Title |
A Study on Verification of Sewage Treatment Expert Operating Experience Using RNN |
DOI |
https://doi.org/10.5370/KIEE.2025.74.4.691 |
Keywords |
RNN; Expert Operation Experience; Wastewater Treatment System; Data Quality and Diversity |
Abstract |
In this paper, we proposed a data quality and diversity guarantee plan and a method to verify the expert operation experience using RNN and a data quality and diversity guarantee plan to 1) solve the two problems of the sewage treatment expert driving experience and 2) the limitation of situation recognition and judgment. As a solution to the lack of data quality and diversity, not all data are used, but a method of filtering and using only the part that is the basis for determining the expert's experience was proposed. In addition, RNN was used to solve the problem of the limitation of situation recognition and judgment. The reason why RNN is used among various neural networks is that RNN is a neural network architecture for deep learning that predicts time series or sequential data, and sewage treatment data is also time series or sequential data. According to the proposed technique, a RNN model with a reliability of more than 8% was implemented, and the effectiveness of the proposed technique was verified by applying two experiences and one incorrect experience of actual operation experts of the target sewage treatment plant. As a result of the verification, both the actual driving expert's experiences 1 and 2 were verified as valid, and for incorrect experiences, there were 3,442 data that matched the expert's experience and 4,444 data that did not match, so the proposed method can verify the expert's experience. |