Condition Monitoring to Enable Predictive Maintenance on a Six-Die Nut Manufacturing Machine through Force Data Analysis

Nut fasteners are produced by machines working around the clock. Companies generally operate with a run-to-failure or planned maintenance approach. Even with a planned maintenance schedule, however, undetected damage to the dies and non-die parts occurring between maintenance periods can cause considerable downtime and pervasive damage to the machine. To address this shortcoming, force data from the fourth and sixth dies of a six-die nut manufacturing machine were analysed using correlation to the best health condition on the force profile and on the force shock response spectrum profile. Fault features such as quality adjustments and damage to both die and non-die parts were detectable prior to required maintenance or machine failure. This detection was facilitated by the determination of health thresholds, whereby the force SRS profile generated a longer warning period prior to failure. The analytical approach could benefit the industry by identifying damage that would normally go undetected by operators, thereby reducing downtime, extending die life, enabling “as needed” maintenance, and optimising machine operation.

A physical investigation of dimensional and mechanical characteristics of 3D printed nut and bolt for industrial applications

Purpose A nut bolt joint is a primary device that connects mechanical components. The vibrations cause bolted joints to self-loosen. Created by motors and engines, leading to machine failure, and there may be severe safety issues. All the safety issues and self-loosen are directly and indirectly the functions of the accuracy and precision of the fabricated nut and bolt. Recent advancements in three-dimensional (3D) printing technologies now allow for the production of intricate components. These may be used technologies such as 3D printed bolts to create fasteners. This paper aims to investigate dimensional precision, surface properties, mechanical properties and scanning electron microscope (SEM) of the component fabricated using a multi-jet 3D printer. Design/methodology/approach Multi-jet-based 3D printed nut-bolt is evaluated in this paper. More specifically, liquid polymer-based nut-bolt is fabricated in sections 1, 2 and 3 of the base plate. Five nuts and bolts are fabricated in these three sections. Findings Dimensional inquiry (bolt dimension, general dimensions’ density and surface roughness) and mechanical testing (shear strength of nut and bolt) were carried out throughout the study. According to the ISO 2768 requirements for the General Tolerances Grade, the nut and bolt’s dimensional examination (variation in bolt dimension, general dimensions) is within the tolerance grades. As a result, the multi-jet 3D printing (MJP)-based 3D printer described above may be used for commercial production. In terms of mechanical qualities, when the component placement moves from Sections 1 to 3, the density of the manufactured part decreases by 0.292% (percent) and the shear strength of the nut and bolt decreases by 30%. According to the SEM examination, the density of the River markings, sharp edges, holes and sharp edges increased from Sections 1 to 3, which supports the findings mentioned above. Originality/value Hence, this work enlightens the aspects causing time lag during the 3D printing in MJP. It causes variation in the dimensional deviation, surface properties and mechanical properties of the fabricated part, which needs to be explored.

Identification of Most Significant Parameter of Impact of Climate Change and Urbanization on Operational Efficiency of Hydropower Plant

The operational performance of hydropower plants (HPPs) is largely affected as the output from the plant entirely depends on the rainfall and demand from consumers both of which are compromised due to the vulnerability in climatic patterns and rapid change in urbanization rate. Although, not all the parameters are equally affected and the present study aims to find the degree of impact on the various correlated parameters on which production efficiency of HPPs varies. In this aspect, a neural network concept was used as decision making tool to identify the most significant parameters with respect to change in climate, urbanization along with machine failure because as a combined effect of the first two parameters, the probability of machine failure will also increase. The result from the study provides an opportunity to mitigate the impact that can be caused as a result of climate change impact and change in rate of urbanization. According to the result it was found that Efficiency of Generators is the most significant parameter of impact of climate change and urbanization on operational efficiency of hydropower plant. The result from the scenario analysis suggested that if the A2 scenario becomes true in 2061-70 there will be a maximum decrease in the OE and if land use scenario: PR story line is found to be adopted in the future world of 2020-30 the change in OE will be the greatest (an increase of 6.056%) compared to any other scenario developed for the impact of urbanization followed by land use change scenario of the 2031-40 decade, which will be equal to an increase of 5.247% compared to the baseline.

A Q-Learning Rescheduling Approach to the Flexible Job Shop Problem Combining Energy and Productivity Objectives

The flexible job shop problem (FJSP) has been studied in recent decades due to its dynamic and uncertain nature. Responding to a system’s perturbation in an intelligent way and with minimum energy consumption variation is an important matter. Fortunately, thanks to the development of artificial intelligence and machine learning, a lot of researchers are using these new techniques to solve the rescheduling problem in a flexible job shop. Reinforcement learning, which is a popular approach in artificial intelligence, is often used in rescheduling. This article presents a Q-learning rescheduling approach to the flexible job shop problem combining energy and productivity objectives in a context of machine failure. First, a genetic algorithm was adopted to generate the initial predictive schedule, and then rescheduling strategies were developed to handle machine failures. As the system should be capable of reacting quickly to unexpected events, a multi-objective Q-learning algorithm is proposed and trained to select the optimal rescheduling methods that minimize the makespan and the energy consumption variation at the same time. This approach was conducted on benchmark instances to evaluate its performance.

Analisis Kerusakan Bucket Elevator M-145 Dengan Metode Fishbone Diagram

Machine failure on the M-145 bucket elevator contributed to downtime in January 2021, as much as 605 minutes or 6.75 hours, starting from January 17-20 2021 with a total of 17 production downtimes. The purpose of this study was to analyze the damage to the M145 elevator to determine the root cause of the damage and in this study used the fishbone diagram method. From the results of research conducted there was damage to the M145 elevator that the cause of the overflow was due to the peeling of the rubber coating on the top pulley of the upper motor pulley so that the belt became loose and caused friction on the conveyor belt, the tightness of the conveyor belt which was rarely checked, from the beginning of the construction of the bucket which caused the material to be indirect. lifted to the top but first stirred at the bottom so that it inhibits the speed of the bucket elevator. In preventing overflow damage on the M145 elevator machine by repairing or replacing the top pulley with a new pulley or repair by providing a good pulley rubber coating, as well as changing the direction of the elevator inlet so that the bucket is no longer stuck with raw materials

Semiconductor Bearing Fault Recognition

Semiconductor manufacturing is consid-ered to be one of the most technologically complicated manufacturing processes. Bearing, being a critical part of the rotating machinery used in the process, plays an essential role as it supports the mechanical rotating body and decreases the friction coefficient. However, extensive use makes this element a target of health degradation, which indirectly causes machine failure. A defective bearing causes approximately 50% of failures in electrical machines. Hence, there arises a dire need for effective fault detection and diagnosis methods to recog-nise fault patterns and help take preventive measures. This paper carries out a comprehensive comparative study of the pre-existing machine learning and deep learning techniques used for diagnosing bearing faults and further devises a novel framework for bearing fault diagnosis based on the results. Unlike the conventional Fault Detection Classifiers (FDC) that operate in the original data space, this algorithm explores the scope for feature extraction and transferability empowered by the deep learning models used.

Theoretical foundations for calculating bridges for endurance using the kinetic theory of durability of structural materials

The economically effective method for determining the unknown parameters of the dependence of the durability of structural materials on the level of acting constant stresses in them and their absolute temperature for various structural materials is proposed, taking into account the data established by Academician of the USSR Academy of Sciences S.N. Zhurkov. It does not require long-term testing of materials, but is based on the use of the results of short-term standard machine failure of two groups of standard samples of materials at two significantly different temperatures. When using these parameters and the Bailey integral criterion for summing up the losses in the durability of materials, it is possible to calculate the endurance of elements of road bridge structures from any structural materials and to determine the residual durability resource of the structure under the predicted subsequent mode of loading it with real temporary vertical loads.

Predicting machine’s performance data using the stacked long short-term memory (LSTM) neural networks

Purpose Machine Performance Check (MPC) is a daily quality assurance (QA) tool for Varian machines. The daily QA data based on MPC tests show machine performance patterns and potentially provide warning messages for preventive actions. This study developed a neural network model that could predict the trend of data variations quantitively. Methods and materials: MPC data used were collected daily for 3 years. The stacked long short-term memory (LSTM)model was used to develop the neural work model. To compare the stacked LSTM, the autoregressive integrated moving average model (ARIMA) was developed on the same data set. Cubic interpolation was used to double the amount of data to enhance prediction accuracy. After then, the data were divided into 3 groups: 70% for training, 15% for validation, and 15% for testing. The training set and the validation set were used to train the stacked LSTM with different hyperparameters to find the optimal hyperparameter. Furthermore, a greedy coordinate descent method was employed to combinate different hyperparameter sets. The testing set was used to assess the performance of the model with the optimal hyperparameter combination. The accuracy of the model was quantified by the mean absolute error (MAE), root-mean-square error (RMSE), and coefficient of determination (R2). Results A total of 867 data were collected to predict the data for the next 5 days. The mean MAE, RMSE, and \({\text{R}}^{2}\) with all MPC tests was 0.013, 0.020, and 0.853 in LSTM, while 0.021, 0.030, and 0.618 in ARIMA, respectively. The results show that the LSTM outperforms the ARIMA. Conclusions In this study, the stacked LSTM model can accurately predict the daily QA data based on MPC tests. Predicting future performance data based on MPC tests will foresee possible machine failure, allowing early machine maintenance and reducing unscheduled machine downtime.