AN ELEVATOR GROUP CONTROL-SYSTEM WITH FLOOR-ATTRIBUTE CONTROL METHOD AND SYSTEM OPTIMIZATION USING GENETIC ALGORITHMS

A new concept is proposed for an elevator group control system which c an change control settings according to individual floor utilization s ituations. The floor-attribute-based control method uses a combination of floor-attribute-based evaluation and car-attribute-based evaluatio n. Computer simulations show it can improve multiple preferential floo r settings at the same time. Additionally, an on-line parameter tuning method using genetic algorithms is proposed. The floor attribute cont rol method needs several dozen control parameters to be tuned accordin g to each building's traffic, and the proposed method can do it. The e litist-preserving strategy is an effective way to keep control stabili ty. A diversity-recovering operation, which initializes population, ex cluding the elitist individual, when traffic flow data have changed, i mproves the following ability for transitions of utilized conditions.