Method To Generate Coincident Design Day For Air-conditioning System Design

At present,the energy shortage and global climate change caused by the rapid growth of global energy consumption are serious problems.The energy-saving design of air-conditioning(AC)system is one of the keys to reduce the building energy consumption.In the design process of the AC system,design cooling load is one of the main design indexes.Due to the irrational outdoor design weather data of the AC system,the design cooling loads of the AC system are generally overestimated,which leads to the oversized energy consumption of the AC system.Aiming at the problem,this dissertation has completed several innovative and engineering research works.Firstly,based on the forming process and mechanism of indoor heat and cooling loads,a calculation model of indoor heat and cooling loads for generating coincident design days was established.A method to evaluate the applicability of the cooling load calculation model based on the accuracy of generating the coincident design days was proposed,and a cooling load calculation model suitable for generating the coincident design days was established.Then,based on the comprehensive clustering method without supervision,a method was established to select the most representative coincident design days from years of measured weather data.Under the overall unsupervised principle,a method of the coincident design day prediction based on support vector machine and decision chain was proposed to predict the applicable coincident design days for all possible rooms.It solves the bottleneck problem that limiting the application of the coincident design days,and enhances the practicability of the coincident design days.Finally,the coincident design day generation method based on rooms is extended to generate the coincident design day for buildings.The results show that the coincident design days reasonably determined or predicted by the proposed method in this dissertation can not only match with the real near-extreme weather conditions,but also can be applied to almost all room conditions,while the deviation of the design cooling load calculated from the predicted coincident design days can be controlled within a rational range.At the same time,although the whole method is constructed based on the room situation,it can be well applied to a broader situation such as the whole building,indicating that the method constructed under the principle of unsupervised has great robustness and universality.In order to solve the basic key problem in the AC system design,the systematic and complete theory and method for coincident design day generation and application are constructed in this dissertation.The purpose of this dissertation is to provide a theoretical basis for outdoor design weather data updating and improvement in the future,and also to provide some reference and inspiration for the subsequent study of coincident design weather data or other outdoor design weather data for the AC system design.