Research

31 pages, 5515 KiB

Open AccessEditor’s ChoiceArticle

UAV and IoT-Based Systems for the Monitoring of Industrial Facilities Using Digital Twins: Methodology, Reliability Models, and Application

by Yun Sun, Herman Fesenko, Vyacheslav Kharchenko, Luo Zhong, Ihor Kliushnikov, Oleg Illiashenko, Olga Morozova and Anatoliy Sachenko

Sensors 2022, 22(17), 6444; https://0-doi-org.brum.beds.ac.uk/10.3390/s22176444 - 26 Aug 2022

Cited by 17 | Viewed by 2500

Abstract

This paper suggests a methodology (conception and principles) for building two-mode monitoring systems (SMs) for industrial facilities and their adjacent territories based on the application of unmanned aerial vehicle (UAV), Internet of Things (IoT), and digital twin (DT) technologies, and a set of [...] Read more.

This paper suggests a methodology (conception and principles) for building two-mode monitoring systems (SMs) for industrial facilities and their adjacent territories based on the application of unmanned aerial vehicle (UAV), Internet of Things (IoT), and digital twin (DT) technologies, and a set of SM reliability models considering the parameters of the channels and components. The concept of building a reliable and resilient SM is proposed. For this purpose, the von Neumann paradigm for the synthesis of reliable systems from unreliable components is developed. For complex SMs of industrial facilities, the concept covers the application of various types of redundancy (structural, version, time, and space) for basic components—sensors, means of communication, processing, and presentation—in the form of DTs for decision support systems. The research results include: the methodology for the building and general structures of UAV-, IoT-, and DT-based SMs in industrial facilities as multi-level systems; reliability models for SMs considering the applied technologies and operation modes (normal and emergency); and industrial cases of SMs for manufacture and nuclear power plants. The results obtained are the basis for further development of the theory and for practical applications of SMs in industrial facilities within the framework of the implementation and improvement of Industry 4.0 principles. Full article

(This article belongs to the Special Issue Complex Data Processing Systems and Computing Algorithms: New Concepts and Applications)

► Show Figures

Figure 1

23 pages, 5869 KiB

Open AccessArticle

Discrete Missing Data Imputation Using Multilayer Perceptron and Momentum Gradient Descent

by Hu Pan, Zhiwei Ye, Qiyi He, Chunyan Yan, Jianyu Yuan, Xudong Lai, Jun Su and Ruihan Li

Sensors 2022, 22(15), 5645; https://0-doi-org.brum.beds.ac.uk/10.3390/s22155645 - 28 Jul 2022

Cited by 7 | Viewed by 1572

Abstract

Data are a strategic resource for industrial production, and an efficient data-mining process will increase productivity. However, there exist many missing values in data collected in real life due to various problems. Because the missing data may reduce productivity, missing value imputation is [...] Read more.

Data are a strategic resource for industrial production, and an efficient data-mining process will increase productivity. However, there exist many missing values in data collected in real life due to various problems. Because the missing data may reduce productivity, missing value imputation is an important research topic in data mining. At present, most studies mainly focus on imputation methods for continuous missing data, while a few concentrate on discrete missing data. In this paper, a discrete missing value imputation method based on a multilayer perceptron (MLP) is proposed, which employs a momentum gradient descent algorithm, and some prefilling strategies are utilized to improve the convergence speed of the MLP. To verify the effectiveness of the method, experiments are conducted to compare the classification accuracy with eight common imputation methods, such as the mode, random, hot-deck, KNN, autoencoder, and MLP, under different missing mechanisms and missing proportions. Experimental results verify that the improved MLP model (IMLP) can effectively impute discrete missing values in most situations under three missing patterns. Full article

(This article belongs to the Special Issue Complex Data Processing Systems and Computing Algorithms: New Concepts and Applications)

► Show Figures

Figure 1

17 pages, 1762 KiB

Open AccessArticle

FDNet: Knowledge and Data Fusion-Driven Deep Neural Network for Coal Burst Prediction

by Anye Cao, Yaoqi Liu, Xu Yang, Sen Li and Yapeng Liu

Sensors 2022, 22(8), 3088; https://0-doi-org.brum.beds.ac.uk/10.3390/s22083088 - 18 Apr 2022

Cited by 5 | Viewed by 2304

Abstract

Coal burst prediction is an important research hotspot in coal mine production safety. This paper presents FDNet, which is a knowledge and data fusion-driven deep neural network for coal burst prediction. The main idea of FDNet is to extract explicit features based on [...] Read more.

Coal burst prediction is an important research hotspot in coal mine production safety. This paper presents FDNet, which is a knowledge and data fusion-driven deep neural network for coal burst prediction. The main idea of FDNet is to extract explicit features based on the existing mine seismic physical model and utilize deep learning to automatically extract the implicit features of mine microseismic data. The key innovations of FDNet include an expert knowledge indicator selection method based on a subset search strategy, a mine microseismic data extraction method based on a deep convolutional neural network, and a feature deep fusion method of mine microseismic data based on an attention mechanism. We conducted a set of engineering experiments in Gaojiapu Coal Mine to evaluate the performance of FDNet. The results show that compared with the state-of-the-art data-driven machines and knowledge-driven methods, the prediction accuracy of FDNet is improved by 5% and 16%, respectively. Full article

(This article belongs to the Special Issue Complex Data Processing Systems and Computing Algorithms: New Concepts and Applications)

► Show Figures

Figure 1

32 pages, 1115 KiB

Open AccessArticle

Clustering at the Disposal of Industry 4.0: Automatic Extraction of Plant Behaviors

by Dylan Molinié, Kurosh Madani and Véronique Amarger

Sensors 2022, 22(8), 2939; https://0-doi-org.brum.beds.ac.uk/10.3390/s22082939 - 12 Apr 2022

Cited by 5 | Viewed by 1959

Abstract

For two centuries, the industrial sector has never stopped evolving. Since the dawn of the Fourth Industrial Revolution, commonly known as Industry 4.0, deep and accurate understandings of systems have become essential for real-time monitoring, prediction, and maintenance. In this paper, we propose [...] Read more.

For two centuries, the industrial sector has never stopped evolving. Since the dawn of the Fourth Industrial Revolution, commonly known as Industry 4.0, deep and accurate understandings of systems have become essential for real-time monitoring, prediction, and maintenance. In this paper, we propose a machine learning and data-driven methodology, based on data mining and clustering, for automatic identification and characterization of the different ways unknown systems can behave. It relies on the statistical property that a regular demeanor should be represented by many data with very close features; therefore, the most compact groups should be the regular behaviors. Based on the clusters, on the quantification of their intrinsic properties (size, span, density, neighborhood) and on the dynamic comparisons among each other, this methodology gave us some insight into the system’s demeanor, which can be valuable for the next steps of modeling and prediction stages. Applied to real Industry 4.0 data, this approach allowed us to extract some typical, real behaviors of the plant, while assuming no previous knowledge about the data. This methodology seems very promising, even though it is still in its infancy and that additional works will further develop it. Full article

(This article belongs to the Special Issue Complex Data Processing Systems and Computing Algorithms: New Concepts and Applications)

► Show Figures

Figure 1

13 pages, 25801 KiB

Open AccessArticle

A Hybrid Bipolar Active Charge Balancing Technique with Adaptive Electrode Tissue Interface (ETI) Impedance Variations for Facial Paralysis Patients

by Ganesh Lakshmana Kumar Moganti, V. N. Siva Praneeth and Siva Rama Krishna Vanjari

Sensors 2022, 22(5), 1756; https://0-doi-org.brum.beds.ac.uk/10.3390/s22051756 - 23 Feb 2022

Cited by 1 | Viewed by 1763

Abstract

Functional electrical stimulation (FES) is a safe, effective, and general approach for treating various neurological disorders. However, in the case of FES usage for implantable applications, charge balancing is a significant challenge due to variations in the fabrication process and electrode tissue interface [...] Read more.

Functional electrical stimulation (FES) is a safe, effective, and general approach for treating various neurological disorders. However, in the case of FES usage for implantable applications, charge balancing is a significant challenge due to variations in the fabrication process and electrode tissue interface (ETI) impedance. In general, an active charge balancing approach is being used for this purpose, which has limitations of additional power consumption for residual voltage calibration and undesired neurological responses. To overcome these limitations, this paper presents a reconfigurable calibration circuit to address both ETI variations and charge balancing issues. This reconfigurable calibration circuit works in two modes: An impedance measurement mode (IMM) for treating ETI variations and a hybrid charge balancing mode (HCBM) for handling charge balance issues. The IMM predicts the desired stimulation currents by measuring the ETI. The HCBM is a hybrid combination of electrode shorting, offset regulation, and pulse modulation that takes the best features of each of these techniques and applies them in appropriate situations. From the results, it is proved that the proposed IMM configuration and HCBM configuration have an optimal power consumption of less than 44

μ

W with a power ratio ranging from 1.74 to 5.5 percent when compared to conventional approaches. Full article

(This article belongs to the Special Issue Complex Data Processing Systems and Computing Algorithms: New Concepts and Applications)

► Show Figures

Figure 1

13 pages, 3117 KiB

Open AccessArticle

Modeling the Voltage Produced by Ultrasound in Seawater by Stochastic and Artificial Intelligence Methods

by Alina Bărbulescu and Cristian Ștefan Dumitriu

Sensors 2022, 22(3), 1089; https://0-doi-org.brum.beds.ac.uk/10.3390/s22031089 - 30 Jan 2022

Cited by 2 | Viewed by 2221

Abstract

Experiments have proved that an electrical signal appears in the ultrasonic cavitation field; its properties are influenced by the ultrasound frequency, the liquid type, and liquid characteristics such as density, viscosity, and surface tension. Still, the features of the signals are not entirely [...] Read more.

Experiments have proved that an electrical signal appears in the ultrasonic cavitation field; its properties are influenced by the ultrasound frequency, the liquid type, and liquid characteristics such as density, viscosity, and surface tension. Still, the features of the signals are not entirely known. Therefore, we present the results on modeling the voltage collected in seawater, in ultrasound cavitation produced by a 20 kHz frequency generator, working at 80 W. Comparisons of the Box–Jenkins approaches, with artificial intelligence methods (GRNN) and hybrid (Wavelet-ARIMA and Wavelet-ANN) are provided, using different goodness of fit indicators. It is shown that the last approach gave the best model. Full article

(This article belongs to the Special Issue Complex Data Processing Systems and Computing Algorithms: New Concepts and Applications)

► Show Figures

Figure 1

31 pages, 10875 KiB

Open AccessArticle

Machine Learning Techniques for Increasing Efficiency of the Robot’s Sensor and Control Information Processing

by Yuriy Kondratenko, Igor Atamanyuk, Ievgen Sidenko, Galyna Kondratenko and Stanislav Sichevskyi

Sensors 2022, 22(3), 1062; https://0-doi-org.brum.beds.ac.uk/10.3390/s22031062 - 29 Jan 2022

Cited by 18 | Viewed by 3682

Abstract

Real-time systems are widely used in industry, including technological process control systems, industrial automation systems, SCADA systems, testing, and measuring equipment, and robotics. The efficiency of executing an intelligent robot’s mission in many cases depends on the properties of the robot’s sensor and [...] Read more.

Real-time systems are widely used in industry, including technological process control systems, industrial automation systems, SCADA systems, testing, and measuring equipment, and robotics. The efficiency of executing an intelligent robot’s mission in many cases depends on the properties of the robot’s sensor and control systems in providing the trajectory planning, recognition of the manipulated objects, adaptation of the desired clamping force of the gripper, obstacle avoidance, and so on. This paper provides an analysis of the approaches and methods for real-time sensor and control information processing with the application of machine learning, as well as successful cases of machine learning application in the synthesis of a robot’s sensor and control systems. Among the robotic systems under investigation are (a) adaptive robots with slip displacement sensors and fuzzy logic implementation for sensor data processing, (b) magnetically controlled mobile robots for moving on inclined and ceiling surfaces with neuro-fuzzy observers and neuro controllers, and (c) robots that are functioning in unknown environments with the prediction of the control system state using statistical learning theory. All obtained results concern the main elements of the two-component robotic system with the mobile robot and adaptive manipulation robot on a fixed base for executing complex missions in non-stationary or uncertain conditions. The design and software implementation stage involves the creation of a structural diagram and description of the selected technologies, training a neural network for recognition and classification of geometric objects, and software implementation of control system components. The Swift programming language is used for the control system design and the CreateML framework is used for creating a neural network. Among the main results are: (a) expanding the capabilities of the intelligent control system by increasing the number of classes for recognition from three (cube, cylinder, and sphere) to five (cube, cylinder, sphere, pyramid, and cone); (b) increasing the validation accuracy (to 100%) for recognition of five different classes using CreateML (YOLOv2 architecture); (c) increasing the training accuracy (to 98.02%) and testing accuracy (to 98.0%) for recognition of five different classes using Torch library (ResNet34 architecture) in less time and number of epochs compared with Create ML (YOLOv2 architecture); (d) increasing the training accuracy (to 99.75%) and testing accuracy (to 99.2%) for recognition of five different classes using Torch library (ResNet34 architecture) and fine-tuning technology; and (e) analyzing the effect of dataset size impact on recognition accuracy with ResNet34 architecture and fine-tuning technology. The results can help to choose efficient (a) design approaches for control robotic devices, (b) machine-learning methods for performing pattern recognition and classification, and (c) computer technologies for designing control systems and simulating robotic devices. Full article

(This article belongs to the Special Issue Complex Data Processing Systems and Computing Algorithms: New Concepts and Applications)

► Show Figures

Figure 1

19 pages, 7049 KiB

Open AccessArticle

Forecasting Day-Ahead Electricity Metrics with Artificial Neural Networks

by Milutin Pavićević and Tomo Popović

Sensors 2022, 22(3), 1051; https://0-doi-org.brum.beds.ac.uk/10.3390/s22031051 - 28 Jan 2022

Cited by 19 | Viewed by 4005

Abstract

As artificial neural network architectures grow increasingly more efficient in time-series prediction tasks, their use for day-ahead electricity price and demand prediction, a task with very specific rules and highly volatile dataset values, grows more attractive. Without a standardized way to compare the [...] Read more.

As artificial neural network architectures grow increasingly more efficient in time-series prediction tasks, their use for day-ahead electricity price and demand prediction, a task with very specific rules and highly volatile dataset values, grows more attractive. Without a standardized way to compare the efficiency of algorithms and methods for forecasting electricity metrics, it is hard to have a good sense of the strengths and weaknesses of each approach. In this paper, we create models in several neural network architectures for predicting the electricity price on the HUPX market and electricity load in Montenegro and compare them to multiple neural network models on the same basis (using the same dataset and metrics). The results show the promising efficiency of neural networks in general for the task of short-term prediction in the field, with methods combining fully connected layers and recurrent neural or temporal convolutional layers performing the best. The feature extraction power of convolutional layers shows very promising results and recommends the further exploration of temporal convolutional networks in the field. Full article

(This article belongs to the Special Issue Complex Data Processing Systems and Computing Algorithms: New Concepts and Applications)

► Show Figures

Figure 1

23 pages, 10560 KiB

Open AccessArticle

LSTM in Algorithmic Investment Strategies on BTC and S&P500 Index

by Jakub Michańków, Paweł Sakowski and Robert Ślepaczuk

Sensors 2022, 22(3), 917; https://0-doi-org.brum.beds.ac.uk/10.3390/s22030917 - 25 Jan 2022

Cited by 15 | Viewed by 5399

Abstract

We use LSTM networks to forecast the value of the BTC and S&P500 index, using data from 2013 to the end of 2020, with the following frequencies: daily, 1 h, and 15 min data. We introduce our innovative loss function, which improves the [...] Read more.

We use LSTM networks to forecast the value of the BTC and S&P500 index, using data from 2013 to the end of 2020, with the following frequencies: daily, 1 h, and 15 min data. We introduce our innovative loss function, which improves the usefulness of the forecasting ability of the LSTM model in algorithmic investment strategies. Based on the forecasts from the LSTM model we generate buy and sell investment signals, employ them in algorithmic investment strategies and create equity lines for our investment. For this purpose we use various combinations of LSTM models, optimized on in-sample period and tested on out-of-sample period, using rolling window approach. We pay special attention to data preprocessing in the input layer, to avoid overfitting in the estimation and optimization process, and assure correct selection of hyperparameters at the beginning of our tests. The next stage is devoted to the conjunction of signals from various frequencies into one ensemble model, and the selection of best combinations for the out-of-sample period, through optimization of the given criterion in a similar way as in the portfolio analysis. Finally, we perform a sensitivity analysis of the main parameters and hyperparameters of the model. Full article

(This article belongs to the Special Issue Complex Data Processing Systems and Computing Algorithms: New Concepts and Applications)

► Show Figures

Figure 1

Journal Menu

Journal Browser

Complex Data Processing Systems and Computing Algorithms: New Concepts and Applications

Share This Special Issue

Special Issue Editors

Special Issue Information

Keywords

Published Papers (9 papers)

Research

Further Information

Guidelines

MDPI Initiatives

Follow MDPI