A Review on Data Reconciliation and Gross Error Detection for Process Plant Energy Management

Abstract

Abstract Process plant is essential for energy management, especially for analysis that requires steady state data, such as Pinch Analysis. The data from the distributed control system (DCS) often recorded with fluctuations. Data processing is needed to obtain representable data for energy management related analysis. Data error occurs in several forms such as gross error, random error, bias error, and systemic error. This error affects the reliability of energy management studies. Data reconciliation (DR) and gross error detection (GED) come in place for the data processing required before the energy management analysis could be done. GED detects measurement variable error which works with DR technique for a new estimate or reconciled value will be gather and use for further data analysis. DR and GED are commonly used in mathematical programming optimization, such as model predictive control (MPC), which has proved its effectiveness in providing good data for further analysis. This means that DR and GED affect energy data manipulation and studies. In this paper, DR and GED on energy data for plant energy enhancement are reviewed. The specific method for DR and GED are classified and discussed in this paper.