Optimized fuzzy controller in precise circuits monitoring the current from resistance shunt in high-voltage DC circuits

Abstract

Abstract Due to the low value of the resistance shunt, the measurement of low value current is a metrological problem. Properly selected processing constant and measurement system allows to minimize the metrological error. The proposed system consists of two identical non-inductive resistance shunts (125 µΩ) for measurement current. One of them is used to estimate the gain, the other is to measure the current. The amplification of the signal from the measuring shunt is controlled digitally by the fuzzy controller algorithm. The role of the intelligent system is to select the processing constant in order to use the dynamic range of the measuring system. The microprocessor parameters affect the efficiency of the intelligent algorithm. For this reason, part of the fuzzy controller should be implemented in the optimized form in order to relieve the microcontroller from calculations. The easiest way is to use operational amplifiers. Mention above the circuits act as a controller function block for fuzzification the output signal from the additional shunt. Takagi Sugeno Kang's inference system was selected in the proposed system due to the lack of a defuzzifier, which influences into its speed of operation. Carrying out computer simulations proved the effectiveness of the proposed system.