convergence errors 中文意思是什麼
-
Compared with the classical bp algorithm, robust adaptive bp algorithm possesses some advantages as following : ( 1 ) increasing the accuracy of the network training by means of using both the relative and absolute residual to adjust the weight values ; ( 2 ) improve the robustness and the network convergence rate through combining with the robust statistic technique by way of judging the values of the samples " relative residual to establish the energy function so that can suppress the effect on network training because of the samples with high noise disturbances ; ( 3 ) prevent entrapping into the local minima area and obtain the global optimal result owing to setting the learning rate to be the function of the errors and the error gradients when network is trained. the learning rate of the weights update change with the error values of the network adaptively so that can easily get rid of the disadvantage of the classical bp algorithm that is liable to entrap into the local minima areas
與基本bp演算法相比，本文提出的魯棒自適應bp演算法具有以下優點： （ 1 ）與魯棒統計技術相結合，通過訓練樣本相對偏差的大小，確定不同訓練樣本對能量函數的貢獻，來抑制含高噪聲干擾樣本對網路訓練的不良影響，從而增強訓練的魯棒性，提高網路訓練的收斂速度； （ 2 ）採用相對偏差和絕對偏差兩種偏差形式對權值進行調整，提高了網路的訓練精度； （ 3 ）在採用梯度下降演算法對權值進行調整的基礎上，通過將學習速率設為訓練誤差及誤差梯度的特殊函數，使學習速率依賴于網路訓練時誤差瞬時的變化而自適應的改變，從而可以克服基本bp演算法容易陷入局部極小區域的弊端，使訓練過程能夠很快的「跳出」局部極小區域而達到全局最優。 -
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 -
Convergence of iterative sequence with errors for asymptotically nonexpansive type mappings
漸近非擴張映象具誤差的迭代序列的收斂性 -
Convergence of generalized lipschitz generalized - accretive operator by ishikawa iteration sequences with errors and applications
迭代序列的收斂性及應用 -
The convergence theorem of the triple ishikawa iterative with errors for nonexpansive mappings in a uniformly convex banach spaces
型的三重迭代序列的收斂性 -
The solution formula of dwell time is deduced by using the pulse iteration method and the smoothing fourier transition method, and the formula can be simplified referring to workpiece shape. according to the anastomotic status of the tool and workpiece surface, we present the new model of the inherent relation among local pressure density, removal rate, convergence ratio and residual errors, and the corresponding formulas are given. the forming cause of the edge effect in the polishing process is analyzed, and the means which reduces the edge effect is put forward
2 、針對計算機控制光學表面成形（ ccos ）的加工方式、誤差收斂特點，研製了去除函數呈高斯分佈的雙轉子結構研拋模；採用脈沖迭代法、平滑因子傅立葉變換法推導出駐留時間的演算法，求解過程中根據工件面形的特點作了各種形式的簡化；提出了工件表面和研拋模的吻合誤差與局部壓強、材料去除率、收斂比之間內在關系的數學模型，並推導出相應的計算公式；分析了邊緣效應產生的原因，在加工過程中採用相對壓力因子對去除函數進行修正，可以消除邊緣效應的影響。 -
Traditional inertial mechanized - platform uses velocities to damp the system attitude to improve the precision of attitude, when the system acceleration is small. referring to the idea, this paper designeda damp kalman filter in strap - down attitude heading reference system ( ahrs ). the new method makes use of 3 - d accelerometer ' s measurements to estimate the system attitude, which is measured to compensate attitude errors. because the acceleration affected the precision of fiher directly, the fuzzy adaptive system was presented. the fuzzy logic inputs are three accelerations and the output is to control the measurement noise covariance matrix. simulations and experimental results prove that the damp algorithm can damp most of schuler oscillation and foucauh oscillation, so that to assure the filter convergence and efficiently improve the precision of strap - down ahrs
在系統機動性不強的情況下，傳統的平臺內阻尼演算法將系統本身的速度信息通過阻尼網路加到系統中，達到提高姿態角精度的目的.將這種平臺內阻尼的思想引入到捷聯慣性航姿系統中，在系統加速度較小的情況下，利用加速度計的輸出估計系統姿態角，通過卡爾曼濾波的形式補償系統姿態誤差.由於加速度的大小直接影響濾波器精度，本文設計了模糊自適應卡爾曼濾波演算法，根據三軸加速度計的輸出調整內阻尼量測誤差方差陣，從而避免了濾波器的發散.模擬和實驗驗證，內阻尼演算法可明顯抑制舒勒周期振蕩和傅科周期振蕩，避免了系統姿態漂移，有效提高了捷聯慣性航姿系統的精度 -
We construct some new iterative algorithms with errors for solving these generalized fuzzy variational inclusions by using the resolvent operator technique for maximal - monotone mappings and prove the convergence of iterative sequences generated by the algorithms
為了利用-極大單調映象的預解運算元技巧求解這類廣義模糊變分包含，我們建立了一些新的含誤差的迭代演算法，並證明了由這類演算法產生的迭代序列的收斂性。 -
The main conclusions are as follows : through the different structure and algorithm application of bp model in the predication of regional groundwater hydrology, the hidden layers number, learning rates, neuron number of hidden layer and training errors of bp model and accelerated bp algorithm which influence the convergence effects and test results of model are compared each other. some application technology related parameters of bp structure design are put forward
論文取得了以下主要成果：通過不同bp網路結構和演算法在區域地下水文預測中的實例研究，重點比較了不同層次結構、隱層單元數、學習速率、訓練收斂誤差等4個基本要素及不同演算法、不同樣本容量等對模型收斂效果、模擬、檢驗與預報結果的具體影響。 -
1. this thesis puts forward a new model ? the model of crop ' s water rising and vaporizing regression nerve networks ( nn ). this model adds a deviation cell based on existing bp ( nn ) the effect of the deviation cell is to amend training errors of the nn, to increase convergence speed of the nn, and to reach expectant objective
本文主要完成了以下幾個方面的內容： 1 、提出了一個新型的模型? ?作物蒸騰蒸發回歸神經網路決策模型，該模型以已有的bp神經網路為基礎，增加了一個偏差單元，該偏差單元的作用是修正網路的訓練誤差、提高網路的收斂速度，從而達到預期的目的。 -
Abstract : iterative learning control is an effective approach to the control of processes that are repetitive in nature. in this paper, an open - closed - loop pi - type iterative learning control scheme for the precise tracking control of a class of discrete nonlinear time - varying systems over a finite time interval is presented. the scheme updates control input with tracking errors of both current and last iterations simultaneously. sufficient and necessary conditions which guarantee the convergence of the scheme are given and then proved with inductive method. finally, the conditions are verified with simulation results
文摘:對于具有重復運動性質的對象，迭代學習控制是一種有效的控制方法.針對一類離散非線性時變系統在有限時域上的精確軌跡跟蹤問題，提出了一種開閉環pi型迭代學習控制律.這種迭代律同時利用系統當前的跟蹤誤差和前次迭代控制的跟蹤誤差修正控製作用.給出了所提出的學習控制律收斂的充分必要條件，並採用歸納法進行了證明.最後用模擬結果對收斂條件進行了驗證 -
The problem to improving the rate of convergence and the accuracy of tracking of ilc for deterministic linear systems is considered, in the meanwhile, the effects of the plant characteristics, various types of disturbances, errors in initial conditions and the " slowly " varying desired trajectories on the convergence and performance of ilc for uncertain linear and nonlinear systems are also investigated
針對確定的線性系統，主要研究能夠提高演算法的收斂速度和跟蹤精度的迭代學習控制技術；針對不確定的線性和非線性系統，主要考慮系統的特性、各種干擾、初始狀態偏移和不確定的未建模動態以及緩慢變化的期望軌跡對迭代學習控制過程收斂性和跟蹤性能的影響。 -
Section 4 is devoted to give some new iterative algorithms with errors. with some noncompact constraint qualifications, we prove the existence of solution for the generalized nonlinear fuzzy quasi - variational - like inclusions involving maximal - monotone mappings and the convergence of iterative sequences generated by the iterative algorithms. in the last, we discuss the convergence and stability of perturbed iterative algorithm for solving a class of generalized nonlinear quasi - variational - like inclusions involving maximal - monotone mappings
第四部分，首先給出一些新的含誤差的迭代演算法，然後證明這類含-極大單調映象的廣義非線性模糊擬似變分包含在非緊條件下的解的存在性和由此類迭代演算法產生的迭代序列的收斂性；第五部分，我們討論了求解一類含-極大單調映象的廣義非線性擬似變分包含的擾動迭代演算法的收斂性和穩定性。 -
According to the theory of fixed - point stability in nonlinear sciences, the convergence property is discussed. it is concluded that the analytical solutions can be figured out by iteration in principle, though iterative solutions exhibit significant errors when ri approaches 0. 2
根據非線性科學中關于映射不動點穩定性的理論，討論了迭代演算法的收斂性，表明迭代演算法在理論上可以收斂到解析解，但在
分享友人