Model-based multi-objective optimization is a useful tool to compute optimal trade-offs between multiple conflicting objectives such as minimizing energy consumption while maximizing productivity. However, the computational cost of solving problems high. In addition, model approximates real process, meaning uncertainty inherently present. To avoid erroneous predictions process performance and u...

This work introduces the concept of parametric Gaussian processes (PGPs), which is built upon the seemingly self-contradictory idea of making Gaussian processes parametric. Parametric Gaussian processes, by construction, are designed to operate in “big data” regimes where one is interested in quantifying the uncertainty associated with noisy data. The proposed methodology circumvents the welles...

Variable speed wind turbines maximize the energy capture by operating the turbine at the peak of the power coefficient, however parametric uncertainties and disturbances may limit the efficiency of a variable speed turbine. In this study, we present a robust backstepping approach for the variable speed control of wind turbines. Specifically, to overcome the undesirable effects of parametric unc...

The literature on non-parametric production analysis has formulated tests for profit maximizing behavior that do not require a parametric specification of technology. Negative test results have conventionally been interpreted as inefficiency, or have been attributed to data perturbations. In this paper, we exploit the possibility that negative test results reveal violations of the underlying ne...

Monitoring machine learning models once they are deployed is challenging. It even more challenging to decide when retrain in real-case scenarios labeled data beyond reach, and monitoring performance metrics becomes unfeasible. In this work, we use non-parametric bootstrapped uncertainty estimates SHAP values provide explainable estimation as a technique that aims monitor the deterioration of de...