backward algorithm 中文意思是什麼
backward algorithm
解釋
向後演算法-
Applying the classical pattern recogtiition theory anci ftiflcial neural networks method, this paper proposes the analog fault diahoes priricip1s with backward - propagation neural network ( i3pnn ) arid self - or ~ anizing feature map ( sofm ) neural network algorithm implementation
本文提出了模擬電路故障診斷前向多層誤差反傳（ bp ）網路和自組織特徵映射（ sofm ）網路的演算法實現方法。 -
The backward error and the structured backward error of the approximate solution are the criteria to judge the stability and the strong stability of the numerical algorithm. condition number is a measure of the sensitivity to the approximate solution for the perturbation of original date
近似解的最佳向後誤差和最佳結構向後誤差的數值分別是判別演算法的穩定性和強穩定性的標準，而條件數則是反映數值問題的解對于該問題數據擾動的敏感程度。 -
This algoritlun is more efficient by the heuristic depth esthoating technique proposed in this paper which gets displacement values used during backward mapping process from a single primary reference image under the observation that, the shape of space surfaces often changes smoothly in comparison with usual forward mapping methods, this algorithm generates derived images with less errors and tackles the problems associated with multiple reference images
在論文第二章中，本文提出了一種多幅圖象合成演算法。該演算法首先進行主參考圖象位移值的正向映射，以獲取目標圖象所對應的位移圖：然後通過一個深度預測演算法，對尚未完整的位移圖作空洞填補：最後由逆向映射過程來生成最終目標圖象。 -
Two image mapping algorithms of the volume rendering, forward mapping and backward mapping are researched on the basis of optical model for volume rendering. combined with the characteristic of a series of section images from laser confocal scanning microscopy, the volume buffer algorithm and maximizing algorithms are proposed in this paper. 4
同時在探討體繪制光學模型的基礎上，研究了體繪制的兩種圖像合成演算法，即由前向後演算法和由後向前演算法，並結合激光共聚焦掃描顯微成像系統序列斷層熒光圖像的特點提出了體緩沖器演算法和最大值演算法。 -
It is shown that using backward selection algorithm select centers can minimize the net work output error in great extent, make network structure more simple and more generalization and approximation. furthermoure, the offline control to pangang hot metal desulfuration process is realized using rbfnn model
理論研究和模擬實驗證明： rols法比起批處理正交最小二乘法（ ols ）所需的計算空間小，訓練速度快；用後向選擇法選擇中心能最大程度地減小網路的輸出誤差，使網路結構簡化，並且具有更強的泛化能力和逼近能力。 -
In the numerical solution algorithm, the method of characteristics, analytic method and galerkin finite element method ( galerkin - fem ) can be chosen to solve the advective equation, diffusion equations, reaction ( source / sink ) equations, propagation equations and pressure poisson equation, respectively. the developed new algorithm has been verified using analytical solution of circular conduit flow in a reynolds number range of 100 < re < 1 000 and experimental data of the laminar flow over a backward - step facing step. the flow properties are well characterized by this three - dimensional numerical model
本論文在評述三維粘性流動數學模型已有研究成果的基礎上，著重在數值計算方法的選擇和定解條件的給定對數學模型計算結果的影響進行了研究，並首次提出了求解三維純對流方程的高精度的擬協調單元法，建立了三維低雷諾數re流動的數學模型，並在圓管流動、臺階突擴矩形管道流動中得到驗證和應用。 -
The system denotes knowledge with status space, manages repository and correlative database by use of dbms, and speculate mechanism makes use of backward promulgate algorithm of three layer artificial neural network
該系統採用狀態空間知識表示方法，應用dbms管理知識庫和相關數據庫，推理機採用三層人工神經網路的反向傳播演算法。 -
Secondly, binary probability hypothesis detection is studied and is utilized as actuators " fdi residual decision. thirdly, multi - layer feed - forward ann and error backward propagation ( bp ) algorithm are studied and are utilized as control surfaces " failure detection and sortation
第三，研究多層前饋神經網路及相應的bp （ errorbackwardpropagation誤差反向傳播）演算法，設計基於模型殘差的故障分類器，並通過其來完成全局舵面的故障檢測與分類。 -
The forward inference algorithm starts from existing facts while the backward inference starts from result fact
基於ann的知識庫，論文提出了兩種不同的推理演算法。 -
The fuzzy system is constructed through the artificial neural learning algorithm. third, the fault diagnosis using neural networks is discussed in this paper, especially the internal backward neural network with deviation elements. its model and learning algorithm are showed in detail
其中詳細討論了帶有偏差單元的遞歸神經網路的模型和學習演算法，在此基礎上合成了模糊神經網路系統，將模糊理論和神經計算原理相結合，使神經網路藉助其大規模的并行分佈處理結構完成模糊診斷的推理過程。 -
As emphasis, we propose a new backward width - flrst search circuit partitioning method with flip - flop as core for synchronous sequential circuits. and then based on it, we develop a new circuit parallel tg algorithm
最後重點對電路并行方法進行了研究，提出了一種新的以觸發器為核且消除大功能塊之間反饋的寬度優先反向搜索同步時序電路劃分方法。 -
For the total tardiness scheduling with precedence constraints, an approximation algorithm with polynomial complexity was presented by transplanting the backward - shift algorithm of the case without precedence constraints
摘要把工件之間不帶前後約束的延誤排序的后移演算法移植到帶有前後約束的情況，提出一個多項式時間的近似演算法。 -
In order to reduce the computations and the memories, restarted version is used in it. by using the norm of the total backward perturbation as the termination of the algorithm, the new method makes up the shortcomings of using residual norm, and then the restarted total minimal backward perturbation ( rtminback ) method is proposed
為減少存儲量，新演算法採用重新開始的循環格式，並將總體向後擾動范數作為判斷演算法終止的準則，克服了殘量范數作為終止準則的不足，得到了循環總體極小向後擾動（ rtminback ）方法。 -
The critical jobs must be scheduled by the backward algorithm, which ensures the necessary resources allocated to the most important jobs ; then the remaining jobs are scheduled with a forward algorithm, which tends to allocate resources where the previous scheduling pass did not use them
演算法中關鍵零件採用反向調度，盡可能地滿足交貨期要求；一般零件採用前向調度方法，利用剩餘的車間資源，盡可能早完工。通過遺傳演算法的全局搜索，得到最佳的調度結果。 -
In this thesis, we first introduce give the definition of hidden markov models. then the methods to solve the three basic problems in the application of hidden markov models are introduced, namely three basic arithmetic : forward - backward algorithm, viterbi algorithm, baum - welch algorithm. also we present commonly model of hidden markov processes dynamic system
在這篇論文中，首先給出了隱馬爾科夫模型的定義，接著介紹了隱馬爾科夫模型實際應用中所面臨的三大基本問題的解決方案，即隱馬爾科夫模型三大基本演算法：前向一後向演算法、 viterbi演算法、 baum ? welch演算法。 -
In section two, the algorithms of vector - valued rational interpolants are stated generally. then a new algorithm of brivate vector - valued rational interpolants by means of complexification of the knots and backward three - term recurrence relations is given
然後利用插值型值點復數化的方法及向量值連分式的向後三項遞推關系式討論並給出了二元向量值有理插值的一種新演算法。 -
Secondly, novel genetic algorithm is developed to optimize the wavelength and power value of the pumping sources in the mutli - wavelength backward pumping fra to get the best gain flatness
第二設計新型的遺傳演算法，結合前面提到的多維并行打靶法來優化多波長後向泵浦fra各個泵浦光的波長、功率值，以得到更好的增益平坦度。 -
Applying this classic pattern recognition theory and artificial networks method, this paper proposes the analog fault diagnosis principles with backward - propagation neural network ( bpnn ) algorithm implementation
本文有針對性的提出了模擬電路故障診斷前向多層誤差反傳（ bp ）網路的演算法實現途徑。
分享友人