Active vibration control, artificial intelligence applications and internet of things applications in turning processes

In recent years, with the development of sensor technologies, communication platforms, cyber-physical systems, storage technologies and internet applications, the way to produce competitive products with high quality and low cost has been opened. In this process called "Industry 4.0", smart manufacturing has been started by integrating the Internet of Things (IoT), artificial intelligence applications, learning robots, cyber-physical systems, big data and analysis, cloud technologies, three-dimensional printers, simulation-digital twin technologies into production. In this thesis study; Based on smart manufacturing practices, it is aimed to obtain high quality products for turning, which is an important process in machining. Within the scope of the thesis, firstly, turning process was performed for S235JR steel and cutting parameters were estimated for the best values of surface roughness and cutting tool tip temperature by using Artificial Neural Network (ANN) model and Multiple Linear Regression Model. In the other part of the estimation studies, ANFIS trained with PSO, ACO and GA metaheuristic algorithms was used to predict tool vibrations and surface roughness during dry turning of AA6013 Al alloy. Secondly, the relationships between surface roughness and tool vibration are investigated for AA6013. The effects of cutting parameters such as depth of cut, feed rate and spindle speed on tool vibration and surface roughness were investigated. The specified cutting parameters are optimized for tool life and best product quality. Third, within the framework of the industrial IoT concept, three axes acceleration data were collected from the tool during turning and a status monitoring/display system was designed to inform the user via SMS in case the frequency reaches undesirable levels. Finally, an active vibration control based on sliding mode control has been implemented in order to reduce and eliminate chatter, which is an undesirable situation for the turning process. In the study, three-axis acceleration data were collected from the tool during the turning of Ti6Al4V. In order to use the collected raw data in control, FIR filtering, FFT analysis and integration were taken in order to adapt it to the mathematical model. After the application of active vibration control, it was observed that chattering was reduced by 85%. The surface quality formed by the amplitude of the graph obtained after the active control was compared and verified with the data obtained from the actual production result.