Research on Elevator Group Control Algorithms

Elevator group control algorithms constitute a vital component of intelligent building transportation management systems. This paper delves into the multidimensional challenges inherent in such algorithms: the uncertainty of passenger demand, striking a balance between energy efficiency and service quality, and the intricacy of the algorithms necessitating real-time execution. It provides a comprehensive review of the evolution and core challenges faced in elevator group control algorithms, with an emphasis on analyzing the characteristics, advantages, and limitations of various methodologies, including expert systems, fuzzy logic control, genetic algorithms, particle swarm optimization, neural networks, and reinforcement learning. The study concludes by casting a forward glance at the prospective trajectories of elevator group control technology, notably the promising prospects of group control algorithms that integrate reinforcement learning with deep neural networks. These advanced algorithms hold significant potential in enhancing transportation efficiency, facilitating energy conservation and emission reduction, as well as optimizing passengers' psychological waiting experience in elevators.