JP2022173848A - Abnormality diagnosis system, abnormality diagnosis method and abnormality diagnosis program - Google Patents

Abstract

To provide an abnormality diagnosis system which expresses a propagation relation of affections between respective parts and easily discriminates a specific part which causes a different state from a normal operation even if there are a large number of facility machines.SOLUTION: An abnormality diagnosis system of a facility to form a travel route of matters by plural parts is constituted of plural states change detection means which are set per parts of the facility and detect state change of the matter, and possible cause estimation means that estimates the parts to be causes in a case the states change detection means detect a change in state. The state change detection means store relation of plural operation data related to parts of objects detected by the detection unit during normal operation, and detect state changes from the fact that the relations of operation data change from the normal operation. The cause estimation means gives points to detection units of each part detecting the state change using plural criteria for evaluation determined in relation to the matter moving between plural parts, and estimates the parts to be the causes of the state change of facility according to the total of points given by plural evaluation criteria.SELECTED DRAWING: Figure 1

Description

The present invention relates to an abnormality diagnosis system, an abnormality diagnosis method, and an abnormality diagnosis program for detecting an abnormality during operation of a device, facility, plant, etc., and notifying/transmitting it to a driver or an operation system.

In a facility or plant that consists of multiple pieces of equipment, there is a method of acquiring operating data from the measured values measured by the sensors installed in each piece of equipment, and then executing an abnormality diagnosis for the facility or plant based on that operating data. Are known.

For example, in Patent Document 1, when operating data is classified into a plurality of categories and recorded, and operating data in a new category different from the data stored as operating data during normal operation is acquired, the previously stored operating data is stored. An abnormality diagnosis method for determining that the operating state is different from the operating state is described. Also, based on this method, a method of discriminating an abnormal point in equipment by considering the direction of material flow is also described.

Patent No. 4430384

However, in the technique disclosed in Patent Document 1, the devices in the facility or plant are diagnosed as a single unit, so when there are many target devices or when the abnormality of one device affects a wide range However, it was difficult to identify the causative equipment.

It is an object of the present invention to express the influence propagation relationship between devices even when there are many devices in a facility or plant, and to easily identify the device (part) that causes an operating state different from normal operation. An object of the present invention is to provide an abnormality diagnosis system, an abnormality diagnosis method, and an abnormality diagnosis program.

In order to solve the above problems, the present invention provides a system for diagnosing an abnormality of a facility that forms a movement path of a substance with a plurality of parts, wherein a plurality of states are set for each part of the facility and detect changes in the state of the substance. and a cause estimating means for estimating a part causing the change when the state change detecting means detects a change in the state, and the state change detecting means detects the object detected by the detection unit. The cause estimating means stores the relationship during normal operation of a plurality of operation data relating to each part, detects a state change from the change in the relationship between the operation data and the normal state, and the cause estimation means relates to the substance moving in the plurality of parts Using the multiple evaluation criteria specified in the above, a score is given to the detection unit of each part that detected the state change, and the cause of the equipment state change is determined according to the total score given by the multiple evaluation standards. Abnormality diagnosis system characterized by estimating the part where

In addition, the present invention provides a method for diagnosing an abnormality of equipment in which a plurality of parts form a movement path of a substance, wherein a plurality of state change detection means are set for each part of the facility and detect a change in the state of the substance. The unit stores the relationship during normal operation of a plurality of operation data relating to the target part detected by the unit, and detects the state change when the relationship of the operation data changes from the normal state, and the state change detection means detects the state. When a change is detected, a detection unit for each part that has detected a state change using a plurality of evaluation criteria determined in relation to a substance that moves in a plurality of parts in order to estimate the part that causes it. and estimating the part that causes the state change of the equipment according to the total score given by multiple evaluation criteria.

In addition, the present invention provides an "abnormality diagnosis program for diagnosing an abnormality in a facility that forms a movement path of a substance by using a computer, which is set for each part of the facility and detects a change in the state of the substance. By means of a plurality of state change detection means, the relationship during normal operation of a plurality of operation data relating to the target part detected by the detection unit is stored, and the state change is detected from the fact that the relationship of the operation data has changed from the normal state. and a plurality of evaluation criteria determined in relation to a substance moving through a plurality of parts in order to estimate a part causing the change when the state change detection means detects a change in the state is used to give a score to the detection unit of each part that has detected a state change, and the part that causes the state change of the equipment is estimated according to the total score given by the multiple evaluation criteria. and a third program for adjusting the allocation of scores in a plurality of evaluation criteria."

According to the present invention, even if there are many devices in a facility or plant, it is possible to express the influence propagation relationship between the devices, and to easily determine the device (part) that caused the operating state different from normal operation. .

1 is a block diagram showing a configuration example of an abnormality diagnosis system 1 according to an embodiment of the present invention; FIG. FIG. 2 is a diagram showing a configuration example of plant equipment to be monitored by the abnormality diagnosis system 1. FIG. The figure which showed the change of the operation data at the time of detecting a state change in detection unit A2. It is the figure which showed an example of the relationship information between detection units. It is the figure which showed an example of the display means. FIG. 4 is a diagram showing changes in operating data when state changes are detected in detection units A2, A4, and A5. FIG. 10 is a diagram showing a display example of a control end and an operation end of control logic; It is the figure which showed the cause estimation result on the display means. It is the figure which showed the cause estimation result on the display means by another method. The figure which shows the processing flow of the abnormality-diagnosis system 1 which concerns on the Example of this invention.

BEST MODE FOR CARRYING OUT THE INVENTION Hereinafter, embodiments for carrying out the present invention will be described in detail with reference to the drawings.

FIG. 1 is a block diagram showing an example configuration of an abnormality diagnosis system 1 according to an embodiment of the present invention. An abnormality diagnosis system 1 as an example is connected to a measurement/control device 3, an external output device 4, and an external input device 5 of a plant 2 so as to be capable of two-way communication. The abnormality diagnosis system 1 configured by a computer includes an operation data database 110 , state change detection means 120 , detection unit relationship information database 130 , cause estimation means 140 , parameter adjustment means 150 and display means 160 .

Of these, the operation data database 110 is connected to the measurement/control device 3 of the plant 2, and stores measurement data such as plant temperature, pressure, and flow rate, manipulated variable data such as valve opening, and control set value data. stored in chronological order. Below, measurement data, operation amount data, and set value data are collectively referred to as operation data.

The state change detection means 120 detects a state change of each device by means of a detection unit set for each plant device (may be a part; hereinafter simply referred to as device). Here, the detection unit uses a plurality of pieces of operation data relating to target equipment and detects state changes from the relationship between these pieces of operation data.

The inter-detection unit relationship information database 130 stores the effect propagation relationship between plant equipment (parts) B corresponding to the detection unit A. FIG. For example, when the detection unit A1 and the detection unit A2 respectively detect changes in the state of the device B1 and the device B2 in the plant, if the device B1 is upstream of the device B2, the effect of the device B1 is the device B2. up to In this case, information indicating that the detection unit A1 affects the detection unit A2 is stored in the inter-detection unit relationship information.

The cause estimating means 140 estimates the device B that caused the state change by using the information that the detecting unit A of the state change detecting means 120 has detected the state change and the inter-detecting unit relationship information database 130 .

The parameter adjusting means 150 adjusts parameters used when the cause estimating means 140 estimates the device B that caused the state change.

The display means 160 displays the cause device B estimated by the cause estimation means 140 on the screen.

Next, the operation of the abnormality diagnosis system 1 will be described in detail. As a premise, the plant configuration example of FIG. 2 will be described as a reference example. The plant in FIG. 2 shows a substance transfer route by a plurality of devices B (from device B1 to device B5 in this example), and the substance from device B1 flows into device B2 mainly composed of reactor V1. , from the device B2 to the device B3 and the device B4 mainly composed of the reactor V2, and from the device B4 to the device B5. Appropriate control valves V- (VF4, VP5, VF7) are disposed at various points in these paths to appropriately control substances flowing in and out.

In the operation data database 110 of FIG. 1, the operation data of each part of the plant of FIG. 2 is detected by each sensor and recorded. A specific example of the operating data here is, for example, in the case of equipment B2 of the plant shown in FIG. It becomes the driving data that can be used. Similarly, the operating data used for the detection unit A4 corresponding to the equipment B4 are the flow rate F4, the pressure P6, the temperatures T4 and T6, and the level L6.

Returning to FIG. 1, the abnormality diagnosis system 1 has a learning phase and a diagnosis phase. The learning phase is a preparatory phase before operating the abnormality diagnosis system 1, and only the state change detection means 120 operates. In the diagnosis phase, the equipment B of the plant is diagnosed, and when the state change detecting means 120 detects a state change of the equipment, the cause estimating means 140 operates. Also, the diagnostic status in the diagnostic phase is displayed on the display means 160 .

First, the learning phase will be explained. In the learning phase, using the operating data stored in the operating data database 110 when the device B is in a normal state, the detection unit A of the state change detection means 120 detects the relationship between the operating data of each part in a normal operating state. Learning is performed using, for example, Adaptive Resonance Theory (hereinafter referred to as ART), which is one of data clustering techniques.

Specifically, a plurality of operational data corresponding to each detection unit are input to ART as multidimensional data. The input here is the operation data illustrated in FIG.

Inputted driving data undergoes data preprocessing such as normalization processing and addition of complements, and is classified into a plurality of categories according to the similarity of the data. Although the number of categories to be classified differs depending on the resolution parameter that determines the level of detail of classification, the number of operating data, data variations, etc., the classified categories are defined as categories representing normal conditions.

For example, if the operating data obtained when the plant is in a normal state are divided into 10 categories from category 1 to category 10, categories 1 to 10 are normal state categories.

Next, the diagnostic phase will be described. In the diagnosis phase, operation data (diagnosis data) to be diagnosed is input to the ART that has learned normal operation data. As a result, data with a high degree of similarity to the learning data are classified into the same category as the learning phase (one of 10 categories from 1 to 10). However, if some kind of abnormality occurs in the equipment and the tendency of the data changes, the data is classified into a new category (eg, 11 as a new category) different from the learning data. In this way, the detection unit can detect changes in the state of equipment from categories classified by ART.

Next, the case where a new category occurs in the diagnosis phase will be described in detail. Here, a case where a new category occurs in the detection unit A2 and a state change is detected will be described. In the case of the plant configuration of FIG. 2, the detection unit A (A1 to A5) is provided for each device B (B1 to B5), and the detection unit A2 is mainly composed of the reactor V1 of FIG. Operation data (flow rate F1, pressure P2, temperature T1, T2, T3, level L2) from sensors related to device B2 are detected, and the abnormality is detected by clustering.

FIG. 3 shows changes in the values of each measurement item of the detection unit A2 at that time. The horizontal axis indicates items of operation data (flow rate F1, pressure P2, temperature T1, T2, T3, level L2), and the vertical axis indicates normalized measurement values. The dashed line shows the characteristics of data in the same category as during normal operation (data immediately before detecting a state change), and the solid line shows the characteristics of data in the new category. The pressure P2 and the temperature T3 measured in the detection unit A2 are different from those during normal operation, which is the reason for the new category.

It should be noted that "E" shown below the operation data in FIG. For example, in the example of the detection unit A2 in FIG. 2, the flow rate F1 and the temperature T1 are data dependent on the device B1 on the upstream side of the facility, and change regardless of the state of the device B2. On the other hand, the temperature T2, the pressure P2, the temperature T3, and the level L2 are operating data that are also affected by the state of the reactor V1, although they also change depending on the equipment B1 on the upstream side. In this way, the operation data to be targeted by all the detection units can be distinguished depending on whether or not they serve as input conditions.

That is, the detection unit A can detect whether or not the state of the equipment B of the plant has changed, and at the same time, can also detect which operation data has changed from the characteristics of the classified data. It can also be determined whether or not the changed operating data is the input condition.

Next, the operation of the cause estimating means 140 in FIG. 1 will be described. In the cause estimating means 140, the time when each detection unit A detected the state change in the state change detecting means 120, the operation data that caused the state change, and the parts of the equipment stored in the inter-detection unit relationship information database 130 The site that caused the abnormality is estimated from the propagation relationship of the influence of

A specific example of information stored in the inter-detection unit relationship information database 130 is shown in FIG. FIG. 4 shows the relationship of the detection units A1 to A5 corresponding to the devices B1 to B5 shown in FIG. The arrows between each detection unit represent the influence propagation relationship. This is the propagation relationship determined from the movement of matter in the corresponding sites. For example, focusing on the equipment B2 of the plant in FIG. 2, the material flowing in from the equipment B1 on the upstream side moves to the equipments B3 and B4 via the reactor V1. Therefore, in the detection unit A as well, the influence propagates from A1 to A2 and from A2 to A3 and A4 as shown in FIG. In this way, the influence propagation relationship between parts of the facility stored in the inter-detection unit relationship information database 130 represents the relationship between detection units that collectively handle a plurality of operation data. can be constructed much more easily than the method of expressing the relationship of

The cause estimating means 140 comprehensively determines the cause of the abnormality based on the propagation relationship of these effects, the time when each detection unit A detected the state change in the state change detecting means 120, and the operation data that caused the state change. Estimate the site where the

FIG. 10 is a flow chart showing the processing contents of the cause estimating means 140 and the parameter adjusting means 150. As shown in FIG. In FIG. 10, processing step S10 is a process performed by the parameter adjusting means 150 in the preparation stage before the cause estimating means 140 is operated, and the parameters for the criteria used by the cause estimating means 140 are adjusted. Processing step S10, which is the parameter adjusting means 150, will be described later in detail.

In processing step S11, data is input and taken in from each detection unit A (A1 to A5), and in processing step S12, points are given to detection unit A (A1 to A5) according to the concept of the evaluation criteria shown below. .

In this embodiment, the detection unit A that detected the state change was scored according to the following four evaluation criteria, and the part with the highest total score was estimated to be the most likely cause of the abnormality. These criteria are criteria established in relation to substances that migrate across multiple sites.

Criterion 1: A higher score is given to a detection unit whose influence propagation relationship is on the upstream side.

Criterion 2: A higher score is given to a detection unit whose detection time is earlier.

Criterion 3: Give a score to the detection unit whose operating data that caused the state change did not include the input condition.

Criterion 4: If the operating data that caused the state change is the control end of the control logic, give a score to the detection unit including the control end.

The validity judgment of these criteria is as follows. First, Criterion 1 is based on the fact that the more upstream the influence propagation relationship, the higher the possibility of being the cause of the abnormality. Criterion 2 is because the earlier the state change is detected, the higher the possibility of the cause. This is because Criterion 3 indicates that the state has changed even though the input conditions in the operation data have not changed, and this is highly likely to be the root cause. Criterion 4 is because there is a high possibility that the root cause is on the control end side of the control logic. This is because the operation data related to the control logic fluctuates at the operating end but does not fluctuate at the control end.

A specific example of Criterion 4 will now be described with reference to FIG. The level L2 of reactor V1 in plant equipment (site) B2 is kept constant by valve V-F4 which regulates the flow rate F4 in equipment (site) B4. At this time, a control logic indicated by a dotted line exists between the level L2 and the flow rate F4, and the level L2 is called a control end, and the flow rate F4 is called an operation end.

In the presence of such control logic, the level L2 is kept constant while the flow rate F4 is fluctuating. Therefore, in Criterion 4, if the operating data that caused the state change is the operation end of the control logic, it is highly likely that the root cause is the control end of the control logic.

Next, specific points are given for each criterion as follows.

Criterion 1:
100 points are given to the detection unit with the most upstream propagation relationship. 50 points are given to the detection unit immediately downstream of the most upstream detection unit and the detection unit that detected the state change. 30 points are given to the detection unit on the second downstream side from the most upstream detection unit and the detection unit that has detected the state change. 20 points are given to the detection unit that is the third downstream detection unit from the most upstream detection unit and that has detected a state change. 10 points are given to the detection unit that is the fourth downstream detection unit from the most upstream detection unit and that has detected a state change. No score is given to detection units on the fifth and subsequent downstream sides from the most upstream detection unit.

Criterion 2:
50 points are given to the detection unit with the fastest detection time. Below, one point is deducted for each one minute delay in detection time. For example, a detection unit that detected a change of state 15 minutes after the earliest detection unit was given 35 points. A detection unit with a delay of 50 minutes or more in the detection time is not scored.

Criterion 3:
A score of 100 is given to a detection unit that does not include an input condition in the operating data that caused the state change.

Criterion 4:
If the operating data causing the state change is the control end of the control logic, give 50 points to the detection unit containing the control end of the corresponding control logic.

It should be noted that the scoring method of this embodiment is an example, and does not limit the present invention.

In processing step S13 of FIG. 10, the total score given to the detection unit based on each criterion is obtained, and the content of this is displayed on the display means 160. FIG. An example of the display means 160 and an example of actually estimating the cause based on these evaluation criteria are shown below. A display example of the display means 160 is shown in FIG. A diagram representing the inter-detection unit relationship information database 130 is shown on the display means 160 . FIG. 5 is a display in which the information of the time when the state change is detected is superimposed on FIG. 4 showing the relationship of the detection units A1 to A5 corresponding to the devices B1 to B5.

Detecting units A2, A4, and A5 colored in the background in FIG. 5 are detecting units that have detected a state change because it has been determined that a new category different from that during normal operation has occurred. The numbers on the upper right of the detection units A2, A4, and A5 represent the time, indicating that the state change was detected at that time. That is, the detection unit A2 detects the state change at 8:30, the detection unit A4 detects the state change at 8:20, and the detection unit A5 detects the state change at 8:40.

In the example of FIG. 5, FIG. 6 shows changes in the values of the measurement items of the detection units A2, A4, and A5. The horizontal axis is the item of the operating data, and the vertical axis is the normalized measured value. The dashed line shows the characteristics of data in the same category as during normal operation (data immediately before detecting a state change), and the solid line shows the characteristics of data in the new category.

"E" under the operation data in FIG. 6 indicates input conditions for the target equipment. "C" under the operation data in FIG. 6 indicates that the control logic exists and is the control terminal, and "O" indicates that the control logic exists and the operation terminal It shows that

Another representation of the control logic is shown in FIG. In FIG. 7, the control end and operation end of the control logic are summarized in a table. According to FIG. 7, control logic No. It can be seen that the control end of 1 is level L2 and the control end is flow rate F4. Also, the control logic No. The control end of 2 is level L6, and the control end is flow rate F7.

According to the series of examples in FIGS. 5, 6, and 7, in the detection unit A2, the pressure P2 and the temperature T3 are different from those during normal operation, which is the reason for the new category. . In the detection unit A4, the measured values of the flow rate F4 and the temperature T4 are different from those during normal operation, which is the reason why the new category is determined. Both the flow rate F4 and the temperature T4 that cause the state change are input conditions for the detection unit A4. Furthermore, the flow rate F4 that caused the state change is the operating end of the control logic. In the detection unit A5, the pressure P8 and the temperature T8 are different from those during normal operation, which is the reason for the new category.

Based on these conditions, if each detection unit is scored according to the evaluation criteria shown above, the results are as follows.

Criterion 1: Detection unit A1 to 0 points, detection unit A2 to 100 points, detection unit A3 to 0 points, detection unit A4 to 50 points, detection unit A5 to 30 points Criterion 2: detection unit A1 to 0 points, detection unit A2 ~40 points, detection unit A3 to 0 points, detection unit A4 to 50 points, detection unit A5 to 30 points Standard 3: detection unit A1 to 0 points, detection unit A2 to 100 points, detection unit A3 to 0 points, detection unit A4-0 points, detection unit A5-100 points Standard 4: detection unit A1-0 points, detection unit A2-50 points, detection unit A3-0 points, detection unit A4-0 points, detection unit A5-0 points Therefore, The total score is 1 to 0 points for the detection unit, 2 to 290 points for the detection unit, 3 to 0 points for the detection unit, 4 to 100 points for the detection unit, and 5 to 160 points for the detection unit. It can be estimated that the detection unit A2 has the highest score.

In this embodiment, when the "Cause Estimation" button shown in FIG. 5 is pressed, an estimation result based on the given score is displayed as shown in FIG. The "possibility" displayed in the rightmost column of FIG. 8 was evaluated from the ratio of the score of each detection unit to the total score of the detection units.

In FIG. 5, the screen when the cause estimation button is pressed may be indicated by color shades or hues as shown in FIG.

Next, the parameter adjusting means 150 will be described. The parameter adjusting means 150 adjusts the parameters (score distribution methods given by criteria 1 to 4) used when the cause estimating means 140 estimates the equipment that caused the state change. The processing of processing step S10 in FIG. 10 corresponds to this.

The adjustment of the parameters is based on the judgment that there is a difference between the result of estimating the equipment that caused the state change by the cause estimating means 140 and the equipment that caused the abnormality in the actual plant. be carried out at the given timing.

Specifically, the method of allocating scores given by Criteria 1 to 4 is optimized so that the total score of equipment that is the cause of an abnormality in an actual plant is the highest.

In Criterion 1, for example, the score allocation is set so as to be a linear function of the distance from the most upstream detection unit and the order of the units, and the intercept and slope of the linear function are optimized.

In Criterion 2, for example, score distribution is set so as to be a linear function of the elapsed time from the detection unit with the earliest detection time, and the intercept and slope of the linear function are optimized.

Criterion 3, for example, optimizes the score given to the sensing unit whose operating data that caused the state change did not include the input condition.

In Criterion 4, for example, if the operating data that caused the state change is the operation end of the control logic, the score to be given to the detection unit including the control end of the corresponding control logic is optimized.

By optimizing the score distribution method given by criteria 1 to 4 (adjusting the parameters), the device (part) that caused an abnormality in the actual plant and the cause estimation means 140 can determine the state change. It is possible to match the result of estimating the device (site) that caused the problem.

As described above, in this embodiment, it is possible to detect changes in the state of equipment (parts) in the plant, and to estimate the equipment (parts) that caused the change in conditions.

In this embodiment, the adaptive resonance theory is used as the clustering technique for the state change detection means 120, but other data clustering techniques such as vector quantization and k-means may be used.

Although the above description has been made mainly from the viewpoint of the system configuration, this can be realized as a method or a program. When it is implemented as a method, it is described as "a method for diagnosing an abnormality in a facility that forms a movement path of a substance with a plurality of parts, wherein a plurality of state change detection means are set for each part of the facility and detect a change in the state of the substance. a state change detecting means for detecting a state change based on a change in the relation of the operating data from a normal state, by storing the relationship during normal operation of a plurality of operation data relating to the target part detected by the detection unit; In order to estimate the site that is the cause of a change in the state when a sensor detects a change in state, each site that detects a change in state uses multiple evaluation criteria established in relation to substances that move through multiple sites. and assigning a score to each detection unit, and estimating the part that causes the state change of the equipment according to the total score given by multiple evaluation criteria. .

When it is realized as a program, it is defined as "an anomaly diagnosis program for diagnosing anomalies in a facility that forms a movement path for substances using a computer, and which is set for each part of the facility and By means of a plurality of state change detection means for detecting changes, the relationship between a plurality of operating data relating to the target part detected by the detection unit and during normal operation is stored, and the relationship of the operating data changes from the normal state. A first program for detecting a state change and a substance moving through a plurality of sites for estimating a site that causes the state change when the state change detection means detects the state change. Using multiple evaluation criteria, a score is given to the detection unit of each part that detected a state change, and the part that causes the state change of the facility is estimated according to the total score given by the multiple evaluation criteria and a third program for adjusting the distribution of scores in a plurality of evaluation criteria."

DESCRIPTION OF SYMBOLS 1 --- Abnormal diagnosis system 2 --- Plant 3 --- Measurement/control device 4 --- External output device 5 --- External input device 110 --- Operation data database 120 --- Status change detection means 130 --- Detection unit relation information database 140 --- Cause estimation means 150 --- Parameter adjustment means 160 ... display means

Claims (14)

An abnormality diagnosis system for equipment that forms a movement path of substances with a plurality of parts,
A plurality of state change detection means set for each part of the facility and detecting a change in the state of a substance; and a cause estimation means for estimating a part causing the change when the state change detection means detects a change in the state. consists of
The state change detection means stores the relationship during normal operation of a plurality of pieces of operation data relating to the target part detected by the detection unit, and detects the state change from the fact that the relationship of the operation data changes from the normal state. ,
The cause estimating means uses a plurality of evaluation criteria determined in relation to a substance moving in a plurality of parts to give a score to the detection unit of each part that has detected the state change, and according to the plurality of evaluation criteria An anomaly diagnosis system characterized by estimating a part that causes a change in the state of equipment according to the total score given. The abnormality diagnosis system according to claim 1,
An anomaly diagnosis system, wherein one of the plurality of evaluation criteria is such that a higher score is given to a detection unit located upstream in a propagation relationship of an influence of a substance moving through a plurality of parts. The abnormality diagnosis system according to claim 1,
One of the plurality of evaluation criteria is an anomaly diagnosis system characterized by giving a higher score to a detection unit having an earlier detection time for a substance moving through a plurality of parts. The abnormality diagnosis system according to claim 1,
An anomaly diagnosis system, wherein one of the plurality of evaluation criteria is to give a score to a detection unit whose operation data that caused a state change does not include an input condition for a substance moving through a plurality of parts. The abnormality diagnosis system according to claim 1,
One of the plurality of evaluation criteria is characterized by giving a score to a detection unit including a control end when the operating data causing the state change is the operation end of the control logic for a substance moving through a plurality of parts. anomaly diagnosis system. The abnormality diagnosis system according to any one of claims 1 to 5,
An anomaly diagnosis system characterized by displaying on a display means that there is a high possibility that a piece of equipment with a high total score is a part that causes a state change. The abnormality diagnosis system according to any one of claims 1 to 6,
An anomaly diagnosis system characterized by having parameter adjustment means for adjusting distribution of scores in a plurality of evaluation criteria. The abnormality diagnosis system according to claim 7,
Score distribution adjustment in multiple evaluation criteria is to set the score distribution so that it is a linear function of the distance from the most upstream detection unit and the order of the units, and optimize the intercept and slope of the linear function. An anomaly diagnosis system characterized by displaying that there is. The abnormality diagnosis system according to claim 7,
Score distribution adjustment in multiple evaluation criteria is to set the score distribution so that it becomes a linear function of the elapsed time from the detection unit with the earliest detection time, and optimize the intercept and slope of the linear function. An abnormality diagnosis system characterized by displaying that. The abnormality diagnosis system according to claim 7,
An anomaly characterized by indicating that the adjustment of the distribution of scores in a plurality of evaluation criteria optimizes the score given to a detection unit whose input condition is not included in the operating data that caused the state change. diagnostic system. The abnormality diagnosis system according to claim 7,
Adjustment of score allocation in multiple evaluation criteria means optimizing the score to be given to the detection unit including the control end of the control logic when the operating data that caused the state change is the control end of the control logic. An abnormality diagnosis system characterized by displaying. A method for diagnosing an abnormality in a facility that forms a movement path for a substance with a plurality of parts,
A plurality of state change detection means are set for each part of the facility and detect a change in the state of a substance, and the detection unit detects a plurality of operation data relating to the target part. Detect state change from the fact that the relationship of the data has changed from the normal state,
When the state change detecting means detects a change in state, a state change is detected using a plurality of evaluation criteria determined in relation to a substance moving through a plurality of regions in order to estimate the site that causes the state change. An anomaly diagnosis method characterized by: giving a score to a detection unit of each part that has detected an abnormality, and estimating a part that causes a change in the state of equipment according to the total score given by a plurality of evaluation criteria. The abnormality diagnosis method according to claim 12,
A method of diagnosing abnormalities, characterized by displaying adjusted distribution of scores in a plurality of evaluation criteria. An anomaly diagnosis program that uses a computer to diagnose anomalies in equipment that forms a substance movement path with a plurality of parts,
A plurality of state change detection means are set for each part of the facility and detect a change in the state of a substance, and the detection unit detects a plurality of operation data relating to the target part. a first program for detecting a state change from a change in the data relationship from a normal state;
When the state change detecting means detects a change in state, a plurality of evaluation criteria determined in relation to a substance moving through a plurality of regions are used to estimate the site that causes the state change. A second program that gives a score to the detection unit of each part that detects the An anomaly diagnosis program comprising a third program for adjusting the distribution of scores in the criteria.

Images