Research And Application Of Energy Efficiency Analysis Method Based On Fuzzy Extreme Machine

With the acceleration of the industrialization process,the demand for energy resources is rising rapidly,which makes human society facing tremendous pressure on energy and environmental protection.Energy efficiency analysis of petrochemical process plays a very important role in sustainable development.Evaluating and predicting the production energy efficiency,understanding the production level and analyzing energy saving potential are the key technologies to improve efficiency and reduce consumption.However,due to the complex production process and the uncertainty data with high-noise,traditional energy efficiency analysis methods are restricted by various conditions and cannot be widely put into use.There are still many deficiencies in theoretical research and practical application in production forecasting,energy efficiency analysis and optimization of complex chemical industries.Therefore,this thesis takes the research and application of energy efficiency analysis methods for complex petrochemical production as the research topic,proposes a novel energy management and optimization framework based on the fuzzy extreme learning machine(FELM)method,makes a reasonable analysis of the energy efficiency in the actual production process,and analyzes the energy saving potential of production equipment.The case study shows that the energy efficiency analysis and prediction framework based on the fuzzy extreme learning machine method is superior to traditional methods in performance,and can provide guidance for actual production effectively.The research content of this thesis is as follows;1.In order to solve the problem of production data volatility and uncertainty,a novel analysis and prediction framework based on fuzzy extreme learning machine method is established.Triangular fuzzy numbers are used to deal with uncertainty data.After extracting the characteristic data,the membership degree is used instead of accurate data as network inputs to obtain the minimum,average,and maximum value of energy efficiency data.And the cross recombination of triangular fuzzy numbers(TFNs)is applied in the training of the network.Moreover,the upper and the lower limits of efficiency values are obtained on the basis of network generalization to analyze the energy conservation and saving potentials.2.The proposed method is applied in energy management and optimization of China ethylene industry in complex petrochemical industries.The least efficient production model,the average production model and the most efficient production situation model are established respectively.Then the contrast experiments are presented to verify the effectiveness and practicability of the method.Finally analyze energy efficiency,provide production optimization guidance for the actual ethylene production,and analyze the energy saving potential.3.The prototype application system for energy efficiency analysis methods is designed and implemented with the detailed description of the system's overall architecture and functional module,to make the energy efficiency analysis process more productive and efficient.