【正文】 ture, the network function is determined largely by the connections between elements. We can train a neural network to perform a particular function by adjusting the values of the connections(weights)between elements. Commonly neural networks are adjusted, or trained, so that a particular input leads to a specific target output based on a parison of the output and the target, until the network output matches the target. Some points on relation curve between teeth number Z2 and the profile factor Yf of worm gear are selected as training sample data, the Fast Back Propagation are adopted to train feedforward networks, the weights and biases of the network are updated. Then neural networks is simulated by the function of Neural Networks Toolbox in MATLAB. Program as follows:Z2=0:10:90。YF=[,,,]n1=5。[W1,b1,W2,b2]= initff(Z2,n1,’tansig’,YF,’purelin’)。fpd=100。mne=20000。sse=。lr=。tp=[fpd, mne, sse, lr]。[W1,b1,W2,b2,te,tr]=trainbpx(W1,b1,’tansig’,W2,b2,’purelin’,Z2,YF,tp)y=simuff(Z2,W1,b1,’tansig’,W2,b2,’purelin’)VI..SOLVING USUAL OPTIMIZATION MATHEMATICAL MODEL BY GENETIC ALGORITHM TOOLBOXOne key to successfully solving many types of optimization problems is choosing the method that best suits the problem. The Genetic Algorithm and Direct Search Toolbox is a collection of functions that extend the capabilities of the Optimization Toolbox and the MATLAB? numeric puting environment. The Genetic Algorithm Toolbox includes routines for solving optimization problems using Genetic algorithm Direct search. These algorithms enable you to solve a variety of optimization problems that lie outside the scope of the standard Optimization Toolbox. Firstly the fitness function with penalty terms is built by penalty strategy with addition type, and the fitness function is programmed in MATLAB language, and above neural networks program fitting the profile factor of worm gear teeth is recalled, then the nonlinear constraints function areprogrammed and the solver functions of Genetic Algorithm Toolbox are adopted. Program as follows:options= gaoptim set (‘PopulationSize’,20)。options=gaoptimset(‘Generations39。,100)。options=gaoptimset(‘CrossoverFraction’, ’MigrationFraction’)。options=gaoptimset(39。SelectionF39。, selectiontournament,’CrossoverF’, cross over scattered,’ Mutation F’,Mutation gaussian)。 nvars=3。lb=[1。2。10]。ub=[2。8。150]。[x, Fval, exit Flag, Output]=ga(@fitnessfun, nvars,[],[],[],[],lb, ub, @yueshufun, options)After function counting 108 times and iterating 326times, the final running output of above programming is:x1=,x2=,x3=,f(X)=1090628.VII..CONCLUSIONThis paper explored the methods available in the Genetic Algorithm and Neural Networks Toolbox. Compared with standard optimization algorithms(f(X)=),the objective function optimum in the genetic algorithm is about16 .37%less than the former. Therefore we saw that the genetic algorithm is an effective solver for non smooth problems. Additionally, we found that the genetic algorithm can be bined with other solvers, such as fuzzy logic and neural networks, to efficiently find a more accurate solution.TABLE IOUTPUT OF STANDARD OPTIMIZATION AND GENETIC ALGORITHM附 錄 B在神經(jīng)網(wǎng)絡(luò)中起重機傳輸遺傳算法最佳化摘要：那失真的適宜數(shù)學(xué)模型在設(shè)計起重機傳輸建立。那方法的二等的綜合評價被那最佳的把割集弄平整使用經(jīng)由，那方法的二等的綜合評價是使用經(jīng)由那最佳的把割集弄平整，因此每個模糊約束那最佳的價值可以是獲得弄平整，并且那模糊的最佳化是被變成那通常的最佳化。神經(jīng)網(wǎng)絡(luò)算法那背面加固增長的將采用到連續(xù)性前饋網(wǎng)絡(luò)如此適合相關(guān)系數(shù)。然后那用罰款期限是構(gòu)成由罰款策略裝配功能、神經(jīng)網(wǎng)絡(luò)計劃是召回、解算機功能的遺傳算法工具箱的matlab軟件是采用到解決那最佳化數(shù)學(xué)模型。索引詞：起重機機構(gòu)；遺傳算法最佳化；神經(jīng)網(wǎng)絡(luò)。模糊的最佳化數(shù)學(xué)模型的牽引機構(gòu)設(shè)計漸開線螺旋狀的蝸輪傳動裝置是采民用在起重機傳輸，哪個有主參數(shù)如下：額定功率Pe=、／min、輸出轉(zhuǎn)矩T2=2 、齒輪比U=、工作負荷因素k=, 那螺旋是機器和經(jīng)加熱處理材45鋼和那由ZQA194構(gòu)成的齒輪的齒輪冠.A指定目標(biāo)函數(shù)為了節(jié)省有色金屬的齒輪冠的螺旋齒輪，那目標(biāo)函數(shù)將應(yīng)指定那那大量的齒輪冠的螺旋齒輪在牽引機構(gòu)向最小的按照圖1傾斜，d0、 di2,b分別是外徑、內(nèi)徑和螺旋齒輪的齒面寬冠，因此那是大量的牙齒冠；由所以那目標(biāo)函數(shù)是m齒輪模數(shù)；d1齒輪分度圓直徑；z1螺旋開始的齒數(shù)。B反面選擇設(shè)計參數(shù)按照等式的那目標(biāo)函數(shù)、,m、d1將應(yīng)雖然設(shè)計參數(shù)選擇，但是簡而言之：C建立模糊約束認為Φ值的隨機特性設(shè)計參數(shù)和一些因素誰的價值很不定的比如負荷性質(zhì)和材料品質(zhì)、那模糊約束是建立、包括那性質(zhì)和邊界約束在內(nèi)。1）極限的開始的螺旋的齒數(shù)：z1=1~2。；2）極限的齒輪的模數(shù):2≤m≤8；3）極限的那導(dǎo)程角螺旋的因為保證蝸輪傳動裝置的效率：3≤γ≤8,tan γ=mz1/d1。4）約束的接觸強度的螺旋齒輪：那材料彈性因素、σh那接觸應(yīng)力的螺旋齒輪；[σh]那模糊的Φ值那容許接觸應(yīng)力的螺旋齒輪。5）約束的牙齒梁強度的螺旋齒輪：那橫梁強調(diào)的輪齒；[σf]模糊的Φ值那容許彎曲應(yīng)力的螺旋齒輪牙齒；Yf那輪廓因素因為螺旋齒輪牙齒。６）約束稠的的的螺旋：那螺旋信息系統(tǒng)支持在...之間二軸承、如果那蝸桿軸彎曲多，那就是說，那牙齒不會適當(dāng)?shù)鼐W(wǎng)孔，那么，那結(jié)果將要成為...的過度磨損和過早損壞所以極限偏轉(zhuǎn)是 ft1螺旋的切向力（N）、F r1螺旋的徑向力（N）、E那彈性模數(shù)（Mpa)、I危險截面的慣性矩的螺旋（mm4）L蝸桿軸承的距離（毫米）、L=9muz1這個的鍵方法是如何決定那最佳的把價值個別的因素弄平整、比如因素通信鏈路分析器系統(tǒng)、因素模糊和因素的不同的影響上去那不同的最佳的把價值弄平整、是認為和那方法的二等的綜合評價是使用以那最佳的把割集弄平整為基礎(chǔ)，因此那最佳的把價值＊的每模糊約束可以是獲得弄平整、簡而言之＊那模糊的優(yōu)化問題是變?yōu)槟峭ǔ５膬?yōu)化問題。神經(jīng)網(wǎng)絡(luò)由...組成簡單的元件并行操作這個元件被生物學(xué)的神經(jīng)的體系當(dāng)做本質(zhì)上鼓舞、那網(wǎng)絡(luò)函數(shù)決意大量地由那關(guān)系在...之間元件我們可以訓(xùn)練一神經(jīng)網(wǎng)絡(luò)執(zhí)行一特定函數(shù)由調(diào)整那Φ值那關(guān)系（重量）在...之間元件通常神經(jīng)網(wǎng)絡(luò)是調(diào)整，否則連續(xù)性，結(jié)果一特別的輸入導(dǎo)致一具體任務(wù)產(chǎn)量以一比較產(chǎn)量的和那靶子為基礎(chǔ)、直到那網(wǎng)絡(luò)產(chǎn)量相配那靶子。一些漲若干點相關(guān)曲線在齒數(shù)和那輪廓因素的螺旋齒輪被選為連續(xù)性樣本數(shù)據(jù)之間、那背面加固繁殖將采用到連續(xù)性前饋網(wǎng)絡(luò)、網(wǎng)絡(luò)的重量和偏見是更新然后神經(jīng)網(wǎng)絡(luò)被那功能的神經(jīng)網(wǎng)絡(luò)工具箱在matlab模擬。計劃如下：Z2=0:10:90。YF=[,, 25,]。n1=5。[W1,b1,W2,b2]= initff(Z2,n1,’tansig’,YF,’purelin’)。fpd=100。mne=20000。sse=。lr=。tp=[fpd, mne, sse, lr]。[W1,b1,W2,b2,te,tr]=trainbpx(W1,b1,’tansig’,W2,b2,’purelin’,Z2,YF,tp)y=simuff(Z2,W1,b1,’tansig’,W2,b2,’purelin’)VI..解決通常的最佳化數(shù)學(xué)模型由遺傳算法工具箱單密鑰到成功地解決許多種優(yōu)化問題是選擇那方法那井衣服那問題那遺傳算法和直接檢索工具箱是許多功能那伸展那做...的能力那最佳化工具箱和那matlab？數(shù)字計算環(huán)境那遺傳算法工具箱包括常規(guī)因為解決優(yōu)化問題與罰款期限是用加法類型造由罰款策略一起使用遺傳算法直接檢索這算法使你解決種種的優(yōu)化問題那謊言超出那標(biāo)準最佳化工具箱范圍。第一那適合功能，并且那適合功能是編制matlab語言、和在神經(jīng)網(wǎng)絡(luò)計劃適合那輪廓因素的螺旋齒輪牙齒是召回上、然后那非直線型限制功能是程序和那解算機功能的遺傳算法工具箱是采用。程序如下：選擇權(quán)gaoptimset（總體大小 ,20）；選擇權(quán)gaoptimset（世代,100）；選擇權(quán)gaoptimset（交叉分數(shù)39。,39。 ）；選擇權(quán)gaoptimset（s選擇完全約束的非晶網(wǎng)、選擇錦標(biāo)賽、交叉分散、變化高斯型曲線）；nvars=3。lb=[1。2。10]。ub=[2。8。150]。[x,Fval, exitFlag, Output]=ga(@fitnessfun, nvars,[],[],[],[],lb, ub, @yueshufun, options)）在功能計算108次并重復(fù)次時以后，那最后的焊道上面程序的產(chǎn)量是：x1=,x2=,x3=,f(X)=1090628.VII..結(jié)論這個紙?zhí)綔y那方法有效范圍那遺傳算法和神經(jīng)網(wǎng)絡(luò)工具箱和...相比標(biāo)準最優(yōu)化算法f(X)=)、那目標(biāo)函數(shù)在遺傳算法的最佳的是 大概16 .37%、我們發(fā)現(xiàn)那遺傳算法可以與...化合其他的解算機、比如模糊邏輯和神經(jīng)網(wǎng)絡(luò)、有效地發(fā)現(xiàn)一更精確的解答。表格1標(biāo)準最佳化的產(chǎn)量和遺傳算法解算機X1X2X3F(X)標(biāo)準最佳化遺傳算法1090628