitoring device 20 includes a CPU 21 , an input / output interface 22 , a display unit 23 , an operation accepting unit 24 , and a storage unit 25 .

[0045] The input / output interface 22 is connected to detectors of various parts of the plant 1 , and receives input of detection values ​​for each of a plurality of evaluation items.

[0046] The display unit 23 is a display for displaying the determination result of the operating state of the plant 1 by the plant monitoring device 20 and the like.

[0047] The operation accepting unit 24 is a device such as a keyboard and a mouse for accepting operations performed by a worker who monitors the plant 1 .

[0048] The CPU 21 is a processor that controls the overall operation of the plant monitoring device 20. The CPU 21 performs various calculations according to a pre-prepared program, thereby functioning as a detection value acquisition unit 211, a first Mahalanobis distance calculation unit 212, a plant state determination unit 213, a first SN ratio calculation unit 214, a second Mahalanobis distance calculation unit 215, a second SN ratio acquisition unit 216, an addition unit 217, and a trend determination unit 218.

[0049] The detection value acquisition unit 211 acquires a detection value bundle, which is a set of detection values ​​for each of a plurality of evaluation items, from the plant 1 via the input / output interface 22. The detection value acquisition unit 211 acquires the detection value bundle every predetermined time (for example, 1 minute) and stores and accumulates it in the storage unit 25.

[0050] The first Mahalanobis distance calculation unit 212 calculates the Mahalanobis distance (hereinafter also referred to as “first Mahalanobis distance” or “first MD”) of the detection value bundles based on a unit space constituted of a plurality of detection value bundles which are past operation data.

[0051] The plant state determination unit 213 determines whether the operating state of the plant 1 is normal or abnormal based on whether the first Mahalanobis distance is equal to or smaller than a predetermined threshold value.

[0052] The first SN ratio calculation unit 214 calculates an SN ratio for each of a plurality of evaluation items (hereinafter, also referred to as “first SN ratio”).

[0053] The second Mahalanobis distance calculation unit 215 increases or decreases each of the plurality of detection values ​​to calculate the Mahalanobis distance (hereinafter also referred to as "second Mahalanobis distance" or "second MD") corresponding to the increased or decreased detection values. In the present embodiment, the second Mahalanobis distance calculation unit 215 increases each of the detection values ​​by a predetermined amount to calculate the second Mahalanobis distance.

[0054] The second SN ratio acquisition unit 216 converts the first SN ratio of each evaluation item into a second SN ratio based on the first Mahalanobis distance and the second Mahalanobis distance, and acquires the second SN ratio.

[0055] The adding unit 217 calculates the added value of the plurality of second SN ratios acquired within a predetermined period for each of the plurality of evaluation items.

[0056] The trend determination unit 218 determines whether the detection value of each of the plurality of evaluation items tends to increase or decrease based on the added value of the second SN ratio.

[0057] The output unit 219 creates the operating state information of the plant 1 and outputs it to the display unit 23 for display. For example, the output unit 219 creates the operating state information including the determination result of the plant state determination unit 213 (information indicating whether the operating state of the plant 1 is normal or abnormal). Furthermore, the output unit 219 may include in the operating state information information the information in which the detected value of each evaluation item to date, the first Mahalanobis distance, the first SN ratio, the second SN ratio, and the like are graphically displayed.

[0058] Furthermore, when a worker performs an operation for remotely controlling the plant 1 via the operation receiving unit 24 , the output unit 219 may output a control signal corresponding to the operation to the control device 40 of the plant 1 .

[0059] The storage unit 25 stores data and the like acquired and generated in the processing of each unit of the CPU 21 .

[0060] (Processing flow of complete plant monitoring device)

[0061] Figure 3 This is a first flowchart showing an example of processing of the plant monitoring device according to the first embodiment of the present invention.

[0062] Below, reference Figure 3 An example of a process in which the plant monitoring device 20 monitors the presence or absence of abnormality in the operating state of the plant 1 and the long-term trend of the plant 1 will be described.

[0063] First, the detection value acquisition unit 211 acquires the detection value of each of the plurality of evaluation items from the detectors provided in each unit of the plant 1 (step S1). For example, when the number of evaluation items is 100, the detection value acquisition unit 211 acquires 100 detection values ​​corresponding to each evaluation item and stores them in the storage unit 25 as one bundle (detection value bundle).

[0064] Next, the first Mahalanobis distance calculation unit 212 calculates the first Mahalanobis distance (first MD) of the detection value bundle acquired in step S1 with reference to the unit space (step S2 ).

[0065] The unit space is a collection of data that serves as a reference for determining the operating status of a complete set of equipment, and is generated by gathering a bundle of detection values ​​when the operating status of the complete set of equipment 1 is normal. The collection period of the detection value bundle used to generate the unit space is a period earlier than the evaluation time point of the operating status of the complete set of equipment 1, and it changes with the passage of time. In this embodiment, the unit space is composed of a bundle of detection values ​​collected from the present to the time when a specified time has passed. That is, the bundle of detection values ​​constituting the unit space changes from old to new over time. The unit space generated and updated in this way is stored in the storage unit 25. As a result, the following situation is suppressed: for example, changes in detection values ​​caused by years of degradation, seasonal changes, etc. affect the Mahalanobis distance, resulting in an erroneous judgment of abnormal operating status in the complete set of equipment status determination unit 213.

[0066] Furthermore, the Mahalanobis distance is a distance weighted by the variance and correlation of the detection values ​​in the unit space, and the lower the similarity between the data groups in the unit space, the larger the value. For example, the average Mahalanobis distance of the detection value bundle (normal data group) constituting the unit space is 1. Furthermore, when the operating state of the plant 1 is normal, the first Mahalanobis distance of the detection value bundle acquired in step S1 is approximately 4 or less. However, if the operating state of the plant 1 becomes abnormal, the first Mahalanobis distance becomes larger according to the degree of abnormality.

[0067] Next, the plant state determination unit 213 determines whether the first Mahalanobis distance calculated in step S2 is less than a threshold value (step S3). In addition, regarding the threshold value, the storage unit 25 stores a value preset according to the characteristics of the plant 1. In addition, the plant monitoring device 20 can accept changes in the threshold value from the operator via the operation acceptance unit 24.

[0068] When the first Mahalanobis distance is less than the threshold value (step S3: Yes), the plant state determination unit 213 determines that the operating state of the plant 1 is normal (step S4). At this time, the output unit 219 can display the operating state information including the determination result that the operating state of the plant 1 is normal on the display unit 23.

[0069] Then, when it is determined that the operating state of the plant 1 is normal, the plant monitoring device 20 executes a series of processes for monitoring the long-term trend of each evaluation item. For example, when there are k evaluation items, the following processes are repeatedly executed for the detection values ​​i (i=1, 2, ..., k) corresponding to the evaluation items 1 to k acquired in step S1 (step S5).

[0070] First, the first SN ratio calculation unit 214 calculates the first SN ratio which is the maximum SN ratio of the detection value i (step S6 ). The method of calculating the SN ratio is known, and thus the description thereof is omitted.

[0071] Next, the second Mahalanobis distance calculation unit 215 determines whether the first SN ratio of the detection value i is a positive number (a value larger than “0”) (step S7 ).

[0072] When the first SN ratio of the detection value i is not a positive number (is "0" or a negative number), the second Mahalanobis distance calculation unit 215 does not perform the process of calculating the second Mahalanobis distance. And, in this case, the second SN ratio acquisition unit 216 sets the second SN ratio of the detection value i to "0" (step S8). And, the second SN ratio is stored and accumulated in the storage unit 25. Then, the plant monitoring device 20 ends the long-term trend monitoring process of the detection value i and proceeds to step S16.

[0073] On the other hand, when the first SN ratio of the detection value i is a positive number, the second Mahalanobis distance calculation unit 215 calculates the second Mahalanobis distance (second MD) of the detection value i (step S9). Specifically, the second Mahalanobis distance calculation unit 215 increases the value of the detection value i by a predetermined amount to calculate the second Mahalanobis distance corresponding to the increased detection value i. At this time, the second Mahalanobis distance calculation unit 215 can set the predetermined amount of each evaluation item according to the detection value bundle constituting the unit space. For example, the second Mahalanobis distance calculation unit 215 extracts the detection values ​​of the same evaluation item as the detection value i from the unit space to calculate their standard deviations. Then, the second Mahalanobis distance calculation unit 215 sets the value of 1 / 10000 to 5 / 10 of the standard deviation (+0.0001σ~+0.5σ) as the predetermined amount added to the detection value i. The predetermined amount is more preferably a value of 1 / 1000 to 1 / 10 of the standard deviation (+0.001σ to +0.1σ), and most preferably a value of 1 / 100 to 1 / 10 of the standard deviation (+0.01σ to +0.1σ).

[0074] Next, the second SN ratio acquisition unit 216 determines whether the second Mahalanobis distance is equal to or greater than the first Mahalanobis distance (step S10 ).

[0075] When the second Mahalanobis distance is greater than or equal to the first Mahalanobis distance (step S10: Yes), the second SN ratio acquisition unit 216 determines that the detection value i tends to be "higher" than the other detection values. Therefore, the second SN ratio acquisition unit 216 acquires a value obtained by converting the first SN ratio of the detection value i into a "positive number" as the second SN ratio of the detection value i (step S11). And, the second SN ratio is stored and accumulated in the storage unit 25.

[0076] On the other hand, when the second Mahalanobis distance is smaller than the first Mahalanobis distance (step S10: No), the second SN ratio acquisition unit 216 determines that the detection value i tends to be "lower" than the other detection values. Therefore, the second SN ratio acquisition unit 216 acquires a value obtained by converting the first SN ratio of the detection value i into a "negative number" as the second SN ratio of the detection value i (step S12). And, the second SN ratio is stored and accumulated in the storage unit 25.

[0077] Next, the adding unit 217 adds the second SN ratios of the detection values ​​i accumulated in the storage unit 25 (step S13 ). That is, the adding unit 217 adds the second SN ratios of all the detection values ​​i calculated from the first startup of the plant 1 to the present.

[0078] Furthermore, the output unit 219 generates a graph showing changes in the added values ​​in which the added values ​​calculated so far are arranged in time series ( Figure 5 ), and displays it on the display unit 23 as the operation status information (step S14).

[0079] Figure 4 This is a diagram showing an example of detection values ​​according to the first embodiment of the present invention.

[0080] For example, it is assumed that the detection value i is a detection value of the compressor efficiency of the compressor 101. In this case, Figure 4 As shown, the detection value acquisition unit 211 acquires and accumulates the detection value of the compressor efficiency at each time point from the time (t0) when the complete set of equipment 1 is first started to the present (t8). In addition, the detection value acquisition unit 211 can calculate the compressor efficiency based on the detection values ​​of the temperature, pressure, flow rate sensors, etc. provided in the compressor 101, thereby acquiring the compressor efficiency. The detection value of each evaluation item will also change depending on the specifications of the complete set of equipment 1, operating conditions, sensor type, etc. Therefore, even if the staff of the complete set of equipment 1 confirms that Figure 4 It is also difficult to understand the long-term trend of the compressor efficiency from the time-series changes in the detected values ​​(raw data) of the compressor efficiency shown.

[0081] Furthermore, for example, although slight damage to the compressor does not immediately have a significant impact, it sometimes gradually reduces the compressor efficiency over a long period of time. At this time, the value of the compressor efficiency contained in the unit space will also change over time, so the unit space will gradually move as the compressor efficiency decreases. In this way, the value of the compressor efficiency will tend to decrease over a long period of time, but it may be difficult to reflect it in the Mahalanobis distance as the unit space is updated. Therefore, in the previous system using the Mahalanobis-Taguchi method (MT method), it is sometimes difficult to detect abnormalities based on such a detection value that gradually decreases or increases over a long period of time.

[0082] Figure 5 This is a diagram showing an example of changes in the added value of the second SN ratio according to the first embodiment of the present invention.

[0083] In summary, whenever the detection value acquisition unit 211 acquires the detection value of the compressor efficiency ( Figure 4 ), the adding unit 217 of the present embodiment will calculate the added value of the second SN ratio of the past compressor efficiency and accumulate it in the storage unit 25. And, in step S14, the output unit 219 will output the following: Figure 5 The change in the added value of the compressor efficiency shown is displayed on the display unit 23. Figure 5 It can be confirmed from the added value change shown that the added value of the second SN ratio tends to decrease over time. That is, the staff of the plant 1 can easily understand that the compressor efficiency tends to decrease in the long term based on the added value change displayed on the display unit 23. Therefore, even if it is determined in steps S3 to S4 that the first Mahalanobis distance is less than the threshold value and the operating state of the plant 1 is normal, the staff can easily understand that the plant 1 is gradually approaching the abnormal side by confirming the long-term trend of each evaluation item of the plant 1.

[0084] In addition, the long-term trend of each evaluation item may be determined by the trend determination unit 218 of the plant monitoring device 20 instead of the staff (step S15 ).

[0085] Figure 6 This is a second flowchart showing an example of processing of the plant monitoring device according to the first embodiment of the present invention.

[0086] like Figure 6 As shown, first, the trend determination unit 218 determines Figure 3 The operator determines whether the added value calculated in step S13 is within a predetermined range (for example, within ±100) (step S150). The predetermined range can be set to any value by the operator through the operation receiving unit 24.

[0087] For example, if the added value is within the predetermined range (step S150 : Yes), the trend determination unit 218 determines that the detection value i will not change significantly in the long term (step S151 ). That is, the trend determination unit 218 determines that the detection value i has not detected an increasing trend or a decreasing trend.

[0088] On the other hand, when the added value exceeds the predetermined range (step S150 : No) and is a positive number, the trend determination unit 218 determines that the detection value i tends to increase in the long term (step S152 ).

[0089] Then, when the added value exceeds the predetermined range (step S150 : No) and is a negative number, the trend determination unit 218 determines that the detection value i tends to decrease in the long term (step S153 ).

[0090] Furthermore, when the added value exceeds the prescribed range (step S150: No), the output unit 219 outputs a warning message including the determination result of the trend determination unit 218 (whether the detection value i tends to increase or decrease) and information that can identify the evaluation item (the name of the evaluation item or the identification number) (step S154). At this time, the output unit 219 can display the warning message on the display unit 23, or send it to a terminal device held by the staff via an e-mail. Thus, the output unit 219 enables the staff to quickly and easily identify which evaluation item has changed in what way in the long-term trend.

[0091] Next, return to Figure 3 When the processing of steps S6 to S16 is completed for all evaluation items 1 to k, the plant monitoring device 20 ends a series of monitoring processing (step S16 ).

[0092] Furthermore, when the first Mahalanobis distance exceeds the threshold value (step S3: No), the plant state determination unit 213 determines that the operating state of the plant 1 is abnormal (step S17). At this time, the plant monitoring device 20 implements the processing when the plant 1 is abnormal (step S18). The content of the processing when the abnormality occurs is the same as that of the conventional system. For example, the first SN ratio calculation unit 214 calculates the expected maximum SN ratio of each evaluation item, and estimates the detection value that is the cause of the increase in the Mahalanobis distance based on the difference in the expected maximum SN ratio of the item with or without the item analyzed by the orthogonal table. The output unit 219 displays the evaluation item name, SN ratio, etc. of the detection value that is the cause estimated by the first SN ratio calculation unit 214 on the display unit 23 as the operating state information. The output unit 219 receives the operation performed by the staff of the plant 1 through the operation receiving unit 24, and outputs a control signal to the control device 40 of the plant 1. In addition, the output unit 219 can automatically output a control signal to stop the plant 1 to the control device 40 of the plant 1.

[0093] (Effect)

[0094] As described above, the complete plant monitoring device 20 involved in the present embodiment obtains the second SN ratio obtained by converting the first SN ratio of each evaluation item based on the first Mahalanobis distance and the second Mahalanobis distance, and calculates the added value of the second SN ratio for each evaluation item. The first Mahalanobis distance is calculated based on the detection value of each evaluation item, and the second Mahalanobis distance is calculated by changing each detection value.

[0095] Thus, the plant monitoring device 20 enables the operator to easily grasp the long-term trend of each evaluation item by the added value of the second SN ratio. Furthermore, the plant monitoring device 20 can express the long-term trend of each evaluation item by the added value of the second SN ratio regardless of the specifications, operating conditions, sensor types, and contents of the evaluation items of the plant 1, thereby improving versatility and generality.

[0096] Furthermore, the plant monitoring device 20 determines whether the operating state of the plant 1 is normal or abnormal based on whether the first Mahalanobis distance is less than a predetermined threshold value. When it is determined that the first Mahalanobis distance is less than the predetermined threshold value and the operating state of the plant 1 is normal, the plant monitoring device 20 further obtains the first SN ratio, the second Mahalanobis distance, and the second SN ratio to monitor the long-term trend of each evaluation item of the plant 1.

[0097] Thus, the plant monitoring device 20 can monitor both abnormalities in the operating state of the plant 1 and the long-term trend of the plant 1 at the same time.

[0098] Then, the plant monitoring device 20 determines whether the detection value of each evaluation item tends to increase or decrease based on the added value of the second SN ratio.

[0099] In this way, the plant monitoring device 20 can automatically diagnose the long-term trend of the plant 1 without the need for the staff to monitor the added value of the second SN ratio in sequence. Furthermore, the plant monitoring device 20 can output the determination result of the long-term trend. Thus, the plant monitoring device 20 allows the staff to quickly and easily identify how the long-term trend of each evaluation item has changed.

[0100] Furthermore, when the added value of the second SN ratio of a certain evaluation item exceeds a prescribed range, the complete set equipment monitoring device 20 determines that the detection value of the evaluation item tends to increase if the added value of the second SN ratio is a positive number, and determines that the detection value of the evaluation item tends to decrease if the added value of the second SN ratio is a negative number.

[0101] In this way, the plant monitoring device 20 can automatically determine the long-term trend of the detection value of each evaluation item, and thus can reduce the monitoring burden on the operator.

[0102] The plant monitoring device 20 calculates the second Mahalanobis distance by increasing each detection value by a predetermined amount, and converts the first SN ratio into a positive number or a negative number to obtain the second SN ratio according to whether the second Mahalanobis distance is greater than the first Mahalanobis distance.

[0103] In this way, the plant monitoring device 20 can determine whether a certain detection value tends to be higher than other detection values ​​based on whether the second Mahalanobis distance increases, and reflect this in the second SN ratio.

[0104] Furthermore, when the first SN ratio is a negative number, the plant monitoring device 20 sets the second SN ratio to zero.

[0105] When the first SN ratio of a certain evaluation item is negative, it can be determined that the effect of the detection value of the evaluation item on the increase of the first Mahalanobis distance is so small that it can be ignored. Therefore, the complete plant monitoring device 20 can simplify the processing by considering the second SN ratio of the evaluation item with such a small influence as zero. And, thereby, the complete plant monitoring device 20 can divide the second SN ratio of the detection value of the evaluation item that affects the first Mahalanobis distance into a positive number or a negative number according to the height relative to the detection value of other evaluation items. As a result, the complete plant monitoring device 20 can provide the staff with data that is easy to intuitively understand whether the detection value tends to increase or decrease according to whether the added value is a positive number or a negative number.

[0106] <Second embodiment>

[0107] Next, refer to Figure 7 A plant monitoring device 20 according to a second embodiment of the present invention will be described.

[0108] The same components as those in the first embodiment are denoted by the same reference numerals, and detailed description thereof will be omitted. In the present embodiment, the processing of the trend determination unit 218 of the plant monitoring device 20 is different from that in the first embodiment.

[0109] (Processing flow of complete plant monitoring device)

[0110] Figure 7 This is a flowchart showing an example of processing of the plant monitoring device according to the modification example of the first embodiment of the present invention.

[0111] The trend determination unit 218 according to this embodiment is Figure 3 In step S15, Figure 7 The series of processing shown is replaced by Figure 6 The series of processing shown.

[0112] In this embodiment, the trend determination unit 218 Figure 3 The sum of the second SN ratios of the detection value i calculated in step S13 is regarded as one detection value. Figure 7 As shown, the trend determination unit 218 calculates the third Mahalanobis distance (third MD) of the added value (step S250). In addition, in the present embodiment, the trend determination unit 218 generates a unit space of the added values ​​of the plurality of second SN ratios calculated from the current time to the elapse of the specified time at each specified time. The trend determination unit 218 calculates the third Mahalanobis distance (third MD) of the added value based on the unit space of the added value. Figure 3 The trend determination unit 218 may generate a unit space including a plurality of detection value bundles collected from the current period to the predetermined time in addition to the added value of the second SN ratio, and calculate the third Mahalanobis distance of the added value calculated in step S13. Figure 3 The third Mahalanobis distance of the detection value bundle consisting of the plurality of detection values ​​(detection value 1 to detection value k) acquired in step S1 and the added value (detection value k+1) calculated in step S13 .

[0113] Next, the trend determination unit 218 determines whether the third Mahalanobis distance is less than or equal to a threshold value (step S251). In addition, regarding the threshold value, the storage unit 25 stores a value preset according to the characteristics of the plant 1. In addition, the plant monitoring device 20 can accept changes in the threshold value from the operator via the operation acceptance unit 24.

[0114] When the third Mahalanobis distance is less than the threshold value (step S251 : Yes), the trend determination unit 218 determines that the detection value i will not change significantly in the long term (step S252 ). That is, the trend determination unit 218 determines that the detection value i has not detected an increasing trend or a decreasing trend.

[0115] On the other hand, when the third Mahalanobis distance exceeds the threshold value (step S251 : No) and the added value is a positive number, the trend determination unit 218 determines that the detection value i tends to increase in the long term (step S253 ).

[0116] Then, when the third Mahalanobis distance exceeds the threshold value (step S251 : No) and the added value is a negative number, the trend determination unit 218 determines that the detection value i tends to decrease in the long term (step S254 ).

[0117] Furthermore, when the third Mahalanobis distance exceeds the threshold value (step S251: No), the output unit 219 outputs a warning message including the determination result of the trend determination unit 218 (whether the detection value i tends to increase or decrease) and information that can identify the evaluation item (evaluation item name or identification number) (step S255). Figure 6 The same as step S154.

[0118] (Effect)

[0119] As described above, in the complete set equipment monitoring device 20 involved in this embodiment, when the third Mahalanobis distance of the sum of the second SN ratios of a certain evaluation item exceeds the prescribed threshold value, if the sum of the second SN ratios is a positive number, it is determined that the detection value of the evaluation item tends to increase, and if the sum of the second SN ratios is a negative number, it is determined that the detection value of the evaluation item tends to decrease.

[0120] In this way, the plant monitoring device 20 can automatically determine whether the long-term trend has changed based on the third Mahalanobis distance, thereby reducing the monitoring burden on the operator. In addition, the plant monitoring device 20 can use a general threshold value (for example, 4) of the Mahalanobis distance to determine whether the long-term trend has changed. As a result, the operator does not need to adjust the threshold value for each evaluation item, thereby further improving the versatility and generality of the plant monitoring device 20.

[0121] (Hardware structure)

[0122] Figure 8 This is a diagram showing an example of the hardware configuration of a plant monitoring device according to at least one embodiment of the present invention.

[0123] Below, reference Figure 8 The hardware configuration of the plant monitoring device 20 according to the present embodiment will be described.

[0124] The computer 900 includes a processor 901 , a main storage device 902 , an auxiliary storage device 903 , and an interface 904 .

[0125] The above-mentioned plant monitoring device 20 is installed in one or more computers 900. In addition, the operations of the above-mentioned functional units are stored in the auxiliary storage device 903 in the form of a program. The processor 901 reads the program from the auxiliary storage device 903 and expands it in the main storage device 902, and executes the above-mentioned processing according to the program. In addition, the processor 901 ensures the storage area corresponding to the above-mentioned storage units in the main storage device 902 according to the program. Examples of the processor 901 include a CPU (Central Processing Unit), a GPU (Graphic Processing Unit), a microprocessor, and the like.

[0126] The program may also be used to implement a part of the function of the computer 900. For example, the program may also perform the function by combining with other programs stored in the auxiliary storage device 903 or with other programs installed in other devices. In addition, in another embodiment, the computer 900 may also have a customized LSI (Large Scale Integrated Circuit) such as PLD (Programmable Logic Device) in addition to the above structure, or a customized LSI (Large Scale Integrated Circuit) such as PLD (Programmable Logic Device) may be used to replace the above structure. Examples of PLDs include PAL (Programmable Array Logic), GAL (Generic Array Logic), CPLD (Complex Programmable Logic Device), and FPGA (Field Programmable Gate Array). In this case, part or all of the functions implemented by the processor 901 may be implemented by the integrated circuit. Such an integrated circuit is also included in an example of a processor.

[0127] Examples of the auxiliary storage device 903 include a HDD (Hard Disk Drive), an SSD (Solid State Drive), a magnetic disk, an optical magnetic disk, a CD-ROM (Compact Disc Read Only Memory), a DVD-ROM (Digital Versatile Disc Read Only Memory), and a semiconductor memory. The auxiliary storage device 903 may be an internal medium directly connected to the bus of the computer 900, or an external storage device 910 connected to the computer 900 via an interface 904 or a communication line. Furthermore, when the program is distributed to the computer 900 via a communication line, the distributed computer 900 may expand the program in the main storage device 902 and execute the above-mentioned processing. In at least one embodiment, the auxiliary storage device 903 is a non-temporary tangible storage medium.

[0128] Furthermore, the program may be used to realize a part of the above functions. Furthermore, the program may be a so-called differential file (differential program) that realizes the above functions by combining with other programs stored in the auxiliary storage device 903 .

[0129] As mentioned above, although embodiment of this invention is described in detail, it is not limited to this, unless it deviates from the technical idea of ​​this invention, and some design changes etc. are possible.

[0130] <First Modification>

[0131] (Processing flow of complete plant monitoring device)

[0132] Fig. 9 This is a flowchart showing an example of processing of the plant monitoring device according to the first modified example of the present invention.

[0133] In the first or second embodiment, the second Mahalanobis distance calculation unit 215 increases the detection value i by a predetermined amount to calculate the second Mahalanobis distance, but the present invention is not limited thereto. Figure 3In step S9, the detection value i is reduced by a prescribed amount to calculate the second Mahalanobis distance. In this case, the second Mahalanobis distance calculation unit 215 calculates the prescribed amount in the same manner as in the above-mentioned embodiments. Specifically, the second Mahalanobis distance calculation unit 215 extracts the detection values ​​of the same evaluation items as the detection value i from the unit space to calculate their standard deviations. Then, the second Mahalanobis distance calculation unit 215 sets a value of 1 / 10000 to 5 / 10 of the standard deviation (+0.0001σ~+0.5σ) as the prescribed amount subtracted from the detection value i. In addition, the prescribed amount is more preferably a value of 1 / 1000 to 1 / 10 of the standard deviation (+0.001σ~+0.1σ), and most preferably a value of 1 / 100 to 1 / 10 of the standard deviation (+0.01σ~+0.1σ).

[0134] Furthermore, the second SN ratio acquisition unit 216 according to this modification executes Fig. 9 Steps S20 to S21 are replaced Figure 3 Steps S10 to S12.

[0135] Specifically, if Fig. 9 As shown in FIG. 1 , when the second Mahalanobis distance is greater than or equal to the first Mahalanobis distance (step S20: Yes), the second SN ratio acquisition unit 216 determines that the detection value i tends to be "lower" than the other detection values. Therefore, the second SN ratio acquisition unit 216 acquires a value obtained by converting the first SN ratio of the detection value i into a "negative number" as the second SN ratio of the detection value i (step S21). And, the second SN ratio is stored and accumulated in the storage unit 25.

[0136] On the other hand, when the second Mahalanobis distance is smaller than the first Mahalanobis distance (step S20: No), the second SN ratio acquisition unit 216 determines that the detection value i tends to be "higher" than the other detection values. Therefore, the second SN ratio acquisition unit 216 acquires a value obtained by converting the first SN ratio of the detection value i into a "positive number" as the second SN ratio of the detection value i (step S22). And, the second SN ratio is stored and accumulated in the storage unit 25.

[0137] The subsequent processing is the same as that in the first or second embodiment.

[0138] (Effect)

[0139] As described above, the plant monitoring device 20 according to the present modification calculates the second Mahalanobis distance by reducing each detection value by a predetermined amount, and converts the first SN ratio into a negative number or a positive number to obtain the second SN ratio depending on whether the second Mahalanobis distance is greater than the first Mahalanobis distance.

[0140] In this way, the plant monitoring device 20 can determine whether a certain detection value tends to be higher than other detection values ​​based on whether the second Mahalanobis distance decreases, and reflect this in the second SN ratio.

[0141] <Second Modification>

[0142] (Processing flow of complete plant monitoring device)

[0143] Fig.10 This is a flowchart showing an example of processing of the plant monitoring device according to the second modified example of the present invention.

[0144] In the second embodiment described above, the trend determination unit 218 calculates the third Mahalanobis distance by treating the added value of the second SN ratio as one detection value, but the present invention is not limited to this. For example, the trend determination unit 218 involved in the second modification may also perform Fig.10 Instead of Figure 3 The processing of steps S14 and S15.

[0145] Get a bunch of detected values ​​every specified time (e.g., 1 minute) and perform Figure 3 The second SN ratios calculated by the processing of steps S2 to S12 each contain an error. As a result, the added value of the second SN ratios also contains an error. When it is assumed that the error δsn of each second SN ratio is the same, if the number of data n added by the adder 217 (the number of second SN ratios acquired so far for the detection value i) increases, the error of the added value of the second SN ratio will become "δsn×√n", and increase to √n times. That is, if the number of data n added is 100, the error of the added value of the second SN ratio increases to 10 times, and if it is 10,000, it increases to 100 times. Therefore, if the second SN ratios are calculated and added every specified time (for example, 1 minute) for a long period of time (for example, 6 months), the added value of the second SN ratio will contain a large error that cannot be ignored. In addition, the increase in the error of the added value is well known as the addition of the error propagation law, so the description is omitted.

[0146] In this way, based on the second SN ratio including the error, the plant monitoring device 20 according to the present modification example performs Fig.10 Step S240 is replaced by Figure 3 Step S14. Here, the trend determination unit 218 calculates the corrected added value obtained by dividing the added value of the second SN ratio of the data number n by √n for error correction. Then, the output unit 219 generates a chart ( Figure 5 ), and displays it on the display unit 23 as the operation status information (step S240).

[0147] Next, the trend determination unit 218 regards the corrected sum of the second SN ratio as one detection value and calculates the third Mahalanobis distance (third MD) (step S250). Specifically, similarly to the second embodiment, the trend determination unit 218 generates a unit space of the corrected sum of the second SN ratio calculated from the present time to the elapse of the predetermined time at each predetermined time. The trend determination unit 218 calculates the third Mahalanobis distance (third MD) based on the unit space of the corrected sum of the second SN ratio. Fig.10 The trend determination unit 218 may generate a unit space including a plurality of detection value bundles collected from the present time to the predetermined time in addition to the corrected added value of the second SN ratio, and calculate the third Mahalanobis distance of the corrected added value of the second SN ratio calculated in step S240. Figure 3 The multiple detection values ​​(detection value 1 to detection value k) obtained in S1 and Fig.10 The third Mahalanobis distance of the detection value bundle consisting of the added value (detection value k+1) calculated in step S240.

[0148] The subsequent processing of steps S251 to S255 is the same as that of the second embodiment.

[0149] (Effect)

[0150] As described above, the plant monitoring device 20 according to the present modification calculates the third Mahalanobis distance of the corrected added value obtained by performing error correction on the added value of the second SN ratio.

[0151] In this way, the plant monitoring device 20 can suppress the influence of the error and calculate the third Mahalanobis distance more accurately. Therefore, the plant monitoring device 20 can more accurately determine whether the long-term trend of the plant 1 has changed.

[0152] <Note>

[0153] The plant monitoring device, the plant monitoring method, and the program described in the above-mentioned embodiments can be understood, for example, as follows.

[0154] According to a first aspect of the present invention, a complete plant monitoring device comprises: a detection value acquisition unit that acquires a set of detection values ​​for each of a plurality of evaluation items, namely, a detection value bundle; a first Mahalanobis distance calculation unit that calculates the first Mahalanobis distance of the detection value bundle based on a unit space created by a past detection value bundle; a first SN ratio calculation unit that calculates the first SN ratio of each of a plurality of the evaluation items; a second Mahalanobis distance calculation unit that increases or decreases each value of a plurality of the detection values ​​to calculate a second Mahalanobis distance corresponding to the increased or decreased detection values, respectively; a second SN ratio acquisition unit that converts the first SN ratio of each of the evaluation items into a second SN ratio based on the first Mahalanobis distance and the second Mahalanobis distance, and acquires the second SN ratio; and an addition unit that calculates the sum of a plurality of the second SN ratios acquired within a prescribed period for each of the plurality of the evaluation items.

[0155] Thus, the plant monitoring device can allow the operator to easily grasp the long-term trend of each evaluation item by the added value of the second SN ratio. Furthermore, the plant monitoring device can express the long-term trend of each evaluation item by the added value of the second SN ratio regardless of the specifications of the plant, the operating conditions, the sensor type, the content of the evaluation item, etc., thereby improving the versatility and generality.

[0156] According to a second aspect of the present invention, the plant monitoring device according to the first aspect further includes a plant state determination unit for determining whether the plant operation state is normal or abnormal based on whether the first Mahalanobis distance is equal to or smaller than a predetermined threshold.

[0157] Thus, the plant monitoring device can simultaneously monitor both the presence or absence of abnormality in the operating state of the plant and the long-term trend of the plant.

[0158] According to a third aspect of the present invention, the plant monitoring device according to the first or second aspect further comprises a trend determination unit for determining whether the detection value of each of the plurality of evaluation items tends to increase or decrease based on the added value of the second SN ratio.

[0159] In this way, the plant monitoring device can automatically diagnose the long-term trend of the plant without the need for the operator to monitor the added value of the second SN ratio in sequence.

[0160] According to the fourth aspect of the present invention, in the complete set equipment monitoring device involved in the third aspect, when the added value of the second SN ratio of the evaluation item exceeds a prescribed range, if the added value of the second SN ratio is a positive number, it is determined that the detection value of the evaluation item tends to increase, and if the added value of the second SN ratio is a negative number, it is determined that the detection value of the evaluation item tends to decrease.

[0161] In this way, the plant monitoring device can automatically determine the long-term trend of the detection value of each evaluation item, thereby reducing the monitoring burden on the operator.

[0162] According to the fifth embodiment of the present invention, in the complete plant monitoring device involved in the third embodiment, the trend determination unit calculates the third Mahalanobis distance of the sum of the second SN ratios of the evaluation item, and when the third Mahalanobis distance exceeds a prescribed threshold value, if the sum of the second SN ratios is a positive number, it is determined that the detection value of the evaluation item tends to increase, and if the sum of the second SN ratios is a negative number, it is determined that the detection value of the evaluation item tends to decrease.

[0163] In this way, the plant monitoring device can automatically determine whether the long-term trend has changed based on the third Mahalanobis distance, thereby reducing the monitoring burden on the staff. In addition, the plant monitoring device can use the general threshold of the Mahalanobis distance to determine whether the long-term trend has changed. As a result, the staff does not need to adjust the threshold for each evaluation item, thereby further improving the versatility and generality of the plant monitoring device.

[0164] According to the sixth aspect of the present invention, in the complete plant monitoring device involved in the fifth aspect, when the number of the second SN ratio obtained by adding the addition unit is set to n, the trend determination unit calculates the third Mahalanobis distance of the corrected addition value obtained by dividing the addition value of the second SN ratio by √n for error correction.

[0165] In this way, the plant monitoring device can suppress the influence of the error of the second SN ratio and calculate the third Mahalanobis distance more accurately. Therefore, the plant monitoring device can more accurately determine whether the long-term trend of the plant has changed.

[0166] According to the 7th mode of the present invention, in the complete plant monitoring device involved in any one of the 1st to 6th modes, the second Mahalanobis distance calculation unit increases each value of the plurality of detection values ​​by a prescribed amount to calculate the second Mahalanobis distance, and the second SN ratio acquisition unit converts the first SN ratio into the second SN ratio in a manner that makes the second SN ratio a positive number when the second Mahalanobis distance is greater than the first Mahalanobis distance, and converts the first SN ratio into the second SN ratio in a manner that makes the second SN ratio a negative number when the second Mahalanobis distance is less than the first Mahalanobis distance.

[0167] In this way, the plant monitoring device can determine whether a certain detection value tends to be higher than other detection values ​​based on whether the second Mahalanobis distance increases, and reflect this in the second SN ratio.

[0168] According to the 8th mode of the present invention, in the complete plant monitoring device involved in any one of the 1st to 6th modes, the second Mahalanobis distance calculation unit reduces each value of the plurality of detection values ​​by a prescribed amount to calculate the second Mahalanobis distance, and the second SN ratio acquisition unit converts the first SN ratio into the second SN ratio in a manner that makes the second SN ratio a negative number when the second Mahalanobis distance is greater than the first Mahalanobis distance, and converts the first SN ratio into the second SN ratio in a manner that makes the second SN ratio a positive number when the second Mahalanobis distance is less than the first Mahalanobis distance.

[0169] In this way, the plant monitoring device can determine whether a certain detection value tends to be higher than other detection values ​​based on whether the second Mahalanobis distance decreases, and reflect this in the second SN ratio.

[0170] According to a ninth aspect of the present invention, in the plant monitoring device according to any one of the first to eighth aspects, when the first SN ratio is a negative number, the second SN ratio acquisition unit sets the second SN ratio to zero.

[0171] When the first SN ratio of a certain evaluation item is negative, it can be determined that the detection value of the evaluation item has a negligible effect on the increase of the first Mahalanobis distance. Therefore, the plant monitoring device can simplify the processing by considering the second SN ratio of such an evaluation item with a small effect as zero.

[0172] According to the 10th aspect of the present invention, the complete set equipment monitoring method includes: a step of obtaining a set of detection values, i.e., a detection value bundle, for each of a plurality of evaluation items; a step of calculating the first Mahalanobis distance of the detection value bundle based on a unit space created by a past detection value bundle; a step of calculating the first SN ratio of each of the plurality of evaluation items; a step of increasing or decreasing each value of the plurality of detection values ​​to calculate the second Mahalanobis distance corresponding to the increased or decreased detection values, respectively; a step of converting the first SN ratio of each of the evaluation items into a second SN ratio based on the first Mahalanobis distance and the second Mahalanobis distance, and obtaining the second SN ratio; and a step of calculating the sum of the plurality of second SN ratios obtained within a specified period for each of the plurality of evaluation items.

[0173] According to the 11th mode of the present invention, the program causes the computer of the complete set equipment monitoring device to execute the following steps: a step of obtaining a set of detection values, i.e., a detection value bundle, for each of a plurality of evaluation items; a step of calculating the first Mahalanobis distance of the detection value bundle based on a unit space created by a past detection value bundle; a step of calculating the first SN ratio of each of the plurality of evaluation items; a step of increasing or decreasing each value of the plurality of detection values ​​to calculate the second Mahalanobis distance corresponding to the increased or decreased detection values, respectively; a step of converting the first SN ratio of each of the evaluation items into a second SN ratio based on the first Mahalanobis distance and the second Mahalanobis distance and obtaining the second SN ratio; and a step of calculating the sum of the plurality of second SN ratios obtained within a specified period for each of the plurality of evaluation items.

[0174] Industrial Applicability

[0175] According to any of the above methods, the versatility and generality are improved while long-term trends can be easily grasped.

[0176] Explanation of symbols

[0177] 1-complete equipment, 20-complete equipment monitoring device, 21-CPU, 211-detection value acquisition unit, 212-first Mahalanobis distance calculation unit, 213-complete equipment state determination unit, 214-first SN ratio calculation unit, 215-second Mahalanobis distance calculation unit, 216-second SN ratio acquisition unit, 217-addition unit, 218-trend determination unit, 219-output unit, 22-input and output interface, 23-display unit, 24-operation acceptance unit, 25-storage unit, 40-control device, 900-computer.

Claims

1. A complete plant monitoring device, comprising: a detection value acquisition unit that acquires a detection value bundle that is a collection of detection values ​​for each of a plurality of evaluation items; a first Mahalanobis distance calculation unit that calculates a first Mahalanobis distance of the detection value bundle based on a unit space created by a past detection value bundle; a first SN ratio calculation unit that calculates a first SN ratio for each of the plurality of evaluation items; a second Mahalanobis distance calculation unit that increases or decreases each of the plurality of detection values ​​to calculate a second Mahalanobis distance corresponding to the increased or decreased detection value; a second SN ratio acquisition unit that converts the first SN ratio of each of the evaluation items into a second SN ratio based on the first Mahalanobis distance and the second Mahalanobis distance, and acquires the second SN ratio; an adding unit for calculating, for each of the plurality of evaluation items, an added value of the plurality of second SN ratios acquired within a predetermined period; and A display unit displays the change of the added value. In the case of increasing each of the plurality of detection values: The second Mahalanobis distance calculation unit calculates the second Mahalanobis distance by increasing each of the plurality of detection values ​​by a predetermined amount. The second SN ratio acquisition unit converts the first SN ratio into the second SN ratio in a manner that makes the second SN ratio a positive number when the second Mahalanobis distance is greater than the first Mahalanobis distance, and converts the first SN ratio into the second SN ratio in a manner that makes the second SN ratio a negative number when the second Mahalanobis distance is less than the first Mahalanobis distance. In the case of reducing each of the plurality of detection values: The second Mahalanobis distance calculation unit calculates the second Mahalanobis distance by reducing each of the plurality of detection values ​​by a predetermined amount. The second SN ratio acquisition unit converts the first SN ratio into the second SN ratio in a manner that makes the second SN ratio a negative number when the second Mahalanobis distance is greater than the first Mahalanobis distance, and converts the first SN ratio into the second SN ratio in a manner that makes the second SN ratio a positive number when the second Mahalanobis distance is less than the first Mahalanobis distance.

2. The plant monitoring device according to claim 1, further comprising a plant state determination unit, The plant state determination unit determines whether the plant operation state is normal or abnormal based on whether the first Mahalanobis distance is equal to or smaller than a predetermined threshold.

3. The plant monitoring device according to claim 1 or 2, further comprising a trend determination unit, The trend determination unit determines whether the detection value of each of the plurality of evaluation items tends to increase or decrease based on the added value of the second SN ratio.

4. The plant monitoring device according to claim 3, in, When the added value of the second SN ratio of the evaluation item exceeds a predetermined range, the trend determination unit If the added value of the second SN ratio is a positive number, it is determined that the detection value of the evaluation item tends to increase. If the added value of the second SN ratio is a negative number, it is determined that the detection value of the evaluation item tends to decrease.

5. The plant monitoring device according to claim 3, in, The trend determination unit calculates a third Mahalanobis distance of the added value of the second SN ratio of the evaluation item, and when the third Mahalanobis distance exceeds a predetermined threshold value, If the added value of the second SN ratio is a positive number, it is determined that the detection value of the evaluation item tends to increase. If the added value of the second SN ratio is a negative number, it is determined that the detection value of the evaluation item tends to decrease.

6. The plant monitoring device according to claim 5, in, When the number of the second SN ratios added by the adding unit is n, the trend determining unit calculates the third Mahalanobis distance of a corrected added value obtained by performing error correction by dividing the added value of the second SN ratio by √n.

7. The plant monitoring device according to claim 1 or 2, in, When the first SN ratio is a negative number, the second SN ratio acquisition unit sets the second SN ratio to zero.

8. A complete set of equipment monitoring method, wherein include: A step of obtaining a set of detection values ​​for each of a plurality of evaluation items, namely, a detection value bundle; a step of calculating a first Mahalanobis distance of the detection value bundle based on a unit space created by a past detection value bundle; a step of calculating a first SN ratio for each of the plurality of evaluation items; The step of increasing or decreasing each of the plurality of detection values ​​to calculate a second Mahalanobis distance corresponding to the increased or decreased detection values, respectively; A step of converting the first SN ratio of each evaluation item into a second SN ratio according to the first Mahalanobis distance and the second Mahalanobis distance, and obtaining the second SN ratio; a step of calculating, for each of the plurality of evaluation items, an added value of the plurality of second SN ratios acquired within a predetermined period; and a step of displaying the change of the added value, In the case of increasing each of the plurality of detection values: In the step of calculating the second Mahalanobis distance, each of the plurality of detection values ​​is increased by a predetermined amount to calculate the second Mahalanobis distance. In the step of obtaining the second SN ratio, when the second Mahalanobis distance is greater than the first Mahalanobis distance, the first SN ratio is converted into the second SN ratio in such a way that the second SN ratio becomes a positive number, and when the second Mahalanobis distance is less than the first Mahalanobis distance, the first SN ratio is converted into the second SN ratio in such a way that the second SN ratio becomes a negative number, In the case of reducing each of the plurality of detection values: In the step of calculating the second Mahalanobis distance, each of the plurality of detection values ​​is reduced by a predetermined amount to calculate the second Mahalanobis distance. In the step of obtaining the second SN ratio, when the second Mahalanobis distance is greater than the first Mahalanobis distance, the first SN ratio is converted into the second SN ratio in a way that makes the second SN ratio a negative number, and when the second Mahalanobis distance is less than the first Mahalanobis distance, the first SN ratio is converted into the second SN ratio in a way that makes the second SN ratio a positive number.

9. A program product, comprising a program, wherein the program causes a computer of a plant monitoring device to execute the following steps: A step of obtaining a set of detection values ​​for each of a plurality of evaluation items, namely, a detection value bundle; a step of calculating a first Mahalanobis distance of the detection value bundle based on a unit space created by a past detection value bundle; a step of calculating a first SN ratio for each of the plurality of evaluation items; The step of increasing or decreasing each of the plurality of detection values ​​to calculate a second Mahalanobis distance corresponding to the increased or decreased detection values; A step of converting the first SN ratio of each evaluation item into a second SN ratio according to the first Mahalanobis distance and the second Mahalanobis distance, and obtaining the second SN ratio; a step of calculating, for each of the plurality of evaluation items, an added value of the plurality of second SN ratios acquired within a predetermined period; and a step of displaying the change of the added value, When each of the plurality of detection values ​​is increased, causing the computer to: In the step of calculating the second Mahalanobis distance, each of the plurality of detection values ​​is increased by a predetermined amount to calculate the second Mahalanobis distance. In the step of obtaining the second SN ratio, when the second Mahalanobis distance is greater than the first Mahalanobis distance, the first SN ratio is converted into the second SN ratio in such a way that the second SN ratio becomes a positive number, and when the second Mahalanobis distance is less than the first Mahalanobis distance, the first SN ratio is converted into the second SN ratio in such a way that the second SN ratio becomes a negative number, When each of the plurality of detection values ​​is reduced, causing the computer to: In the step of calculating the second Mahalanobis distance, each of the plurality of detection values ​​is reduced by a predetermined amount to calculate the second Mahalanobis distance. In the step of obtaining the second SN ratio, when the second Mahalanobis distance is greater than the first Mahalanobis distance, the first SN ratio is converted into the second SN ratio in a way that makes the second SN ratio a negative number, and when the second Mahalanobis distance is less than the first Mahalanobis distance, the first SN ratio is converted into the second SN ratio in a way that makes the second SN ratio a positive number.

Images