( 3 ) features of turbopump test data both in time domain and in frequency domain are analyzed using statistical estimation methods . the value of applying these features into nd is discussed ( 3 )應用統(tǒng)計特征估計方法,分別對渦輪泵試車數據的時域特征和頻域特征進行了統(tǒng)計分析,并討論了這些特征在新異類檢測中的應用價值。
Statistical learning theory is a newly developed theory for studying the statistical estimation and prediction problem based on small number of samples . it studies the nature of machine learning , so more and more people are interested in it 統(tǒng)計學習理論是在研究小樣本統(tǒng)計估計和預測的過程中發(fā)展起來的一種新興理論,它試圖從更本質上來研究機器學習問題,因此引起了人們越來越多的重視。
Under the fractal application , the transitional state rendering method based on t - fbm constraint model is proposed . in this model , the transitional deformation energy is defined as the bi - association between the initial state and the transitional fractal detail , and the fbm constraint is realized by the threshold statistical estimation . furthermore , the regional buffering control and regional harmonic control can offer more abundant and flexible control ways for the rendering of fractal transitional state 針對過渡狀態(tài)可視化的分形應用,提出了基于t一fbm約束的繪制方法,將過渡變形能作為與初始形狀和過渡分形細節(jié)之間的雙向關聯(lián),以閉值估計一作為fbm約束因子的具體實現,以區(qū)域緩沖與區(qū)域調和作為不同約束下的控制手段,很好地在繪制結果中反映出分形性過渡特性的分布情況。