Sensor and actuator bias estimation using multi model approach. (January 2017)

• Record Type: Journal Article
• Authors: Manimozhi, M.; Saravanakumar, R.
• Is Part Of: Computers & electrical engineering. Volume 57(2017)
• Page Start: 118
• Page End: 133
• Publication Date: 2017-01
• Subjects: Kalman filter -- H– infinity filter -- Process monitoring -- Actuator bias -- Sensor bias -- Conical tank process -- CSTR process; Computer engineering -- Periodicals; Electrical engineering -- Periodicals; Electrical engineering -- Data processing -- Periodicals; Ordinateurs -- Conception et construction -- Périodiques; Électrotechnique -- Périodiques; Électrotechnique -- Informatique -- Périodiques; Computer engineering; Electrical engineering; Electrical engineering -- Data processing; Periodicals; Electronic journals; 621.302854
• DOI: 10.1016/j.compeleceng.2016.11.003
• Languages: English
• ISSNs: 0045-7906

Abstract

Highlights: Adaptive Linear Kalman Filter (ALKF) is designed to estimate non-linear process states. Multi Model Adaptive Linear Kalman Filter (MMALKF) is designed to detect multiple sensor and actuator faults. These faults may occur either sequentially or simultaneously. MMALKF estimated process states in the absence of unknown noise statistics and unmodeled dynamics. Multi Model Adaptive Linear H ∞ Filter (MMALH ∞ F) is designed for detecting multiple sensor and actuator faults in the presence of unknown noise statistics and unmodeled dynamics. MMALH ∞ F detects and diagnoses the simultaneous and sequential occurrence of sensor and actuator faults in the presence of unknown noise statistics and unmodeled dynamics. Abstract: The objective of this paper is first to design an Adaptive Linear Kalman Filter (ALKF) to estimate nonlinear process states and to compare the performance of the ALKF with the Extended Kalman Filter (EKF). The designed ALKF is next used to detect sensor and actuator biases which may occur either sequentially or simultaneously using a Multi Model ALKF (MMALKF). Finally the Multi Model Adaptive Linear H ∞ Filter (MMALH ∞ F) is designed to increase the robustness of bias Detection in the presence of unknown noise statistics and unmodeled dynamics. The proposed estimator is demonstrated on the variable area tank process and Continuously Stirred Tank Reactor (CSTR) process to show its efficacy. Graphical abstract: Structure of the proposed MMALKF