Robust model‐based fault detection and isolation for nonlinear processes using sliding modes

Abstract

This paper proposes a robust fault detection and isolation system for nonlinear processes that can be formulated as differential algebraic equations. For open-loop stable or closed-loop stabilized systems that operate under strict nonlinear detectability conditions, a methodology to design a nonlinear state estimator based on sliding mode theory was proposed. The extended observer can handle both parameter estimation and parameters with uncertainties. As a result, the state estimator is able to follow the faulty system, detecting faults by examining changes in the controlled outputs with respect to setpoint and then probing variations in the parameters estimated. Once the fault has occurred, the isolation mechanism uses the information provided from the state estimator, generated from recovery actions in the presence of a fault. These differences from normal operation trends can be derived through statistical analysis and then can be used to identify faults. A steam generator system was used to validate this approach, where process faults were considered. The proposed robust fault detection and isolation method shows significant advantages when applied to nonlinear model systems with parameter uncertainties or with complex nonlinearities. The complex nonlinearities can be simplified with algebraic nonlinear functions that have bounded uncertainties in their parameters.can be applied to nonlinear chemical processes by using a state estimator capable of dealing with both bounded uncertainties and estimation of the model parameters.There are multiple alternatives that can be used for solving the fault detection and isolation (FDI) problem, including model-based [6][7][8], data-driven [9, 10] or combined approaches [11]. A successful model-based fault isolation technique depends on a sufficiently accurate model of the system. Uncertainties in the model lead to false alarms or failures in detection. These issues make applying model-based techniques to nonlinear processes difficult. The majority of chemical processes have many uncertainties in their parameters due to their complexity. Research that addresses this problem consists of designing observers capable of parameter estimation [12]. However, restrictions surrounding the observability of these parameters limit this approach. Another alternative is represented by the design of RFDI techniques that deal with parameter uncertainty and are based on the sliding mode concepts, in which sliding observers are formulated for restrictive nonlinear systems. Examples of these FDI techniques based on residual generation, defined as the difference between the measured signals and the variables estimated, can be found in [13][14][15]. Also, linear sliding observers are used in [16] by transforming the nonlinear model. A disadvantage of FDI model-based techniques lies in the inability of distinguishing or isolating faults when the complexity of the nonlinear model increases.Data-driven approaches have the ability to handle a large number of measurements by compressing them into ...

Design of Sliding Mode Controller For The Estimation Of Sensor/Actuator Fault With Unknown Input Observer For Satellite Control System.

“…where S + 1 is a generalized inverse of S 1 , i.e., (20) and letting X = PZ and Y = PK 0 , we can get the LMIs (39) and (40) directly.…”

Multiway PCA for Early Leak Detection in a Pipeline System of a Steam Boiler—Selected Case Studies

“…The data recorded from a real plant are then compared with the model outputs (using, e.g., the least-squares algorithm) to detect the leakage. Some approaches [13,14] employ state estimators and observers to detect faults by tracking the dynamics of errors in the presence of model uncertainties (e.g., caused by process faults).…”

A Modified Extended Kalman Filter for a Two-Antenna GPS/INS Vehicular Navigation System

“…In vehicular navigation application, relatively accurate PDFs of state estimate predictions and measurements are hardly generated by the filter itself due the frequently occurring measurement outliers without boundedness and periodicity. Hence, one of the main factors affecting filter performance is the controller and measurement outlier or fault detection, isolation, and reconfiguration [ 16 , 17 , 18 ]. To this end, a considerable portion of scholars have developed optimal adaptive Kalman filters (AKFs) to determine the most appropriate weight between and based on the maximum likelihood criterion where the Kalman gain factors for accuracy improvement are based on the treatment of variable error characteristics [ 19 , 20 , 21 ].…”