Research On Counter-gravity Casting Furnace Health Status Assessment Technology

With the transformation of China from a manufacturing power to a manufacturing power,domestic machinery and equipment have been greatly improved in all aspects,but the technical level of high-end machinery and equipment still lags far behind the international advanced level.Among them,the counter gravity casting equipment,represented by low pressure and differential pressure,is the main technical equipment for manufacturing complex thin-walled light alloy components with high metallurgical quality.As a necessary means for quantitative control of the operation state of the casting equipment,the main task of the health state assessment of the counter gravity casting equipment is to measure whether the casting equipment meets the design and use requirements in the working process,to evaluate the health state of the casting equipment in real time,to ensure the efficient and reliable operation of the equipment and the qualified and stable casting quality.In order to make the casting structure compact and high performance,improve the manufacturing capacity of high-quality molding and structural complex,wall thickness mutation and other structural parts,in-depth study of the health status of the counter gravity casting equipment is of great significance to improve the product quality stability and reduce the abnormal downtime in the operation process.Therefore,from the perspective of industrial big data analysis,this paper studies the evaluation method of the health status of the counter gravity casting equipment.The main research contents are as follows:(1)The overall research framework of health assessment of counter gravity casting furnace was established.Combing the structure of the counter gravity casting furnace,analyzing and summarizing the casting process and the mechanism of the counter gravity casting,on this basis,this paper puts forward the counter gravity casting furnace health assessment technology based on the industrial big data analysis technology.The main contents include:first,preprocessing the data and Feature Engineering,and then monitoring the status of the casting equipment by optimizing the machine learning classification algorithm Thirdly,based on the deep learning algorithm,the health indicators of the counter gravity casting furnace are constructed.Finally,the health status of the counter gravity casting furnace is evaluated by the time series algorithm combined with the health evaluation method.(2)The state monitoring model of SVM based on genetic algorithm optimization is constructed.According to the characteristics of complex and changeable working conditions,bad tendency of characteristic parameters and large amount of data in the casting process of casting furnace,a prediction model of condition monitoring of counter gravity casting furnace based on support vector machine is proposed and optimized by genetic algorithm.Through genetic algorithm,the kernel function of SVM is optimized to improve the convergence speed and classification accuracy of the model.By using the support vector machine model,the mapping relationship between data features and fault state is established,so as to realize the condition monitoring of casting equipment.(3)The health index of the counter gravity casting furnace was constructed by using the self encoder algorithm.In view of the problem that the current condition monitoring method needs to master a lot of prior knowledge and engineering practice experience to judge,a construction method of anti gravity casting furnace health index based on self encoder is proposed.Using the nonlinear transformation of deep learning,we can mine deeper information from the data with poor trend.Using self encoder to realize the construction of health status evaluation index in the casting process,considering the characteristics of health status degradation,the correlation evaluation and monotony evaluation are introduced to form the evaluation index to evaluate the health index,which provides support for the follow-up health status evaluation of casting furnace.(4)Based on the long short term memory(LSTM),a health assessment model of the counter gravity casting furnace was established.Based on the health indicators obtained in(3),using the LSTM network codec framework,the deep sensitive feature extraction of the input data is realized.At the same time,combined with the memory ability of LSTM to the time-series data,a health status evaluation method based on the spatial distance of the health status indicators is proposed to realize the dynamic evaluation of the health status of the anti gravity casting furnace.Based on the engineering case analysis of the health assessment of the counter gravity casting furnace,the effectiveness of the health assessment method of the counter gravity casting furnace is verified by comparing the actual working state with the assessment results.