| Under the promotion of the strategy of strengthening the country by sea,the ship manufacturing industry has put forward higher requirements on the performance of diesel engines.The diesel engine is the core of a ship,with a complex structure and harsh operating environment.The machining accuracy of its components determines the reliability as well as the stability of the diesel engine operation.During the machining process of the parts,the machining resources change dynamically with time,which in turn affects the final quality of the product.Product quality prediction,as an important part of the machining process,has become the key to achieving intelligent manufacturing systems.To address the problem of process quality prediction of parts in the machining process,this thesis proposes a machining quality prediction method based on digital twin technology.Firstly,the key processes in the machining process of the part are analyzed to determine the factors affecting the machining quality.Secondly,the machining process is divided into time nodes,a digital twin model of the machining process is established,the Long Short-Term Memory(LSTM)algorithm is introduced to predict the machining quality at each time sequence stage,a constraint relationship between machining quality and process element characteristics is created based on the prediction results,and the machining data is adjusted based on the time nodes.The process data is adjusted based on the time nodes to realize the closed-loop operation of processing-prediction-control.Finally,the effectiveness of the proposed method is verified with a marine diesel engine body.The main research elements of the thesis are as follows:(1)Analysis of factors affecting the machining quality of key parts of marine diesel enginesIn view of the characteristics of marine diesel engine key parts such as many types,small batches and many machining processes,the principles of the design of the machining process scheme for key parts are analyzed.Through the relationship between the machining quality characteristics and the process,the key process selection method of the parts is determined.At the same time,the factors affecting the machining accuracy of the part are analyzed,the causes of single-process errors and multi-process errors and the transmission methods are explained,and a model of the error transmission process is constructed.(2)Machining quality control model based on the optimal configuration of process elementsWith the digital twin processing quality prediction model as the carrier and the process processing quality characteristics as the research object,through real-time sensing and acquisition of processing quality status and prediction data of each process,the processing process is divided into time series,and real-time adjustment is made to the process data of each time series respectively,so as to continuously update and optimize the processing process and realize real-time tracking,analysis and dynamic control of processing quality.(3)Digital twin-based machining quality prediction modelThrough the digital twin’s virtual-real fusion technology,the simulation of the whole process and elements of processing resources in the processing process is used to obtain real-time data in the processing process;then,the correlation mechanism between processing quality characterization parameters and process elements is constructed to determine the relationship between process elements on the processing quality of the process.According to the method of mutual fusion of real-time state,historical data and simulation data of machining resources in the machining process,the machining quality prediction model is constructed by combining short-time neural network algorithm.(4)Marine diesel engine key parts machining quality prediction systemIn order to realize the machining quality prediction of key parts of marine diesel engines,the machining quality prediction system is developed,including process data management module,machining quality monitoring module,machining quality prediction module and machining quality dynamic control module,taking the connecting rod,piston and body of marine diesel engines as validation examples.The effectiveness and practicality of the method proposed in this thesis is verified by combining the application effects of each functional module. |
