An Industrial Digitalization Platform for Condition Monitoring and Predictive Maintenance of Pumping Equipment

Short, Michael; Twiddle, J.A.

doi:10.3390/s19173781

Cited by 43 publications

(23 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…It has become necessary to apply the Industry 4.0 paradigm in CPS systems for Internet access. Gateway devices may be used for that purpose (Short and Twiddle 2019). However, it is best to provide Industry 4.0 integrity at the point of the system design ('Intelligent gateway for Industry 2020).…”

Section: Related Workmentioning

confidence: 99%

FPGA based Industrial Ethernet Network Analyser for Real-time Systems Providing Openness for Industry 4.0

Stój

Ziębiński

Cupek

2021

Enterprise Information Systems

View full text Add to dashboard Cite

In modern cyber-physical systems, there is a high degree of connectivity that is imperative to satisfy the Four Industrial Revolution paradigm (Industry 4.0). While creating new systems according to the Industry 4.0, there are several protocols and solutions that can be applied. However, the introduction of Industry 4.0 to the existing industrial systems is not so straightforward. One point of consideration is the necessity of system modification, and another is the potential security threats a system could face. A solution presented in this paper addressing these issues is a FPGA-based network analyser with network datagram filtering and process data extraction. The analyser is transparent and may be applied without alterations to the monitored system. Moreover, there is no risk of influencing the system operation either by the data extraction device itself or by unauthorised access from the Internet, because the FPGA is a passive device operating as a data diode. The presented solution allows for the integration of existing Ethernetbased systems with cloud services for further processing of the data obtained from the monitored system. The FPGA makes process data available for the enterprise information systems.

show abstract

Section: Related Workmentioning

confidence: 99%

FPGA based Industrial Ethernet Network Analyser for Real-time Systems Providing Openness for Industry 4.0

Stój

Ziębiński

Cupek

2021

Enterprise Information Systems

View full text Add to dashboard Cite

show abstract

“…To learn, first the weights w i and bias b are randomly initialized. A widely used activation function is the rectified linear unit (ReLU), as calculated in Equation (2).…”

Section: How Neural Network Learning Is Performedmentioning

confidence: 99%

“…Prognostic and health management (PHM) has become increasingly important in maintaining the integrity of automotive, aerospace, and manufacturing systems [1][2][3][4]. Machines are now fully equipped with sensors that constantly gather information regarding their status.…”

Section: Introductionmentioning

confidence: 99%

Deep Learning with Dynamically Weighted Loss Function for Sensor-Based Prognostics and Health Management

Rengasamy

Jafari

Rothwell

et al. 2020

Sensors

View full text Add to dashboard Cite

Deep learning has been employed to prognostic and health management of automotive and aerospace with promising results. Literature in this area has revealed that most contributions regarding deep learning is largely focused on the model’s architecture. However, contributions regarding improvement of different aspects in deep learning, such as custom loss function for prognostic and health management are scarce. There is therefore an opportunity to improve upon the effectiveness of deep learning for the system’s prognostics and diagnostics without modifying the models’ architecture. To address this gap, the use of two different dynamically weighted loss functions, a newly proposed weighting mechanism and a focal loss function for prognostics and diagnostics task are investigated. A dynamically weighted loss function is expected to modify the learning process by augmenting the loss function with a weight value corresponding to the learning error of each data instance. The objective is to force deep learning models to focus on those instances where larger learning errors occur in order to improve their performance. The two loss functions used are evaluated using four popular deep learning architectures, namely, deep feedforward neural network, one-dimensional convolutional neural network, bidirectional gated recurrent unit and bidirectional long short-term memory on the commercial modular aero-propulsion system simulation data from NASA and air pressure system failure data for Scania trucks. Experimental results show that dynamically-weighted loss functions helps us achieve significant improvement for remaining useful life prediction and fault detection rate over non-weighted loss function predictions.

show abstract

“…A detailed description of digitalization processes of factories, according to the Industry 4.0 paradigm, is proposed by Belli et al [37], together with an application developed for monitoring the production and the quality control. Short and Twiddle [38] developed the digitalization of a water industry application, while Li et al [39] developed an AR application for the control and adjustment of robots during a human-robot interaction based working task. They implemented the DT of human hands by means of LeapMotion sensor and a Kinect V2 camera.…”

Section: Previous Researchesmentioning

confidence: 99%

Towards Digital Twin Implementation for Assessing Production Line Performance and Balancing

Fera

Greco

Caterino

et al. 2019

Sensors

View full text Add to dashboard Cite

The optimization of production processes has always been one of the cornerstones for manufacturing companies, aimed to increase their productivity, minimizing the related costs. In the Industry 4.0 era, some innovative technologies, perceived as far away until a few years ago, have become reachable by everyone. The massive introduction of these technologies directly in the factories allows interconnecting the resources (machines and humans) and the entire production chain to be kept under control, thanks to the collection and the analyses of real production data, supporting the decision making process. This article aims to propose a methodological framework that, thanks to the use of Industrial Internet of Things-IoT devices, in particular the wearable sensors, and simulation tools, supports the analyses of production line performance parameters, by considering both experimental and numerical data, allowing a continuous monitoring of the line balancing and performance at varying of the production demand. A case study, regarding a manual task of a real manufacturing production line, is presented to demonstrate the applicability and the effectiveness of the proposed procedure.

show abstract

An Industrial Digitalization Platform for Condition Monitoring and Predictive Maintenance of Pumping Equipment

Cited by 43 publications

References 16 publications

FPGA based Industrial Ethernet Network Analyser for Real-time Systems Providing Openness for Industry 4.0

FPGA based Industrial Ethernet Network Analyser for Real-time Systems Providing Openness for Industry 4.0

Deep Learning with Dynamically Weighted Loss Function for Sensor-Based Prognostics and Health Management

Towards Digital Twin Implementation for Assessing Production Line Performance and Balancing

Contact Info

Product

Resources

About