MANAGEMENT | PROCESS CONTROL
“There is no doubt that the most crucial mechanical component in the twin-screw extruder is the gearbox. Unexpected and usually catastrophic failure of this component can take weeks, some- times months, to repair.” For this reason, Okimoto says many extruder
Above: Increased demand for process data and automation is driven by compliance demands and scarcity of skilled staff, according to Maag
Maag Integration Platform options include a
variety of sensor-based solutions that support the machine communicating its status and shed light on the inner workings of the ‘black box’ com- pounding process. This allows data-based deci- sions to be taken in real time to help reduce energy and resource consumption, increase product quality, enable predictive maintenance, and facilitate better production planning. Stadler says that Maag’s future developments will include a variety of machine learning applications, built with the aim to support customers with automatic process setting improvement.
Predicting failures Japanese extruder maker JSW has developed a predictive alert system application for the extruder gearbox. “All equipment downtime due to machine failure becomes a critical factor for decreasing productivity on twin-screw extruders,” says Tasuku Okimoto, Chief Engineer of IoT development.
users take additional precautions for maintaining gearbox reliability, including periodic oil changes, daily vibration measurement and scheduled overhauls. “However, for various reasons gearbox- es still fail occasionally,” he says. “To minimise the risk of unexpected failure of the extruder gearbox, we have been developing a technology to detect actual micro-deterioration within the gears and bearings, and alert this before it fails completely.” The JSW Gearbox Alert System is a predictive
alert system for the gearbox of its TEX twin-screw extruder. The system monitors the signals from multiple vibration sensors mounted on various locations around the gearbox. Evaluation of abnormalities of the internal components are undertaken by frequency domain and time domain statistical analysis. The prototype of this predictive alert system was installed for a number of selected extruder users some years ago, and the company has now been acquiring data over an extended period of time under actual continuous operational conditions. Figure 1 shows vibration levels over time obtained by the prototype detection system. The transition of RMS (root mean square) of vibration shows an increasing tendency for vibration after point A. As well as vibration RMS, the system uses multiple analytical methods, including AI-based decision logic, to determine accurately when to provide an
www.xindacorp.com
info@xindacorp.com
IMAGE: MAAG
Page 1 |
Page 2 |
Page 3 |
Page 4 |
Page 5 |
Page 6 |
Page 7 |
Page 8 |
Page 9 |
Page 10 |
Page 11 |
Page 12 |
Page 13 |
Page 14 |
Page 15 |
Page 16 |
Page 17 |
Page 18 |
Page 19 |
Page 20 |
Page 21 |
Page 22 |
Page 23 |
Page 24 |
Page 25 |
Page 26 |
Page 27 |
Page 28 |
Page 29 |
Page 30 |
Page 31 |
Page 32 |
Page 33 |
Page 34 |
Page 35 |
Page 36 |
Page 37 |
Page 38 |
Page 39 |
Page 40 |
Page 41 |
Page 42 |
Page 43 |
Page 44 |
Page 45 |
Page 46 |
Page 47 |
Page 48 |
Page 49 |
Page 50 |
Page 51 |
Page 52 |
Page 53 |
Page 54 |
Page 55 |
Page 56 |
Page 57 |
Page 58 |
Page 59 |
Page 60 |
Page 61 |
Page 62 |
Page 63 |
Page 64 |
Page 65 |
Page 66 |
Page 67 |
Page 68 |
Page 69 |
Page 70 |
Page 71 |
Page 72 |
Page 73 |
Page 74
