US 12,393,750 B2
Computer-readable recording medium storing machine learning program, information processing apparatus, and machine learning method
Takashi Yamazaki, Kawasaki (JP); Shohei Yamane, Kawasaki (JP); Hiroaki Yamada, Kawasaki (JP); Masatoshi Ogawa, Zama (JP); Yoichi Kochibe, Mihama (JP); Toshiyasu Ohara, Nakano (JP); and Takashi Kobayashi, Machida (JP)
Assigned to Fujitsu Limited, Kawasaki (JP)
Filed by FUJITSU LIMITED, Kawasaki (JP)
Filed on Sep. 8, 2021, as Appl. No. 17/468,810.
Claims priority of application No. 2020-180264 (JP), filed on Oct. 28, 2020.
Prior Publication US 2022/0129607 A1, Apr. 28, 2022
Int. Cl. G06F 30/27 (2020.01); G06F 30/367 (2020.01)
CPC G06F 30/27 (2020.01) [G06F 30/367 (2020.01)] 5 Claims
OG exemplary drawing
 
1. A non-transitory computer-readable recording medium storing a machine learning program for causing a computer to execute a process comprising:
acquiring a shape image that indicates a shape of a circuit based on circuit information before shape simplification processing, the shape of the circuit including at least any one of a slit, a slot, a hole, or an opening;
acquiring a current distribution image that is an image indicating a current distribution in an equivalent circuit model by performing an equivalent circuit simulation on the equivalent circuit model, the equivalent circuit model being a circuit model derived by applying the shape simplification processing to the circuit defined by the circuit information to remove the at least any one of a slit, a slot, a hole, or an opening from the circuit;
acquiring an electromagnetic interference (EMI) value by electromagnetic field analysis based on the circuit information;
generating training data by combining the current distribution image, the shape image, and the EMI value into the training data;
generating an EMI prediction model by machine learning based on the generated training data that includes the current distribution image, the shape image, and the EMI value;
inputting, to the generated EMI prediction model, prediction target data including a geometric shape diagram and a current distribution image, to cause the generated EMI prediction model to output a prediction result of the EMI value for the prediction target data.