Due to the complex production process, long manufacturing process, and large production volume, sand casting products have problems such as large quality fluctuations and difficult quality traceability. In recent years, although with the introduction of information management systems such as ERP and MES, casting enterprises have been able to collect more data, the existing information has not been transformed into knowledge, or only a small amount of data has been utilized. Therefore, a data mining method based on Huazhu ERP system is proposed, which deeply mines and analyzes all collected data on the basis of information management. The mining results are presented in the form of data tables, scatter plots, pie charts, etc. At the same time,sand casting based on the correlation between data, neural networks were used to predict the quality of products with different process parameters, in order to improve the quality of enterprise products Control ability, improve enterprise efficiency.
Introduction
With the continuous development of computer technology, digitalization, networking, and intelligent technology have become the main force in the development of manufacturing industry, and the main driving force of the new round of industrial revolution, providing a new technological path for quality control in manufacturing industry. Sand casting is one of the important branches of casting, occupying an important position in the automotive industry such as engine cylinder head production. Due to the complex production process, long manufacturing process, and large production volume of sand casting, its products have problems such as large quality fluctuations and difficult quality traceability. sand casting In recent years, under the new situation of industrial upgrading, environmental protection, and customer demand, sand casting enterprises have also been transforming towards automated and information-based production. Many data collection devices have been used in front-line production and integrated with information management systems. These information systems record a large amount of real production data such as equipment, processes, environment, and quality in casting production,sand casting which is a wealth that includes the relationship between production parameters and casting quality. However, the existing “wealth” has not been transformed into information and knowledge, or only a small amount of data is utilized, resulting in an embarrassing situation of “data explosion and knowledge shortage” within the enterprise. If we can conduct in-depth mining and segmentation of all collected data on the basis of information management Analyzing and identifying the coupling relationship between various links and parameters will undoubtedly help casting enterprises better control quality and trace defects. It can be seen that mining data from sand casting enterprises is of great significance. Data mining is the process of analyzing large amounts of existing data to reveal meaningful new systems, trends, and patterns. It is the process and technique of discovering potential and valuable information from random, massive, noisy, incomplete, and fuzzy large databases, and is also a decision support process. In the field of artificial intelligence, it is commonly referred to as knowledge discovery in databases, and some people consider data mining as a fundamental step in the process of knowledge discovery in databases. As an emerging technology,sand casting data mining is currently at a critical stage in its lifecycle, requiring time and effort to research, develop, and ultimately be accepted by people. At present, many fields have been analyzed through data mining techniques, such as financial management, insurance, and management system research, while data mining research in the foundry industry is still in its infancy.
Data mining methods include neural network methods, decision tree algorithms, association analysis, rough set methods, fuzzy set methods, statistical analysis methods, covering positive and rejecting negative examples methods, visualization techniques, sand casting etc. At the same time, through data mining models, the process parameters of each product are also interrelated and can form a mapping relationship with the final quality results. Therefore, the relevant process parameters of each product can be used as inputs, and the final quality results can be used as outputs to train neural network models, sand casting which can predict the quality of products using different process parameters.
Conclusion
We have conducted research on the problem of “data explosion and knowledge deficiency” in current casting enterprises due to the lack of data mining and analysis work or the limited utilization of data after introducing information management systems.
(1) A system based data mining model is proposed based on the production practice and system usage of Y Company, a sand casting enterprise. The model uses association analysis method with casting numbers as input objects, and forms associations between orders, planned production, production parameters, quality registration and other tables in the form of primary and foreign keys through SQL Server database, thereby linking various business processes of the system.
(2) By using the system to perform statistical analysis on data, deeper information about customers, defect distribution status, production capacity, and other aspects can be obtained, which can be presented in the form of data tables and various statistical charts.
(3) On the basis of this data mining model, it is also possible to trace the quality problems of casting products, and in the future, neural networks can be used to predict the process quality of different parameters, so as to fully utilize the role of data, improve the control level of product quality in enterprises, and promote better development of enterprises.