Research On Data Reconciliation In Thermal Process And Its Application In Engineering

Reliable process data is the foundation of efficient process control , optimization of process circulate and calculate of process performance. However, in the fact, process measurements inevitable contain errors, include random errors and gross errors. By using the redundancy in process measurements, data reconciliation can be used to eliminate measurements with gross errors and reduce the effect of random errors, so it can enhance the reliability, the consistency and the integrality of the process data. In this thesis, based on the analyzing the present methods of data reconciliation, the problems in data reconciliation are studied, An approach for estimation of data reconciliation is proposed based on neural network and Time Series Analysis simulation results demonstrate the efficiency of the proposed method.