TOP > 外国特許検索 > CONTINUOUS VALUE OPTIMIZATION PROBLEM GLOBAL SEARCH DEVICE AND PROGRAM

# CONTINUOUS VALUE OPTIMIZATION PROBLEM GLOBAL SEARCH DEVICE AND PROGRAM

外国特許コード F180009619 5459 2018年11月20日 世界知的所有権機関（ＷＩＰＯ） 2017JP023050 WO 2018042840 平成29年6月22日(2017.6.22) 平成30年3月8日(2018.3.8) 特願2016-171776 (2016.9.2) JP CONTINUOUS VALUE OPTIMIZATION PROBLEM GLOBAL SEARCH DEVICE AND PROGRAM In the present invention, a setting unit (6) sets a plurality of individual piece variables in accordance with an evaluation function in a search space. An extreme value conversion unit (7) derives a gravitational force potential that causes a gravitational force to act on at least two individual pieces selected from among the plurality of individual pieces set by the setting unit, and converts to an extreme value an effective evaluation value (Heff) obtained by adding: an evaluation addition value obtained by adding evaluation values of an evaluation function based on the plurality of individual piece variables; and a gravitational force value obtained by adding up the derived gravitational force potential between the selected individual pieces, and multiplying the addition value by a gravitational force coefficient (g/2) acting between the selected individual pieces. A derivation unit (8) changes the plurality of individual piece variables such that the effective evaluation value is converted to an extreme value by the extreme value conversion unit, while increasing the gravitational force coefficient gradually from an initial value, and derives various values using the individual piece variable, the optimized value of the variable, the evaluation value corresponding to the variable, or the optimized value of the evaluation value, on the basis of at least one of the plurality of individual pieces when a prescribed end condition is satisfied. BACKGROUND ARTFor example, in a multi-variable search space, i.e. a method to determine the optimum value of the parameter has been proposed (for example, see Patent Document 1). According to the technique described in Patent Document 1, parameters for the optimum value calculation within the search space and generate a plurality of individual elements, and calculates the evaluation value of these individuals, individuals of poor evaluation value from a selected individual. Then, the individual evaluation value of the selected individual may be provided near at a predetermined ratio. In addition, the evaluation value of the best individual from the individual evaluation values of the poor Euclidean distance within the constant is moved to an arbitrary region. Then, a better evaluation value of the best individual of the selected individual evaluation value is updated at any time, the evaluation value converges a plurality of individual, end determination condition is satisfied on determining the individual having the best evaluation value as an optimum value of the parameters included in the output. DENSO CORPORATION National university corporation Kyoto University OKADA, Shuntaro TERABE, Masayoshi OHZEKI, Masayuki G05B  13/02      電気式 G06N  99/00      このサブクラスの他のグループに分類されない主題事項 National States: AE AG AL AM AO AT AU AZ BA BB BG BH BN BR BW BY BZ CA CH CL CN CO CR CU CZ DE DJ DK DM DO DZ EC EE EG ES FI GB GD GE GH GM GT HN HR HU ID IL IN IR IS JO KE KG KH KN KP KR KW KZ LA LC LK LR LS LU LY MA MD ME MG MK MN MW MX MY MZ NA NG NI NO NZ OM PA PE PG PH PL PT QA RO RS RU RW SA SC SD SE SG SK SL SM ST SV SY TH TJ TM TN TR TT TZ UA UG US UZ VC VN ZA ZM ZWARIPO: BW GH GM KE LR LS MW MZ NA RW SD SL SZ TZ UG ZM ZWEAPO: AM AZ BY KG KZ RU TJ TMEPO: AL AT BE BG CH CY CZ DE DK EE ES FI FR GB GR HR HU IE IS IT LT LU LV MC MK MT NL NO PL PT RO RS SE SI SK SM TROAPI: BF BJ CF CG CI CM GA GN GQ GW KM ML MR NE SN ST TD TG
ライセンスをご希望の方、特許の内容に興味を持たれた方は、下記までご連絡ください。