An Improved Learning Based Multilayer Height Control Strategy in LMD Process
| Autor: | Sarbani Mukherjee, A. K. Lohar, Santu Kr. Giri, Apurba Das |
|---|---|
| Rok vydání: | 2017 |
| Předmět: |
0209 industrial biotechnology
Computer science Iterative learning control Process (computing) Particle swarm optimization 02 engineering and technology Solid modeling 021001 nanoscience & nanotechnology 020901 industrial engineering & automation Control theory Process control Deposition (phase transition) 0210 nano-technology MATLAB computer computer.programming_language |
| Zdroj: | 2017 14th IEEE India Council International Conference (INDICON). |
| DOI: | 10.1109/indicon.2017.8488059 |
| Popis: | The Laser Metal Deposition (LMD) is an emerging manufacturing technology based on additive manufacturing principle. To achieve a desired geometry of overall part, accurate modeling of the process is important before approaching to control the process. The deposition of successive layers in metal 3D printing is sensitive to variations in process parameters and disturbances in the deposition process. In this work the layer height is controlled using a semi-emperical model of the LMD process relating the scan velocity to layer thickness. The control algorithm in the proposed process uses an optimization method, Particle Swarm Optimization (PSO) to estimate the model parameter which identifies the process model correctly. A simple proportional-integral (PI) controller is used here along with an iterative learning controller (ILC) in the feed-forward path of the closed loop control system in order to enhance the tracking performance of the controller. The performance of the proposed control algorithm is evaluated through simulations in MATLAB/Simulink platform and validated in the experimental setup. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|