Sebastian Clever

Model-based fault detection and diagnosis with special input excitation applied to a modern diesel engine

Abstract Due to the rising complexity of many technical processes modern diagnosis systems have to supervise a multitude of hydraulic, mechanical, electromechanical and mechatronic components. Therefore model-based methods of fault-detection and diagnosis have been developed. These methods use mathematical process models to relate data of several measurable variables. Thus the diagnosis quality depends on the available sensor data. In order to obtain additional information with the given sensor configuration special input excitation signals can be used. This paper will describe a method to locate faults in multivariable systems using such input excitation and its application to the intake air system of a modern common rail Diesel engine. The presented method uses the knowledge of fault effects on the measured output, when the inputs are successively excited quasi-stationary, to determine the location of the fault. It has been applied successfully to differentiate air mass sensor faults from other process faults.

Model-based fault detection and diagnosis for common-rail injection systems

In order to guarantee the reliability and availability of electronically controlled multi injection systems of modern diesel engines, an improved fault detection and diagnosis gains in importance. For this purpose especially model-based algorithms with signal- and process models can be applied. Thus, Technische Universitat Darmstadt developed a model-based fault detection module for common-rail injection systems among other things in the FVV project “Diagnosis Systems for Diesel Engines” (No. 885 and 952).

Modellgestützte Fehlererkennung und Diagnose für Common-Rail-Einspritzsysteme

Moderne Pkw-Dieselmotoren sind mit verschiedensten komplexen mechatronischen Komponenten ausgestattet. So kommen neben Abgasruckfuhrsystemen zur Minderung der Stickoxidemissionen und Turboladern zur Steigerung der Leistungsdichte insbesondere elektronisch gesteuerte Mehrfach-Einspritzsysteme zum Einsatz. Deshalb erhoht sich die Bedeutung einer umfassenden Fehlererkennung und Diagnose, zusatzlich zu den bekannten OBD- und EOBD-Vorgaben, im Rahmen steigender Anforderungen an die Zuverlassigkeit und Verfugbarkeit. Um diese Anforderungen zu erfullen, bieten sich modellbasierte Verfahren an. Ihr Stand in verschiedenen Gebieten wird zum Beispiel in [1–6] beschrieben. Mit dem Einsatz modellgestutzter Fehlererkennungsverfahren bei Verbrennungsmotoren beschaftigen sich beispielsweise [7–18].

A Fault Diagnosis Toolbox Applying Classification and Inference Method

Abstract This paper presents a MatLab/SIMULINK based toolbox for fault diagnosis. It has been witnessed that while the fault detection methods must be tailored specifically to the process that is to be supervised, the fault diagnosis methods on the other hand are very similar in most applications. Therefore, a MatLab/SIMULINK-based toolbox was developed by the authors which shall be presented in this paper and is available for download. The software is designed such that it can be used with the Real-Time Workshop and can thus be compiled and downloaded to a wide range of rapid control prototyping system. Depending on the type and availability of a-priori knowledge, one can either employ classification or inference methods. Both approaches are supported by development environment. Classification is used whenever there are experimental data available which describe the influence of the faults on the symptoms. The available implementations encompass the Bayes classifier, the k-nearest neighbor and the polynomial classifier. Inference methods are used, whenever rules or expert knowledge describing the influence of the fault on the symptoms are available. In the paper, a Fuzzy-Logic based inference engine is presented, where the symptoms are first fuzzified to account for the uncertainty in the reaction of residuals. Then, the individual symptoms are combined using Fuzzy-Logic AND and OR operators respectively. The mapping of the fuzzy outputs to the diagnosed fault is accomplished by determining the maximum fault possibility among all fault possibilities. The different diagnostic engines have already successfully been applied to a wide range of prototype fault management realizations at the institute and have proven very capable.

Probabilistic color optical flow

Usually, optical flow computation is based on grayscale images and the brightness conservation assumption. Recently, some authors have investigated in transferring gradient-based grayscale optical flow methods to color images. These color optical flow methods are restricted to brightness and color conservation over time. In this paper, a correlation-based color optical flow method is presented that allows for brightness and color changes within an image sequence. Further on, the correlation results are used for a probabilistic evaluation that combines the velocity information gained from single color frames to a joint velocity estimate including all color frames. The resulting color optical flow is compared to other representative multi-frame color methods and standard single-frame grayscale methods.

Neural-Net Model Based Fault Diagnosis for Hydraulic Servo Axes

Abstract In this paper, fault detection methods for hydraulic systems based on a parity equation approach with neural net models are presented. Hydraulic systems are used in manifold applications in industry. They are however not yet the subject of intense research in the area of fault detection and diagnosis, which can be mainly attributed to their strong nonlinear behavior, which exacerbates the physical modeling extensively. To avoid the difficulties associated with the physical modeling, a data-driven modeling approach based on the LOLIMOT neural network will be presented in this paper. Different subsystems of the hydraulic servo axis will be modeled using different sensor configurations. Experimental data from a real testbed allow to compare the model fidelity of the different resulting neural nets and can also be used to illustrate the capabilities of the parity-equation based fault detection approach, which in general allows the detection of tiny faults, such as sensor offset faults in the area of a few percent of the maximum sensor readout.

Operation Point Dependent Diagnosis - WAI

Abstract As fault detection is becoming increasingly relevant for modern passenger car Diesel engines, it is now time to take a closer look at the engine specific fault diagnosis problems. Many of the applied fault detection algorithms can only be computed in certain operation ranges of the engine. Whenever the engine is operated outside these ranges, the corresponding symptoms are usually not actualized. As a matter of course this has to be considered in the design of a diagnosis system. However, since there are many well-developed diagnosis system structures, a method for the consideration of operation ranges should not impact on them. Hence, the diagnosis system structures are kept interchangeable and can be chosen from the available systems independently. The paper at hand focuses on the prevention of false alarms by considering operation ranges in a certain class of well-known diagnosis algorithms. For this purpose special symptoms and artificial fault measures are introduced. It can be shown that operation range dependent false alarms will no longer occur, if the diagnosis system is constructed using the new symptoms and fault measures. Since only new symptoms and fault measures are introduced, the choice of the diagnosis system structure can remain independent from the consideration of the operation ranges.