Energy Conversion and Management 225 (2020) 113339 Available online 16 September 2020 0196-8904/© 2020 Elsevier Ltd. All rights reserved. Contents lists available at ScienceDirect Energy Conversion and Management journal homepage: www.elsevier.com/locate/enconman Recovery of energy losses using an online data-driven optimization technique Turaj Ashuri a,∗,1 , Yaoyu Li b,2 , Seyed Ehsan Hosseini c ,3 a Kennesaw State University, Southern Polytechnic College of Engineering and Engineering Technology, Marietta, GA 30060, USA b University of Texas at Dallas, Erik Jonsson School of Engineering and Computer Science, Richardson, TX 75080, USA c Arkansas Tech University, College of Engineering and Applied Sciences, Russellville, AR 72801, USA ARTICLE INFO Keywords: Data-driven optimization Energy loss recovery Annual energy production Levelized cost of energy Ultimate and fatigue structural loads Extremum seeking controls Hypothesis testing ABSTRACT This paper investigates the energy loss due to rotor aging for a wind turbine and its recovery. Energy loss for a wind turbine is caused by formation of dust and bugs, and corrosion and erosion of the surface of the rotor by sand and rain. We present a real-time data-driven optimization algorithm that uses dither and demodulation signals to recover the energy loss by extracting online the sensitivity of the function of interest to optimize. We use a viscid–inviscid model to represent aerodynamic performance loss of the rotor due to aging. We employ time-domain aeroservoelastic simulation of the CART3 wind turbine of the National Renewable Energy Laboratory to provide a comprehensive assessment of aging and its recovery. We also investigate the impact of aging on structural loads and levelized cost of energy. Using annual energy production and WindPACT cost models of the National Renewable Energy Laboratory, the levelized cost of energy enables an overall assessment of aging and its recovery. To evaluate the impact of the energy loss recovery on structural loading, ultimate and fatigue loads for power production design load case are used. The results of this study show 6.9% reduction in the annual energy production due to aging for a class C wind condition based on IEC61400 standard. This energy loss increases the levelized cost of energy by 7.5%. Our online optimization algorithm can recover 1.7% of the energy losses, and it results in 2.0% reduction in the levelized cost of energy. Results of the ultimate loads indicate that an aged rotor reduces structural loading on most components except the main shaft where the bending moment shows an increase of 5.3%. Rotor aging also reduces the fatigue loads in most components except the tower bottom fore-aft moment with an increase of 5.5%. The statistical inference of the results shows that the proposed optimization algorithm is effective in recovering aging related energy losses of wind turbines with a confidence level of 84% based on the sampled data in this study. 1. Introduction In 2018, renewable energy resources made up 26% percent of global electricity generation. In spite of progress in renewable energies uptake, the world is not meeting the targets of the Paris Agreement goals [1]. Wind energy is the most widely used resource among other renewable energies, but it still has higher cost of energy generation to compete economically with conventional energy resources [2–4]. Further research and development of the factors that negatively impact the economy of wind energy is needed [5]. Energy loss is among the cost increasing factors in a wind turbine that happens by formation of dust and bugs, and corrosion and erosion of the surface of the rotor by sand and rain [6]. This mechanism is referred to as aging, and it increases the levelized cost of energy, as well as the alteration of the ∗ Corresponding author. E-mail address: tashuri@kennesaw.edu (T. Ashuri). 1 Associate Professor. 2 Professor. 3 Assistant Professor. structural loads. Therefore, development of innovative solutions such as advanced control algorithms is crucial to mitigate energy losses. Control algorithms that rely on a model are not capable to cope with this problem since they have frozen optimized parameters for the initial configuration of the system [7–10]. Control algorithms that are model- free can address partially the issue by adapting the system operation for optimal performance where the optimal input parameters vary over time. Extremum seeking controls (ESC) is such a model-free adaptive control algorithm that is capable of extracting online the sensitivity of the input–output function of interest using demodulation and dither signals, and a pair of high pass and low pass filters [11–14]. This sensitivity is then used to optimize a performance index with respect to a single or multiple control variables. https://doi.org/10.1016/j.enconman.2020.113339 Received 5 April 2020; Received in revised form 11 August 2020; Accepted 12 August 2020