两阶段混合算法的立体车库车位分配建模与仿真

针对自动化立体车库车位分配时顾客排队队长过长以及堆垛机能耗过高等问题，以减少顾客平均排队队长、堆垛机运行能耗等为目标，在保证车库运行效率前提下，通过介绍神经网络算法和果蝇算法原理与方法，提出基于2阶段混合算法的立体车库车位分配决策模型：第1阶段利用神经网络算法预测顾客停留时间；第2阶段利用果蝇算法实现车位最优分配。以车辆到达间隔时间服从泊松分布情况下建立立体车库数学模型，并以顾客平均等待时间、平均等待队长、平均服务时间、顾客的平均能耗作为立体车库效率指标评价立体车库的性能，采用MATLAB编制仿真程序。通过比较车辆就近分配原则和基于果蝇算法车位分配模型下的效率指标，证明了果蝇算法进行车位分配在保证车库运行效率和降低车库运行能耗上的有效性。

Two stage hybrid algorithm modeling and simulation of stereo garage parking location

When the automatic stereo garage parking space is allocated, the customer queue length is too long and the stacking machine energy consumption is too high. Based on the principle and method of introducing the neural network algorithm and the fruit fly algorithm, this paper proposes the stereo garage parking allocation based on three-phase hybrid algorithm, in order to reduce the average queue length and the running energy consumption of the stacker. The first stage uses the neural network algorithm to predict the customer stay time and the second stage uses the fruit fly algorithm to achieve the optimal allocation of parking spaces. The stereo garage mathematical model was established under the condition of vehicle arrival interval and the performance of the stereo garage was evaluated by the average waiting time of the customers, the average waiting queue length, the average service time and the average energy consumption of the customers. The simulation program using MATLAB is written. By comparison of efficiency index under the principle of near vehicle distribution and under the flock allocation model based on the fruit fly algorithm, it is proved that the fruit fly distribution is effective in ensuring the efficiency of garage operation and reducing the energy consumption of the garage.