Dominik Füssel

Integrated control, diagnosis and reconfiguration of a heat exchanger

Heat exchangers play an important role in chemical and process industries. In order to improve reliability and control performance, intelligent concepts for control, supervision and reconfiguration are necessary. In the paper, an approach is presented which integrates model-based adaptive control and reconfiguration based on fault detection/diagnosis applied to a heat exchanger plant. The adaptive controller and the fault detection scheme are based on a fuzzy model of the process (Takagi-Sugeno type). The fault diagnosis is performed using a self-organizing fuzzy structure.

Closed loop fault diagnosis based on a nonlinear process model and automatic fuzzy rule generation

Abstract In this contribution a new approach for fault detection and diagnosis (FDD) for nonlinear processes is presented. A nonlinear fuzzy model with transparent inner structure is used for the generation of relevant symptoms. The resulting symptom patterns are classified with a new self-learning classification structure based on fuzzy rules. The approach is successfully applied to a electro-pneumatic valve in a closed control loop.

Component-based multi-model approach for fault detection and diagnosis of a centrifugal pump

A model-based approach for fault detection and diagnosis of nonlinear processes is presented. However, the supervision of nonlinear systems is often very difficult in view of the lack of accurate models. Neuro-fuzzy models may help to cope with this problem since they can be trained from measured data. In this paper the application of a multi-model approach for fault detection and diagnosis of centrifugal pumps is presented. For this purpose the process is decomposed in several sub-processes. The supervision scheme allows the detection of several faults both in the hydraulic and mechanical subsystems.

Supervision, Fault-Detection and Fault-Diagnosis Methods

The operation of technical processes requires increasingly advanced supervision and fault diagnosis to improve reliability, safety and economy. This contribution describes advanced methods of fault detection and diagnosis. It begins with the consideration of a knowledge-based procedure, which is based on analytical and heuristic information. Then different methods of fault detection are considered, which extract features from measured signals and use process and signal models. These methods are based on parameter estimation, state estimation and parity equations. By comparison with the normal behaviour, analytic symptoms are generated. Human operators may be a further source of information and support the generation of heuristic symptoms. For fault diagnosis, all symptoms have to be processed in order to determine possible faults. This can be performed by classification methods or approximate reasoning, using probabilistic or possibilistic (fuzzy) approaches based on if-then rules. The application of these methods is shown for fault detection and diagnosis of a machine tool drive and a d.c.motor. Emphasis is given to the application of fuzzy logic in various parts of the diagnosis system.

Combining neuro-fuzzy and machine learning for fault diagnosis of a DC motor

An approach for the diagnosis of faults in dynamic systems based on a neuro-fuzzy scheme is presented. The simple structure that represents fuzzy rules in a neural network uses a rule extraction mechanism varying from most other approaches as it is based on concepts of machine learning. An additional, straightforward optimization eventually enhances the performance of the diagnosis. The approach is especially designed for the needs of technical fault diagnosis using parity space, observer and parameter estimation techniques. It evaluates parameter as well as parity space residuals and other information from the faulty process. Priory knowledge can easily be included as rules due to the simple structure of the scheme. The approach is tested on an electrical DC motor test bench to which several different faults can be applied.

Hierarchical motor diagnosis utilizing structural knowledge and a self-learning neuro-fuzzy scheme

Fault diagnosis requires a classification system that can distinguish between different faults based on observed symptoms of the process under investigation. Since the fault symptom relationships are not always known beforehand, a system is needed which can be learned from experimental or simulated data. A fuzzy logic based diagnosis is advantageous. It allows an easy incorporation of a-priori known rules and also enables the user to understand the inference of the system. In this contribution, a new diagnosis scheme is presented and applied to a DC motor. The approach is based on a combination of structural a-priori knowledge and measured data in order to create a hierarchical diagnosis system that can be adapted to different motors. Advantages of the system are its high degree of transparency and an increased robustness.

Generation of diagnostic trees by means of simplified process models and machine learning

Abstract Fault diagnosis by means of diagnostic trees is of considerable interest for industrial applications. The drawbacks of this approach are mostly related to the knowledge elicitation through laborious enumeration of the tree structure and ad hoc threshold selection for symptoms definition. These problems can be alleviated if a more profound knowledge of the process is brought into play. The main idea of the paper consists of modeling the nominal and faulty states of the plant by means of interval-like component models derived from first-principles laws, e.g. the conservation law. Such a model serves to simulate the entire system under different fault conditions, in order to obtain the representative patterns of measurable process quantities, i.e. training examples. To match these patterns by diagnostic rules, multistrategy machine learning is applied. As a result, binary decision trees that relate symptoms to faults are obtained, along with the thresholds defining the symptoms. This technique is applied to a laboratory test process operating in the steady state, and is shown to be suitable for handling incipient single faults. The proposed learning approach is compared with two related machine learning methods. It is found that it achieves similar classification accuracy with better transparency of the resulting diagnostic system.

Fault Detection for Nonlinear Processes Based on Local Linear Fuzzy Models in Parallel and Series-Parallel Mode

Abstract In this contribution, a new approach for model-based fault detection and isolation (FDI) of sensor faults for non linear processes is presented. A local linear fuzzy model of the process is used for the generation of structured residual equations similar to the parity space approach. The model is run in parallel and series-parallel to the process which leads to residuals with different sensitivities. The practical applicability is illustrated on an industrial scale thermal plant. Here, five different faults can be detected and isolated continuously over all ranges of operation.

Diagnose von Verbrennungsaussetzern in Ottomotoren durch Messung des Abgasdrucks

Zur zylinderselektiven Uberwachung der Arbeitsspiele von Ottomotoren wird der dynamische Verlauf des gemessenen Abgasdrucks analysiert. Es wird eine Methode beschrieben, die Verbrennungsaussetzer durch ein mehrstufiges Diagnosesystem erkennt und einzelnen Zylindern zuordnet. Dabei werden auch im Bereich hoher Drehzahlen und geringer Lasten bei vielzylindrigen Motoren gute Erkennungsraten erzielt. Die Untersuchungen wurden im Rahmen einer Forschungskooperation zwischen der TU-Darmstadt und dem Arbeitskreis 2 der deutschen Automobilhersteller durchgefuhrt