| In the current context of global energy shortage and continuous deterioration of ecological environment,remanufacturing,as a new manufacturing mode for sustainable development,has become an effective way to promote energy saving and emission reduction and develop circular economy.As the remanufacturing mode is affected by many uncertain factors,it makes it difficult to reasonably schedule the remanufacturing process of used parts in the actual workshop,thus causing excessive consumption and waste of energy.Therefore,in order to reduce the energy consumption generated in the remanufacturing process of used machine tool spindles,this thesis takes into account the uncertainty environment and conducts research on the remanufacturing failure characteristics decision method,the uncertainty energy saving scheduling problem and the model solving algorithm,using the proposed research model to achieve the purpose of energy consumption reduction.The details of the research are as follows:(1)A failure feature classification model for autonomous decision making is proposed for the classification of failure features of scrap spindles under uncertainty environment.According to the three key uncertainty factors in the remanufacturing stage,the failure form features of the scrap spindle are extracted,and the normalization method is used to eliminate the difference in magnitude between the feature data;based on the BP neural network,the Softmax function is introduced into its network structure to establish the spindle remanufacturing failure feature classification decision model.The BP model is validated by spindle feature data,and the results show that the model can judge and categorise the failure form of each scrap spindle,and then determine the number of spindles entering the reworking stage,and also clarify the process route of the reworking stage,estimate the time of the reworking stage,and eliminate the influence of uncertainty factors on the reworking stage of the spindle.(2)Modelling the spindle remanufacturing scheduling problem based on the uncertainties that exist in the spindle remanufacturing process.Combining the manufacturing model of the remanufacturing workshop and the sources and components of energy consumption in the remanufacturing stage,the calculation formula for each energy consumption is designed;based on three uncertainty factors,namely the number of scrap parts,the remanufacturing process route and the remanufacturing time,a quantitative parametric expression is made;with the minimisation of total energy consumption as the optimisation objective and uncertainty and machining conditions as constraints,the uncertainty energy-saving scheduling in the remanufacturing stage is proposed Mathematical model.The model is validated using practical cases,and the results show that the established uncertainty scheduling model can correctly calculate the machining time and machining energy consumption in the re-processing stage.(3)A Genetic-Simulated Annealing Algorithm(GSAA)is proposed by combining the genetic algorithm with the simulated annealing algorithm to address the problem that the genetic algorithm can hardly meet the requirements of the uncertain scheduling model solution.In order to match the algorithm with the shop floor scheduling problem,a chromosome pair encoding method containing process and processing equipment information is designed;to address the problems of premature convergence and insufficient local search capability of the genetic algorithm,a POX(Precedence Operation Crossover)crossover operator based on the Metropolis criterion and a variation of the fused simulated annealing The algorithm is designed to achieve dynamic adjustment of the crossover stage and further screening of the variation results,while introducing an elite retention strategy to avoid the loss of outstanding chromosomes.The analysis and validation of the GSAA is carried out through a spindle repair and re-processing case.The experimental results show that the proposed algorithm outperforms the initial scheduling scheme in terms of processing time,processing energy consumption and equipment idle rate compared with the genetic algorithm,which verifies that the GSAA in this study has a high solution performance. |
