Studies On Data Reconciliation Methods For Chemical Process

Reliable chemical process data is the foundation of efficient process control, operation and optimization. However, data in chemical process is inherently inaccurate. Data reconciliation technology is one of the main means to acquire high quality data. Its aim is to improve the reliability, precision and consistency of data for process industries. Many works have been done in this field and much progress has been made in the last few years. But there are still many problems to solve in date reconciliation areas.In this paper some main problems, such as steady bilinear data reconciliation, the integral method for dynamic linear data reconciliation and the method based on finite-element orthogonal collocation for nonlinear dynamic data reconciliation are studied in detail. The main contents are as follows:1. Studies on the techniques developed for steady bilinear data reconciliation problems. Among the techniques, the method based on independent flows is well known in terms of both accuracy and efficiency. However, the independent flow method has some limitation. In this paper, the principle of the independent flow method is studied first. Then two data reconciliation methods based on the independent flows technique are proposed to overcome the limitation of the independent flow method. The first method, for process with no reaction unit, uses the relationships between independent flows and independent component flows to handle the normalization constraints on variables. The second method, by adding fictitious stream at reaction nodes and using the reaction constraints, can account for the reaction units in chemical process efficiently. To demonstrate and verify the proposed methods two examples are presented. The...