Piezoelectric Energy Harvester Response Statistics

Abstract

Safety and reliability are essential engineering concerns for energy-harvesting installations. In the case of the piezoelectric galloping energy harvester, there is a risk that excessive wake galloping may lead to instability, overload, and thus damage. With this in mind, this paper studies bivariate statistics of the extreme, experimental galloping energy harvester dynamic response under realistic environmental conditions. The bivariate statistics were extracted from experimental wind tunnel results, specifically for the voltage-force data set. Authors advocate a novel general-purpose reliability approach that may be applied to a wide range of dynamic systems, including micro-machines. Both experimental and numerically simulated dynamic responses can be used as input for the suggested structural reliability analysis. The statistical analysis proposed in this study may be used at the design stage, supplying proper characteristic values and safeguarding the dynamic system from overload, thus extending the machine’s lifetime. This work introduces a novel bivariate technique for reliability analysis instead of the more general univariate design approaches.

Offloading operation bivariate extreme response statistics for FPSO vessel

“…In this study, authors employed a general Monte Carlo (MC) based technique, able to tackle inherent nonlinear effects without significant simplifications except those inherent in the numerical model itself. The target bivariate extreme response distribution was obtained by combining MC simulations with the ACER2D extrapolation method [12][13][14][15][16] . Unlike univariate statistical techniques, the ACER2D method accounts for bivariate response correlation.…”

Improving extreme offshore wind speed prediction by using deconvolution

“…The countless strategies endeavoured included the method presented by Mohande [ 26 ], which predicts wind speeds using neural networks or the different approaches to analyse wind speed estimates by Yan et al, as in Refs. [ [1] , [2] , [3] , [4] , [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] ]. Similarly, the authors of this paper have also previously used different statistical approaches to estimate better extreme statistical values such as, i.e.…”

“…Similarly, the authors of this paper have also previously used different statistical approaches to estimate better extreme statistical values such as, i.e. wind speed, wave height, response and load [ [1] , [2] , [3] , [4] , [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] ]. Furthermore, the authors developed a novel reliability approach to forecast COVID-19 epidemic levels, indicating the approach's versatility and reliability in various fields [ 14 , 16 , 26 ].…”

“…Similarly, other authors, as in Refs. [ [1] , [2] , [3] , [4] , [5] , [6] , [7] , [8] , [9] , [10] , [11] , [12] , [13] , [14] , [15] , [16] , [17] , [18] , [19] , [20] , [21] , [22] , [23] , [24] , [25] , [26] ], have attempted to use novel statistical approaches to estimate wind speed or pattern more accurately. Meanwhile, two recent and pertinent scientific articles [ 8 , 26 ] illustrated the importance of wind forecasting and strategies adopted to maximise efficiency.…”

“…The convolution technique also does not assume any specific extrapolation functional class. This is in contrast to many traditional extrapolation methods widely used in engineering practice where an assumed probability function is used [ [1] , [2] , [3] , [4] , [5] , [6] , 30 ]. Some other examples also include the Pareto-based distribution peak over the threshold (POT) [ 1 ], The Naess-Gaidai (NG) method fitting procedure was used in averaged conditional exceedance rate (ACER) method [ [7] , [8] , [9] , [10] , [11] ], bivariate ACER fitting method [ [12] , [13] , [14] , [15] , [16] , 47 ], Weibull distribution-based fitting method [ 38 ] and Gumbel distribution based fitting method [ 17 ].…”