JP6482817B2 - Plant monitoring support system and plant monitoring support method - Google Patents

Description

  The present invention relates to a technology for a plant monitoring support system and a plant monitoring support method for monitoring a water treatment plant or the like.

  Currently, various industries operate various plants to produce various products. If an abnormality occurs in these plants, the production of the products stops and repair costs are incurred. In addition, the production of the product is stopped, which can cause a large loss. It is practically difficult to completely eliminate plant abnormalities and failures. Therefore, it is necessary to be able to grasp the sign of abnormality at the earliest possible time based on the measured value and clarify the root cause.

  In relation to such needs, a technique described in Patent Document 1 is known. In the technique described in Patent Document 1, ART (Adaptive Resonance Theory) is used as one of versatile abnormality prediction techniques. In this method, a normal category is constructed in advance based on past normal measurement values. In ART, when the latest measured value appears outside the normal data category, it is regarded as abnormal and an alarm is issued. A feature of ART is that a measurement value at the time of abnormality is not required as teacher data. Therefore, it is not necessary to intentionally generate an abnormality, and it can be said that this method is suitable for a plant that cannot perform various tests.

  For example, the technique described in Patent Literature 1 displays a method for dealing with an abnormality based on a category number selected from an abnormal operation knowledge database. This display is a trend graph of items used as input, information on the presence / absence of abnormality, and a list of countermeasures.

  On the other hand, a technique described in Patent Document 2 is also known as a plant monitoring technique. The technique described in Patent Document 2 relates to a plant state monitoring device that displays a monitoring result of a plant, and does not require prior knowledge about a plant to be applied, and aims to contribute to a reduction in operating cost. .

  Even in the technique described in Patent Document 2, ART is applied as a data classification function of the plant state monitoring device. The technique described in Patent Document 2 obtains the closest existing category (normal category N) for the data X belonging to the new category. The technique described in Patent Document 2 obtains the distance between the value of the data X and the center of the normal category N, and obtains the data item having the maximum contribution rate to this distance as the main factor data item.

JP 2014-63337 A JP 2013-140495 A

  In the technique described in Patent Document 1, after determining that an abnormality has occurred, the cause item and countermeasure are displayed. As a result, the occurrence of an abnormality is suddenly issued and it is difficult to prepare in advance. In addition, in the technique described in Patent Document 1, when there are a plurality of items related to the cause of the abnormality, when a certain abnormality is detected, the user knows which item is greatly affected by the abnormality. Therefore, it is desired to find the cause of the abnormality. However, in the technique described in Patent Document 1, it is not considered to display which items are affected abnormally.

  On the other hand, in the technique described in Patent Document 2, similarly to the technique described in Patent Document 1, after determining that an abnormality has occurred, the cause item is displayed. As a result, the occurrence of an abnormality is suddenly issued and it is difficult to prepare in advance. In the technique described in Patent Document 2, it is disclosed that the main factor item having the maximum contribution rate is displayed. However, it is not disclosed that information related to items other than the main factor item having the maximum contribution rate is displayed. Therefore, the user cannot know how items other than the main factor item having the maximum contribution rate have an effect on the abnormality.

  As described above, simply using the ART technique will later reveal that the target state has changed to a state different from the normal state, and there is a possibility that preparation for countermeasures against occurrence of an abnormality will not be in time.

  This invention was made in view of such a background, and this invention makes it a subject to show the sign of abnormality of a plant.

In order to solve the above-described problem, the present invention provides a measurement value acquisition unit that acquires a plurality of measurement values in a plurality of items newly from a plant, and a space that has each of the items related to the acquired measurement values as axes. By associating the plurality of measurement values, the acquired measurement values are plotted in the space, and by setting a predetermined region in the space, a category range corresponding to the region is set in the space. , By changing the size of the area, changing the range of the category, and classifying the measurement values that have already been acquired into categories, and the newly acquired measurement in any category range An alarm level calculation unit that calculates an alarm level over time depending on whether or not a value is out of the category, and a distance between the newly acquired measurement value and the category The contribution rate calculation unit that calculates the contribution rate corresponding to the time when the alarm level is calculated, and the alarm level and the contribution rate calculated in a predetermined period are displayed on the display unit in association with each other over time. And a display processing unit.

  According to the present invention, it is possible to show a sign of occurrence of a plant abnormality.

It is a figure which shows the structural example of the plant monitoring assistance system which concerns on this embodiment. It is a figure which shows the structural example of the plant monitoring assistance apparatus which concerns on this embodiment. It is a figure which shows the apparatus structural example of a desalination plant. It is a figure which shows the example of categorization of a measured value. It is a figure which shows the correlation between categories after a classification | category. It is a figure showing the time change of the category classification | category result at the time of learning and diagnosis with respect to the measured value obtained from a certain desalination plant. It is a figure showing change of lead time and the number of false detections at the time of performing learning and diagnosis by changing adjustment parameter ρ. It is a schematic diagram for demonstrating the calculation method of the alarm level which concerns on this embodiment. It is a schematic diagram for demonstrating the calculation method of the contribution rate which concerns on this embodiment. It is a flowchart which shows the procedure which the plant monitoring assistance system which concerns on this embodiment performs. It is a figure (the 1) which shows the example of the plant monitoring assistance screen which concerns on this embodiment. It is FIG. (2) which shows the example of the plant monitoring assistance screen which concerns on this embodiment. It is FIG. (1) which shows the example of the auxiliary | assistant screen which concerns on this embodiment. It is FIG. (3) which shows the example of the plant monitoring assistance screen which concerns on this embodiment. It is FIG. (4) which shows the example of the plant monitoring assistance screen which concerns on this embodiment. It is FIG. (2) which shows another example of the auxiliary | assistant screen which concerns on this embodiment.

Next, modes for carrying out the present invention (referred to as “embodiments”) will be described in detail with reference to the drawings as appropriate.
In the present embodiment, a desalination plant that desalinates seawater is assumed as a plant to which the plant monitoring support system is applied. However, if the operation state can be acquired as a measured value, it is not a desalination plant. May be. For example, the present invention may be applied to a plant, apparatus, or system that can acquire measurement values, such as a water and sewage plant or a medical system.

(System configuration)
FIG. 1 is a diagram illustrating a configuration example of a plant monitoring support system according to the present embodiment.
The plant monitoring support system 10 includes a plant monitoring support device 1, a desalination plant 3, and a monitoring control device 4.
The plant monitoring support apparatus 1 and the monitoring control apparatus 4 are connected via a communication network such as the Internet or a cloud.
The desalination plant 3 is a plant that desalinates seawater.
The monitoring control device 4 acquires measurement values such as flow rate, pressure, temperature, and water quality from the desalination plant 3 and sends the acquired measurement values to the plant monitoring support device 1. Moreover, the monitoring control apparatus 4 is a pump output and piping of the desalination plant 3 so that the measured value acquired from the desalination plant 3, for example, the production amount of fresh water and the discharge pressure of a pump may become those target values. The desalination plant 3 is monitored by outputting an operation amount such as an opening degree of a valve installed in the desalination plant 3 to the desalination plant 3.

The plant monitoring support apparatus 1 includes a measurement value acquisition unit 112, a categorization unit 113, an alarm level calculation unit 114, a contribution rate calculation unit 115, a display processing unit 116, a measurement value DB (Data Base) 121, and a monitoring support DB 122. A display device 2 is connected to the plant monitoring support device 1.
The measurement value acquisition unit 112 acquires the measurement value in the desalination plant 3 from the monitoring control device 4 and stores it in the measurement value DB 121.

The categorization unit (alarm level calculation unit) 113 classifies the states of the devices constituting the desalination plant 3 into a plurality of categories based on the normal measurement values stored in the measurement value DB 121. Furthermore, the categorizing unit 113 performs categorization calculation by adding the latest data, and evaluates whether the latest measured value is outside the normal data category.
The alarm level calculation unit 114 calculates the alarm level of the latest measurement value based on the processing result of the categorization unit 113. Further, the alarm level calculation unit 114 stores the calculated alarm level in the monitoring support DB 122. The alarm level will be described later.

When the latest measurement value is located outside the existing category, the contribution rate calculation unit 115 calculates a contribution rate that is the degree of influence of the measurement value item on the latest measurement value. The contribution rate calculation unit 115 stores the calculated contribution rate in the monitoring support DB 122. The contribution rate will be described later.
The display processing unit 116 acquires the alarm level calculated by the alarm level calculation unit 114 and the contribution rate calculated by the contribution rate calculation unit 115 from the monitoring support DB 122 and displays them on the display device 2 as changes over time in the alarm level and the contribution rate. To do. That is, the display processing unit 116 displays each screen to be described later with reference to FIGS.

(Plant monitoring support device)
FIG. 2 is a diagram illustrating a configuration example of the plant monitoring support apparatus according to the present embodiment.
The plant monitoring support apparatus 1 includes a memory 110, a CPU (Central Processing Unit) 130, a measurement value DB 121 as a storage device, a monitoring support DB 122, and a transmission / reception apparatus 140. A display device 2 is connected to the plant monitoring support device 1.
The transmission / reception device 140 is connected to the monitoring control device 4 and acquires the measurement value of the desalination plant 3 (FIG. 1) from the monitoring control device 4.

In addition, the program stored in the storage device is expanded in the memory 110, and the program expanded in the memory 110 is executed by the CPU 130, whereby the processing unit 111 and the measurement value acquisition unit 112 constituting the processing unit 111, A categorization unit 113, an alarm level calculation unit 114, a contribution rate calculation unit 115, and a display processing unit 116 are embodied.
The measurement value acquisition unit 112, the categorization unit 113, the alarm level calculation unit 114, the contribution rate calculation unit 115, the display processing unit 116, the measurement value DB 121, and the monitoring support DB 122 have already been described with reference to FIG. The description of is omitted.

In addition, each part 112-116 in the plant monitoring assistance apparatus 1 does not need to be integrated. That is, an apparatus including at least one of each of the units 112 to 116 may be installed separately from the plant monitoring support apparatus 1.
Further, at least one of the measurement value DB 121 and the monitoring support DB 122 may be installed outside the plant monitoring support apparatus 1.

FIG. 3 is a diagram illustrating a device configuration example of a desalination plant.
Here, the desalination plant 3 is, for example, a seawater desalination plant 3 in which the water to be treated is seawater, and FIG. 3 shows only its main part.
In the desalination plant 3, the seawater pumped up by the seawater supply pump 201 is membrane-separated into fresh water and concentrated water enriched in salt by the reverse osmosis membrane module 202. The membrane-separated seawater is discharged through the pipe 204. On the other hand, the concentrated water after membrane separation is carried to the power recovery device 203.
A part of the seawater pumped up by the seawater supply pump 201 is carried to the power recovery device 203. The power recovery device 203 uses the concentrated water discharged from the reverse osmosis membrane module 202 to pressurize the seawater that has been conveyed from the seawater supply pump 201 to the power recovery device 203.
The pressurized seawater is merged with the seawater conveyed from the seawater supply pump 201 by the reverse osmosis membrane module 202 and then conveyed to the reverse osmosis membrane module 202. By providing the power recovery device 203, the pressure of the concentrated water can be recovered and transmitted to the seawater conveyed from the seawater supply pump 201 to the power recovery device 203, so that energy efficiency can be realized.
Further, the concentrated water whose pressure is recovered by the power recovery device 203 is discharged through the pipe 205.

  “T1” to “T3” installed in the desalination plant 3 are thermometers, “P1” to “P6” are pressure gauges, and “W1” to “W5” are water quality meters, “F1” to “F4” are flow meters.

<Categorization>
Next, normal data categorization in the plant monitoring support system 10 according to the present embodiment will be described. Here, the category is a group of data having similarity. In the present embodiment, ART, which is one of clustering techniques, is used as an example.

FIG. 4 is a diagram illustrating an example of categorization of measurement values. FIG. 5 is a diagram showing the correlation between categories after classification. Reference is made to FIG. 1 as appropriate.
FIG. 4A shows “measured value A” (solid line) and “measured value B” in a period during which the desalination plant 3 is operating normally (normal period) and a period during which the operating state is diagnosed (diagnosis period). It is a figure which shows the example of the time change of (broken line). Here, the “measured value A” and the “measured value B” are measured values obtained from the sensors such as a thermometer, a flow meter, a pressure gauge, and a water quality meter in the desalination plant 3 via the monitoring control device 4. is there.

  The normal period is a certain test period. For example, in a normal period, a sufficient amount of measurement values are generated to generate a category. When the test period ends and the desalination plant 3 begins to operate, it becomes a diagnosis period, and each time a measured value is acquired, categorization (alarm level calculation) and contribution rate calculation described later are performed. Is called. However, even if it is a diagnosis period, if a period in which the measurement value is normal continues, it becomes a normal period after categorization, alarm level calculation, and contribution rate calculation.

  The categorizing unit 113 first inputs “measurement value A” and “measurement value B” in the normal period, and learns the correlation between “measurement value A” and “measurement value B”. At this time, as shown in FIGS. 4A and 4B, (1) “Measured value A” is large and “Measured value B” is the correlation between “Measured value A” and “Measured value B”. (2) “Measurement value A” and “Measurement value B” are both small, (3) “Measurement value B” is large, and “Measurement value A” is small. It shall be extracted. The categorizing unit 113 classifies each of (1) to (3) as category numbers 1 to 3. FIG. 4B shows a change in category together with a change in time. Here, in order to simplify the description, a case where categorization is performed based on a magnitude relationship between “measurement value A” and “measurement value B” is not limited to this, but “measurement value A” and “measurement value” are not limited thereto. A category may be generated by comparing the difference of “B” with a predetermined threshold.

Next, the operation of the alarm level calculation unit 114 according to the present embodiment will be described.
The categorizing unit 113 divides each measurement value into categories, and determines whether or not the newly acquired measurement value for the diagnosis period (measurement value to be processed) is classified into the category to which the measurement value for the normal period belongs. To do. For example, since “measured value A” and “measured value B” in the first diagnosis period shown in FIG. 4A are similar to the characteristics of “category 2” already learned, they are classified as “category 2”. Is done. However, the subsequent correlation between “measurement value A” and “measurement value B” is “large” for each data, and is not similar to the characteristics of any of the learned “category 1” to “category 3”. Therefore, it is classified as a new category. As a result, as shown in FIGS. 4B and 5, “Category 4” is generated as a new category corresponding to the measurement value 501 in addition to the learned “Category 1” to “Category 3”. The

(Alarm level)
For categorization, the category size (value corresponding to the radius of each category divided by a circle in FIG. 5) needs to be set in advance. When using ART, an adjustment parameter ρ (a value corresponding to the reciprocal of the radius of the category, 0 <ρ <1) is set. Increasing ρ tends to reduce the size of each category and increase the number of categories. On the other hand, when ρ is set small, the size of each category increases, and the number of categories tends to decrease.

FIG. 6 is a diagram showing a change over time in the category classification result when learning and diagnosis are performed on a measurement value obtained from a desalination plant.
This device is known to have failed at time t2. A new category appears before the failure occurs, that is, from time t1. From this, it is considered that a sign of abnormality appeared at time t1. Therefore, the time (t2-t1) is defined as the lead time 601 until the failure occurs.

FIG. 7 is a diagram showing changes in the lead time and the number of false detections when learning and diagnosis are performed by changing the adjustment parameter ρ.
As shown in FIGS. 7A and 7B, when the adjustment parameter ρ is set large, that is, when the category radius is set small, the lead time becomes long, but the number of false detections tends to increase. On the other hand, if the adjustment parameter ρ is set to be small, that is, the category radius is set to be large, the lead time is shortened, but the number of false detections tends to be reduced. That is, the greater the value of the adjustment parameter ρ, the lower the accuracy, but the earlier the occurrence of an abnormality can be detected, and the smaller the value of the adjustment parameter ρ, the more immediately the occurrence of an abnormality can be detected with higher accuracy. It will be possible.
Using this characteristic, the categorizing unit 113 changes the value of ρ within a predetermined range, for example, by 10 steps, and outputs the number of times determined to be abnormal as an “alarm level”.

This will be described with reference to FIG.
FIG. 8A shows the case where the adjustment parameter ρ is “large”, that is, the category radius is “small”. In this case, the newly measured value 501 is outside the range of “Category 1” to “Category 3” and is included in “Category 4” which is a new category.
FIG. 8B shows the case where the adjustment parameter ρ is “medium”, that is, the category radius is “medium”. In this case, the measured value 501 is located on the “category 1” line.
FIG. 8C shows the case where the adjustment parameter ρ is “small”, that is, the category radius is “large”. In this case, the measurement value 501 is completely included in “Category 1”.
In the case of the example of FIG. 8, the alarm level calculation unit 114 determines that the alarm level is “medium”.

  In the example of FIG. 8, the adjustment parameter ρ is changed in three stages of “large”, “medium”, and “small”, but the value of ρ may be changed in 10 stages. In this case, the alarm level calculation unit 114 determines that the alarm level = 0 if it has not been determined to be abnormal once, and determines that the alarm level = 10 if it has been determined to be abnormal at all stages.

  That is, the alarm level is whether or not the measurement value to be processed is outside the range of the existing category when the category radius is changed and the category is set at which radius. That is, if the alarm level is a small value, it is estimated that the measured value is far from the abnormal value because the measured value is within the existing category even if the radius of the category is reduced. Conversely, if the alarm level is a large value, the measured value is within the existing category while the category radius is large, but if the category radius is reduced, the measured value will be output outside the existing category. The measured value is estimated to be close to the abnormal value.

(Contribution rate)
FIG. 9 is a schematic diagram for explaining a contribution rate calculation method according to the present embodiment.
The contribution rate calculation unit 115 calculates the influence (contribution rate) of “measurement value A” and “measurement value B” on the measurement value to be processed by the method shown in FIG. Note that the contribution rate is calculated when the alarm level is fixed, that is, when the adjustment parameter ρ is changed stepwise and the measurement value to be processed crosses the boundary line of the existing category. .

The contribution rate calculation unit 115 determines which category of the existing categories (“category 1” to “category 3” in the example of FIG. 9) for the measurement value 501 classified into the new category (“category 4” in FIG. 5). To find the closest to. The measurement values classified into the new category should have been included in the existing category when the adjustment parameter ρ is larger than the current adjustment parameter ρ (when the radius of the category is small). Therefore, when the adjustment parameter ρ is changed, there is a high possibility that the existing category to which the measurement value to be processed belongs most is the nearest existing category in the measurement value to be processed.
The existing category is a category into which the measurement values in the normal period are classified.

That is, the contribution rate calculation unit 115 performs processing when the adjustment value ρ (or adjustment parameter ρ that is one step larger) than the adjustment parameter ρ when the measurement value to be processed is outside the existing category. The existing category in which the target measurement value is included is set as the re-neighbor category.
Referring to the example of FIG. 8, the measured value 501 is included in the new category “Category 4” when the adjustment parameter ρ is “Large”, but when the adjustment parameter ρ is “Medium” and “Small”. Are included in the existing category “Category 1”.
Therefore, in the case of the example of FIG. 8, the contribution rate calculation unit 115 sets the re-neighbor category of the measurement value 501 to “category 1”.

  In the present embodiment, when the measurement value to be processed is an adjustment parameter ρ that is larger than the adjustment parameter ρ when the measurement value goes out of the existing category, the existing category is searched for. In this case, the nearest neighbor category is searched, but the present invention is not limited to this. For example, an existing category having the closest distance between the measured value and the center of each existing category may be set as the nearest category.

  Next, the contribution rate calculation unit 115 calculates a difference between the measurement value to be processed and each item of the nearest category. In the example of FIG. 9, each item is “measurement value A” and “measurement value B”. This difference may be the distance from the center point of the nearest category or the distance in the normal direction from the hypersphere that is the boundary of the nearest category (on the circumference in the example of FIG. 9 that is two-dimensional).

It can be said that an item with a large difference value is highly likely to be a cause of an abnormal measurement value. Therefore, the difference between the items from the nearest category to the measurement value to be processed is defined as “contribution rate”. In the example of FIG. 9, reference numeral 701 is the contribution ratio of “measurement value A” to the measurement value 501, and reference numeral 702 is the contribution ratio of “measurement value B” to the measurement value 501.
In the present embodiment, it is assumed that the contribution ratio is a percentage of the distance from the nearest category of the measurement value to be processed in each item.

  For example, there are three items “measurement value A” to “measurement value C”, and the distance between the measurement value to be processed and the nearest category is set to “distance A” in the “measurement value A” direction. , “Measurement value B” direction distance is “Distance B”, “Measurement value C” direction “Distance C”, the contribution ratio of “Measurement value A” to “Measurement value C” is expressed by the following equation: It is.

Contribution rate of “measured value A” = (100 × distance A) / (distance A + distance B + distance C)
Contribution rate of “measured value B” = (100 × distance B) / (distance A + distance B + distance C)
Contribution rate of “measured value C” = (100 × distance C) / (distance A + distance B + distance C)

  By the way, the total contribution rate for all items is 100%.

  The display processing unit 116 displays the contribution rate of all the measurement values to be processed, for example, “measurement value A” and “measurement value B”, on the driving status display unit 122 in real time or at regular time intervals. indicate. By doing in this way, the operator can confirm the present contribution rate. In addition, as will be described later, by associating the measurement location of the measurement value with the contribution rate, it is possible to support the prioritization of targets to be subjected to on-site confirmation and precise inspection by the operator. As a result, measures can be taken to prevent the occurrence of abnormality at an earlier stage.

(flowchart)
FIG. 10 is a flowchart showing a procedure performed by the plant monitoring support system according to the present embodiment. In the present embodiment, the adjustment parameter ρ is changed from a small value to a large value. That is, the radius of the supersphere constituting the category is changed from a large value to a small value, but may be reversed.
First, the measured value acquisition part 112 acquires the measured value of the desalination plant 3 via the monitoring control apparatus 4 (S101).

Next, the categorizing unit 113 sets the adjustment parameter ρ to an initial value, that is, the smallest value (S102).
Furthermore, the categorizing unit 113 categorizes the measurement values (measurement values acquired in advance) in the normal period with the current adjustment parameters (S103).

Then, the alarm level calculation unit 114 determines whether or not the measurement value currently being processed is within the existing category (S104).
As a result of step S104, when it is in the existing category (S104 → Yes), the alarm level calculation unit 114 determines whether or not the current value of the adjustment parameter ρ is a set maximum value (S105). ).
As a result of step S105, if the current value of the adjustment parameter ρ is not the maximum value (S105 → No), the value of the adjustment parameter ρ is increased by one step (S106), and the process returns to step S103.

  As a result of step S105, when the current value of the adjustment parameter ρ is the maximum value (S105 → Yes), the alarm level calculation unit 114 sets the alarm level to “0” (that is, the lowest level) (S107), and the contribution rate The calculation unit 115 calculates the contribution rate (S109).

On the other hand, if the result of step S104 is that the current category is not within the existing category (S104 → No), the alarm level calculation unit 114 sets the alarm level of the measurement value to be processed to the previous adjustment parameter ρ. The alarm level associated with the adjustment parameter ρ at the stage is determined (S108).
Then, the contribution rate calculation unit 115 calculates the contribution rate of the measurement value to be processed (S109). The contribution rate is calculated by the method described in FIG.

After calculating the contribution rate, the display processing unit 116 displays the calculated alarm level and the change over time in the contribution rate on the display unit (S110). The screen displayed in step S110 will be described later.
Thereafter, the processing unit 111 returns the processing to step S101, and calculates an alarm level and a contribution rate at the next time.

(Screen example)
Next, a plant monitoring support screen according to the present embodiment will be described with reference to FIGS. Both the alarm level value and the contribution rate value are displayed as changes over time so that they can be used for judgment. Examples of plant monitoring support screens are shown in FIG. 11 to FIG. 16, but other formats can be used as long as the change over time of the alarm level and the contribution rate or the change over time of the product of the alarm level and the contribution rate is displayed on the same screen. Good.

FIG. 11 is a diagram illustrating an example of a plant monitoring support screen according to the present embodiment.
In the diagrams shown in FIGS. 11 to 13, 15, and 16, the measurement value items are six items of “measurement value A” to “measurement value F”.
The plant monitoring support screen 1100 in FIG. 11 shows the change over time in the alarm level and the contribution rate in a two-dimensional graph (on a plane). That is, the plant monitoring support screen 1100 is a graph with the alarm level and the contribution rate as two axes.
In the plant monitoring support screen 1100 illustrated in FIG. 11, the series 1101 indicates “measured value A”, and the series 1102 indicates “measured value B”. A series 1103 indicates “measured value C”, and a series 1104 indicates “measured value D”. Further, the series 1105 indicates “measured value E”, and the series 1106 indicates “measured value F”. In this case, the items are “measured value A” to “measured value F”.
In FIG. 11, the alarm level and the contribution rate in each item are represented by symbols (circles).

  Furthermore, the black symbols indicate that the data is based on the latest measured values. A symbol indicated by a fine dot indicates that the data is a measurement value immediately before the latest measurement value. In addition, a symbol indicated by a coarse dot indicates that the data is based on a measurement value immediately before the measurement value of the symbol indicated by a fine dot. A white symbol indicates data based on the measurement value at the oldest time.

  After that, when a new measurement value is acquired and the alarm level and contribution ratio of the acquired measurement value are calculated, the symbol for this new measurement value is displayed as a black-filled symbol and is now black-filled. The symbol is shown with fine dots. Similarly, when a new measurement value is acquired, a symbol currently indicated by a fine dot is indicated by a coarse dot, and a symbol currently indicated by a coarse dot is displayed by a white symbol. The symbols shown in white are not displayed.

  In FIG. 11, four-stage changes are shown as changes over time, but the number is not limited to four stages. Although six items are displayed, the number is not limited to six.

  Moreover, since the alarm level is determined by the points plotted (associated) in the hyperspace with the item as an axis, the measurement values acquired at the same time have the same value for each item, and the contribution rate Are different. Incidentally, the contribution rate is a value calculated by the above-described calculation formula.

For items in which the symbol moves to the right of the screen at the same time as the alarm level becomes high, it can be said that the measurement value item whose contribution rate increases is the most likely candidate for the cause of the abnormality. Even if the contribution rate is high, an item whose contribution rate decreases by moving to the left of the screen at the same time as the alarm level becomes high is unlikely to be a candidate for the cause of abnormality.
Then, by referring to the past symbols, the user can sensuously grasp changes over time in the alarm level and the contribution rate up to the present time, and can appropriately implement a precaution.

For example, in the plant monitoring support screen 1100 shown in FIG. 11, as the alarm level increases, the series 1101 (“measured value A”) and the series 1104 (“measured value D”) move to the right side of the screen. That is, the contribution rate is rising. In particular, since the series 1101 (“measurement value A”) shows a high contribution rate, it can be seen that the user needs to pay attention to the place where the “measurement value A” is acquired.
On the other hand, as shown in the series 1102 (“measured value B”), even if the contribution ratio is high, the contribution ratio decreases with time if it moves to the left side of the screen with time. The measured value is considered to be no problem.

In the example of FIG. 11, each item shown on the plant monitoring support screen 1100 is not distinguished on the display, but the color, the shape, or the size is changed for each item. Also good. That is, the symbol color of each item may be changed and displayed such that the symbol of “measurement value A” is red, the symbol of “measurement value B” is purple,.
Furthermore, in the example of FIG. 11, the change in time is represented by dots, but the change in time may be represented by the brightness of the symbol. That is, the symbol related to the latest measurement value may be displayed darker and may be displayed lighter as it becomes the past, or the symbol size may be reduced as it becomes the past. In FIG. 11, a line connecting each symbol may not be displayed.

FIG. 12 is a diagram illustrating another example of a plant monitoring support screen according to the present embodiment.
The plant monitoring support screen 1200 shown in FIG. 12 is a radar chart graph showing the change over time in the product of the alarm level and the contribution rate. The plant monitoring support screen 1200 includes a plurality of types of items (six types in the example of FIG. 12) displayed on different axes, and is indicated by the types of lines for displaying changes over time. A line 1201 indicated by a solid line indicates the latest (current or the latest time from the present). Next, a line 1202 indicated by a one-dot chain line indicates that the time is in the past from the line 1201. A line 1203 indicated by a broken line indicates that the time is in the past of the line 1202. A line 1204 indicated by a fine broken line indicates that it is the past.

In the example of FIG. 12, the symbol of the item displayed on the plant monitoring support screen 1200 is not distinguished on the display, but the color, shape, or size of the symbol is changed for each item. Or you may.
In addition, in the plant monitoring support screen 1200 of FIG. 12, a change with time is shown depending on the line type. However, not limited to this, the line indicating the past is displayed lighter and the line indicating the latest is displayed darker. It may be displayed so as to become thinner as it becomes the past. Alternatively, the colors of the lines 1201 to 1204 may be changed from the present to the past.

  In the plant management support screen 1200, the value of each axis indicates the product of the alarm level and the contribution rate. Therefore, it shows that the alarm level increases as the area of the radar chart increases. Further, the operator knows sensuously that the item with a large axis value of the radar chart is the item with a large contribution rate, and is most likely as a cause of abnormality. Further, by changing the line of the radar chart, the operator can sensuously grasp the change over time of the contribution rate up to the present time, and can appropriately implement a precaution.

  In the plant monitoring support screen 1200 of FIG. 12, it can be seen that the contribution ratios of “measured value C” and “measured value F” are increased.

  Here, in FIG. 11 and FIG. 12, the items are exemplified as six types, but the number of items may increase depending on the target plant. In such a case, the display processing unit 116 may preferentially display items with a large contribution rate and may not display items with a small contribution rate. In this way, the operator can easily confirm important information and can know in advance signs of occurrence of abnormality.

FIG. 13 is a diagram illustrating an example of an auxiliary screen according to the present embodiment.
An auxiliary screen 1300 shown in FIG. 13 is a screen displayed together with the plant management support screen 1100 and the plant management support screen 1200 of FIG.
The auxiliary screen 1300 shown in FIG. 13 shows a configuration diagram of the desalination plant 3 (FIG. 3) to be monitored, and “measurement value A” to “measurement” shown in FIG. 11 and FIG. Locations where the value “F” is acquired are indicated by reference numerals 1301 to 1306.
That is, the “measured value A” is the temperature obtained from the thermometer “T2”, the “measured value B” is the pressure value obtained from the pressure gauge “P6”, and the “measured value C” is the water quality meter “W5”. It is a measurement value about the water quality obtained from "." “Measured value D” is a pressure value obtained from the pressure gauge “P3”, “Measured value E” is a flow rate obtained from the flow meter “F2”, and “Measured value F” is a pressure gauge “P2”. Is the pressure value obtained from

In this way, by displaying the auxiliary screen 1300 together with the plant management support screen 1100 and the plant management support screen 1200 of FIG. 11, the measurement locations of a plurality of items can be grasped at a glance, and what kind of phenomenon occurs as a whole. It will be easier to figure out if you are. As a result, it is possible to reduce correspondence errors and delays.
Moreover, since the location of the desalination plant 3 applicable to the item with the high alarm level and the contribution rate can be specified easily, response can be speeded up.

14 to 15 are diagrams illustrating another example of the plant management support screen according to the present embodiment, and FIG. 16 is a diagram illustrating another example of the auxiliary screen according to the present embodiment.
In the plant management support screen 1400 shown in FIG. 14, the change over time of the alarm level is shown in the upper part, and the change over time in the contribution rate is shown in the lower part. In the change in the contribution rate with time in the lower part of FIG. 14, the change with time in the contribution rate in “measured value A” (solid line) and “measured value B” (broken line) is shown as a trend graph. Note that the times are associated in the upper and lower graphs. The contribution rate is calculated by the above formula.

  According to the plant management support screen 1400 shown in FIG. 14, it can be seen that the alarm level increases with time. And it turns out that the contribution rate of "measurement value B" (broken line) is rising with the raise of an alarm level. Thereby, it becomes easy for the user to visually confirm the temporal change of the alarm level and the contribution rate.

In FIG. 14, the graph is divided by the alarm level and the contribution rate. For example, even if a graph summarizing the alarm level and the contribution rate is displayed by setting the vertical axis to alarm level × contribution rate, Good.
In FIG. 14, two items “measured value A” and “measured B” are displayed, but the number of items to be displayed is not limited to two.

  11 and 12 are easy to grasp sensuously, but in order to grasp the situation more quantitatively, a trend graph as shown in FIG. 14 may be useful for understanding. Therefore, as shown in FIG. 14, it is effective to show the change over time of the alarm level and the contribution rate with a trend graph as shown in FIG.

Furthermore, as shown in FIG. 15, it is also possible to display on-site image information on the same screen for items for which the contribution rate is displayed.
As shown in FIG. 15, when the symbol of FIG. 11 is clicked on the screen or the mouse is placed on the symbol, an image 1501 of field equipment corresponding to the symbol is displayed. As a result, it is possible to immediately share an image of equipment having a high contribution rate when the alarm level is raised, and to reduce misunderstandings. In some cases, the cause of the abnormality may be identified by viewing the image. In addition, by adopting a form in which audio information can be output simultaneously with an image, there is a high possibility that the cause of the abnormality can be specified more accurately due to abnormal sounds in the vicinity of the site.
In FIG. 15, the plant management support screen 1100 illustrated in FIG. 11 has been described as an example. However, an image 1501 may be displayed on the plant management support screens 1200 and 1400 illustrated in FIG. 12 and FIG. 14.
An image 1501 in FIG. 15 is a still image, but may be a moving image.

Further, the auxiliary screen 1600 in FIG. 16 highlights an acquisition location of an item whose contribution rate is equal to or greater than a predetermined value. In the example of FIG. 16, as an emphasis display, an acquisition location of an item whose contribution rate is equal to or greater than a predetermined value blinks (reference numeral 1601). FIG. 16 shows that the contribution rate of “measured value A” is equal to or greater than a predetermined value.
By doing in this way, it becomes easy for a user to identify the location where the contribution rate is high in the desalination plant 3.
Alternatively, a symbol having a high contribution rate may be displayed large.
In FIG. 16, locations where the contribution rate is greater than or equal to a predetermined value are blinking, but other locations where the alarm level × contribution rate is greater than or equal to the predetermined value are also blinked. It may be.

  According to the plant management support screen 1500 in FIG. 15 and the auxiliary screen 1600 in FIG. 16, it is easy to identify a location where the contribution rate (or alarm level) is increased in the desalination plant 3. It is possible to shorten the time required for planning countermeasures and making decisions. As a result, it is possible to speed up the response.

(Summary)
As shown in the present embodiment, the display processing unit 116 displays the calculated alarm level on the display device 2 in real time or at regular time intervals if possible, thereby notifying the operator of the alarm level. Can confirm the relationship between the current measured value and the abnormal value.
Furthermore, the display processing unit 116 displays the change over time in the alarm level and the contribution rate on the display device 2, so that the operator can easily grasp the change over time in the alarm level and the contribution rate. That is, the operator can confirm whether or not the measured value is approaching the abnormal value.
As a result, the occurrence of an abnormality is not suddenly issued, and the operator can perform advance preparations such as on-site confirmation and precise inspection before the occurrence of the abnormality. As a result, it is possible to minimize damage at an initial stage before a serious failure occurs. For example, in the desalination plant 3 shown in FIG. 3, the reverse osmosis membrane module 202 may be blocked by fouling, but the operator performs flushing or chemical cleaning at the initial stage to replace the membrane module or stop the plant. Damage can be minimized. Similarly, the bearing of the high-pressure pump 202 may also be damaged, but the investigator must replace the grease in the initial stage to minimize damage to the bearing and bearings, resulting in plant shutdown and water shutdown. Can do.

  Since the plant monitoring support system 10 according to the present embodiment can be realized if information can be transmitted electronically, the plant monitoring support device 1, the desalination plant 3, and the monitoring control device 4 are connected via a communication network such as the Internet or the cloud. The information may be sent and received and processed. In the case of adopting such a form, since it is possible to detect an abnormality for a plurality of desalination plants 3, centralized management with a small number of people can be realized. Furthermore, since it is not necessary for each desalination plant 3 to have software for realizing the plant monitoring support system 10, processing on the desalination plant 3 side is almost unnecessary, and software maintenance management becomes extremely easy.

The present invention is not limited to the above-described embodiment, and includes various modifications. For example, the above-described embodiment has been described in detail for easy understanding of the present invention, and is not necessarily limited to having all the configurations described. Further, it is possible to add, delete, and replace other configurations for a part of the configuration of the embodiment. For example, the change with time of either the alarm level or the contribution rate may be displayed.
Moreover, the plant monitoring support system 10 according to the present embodiment may be used for business via a network.

Each of the above-described configurations, functions, the units 111 to 116, the DBs 121 and 122, etc. may be realized by hardware by designing a part or all of them, for example, with an integrated circuit. Further, as shown in FIG. 2, the above-described configurations, functions, and the like may be realized by software by interpreting and executing a program that realizes each function by a processor such as the CPU 130. In addition to storing information such as programs, tables, and files for realizing each function in an HD (Hard Disk), a memory 110, a recording device such as an SSD (Solid State Drive), or an IC (Integrated Circuit) card It can also be stored in a recording medium such as an SD (Secure Digital) card or a DVD (Digital Versatile Disc).
In each embodiment, control lines and information lines are those that are considered necessary for explanation, and not all control lines and information lines are necessarily shown on the product. In practice, it can be considered that almost all configurations are connected to each other.

DESCRIPTION OF SYMBOLS 1 Plant monitoring assistance apparatus 2 Display apparatus 3 Desalination plant 4 Monitoring control apparatus 10 Plant monitoring assistance system 111 Processing apparatus 112 Measurement value acquisition part 113 Categorization part (alarm level calculation part)
114 Alarm level calculation unit 115 Contribution rate calculation unit 116 Display processing unit

Claims (12)

A measurement value acquisition unit for acquiring a plurality of measurement values in a plurality of items newly from the plant;
Plotting the acquired measurement values in the space by associating the plurality of measurement values with a space having each item relating to the acquired measurement values as an axis, and setting a predetermined region in the space Then, by setting a category range corresponding to the region in the space and changing the size of the region, the measurement values already acquired are classified into categories while changing the category range. A categorization section;
An alarm level calculation unit that calculates an alarm level over time according to whether a newly acquired measurement value is out of the category in which category range;
A contribution rate calculation unit that calculates a contribution rate that is a value related to a distance between the newly acquired measurement value and the category in association with a time at which the alarm level is calculated;
A display processing unit for displaying the alarm level and the contribution rate calculated in a predetermined period in association with each other on a display unit;
A plant monitoring support system comprising: The contribution rate calculation unit calculates the contribution rate for each item,
The display processing unit displays the calculated alarm level and the calculated contribution rate for each item over time on a plane having the alarm level and the contribution rate as axes. The plant monitoring support system according to 1. The contribution rate calculation unit calculates the contribution rate for each item,
2. The plant monitoring according to claim 1, wherein the display processing unit displays, on a corresponding axis, information on an alarm level and the contribution rate in each item in a radar chart having each item as an axis. Support system. The display processing unit
The data relating the alarm level and the contribution rate displayed on the display unit and the image relating to the acquisition location of the newly acquired measurement value in the plant are associated and displayed on the display unit. The plant monitoring support system according to claim 1, wherein The display processing unit
When associating the alarm level and the contribution rate displayed on the display unit with the image relating to the newly acquired measurement value acquisition location in the plant and displaying the association on the display unit, When at least one of the alarm level and the contribution rate is equal to or higher than a predetermined value, an acquisition location of the measurement value where at least one of the alarm level and the contribution rate is equal to or higher than a predetermined value is highlighted. The plant monitoring support system according to claim 4. The display processing unit
The plant monitoring support system according to claim 1, wherein the change with time of the contribution rate and the change with time of the alarm level are displayed as a trend graph. A plant monitoring support system that monitors plant abnormalities
The measurements in the newly plurality of items from the plant, a plurality of acquisition,
Plotting the acquired measurement values in the space by associating the plurality of measurement values with a space having each item relating to the acquired measurement values as an axis, and setting a predetermined region in the space Then, by setting a category range corresponding to the region in the space and changing the size of the region, the measurement values already acquired are classified into categories while changing the category range. ,
In which category range, the alarm level is calculated over time depending on whether the newly acquired measurement value is out of the category,
A contribution rate that is a value related to a distance between the newly acquired measurement value and the category is calculated in association with a time at which the alarm level is calculated,
The alarm level and the contribution rate calculated in a predetermined period are displayed in association with each other on a display unit over time. The plant monitoring support system includes:
Calculate the contribution rate for each item,
The plant monitoring according to claim 7, wherein the calculated alarm level and the calculated contribution rate for each item are displayed over time on a plane having the alarm level and the contribution rate as axes. Support method. The plant monitoring support system includes:
Calculate the contribution rate for each item,
8. The plant monitoring support method according to claim 7, wherein, in a radar chart having each item as an axis, information related to an alarm level and the contribution rate in each item is displayed on a corresponding axis. The plant monitoring support system includes:
The data relating the alarm level and the contribution rate displayed on the display unit and the image relating to the acquisition location of the newly acquired measurement value in the plant are associated and displayed on the display unit. 8. The plant monitoring support method according to claim 7, wherein The plant monitoring support system includes:
When associating the alarm level and the contribution rate displayed on the display unit with the image relating to the newly acquired measurement value acquisition location in the plant and displaying the association on the display unit, When at least one of the alarm level and the contribution rate is equal to or higher than a predetermined value, an acquisition location of the measurement value where at least one of the alarm level and the contribution rate is equal to or higher than a predetermined value is highlighted. The plant monitoring support method according to claim 8. The plant monitoring support system includes:
The plant monitoring support method according to claim 7, wherein the change over time of the contribution rate and the change over time of the alarm level are displayed as a trend graph.

Images