Automating data transfer and implementing data mining techniques to analyze and predict the performance of industrial heating and curing Equipment (Masterarbeit)

Kurzfassung

Digitalization is one of the most important and vastly growing fields among almost all forms of the industries. It can give competitive advantages to the organizations by achieving results faster, better and in more efficient manner. The work in this thesis involves digitalizing the data of pyrolysis and siliconisation of various products produced at DLR. This data was previously stored in form of excel and CSV files. A PostgreSQL database was developed to save all of the data at one place in digitalized and managed form. The main purpose of this work was to automate data analysis and product development. The data analysis also evaluated the performance of the pyrolysis and siliconisation ovens. The database that was developed to store and deal with this data had a proper structure. The data of all products was stored by following that structure. This made it easier to access different attributes of the data of more than one product at the same time. The data of previous years stored in form of excel and CSV files was read and inserted to the database using Python. Furthermore, a user interface was also developed to analyze and handle the data stored in the database. Various revelations were discovered while analyzing and storing this data. The ovens that were used for pyrolysis and siliconisation of products performed well while heating the products. But they had lesser control while cooling these products. There were also a few products placed in the oven which did not have proper names or their appropriate names were not mentioned in their respective sheets. This made it difficult to track the history of the products and analyze them. Overall, this work is of significant importance because it is easier to monitor the errors in the data and perform analysis. The data is going to be stored in future will also be checked by the database and any possibility of error in its name or structure will be automatically pointed out. The efficiency of the ovens can also be continuously monitored.