In order to find out the best one within the final solutions , fuzzy comprehensive evaluation is introduced into the thesis 為了在幾個(gè)候選解中找到最優(yōu)解,本論文引入了二級(jí)模糊綜合評(píng)判方法。
In such problems , there are too many test cases to evaluate all the candidates , and a random sampling of test cases can be uninformative 在這類(lèi)問(wèn)題中,對(duì)于所有的候選解而言,由于其數(shù)目太多而無(wú)法對(duì)它們進(jìn)行評(píng)估,而且隨機(jī)的測(cè)試事例采樣也不能提供有用的信息。
This makes accurate evaluation of candidate solutions difficult . by contrast , accurate and efficient evaluations can be obtained by comparing and testing different solutions to expose their advantages and flaws 相反,通過(guò)對(duì)不同候選解的比較和測(cè)試,則可能獲得這些候選解的準(zhǔn)確和有效的評(píng)價(jià),從而揭示候選解中的優(yōu)缺點(diǎn)。
We implement the simulated - annealing algorithm and use it to partition the graph gained from the user design . at last , we introduce and illustrate the software part , including the graph input style , the data structure of the result , the method to create candidate result and the cost of result , we also give some experiment data 本文實(shí)現(xiàn)了模擬退火算法,用于對(duì)系統(tǒng)所轉(zhuǎn)化的圖的劃分,在本文的最后給出了具體的軟件實(shí)現(xiàn)的詳細(xì)說(shuō)明,包括圖的輸入,解的格式,候選解的生成,解的造價(jià)等等,給出了部分的實(shí)驗(yàn)數(shù)據(jù)。
Besides , it is not fit with the precise adjustment and is difficult to conform the place . a new adaptive genetic algorithm with bp algorithm to optimize weight is backed up . the algorithm which combines the merits of the global convergence of genetic algorithm with fast local researching of bp algorithm not only intensifies the gradual convergence and evolution ability but also advance the speed of convergence , precision of training and generalization 針對(duì)傳統(tǒng)遺傳算法的搜索過(guò)程帶有一定的盲目性,其收斂特性不穩(wěn)定且收斂速度緩慢,特別是在系統(tǒng)規(guī)模較大時(shí),優(yōu)化效果的明顯改善往往需要相當(dāng)長(zhǎng)的時(shí)間,而且不適合候選解的精調(diào),難以確定解的確切位置,提出一種新型自適應(yīng)性遺傳算法,并在此基礎(chǔ)上,用bp算法優(yōu)化前向神經(jīng)網(wǎng)絡(luò)權(quán)值,綜合了兩種算法的優(yōu)點(diǎn),即遺傳算法的全局收斂性和bp算法局部搜索的快速性,強(qiáng)化了遺傳算法的漸進(jìn)收斂和進(jìn)化能力,全面改善了算法的收斂性,提高了收斂速度及訓(xùn)練精度,也擴(kuò)展了泛化能力。