QUALITATIVE INTERPRETATION AND COMPRESSION OF PROCESS DATA USING CLUSTERING METHOD

This paper presents a new qualitative data interpretation and data com pression method, which is based on modified adaptive k-means clusterin g algorithm. Conventional qualitative data interpretation methods that are based on control charts, such as Shewart, CUSUM, and EWMA control charts, are focused upon detection of changes from steady state value , so they are not suitable for describing transient or dynamic behavio r. But the proposed method continuously updates its detection limit, o r center of duster, so it can handle transient or dynamic behavior. Th us it can be applied to fault diagnosis of chemical process by combini ng with cause-effect digraph model, RCED(Reduced Cause Effect Digraph) . The usefulness of the proposed data interpretation method and cause- effect digraph model are illustrated using their application to the wa ter supply unit of a utility boiler plant. The proposed data interpret ation method can be used for not only change detection but also data c ompression. As the proposed method can store the value that minimizes the SSE between the retrieved data and the original data, it shows bet ter data compression result compared to other conventional data compre ssion methods such as Box Car, Backward Slope, Combined Box Car and Ba ckward Slope algorithm and Swinging Door Trending. (C) 1998 Elsevier S cience Ltd. All rights reserved.