A Soft-Sensor Method Based on Fuzzy Rules for Pulverized Coal Mass Flow Rate Measurement in Power Plant

A soft sensor method based on fuzzy rules for pulverized coal mass flow rate measurement in power plant is introduced, which used to resolve the problems of the electro dynamic sensor´s deficiency in absolute mass flow rate measurement and the effect of flow regime on the output of the sensor. Abundant experimental data captured by special electrostatic sensor is pre-processed through noise reduction and smooth filtering model, and the characteristic data is partitioned into some local region space by the fuzzy clustering algorithm, and a non-linear sub-model was established for each local region by using the radial basis function (RBF) neural network. The measurement result value can be described by a set of fuzzy rules based sub-models. The soft method based on non-linear processing can effectively reduce the influence of flow regime on the measurement results.