The invention discloses an intelligent adjusting method for an
edible oil alkali refining process based on
big data analysis. The method mainly comprises the following steps: 1, aligning
crude oil quality, acid-base reaction data,
centrifuge parameters and quality
test data into a splicing table according to time
delay; 2, forming a
decision table after redundancy and contradiction records are removed; 2, training an Xgboost model based on the
decision table to calculate an oil yield; 3, optimizing the optimal oil yield interval of the key parameters by adopting a Kmeans clustering
algorithm;4, optimizing the optimal parameter combination in the optimal oil yield interval by adopting an adaptive
simulated annealing genetic algorithm and taking the highest oil yield as a target; 5, adopting a
rough set algorithm to integrate field expert rules to form adjustment basic rules; and 6, performing basic rule constraint on the optimal parameters given by the
genetic algorithm, and performingbasic rule filtering on the optimal parameter combination given by the
genetic algorithm. According to the invention, the stable operation of the alkali refining process is ensured, the production cost is reduced, and the benefit maximization is realized.