In this thesis, two approaches for EGCS(Elevator Group Control System) are proposed based on fuzzy logic. First, an important weighting factor to schedule the EGCS is studied and a fuzzy model is proposed. A two-stage fuzzy model is proposed to determine the neighborhood-weight with the aid of expert knowledge. Second, an FEGCS(Fuzzy Elevator Group Control System) is designed and implemented on an on-board system. A fuzzy classifier of the passenger traffic in buildings and a fuzzy inference engine to assign hall-calls to suitable elevators are proposed.
Elevator Group Control Systems(EGCS) are the control systems that man-age multiple elevators systematically in order to transport the passengers efficiently. The EGCSs assign hall-calls to elevators in response to passenger``s calls. Many factors such as arriving time, capacity of elevator and position of other elevators are considered to optimize performance. In this thesis, three typical optimization targets(average waiting times, the percentage of passengers waiting more than 60 seconds, and power consumption) are observed. The importance of optimization targets is defined interactively by the building manager in modern EGCS.
First, the property of the neighborhood-weight of the conventional EGCS is studied and optimized by the proposed fuzzy model. The neighborhood-weight which determines the load biases of elevators is a control parameter closely related to the system performance. This paper proposes a method based on a fuzzy model to determine the neighborhood-weight. The proposed method uses a two-stage fuzzy inference model that is built by the study of neighborhood-weight properties and expert knowledge. The proposed method shows more desirable results than the conventional method in simulations that use real traffic data.
Second, an FEGCS is designed and implemented to generate the assignment strategy and to assign hall-calls by predefined optimization targets. This thesis shows overall structure of t...