On the Problem of State Recognition in Injection Molding Based on Accelerometer Data Sets

Abstract

The last few decades have been characterised by a very active application of smart technologies in various fields of industry. This paper deals with industrial activities, such as injection molding, where it is required to monitor continuously the manufacturing process to identify both the effective running time and down-time periods. Supervised machine learning algorithms are developed to recognize automatically the periods of the injection molding machines. The former algorithm uses directly the features of the descriptive statistics, while the latter one utilizes a convolutional neural network. The automatic state recognition system is equipped with an 3D-accelerometer sensor whose datasets are used to train and verify the proposed algorithms. The novelty of our contribution is that accelerometer data-based machine learning models are used to distinguish producing and non-producing periods by means of recognition of key steps in an injection molding cycle. The first testing results show the approximate overall balanced accuracy of 72–92% that illustrates the large potential of the monitoring system with the accelerometer. According to the ANOVA test, there are no sufficient statistical differences between the comparative algorithms, but the results of the neural network exhibit higher variances of the defined accuracy metrics.

Development of an Open-Source Injection Mold Monitoring System

“…Additionally, cycles in which part extraction took longer than normal can be identified and coincide with the extraction pin initially jamming due to the added friction. The data collected from the 3D accelerometer could be used to identify and monitor different machinery movements and respective process phases, similar to the results presented by Moreira et al [ 8 ] and Brunthaler et al [ 22 ]. As broadly verified in those works, the most noticeable vibrations in amplitude, and with a significant duration, were observed for the part ejection phase.…”

A High-Linearity Closed-Loop Capacitive Micro-Accelerometer Based on Ring-Diode Capacitance Detection

“…The capacitive accelerometer can detect the capacitance change caused by proof mass displacement under the action of inertia force and solve the acceleration information. The MEMS capacitive accelerometer, with the characteristics of low cost, high precision, low power consumption, and small size [ 1 ], is widely used in various environments [ 2 , 3 ]. Due to its small size, the capacitance of the MEMS capacitive accelerometer is in the order of pF, and the capacitance detection accuracy needs to reach the order of aF [ 4 ], which strives for greater requirements for the signal-to-noise ratio of the sensitive structure and the detection circuit.…”