Explore chapters and articles related to this topic
Obtaining full-field response for optimal sensor placement
Published in Nigel Powers, Dan M. Frangopol, Riadh Al-Mahaidi, Colin Caprani, Maintenance, Safety, Risk, Management and Life-Cycle Performance of Bridges, 2018
Virtual sensing (VS) techniques can be used to estimate the quantities of interest using the available measurements. Virtual sensing can be either analytical (model-based) or empirical (data-driven) (Lin et al. 2007).
Virtual sensing network for statistical process monitoring
Published in IISE Transactions, 2023
Alexander Krall, Daniel Finke, Hui Yang
Virtual sensing entails the processing and transformation of nonlinear signals using a model or transfer function. This, in turn, enables a fine-grained examination into system dynamics and further extracts useful information for change detection in the undercurrents of nonlinear dynamical systems. Virtual Sensors (VSs) can be used alongside or in lieu of physical sensors to mitigate practical or analytical constraints in the real world. Nonetheless, the notion of virtual sensing is rather broad, because the scope of transformation modeling is large. In this investigation, we focus on virtual sensing within the context of placing sensors at different locations of signaling trajectories to monitor evolving dynamics within the signal space. In this regard, VS can be treated as imaginary sensors that sense the flux dynamics of signals, also referred to as virtual flux sensing.
Knowledge integration via the fusion of the data models used in automotive production systems
Published in Enterprise Information Systems, 2019
Rafal Cupek, Adam Ziebinski, Marek Drewniak, Marcin Fojcik
For example, data fusion can be used to create a learning model to synchronise the semantics from existing manufacturing knowledge graphs (Lamolle, Menet, and Le Duc 2015) and operational data (Ringsquandl et al. 2017). Multi-sensor data fusion (Mitchell 2007) permits the working status of the process and machinery to be acquired by integrating sensors into manufacturing systems. Virtual sensing is a non-invasive method, which is used to measure the parameters in dynamic systems, is based on computational models and is widely used to optimise an operation (Ploennigs et al. 2011) or product quality in industry (Huang et al. 2015). Artificial intelligence methods are also used to prepare virtual sensing models (Bustillo, Correa, and Reñones 2011). An example of multi-sensory fusion that is based on the virtual sensing method and artificial intelligence using the kernel principal component analysis model was described in (J. Wang et al. 2017). The high degree of the complexity of manufacturing systems provides much measurement data, which then affects many additional factors. This significantly inhibits the modelling of cause-effect relations. In these cases, the best results have been obtained using Supervised Machine Learning and Cluster Analysis (Wuest, Irgens, and Thoben 2014). In order to obtain better performance, some solutions are realised as embedded real-time systems (Rodriguez-Donate et al. 2010) that use Field Programmable Gate Arrays (Ziębiński and Świerc 2009).