Edge Computing for Data Anomaly Detection of Multi-Sensors in Underground Mining

Abstract

There is a growing interest in safety warning of underground mining due to the huge threat being faced by those working in underground mining. Data acquisition of sensors based on Internet of Things (IoT) is currently the main method, but the data anomaly detection and analysis of multi-sensors is a challenging task: firstly, the data that are collected by different sensors of underground mining are heterogeneous; secondly, real-time is required for the data anomaly detection of safety warning. Currently, there are many anomaly detection methods, such as traditional clustering methods K-means and C-means. Meanwhile, Artificial Intelligence (AI) is widely used in data analysis and prediction. However, K-means and C-means cannot directly process heterogeneous data, and AI algorithms require equipment with high computing and storage capabilities. IoT equipment of underground mining cannot perform complex calculation due to the limitation of energy consumption. Therefore, many existing methods cannot be directly used for IoT applications in underground mining. In this paper, a multi-sensors data anomaly detection method based on edge computing is proposed. Firstly, an edge computing model is designed, and according to the computing capabilities of different types of devices, anomaly detection tasks are migrated to different edge devices, which solve the problem of insufficient computing capabilities of the devices. Secondly, according to the requirements of different anomaly detection tasks, edge anomaly detection algorithms for sensor nodes and sink nodes are designed respectively. Lastly, an experimental platform is built for performance comparison analysis, and the experimental results show that the proposed algorithm has better performance in anomaly detection accuracy, delay, and energy consumption.

Steam++ An Extensible End-to-end Framework for Developing IoT Data Processing Applications in the Fog

“…In the sequence, we present a discussion of the main aspects of related works. [21] 2020 Framework ✓ ✓ ✓ PCA, R-PCA Liu et al [8] 2020 Algorithm ✓ ✓ Chi-Square Distance Greco et al [22] 2019 Architecture ✓ ✓ HTM, Node-RED, Flink, Kafka Liu et al [14] proposed an anomaly detection method using K-Means and C-Means over a sliding window, executed on a sink node on the network edge. They monitored multiple sensors in realtime inside an underground mine.…”

“…Most of the existing methods to process sensor data rely on cloud architecture and Stream Processing (SP) or Complex Event Processing (CEP) services, bringing some problems to industrial plants [12,13]. Sometimes, companies operating in remote places such as the countryside, offshore or underground do not have reliable and stable Internet access [14]. Usually, IIoT applications execute real-time analysis in product supply chain management, performance evaluation, and simulation [1].…”

“…Analysing the literature, we found challenges addressed to IoT, such as inaccurate data, lacking Internet access, and real-time applications [14]. However, a significant challenge is the development of infrastructure containing a common framework.…”

“…Most proposed frameworks cannot be reused for all types of data since they were designed specifically for a particular domain [1]. The studies presented in this paper approached anomaly detection in the network edge applied to a specific area, including underground mining [14], automotive assembly plant [18], water quality monitoring [19], and industrial machinery monitoring [8]. Authors used diverse techniques, such as K-means and C-means [14,18], Confidence Interval and Interval Variance [20], FFT over vibration and ANN [18], One-Class Classifier SVM, Isolation Forest and Elliptic Envelope [19], Principal Component Analysis (PCA) and R-PCA [21], Chi-Square Distance [8] and Hierarchical Temporal Memory [22].…”

Analysis of Machine Learning Algorithms for Anomaly Detection on Edge Devices

“…Edge anomaly detection for sensor networks appears in many research areas of the IoT in industrial solutions [ 20 ]. Many traditional clustering methods, such as K-means and C-means have been proposed for data analysis and prediction, but they are not directly useful for IoT applications in underground mining systems.…”