Classification and feature selection method based on multi-objective optimization

• Main Author: PAN SHIXIAN
• Format: Patent
• Language: Chinese; English
• Published: 25.10.2022
• Subjects: CALCULATING; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; PHYSICS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses a classification and feature selection method based on multi-objective optimization, and the method comprises the steps: 1, simulating the scheduling of a cross-basin water transfer project based on a water balance principle, and building a multi-year average water transfer simulation model; 2, constructing a multi-target optimization scheduling model with the maximum average total water supply amount for many years, the maximum total power generation amount of the parallel reservoirs and the minimum energy consumption of the pump station as targets; 3, constructing a water shortage index, an economic index and an ecological index, and fusing the three indexes to construct a novel water resource distribution comprehensive index; 4, introducing the water resource distribution comprehensive index into a simulated water resource configuration model, calculating the configured water resource condition through the model, carrying out the maximum water transfer potential calculation of the c