Anomaly Detection for Hydraulic Power Units—A Case Study

Abstract

This article aims to present the real-world implementation of an anomaly detection system of a hydraulic power unit. Implementation involved the Internet of Things approach. A detailed description of the system architecture is provided. The complete path from sensors through PLC and the edge computer to the cloud is presented. Some technical information about hydraulic power units is also given. This article involves the description of several model-at-scale deployment techniques. In addition, the approach to the synthesis of anomaly and novelty detection models was described. Anomaly detection of data acquired from the hydraulic power unit was carried out using two approaches, statistical and black-box, involving the One Class SVM model. The costs of cloud resources and services that were generated in the project are presented. Since the article describes a commercial implementation, the results have been presented as far as the formal and business conditions allow.

Edge and Fog Computing for the Internet of Things

“…While some of them focus on specific aspects, some others are more devoted to the description of novel applications and innovative ways to face specific problems. This is the case of the first paper [1], which focuses specifically on the field of hydraulic power units. In this context, the scope of this paper is to describe a complete architecture from the sensors used to acquire a large set of parameters, to the edge computing tools, up to the cloud infrastructure.…”