Research On Energy Demand Modeling And Intelligent Computing Of Machining Process For Low Carbon Manufacturing

Manufacturing industry accounts for a significant part of world’s consumption of energy and generation of environmental pollutions and CO2while transforming resources into product or service. Machining processes, as a major process of manufacturing industries, plays an important role in energy saving and emission reduction of manufacturing industries. Accurate energy estimation of machining processes is the foundation of energy improvement and optimization, and also is the basis of low carbon manufacturing. To solve the problem of energy demand modeling and computation of machining processes, several research points were focused on in this thesis:decomposition of energy demand of machining process, energy demand modeling of activity, energy demand modeling of activity transition, and energy intelligent computing of machining processes.To deal with the problem of decomposition of energy demand of machining process, the energy demand characteristics of machining process was firstly analyzed. According to the energy demand characteristic differences, machining process was decomposed into activities and activity transitions, and then the energy demand of machining was decomposed into of activity energy demand and activity transition energy demand. Consequently, the decomposition of energy demand of machining process was realized. The explicit definition of activity/activity transition and corresponding energy calculation parameters were given, which lay the foundation for the following research.To solve the problem of energy demand modeling of machining activity, a Therblig based energy demand modeling method of activity was proposed. Activity was firstly decomposed into Therblig and fourteen basic Therbligs were defined. According to the power characteristics, therbligs were grouped into four types:constant power non-material cutting Therblig, variable power non-material cutting Therblig, constant-MRR (Material Removal Rate) material cutting Therblig, and variable-MRR material cutting Therblig. Then power models of the above four types of Therbligs were established by adopting existing model, improving existing model and establishing new model. The mapping relationship between activity and Therblig was established by taking machining state as a bridge. Then the predictive power curve of activity was established based on the Therblig power and activity-Therblig mapping relationship. Moreover, the energy demand model of activity for machining processes was built and the ex-ante energy demand estimation of activity was realized.To solve the problem of energy demand modeling of machining activity transition, a finite state machine based energy demand modeling method of activity transition was proposed. According to the characteristic of activity transition, finite state machine was adopted to express the activity transition and activity transition was transformed into state transition. The key state transitions were determined according to the Pareto principle and the energy demand models of the key state transitions were established. The state transition diagram of machining process was established and the key state transitions and corresponding execution times in the whole machining process were also obtained. Then the energy demand model of activity transition of the whole machining process was built combined with the established energy model of the key state transitions. Moreover, the predictive power curve of machining process (activity+activity transition) was established through embedding the transiton power in the predictive power curve of activity. The ex-ante energy demand estimation of activity transition of the whole machining process was realizedTo solve the problem of energy intelligent computing of machining processes, an energy intelligent computing method of machining processes based on energy calculation parameter extraction and inheritance was researched. The extraction method of energy calculation parameter of machining process was proposed to obtain the energy calculation parameter of activity and activity transition. The definition of energy calculation parameter inheritance was given. Then completeness inheritance and continued effectiveness inheritance rules between activity and machining state and matching inheritance and replication inheritance rules between machining state and Therblig were set up. Based on the above inheritance rules, the transmission of energy calculation parameter was realized. Then the intelligent computing of energy demand of activity and activity transition were researched based on the energy calculation parameter inheritance. The intelligent computing of energy demand of machining processes was realized through combining the energy intelligent computing of activity and that of activity transition. The feasibility and validation of the proposed method were demonstrated by taking the machining process of a typical mechanical part as an example.Finally, the research contents and the innovation points of this thesis were summarized, and the future research directions were discussed.