Volker Krebs

Predictive Powertrain Control for Heavy Duty Trucks

Abstract A truck driver controls his vehicle with the objective of maintaining a desired velocity while keeping the fuel consumption as low as possible. In order to achieve these goals he continuously estimates oncoming operation points of thepowertrain and chooses the inputs (driving torque, brake level, gear) in an optimal manner based on this estimation. A navigation system combined with a 3D digital road map is able to provide information about the roadway not only for the current position but also for a coming distance if the route is known. By imitating the driver, the Model Predictive Control method is used to apply this information for the implementation of a predictive gearshift program and a predictive cruise controller. Thereby in both systems the input variables are determined by the successive solution of discrete/continuous optimal control problems. The present paper first points out the idea of a predictive powertrain control for a heavy duty truck, followed by the description of its algorithmic realization.

PREDICTIVE CONTROL OF DRIVETRAINS

Abstract Telematics allow a prediction of the future driving conditions of a car along some time horizon. This prediction offers a knowledge of the torque request caused by the route ahead and can be used for implementing sophisticated operating strategies for Hybrid Electric Vehicles (HEVs). The intention of the present paper is to describe a method which minimizes the fuel consumption of the system beyond the prediction horizon. Therefore the strategy determines the best operating conditions of the combustion engine and the electric motor with respect to the predicted torque request and the SOC of the battery.

Predictive powertrain control for hybrid electric vehicles

Abstract This paper presents the use of telematics information for a predictive powertrain control in hybrid electric vehicles. By estimating the inclination and velocity profile of the road ahead the aggregates of the hybrid drivetrain can be controlled in a fuel-optimising fashion. For this purpose model predictive control is used. As the system is highly nonlinear Bellmans dynamic programming is used for solving the optimisation problem. The main focus of this paper is to describe how the computational effort of a hybrid drivetrain control algorithm can be reduced.

Brief paper: Input signal design for identification of max-plus-linear systems

The present contribution addresses the problem of designing an adequate persistent excitation for state space identification of max-plus-linear systems. The persistent excitation is designed using the same techniques that have recently been developed for model predictive control for max-plus-linear systems. The application of this method for input signal design allows to incorporate additional objectives which are desirable for the input signals and the resulting process behaviour such that an optimal persistent excitation is obtained.

A comparison of two methods for the numerical inversion of Laplace transforms

Abstract Two popular methods for numerical Laplace inversion are used to invert some Laplace transforms arising in chemical engineering problems. These methods have been chosen as they theoretically have general applicability and are relatively simple to implement. The first method devised by Zakian and known as I MN approximants inverts the transform by approximating a ∂ function by a ∂ convergent sequence. The second method developed by Honig and Hirdes is a sophisticated application of the trapezoidal rule to Bromwichs integral. In general, both methods give satisfactory results However, the latter method is to be preferred when high accuracy is desired with a FORTRAN compiler having no more than double precision facilities. Zakians method is generally faster but requires quadruple or higher precision facilities for results of similar accuracy to those obtained with the method of Honig and Hirdes employing double precision. We propose a method for checking the accuracy of the results and give an extended table of values of M and N for I MN approximants of full grade.

IDENTIFICATION OF HYBRID SYSTEMS USING A PRIORI KNOWLEDGE

Abstract An approach for the identification of a class of hybrid systems is presented. The identification problem for hybrid systems is formulated as an optimization problem and two possible ways for an approximative solution of the problem are discussed. As a result of this discussion a top-down algorithm for the approximative solution is developed. The proposed algorithm enables the user to incorporate a priori knowledge in very different scales. The feasibility and performance of the procedure is demonstrated at a two tank laboratory system.

Paper: Command and stability systems for aircraft: A new digital adaptive approach

Ensuring good handling qualities of aircraft in the whole flight envelope is a difficult and time consuming task, because aircraft parameters are subjected to drastic changes. Adaptive control is a very promising technique to solve these problems and to improve the performance deficiencies of conventional stability systems with preprogrammed control gains. The approach used in this paper is based on digital control and on a special stabilization concept which requires only three dynamic parameters of the aircraft to meet standard handling qualities criteria. These three parameters cannot be measured directly. Therefore a recursive, weighted least squares algorithm is used, which delivers sufficient fast and accurate on-line parameter estimates. The adaptive control system has proven its very satisfactory performance in several digital and hybrid simulations including fast parameters variations and turbulence conditions. Real flight test data have been used to verify the performance of the identification process and recently flight tests of the complete adaptive system have been successfully carried out with the DO 28 D SK YSERVANT aircraft.

Nonlinear Torque Control of a Spark Ignition Engine

Abstract In vehicles the produced engine torque is closely related to the manifold pressure, which is determined by the position of the throttle. In order to satisfy new comfort and safety requirements or to reduce NOx and HC emissions using variable cam timing an intervention of a torque control is needed. We propose two controller design approaches for this task. First, a new continuous nonlinear controller design method, called modified optimal control, is applied. State of the art controllers are implemented in digital signal processors. We therefore derive a hybrid nonlinear time discrete controller, too. Both controllers achieve appropriate dynamics and asymptotic stability of the closed loop in the whole operative range.

Modellierung, Simulation und Analyse hybrider dynamischer Systeme mit Netz-Zustands-Modellen

Prof. Dr.-Ing. Volker G. Krebs ist Leiter des Instituts für Regelungsund Steuerungssysteme der Universität Karlsruhe (TH). Forschungsgebiete: Ereignisdiskrete und hybride Systeme, nichtlineare Regelungen, wissensbasierte Systeme und qualitative Modellierung. Adresse: Universität Karlsruhe (TH), Institut für Regelungsund Steuerungssysteme, Kaiserstr. 12, D-76131 Karlsruhe. Tel.: 0721/608-3180, E-mail: krebs@irs.etec.uni-karlsruhe.de

Robust Consistency-Based Diagnosis of Nonlinear Systems by Set Observation

Abstract Achieving robustness and a high fault sensitivity simultaneously is one of the most important goals of diagnosis system design. The idea of the so-called passive approach, which has been given relatively little attention in literature so far, is to include the effects of measurement and model uncertainties in the residual. In the subsequent residual evaluation, these uncertainties can then be accounted for such that false alarms can be precluded. Following this passive approach, we present a new model-based diagnosis algorithm based on state-set observation of nonlinear continuous-time systems. A set-valued observer following the well-known predictor-corrector scheme is used to calculate a set of system states. These sets are consistent with the underlying system model as well as with the discrete measurements including both model and measurement uncertainties. In the prediction step, a validated ODE solving method is applied for calculation of the consistent state set. To the authors knowledge, such a nonlinear continuous-time set-valued observer has not yet been used for diagnosis tasks. The performance of the method is demonstrated using measured data of fault-free and faulty operation of an inverted pendulum as a benchmark system.