點態收斂 的英文怎麼說

中文拼音 [diǎntàishōuliǎn]
點態收斂 英文
pointwise convergence
  • 點 : Ⅰ名詞1 （液體的小滴） drop （of liquid） 2 （細小的痕跡） spot; dot; speck 3 （漢字的筆畫「、」）...
  • 態 : 名詞1. （形狀; 狀態） form; condition; appearance 2. [物理學] （物質結構的狀態或階段） state 3. [語言學] （一種語法范疇） voice
  • 收 : Ⅰ動詞1 （把攤開的或分散的事物聚集、合攏） put away; take in 2 （收取） collect 3 （收割） harvest...
  • 斂 : Ⅰ動詞1 （收起; 收住） hold back; keep back 2 （約束） restrain 3 （收集; 徵收） gather; collect; ...
  • 收斂 : 1 （減弱或消失） weaken or disappear 2 （約束言行） restrain oneself 3 [數學] convergence; constr...
  1. This article puts forward a solution named divide - assemble by deducing the size of bp neural network to overcome entering the local best point, the dividing process is that a big bp neural network is divided into several small bp neural networks, every small bp neural network can study alone, after all small bp neural networks finish their study, we can assemble all these small bp neural networks into the quondam big bp neural networks ; on the basis of divide - assemble solution, this article discusses the preprocessing of input species and how to deduce the size of bp neural network further to make it easy to overcome entering the local best point ; for the study of every small bp neural network, this article adopts a solution named gdr - ga algorithm, which includes two algorithms. gdr ? a algorithm makes the merits of the two algorithms makeup each other to increase searching speed. finally, this article discusses the processing of atm band - width distribution dynamically

    本文從bp網的結構出發，以減小bp神經網路的規模為手段來克服陷入局部極小點，提出了bp神經網路的拆分組裝方法，即將一個大的bp網有機地拆分為幾個小的子bp網，每個子網的權值單獨訓練，訓練好以後，再將每個子網的單元和權值有機地組裝成原先的bp網，從理論和實驗上證明了該方法在解決局部極小值這一問題時是有效的；在拆分組裝方法基礎上，本文詳細闡述了輸入樣本的預處理過程，更進一步地減小了bp網路的規模，使子網的學習更加容易了；對于子網的學習，本文採用了最速梯度? ?遺傳混合演算法（即gdr ? ? ga演算法） ，使gdr演算法和ga演算法的優點互為補充，提高了收斂速度；最後本文闡述了用以上方法進行atm帶寬動態分配的過程。
  2. Secondly, after analyzing the strategic relations between the manufacturer and the supplier, the dissertation brings forward that the manufacturer must category the materials and select different supplier to built relations, explains the features and the advantages and disadvantages of strategic partnerships, and bring forward that the manufacturer must give some special request for the supplier strategic partner. thirdly, the dissertation analyses and summarize the existing methods of supplier selection, and point out their disadvantages. then, the dissertation apply the extenics to the dynamic selection model of the strategic supplier, building the evaluation indexes with implication systems, selecting strategic supplier with rhombus thinking method, looking for the suppliers with the extension of the matter - element, selecting the right supplier with the extension evaluating methods, replacing the dissatisfying features with the transformation matter - element method and making it more better

    本文首先對供應鏈管理和戰略合作夥伴關系的國內外研究現狀和發展趨勢進行了研究；其次，對製造商和供應商的戰略合作夥伴關系進行分析，指出了製造商需要對采購物料分類，針對戰略型物料選擇戰略合作夥伴建立戰略合作夥伴關系，闡明了建立戰略合作夥伴關系的特徵及優缺點，並提出了製造商對供應商戰略合作夥伴的具體要求；再次，對目前存在的供應商選擇方法進行了深入的分析和總結，指出了不足之處，並提出了應用可拓理論進行供應商戰略合作夥伴選擇的思路；然後，應用可拓理論建立供應商戰略合作夥伴選擇動態模型，其中，用蘊含系方法建立評價指標體系，用菱形思維建立選擇模型，利用物元的可擴性來尋找待評供應商，用可拓評價進行收斂，並用物元變換對已經選中供應商的不滿意特徵進行變換，使其完善；最後對r企業的供應商戰略合作夥伴選擇進行了本文模型的實證部分，使模型不僅具有理論意義，更具有實踐意義。
  3. In particular, as another important contribution, in the third algorithm the strong linear independence condition is slacked to a much milder assumption, by which the wachter - biegler phenomenon can be significantly avoided, and however all desirable convergence properties remain uneffected

    在這種情況下，病態wachier一biegler現象（參見[ 4 ）就會在演算法中發生。 tits等最近在2 ]中提出了一種雙重內點演算法，在保證收斂性質不受影響的前提下，該演算法大大減弱了以上線性無關條件。
  4. Among the adaptive beam - forming algorithms, the least mean square algorithm is widely used because it has a simple configuration and it is apt to come true and have nice convergence. on the other hand, it has a disadvantage that it converges slowly and there is a conflict between the fixed step and the convergence pace or the error in stabilization. so people have developed many improved least mean square algorithms which generally start from convergence, stabilization, misadjustment, and robustness and come to a formula about variational step in the end

    在自適應波束形成演算法中，最小均方（ lms ）演算法因結構簡單，易於實現，能穩定收斂而得到廣泛應用，但它也存在收斂速度受限的缺點：固定步長因子無法解決收斂速度和穩態誤差之間的矛盾。因此，人們提出了各種改進的最小均法演算法來解決這一問題。改進的最小均方演算法通常從如何改進收斂速度、穩態誤差、失調量和魯棒性等指標上出發，最後在新演算法最終表達式中的步長公式上變化。
  5. Pointwise estimate forthe rate of convergence of fourier - laplace series

    級數收斂速度的點態估計
  6. Reinforcement learning algorithms that use cerebellar model articulation controller ( cmac ) are studied to estimate the optimal value function of markov decision processes ( mdps ) with continuous states and discrete actions. the state discretization for mdps using sarsa - learning algorithms based on cmac networks and direct gradient rules is analyzed. two new coding methods for cmac neural networks are proposed so that the learning efficiency of cmac - based direct gradient learning algorithms can be improved

    在求解離散行為空間markov決策過程（ mdp ）最優策略的增強學習演算法研究方面，研究了小腦模型關節控制器（ cmac ）在mdp行為值函數逼近中的應用，分析了基於cmac的直接梯度演算法對mdp狀態空間離散化的特點，研究了兩種改進的cmac編碼結構，即：非鄰接重疊編碼和變尺度編碼，以提高直接梯度學習演算法的收斂速度和泛化性能。
  7. The following algorithms have been proposed and tested in the thesis : 1 frequency selective fading : combine the isomorphism between the input space and the output space and propose a new approach to blind equalization of the channel. compared with conventional methods, the new approach offers lower computational complexity, better performance, and more robust against the over - determination of the system order ; 2 time selective fading : a new approach to the equalization of time selective channel based on the zero - forced equalizer is proposed which is more simple in its structure of algorithm ; 3 time - varying channel : using the instantaneous mean value changes of the output signal to extract the information of channel variations and model it using ar model, kalman filter is then employed to track channel variations, it bears faster ability in tracking the variation of tv channels ; based on the isomorphism between the inputs and the outputs and some of the approaches using in mimo system, a new algorithm of equalization of simo time - varying channel is proposed, which also share the merits of being robust against the over - determination of the system order ; model the time - varying channel using the multi - resolution decomposition wavelets, and then a blind identification method based " on the model is proposed ; at last, a new model for equalization and identification of mimo system is proposed

    主要工作在以下幾個方面： 1 、針對頻率選擇性衰落通道：結合輸入輸出空間同構關系提出一種新的頻率選擇性通道均衡方法，與傳統方法相比，該方法計算量更小，收斂速度更快，性能更優，且對系統階次的過確定表現穩健，具有實際均衡應用價值； 2 、針對時間選擇性衰落通道：提出一種基於迫零均衡的時間選擇性通道均衡方法，演算法結構簡單； 3 、針對時變色散通道：利用瞬態均值曲線提取通道時變信息，對之ar建模，利用卡爾曼濾波器跟蹤時變通道抽頭變化，可以快速跟蹤通道變化；基於輸入輸出空間之間的同構關系以及多輸入多輸出系統的處理方法，提出了新的單輸入多輸出色散時變通道均衡與識別演算法，同樣具有對通道階次過確定保持穩健的優點；結合小波多解析度分析提出一種基於小波模型的通道盲識別演算法；研究時變的多輸入多輸出系統的盲均衡與盲反卷積問題，給出一種時變系統處理模型。
  8. At present, it is still in the middle stage of kuznets curve ; the differences of income in the interior of the east and the west are bigger than the middle, and they have been in the scope of income inequality which is regarded to be bigger by the international ; the phenomenon of " the club convergence " of income growth of the fanner between the east and the middle is obvious, but this can not appear between the east and the west, the west and the middle ; the differences in the interior of the high income area and the low income area are smaller than in the interior of the other areas in which the differences are increasing, this conforms to " the u " theory. by adopting theil model to analysis the general farmers incomes differences based on various regions, it is showed clearly that the farmers income inequality among different areas is the main reason to result to rural income differences. this part contents chapter 1, chapter 2 and chapter 3

    結果表明，無論從那個角度來看，改革以來，我國地區間農民收入均呈現差異擴大的態勢，而且擴大的速度快，幅度大，且目前仍處在庫茲涅茲所提出的「 u型」曲線的中期階段；東部與西部地區內部各省之間收入差異較大，目前已進入國際上公認的較大不平等的范圍，而中部內部差異較小，目前仍處于合理范圍之內；同時農民收入增長在東部與中部之間出現了新經濟增長理論所揭示的「俱樂部收斂」現象，而這一現象在東部與西部、中部與西部之間卻沒有發生；高收入與低收入水平區內部差異小，而介於中間收入水平的區域內部差異較大，且保持著較強的擴大趨勢，這一點基本上與「 u型」理論的結論相符。
  9. The results show that, when and 02 are in various intervals, the trajectories for systems ( i ) are either ultimately periodic or convergent to an equilibrium. the difference between system ( i ) and system ( ii ) only lies on the sign of self - feedback coefficient, but their asymptotic behavior of trajectories are much different. for the given initial value, the solutions for system ( ii ) all converge to the certain equilibrium

    結果表明：當_ 1和_ 2在不同的取值范圍內，系統（ ）的解或收斂到某一平衡點，或無限逼近於某個周期解，且部分_ 1 ， _ 2下的極限環全局穩定；系統（ ）與系統（ ）的區別僅在於自反饋系數的符號為負，但由此其解軌線的極限性態發生了很大的變化，對于確定的初值，系統（ ）的解均收斂到確定的平衡點。
  10. Point estimation about the convergence order of the summation operator of bernstein type

    型求和運算元收斂階的點態估計
  11. The problem is then fed to a nonlinear programming routine to solve for all parameters. the flight dynamics model with yawing motion of uh - 60a is given, scale factors of state variables and control variables are applied based on the consideration of algorithm convergence. then, the optimization of trajectories is expressed as optimal control problem with a performance index constrained by helicopter performance limits, faa regulation and so on

    首先，建立帶偏航運動的直升機點質量模型，然後對狀態變量和控制變量進行無量綱處理，因為狀態變量和控制變量量值相差較大，這使得優化演算法收斂效果變差，所以需要選擇適當的縮放因子，最後把直升機單臺發動機停車后的飛行軌跡表示成一個最優控制問題，即在考慮直升機性能限制和far規范后，指定一個優化性能指標，選擇合適的路徑約束，然後再解最優控制問題。
  12. Because the ins error equation is unstable, some initial states error will cause error floating and error accumulating, if the filter observations were only position error, kalman filter will converge very slowly, and some states error ( such as yaw error ) will be great. since the milemeter altimeter and piloting could only output position information, this paper put forward a method, firstly estimateing states and then kalman filtering, to improve filtering effect. simulation proved that this method could effectively reduce the system states error, quicken filtering convergence and improve filtering precision

    由於慣導系統（ lsins ）的誤差方程是發散的，某些初始狀態的誤差會引起誤差的漂移和積累，當觀測量只有位置誤差時，卡爾曼濾波的收斂速度很慢，某些狀態（如方位角）誤差很大，而以上除慣導外的其它導航傳感器直接提供的只是位置信息，為了改善濾波器性能，本文根據里程計等傳感器的特點，提出了首先對狀態做出估計，然後在狀態估計的基礎上，進行卡爾曼濾波的方法。
  13. Later on, after elaborating the disadvantages of the old methods in detecting and recognizing moving objects, a series of corresponding approaches are proposed, such as grid scan, local tracking bug and dynamic window in object tracing to reduce the huge data needed to be processed, maximum and minimum for selecting a proper segmentation threshold and improved conversion from rgb model to hsv and so on to decrease the influence of inhomogeneous lighting and the color noise, a bilinear interpolation in each quadrant to eliminate the bad effect on the recognition precise because of the distortions of the camera. after that, much emphasis is given on application study in pattern recognition with a feed - forward neural network. both the basic bp algorithm and improved bp algorithm in the study process are described in detail, and the later is used to quicken convergence speed and improve validity of the network

    然後，分析和闡明了傳統的運動目標檢測方法的不足，並在此基礎上結合研究中的實際實驗環境，提出了一系列解決方法，包括針對降低龐大數據量而提出的網格掃描、局部「跟蟲」追蹤和動態窗口掃描等目標檢測方法，針對實驗環境中光照不均和顏色干擾提出基於人機交互的最大最小值閾值選取方法和引入改進的rgb模型到hsv模型的轉換方法，為消除圖像畸變對識別精度的惡劣影響而採用的通過控制點進行雙線性插值進行畸變校正的方法；緊接著，概述了神經網路的發展歷史和幾種常用神經網路模型的特點，重點研究了前饋型神經網路在模式識別中的應用問題，詳細闡述了基本的bp演算法和學習過程中bp演算法的改進，從而使網路收斂速度更快，解決問題更有效，並在此基礎上，設計了一個基於bp神經網路的運動目標識別系統，給出了實驗結果。
  14. This algorithm improves confidence in se by estimating parameters and states at the same time. simulation results on test power systems which range in size from 4 to 118 buses, have shown the virtues as follows : getting unbiased estimation without detecting and identifying bad data in measurements ; solving state and parameter estimation for power system with good convergence and excellent robust property ; increasing the numbers of iterations a little bit with the test systems expanded ; estimating many transformer taps simultaneously and remaining the main state estimation ; keeping the estimated relative error within + 0. 1 % and processing efficiently equality constraints and ill condition with polynomial complexity

    對ieee ? 4 118節點系統和廣西主網進行的模擬結果表明： l1范數估計具有不良數據拒絕特性，當量測量中存在不良數據時，該演算法在不經檢測和辨識不良數據情況下仍是無偏估計，具有良好收斂性，所需迭代次數隨著問題規模擴大而增長極小；能夠同時估計多個變壓器抽頭，並保持狀態估計主體；在滿足可觀測性條件下，估計的相對誤差保證在0 . 1以內；能夠有效處理等式約束和病態條件，並具有多項式時間性。
  15. The article draws some value conclusion through observation of steady state, comparative static characteristics, transitional dynamics, saddle point stability, convergence patterns, convergent speeds and other basic parameters

    考察了模型的穩態特徵、增長率決定因素、比較靜態性質、轉移動態行為、鞍點穩定性、收斂特徵、收斂速度等基本性態，得出了有意義的結論。
  16. The computer simulation result shows that the method algorithm converges fast, ram requires less and compensation effect is good

    計算機模擬結果表明，該方法具有收斂速度快、需要動態存儲器（ ram ）少、補償效果好等特點。
  17. This method can guarantee the solution matrix of sylvester equation to be inverse and the sum of the input gain norm and the observer gain norm is the minimum. for the linear systems with unknown parameters, we identify the parameters using hopfield network, then design the observers using the identified parameters, the exponential convergence of adaptive observer is also proved. for the linear time - varying systems, a new network to solve the time - varying sylvester equation is proposed, we analysis it ' s convergence and robustness, then, deign the linear time - varying observer using this network model, and we discuss the convergence of the observer and ruboustness to unknown match parameters

    同時保證了sylvester方程的解矩陣的可逆性和觀測器的增益矩陣與輸入矩陣范數的和最小；在設計線性時不變自適應觀測器時，首先利用系統的輸入、輸出數據設計一個hopfield網路參數估計器，進一步設計狀態觀測器，證明了參數估計器和狀態觀測器的指數收斂性；為了仍然從神經優化計算的角度設計線性時變系統的狀態觀測器，最後介紹了一種求解時變sylvester矩陣方程的神經網路模型，分析了它的收斂性和魯棒性，然後利用該網路設計時變狀態觀測器，進一步討論該觀測器的在系統存在未建模不確定和外部噪聲時的魯棒性；最後給出了一種基於分離性原理和hopfield網路觀測器的狀態反饋閉環系統的結構，分析了該閉環系統的特點；對于每一種設計方法都給出了相應的數值模擬例子來進一步表明所提方法的可行性和有效性。
  18. Some improvement are made on the ground of direct method, such as adding the outer iteration and adopting dynamic convergent criterion to accelerate it, introducing the iteration step factor, adding the flux criterion to main iteration, etc. compared with the other two computation methods, the modified direct method has the advantage of wide calculational range and high precision

    在直接法的基礎上作了幾點改進，如引入參數使得主迭代中增加了外迭代並採用動態收斂判據加速迭代過程，引入步長調整因子f ，給主迭代收斂條件增加了通量判據等。
  19. The main contributions of this thesis are as follows : ( 1 ) the conventional minimum output energy ( moe ) algorithm has the global convergence but its stable excess mean square error is high and it ca n ' t converge to the minimum mean square error ( mmse ). in this thesis an equivalent decision feedback moe ( df - moe ) algorithm variable step - size is presented and analyzed

    主要工作可以概括如下： （ 1 ）針對最小輸出能量檢測器（ moe ）具有全局收斂，但是穩態剩餘均方誤差比較大，無法收斂到最佳mmse檢測器的特點，提出並分析了一種等效的判決反饋變步長moe檢測演算法（簡稱df - moe ： decidedfeedbackmoe ） 。
分享友人
分享到 Line

分享到臉書