No description available.

No description available.

No description available.

This report treats two assignments that concerns the biofuel furnace at Billerud Skärblacka AB and how these assignments were solved. The assignments were: reducing the number of CO-spikes and tuning the overfire air control loop. To reduce the number of CO-spikes the possibility that skewness in mass distribution affects the number of CO-spikes were study. The skewnwss of mass distribution was controled and adjusted with two experiments. The conclusion that was made is that the skewness of mass distribution were not sufficient enough to affect the number of CO-spikes. Most likely because there are other factor that affects the number of CO-spikes more than the skewness of mass distribution. The objective with tuning the overfire air control loop was to find a faster and better interplayed control loop. To solve the assignment the first thought was to make a model of the furnace. But this idea was too much work for a single master thesis. Therefore a Simulink-model and a control strategy were produced instead. During the work with tuning the overfire air control loop the changes were issued from the control strategy and the Simulink-model were used for the tuning of the air valves. The result from tuning the overfire air control loop were faster regulators and a better tracking of the reference value for the secondary and tertiary air valve along with a reduced levels of carbon monoxide.

More strict control quality requirements are raised to the synthesis of modern algorithmic control systems which can not be satisfied by using classical methods of systems synthesis. Also, the usage of them sometimes becomes impossible, e.g. in cases when a mathematical model of the control object is described by means of complex differential equations or in cases when the model itself is unknown. By applying the suggested synthesis methods of quasi-optimal and variable structure systems as well as algorithms, it is possible to avoid disadvantages common with classical synthesis methods.

No description available.

The control law that is investigated in this thesis is referred to as the acceleration feedback control law, first introduced by (Luo and Saridis 1983). The study of the error performance of this control law under non-ideal conditions is the subject of this thesis. A design methodology evolves from the theoretical foundations of the acceleration feedback control law. The design methodology yields a robust, high performance closed loop robot manipulator system. Design experiments are performed on a PUMA 600 robot manipulator and the closed loop performance is analyzed by simulation study. / The control law objective is to ensure that the manipulator joint coordinate trajectory is maintained with respect to a desired joint trajectory. The stability and error performance properties of the acceleration feedback control law are analytically examined by a Lyapunov stability analysis. The effects of integral control and computed torque augmentation are also examined in the stability and error performance analysis. The robustness properties, in the presence of high frequency unstructured uncertainties, are examined by classical frequency domain techniques. The theoretical results are verified by simulation. The simulation model test-bed is based on a PUMA 600 robot manipulator which is used throughout the thesis to support theoretical results. / The goal of this thesis is in presenting the acceleration feedback control law with an accompanying design methodology as a practical robot manipulator control law. The design methodology allows robot manipulator system designers to implement this control law as a simple, inexpensive, microprocessor based servomechanism at each joint. The theoretical development not only accounts for the intrinsic design trade-offs but also assures designers that the implementation will yield a robust, high performance closed loop system. This detailed design methodology is presented and a sample design is carried out and verified by simulation. The effects due to sampling, quantization, friction and frequency uncertainty are included in the simulation study.

No description available.

No description available.

No description available.