A methodology to improve soft sensors performances in spatial forecast of environmental parameters is introduced. To this aim, we substitute a single soft sensor based on a single neural network with a mixture of several soft sensors. This mixture is formed by different neural network models and a selector. The performance of this new architecture has been statistically analyzed from a metrological point of view. In comparison with traditional soft sensor approach, this model gives better results for each evaluator and in any measurement condition. We call this architecture HyperSensor. The HyperSensor wraps the best characteristics of different neural network models through a gating network, which selects the best performing soft sensor according to the current input
Multi Soft-Sensors Data Fusion In Spatial Forecasting Of Environmental Parameters
U Maniscalco;G Pilato
2012
Abstract
A methodology to improve soft sensors performances in spatial forecast of environmental parameters is introduced. To this aim, we substitute a single soft sensor based on a single neural network with a mixture of several soft sensors. This mixture is formed by different neural network models and a selector. The performance of this new architecture has been statistically analyzed from a metrological point of view. In comparison with traditional soft sensor approach, this model gives better results for each evaluator and in any measurement condition. We call this architecture HyperSensor. The HyperSensor wraps the best characteristics of different neural network models through a gating network, which selects the best performing soft sensor according to the current inputI documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.