策略迭代法 的英文怎麼說

中文拼音 [cèlüèdiědàifǎ]
策略迭代法 英文
policy iteration
  • 策 : Ⅰ名詞1 （通「冊」 古代寫字用的竹片或木片） bamboo or wooden slips used for writing on in ancient ...
  • 略 : Ⅰ名詞1 （簡單扼要的敘述） summary; outline; brief account; résumé: 節略 memorandum; aide mémoire;...
  • 迭 : Ⅰ動詞（輪流; 替換） alternate; change Ⅱ副詞1 （屢次） repeatedly; again and again 2 （及） in tim...
  • 代 : Ⅰ動詞1 （代替） take the place of; be in place of 2 （代理） act on behalf of; acting Ⅱ名詞1 （歷...
  • 法 : Ⅰ名詞1 （由國家制定或認可的行為規則的總稱） law 2 （方法; 方式） way; method; mode; means 3 （標...
  • 策略 : 1 （制定的行動方針和斗爭方式） tactics 2 （講究斗爭藝術） tactful 3 [數學] （對策） policy; strat...
  1. For problems whose objective function or constraint functions have sharp curves on their contour maps ( such as the rosenbrock ' s function which has banana shape contours ), monotonicity may cause a series of very small steps, causing a huge number of iteration to reach their solutions. by using the nonmonotone technique, we get the sequence of successful interative point which should make the objective function mono - tonically decreasing. hence, we use both trust region strategy and line search technique and make each iterate generate an acceptable trial step in interior feasible set as next interative point

    我們利用非單調技術得到使目標函數非單調下降的迭代點，因為非單調克服高度非線性化函數的求解問題，從而避免了只使用單調搜索在「峽谷」現象局部最優解被卡的情況，我們用信賴域策略和非單調線搜索技術相結合的方法，使演算法產生的迭代步落在可行域內點，同時又在信賴域內滿足接受準則。
  2. The control equation consist of completely coupled deformation equation, seepage equation, conduction and convection equation of heat, which describe the reservoir non - liner performance. 2, present the detailed strategy and methods to solve this mathematics model, the basic strategy as follows : regard the deformation equation ? seepage equation conduction and convection equation of heat as separate system, and solve the equation by coupling and iterative method ; disperse the control equation in the geometry field by the finite element method ( galerkin ), and in the time field by the finite difference method : programme the computer program on this task ; when the solving, take the combinative measures of the thick and thin mesh ; successfully carry out the numerical simulation in vast 3d heat extraction system of hdr

    2 、提出了高溫巖體地熱開發的固、流、熱多場耦合數學模型的數值解法，其基本的求解策略是：將固體變形，流體滲流與溫度場方程看成獨立的子系統，耦合迭代求解；利用有限元離散（ galerkin ）方法將控制方程在幾何域上離散，並用差分法得到時間域上的離散方程，並在此基礎上，編制了相應的計算機源程序；有限元求解中，為減小邊界效應的影響，在計算中採取粗細網格結合的方法，順利地實現了高溫巖體地熱開發三維巨系統的數值模擬。
  3. This thesis suggests a process considered minimizes the population size as similar individuals occur in the fitter members of the population, which helps reduce the execution times for ga by removing the redundancy associated with the saturation effect found in the later generation. this thesis uses a method that adds dynamic penalty terms to the fitness function according to the optimal degree of solutions, so as to create a gradient toward a feasible suboptimal or even optimal solutions. on the basis of the difference of the biggest and the smallest of fitness of individual, modifying the fitness function in order to convergence is a satisfaction

    動態調節種群大小，去掉遺傳演算法在迭代後期搜索產生的過多相似個體，達到減少計算時間的目的；按照解的優劣程度給適應度函數增加一個在ga搜索過程中動態改變的可變罰函數，給搜索最優解創造一個梯度，使遺傳演算法收斂到可行的較優解或最優解；根據適應度值最大和最小個體的差修正適應度函數，使適應度函數值適中不容易造成收斂太快、局部收斂或根本不收斂而變成隨機搜索；為了避免「近親繁殖」採用競爭擇優的交叉操作；利用并行遺傳演算法的思想，提出一種自適應多子種群進化策略；提出人口汰新政策來解決類似甚至相同的個體的情況發生。
  4. In order to adapt the maintenance activities well, this paper makes some ameliorations, as follows : ( 1 ) it is considered not so pertinent in this paper to take road user cost into account of total cost, and an advice of using pavement serviceability performance for benefit is given. ( 2 ) when evaluating the average of the road network serviceability, this paper put forward to using weighed geometric mean to calculate pqi. ( 3 ) bringing in a new idea of checking the reasonability of classifying pavement state and typical measure by transfer probability

    為盡可能使其反映實際的養護規律，本文在前人研究的基礎上，提出以下改進： （ 1 ）在建立費用模型時，認為在我國現階段，考慮用戶費用有不妥之處，提出以路面使用性能作為用戶費用的映象，以路面使用性能作為效益的評價指標； （ 2 ）在評價路網平均服務水平時，提出用加權幾何平均值來計算路面使用性能綜合評價指標pqi ； （ 3 ）提出了利用轉移概率檢查狀態的劃分和典型措施的合理性的新思想； （ 4 ）在求解markov決策模型時，通常使用的方法是線性規劃法，本文闡述了另外兩種求解的方法：策略迭代法和整數規劃法。
  5. The recursive modification algorithm is mainly based on the ideal of stochastic approximation and truncation technique to obtain modification strategy for estimated parameters, it is proved that these modification parameters turn out to be a constant vector

    其中迭代的參數修正策略主要是基於隨機逼近思想和截尾技術給出了修正向量的構造演算法。
  6. In order to solve optimization calculation efficiency problem of large structures with more than ten thousand degrees of freedom, based on the evolutionary structural optimization method, combining with the iteration method of structural displacement calculations, a strategy and its algorithm of large three - dimension structural topology optimization designs were proposed

    摘要為了提高擁有數萬個單元以上的三維機械結構拓撲優化的計算效率，基於漸進結構優化方法，並結合結構位移計算的迭代方法，建立一套三維機械結構拓撲優化的求解策略和演算法。
  7. In this paper, we consider the optimal parameter vector a of the modified incomplete gauss _ seidel method ( migs ). we prove that the spectral radius function of the iterative matrix t of migs with parameter vector is strictly monotonic decreasing with respect to a satisfying 0 e if the classical gauss _ seidel method converges for a z _ matrix. some properties of the left and right eigenvectors corresponding to the largest eigenvalue in modulus are given, too. these results are useful to find an optimal parameter for migs

    目前主要的方法有兩類：一是充分利用所給矩陣a的特點，採用適當的主元素選取策略，使分解出的因子盡可能地保持稀疏性；二是迭代法。對于第二種方法，迭代矩陣的選取具有決定作用。只有選取的迭代矩陣的譜半徑小於1才能保證迭代法收斂。
  8. In the solution, the 0 - 1 integer and real number mixed encoding technique was employed to describe an artificial fish ; behaviors of a fish were dispatched by its body energy status ; the following behavior was described by the greedy method where moving step is direct ratio to a fish ' s hungry degree ; the lowest survival body energy controlling technique was used to realize escaping policy from locally optimum positions ; the maximum iterating times and the changing degree of the optimum solutions during iterating were used to control the terminating time

    在解算過程中，人工魚個體採用0 - 1整數和實數混合編碼方法描述；用人工魚體能累計和消耗程度來調度其行為；採用與饑餓程度成正比的移動步距的貪婪法描述個體追尾行為；採用最低生存體能控制來實現局部最優解逃逸策略；採用最大迭代次數和迭代過程中最優解平均值變化程度來控制迭代終止時機。
  9. The improvements in this thesis include the hybrid code method, the method of generation of initial populations, substituting the children for parents by combination of the simulated annealing algorithm and niche, adding some new chromosomes to ensure the population ’ s diversity and using the adaptive probability of crossover and mutation

    本文的改進方案主要涉及混合編碼方式，初始種群的產生方式，採用結合模擬退火演算法和小生境思想的替代策略，迭代過程中添加新染色體，採用自適應交叉、變異概率等。
  10. This paper revealed an optimized strategy to search for systolic array transformations for nested loops, when the upper and lower bound of inner index was affine functions of there outer loop indexes

    對于多重循環中，當內層迭代的上下界為外層迭代的仿射函數時，介紹了搜索脈動變換的一個優化策略和相應的自動化演算法。
  11. The objective function is the sum of the daily cost for pc and the possibility - weighted cost for ec. in this thesis, a two - layer tsc optimization strategy is proposed. unstable contingencies are firstly classified into subsets according to their ums

    進而提出大電網穩控協調的「解禍迭代一聚合協調」全局優化策略，構造了大電網暫態穩定控制協調優化的統一演算法流程。
  12. In the parallel computation, there are two strategies, namely sbs and ebe ; for the nonlinear problem solution, there are direct and iterative algorithms, but for dynamic analysis there are explicit and implicit integration, and high precision direct integration

    有限元并行計算有sbs 、 ebe兩種策略，非線性求解有直接與迭代解法，動力時程分析有顯式、隱式和精細時程積分法。
  13. In the thesis a new count expression of mesh fractal dimension is introduced, on the basis of it the storage and operation counts of directs solution methods, assembly global stiffness iterative methods and ebe iterative methods have been analyzed and compared, a conclusion has been obtained that the storage and operation counts of direct solution methods is the most, the storage counts of the ebe iterative methods is the least. estimating compute scale for ebe iterative methods at shuguang 2000 parallel computer, it is make out that compute scale is large in the extreme if make the most of the memory of every processor. by analyzing the parallel compute performance of 2d upend - step flow calculation, it is proved that ebe iterative methods can achieve upper speedup rations

    本文還提出了一種網格分維新的計算公式，並基於網格分維對直接解法，組集總剛矩陣的迭代解法和基於ebe的迭代解法等三種方法的存儲量和計算量進行了分析和比較，得出了直接解法無論存儲量和計算量均最大，基於ebe迭代解法存儲量最小的結論；在曙光2000并行機上估計了基於ebe迭代解法的計算規模，說明只要充分利用各處理機的內存，可解問題的規模是非常巨大的；對二維倒臺階流并行計算進行了性能分析，表明基於ebe策略的迭代解法能獲得較高的加速比。
  14. Discussion and testifying were made to the convergence of the algorithm under the condition of having constrains in objects " outputs. one new algorithm, which can maintain the convergent speed under such constraint conditions, was presented. using iterative learning control in real job of industrial robotic manipulators, we h

    本文對對象輸出有限制情況下迭代學習演算法的收斂性做了討論和證明，並且提出了一種在這種情況下，能相對維持收斂速率的迭代學習律的改進策略。
  15. To solve the problem that the ph distribution proposed changes the state space of system, the value iteration algorithm for the semi - markov decision process is improved to get the optimal inspection and maintenance policy

    將位相型（ ph ）分佈引入模型后，決策過程的狀態空間發生變化，為了獲得適用於原有模型假設的檢測與維修優化策略，提出了一種改進的值迭代演算法。
  16. To make the system perfectly recur the ideal response, the input signal should be iteratively compensated. this paper studied the method of identifying the transfer function and the arithmetic to iterate the input. the results shows that the controlling measures adopted in the

    為了使系統的輸出能很好的復現期望的響應信號，本文研究了利用遠程參數控制對系統頻響函數辨識以及對驅動信號的迭代演算法，從理論上說明了該控制方法實施的可能性，為以後振動臺控制策略方面的研究打下了一個良好的基礎。
  17. An evolutionary iterated local search algorithm with the features of multiple neighborhood structures and population - based evolution is designed to solve the model

    提出了一種基於多個鄰域結構和種群進化策略的進化迭代局部搜索演算法，用於求解該模型。
分享友人
分享到 Line

分享到臉書