| Summary: | 碩士 === 國立臺北科技大學 === 冷凍空調工程系所 === 104 === The chiller loading distribution methods include Average Loading(AVL)method, Simulated Annealing(SA) method, Particle Swarm Optimization(PSO) method, Evolution Strategy(ES) method, Genetic Algorithm(GA), Lagrangian Multiplier(LGM) method at present. Each of these methods has its own shortcomings. For example, AVL method may be the most popular used method but not the optimal one, the Lagrangian Multiplier(LGM) method can’t achieve the optimal solution when convex function and non-convex function exist together in the KW-PLR curve, Genetic Algorithm(GA) processes is too complex.
This study used regression analysis to construct a power consumption model of the chiller, and new method Fruit Fly Optimization Algorithm(FOA) was used for optimal chiller loading. This study compare three case use different methods for optimal chiller loading. The results showed the energy consumption can get lowest when load ratios was 90%~75%, the highest saving was 4.93% in the case1. The energy consumption can get lowest when load ratios was 90%~70%, the highest saving was 2.09% in the case2. A little less than Evolution Strategy(ES) method when load ratios was 90%~70%, the highest saving was 9.41% in the case3. The reason may be related to chiller’s performance, but the results indicate Fruit Fly Optimization Algorithm(FOA) used for optimal chiller loading was very well.
The Fruit Fly Optimization Algorithm(FOA) was proposed in recent years. The method and logic was very simple and easy to used. The test results show that the FOA can be very suitable for optimization.
|
|---|
