Hybridization of model-specific and model-agnostic methods for interpretability of Neural network predictions: Application to a power plant
Authors:
- Tina DaneshTDTina Danesh
- Rachid OuaretRORachid Ouaret
- Pascal FloquetPFPascal Floquet
- Stephane NegnySNStephane Negny
- 24 May 2023
- journal article
- research article
- Published by Elsevier in Computers & Chemical Engineering
- Vol. 176
Abstract
No abstract availableKeywords
This publication has 37 references indexed in Scilit:
- First and second order sensitivity analysis of MLPNeurocomputing, 2010
- Variance based sensitivity analysis of model output. Design and estimator for the total sensitivity indexComputer Physics Communications, 2010
- Simulation of thermodynamic systems using soft computing techniquesInternational Journal of Energy Research, 2005
- Influence of ambient temperature on combined-cycle power-plant performanceApplied Energy, 2005
- A review of process fault detection and diagnosisComputers & Chemical Engineering, 2003
- A review of process fault detection and diagnosisComputers & Chemical Engineering, 2003
- Statistical inference, the bootstrap, and neural-network modeling with application to foreign exchange ratesIEEE Transactions on Neural Networks, 2001
- Multilayer feedforward networks with a nonpolynomial activation function can approximate any functionNeural Networks, 1993
- Universal Approximation Using Radial-Basis-Function NetworksNeural Computation, 1991
- Approximation capabilities of multilayer feedforward networksNeural Networks, 1991