Energy Consumption Prediction And Optimal Scheduling Of Public Buildings Based On Centerless Network

Nowadays,the problem of energy crisis and environmental pollution has become a great challenge all over the world,building energy consumption has also increased dramatically with the rapid development of economy.so,The optimization of building energy conservation has become an important means to alleviate the energy problem.Traditional building energy saving technology is to study the balance between indoor environment thermal comfort requirement and building energy consumption,but often at the expense of indoor environment thermal comfort degree as the cost.so,The key scientific problem that the building energy-saving needs to solve is to meet the thermal comfort needs while achieving the purpose of energy saving.In this paper,a parallel optimization scheduling strategy based on Centerless network platform for building energy efficiency and comfort is studied.The goal of the optimal scheduling strategy is to achieve the minimum energy consumption while maximizing the comfort requirements of each subspace,to achieve comfort balance among indoor space users under certain load requirements.The main contents of this paper include:(1)The equivalent thermal parameter model of air conditioning load is constructed.The unknown parameters in the model are identified by recursive least square method.The identification data is composed of indoor and outdoor environmental parameters and building load data,and the data is obtained based on the TRNSYS simulation platform.The validity of the model is verified by simulation experiments.(2)The prediction model of building energy consumption based on HCMAC neural network is constructed.This model improves the determination strategy of HCMAC neural network nodes by combining particle swarm optimization with improved K-means clustering algorithm to realize the accurate prediction of building load.The data needed in theexperiment can be obtained by simulating the actual running rule of the building on the TRNSYS simulation experimental platform.The simulation results show that the prediction model is better than other prediction models based on HCMAC neural network.(3)The thermal comfort measurement model based on HCMAC neural network is constructed.PMV evaluation index is selected as the quantitative index of human thermal comfort.The model predicts the PMV values of each subspace according to the IKHCMAC neural network prediction model proposed in the third chapter,In order to eliminate the gap between simulation data and actual data,a Gauss white noise v with a mean of 0 and a variance of 1 is added to the simulation data.The validity of the model is verified by simulation experiments.(4)A building energy optimization scheduling strategy based on centerless network platform is studied.The optimization goal of the optimal scheduling strategy is to achieve the minimum energy consumption while maximizing the comfort requirements of each subspace under certain load requirements,so that the comfort of each subspace can be balanced.The optimization algorithm adopts an improved multi-objective genetic algorithm based on K-Means.The simulation results show that the optimal scheduling strategy can better solve the problem of parallel optimization between building comfort balance and building energy consumption.