Kernel PCA based faults diagnosis for wastewater treatment system

  • Authors:

  • Byong-Hee Jun;Jang-Hwan Park;Sang-Ill Lee;Myung-Geun Chun

  • Affiliations:

  • School of Fire & Disaster Prevention, Kangwon National University, Samcheok, Korea;School of Electrical and Electronic Engineering, Chungju National University, Korea;Dept. of Environmental Engineering, Chungbuk National University, Cheongju, Korea;Dept. of Electrical and Computer Engineering, Chungbuk National University, Korea

  • Venue:
  • ISNN'06 Proceedings of the Third international conference on Advances in Neural Networks - Volume Part III
  • Year:
  • 2006

Quantified Score

Hi-index 0.00

Visualization

Abstract

A Kernel PCA based fault diagnosis system for biological reaction in full-scale wastewater treatment plant was proposed using only common bio-chemical sensors such as ORP (Oxidation-Reduction Potential) and DO (Dissolved Oxygen). SBR (Sequencing Batch Reactor) is one of the most general sewage/wastewater treatment processes and, particularly, has an advantage in high concentration wastewater treatment like sewage wastewater. During the SBR operation, the operation status could be divided into normal status and abnormal status such as controller malfunction, influent disturbance and instrumental trouble. For the classification and diagnosis of these statuses, a series of preprocessing, dimension reduction using PCA, LDA, K-PCA and feature reduction was performed. Also, raw data obtained from SBR were transformed to synthetic data or fusion data and the performance were compared with each other. As the results, the fault recognition rate using fusion data showed the better result than that of raw data of [ORP] or [DO] and the combination method of K-PCA with LDA was superior to other methods such as PCA and LDA.