Fault detection

The healthy measure, obtained via the diagnostic system described above, is used to feed a bank of observers providing process/actuator fault detection and isolation. One observer detects the occurrence of an actuator or process fault, while the other Np observers, each one corresponding to a fault type, are used for isolation and identification. 

Moreover, in the absence of faults and in the presence of uncertainties and disturbances, a bound on the output estimation error can be found in a similar way as in Sect. 6.4, i.e., 

A fault is declared when the norm of the residual vector 

In the presence of a fault occurring at t = tf, the state estimation error can be expressed as 

a sufficient condition for correct detection of the fault is given by 

---

MaintainabiUty is a characteristic of design, installation, and operation, usually expressed as the probabiUty that a system can be restored to specified operable conditions within a specified interval of time when maintenance is performed in accordance with prescribed procedures. The ease of fault detection, isolation, and repair are all influenced by system design and are principal factors contributing to maintainabiUty. Also contributing is the supply of spare parts, the supporting repair organization, and preventative maintenance practices. MaintainabiUty must be designed into the equipment. Some factors to consider foUow. 

Serviceability. ServiceabiUty is defined as the degree of ease (or difficulty) with which a system can be repaired. This measure specifically considers fault detection, isolation, and repair. RepairabiUty considers only the actual repair time, and is defined as the probabiUty that a failed system is restored to operation in a specified interval of active repair time. Access covers, plug-in modules, or other features to allow easy removal and replacement of failed components improve the repairabihty and serviceabihty (see also Electrical connectors). 

J. R. Whiteley andj. F. Davis, "QuaHtative Interpretation of Sensor Patterns using a Similarity-Based Approach," paper presented at the IFAC Symposium on On-Eine Fault Detection and Supervision in the Chemical Process Industries, Newark, Del., Apr. 1992. 

Isermann R., Process Fault Detection Based on Modeling and Estimation Methods—A Survey, Automatica, 20(4), 1984, 387 04 (Fault detection survey article)... 

Narashimhan, S., R.S.H. Mah, A.C. Tamhane, J.W. Woodward, and J.C. Hale, A Composite Statistical Test for Detecting Changes of Steady States, AlChE Journal, 32(9), 1986, 1409-1418. (Fault detection, steady-state change)... 

Wei, C.N., Diagnose Process Problems, Chemical Engineeiing Piogiess, September 1991, 70-74. (Parameter estimate monitoring for fault detection)... 

The three vertices are the operating plant, the plant data, and the plant model. The plant produces a product. The data and their uncertainties provide the histoiy of plant operation. The model along with values of the model parameters can be used for troubleshooting, fault detection, design, and/or plant control. 

These measurements with their inherent errors are the bases for numerous fault detection, control, and operating and design decisions. The random and systematic errors corrupt the decisions, amplifying their uncertainty and, in some cases, resulting in substantially wrong decisions. 

History The histoiy of a plant forms the basis for fault detection. Fault detection is a monitoring activity to identify deteriorating operations, such as deteriorating instrument readings, catalyst usage, and energy performance. The plant data form a database of historical performance that can be used to identify problems as they form. Monitoring of the measurements and estimated model parameters are typic fault-detection activities. 

Design In this context, design embodies all aspec ts requiring a model of the plant operations. Examples can include troubleshooting, fault detection, control corrections, and design development. 

Focus For the purposes of this discussion, a model is a mathematical representation of the unit. The purpose of the model is to tie operating specifications and unit input to the products. A model can be used for troubleshooting, fault detection, control, and design. Development and refinement of the unit model is one of the principal results of analysis of plant performance. There are two broad model classifications. 

Intended Use The intended use of the model sets the sophistication required. Relational models are adequate for control within narrow bands of setpoints. Physical models are reqiiired for fault detection and design. Even when relational models are used, they are frequently developed bv repeated simulations using physical models. Further, artificial neural-network models used in analysis of plant performance including gross error detection are in their infancy. Readers are referred to the work of Himmelblau for these developments. [For example, see Terry and Himmelblau (1993) cited in the reference list.] Process simulators are in wide use and readily available to engineers. Consequently, the emphasis of this section is to develop a pre-liminaiy physical model representing the unit. 

Parameter estimation is a procedure for taking the unit measurements and reducing them to a set of parameters for a physical (or, in some cases, relational) mathematical model of the unit. Statistical interpretation tempered with engineering judgment is required to arrive at realistic parameter estimates. Parameter estimation can be an integral part of fault detection and model discrimination. 

Fault detection is a monitoring procedure intended to identify deteriorating unit performance. The unit can be monitored by focusing on values of important unit measurements or on values of model parameters. Step changes or drift in these values are used to identify that a fault (deteriorated performance in unit functioning or effectiveness) has occurred in the unit. Fault detection should be an ongoing procedure for unit monitoring. However, it is also used to compare performance from one formal unit test to another. 

Parameter Estimation Relational and physical models require adjustable parameters to match the predicted output (e.g., distillate composition, tower profiles, and reactor conversions) to the operating specifications (e.g., distillation material and energy balance) and the unit input, feed compositions, conditions, and flows. The physical-model adjustable parameters bear a loose tie to theory with the limitations discussed in previous sections. The relational models have no tie to theory or the internal equipment processes. The purpose of this interpretation procedure is to develop estimates for these parameters. It is these parameters hnked with the model that provide a mathematical representation of the unit that can be used in fault detection, control, and design. 

These observations lead to the principal questions toward which fault detection is addressed. 

A fault may interfere with the effectiveness or the func tioning of the unit (Watanabe, K., and D.M. Himmelblau, Incipient Fault Diagnosis of Nonhnear Processes with Multiple Causes of Faults, Chemical Engineering Science, 39(3), 1984, 491-508). The first question addresses the effectiveness. The second two address the functioning. Fault detection is a unit monitoring activity, done automatically or periodically, to determine whether the unit operation has changed. 

The pumose of fault detection is to interpret the set of measurements to determine whether the operation of the unit has changed. This interpretation is done by monitoring the set of the measurements or by monitoring values for the significant unit parameters. It is done automatically as part of the computer control of the unit or periodically as when comparing one unit test to a subsequent one. 

Periodic fault detection is readily done by analysts without extensive software support. Process monitoring such as the examination of the traces discussed above are one exam e. However, the number of measurements in a single set have such complex interactions that it is... 

In Fig. 30-25, representation of the fault detection monitoring activity, there appears to be two distinct time periods of unit operation with a transition period between the two. The mean parameter value and corresponding sample standard deviation can be calculated for each time. These means can be tested by setting the null hypothesis that the means are the same and performing the appropriate t-test. Rejecting the null hypothesis indicates that there may have been a shift in operation of the unit. Diagnosis (troubleshooting) is the next step. 

FIG. 30-25 Trend in model parameter developed during fault detection parameter estimation. 

Relays have inputs from several current transformers (CTs) and the zone of protection is bounded by these CTs. While the CTs provide the ability to detect a fault inside the zone, circuit breakers (CBs) provide the ability to isolate the fault by disconnecting all of the power equipment within the zone. Thus, a zone bonndai y is usually defined by a CT and a CB. When the CT is part of the CB it becomes a natural zone boundaiy. When the CT is not an integral part of the CB, special attention must be paid to the fault detection and fault interruption logic. The CT still defines the zone of protection, but communication channels must be used to implement the tripping function. Figure 1 shows the zones of protection in a typical system. 

Vaidyanathan, R., and Venkatasubramanian, V., Process fault detection and diagnosis using neural networks II. Dynamic processes. AIChE Ann. Meet., Chicago, IL (1990). 

This chapter provides a complementary perspective to that provided by Kramer and Mah (1994). Whereas they emphasize the statistical aspects of the three primary process monitoring tasks, data rectification, fault detection, and fault diagnosis, we focus on the theory, development, and performance of approaches that combine data analysis and data interpretation into an automated mechanism via feature extraction and label assignment. 

Iserman, R., Process fault detection based on modeling and estimation methods—a survey, Autonuttica 20, 387-404 (1984). 

Ulerich, N. H., and Powers, G. J., On-Line hazard aversion and fault detection Multiple loop control example. AIChE Fall National Meeting, New York, 1987. 

A very important issue related to the performance of the whole wind-hydrogen system is the quality of its electrical equipment. There is a strong dependence on the reliability of the power electronics used for the AC/DC conversion as well as for the sensitivity of the fault detection devices that often result in a lower overall efficiency rate. 

Soft sensors Fault detection Data reconciliation Statistical analysis Parameter estimation... 

Fault Detection-Identification in Dynamic Processes Problem Formulation... 

Finally, the selection of the critical value for the test is a trade-off between the sensitivity to fault detection and the probability of giving a false alarm. 

One of the disadvantages of replacing fault detection tests on components of the innovation vector by a test on the vector itself is that there is no longer a simple indication of the origin of the fault. In order to solve this problem the following procedure was implemented. 

---