李一凡
摘要:隨著我國(guó)建筑施工項(xiàng)目規(guī)模的不斷擴(kuò)大,對(duì)施工過(guò)程的管理提出了更嚴(yán)峻的要求。為了提高大型項(xiàng)目的可靠性,本文引進(jìn)施工系統(tǒng)可靠性作為綜合指標(biāo),對(duì)施工過(guò)程系統(tǒng)的可靠性進(jìn)行優(yōu)化分配。以裝配式建筑為實(shí)例,運(yùn)用教學(xué)算法優(yōu)化施工系統(tǒng)可靠度,針對(duì)教學(xué)算法在解決高維復(fù)雜問(wèn)題時(shí)易失去種群多樣性和陷入局部最優(yōu)的缺點(diǎn),在基本教學(xué)算法的基礎(chǔ)上引入信息熵,提出了基于信息熵改進(jìn)的教學(xué)因子。最后通過(guò)將改進(jìn)后的算法應(yīng)用到建筑項(xiàng)目施工系統(tǒng)可靠性?xún)?yōu)化中,結(jié)果表明改進(jìn)后的教學(xué)算法比基本教學(xué)算法更容易跳出局部最優(yōu),具有較強(qiáng)的全局搜索能力。
Abstract: With the expansion of the scale of construction projects, it has more stringent requirements on the management of the construction process. In order to improve the reliability of large-scale projects, this paper introduces the construction system reliability as a comprehensive index to optimize the reliability of the system. Taking prefabricated construction as an example, the teaching-learning-based algorithm is used to optimize the reliability of construction system. According to the shortcomings of the teaching-learning-based algorithm in solving high-dimensional complex problems, information entropy is introduced in this algorithm and the teaching factors based on improved information entropy are put forward. Finally, the improved algorithm is applied to optimize the reliability of construction project system, and the results show that the improved algorithm is easier to jump out of local optimum problem than the basic teaching-learning-based algorithm and has strong global search ability.
關(guān)鍵詞:施工系統(tǒng)可靠性;教學(xué)算法;可靠性?xún)?yōu)化;信息熵
Key words: construction system reliability;teaching-learning based on algorithm;reliability optimization;information entropy
中圖分類(lèi)號(hào):TU765 文獻(xiàn)標(biāo)識(shí)碼:A 文章編號(hào):1006-4311(2018)13-0181-03
0 引言
施工可靠性是基于施工的特性和系統(tǒng)工程原理所定義的,是探索項(xiàng)目施工生產(chǎn)系統(tǒng)的可靠性,是研究施工過(guò)程實(shí)現(xiàn)目標(biāo)體系的可靠性[1-3]。由于工程項(xiàng)目施工系統(tǒng)與制造業(yè)生產(chǎn)系統(tǒng)二者本身就具有趨同性,同樣都具有復(fù)雜的生產(chǎn)流程,基于這種啟示,將施工系統(tǒng)可靠性應(yīng)用于裝配式住宅施工領(lǐng)域是一個(gè)值得嘗試的研究方向[4]。為了提高大型項(xiàng)目的可靠性,本文引進(jìn)施工系統(tǒng)可靠性作為綜合指標(biāo),對(duì)施工過(guò)程系統(tǒng)的可靠性進(jìn)行優(yōu)化分配[5-6]。針對(duì)該優(yōu)化模型的特點(diǎn),引用一種新型的群智能算法——教學(xué)優(yōu)化算法進(jìn)行求解[7-9]。因此,以系統(tǒng)可靠性理論為基礎(chǔ),基于施工可靠性這一新的視角,對(duì)工程項(xiàng)目施工系統(tǒng)進(jìn)行研究,具有重要的理論及現(xiàn)實(shí)意義[10]。
1 基于信息熵改進(jìn)的TLBO算法
2 算法性能測(cè)試
通過(guò)表1標(biāo)準(zhǔn)測(cè)試函數(shù)進(jìn)行算法性能測(cè)試。
改進(jìn)的TLBO算法在這7個(gè)高維復(fù)雜函數(shù)上都達(dá)到了很高的精度,其在函數(shù)f1、f3、f4、f6和f7都能達(dá)到全局最優(yōu)解,而基本TLBO算法僅在f7達(dá)到了全局最優(yōu)解,其余都陷入了局部最優(yōu)解。圖1和圖2為兩種算法分別在f6和f7中的尋優(yōu)收斂曲線,通過(guò)對(duì)比可以看出,改進(jìn)的TLBO算法收斂速度有顯著提高并且能夠跳出局部最優(yōu),達(dá)到了預(yù)期的效果。
3 工程實(shí)例應(yīng)用
3.1 工程背景
現(xiàn)以江蘇省某裝配式住宅項(xiàng)目為例進(jìn)行分析。本工程主體結(jié)構(gòu)形式為剪力墻結(jié)構(gòu),主體結(jié)構(gòu)部分內(nèi)、外墻板均采用預(yù)制裝配式墻體,結(jié)構(gòu)樓板、樓梯部分采用疊合板,樓梯、陽(yáng)臺(tái)均采用預(yù)制裝配式形式,空調(diào)板、節(jié)點(diǎn)、接縫等采用現(xiàn)場(chǎng)現(xiàn)澆的方式。各子系統(tǒng)的基本信息如表3所示,施工網(wǎng)絡(luò)計(jì)劃圖如圖3所示。假定建設(shè)項(xiàng)目所需預(yù)制構(gòu)件數(shù)量能滿(mǎn)足施工要求,預(yù)制構(gòu)件的運(yùn)輸、固定、堆放等都能保證裝配施工順利進(jìn)行。
3.2 問(wèn)題描述
施工項(xiàng)目在一定的資源下,施工可靠性?xún)?yōu)化問(wèn)題描述如下:
3.3 施工可靠度計(jì)算
采用基于歷史資料數(shù)據(jù)統(tǒng)計(jì)分析的方法,得出各子系統(tǒng)的可靠度和基本費(fèi)用如表4所示。
通過(guò)Matlab進(jìn)行仿真計(jì)算,運(yùn)行改進(jìn)教學(xué)算法,對(duì)參數(shù)記性設(shè)置,種群數(shù)Ps=100,教學(xué)因子最大值TFmax=2,最小值TFmin=1,迭代次數(shù)500次,所得最優(yōu)個(gè)體的成績(jī)即為施工系統(tǒng)可靠性的最優(yōu)分配方案。如表5所示。
同時(shí)與用遺傳算法求解出的結(jié)果進(jìn)行對(duì)比分析,可以看出,改進(jìn)后的教學(xué)算法計(jì)算所得的子系統(tǒng)可靠度基本均優(yōu)于遺傳算法所得結(jié)果,依據(jù)施工系統(tǒng)可靠性原理算出系統(tǒng)可靠度以及系統(tǒng)成本,對(duì)比顯示,ITLBO算法的系統(tǒng)成本較低且具有較優(yōu)的系統(tǒng)可靠度。
4 結(jié)束語(yǔ)
施工系統(tǒng)可靠性的優(yōu)化分配問(wèn)題對(duì)于工程的實(shí)際管理和質(zhì)量保障具有重要意義。本文利用施工系統(tǒng)可靠性的計(jì)算方法,以裝配式建筑施工過(guò)程為例,采用新穎的TLBO算法,通過(guò)引入信息熵改進(jìn)教學(xué)因子,克服了基本教學(xué)算法在解決高維復(fù)雜問(wèn)題時(shí)易失去種群多樣性和陷入局部最優(yōu)的缺點(diǎn),且與已有算例相比具有較高的收斂速度和更精確的最優(yōu)解。同時(shí)也拓寬了教學(xué)算法在工程領(lǐng)域方面的應(yīng)用,促進(jìn)了施工可靠性在工程管理過(guò)程中的發(fā)展,為大型建筑工程的可靠性保障提供了有力的理論支持。
參考文獻(xiàn):
[1]鄧鐵軍.結(jié)構(gòu)工程施工系統(tǒng)可靠性理論方法及其應(yīng)用的研究[D].長(zhǎng)沙:湖南大學(xué),2007.
[2]阮旻智,李慶民,王紅軍.人工免疫粒子群算法在系統(tǒng)可靠性?xún)?yōu)化中的應(yīng)用[J].控制理論與應(yīng)用,2010,27(9):1253-1258.
[3]趙平,吳昊.差分進(jìn)化混合粒子群算法求解裝配式住宅項(xiàng)目進(jìn)度優(yōu)化問(wèn)題[J].計(jì)算機(jī)工程與科學(xué),2016,38(7):1495-1501.
[4]RAO R V,SAVSANI V J,VAKHARIA D P.Teaching-learning based optimization: a novel method for constrained mechanical design optimization problems[J]. Computer-Aided Design,2011,43(3):303-315.
[5]Rao R V, Patel V. Multi-objective optimization of heat exchangers usinga modified Teaching-learning-based optimization algorithm[J]. Applied Mathematical Modeling, 2013, 37(3) : 1147-1162.
[6]RAO R V, Patel V.Multi-objective optimization of two stage thermoelectric coolers using a modified teaching-learning-based optimization algorithm [J]. Engineering Applications of Artificial Intelligence, 2013, 26(1) : 430-445.
[7]RAO R V,PATEL V.An elitist teaching-learning-based optimization algorithm for solving complex constrained optimization problems[J].International Journal of Industrial Engineering Computations,2012,3(4):535-560.
[8]TOGAN V. Design of planar steel frames using teaching-learning based optimization[J].Engineering Structures,2012,34(1) :225-232.
[9]Tayfun Dede.Application of Teaching-Learning-based on Optimization Algorithm for the Discrete Optimization of Truss Structures[J].Structural Engineering,2014,18(6):1759-1767.
[10]W Cheng,F(xiàn) Liu,L J Li. Size and geometry optimization of trusses using teaching-learning-based optimization[J].International Journal of Optimization in Civil engineering,2013,3(3):431-444.