Internal Combustion Engine sensor network analysis using graph modeling. (September 2017)

• Record Type: Journal Article
• Authors: Corsini, A.; Bonacina, F.; Feudo, S.; Marchegiani, A.; Venturini, P.
• Is Part Of: Energy procedia. Volume 126(2017)
• Page Start: 907
• Page End: 914
• Publication Date: 2017-09
• Subjects: Sensor Network -- Graph Modeling -- Big Data Analytics -- Complex Systems; Power resources -- Congresses; Power resources -- Periodicals; Power resources; Conference proceedings; Periodicals; 333.7905
• DOI: 10.1016/j.egypro.2017.08.160
• Languages: English
• ISSNs: 1876-6102

Abstract

Abstract: In recent years there has been a rapid development in technologies for smart monitoring applied to many different areas (e.g. building automation, photovoltaic systems, etc.). An intelligent monitoring system employs multiple sensors distributed within a network to extract useful information for decision-making. The management and the analysis of the raw data derived from the sensor network includes a number of specific challenges still unresolved, related to the different communication standards, the heterogeneous structure and the huge volume of data. In this paper we propose to apply a method based on complex network theory, to evaluate the performance of an Internal Combustion Engine. Data are gathered from the OBD sensor subset and from the emission analyzer. The method provides for the graph modeling of the sensor network, where the nodes are represented by the sensors and the edge are evaluated with non-linear statistical correlation functions applied to the time series pairs. The resulting functional graph is then analyzed with the topological metrics of the network, to define characteristic proprieties representing useful indicator for the maintenance and diagnosis.