Method and system for evaluating maturity of artificial intelligence model in electric power field

• Main Authors: ZHOU DAZHOU, REN JIAYING, YANG YUECHEN, LIU SHENQI, WANG QI, LIU HAN, WU XIA, WANG YONG, LIANG TIAN, BAO LINA, DENG HUA, HAN YUANKAI, ZHAN JIE, XU KANG
• Format: Patent
• Language: Chinese; English
• Published: 08.03.2024
• 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 power field artificial intelligence model maturity assessment method and system, and the method comprises the steps: obtaining the performance index data of a to-be-assessed model, and the performance index data comprises the discovery rate, the false detection ratio and the processing rate index of the to-be-assessed model; calculating an evaluation score of each index based on the index values; and performing weighted summation on the evaluation score of each index to obtain a maturity evaluation result of the model. The method comprises the following steps: selecting a discovery rate, a false detection ratio and a processing rate of a model as evaluation indexes, and calculating an evaluation index value by using the number of correct frames and standard frames in a model identification process; the selection of the evaluation index better reflects the real demand of the actual application on the model, and solves the problem that the existing evaluation index cannot intuitively j