JP2661162B2 - Water quality estimation method - Google Patents

Description

The present invention relates to a method for estimating the quality of treated water treated by, for example, an activated sludge method based on microbial data.

B. Summary of the Invention The present invention provides a method for estimating the quality of water to be measured based on microbial data contained in the water.Specifying, for example, a BOD concentration band corresponding to the water quality, the BOD concentration band in each BOD concentration band for each indicator microorganism. By defining membership values indicating the possibility of appearance in advance, picking up the membership value of each BOD concentration band for microorganisms that appeared in the water to be measured, and estimating the BOD concentration based on that value, It is intended to be able to express the treated water quality with specific numerical values while reflecting the knowledge in the prediction of the treated water quality.

C. Conventional technology The activated sludge method widely applied to the treatment of municipal sewage is a system that purifies sewage using the metabolic reaction of microorganisms. Here, the sewage treatment plant is regulated by its discharged water quality (treated water quality), and thus treated water quality is the most important indicator in process management. By the way, the treatment efficiency (water quality of treated water) of the system largely depends on microorganisms appearing in the system. Conventionally, operators at treatment plants have examined the frequency of appearance of microorganisms called indicator by microscopic observation,
Diagnosis of treatment plant condition has been performed. Long-term studies have shown which water quality each indicator microorganism is likely to appear in, and a skilled operator can predict rough treated water quality from the frequency of appearance of indicator microorganisms.

D. Problems to be solved by the Invention However, the knowledge linking the frequency of occurrence of such indicator microorganisms and the quality of treated water is a very vague concept. For example, certain microorganisms have been found to emerge when water quality is good. Here, "good water quality" is not based on concrete figures, but is vague and intuitive.
Human knowledge generally has many vague expressions as shown in the above example, but actual system management requires specific numerical values. The present invention provides a method for estimating water quality that can express treated water quality by specific numerical values while reflecting knowledge about microorganisms in the prediction of treated water quality, and that enables prediction of future water quality. Aim.

E. Means for Solving the Problems In order to achieve the above object, the present invention classifies the number of appearances of a plurality of types of indicator microorganisms serving as indicators of water quality per unit water amount into a plurality of appearance ranks, The oxygen demand, which is an indicator of the water quality of the measurement quality, is divided into a plurality of bands, and the maximum number that can appear in all the bands for each indicator microorganism is determined in advance, and each oxygen demand zone is determined for each indicator microorganism. The membership value indicating the likelihood of appearance is determined in advance by using the reciprocal of the maximum number of each of the indicator microorganisms that can appear as a factor, and the number of indicator microorganisms per unit water amount in the measured water is measured. From the plurality of appearance ranks for the indicator microorganisms, the appearance rank to which each belongs is determined, and the membership value of each oxygen demand band is picked up for the appeared indicator microorganisms, and this membership value is calculated. The membership values are weighted by the appearance rank to which the indicator microorganism belongs, and these membership values are summed up for each oxygen demand band to obtain the center of gravity of the membership value in the water quality direction. It is characterized in that it is estimated as the oxygen demand.

F. Embodiments The present invention focuses on the fact that fuzzy theory exists as a mathematical method for linking human knowledge (fuzzy expression) with concrete numerical values, and uses this fuzzy theory to, for example, activate sludge based on microbial data. The idea is to estimate the quality of treated water.

FIG. 1 is a diagram schematically showing an example of phage inference in the method of the present invention. This fuzzy inference is slightly different from ordinary fuzzy inference. First, indicator microorganisms serving as indicators of water quality are selected from the microorganism species in the treated water as shown in FIG. And the water quality is good,
In order to divide into "bad" and "between",
The oxygen demand of the treated water, for example, the BOD concentration (biochemical oxygen demand) is divided into three zones, and each concentration zone is associated with water quality. As indicator microorganisms, microorganism A,
Assuming B and C, as shown in FIG. 2, it is assumed that the microorganism A appears when the treated water quality is good and the microorganisms B and C appear when the treated water quality is bad.

Here, in the embodiment of the present invention, the number of microorganisms appearing per unit water amount of the treated water is classified into a plurality of appearance ranks, for example, six (0 to 5) as follows.

On the other hand, while grasping the maximum number that can appear for each microorganism from past data, a membership value indicating the possibility of occurrence in each BOD concentration band for each microorganism is determined in advance using the reciprocal of the maximum number as a factor. . The number of appearing microorganisms varies depending on the type. The maximum number when a certain microorganism is prioritized also differs depending on the type. ("Biology of wastewater treatment" (Sudo) Industrial Water Research Committee S.
Issued 52.6.12, "Diagnosis of treatment functions from the viewpoint of biota" (Sudo and Inamori), S.58.4.21 issued by the Industrial Water Research Committee. The microorganisms handled in this example are the 18 types shown in Fig. 3, Even with these types, the range of the number of appearances is wide. For example,
Assuming that the maximum number of microorganisms per unit water volume of a and b is 10,000 and 500, respectively, and as a result of the measurement, even if 100 appear in both a and b, a and b The weights are different. That is, the water quality cannot be evaluated only by the number of occurrences. Therefore,
In the present invention, the maximum number of each microorganism is determined with reference to past data and the like, and the reciprocal thereof is used as a factor of the membership value to express the weight as an index of water quality of one individual.

FIG. 1 is an explanatory diagram showing the treated water quality (BOD concentration (mg / L)) on the horizontal axis and the membership value on the vertical axis.
For example, in the range of water quality (BOD) assumed that the possibility of the appearance of microorganism A is good, this value is always 1 / (this value is 1 /
(Defined by the maximum number of microorganisms A), but is always zero in other ranges. The probability of appearance of microorganisms B and C is always 1 / maximum number of microorganisms in a water quality range assumed to be bad, but is always zero in other water quality ranges. Although the membership value is often defined by a triangle or a trapezoid, a rectangular value is used here for simplicity.

Now, as a result of observing the treated water, assuming that microorganisms A, B, and C are contained, the number per unit water volume of each of microorganisms A, B, and C was measured, and the results are shown in the above table for each microorganism. The appearance rank to which the user belongs is determined from the appearance ranks.
For example, the number of microorganisms A, B, and C (per 1 mL) is 400 to 999, respectively.
Between, between 1 and 99 and between 100 and 399, the microorganism A,
The appearance ranks of B and C are 3, 1, and 2, respectively. Then, the membership value of each BOD concentration band is picked up for the appearing microorganisms (here, A, B, C).

These membership values are then modified by the rank of appearance of the organism. The rank of appearance of microorganisms is evaluated on a six-point scale from 0 to 5, but the membership value is modified by multiplying the rank of appearance by five. First
In the microorganisms A, B, and C in the figure, the membership values are 3/5, 1/5, since they appear at ranks 3, 1, and 2, respectively.
It will be multiplied by 2/5. The result of this modification is a rectangle indicated by hatching in the figure. The purpose of these series of operations is to weight each microorganism. Therefore, the greater the tendency that the maximum number of each microorganism that can appear is small and that the appearance rank is large, the greater the certainty of the water quality indicated by the microorganism.

As a next step, superposition (synthesis) of the previously corrected membership values (hatched portions) is performed for each BOD concentration band. This is thought to be simply the combination of the hatched portions in a graphical manner. The result is the figure at the bottom of FIG. 1 (hatched).

Finally, the inferred treated water quality is calculated from the combined membership value. For this, the center of gravity calculation is used. That is,
The center of gravity of the combined membership value in the horizontal axis direction (water quality) is obtained, and the BOD concentration at this position of the center of gravity is estimated as the treated water quality concentration. This ends the water quality inference by fuzzy inference.

In this embodiment, the indicator microorganism serving as an indicator of water quality is the third microorganism.
All 18 types in the figure are shown. Therefore, if all 18 microorganisms were observed in the water to be measured, the appearance rank was determined from the number of each occurrence, and the membership value was calculated using the reciprocal of the maximum number as a factor, and the values were summed. To find the center of gravity. Generally, there are few examples in which all microorganisms are observed. In many cases, only a few types are observed, and many types of microorganisms other than the types shown in FIG. 3 may appear. In this case, the membership value is calculated only for those appearing among the 18 types of indicator microorganisms, and the values are summed to obtain the center of gravity. Naturally, the indicator microorganism of the present invention is not limited to the 18 types shown in FIG. The 18 types of this embodiment exemplarily list those which are generally considered to be indicators of water quality.

An example using specific numerical values will be described below. The 18 types of indicator microorganisms shown in Fig. 3 are used for fuzzy inference of water quality. FIG. 2 shows the water quality of each microorganism. Data were taken from the sewage test method 1984.
In the figure, for example, it is considered that Tetrahymena appears between good-medium water quality and medium-bad water quality. Also, beetles and the like appear only when the water quality is good. In the example of FIG. 2, good water quality is defined as X1 to X2, intermediate water quality is defined as X2 to X3, and bad water quality is defined as X3 to X4. The values of these four points (X1, X2, X3, X4)
Biota data and corresponding treated water quality ("Biology of wastewater treatment" (Sudo), "Diagnosis of treatment functions from the viewpoint of biota")
(Sudo and Inamori) Using the optimization (simplex method) method, X1 = -7.1, X2 = 31.6, X3 = 51.
6, X4 = 63.8 mg BOD / L. Although the value of X1 is negative, this is inevitable because the water quality is calculated by the center of gravity calculation. For example, consider a case where only the rotifer has appeared. Then, the composite membership value is the membership value of the rotifer. When calculating the center of gravity with this, the center of gravity is the midpoint between X1 and X2. Therefore, even if X1 is minus, the center of gravity (estimated water quality) is 12.25 mgBOD / L.

G. Effects of the Invention According to the present invention, for example, BOD
Specify the concentration band, define in advance the membership value indicating the possibility of appearance in each BOD concentration band for each indicator microorganism, and pick up the membership value of each BOD concentration band for the microorganisms that appeared in the measured water. Since the BOD concentration is estimated based on the value, the water quality can be represented by specific numerical values while reflecting knowledge about microorganisms in the prediction of the water quality. In addition, the actual water quality appears as a result of microbial fluctuations in the system, and in the present invention, since the microbial fluctuations are used as data in view of the appearance of microorganisms,
Water quality fluctuations can be quickly detected, and future water quality can be predicted. On the other hand, in the method of monitoring only the actual water quality, when a change is observed, the response cannot catch up, and there is a case where it cannot be recovered.

[Brief description of the drawings]

FIG. 1 is a schematic diagram showing the concept of the method of the present invention, and FIGS. 2 and 3 are explanatory diagrams showing water quality ranges in which indicator microorganisms appear.

Claims (1)

(57) [Claims] 1. The method according to claim 1, wherein the number of appearances of a plurality of types of indicator microorganisms serving as an indicator of water quality per unit water amount is divided into a plurality of appearance ranks, and the oxygen demand as an indicator of the water quality of the measured water is divided into a plurality of bands. Then, for each indicator microorganism, the maximum number that can appear in the entire band is determined in advance, and for each indicator microorganism, a membership value indicating the appearance probability in each oxygen demand zone is determined by the indicator microorganism. With the reciprocal of the maximum number of occurrences being predetermined as a factor, the number of indicator microorganisms per unit water amount in the measured water is measured, and each indicator microorganism belongs to each of the plurality of occurrence ranks. Find the appearance rank, pick up the membership value of each oxygen demand band for the appeared indicator microorganism, and weight this membership value by the appearance rank to which the indicator microorganism belongs. And summing up these membership values for each oxygen demand band to determine the center of gravity of the membership value in the water quality direction, and estimating the oxygen demand at the center of gravity as the oxygen demand of the measured water. Water quality estimation method.