Michiel J. Van Nieuwstadt

Real Time Trajectory Generation for Differentially Flat Systems

This paper considers the problem of real-time trajectory generation and tracking for nonlinear control systems. We employ a two-degree-of-freedom approach that separates the nonlinear tracking problem into real-time trajectory generation followed by local (gain-scheduled) stabilization. The central problem which we consider is how to generate, possibly with some delay, a feasible state space and input trajectory in real time from an output trajectory that is given online. We propose two algorithms that solve the real-time trajectory generation problem for differentially flat systems with (possibly non-minimum phase) zero dynamics. One is based on receding horizon point to point steering, the other allows additional minimization of a cost function. Both algorithms explicitly address the tradeoff between stability and performance and we prove convergence of the algorithms for a reasonable class of output trajectories. To illustrate the application of these techniques to physical systems, we present experimental results using a vectored thrust flight control experiment built at Caltech. A brief introduction to differentially flat systems and its relationship with feedback linearization is also included.

Modeling of a Urea SCR Catalyst With Automotive Applications

A zero order lumped parameter control oriented model of a Selective Catalytic Reduction (SCR) catalyst is presented. The lumped parameter model is developed using two approaches. in the first approach it was assumed that the catalyst behaves as an Isothermal Continuously Stirred Tank Reactor (ICSTR). The second approach involved deriving the lumped parameter model from a 1D model using the method of weighted residuals. Both approaches led to a three state model, with the gas phase concentrations of NOx and NH3 and the surface coverage fraction as the states. The model depends on chemical properties specific to the SCR catalyst; consequently model validation requires knowledge of these parameters, either via laboratory-based experiments or as supplied by the catalyst supplier. We present an alternate approach that allows estimation of the essential parameters through a minimization of the l2 errors between measured and simulated results.Copyright

Rapid Hover-to-Forward-Flight Transitions for a Thrust-Vectored Aircraft

The use of differential e atness for computation of a nominal trajectory for fast transition between e ight modes of autonomous vehicles is investigated. Differential e atness of an approximate model of the longitudinal dynamics of a thrust-vectored aircraft is used to achieve fast switching between e ight modes. We conclude that steering to the trimmed state of the full model is of crucial importance for good performance. Simulations and experimental data for a thrust-vectored e ight-control experiment at Caltech are provided to validate the approach.

Thermodynamics-Based Mean Value Model for Diesel Combustion

A thermodynamics-based computationally efficient mean value engine model that computes ignition delay, combustion phases, exhaust temperature, and indicated mean effective pressure has been developed for the use of control strategy development. The model is derived from the thermodynamic principles of ideal gas standard limited pressure cycle. In order to improve the fidelity of the model, assumptions that are typically used to idealize the cycle are modified or replaced with ones that more realistically replicate the physical process such as exhaust valve timing, in-cylinder heat transfer, and the combustion characteristics that change under varying engine operating conditions. The model is calibrated and validated with the test data from a Ford 6.7 liter diesel engine. The mean value model developed in this study is a flexible simulation tool that provides excellent computational efficiency without sacrificing critical details of the underlying physics of the diesel combustion process.

Two-Stage Turbocharger Modeling for Engine Control and Estimation

The increasingly stringent emissions regulations and needs for higher power density for both turbo-diesel passenger vehicle and commercial vehicles have demanded significant alterations to the basic architecture of turbochargers. An attractive option for providing a high-boost system is the use of two-stage turbocharger which consists of two different size turbochargers connected in series that may or may not utilize bypass regulation. The exhaust mass flow is expanded by the high pressure turbine to the low pressure turbine, and on the other side the air flow is compressed through the low pressure compressor to the high pressure compressor. This increases the complexity of the air-charging system and requires new methodologies for modeling and control. A two-stage turbocharger model is presented in this paper. The total efficiency of the two-stage compressor, which poses the biggest problem in two-stage turbocharger modeling, was derived based on a second law analysis. A new parameter, compressor temperature ratio, was introduced as a linkage between the two stage compressors and also used to predict the two-stage compressor outlet temperature. Extrapolation to lower turbocharger speeds and compressor flow rates by using curve fitting methods was also discussed. The model for a two-stage turbine with a bypass valve is derived in the same way. Engine dynamometer tests have been performed to identify the model parameters and to validate the model structure. The test results show a good agreement between the model predictions and test data. In conclusion, this two stage turbocharger model is suitable for turbocharger control design and the estimation of some key turbocharger parameters.Copyright

NOx Prediction in Diesel Engines for Aftertreatment Control

Modern Diesel engines are faced with two major emission challenges in their quest to become an environmentally compatible source of motive power, Nitrogen Oxides (NOx ) and Particulate Matter (PM). Advanced techniques, such as High Pressure Common Rail (HPCR) fuel injection combined with multiple injections per cycle, are commonly employed to minimize in-cylinder production of NOx and PM. However, to meet the EPA mandated standards it is essential that an aftertreatment system be used. Typical Diesel aftertreatment systems will employ some form of a NOx reducing catalyst and a particulate trap for PM removal. Lean NOx traps and Selective Catalytic Reduction (SCR) are examples of aftertreatment techniques frequently used in Diesel engine applications. Whatever the method of choice, knowledge of the feed-gas NOx concentration is essential for not only assessing the performance of the NOx reduction catalyst but also for defining the control strategy for the aftertreatment system with respect to the management of the reductant quantity to be injected. In the absence of a dynamic NOx emission model the control algorithm has to depend on either a NOx sensor upstream of the catalyst or a static map of the feedgas NOx level as some function of engine influence factors. While NOx sensors add to the overall system cost, creating an accurate and representative NOx map over the entire engine operating range can be a challenging task. A dynamic NOx model would, in theory solve, both of these problems, however it is essential that the model be simple and implementable in real time. A model that uses inputs that are not available from the standard measurement set is of little use for real time control applications as is a model that predicts the temporal and spatial NOx evolution in the engine combustion chamber as such models tend to be computationally expensive. However, it is essential that the model behave like a fast NOx sensor in predicting cycle averaged NOx emission. In this paper we present an approach to developing such a model and present results from model validation against vehicle data. The basic structure of the model relies on well-known mechanisms that describe the NOx creation and decomposition chemical kinetics. Simplifying assumptions are made to allow available measurements to be used as inputs to the model. This leads to a parametric model where the unknown parameters are estimated using Nelder Mead optimization routine available in Matlab®. Model validation against vehicle data is also presented.Copyright

Experiments in Active Diesel Particulate Filter Regeneration

Diesel particulate filters (DPFs) are a technology likely to be deployed to meet future stringent emission levels for heavy and light duty diesel powertrains in North America and Europe. This paper discusses experimental results in the active regeneration of DPFs. Attention is given to the system components, the information based on which regeneration is triggered, and the means to achieve a regeneration. The paper will report on successful regenerations under several extreme conditions.

Control Design of an Automotive Urea SCR Catalyst

The leading aftertreatment technologies for NOx removal from the exhaust gas of lean burn engines, Diesels in particular, are urea based Selective Catalytic Reduction (SCR), Lean NOx Traps (LNT) and Active Lean NOx Catalysts (ALNC). It is generally believed that the SCR technique has the potential of providing the best NOx conversion efficiency relative to the other techniques. Nonetheless, it is crucial that the high conversion efficiencies be achieved with a minimum slippage of unreacted ammonia as tail pipe emissions. This necessitates a precise control over the urea injection process. The complex behavior of the catalyst substrate with respect to adsorption and desorption of ammonia in conjunction with a lack of “stored ammonia” sensing capabilities makes the control problem challenging. In this paper we present a model-based control design approach using a lumped parameter model of an SCR system that includes the essential dynamics of the plant. The model includes the adsorption, desorption and surface coverage dynamics, along with the NOx reduction and ammonia oxidation dynamics based on the relevant chemical reaction rates.© 2002 ASME

In-cylinder pressure sensor–based NOx model for real-time application in diesel engines

A real-time implementable, zero-dimensional model for predicting engine-out emissions of nitrogen oxides using in-cylinder pressure measurements is developed. The model is an extension of existing works in open literature that align well with the objectives of real-time implementation. The proposed model uses a simplified Zeldovich NOx mechanism that uses combustion-related parameters derived from simplified thermodynamic and combustion sub-models. The performance of the model is discussed for both a heavy-duty and a light-duty diesel engines. The model behavior is evaluated under input uncertainty so as to provide realistic performance bounds.

Piston temperature model oriented to control applications in diesel engines

In this paper, the development of a control-oriented piston temperature model for diesel engines is discussed. Using the underlying energy balance at the piston, a one-state piston temperature model was developed based on a thermal resistance concept. The model is composed of five sub-models: an engine model, a heat distribution model, a piston temperature model, an initial piston temperature model, and a maximum piston temperature model. In the engine model, the combustion heat transferred to the engine is calculated based on the energy balance in the cylinder chamber. The heat distribution model, which is a main feature in this model, determines the heat transferred to the piston using two maps as a function of engine speed and fuel depending on the piston cooling jet (PCJ) operation. The energy balance at the piston is applied to calculate the mean piston temperature, and the initial piston temperature is determined by the arbitration between the piston and the oil temperatures. The maximum piston temperature is estimated using a simple linear correction to the mean piston temperature. Integrating all sub-models in the Simulink platform, the model was identified and validated using piston temperature measurements under steady-state fuel steps as well as transient tests. There is a good agreement between the modeled and the measured piston temperatures with less than 4.1°C of root-mean-square-error (RMSE) over transient emissions cycles (FTP-75, LA92, and HWEFT). The modeled piston temperature can be used as an input to the control strategy of variable cooling devices, such as a variable displacement oil pump.