Immune genetic algorithm-based adaptive evidential model for estimating unmeasured parameter: Estimating levels of coal powder filling in ball mill

摘要

To estimate the unmeasured parameter from experts and running data, in this paper, a novel method named "immune genetic algorithm-based adaptive evidential classification rule (IGA-EC)" was proposed. The IGA-EC model was realized by two strategies: (1) a new parametric distance metric was applied instead of Euclidean distance to enhance the robust adaptive ability of the traditional evidence-theoretic classification rule; and (2) the powerful evolutionary algorithm immune genetic algorithm was used to parallel search the global optimal solutions of the parameters involved in the proposed model. To validate IGA-EC model, some experiments were conducted based on some popular data sets, and the experimental results show that the proposed method was powerful with respect to the accuracy. Finally, the IGA-EC model was used to estimate the unmeasured parameter level of coal powder filling in the ball mill in power plant. From the analysis of the estimating results, it suggests that the proposed method was applicable for estimating the level of coal powder, and the proposed method can also be applied for estimating other unmeasured parameters in industry.