JPS615314A - Backup method of process operation - Google Patents

Abstract

PURPOSE:To enable an operator of the process operation by making use of the knowledge of a process specialist, obtaining the data useful to the operation through the interactive operations with a calculation means, and referring the data operation. CONSTITUTION:An operator 110 uses a knowledge base system 100 to set 116 the operation conditions of a water purifying process 114. The on-line data obtained from the process 114 is fetched by a computer via a data reading part 118 and a data transmission means, etc., stored as the data base 120, and used suitably at a inference part 104 of the system 100. The system 100 includes the knowledge of a specialist 106 in the operation of a water purifying plant. The knowledge base 102 undergoes proper additions and corrections when necessary through a data collection/correction system 108.

Description

DETAILED DESCRIPTION OF THE INVENTION [Technical Field of the Invention] The present invention relates to, for example, a process operation support method for stably maintaining a water treatment process in a water purification plant.

[Prior art]

In general, water treatment plants deal with suspended matter such as sediment contained in raw water taken from rivers and lakes, microorganisms such as algae and bacteria, off-flavors and tastes generated by these microorganisms, and polluted organic matter derived from humus and sewage. .

Water treatment is performed to separate or decompose various contaminants such as ammonia nitrogen from sewage and iron ions and manganese ions eluted from bottom mud.

Water treatment processes at water treatment plants include oxidation sterilization, coagulation sedimentation, and filtration. In a typical water purification process, pre-chlorine is first injected into the water near the landing well for purposes such as sterilization and oxidative decomposition of reducing inorganic ions.

Next is injected into the water. Next, in the subsequent floc formation pond, coarsening of the coagulated flocs is promoted, and these flocs take in suspended matter and settle in the sedimentation basin, whereupon the suspended matter is separated into solid and liquid. In the subsequent rapid filtration basin, small amounts of fine particles that did not settle in the settling basin are captured in the sand layer, and solids are almost completely removed from the water. At the subsequent stage, post-chlorine is injected to maintain the disinfection effect of the tap water in the delivery water. In addition to the above-mentioned processes, biological oxidation treatment, ozone treatment, and activated carbon treatment may be performed as necessary.

In the water treatment process mentioned above, it is necessary to maintain appropriate injection rates of pre-chlorine, coagulant, alkaline agent, ozone, powdered activated carbon, etc., and to set the operating schedule of the offshore pond, such as setting the filtration duration. It is necessary to determine the conditions. There are two main ways to drive it:
The first is a method of online control using a computer, and the second is a method of online control using a computer.
is a method in which plant operators manually change set values while taking into account water quality measurements and other conditions. In the first method of direct computer control, the number of water quality sensors that can be measured online is limited, so a simple control system can perform good control, but if the process is complex, modeling becomes difficult. . If this is forcibly converted into a control equation and turned into a black box, even if the control results go bad, the operator on site will be unable to do anything, leading to a situation where distrust is only amplified. In addition, in the second manual operation method, a skilled operator can perform optimal plant operation based on years of experience and intuition, but other operators cannot utilize that experience and intuition. .

As described above, the conventional direct control method using a computer has the disadvantage that it cannot deal with phenomena that cannot be measured by online sensors. In addition, the method of manual operation by a skilled operator has the disadvantage that the operator's years of experience accumulated in his or her brain cannot be stored in a form that can be used by other operators.

[Summary of the invention]

The present invention has been made in view of the above points, and allows an operator who operates a process to utilize the knowledge of an expert in the process, and can easily use it as a reference for operation through interactive operation with the calculation means. The purpose of this is to provide a process operation support method that enables the acquisition of data.

In other words, the present invention calculates various problems occurring in the target process, rules indicating matters to be determined in solving the problem, and answers for solving the problem in a chain based on the knowledge of experts. The objective is to be achieved through a process operation support method in which the operator makes decisions regarding the rules, finds an answer to solve the problem, and uses this as a reference to operate the process. That is.

[Embodiments of the invention]

DESCRIPTION OF THE PREFERRED EMBODIMENTS The process operation support method according to the present invention will be described in detail below based on embodiments shown in the accompanying drawings.

FIG. 1 shows an embodiment of the process operation support method according to the present invention. In the system configuration shown in this figure, the central knowledge base system (100) is:
It is composed of a knowledge base (102) and an inference section (104). These are constructed with the knowledge of seasoned experts (106) who have many years of experience and intuition for water purification plant operations.

Further, the knowledge base (102) is adapted to be appropriately added to and modified as necessary by a data collection and modification system (108).

The plant operator (110) accesses the knowledge base system (110) via the man-machine interface (112).
100) can be used at any time. The operator (110) uses the knowledge base system (100) to set operating conditions for the water purification process (114) (see (116) in FIG. 1). On the other hand, the online data obtained from the water purification process (114) is imported into the computer by the data reading unit (118) via a data transmission means, etc., stored as a database (120), and sent to the knowledge base system (100) as appropriate. Reasoning part (10
4).

Next, Tables 1 to 3 illustrate the phenomena (hereinafter simply referred to as "problems") to be solved in this implementation. Of these, Table 1 shows a list of problems in water treatment systems.

Table 1 and Table 2 list problems in sludge treatment systems, and Table 3 lists problems in water flow systems.

As shown in Table 2, Table 3, and Tables 1 and 3, the problem areas are broadly divided into water treatment systems, sludge treatment systems, and water flow systems, and each system is further divided into unit processes. There is. Examples of problems included in each unit process are shown along with problem numbers.

The rules for dealing with the problems exemplified above are:
For example, the answers shown in Table 4 are prepared, and the answers shown in Table 5 are prepared, for example. These rules and answers are given by question number (A) in Table 1.
31) [P resistance arrival time is fast. ” is the problem.

Other problems are similarly prepared in the knowledge base system (100) based on the knowledge of experts.

Table 4 Whistle Table 4 (Continued) Table 5 Table 5 (Continued) Next, with reference to FIG. 2, the specific operation of the above embodiment will be explained. Figure 2 shows the problem number (A3) mentioned above.
1) (See Table 1) This shows an example of the problem solving rules. The numbers shown in this figure indicate the rules or answers shown in Table 4 or Table 5. In addition, "Y" in the figure indicates affirmation (YES) to the rules in Table 4.
``NJ'' represents a negative (NO).

As mentioned above, the water treatment process at a water treatment plant consists of several unit processes, which are closely related to each other and constitute the entire water treatment system. The ice quality of raw water taken from rivers etc. can vary greatly depending on weather conditions and changes in water quality in the basin2, and it is necessary to constantly and appropriately modify the various chemical injection membranes in response to this. However, determining these plant operating conditions is not always easy.

For example: If the concentration of polluted organic matter in raw water has increased, the operating conditions can be corrected by increasing the flocculant injection rate, changing the general setting of the settling tank, or changing the ozone There are several options, such as injecting powdered activated carbon, powdered activated carbon, etc. It is extremely difficult to make these selections using online control by a computer, since the types of measurement sensors are limited. In order to solve such complex problems, it is necessary to find a comprehensive solution that includes not only water quality measurement values, but also the operator's judgment of the situation and the water treatment plant's operation method. The above-mentioned situation of "the concentration of polluted organic matter in raw water has increased" will be explained below. There are several ways to deal with this situation, but the operator is making various judgments about the situation in his head. For example, ``I have a large amount of powdered activated carbon in stock, so I would like to use activated carbon in preference to other chemicals.''

``Powdered activated carbon is expensive, so I don't want to use it when the pollutant concentration is not very high.'' ``The flocculant injection rate has reached its upper limit, so I would like to use some other chemical.'' ``The odor concentration in the raw water is high.'' Therefore, I would like to use ozone, which is also effective for this purpose.'', ``Injecting powdered activated carbon will increase the amount of sludge generated, so it is better to remove the sludge that has accumulated at the bottom of the settling tank in advance.'', ``Dissolved organic sludge The operator is thinking in his/her mind, ``Since there are a lot of substances, it will be difficult to remove them by agglomeration treatment,'' and ``I want to reduce the amount of truffle 10 methane produced as much as possible.''

In this embodiment, as described above, the thinking process based on situational judgment that conventionally performed in the brain of a skilled operator (110) is stored in the computer, and the knowledge to find solutions to various problems is acquired. m-base system (
100) Water treatment process of water treatment plant (114)
The purpose is to provide operational support.

As already mentioned, Tables 1 to 6 provide examples of some of the problems that may occur in the water treatment process (114) at water treatment plants, but the causes and causes of each problem are not known. The solution cannot be determined uniquely.

The solution rules for these problems are stored inside the calculation as the knowledge base (102) in Figure 1, so that they can be modified or rules added as needed due to changes in the operation policy of the water treatment plant. ing. Operator (110
) is a man-machine interface (112) such as a CRT.
) Through this, the inference unit (104) can be used to solve problems using the knowledge base (102). An example of problem solving in the case of FIG. 2, which is shown as the solution rule for problem number (A31) in Table 1, is shown below. Further, the relevant portion is shown in detail in FIG.
In FIGS. 2 and 3, the judgment of rule (0) is first made. If no abnormality is recognized in the quality of the filtered water compared to normal times, then the judgment according to rule (1) is made. Here, if it is recognized that there is no possibility of a measuring instrument error such as a water head loss needle, the judgment according to rule (3) is made. That is, it is determined whether or not there is a difference in blockage depending on the arrangement of the filter basins, and if there is no difference, the determination according to rule (7) is performed.

In rule (7), it is determined whether or not the settling tank outlet turbidity is lower than a set value, and if it is lower, the process moves to rule (8). In rule (8), it is determined whether the sedimentation tank - is within the range of 6.0 to ZO, and if it is within the range, the process moves to rule ffO+. In rule 0Cj, it is determined whether the raw water turbidity is significantly higher than normal, and if it is low, the flow moves to rule aυ. Rule 0υ determines whether or not algae are proliferating in large quantities at the intake water temperature, and if they are not proliferating, move to the old rule (rule 16).
), it is determined whether the concentration of polluted organic matter, ammonia nitrogen, etc. in the raw water is high, and if it is low, the rule (
17).

In rule OD, it is determined whether the value of -ALT ratio (flocculant injection rate/raw water turbidity) is large, and if it is large, the process moves to rule 081. In rule 118I, it is determined whether or not the aggregation aid can be injected, and if not, the process moves to rule a9. In rule a51, settling basin P
It is determined whether H is less than or equal to 6.7, and if it is less than or equal to 6.7, an answer (111) is obtained. This answer (111
), as shown in Table 5, ``In order to densify the flocs, an alkaline agent is injected to bring the precipitated Asahikawa to 6.7 to Z2.''

As above, answer (11) to question (A31)
1) is obtained. Similarly, for other problems, the operator (110) selects the problem number he wishes to solve, and then sequentially selects the problem solving rule options (YES or NO). A solution to the problem will be found. Based on the answers found, the operator (111) appropriately changes the operation of the water purification plan (114) or changes or corrects the set values (see FIG. 1 (116)).

As explained above, one operator (110) can obtain solutions to problems to be solved in the operation of a water treatment plant by interacting with a computer that has a built-in knowledge base system (IOO). By modifying the operating conditions based on this information, the water treatment process at the water purification plant can be maintained in an appropriate state at all times.

In addition, in the above-mentioned example, the water treatment process of a water purification plant was explained as an example, but the present invention is not limited to this in any way, and can be applied to other water treatment processes or processes of chemical plants. It's something you get.

〔Effect of the invention〕

As explained above, according to the process operation support method according to the present invention, instead of using a conventional control method using numerical methods, the process operation support method is based on a chain of questions, rules, and answers constructed based on the knowledge of experts. Since we decided to obtain the necessary data to solve problems in a fixed manner, we can easily utilize our specialized experience and intuition.

This has the advantage that the operator can properly and quickly solve any problems that arise.

[Brief explanation of drawings]

FIG. 1 shows a part of the process support method according to the present invention.
FIG. 3 is an explanatory diagram showing in detail a part of the procedure shown in the figure. In the figure, (100) is a knowledge base system, (10
8) is a data collection/correction system, (110) is an operator, (112) is a man-machine interface, (
114) is a water purification process, (is) is a data reading unit, and (120) is a database. In each figure, the numerals ``[-'' and ``-'' indicate the same or equivalent parts.

Claims (2)

[Claims] (1) Prepare in a computer means various problems that occur in the process to be operated, rules that indicate the judgments that should be made to deal with the problems, and answers to solve the problems, and prepare the rules and answers. Based on the knowledge of experts in the process, the operator makes a judgment on the rules according to the chain of rules and answers that correspond to the problem that has arisen, and thereby finds the necessary answer. A process operation support method characterized by: (2) The process operation support method according to claim 1, wherein the questions, rules, and answers are changed as necessary.