In order to control filamentous bulking, it is very important and beneficial that plant operators can precisely identify filamentous microorganisms in the activated sludge, and that they utilize the information of the identification as an index of the operational conditions at the plant.
In this study, the computer aided system for the identification of the filamentous microorganisms was tried to develop by applying the knowledge engineering (or artificial intelligence ; AI) techniques. This knowledge-based system is designed to support operators at the plant for easy and precise identification of the filaments in their wastewater treatment plants.
The prototype system described in this paper has shown the reliability of being in agreement with human expert in approximately 67% of the cases investigated. It is expected that the reliability (or intelligent ability) can be improved by refining the knowledge base.
This study has given the instance to evaluate whether the knowledge engineering techniques is useful and promising in the field of wastewater treament.

View full abstract

For enumerating the sulfate-reducing bacteria, the most probable number method was recommended because of its reliability. The modified ISA medium (Mara and Williams, 1970) facilitated the early growth and metabolism of the bacteria, and then could shorten their incubation period. The sulfate-reducing bacteria in natural habitats, i.e., mixed culture with other heterotrophic bacteria, were able to enumerate easily by using the semi-liquid agar medium without reducing agents. Because the medium did not precipitate after being autoclaved, it was suitable for forming the distinct black colonies. Anaerobic pouches were available for the isolation as well as the enumeration.
The rate of bacterial sulfate reduction was determined with a radiotracer of ³⁵S. In the case of aquatic sediment, a whole core sample or the sub-sampled mud suspension was used for the incubation. Two types of procedure for recovering acid-soluble sulfides were introduced : a stripping method with an inert gas or an absorbing method with a filter paper. The latter, simple and rapid procedure was more suitable for routine works. These methods for the quantification of the bacterial number and activity, described herein, are considered to be practically used to research in the applied field, such as sewage treatment.

View full abstract

A step process containing nitrification and denitrification in a column packed soil was studied by using a synthetical wastewater contained NH₄Cl, NaHCO₃, and KH₂PO₄. An influent was infiltrated by down flow to the soil column, nitrification occured in upper layer which the depth was 20 cm from top of the soil column, and denitrification occured in lower layer. The water level in the column was constantly kept at the depth of 20 cm from top of the column.
When the influent contained methanol as organic carbon was infiltrated into the upper layer to be unsaturated (aerobic) zone, a little nitrification occured and the nitrification was suppressed rather than development, and also denitrification did not occure at the lower layer to be saturated (anaerobic) zone. In the case of no methanol in the influent, a complete nitrification occured at the upper layer. When the methanol solution was injected at the boundary between upper and lower layers, the denitrification was promoted and nitrogen removal was 60-80%.
The rates of nitrification and denitrification were determined by finite-difference procedure as an inverse problem for a reactive convectio-dispersive equation using measured concentration profiles of NO₃-N. It was found that the nitrification was carried out at upper layer from 0 to 10 cm depth and the denitrification at lower layer from 15 to 35cm.

View full abstract

[in Japanese]

View full abstract