Research On Energy Efficiency Condition Evaluation And Diagnosis System For Coal-fired Power Units

The coal-fired power companies in China are transiting from high energy consumption,high emission and low efficiency production mode to low energy consumption,low emission and high efficiency production mode.Thus,one of the most attractive things facing China industry is the energy saving of coal-fired power units.However,renewable energy sources particularly wind and solar increase their share of the nation’s generating capacity,while those of coal decline.Unfortunately,renewable energy outputs are random,unstable and hard to predict,which makes peak shaving difficult.The peak shaving problem has become a big barrier to the development of energy saving operation for coal-fired power units.Therefore,the research on the energy efficiency condition evaluation and diagnosis technology for coal-fired power units is of great significance for the healthy and sustainable development of coal-fired power units and the power industry.This paper is focus on the research on the theory and technology of energy efficiency condition evaluation and diagnosis for the on-duty coal-fired units.The energy efficiency condition was proposed based on which a series of methods and key technologies of energy efficiency evaluation and diagnosis were carried out.Firstly,the parameters influence on the energy efficiency condition and their relationship were analyzed in detail.A hierarchical index system was constructed to describe the energy efficiency condition.Secondly,the energy efficiency benchmark condition was characterized and determined.And then,the real-time evaluation model of energy efficiency condition was proposed by the historical running data.Thirdly,the theoretical method of energy efficiency diagnosis was studied.Finally,the energy efficiency condition anomaly diagnosis model was constructed.The practical and feasible technical framework was formulated to describe the energy utilization comprehensively and objectively.It was also help to know the gap between the advanced and actual level of energy efficiency,and to make the operation adjustment scheme and maintenance strategy.First of all,based on the theory of system science,a hierarchical system components tree of thermal power system with "system-subsystem" was built.Combining the domain knowledge and the first law of thermodynamics,the main index parameters that can reflect the energy efficiency condition of the coal-fired unit were selected.All indexes were integrated from the low to the high level to make the index system of energy efficiency condition,which provides the foundation for the evaluation and diagnosis of energy efficiency condition.What is more,the mutual relationship among the energy efficiency indexes was clearly expressed based on the directed graph.Secondly,combining data statistics method and data mining technology,two aspects of research and application were carried out based on the energy efficiency condition index system:(1)the energy efficiency benchmark condition of the unit was determined based on the data mining method taking into account the complex and changeable operating conditions.The implicit energy saving model was extracted from the large-scale historical data base.Based on the Fuzzy C-means clustering algorithm,the multi indexes synchronous clustering of history data was carried out after steady condition detection and working condition classification.The energy efficiency benchmark condition was determined according to the clustering results.(2)A real-time evaluation model of energy efficiency condition based on multiple feature fusion was proposed.The energy efficiency condition index was proposed to estimate the deviation between the actual condition and the estimated condition,thus,the energy efficiency condition was classified(excellent,good and poor).Then,the knowledge base of the abnormal event diagnosis was constructed based on the energy efficiency diagnosis theory,combining the signed directed graph with the domain knowledge.Operational adjustment cause and the maintenance cause of the anomaly were analyzed.Adjustment strategy for operation optimization,the location,mode and reason of the abnormal occurrence were located;the corresponding processing decision was given.Finally,the energy efficiency evaluation and diagnosis technology framework for coal-fired power unit was applied to engineering practice.An energy efficiency evaluation and diagnosis system was developed based on the 1000MW ultra-supercritical wet cooling unit of Zheneng GuoHua Ninghai power plant.Database Oracle 10.Og and B/S modes was adopted to develop the system.