JPH1139162A - Engineering equipment - Google Patents

Abstract

PROBLEM TO BE SOLVED: To automatically enable analysis, extraction and adaptation of a control rule in on-line by providing a rule adaptation means to adapt the control rule extracted by a rule extraction means to an active system device. SOLUTION: A method to acquire data is defined and next, timing, interval and frequency, etc., of extracting data are set on an automatic crustering definition screen. After crustering technique is selected, a required data item is set. After methods to acquire and to analyze the data are defined, these definitions are transmitted to a programmable controller 4. A value of a specified register is stored in a buffer based on the transmitted definitions by the programmable controller 4. The required data from the programmable controller 4 is acquired by every specified cycle based on the set definitions by an engineering device 1. Next, the data is analyzed and the control rule is extracted in the on-line only when a condition is satisfied based on the stored data by the engineering device 1.

Description

DETAILED DESCRIPTION OF THE INVENTION

[0001]

The present invention relates to an engineering apparatus for monitoring and maintaining a system such as a water purification plant system.

[0002]

2. Description of the Related Art Conventionally, there are many systems operated by human empirical knowledge, and the automation of this type of system is expected to receive more and more attention in the sense of releasing humans from physical hard labor. . Fuzzy control, for example, reflects human empirical knowledge. In a system using this fuzzy control, control rules are extracted by analyzing data obtained during operation of the system. For example, in the case of a water purification plant system, it is known that the occurrence of abnormal bacteria is related to the water temperature and the water flow.However, since the specific water temperature and water flow slightly vary depending on Control rules are extracted. The engineering apparatus that monitors and maintains the system has a function of collecting data when the system is operating and storing the data in a storage device such as a magnetic disk device in order to extract a control rule. A large computer with a high processing capacity is used to extract the actual control rules, and the data is analyzed.

[0003]

However, such a conventional engineering apparatus has the following problems. Data clustering is necessary to embody human empirical knowledge, but this data clustering requires the selection of data items and a sensoring technology for acquiring data from the system. Must be established. In addition, in order to create a new control rule, data accumulated in the storage device of the engineering device must be fetched and organized, and analyzed using a computer, which requires a huge amount of work. Further, when applying a control rule obtained by analysis by a computer to a system, the system must be temporarily stopped, resulting in a decrease in the operating rate of the system. Therefore, an object of the present invention is to provide an engineering apparatus that can analyze data in real time while a system is operating, extract control rules, and apply the obtained control rules to the system.

[0004]

In order to achieve the above-mentioned object, an engineering apparatus according to the present invention is an engineering apparatus for monitoring and maintaining a system apparatus, wherein specified data is collected and stored while the system apparatus is operating. A data accumulating unit, a data analyzing unit for analyzing data collected by the data accumulating unit if a predetermined condition is satisfied, and a rule extracting unit for extracting a control rule from an analysis result by the data analyzing unit. And a rule application unit for applying the control rules extracted by the rule extraction unit to the operating system device. According to this configuration, data is collected while the system is operating and data that satisfies a predetermined condition is analyzed, and a control rule is extracted. Then, the extracted control rules are applied to the operating system. Therefore, it is not necessary to analyze data and extract control rules by an external computer, so that labor can be saved significantly. In addition, since control rules are automatically extracted and applied to the water purification system, the ability to apply to changes in the system environment and disturbances can be improved, and control can be performed even when modeling of the control target is incomplete. In addition, control rules can be extracted and applied in real time. Further, there is no need to stop the system in order to apply the control rules to the system, so that the operating rate of the system does not decrease.

[0005]

Embodiments of the present invention will be described below with reference to the drawings. FIG. 1 is a block diagram showing a configuration of an embodiment of an engineering apparatus according to the present invention.
FIG. 2 is a block diagram showing a configuration of a water purification system using the engineering device according to this embodiment. In FIG. 2, an engineering device 1 is installed in a monitoring room 2 of a water purification system, and is connected to a programmable controller 4 via a transmission line 3. The engineering device 1 gives a command to the programmable controller 4 and fetches data from the programmable controller 4. Water purification sensor 6 has a water flow sensor 7,
A water temperature sensor 8 and a bacteria detection sensor 9 are provided, and their outputs are taken into the programmable controller 4 via the remote I / O 5. In this case, the water flow sensor 7 and the water temperature sensor 8 are analog input type sensors, and the bacteria detection sensor 9 is a digital input type sensor. The bacteria detection sensor 9 is turned on when bacteria exceed a predetermined amount. Programmable controller 4 and remote I / O5
Are connected via a transmission line 19. The engineering device 1 includes a main body 10, a keyboard 11, a mouse 12, and a CRT 13, as shown in FIG.
In addition to the conventional monitoring and maintenance functions, it has functions such as a transmission function, a data storage function, a data analysis function, a rule extraction function, and a rule application function. The main body 10 includes a monitoring / maintenance unit 15, a learning unit 16, and a transmission unit 17. The learning unit 16 includes a data storage unit (such as a hard disk device) 2
0, a data analysis unit 21, a rule extraction unit 22, and a rule application type unit 23. Here, FIG. 3 is a diagram showing an automatic clustering definition screen in the engineering device 1. On this screen, the control method of the data storage function is defined. In this figure, “clustering group name” is a definition name, and corresponds to a “file name” in a file. The “group number” is for specifying a group number, and can be specified from 1 to N. The “data extraction timing” is for designating a timing at which data is extracted. The H-SCAN stores data for each high-speed scan, the L-SCAN stores data for each low-speed scan, and stores data in a data program. There is a Program that defines the function to be stored.
The “data extraction interval” is used to specify the data extraction interval, and it is possible to specify how many times the data is to be extracted (except for Program). The “number of extractions” is used to specify the number of times data is accumulated. The "data extraction start condition" is for designating a condition for taking in data from the programmable controller 4, and is triggered by the set data to start storing data. In the case shown in this figure, the input by the push button is designated. The “data extraction stop condition” is for designating a condition for stopping the acquisition of data from the programmable controller 4, and is triggered by the set data to stop the data extraction. In the case shown in this figure, the input by the stop push button is designated. “Extracted data specification” is for specifying data to be extracted. For example, when the water flow sensor 7 assigned as the sensor B is designated, the output of the water flow sensor 7 is taken. The definition on this automatic clustering definition screen can be set even while the system is operating. Further, the automatic clustering definition set on the automatic clustering definition screen is sent to the programmable controller 4. FIG. 4 is a diagram showing an automatic clustering selection screen. On this screen, a data analysis method and a control rule extraction method are defined. In this figure, “Clustering selection” is for designating a data analysis method.
There are an axis histogram, a two-axis histogram, one-dimensional clustering, two-dimensional clustering, and three-dimensional clustering. Select the clustering method using the keyboard 11
The operation can be arbitrarily performed by operating the mouse or the mouse 112 and can be instructed to the engineering apparatus 1.
The instructed result is displayed on the CRT 13 by the MMI function of the engineering device 1. Based on the result of the clustering, the operator extracts a control rule existing for the event. "Selection data" is for selecting data. The “rule application method” is for selecting whether to automatically apply the control rule based on the analyzed data or to manually apply the control rule. FIG. 5 is a diagram showing an automatic rule extraction and application screen. On this screen, you can monitor the status of extraction of control rules and the status of application to water purification systems. Hereinafter, the operation of the engineering apparatus will be described. It has been found that the occurrence of bacterial abnormalities in the water purification plant system is related to the water temperature and the water flow. Rules cannot be extracted. Therefore, as shown in the flowchart of FIG. 6, first, a data acquisition method is defined on the automatic clustering definition screen. In this case, the operator designates, as the extracted data, the water flow and the water temperature considered as the cause factors and the occurrence of the bacterial abnormality indicating that the event has occurred. Next, a data extraction timing, an extraction interval, an extraction frequency, a data extraction start condition, and a data extraction stop condition are set. After performing the automatic clustering definition, specify the data analysis method and control rule application method on the automatic clustering selection screen. That is, the operator specifies a method of analyzing the accumulated data. The operator selects a clustering method prepared in the engineering device 1. Then, after selecting the clustering method, necessary data items are set. In this case, assuming that a two-dimensional cluster is selected, X and Y axis data are designated as cause data, and an item of event data is designated as event data. After defining a data acquisition method, a data analysis method, and control rule application conditions, the definitions are sent to the programmable controller 4. When receiving the definitions, the programmable controller 4 accumulates the value of the designated register in the buffer based on the definitions. The engineering apparatus 1 acquires IW and OW data from the programmable controller 4 at regular intervals based on the definition set on the automatic clustering definition screen. In this case, IWxxxx is arbitrarily assigned as data input, data acquisition is started by a data acquisition trigger specified in advance, and data acquisition stop trigger is
Stop data acquisition. While the data is being acquired, the acquired data is stored in a hard disk device in the engineering apparatus 1. Next, the engineering device 1
Performs online data analysis and control rule extraction only when conditions are satisfied, based on stored data. At this time, data analysis and control rule extraction are performed based on the settings on the automatic clustering selection screen. Then, after analyzing the data and extracting the control rules, the extracted control rules are applied to the water purification system. The current control rule extraction application status can be monitored on the automatic rule extraction and application screen. For example, when the automatic clustering trend graph is selected, the result of the two-axis clustering is displayed as a graph (see FIG. 5). The portion surrounded by the line represents a cluster, and shows the relationship between the water temperature and the water flow when bacteria have abnormally occurred. Through the above operations, the operator can easily analyze an event for which a control rule is to be extracted and check characteristics of the event. As described above, in this embodiment, data is collected during the operation of the water purification system, and data that satisfies a predetermined condition is analyzed to extract a control rule. Then, the extracted control rules are applied to the operating water purification system. Therefore, there is no need for other computers or workers,
Significant labor savings are possible. In addition, since control rules are automatically extracted and applied to the water purification system, the ability to apply to changes in the system environment and disturbances can be improved, and control can be performed even when modeling of the control target is incomplete. In addition, control rules can be extracted and applied in real time. Further, there is no need to stop the system in order to apply the control rules to the system, so that the operating rate of the system does not decrease. In the above embodiment, the engineering device 1 is applied to the water purification system. However, not only the water purification system but also the events for which the causal relationship is not understood are clarified as the causal relationship and characteristics, and the failure is defined as a failure. It can be applied to all applications for which prediction is made.

[0006]

As described above, according to the present invention,
In an engineering device for maintaining and monitoring the system, control rule analysis and extraction can be automatically applied online, so that other computers and operators are not required, and labor can be greatly reduced. In addition, since control rules are automatically extracted and applied to the system, the ability to apply to changes in the system environment and disturbances can be improved, and control can be performed even when modeling of the control target is incomplete. Furthermore, control rules can be extracted and applied in real time. Furthermore, since there is no need to stop the system in order to apply the control rules to the system, there is no reduction in the operating rate of the system.

[Brief description of the drawings]

FIG. 1 is a block diagram showing a configuration of an embodiment of an engineering device according to the present invention.

FIG. 2 is a block diagram showing a configuration of a water purification system using the engineering device of the embodiment.

FIG. 3 is a diagram showing an automatic clustering definition screen in the water purification system of the engineering device according to the embodiment.

FIG. 4 is a diagram showing an automatic clustering selection screen in the water purification system of the engineering device according to the embodiment.

FIG. 5 is a diagram showing an automatic rule extraction and application screen in the water purification system of the engineering device according to the embodiment.

FIG. 6 is a flowchart showing an operation of the engineering device according to the embodiment.

[Explanation of symbols]

 Reference Signs List 20 data storage means 21 data analysis means 22 rule extraction means 23 rule application means

Claims (1)

[Claims] 1. An engineering apparatus for monitoring and maintaining a system device, comprising: a data storage unit for collecting and storing specified data while the system device is operating; Data analysis means for analyzing the condition if the following conditions are satisfied; a rule extraction means for extracting a control rule from an analysis result by the data analysis means; and a system operating the control rule extracted by the rule extraction means. An engineering apparatus comprising: a rule application unit applied to the apparatus.