Qiyou Jiang

Gross error modeling and detection in plant linear dynamic reconciliation

This paper presents a method to identify and estimate gross errors in plant linear dynamic data reconciliation. An integral dynamic data reconciliation method presented in a previous paper (Bagajewicz and Jiang, 1997) is extended to allow multiple gross error estimation. The dynamic integral measurement test is extended to identify hold-up measurements as suspects of gross error. A series of theorems are used to show the equivalencies of gross errors and to discuss the issue of exact identification. A serial approach for gross error identification and estimation is then presented. Gross errors are identified without the need for measurement elimination. The strategy is capable of effectively identifying a large number of gross errors.

Simultaneous estimation of biases and leaks in process plants

In this paper, a recursive strategy is applied to identify gross errors (biases and leaks) and estimate their magnitude in linear steady state processes. A recursive search scheme is used first to isolate the candidate sources of gross errors and then simultaneous identification/estimation of gross errors is accomplished. A recently proposed equivalency theory is used to correctly assess results. Comparative studies of performance and accuracy of estimation are performed for some process networks when compared to existing techniques. Simulation results show that the proposed approach has higher performance to identify gross errors as well as to estimate their magnitudes.

Comparison of steady state and integral dynamic data reconciliation

This paper is devoted to the comparison of the performance of integral approach to dynamic data reconciliation and steady state data reconciliation. It is shown that in the absence of biases and leaks, the performance of both approaches is similar. Moreover, it is proven that once the appropriate variance is chosen, both methods are identical in the absence of accumulation terms. Finally, an analysis is made on how large the discrepancies are when there are accumulation terms.

REMOVING SINGULARITIES AND ASSESSING UNCERTAINTIES IN TWO EFFICIENT GROSS ERROR COLLECTIVE COMPENSATION METHODS

Abstract Two collective estimation strategies, the Unbiased Estimation Technique (Rollins and Davis, 1992) and the recursive Generalized Likelihood Ratio (Keller et at.,1994), have been shown lo be very efficient in detecting and estimating multiple gross errors in linear process systems. However, these strategies run into singularities and uncertainties that prevent them from being used in automatic schemes. This paper uses a recently presented theory on gross error equivalency to explain when and how these singularities and uncertainties take place. The procedures presented by these two methods are modified to prevent the singularities from appearing and allowing their automatic implementation.

Performance evaluation of PCA tests for multiple gross error identification

Abstract In this paper, the performance of the Principal Component Test, including both the Principal Component Nodal Test (PCNT) and Principal Component Measurement Test (PCMT), is evaluated when used for the identification of multiple gross errors. A few existing gross error identification techniques are modified to replace the nodal, global and measurement tests they use, by a principal component test. Comparative analysis indicates that PCA tests do not significantly enhance the ability in identification features of these strategies, performing worse in some cases.

PERFORMANCE EVALUATION OF PCA TESTS IN SERIAL ELIMINATION STRATEGIES FOR GROSS ERROR IDENTIFICATION

In this paper, the performance of the Principal Component Measurement Test (PCMT) is evaluated when used for the identification of multiple biases. A serial elimination strategy is implemented where a statistical test based on principal component analysis is used to identify the measurement to eliminate. A simulation procedure involving random measurement errors and fixed gross error sizes is applied to evaluate its performance. This performance is compared with the one obtained using serial elimination using the conventional Measurement Test (MT), as it is performed in some commercial simulators. The analysis indicates that principal component tests alone, without the aid of other collective tests, do not significantly enhance the ability in identification features of this strategy, performing worse in some cases. A few cases of severe failure of this strategy are shown and a suggestion to test other strategies is offered.

Serial Identification with Collective Estimation of Gross Errors and Leaks in Process Plant Measurement

Abstract In this paper we review a model for collective compensation, we present the theory of gross error equivalency and a serial identification with collective compensation method (SICC) to determine gross error in steady state data reconciliation. A comparison of this strategy with other collective estimation strategies is shown. Finally, a new method based on collective identification and collective compensation using a MILP model is presented.