Well position optimization method based on statistical error active learning

• Main Authors: WANG MIN, YU HONGTIAN, LI BINQIAN, FU KE, MIN FAN, ZHOU LEI
• Format: Patent
• Language: Chinese; English
• Published: 07.01.2020
• Subjects: CALCULATING; COMPUTING; COUNTING; DATA PROCESSING SYSTEMS OR METHODS, SPECIALLY ADAPTED FORADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORYOR FORECASTING PURPOSES; HANDLING RECORD CARRIERS; PHYSICS; PRESENTATION OF DATA; RECOGNITION OF DATA; RECORD CARRIERS; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE,COMMERCIAL, FINANCIAL, MANAGERIAL, SUPERVISORY OR FORECASTINGPURPOSES, NOT OTHERWISE PROVIDED FOR

The invention discloses a well position optimization method based on statistical error active learning, belongs to the field of artificial intelligence, and has the beneficial effects that according to theoretical analysis and related experimental results, it is indicated that well position optimization by using an artificial intelligence method is indeed effective. The actual error model established by the invention proves a default nature that the closer the distance is, the higher the possibility of label consistency is in the presence in clustering all the time. According to the method, the clustering internal evaluation indexes are reasonably used, clustered data sets are divided into two classes, and different schemes are selected according to unique characteristics of the data sets,so that the active learning process does not completely depend on the clustering result. According to the method, the model established by reasonably utilizing the clustered distribution provides a theoretical basis for the