Increasing the efficiency and performance of internal combustion engines is always of interest in the automotive industry. One limiting factor to achieve this in gasoline combustion engines is the ignition timing which can not always be set where optimal ignition efficiency and performance is obtained. This is due to the knock phenomenon which is an abnormal combustion process that can damage the engine. Due to knock, a feedback controller which sets the ignition timing at the best possible value without the risk of harming the engine is required. In this thesis, a statistically driven knock intensity simulation environment based on the Burr Type XII distribution model was set up. In the simulation environment, different stochastic knock feedback controllers were implemented along with background noise estimation techniques used in the knock detection system. The feedback controllers were evaluated against the conventional knock controller commonly used in today’s engines in terms of ignition angle and transient response. The results from the simulation environment showed that a more advanced mean ignition angle can be achieved with stochastic based knock control strategies with the same knock-rate and without lessening the fast transient response achieved from the conventional strategy. To evaluate the results, some of the controllers were implemented in a four cylinder two-liter four stroke Volvo engine with similar results.

<p>Knock is a phenomenon that limits how effciently an engine can operate. Severe knock is harmful to the engine and must therefore be avoided. Controlling the knock intensity is complicated by a phenomenon called cycle to cycle variations. Because of these variations, the knock intensity must be considered a stochastic variable and the control is made on a mean value from a large number of cycles. </p><p>The SVC (Saab Variable Compression) concept adds the compression ratio as an extra degree of freedom. At Vehicular systems, research is done on how to put this additional variable to its best use. </p><p>A controller is developed that control the engine to a desired knock intensity and torque, using the ignition angle and the pedal position. The controller is implemented as two separate controllers in Matlab and Simulink. These are merged together with a Stateflow chart. A confidence interval calculation is implemented for the mean value of the knock intensity. A program is also developed to process a large number of operating points and make measurements in all of them. </p><p>The conclusion is that the basic construction of the controller and the script are fi;lling their functions but that there are some improvements left to be done. The controller is rather slow and the calculations of the confi;dence interval needs further refi;nement.</p>

Knock is a phenomenon that limits how effciently an engine can operate. Severe knock is harmful to the engine and must therefore be avoided. Controlling the knock intensity is complicated by a phenomenon called cycle to cycle variations. Because of these variations, the knock intensity must be considered a stochastic variable and the control is made on a mean value from a large number of cycles. The SVC (Saab Variable Compression) concept adds the compression ratio as an extra degree of freedom. At Vehicular systems, research is done on how to put this additional variable to its best use. A controller is developed that control the engine to a desired knock intensity and torque, using the ignition angle and the pedal position. The controller is implemented as two separate controllers in Matlab and Simulink. These are merged together with a Stateflow chart. A confidence interval calculation is implemented for the mean value of the knock intensity. A program is also developed to process a large number of operating points and make measurements in all of them. The conclusion is that the basic construction of the controller and the script are fi;lling their functions but that there are some improvements left to be done. The controller is rather slow and the calculations of the confi;dence interval needs further refi;nement.

Engine knock has long been a well recognized phenomenon in the automotive industry. Detecting engine knock opens up the possibility for an indirect feedback of the engine's internal combustion without installing a pressure transducer inside the cylinder. Knock detection has mainly been used for spark advance control, making it possible to control the engine close to its knock limit in search for the optimal ignition timing. This application has to a lesser extent been applied to lightweight two-stroke engines, which is the focus of this study. The investigation features a modern chainsaw engine whose knock characteristics were first determined with a pressure transducer. The structural vibrations originating from the engine knock are filtered out of the signal from a remote located accelerometer. The knock intensity is compared with the signal from the pressure transducer which shows a correlation with an accepted extent between the two sensors. Parameters that affect the knock intensity have also been investigated. These include engine temperature, different types of fuel and ignition timings.