电梯群控系统节能优化调度控制

Energy Saving Optimal Dispatching Control of Elevator Group Control System

针对传统的电梯群控系统控制目标单一、自适应性差、节能效果不显著的问题,提出一种模糊控制结合神经网络算法的电梯群控系统来进行优化调度。电梯群控系统具有多目标的特点,建立由多个评价指标组成的综合评价函数,评价函数值即电梯的可信度作为整个系统的输出,选择可信度大的电梯进行派梯。从时间和能源角度考虑,选择了影响较大的输入变量,利用模糊控制中的隶属函数求出输入变量的隶属度从而将系统简单化处理。针对电梯群控系统具有非线性、难以建模的特点,利用BP神经网络建立起输入和输出之间的非线性联系。仿真结果表明,训练之后的神经网络输出值精度高,具有一定的泛化能力,可以实现电梯调度控制。通过与其它电梯群控算法进行对比,可知模糊控制结合神经网络算法确实能使电梯群控系统优化。

In view of the fact that the traditional elevator group control system has the disadvantages of single con- trol target, poor adaptability and less energy saving effect, an elevator group control system based on fuzzy control combined with neural network algorithm is proposed to optimize the scheduling. The elevator group control system has the characteristics of multi objectives, so a comprehensive evaluation function consisting of multiple evaluation indexes was established. The evaluation function value acted as the credibility of the elevator and was taken as the output of the whole system. The high confidence elevator was chosen for dispatching. From the point of view of time and energy, 3 important input variables were chosen, and membership function of fuzzy control was used to obtain the membership degree of input variables, so that the system was simplified. Because the elevator group control system was nonlinear and difficult to model, the nonlinear relation between input and output was established by using BP neural network. The simulation results show that the output value of training neural network is of high precision, and has a certain generalization ability, and can realize elevator scheduling control. By comparing with other elevator group control algorithms, it can be seen that the fuzzy control combined with neural network algorithm can really optimize elevator group control system.