Research On Model And Prediction Of Energy Consumption In Manufacturing Process Of Large-scale Mechanical Product Based On Knowledge

There are great potentialities of energy saving in product manufacturing process, for one hand, energy saving by technology improvement should be strengthened to reduce energy consumption of per unit product, for the other hand, fine energy management is also an urgent need to make energy saving effect come to a great extent. Nowadays, energy management method in discrete manufactures is not perfect and should be transformed from experience management to modern management. After energy consumption modeling in product manufacturing process, the source and component of energy consumption may be explicit to reduce or eliminate energy waste in the enterprise, the established model could serve as the guidance for relating energy consumption data with the mechanical process to be the basis for energy consumption prediction based on knowledge, and the model also could serve as the guidance for the development of energy management system to make the energy management work scientific, rational and practicable. With the development of artificial intelligence and computer technology, for the problem that the energy consumption result could not be calculate accurately by mathematical method, it is possible to solve it by the knowledge-based method to estimate the energy consumption in product manufacturing process. The result of calculation and prediction of energy consumption could be used to aid process planning evaluation and workshop scheduling, and it is the basis for fine energy management and the research of green manufacturing. Research on energy consumption model and research on estimation of energy consumption in manufacturing process based on knowledge were made in this dissertation and the research works include:An energy consumption model in product manufacturing process is built up. After some concepts in manufacturing process analyzed, the research system faced energy consumption of the dissertation is presented. Then, energy consumption in product manufacturing process is analyzed, and the energy consumption data model is constructed, the procedure includes: the questions that energy consumption model must be able to answer are presented; the terms of energy consumption are defined, the terms include definitions of energy consumption in machining procedure, definitions of energy consumption in part mechanical manufacturing process and product manufacturing process etc; the definitions and constraints on the terms and energy consumption control are specified in a formal way etc. The effect of the energy consumption model is analyzed, and the model could support fine energy management effectively.A semantic model of energy consumption knowledge and its semantic representation are proposed. Energy consumption knowledge comes from energy consumption information collected, which is associated with process plan and operating parameters afterwards. By explicating the concepts in energy consumption and their relationship, the semantic models of energy consumption knowledge in product manufacturing process and part machining process are presented. After comparing several semantic representation languages, OWL is used to describe entities, properties and relationships of concepts which are related with energy consumption. Axioms and constraints for system of energy consumption concepts are represented in First Order Logic by example of manufacture object. Then, information hierarchy of mechanical machining process is presented.The estimation method of energy consumption in mechanical process layer based on CBR is presented. First, the CBR-based process model of estimation of energy consumption in technical process is presented. Then, a case retrieval process is presented in view of the characteristics of energy consumption, and the retrieval process includes three main steps: retrieval based on affiliation product, part-oriented retrieval, matching and retrieval based on machine tools. The energy consumption results of retrieval similar cases are used to estimate energy consumption of new part by using layouts of article characteristics technology. The feasibility and effectiveness of the method is finally verified by the machining process of turbine rotor.The estimation method of energy consumption in machining procedure layer based on neural network is presented. First, after analyzing the expression of energy consumed per unit volume of metal removed, we know that energy consumption is related with real-life condition and is difficult to calculate accurately, and the choice of cutting parameter is the key factor of energy consumption for certain machine and tool. After analyzing the essence of neural network, neural network model of energy consumption prediction is built up, the choice and unitary of input variables and output variables are illustrated, then, the node number of latent layer and transfer function are selected. The combination of cutting parameter and the corresponding history data of energy consumption are served as training data set, and the nonlinear mapping relation of the combination of cutting parameter with corresponding energy consumption is setup, thus, energy consumption of the new combination of cutting parameter is estimated. The effectiveness of the prediction method is verified by the example of rough machining process of a guide blade.A prototype system of energy management in discrete manufacture is developed based on the guidance of energy consumption model and semantic model of energy consumption knowledge. After analyzing the characteristic of energy management in discrete manufacture, overall structure of system is presented which is constructed of three layers including data collection layer, data management layer and user service layer based on the design idea of function independence and modularization. The work procedure of prototype system is illustrated, and the system is realized from five aspects including product energy consumption management, machine tool energy consumption management, energy consumption information management, energy consumption estimation and system administration. By using the prototype system, it is benefit to make energy consumption clear, to find the way to reduce energy consumption, to promote the collection of energy consumption data. Lastly, the application of the method proposed in this paper in an enterprise is analyzed.