基于混合遗传算法的通风网络解算研究

Study on ventilation network solution based on hybrid genetic algorithm

通过论述矿山常用的风网解算模型,重点对风网解算模型和方法进行了分析,指出了其存在的不足和应用局限性,建立了通风网络的优化解算模型,并提出了一种基于混合遗传算法的风网解算新方法。该方法充分发挥了遗传算法的群体搜索和全局收敛性,有效地克服了经典风网解算方法对分支风量初始值敏感问题;同时在遗传算法中引入经典算法(最速下降法)作局部搜索,克服了遗传算法收敛速度慢和精度差的缺点。最后通过一个矿山风网实例验证(已被其它算法解算过),证明了本文设计的混合遗传算法是一个有着可靠的收敛性、较高的收敛速度和精度,是解算矿山通风网络的一种成功算法。

Basing on the discussion of general model of ventilation network, the solving systems of the ventilation network was analyzed, and the deficiencies and limits of the general model of ventilation network were pointed out. A new optimization model of ventilation network was built and a new method based on hybrid genetic algorithm method was put forward, which combined the advantages of genetic algorithm and the classical algorithms to give full play of the advantage of group search, higher solving precision and globe convergence, and it overcome the problem of high sensitivity to the initial values of branches. The classical algorithm （steepest descent method） was used for local search at the same time to prevent the fault of the low convergence rate and low accuracy of genetic algorithm. This new method was verified by a real ventilation network which calculated by other method. It shows that this method is a reliable convergence, high versatility and solution method which can be used in complex ventilation network system.