Study On Building Air-Conditioning System Dynamic Energy Evaluation And Diagnosis Method And Its Application

The building air-conditioning system is the largest energy-consuming sector in buildings,and it is the key goal of building energy conservation.Reasonably evaluating the energy consumption level of building air-conditioning system is the fundamental prerequisite to carry out the energy-saving work,but also an important basis for assessing the energy conservation effect.In addition,accurately identifing and diagnosing the energy consumption problem in building air-conditioning systems is a powerful way to conduct targeted energy conservation and enhance energy efficiency.In view of the key bottleneck problem of dynamic energy evaluation and diagnosis in building air conditioning system,this thesis elaborates the connotation of the evaluation and diagnosis of energy consumption in building airconditioning system.The dynamic energy consumption characteristics of the air-conditioning system are taken into a fully consideration.Furthermore,the theories and methods of dynamic energy evaluation and diagnosis of building air-conditioning system are intensively investigated.Two typical building air-conditioning systems,i.e.the chiller and the variable refrigerant flow(VRF)air-conditioning system,are studied respectively to verify the reasonability and effectiveness of the proposed energy evaluation and diagnosis method.Finally,a relatively complete framework of dynamic energy evaluation and diagnosis for building air-conditioning system is obtained.The main research contents and results of this thesis are listed as follows:(1)The thesis proposes the framework of dynamic energy evaluation and diagnosis of building air-conditioning systems.Taking the building energy conservation policies and regulations of different countries in the world as a starting point,this thesis analyzes the practices of building energy conservation in various countries detailly.A top-down practical system of building energy evaluation and diagnosis is constructed.Then,the global practices and research progresses are combed according to all levels of the top-down practical system.By summarizing the existing challenges of building energy evaluation and diagnosis,this thesis elaborates the connotation of the building air-conditioning system energy evaluation and diagnosis,which reflects the innovation and advancement of this research.In addition,combined with the characteristics and advantages of different data mining algorithms,this thesis constructs dynamic energy evaluation and diagnosis framework for building air-conditioning systems,respectively.The framework are established using data mining techniques based on numerous historical operating data of building air-conditioning systems.Besides,several methods are proposed in this study,i.e.the data processing methods of building air-conditioning systems,the energy consumption mode classification method based on decision tree,the energy benchmarking method,the energy consumption mode validation method based on variance analysis and Post hoc test,energy consumption rating calculation method,energy grading strategy and abnormal energy diagnosis method.Finally,the theoretical basis and practical application of each step in the framework are expounded in detail.(2)Energy evaluation and diagnosis of the chiller in office buildings.Firstly,a ”Large office”model is established based on the EnergyPlus software.The operating data of the chiller is obtained from the model.Based on the domain knowledge and decision tree analysis,the laws that how different factors such as meteorological conditions,personnel activities and operation time affect the building air-conditioning system energy consumption are revealed when the system is working under normal operating conditions.Besides,four energy consumption modes of chiller are classified by the decision tree.Accordingly,four dynamic energy benchmarks are established.The energy evaluation results of energy baseline and dynamic energy benchmarks are comparatively analyzed.Secondly,the “Large office”models with multiple faults and various faulty severe levels are further constructed.The faulty operationing data of the chiller is obtained.Based on the established four dynamic energy benchmarks,the system energy consumption under different faulty conditions are evaluated,which also verified the effectiveness of the proposed dynamic energy benchmarks.Simultaneously,the waste degrees of the system energy consumption at various faulty cases are given quantitatively.(3)Energy evaluation and diagnosis of the VRF system.Firstly,an experimental VRF platform is set up to carry out both normal and faulty experiments.The operating data of the VRF system under various operational conditions of both normal and faulty cases are obtained.Nine energy consumption modes are obtained based on the decision tree analysis.Accordingly,nine dynamic energy benchmarks are established.The system energy consumptions of both normal and faulty cases are quantificationally evaluated using the energy consumption rating and grade.In addition,the influence degree of various faults to the system energy consumption are analyzed in this study.Secondly,an energy diagnosis strategy is proposed based on various data mining algorithms for the VRF system,since the system complex structure led some faults ignored as well as similar energy consumption variation under the faulty and normal conditions.The validation results based on the experimental data show that the accuracies of abnormal energy detection are higher than 90% in all cases and it is 100% in some cases.Therefore,it indicates that the method is effective in abnormal energy identification in VRF systems.Based on the background of building energy conservation,this thesis proposes a dynamic energy evaluation and diagnosis framework for building air-conditioning systems.The mian innovation points of the thesis are the methodology achieves a reasonable,fair and quantitative evaluation of the energy consumption on the building air-conditioning system,as well as the quantitative assessment and diagnosis on the abnormal energy under faulty conditions.Three problems in previous methods,i.e.low reliability of the evaluation results,unquantifiable evaluation,and less relation between fault diagnosis and energy saving,are addressed in this study.For different air-conditioning systems,a method for assessing the energy consumption of chillers in office buildings with limited information was proposed.In addition,an energy evaluation and diagnosis method is proposed to assess the complex and inconsistent energy consumption of VRF system.Besides,the quantitative evaluation of building air-conditioning system energy consumption under both normal and faulty cases are achieved.The proposed dynamic energy evaluation and diagnosis framework can provide certain references for reasonablely assessment of the energy level of building air-conditioning systems,accurately identification and diagnosis of energy problems in building air-conditioning systems,justly evaluation of the effectiveness of the building energy conservation,carry out of targeted energy-saving work and improvement of energy-saving work efficiency.Further,it provides a powerful support and reference for the application of big data analysis,fault diagnosis etc.in actual building air-conditioning systems.