Matches in SemOpenAlex for { <https://semopenalex.org/work/W2326169622> ?p ?o ?g. }
Showing items 1 to 67 of
67
with 100 items per page.
- W2326169622 endingPage "284" @default.
- W2326169622 startingPage "272" @default.
- W2326169622 abstract "摘要: 支持向量回归(Support vector regression, SVR)的学习性能及泛化性能取决于参数设置.在常规方法中,这些参数以固定值形式参与运算,而当面对复杂分布的数据集时, 可能无法挑选出一组能够胜任各种分布情况的参数,参数设置需要在过拟合和欠拟合之间进行取舍. 因此,本文提出一种能够根据样本分布进行参数自我调整的柔性支持向量回归算法(Flexible support vector regression, F-SVR).该算法根据样本分布的复杂度,将训练样本划分为多个区域,在训练过程中, F-SVR为不同 区域设置不同的训练参数,有效避免了过拟合与欠拟合.本文首先采用一组人工数据对所提算法有效性进行验证,在实验中, F-SVR在 保持学习能力的同时,具备较传统方法更优秀的泛化性能.最后,本文将该算法运用至高频电源故障的实际检测,效果良好. 关键词: 支持向量回归 / 柔性 / 故障检测 / 电源" @default.
- W2326169622 created "2016-06-24" @default.
- W2326169622 creator A5010751143 @default.
- W2326169622 creator A5013359542 @default.
- W2326169622 creator A5021185593 @default.
- W2326169622 creator A5026793625 @default.
- W2326169622 creator A5079527716 @default.
- W2326169622 date "2014-03-21" @default.
- W2326169622 modified "2023-09-27" @default.
- W2326169622 title "Flexible Support Vector Regression and Its Application to Fault Detection" @default.
- W2326169622 doi "https://doi.org/10.3724/sp.j.1004.2013.00272" @default.
- W2326169622 hasPublicationYear "2014" @default.
- W2326169622 type Work @default.
- W2326169622 sameAs 2326169622 @default.
- W2326169622 citedByCount "1" @default.
- W2326169622 countsByYear W23261696222014 @default.
- W2326169622 crossrefType "journal-article" @default.
- W2326169622 hasAuthorship W2326169622A5010751143 @default.
- W2326169622 hasAuthorship W2326169622A5013359542 @default.
- W2326169622 hasAuthorship W2326169622A5021185593 @default.
- W2326169622 hasAuthorship W2326169622A5026793625 @default.
- W2326169622 hasAuthorship W2326169622A5079527716 @default.
- W2326169622 hasConcept C105795698 @default.
- W2326169622 hasConcept C12267149 @default.
- W2326169622 hasConcept C124101348 @default.
- W2326169622 hasConcept C127313418 @default.
- W2326169622 hasConcept C152745839 @default.
- W2326169622 hasConcept C154945302 @default.
- W2326169622 hasConcept C165205528 @default.
- W2326169622 hasConcept C172707124 @default.
- W2326169622 hasConcept C175551986 @default.
- W2326169622 hasConcept C33923547 @default.
- W2326169622 hasConcept C41008148 @default.
- W2326169622 hasConcept C83546350 @default.
- W2326169622 hasConceptScore W2326169622C105795698 @default.
- W2326169622 hasConceptScore W2326169622C12267149 @default.
- W2326169622 hasConceptScore W2326169622C124101348 @default.
- W2326169622 hasConceptScore W2326169622C127313418 @default.
- W2326169622 hasConceptScore W2326169622C152745839 @default.
- W2326169622 hasConceptScore W2326169622C154945302 @default.
- W2326169622 hasConceptScore W2326169622C165205528 @default.
- W2326169622 hasConceptScore W2326169622C172707124 @default.
- W2326169622 hasConceptScore W2326169622C175551986 @default.
- W2326169622 hasConceptScore W2326169622C33923547 @default.
- W2326169622 hasConceptScore W2326169622C41008148 @default.
- W2326169622 hasConceptScore W2326169622C83546350 @default.
- W2326169622 hasIssue "3" @default.
- W2326169622 hasLocation W23261696221 @default.
- W2326169622 hasOpenAccess W2326169622 @default.
- W2326169622 hasPrimaryLocation W23261696221 @default.
- W2326169622 hasRelatedWork W1974448437 @default.
- W2326169622 hasRelatedWork W1993592567 @default.
- W2326169622 hasRelatedWork W2020422030 @default.
- W2326169622 hasRelatedWork W2160656349 @default.
- W2326169622 hasRelatedWork W2350413141 @default.
- W2326169622 hasRelatedWork W2995434426 @default.
- W2326169622 hasRelatedWork W3126246916 @default.
- W2326169622 hasRelatedWork W3138962322 @default.
- W2326169622 hasRelatedWork W3208702349 @default.
- W2326169622 hasRelatedWork W84219535 @default.
- W2326169622 hasVolume "39" @default.
- W2326169622 isParatext "false" @default.
- W2326169622 isRetracted "false" @default.
- W2326169622 magId "2326169622" @default.
- W2326169622 workType "article" @default.