Disturbance monitoring and compensation in soft sensing of mill load in thermal power plant

In this paper, a monitoring method for dynamic disturbance in soft sensing of mill load in thermal power plant is proposed. Firstly, the proposed approach divides dynamic disturbance into detectable disturbance and undetectable disturbance according to whether the disturbance can be monitored or not. Then, the corresponding correction methods are proposed respectively. For detectable disturbance signals, the corresponding error correction model is established respectively, and the mill load of the soft sensing is compensated online when disturbance occurs. For undetectable disturbance, online monitoring and adaptive correction are realized based on the optimal order in the fractional order domain. Finally, for showing the efficacy and feasibility of our proposed algorithm, some comparing experiments are conducted, which are all favorable for our viewpoints. That is the method we proposed enables the monitoring of detectable disturbance and undetectable disturbances, and automatically compensates the value of soft sensors.