Enhanced Performance for Electrostatic Precipitators by Means of Conventional and Fuzzy Logic Control

Modern high voltage power supplies based on fast switching devices like IGBTs offer the performance to adjust voltage and current regarding the process conditions. Optimum performance can be achieved if the control makes use of the enhanced properties of the power supply. In critical precipitation processes the control needs to be high dynamical and accurate. Furthermore, with IGBT inverter type high voltage power supplies it is possible to increase the electrical power significantly, in many applications a factor 2 can be achieved, which leads to a significant improvement of the dedusting. Unfortunately, then the energy consumption of the precipitator increases leading to a higher amount of electricity to be generated in the power station and finally this results in higher CO 2 emissions of the power station. With central and decentralized control algorithms the total energy consumption of an electrostatic precipitator can be significantly reduced by continuous adaption of the control to the process conditions. Retrieving measured data from inside an electrostatic precipitator is always difficult and unreliable due to the high temperatures, dust and high voltage. Therefore, a rule based control system has been developed which can operate with a few input signals only. Mainly the measured values of voltage and current are sufficient for the control system and additional for energy savings the stack emission signal is required.