收斂帶 的英文怎麼說
中文拼音 [shōuliǎndài]
收斂帶
英文
convergence strip-
Globally convergent of a trust region algorithm with memory
帶記憶信賴域方法的收斂性分析Global convergence of non - quasi - newton merhods with a class of inexact line searching
帶一類非精確線搜索的非擬牛頓演算法的整體收斂性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。A quadratic convergent iteration method with adjusted parameters without derivative for solving nonlinear equations is discussed in this paper
摘要討論一種解非線性方程的具有變參數的不帶導數的二階收斂迭代法。The main characteristics and advantages are : on one hand, we adopted the low - complexity bussgang algorithm, and did blind estimation to ofdm sub - channel according to the mean square error criterion ( mse ) and peak distortion criterion ; on the other hand, we used the differential detection, which accelerates the convergence speed and avoid the error transmission problem resulted from the bussgang algorithm
這兩種演算法一方面採用了低復雜度的bussgang自適應演算法，分別依據均方誤差準則和峰值失真準則對ofdm系統子通道進行盲估計；另一方面通過引入差分檢測技術，加快了演算法收斂速度，克服了bussgang演算法帶來的誤差傳播問題。Abstract : in this paper a new identification model constructed by neural networks with modified inputs and stable filters is presented for continuous time nonlinear systems in order to reduce the inherent network approximation errors. an adaptive law with projection algorithm is employed to adjust the parameters of networks. under certain conditions, convergence of the identification error is proved
文摘:在用神經網路進行系統建模時，建模誤差的存在是難免的.為了減小這種誤差，本文對連續時間非線性系統提出了一種新的神經網路辨識模型，它是由帶有輸入修正的神經網路和穩定濾波器組合而成.文中給出了權值的學習演算法，即權值是根據辨識誤差的投影演算法來改變，證明了在一定條件下辨識誤差的收斂性Finally, in the third section, by constructing some functional which similar to the conservation law of evolution equation and the technical estimates, we prove that in the inviscid limit the solution of generalized derivative ginzburg - landau equation ( ggl equation ) converges to the solution of derivative nonlinear schrodinger equation correspondently in one - dimension ; the existence of global smooth solution for a class of generalized derivative ginzburg - landau equation are proved in two - dimension, in some special case, we prove that the solution of ggl equation converges to the weak solution of derivative nonlinear schrodinger equation ; in general case, by using some integral identities of solution for generalized ginzburg - landau equations with inhomogeneous boundary condition and the estimates for the l ~ ( 2 ) norm on boundary of normal derivative and h ~ ( 1 ) ' norm of solution, we prove the existence of global weak solution of the inhomogeneous boundary value problem for generalized ginzburg - landau equations
第三部分：在一維情形，我們考慮了一類帶導數項的ginzburg ? landau方程，通過構造一些類似於發展方程守恆律的泛函及巧妙的積分估計，證明了當粘性系數趨于零時， ginzburg ? landau方程的解逼近相應的帶導數項的schr （ ? ） dinger方程的解，並給出了最優收斂速度估計；在二維情形，我們證明了一類帶導數項的廣義ginzburg ? landau方程整體光滑解的存在性，以及在某種特殊情形下， gl方程的解趨近於相應的帶導數項的schr （ ? ） dinger方程的弱解；在一般情形下，我們討論了一類ginzburg ? landau方程的非齊次邊值問題，通過幾個積分恆等式，同時估計解的h ~ 1模及法向導數在邊界上的模，證明了整體弱解的存在性。Considering the one - sidedness and inaccuracy of knowledge discovery only from single - color database, an approach is proposed to discover knowledge from 1331 groups of mix - color database with partial least - square regression, based on measuring and learning 400 groups of single - color database. by this method, the mean error decreases when converting from rgb to cmyk, the precision of color matching is improved, and the automatic and general problem in color matching is further solved
本文基於統計學習理論構造了一種快速自適應隨機搜索演算法，證明了演算法的收斂性.給出了一種簡易實用的寬帶天線匹配設計新方法.應用該自適應演算法進行天線匹配設計，不僅演算法簡單，易於編程實現;而且能夠快速設計出具有較好性能的匹配網路，非常適用於各種短波、超短波天線的匹配設計問題This paper studies nonlinear dynamic problems of tall building structures, first, constitutes linear dynamic equation and elasto - plastic dynamic equation of structure by using qr method, later, solves the dynamic equation by using spline unconditional stable algorithm, last, programs corresponding computer programs with c program language, and calculates some example and a tall building in constructing the courses and the results prove that qr method is not only simple in calculating and fast in constringency rate, but also precise, that qr method is a effective and economic new method
本文研究高層建築結構彈塑性動力問題，先利用qr法建立了結構線性動力方程及彈塑性動力方程，然後利用樣條無條件穩定演算法求解了動力方程，最後利用c語言編寫了相應的計算程序，計算了一些例題和分析了一個工程實例。 qr法在理論上及方法上不用有限元法及流動法則，避免了這些傳統方法在分析非線性問題時所帶來巨大困難的缺陷。計算結果表明，這種方法不僅計算簡便，而且精度高，收斂速度快，是一種經濟有效的新方法。In chapter 3, we give an equivalent form of semi - infinite programming, and a locally convergent ssle method is proposed for sip. we only need solve a linear system equations and a subproblem with a parameter per step, also a modified algorithm which saves cost of computations is given, at the end of the paper, we give a proof of the convergence for the algorithms
第三章通過適當的變形，得到半無限規劃問題的一個等價形式，並給出一個局部收斂的序列線性方程組演算法，這個演算法在每一步，只需求解一個線性方程組和一個帶參數的非線性子問題，證明了演算法的收斂性，同時，給出了一個修正演算法，與前面演算法相比較，修正演算法節約一定的計算量，同樣具有較好的收斂性。The paper puts forward the clustering algorithm includes : clustering based on grid and iterative, enhanced clustering algorithm base on density and k - medoids, enhanced k - means algorithm ( optimize chooseing consult _ points in iterative process ), enhanced clustering algorithm base on distance. they can overcome many limitations ( some traditional algorithms terminate in local optimization. many results of cluster are roundness, too many times in partition iterative process ), which are related to the static architecture of traditional model
在傳統聚類演算法的基礎上，結合我們科學數據挖掘的應用對象?分子動力學數據，提出了迭代網格聚類演算法， k -平均和基於密度結合的聚類演算法，迭代過程中優化選擇中心點的k -平均方法，以及改進型的基於距離的聚類演算法等模式識別方法，能夠解決傳統演算法帶來的諸多問題（比如一些傳統的聚類演算法常常收斂于局部最優，發現都模式都趨近於球形，劃分方法中迭代次數過多帶來的效率問題） 。Trust region methods has drawn our attention increasingly as one approximating method guarantees general convergence. its basic step is as follows : first, we set trust - region radius, and then we get a trial step produced by constraint quadratical model
信賴域方法是近年來日益受到關注的一種新的保證演算法整體收斂性的逼近方法，信賴域方法的基本步驟是首先指定一個信賴域半徑，然後用帶約束的二次模型來確定搜索方向與步長大小。Astringent. the characteristic flavour of wines which produces an unpleasant chemical stimulus in the mouth, due to an excessive level of ethyl acetate
澀的、收斂性的：葡萄酒中乙酸乙酯含量過高時的典型特徵，給口腔帶來一種不適的化學刺激感。Multi - channel deconvolution, based on multi - channel signal analysis, is presented to recover receiver function from three - component teleseismic p waveforms clustered within certain backazimuth and epicenter distance, to increase the resolution of receiver function. wavelet inversion is also introduced to broadband receiver function, to solve for nonuniqueness cased by linear approximation of nonlinear problem, and broadband receiver function is decomposed into multi - resolution versions, and the inversion result for the low - resolution version is iteratively taken as the initial model for the high - resolution, and gradually obtain the inversion model for broadband receiver function
為克服廣義線性反演對初始模型的依賴性，並保持其快速收斂的特點，本文首次將小波變換引入到接收函數波形反演，將接收函數分解到不同的分辨尺度上，分別進行廣義線性反演，並將低解析度的反演結果作為高解析度反演的初始模型，漸近地獲取寬頻帶接收函數的模型，克服了非線性問題線性化帶來的非唯一性。First of all, the characteristics of different adaptive algorithms and equalizers are theoretically researched. then digital microwave communication system model is build by simulink. based on this model, the convergent behavior and bit error rate performance of three equalizers ( nlms, kalman and rls ) are compared in multi - path fading channel
為了達到這個目的，首先需要從理論上對各種自適應均衡演算法以及自適應均衡器的特性進行深入研究;然後利用simulink建立接近實際環境的數字微波通信系統，在多徑衰減通道環境下比較了三種基帶均衡器（ nlms ， kalman和rls ）的收斂特性以及系統誤碼特性，並比較了nlms基帶均衡器和通帶均衡器的性能。Third, by means of smale, s point estimates theorem, the existence and the convergence theorem of broyden, s iteration is given, the method has less calculation and the speed of super linear convergence for solving the nonlinear equations with nondifferential terms under the point estimates or weak condition
第三，本文利用smale提出的點估計理論，給出了在點估計和弱條件下，用計算量小、具有超線性收斂速度的broyden方法來求解帶不可微項的非線性方程，給出了存在性收斂性定理及相應的證明。We also compute naca0012 airfoil and certain three - foil flow field separately with explicit 4 - step runge - kutta scheme and the implicit lu - sgs scheme and results show the latter is much more efficient. in the end, the capability of the algorithm to deal with multi - bodies in relative movements is demonstrated by a dropping drank problem. it turns out that the unstructured overset grid is valuable and promising in practical application
同時為了驗證隱式lu - sgs格式的收斂效果，我們對基於單塊網格的naca0012翼型與某三段翼型在不同的來流條件下，分別用帶隱式的lu - sgs格式與顯式的四步runge - kutta格式的解算器進行了數值模擬。Secondly, ground on the duct noise control, the delayed - lms algorithm, which is actulized easily, is brought forward, then the simulation on pc is achieved. it is shown from the computer simulation that for the single frequency noise, there will be a good effect when choosing the suitable parameters ; but for the narrowband noise, the effect will become worse with increasement of the bandwidth. this result is helpful to analyze the parameters influencing anc
結果表明： （ 1 ）隨著濾波器長度的增大，系統收斂速度變快； （ 2 ）隨著值的增大，系統的收斂速度增大，但是當值超過某一范圍時會引起系統的發散； （ 3 ）次級通路的延遲數與初級噪聲頻率有關； （ 4 ）對于窄帶信號， dlms演算法可以取得一定的降噪量，但是隨著窄帶信號帶寬的增大，降噪效果明顯變差。Equidistant interpolation can give rise to convergence difficulties when the number of interpolation points becomes large. this difficulty is often referred to as runge ' s phenomenon
等距點插值會帶來收斂困難當插值點數量增加。這一困難被稱為龍格現象。In the first chapter, we narrate the characteristic of convertible bond, give some clues about development and actuality of the market and its pricing theory ; in the second chapter, we introduce modeling idea and some material problems in the model in detail, draw the yield curve which is very important to the model by spline method ; in the third chapter, we first explain the basic idea and convergent speed of monte carlo method, then, give the mathematical description for financial market, prove equivalence of non - arbitrage market, existence of risk neutral probability measure in the market and the price process of underlying asset is a martingale ; in the forth section, we introduce how to simulate stock price path by monte carlo method in detail, based on foregoing result, we prove the path is a martingale, thereby, the model is logical
本文第一章先對可轉債的特點、市場發展和現狀及其定價理論的發展和現狀作一概述;第二章詳細介紹了建模思想和模型中的一些具體問題，利用spline方法繪出了在模型中具有重要作用的收益曲線;第三章首先敘述了montecarlo方法的基本思想和有關其收斂速度的一些性質，然後從數學的角度給出了對金融市場的描述，證明了市場無套利、市場存在風險中性概率測度及標的資產價格過程為鞅的等價性;在第四節中，對用montecarlo方法模擬的帶跳股價路徑作了詳細介紹，並利用前兩節的結論證明了模擬的帶跳股價路徑為一個鞅過程，從而保證了模型在理論上的合理性。分享友人