Research On Optimizing Chiller Loading By Improved Cuckoo Search Algorithm

As the urbanization process accelerates year by year,the energy consumption of commercial buildings continues to rise,and the energy consumption of air conditioning has accounted for more than 40% of the total energy consumption of buildings.Among them,the chiller as the host of the air conditioning system,its operating energy consumption accounts for about 60% of the energy consumption of the air conditioning system.The key point of reducing the energy consumption of the air conditioning system is how to effectively control and optimize the operation energy consumption of the chiller.The cuckoo search algorithm has been successfully applied to the operation and scheduling of water resources and power grids,and has demonstrated superior performance over other intelligent algorithms.This thesis takes multiple parallel chillers as the research object,and solves the problems of chiller load distribution and energy saving through the research of cuckoo search algorithm.The specific work has the following 4 aspects:(1)The six energy consumption models of chiller units commonly used in the current study are summarized.Considering that the research focus of this thesis is to reduce the energy consumption of the unit through load allocation optimization,the semi-empirical model that is only related to the partial load rate is selected as the energy consumption model of the chiller in this paper;Then,the two energy efficiency evaluation indexes COP and IPLV of the chiller are studied.This paper analyzes the difference between the two and selects COP as the energy efficiency index;(2)Analyzing the cuckoo search algorithm.From the perspective of dynamic adjustment of parameters,fusion between algorithms,and improvement of paranoid ability of the algorithm,three improvements are proposed: adaptive cuckoo algorithm,particle swarm based cuckoo algorithm,and population particle attracting strategy algorithm.Five standard test functions were used to test the effect of different improved algorithms.The results show that the cuckoo algorithm based on the particle attraction strategy is optimal in terms of convergence speed and accuracy.(3)The building's cooling load is modeled and analyzed by TRNSYS software.Shenyang,Hangzhou and Kunming were selected as typical cities.Through sensitivity analysis,the effect of window-wall ratio on the building's cooling load was quantitatively studied.The sensitivity of the cooling load to the change of the southward and northward's window-wall ratio is: in the spring and summer transition season,when the southward window-wall ratio changes: Shenyang>Hangzhou>Kunming,when the northward window-wall ratio changes: Hangzhou>Shenyang>Kunming;In the summer and summer-autumn transition season,when the southward and northward's window-wall ratio changes: Hangzhou>Shenyang>Kunming.(4)Three different types of chillers in parallel are selected to verify the improved cuckoo algorithm.Through TRNSYS modeling,the hourly cooling load of the building is obtained,and the improved cuckoo algorithm and three commonly used optimization algorithms are used to compare the hourly cooling load distribution.When three small units with the same capacity are connected in parallel,the energy saving is 2.63% compared with the energy consumption of the genetic algorithm;when two large capacity and two small capacity units are connected in parallel,the energy saving is compared with the energy consumption of the particle swarm algorithm 1.04%;when six large-capacity units are connected in parallel,the energy consumption is 1.38% compared with the energy consumption of the simulated annealing algorithm.Compared with the three intelligent algorithms,the improved cuckoo algorithm has obvious improvement in convergence accuracy and speed.