Jinkyoo Park

Electromagnetic energy harvester with repulsively stacked multilayer magnets for low frequency vibrations

This paper investigates the applicability of an electromagnetic generator with repulsively stacked magnets for harvesting energy from traffic-induced bridge vibrations. First, the governing equation for electro-mechanical coupling is presented. The magnetic field for repulsive pole arrangements is discussed and the model is validated from a magnet falling test. The detailed design, fabrication, and test results of a prototype device are presented in the paper. An experimental vibration shaker test is conducted to assess the performance of the energy harvester. Field test and numerical simulation at the 3rd Nongro Bridge in South Korea shows that the device can generate an average power of 0.12 mW from an input rms acceleration of 0.25 m s−2 at 4.10 Hz. With further frequency tuning and design improvement, an average power of 0.98 mW could be potentially harvested from the ambient vibration of the bridge.

Wind farm power maximization based on a cooperative static game approach

The objective of this study is to improve the cost-effectiveness and production efficiency of wind farms using cooperative control. The key factors in determining the power production and the loading for a wind turbine are the nacelle yaw and blade pitch angles. However, the nacelle and blade angles may adjust the wake direction and intensity in a way that may adversely affect the performance of other wind turbines in the wind farm. Conventional wind-turbine control methods maximize the power production of a single turbine, but can lower the overall wind-farm power efficiency due to wake interference. This paper introduces a cooperative game concept to derive the power production of individual wind turbine so that the total wind-farm power efficiency is optimized. Based on a wake interaction model relating the yaw offset angles and the induction factors of wind turbines to the wind speeds experienced by the wind turbines, an optimization problem is formulated with the objective of maximizing the sum of the power production of a wind farm. A steepest descent algorithm is applied to find the optimal combination of yaw offset angles and the induction factors that increases the total wind farm power production. Numerical simulations show that the cooperative control strategy can increase the power productions in a wind farm.

An intelligent machine monitoring system for energy prediction using a Gaussian Process regression

Recent advances in machine automation and sensing technology offer new opportunities for continuous condition monitoring of an operating machine. This paper describes an intelligent machine monitoring framework that integrates and utilizes data collection, management, and analytics to derive an adaptive predictive model for the energy usage of a milling machine. This model is designed using a Gaussian Process (GP) regression algorithm, which is a flexible regression method that also provides an uncertainty estimate. To improve computational efficiency, we propose a Collective Gaussian Process (CGP) in which the overall energy prediction is made by constructing local GP models weighted by probability distribution functions obtained using the Gaussian Mixture Model (GMM) technique. Finally, we demonstrate the ability of the proposed monitoring framework to construct an energy prediction model to predict the energy used to machine a part.

A Generalized Data-Driven Energy Prediction Model with Uncertainty for a Milling Machine Tool using Gaussian Process

Using a machine learning approach, this study investigates the effects of machining parameters on the energy consumption of a milling machine tool, which would allow selection of optimal operational strategies to machine a part with minimum energy. Data-driven prediction models, built upon a nonlinear regression approach, can be used to gain an understanding of the effects of machining parameters on energy consumption. In this study, we use the Gaussian Process to construct the energy prediction model for a computer numerical control (CNC) milling machine tool. Energy prediction models for different machining operations are constructed based on collected data. With the collected data sets, optimum input features for model selection are identified. We demonstrate how the energy prediction models can be used to compare the energy consumption for the different operations and to estimate the total energy usage for machining a generic part. We also present an uncertainty analysis to develop confidence bounds for the prediction model and to provide insight into the vast parameter space and training required to improve the accuracy of the model. Generic parts are machined to test and validate the prediction model constructed using the Gaussian Process and we consistently achieve an accuracy of over 95 % on the total predicted energy.Copyright

Power evaluation of flutter-based electromagnetic energy harvesters using computational fluid dynamics simulations

H- and T-shaped cross sections are known to be susceptible to rotational single-degree-of-freedom aerodynamic instabilities. Here, such self-excited aerodynamic response of a T-shaped cantilever structure is used to extract energy, which is then converted into electric power through an electromagnetic transducer. The complex fluid–structure interaction between the cantilever harvester and wind flow is analyzed numerically and experimentally. To study the dynamic response of the cantilever and estimate the power output from the harvester, numerical simulations based on the vortex particle method are performed to determine the aerodynamic damping of the harvester section and to analyze the stability behavior of the section. The estimated aerodynamic damping parameter together with the mechanical and electrical damping parameters in the harvester are used to find the critical wind speed of flutter onset as well as the optimum load resistance. Wind tunnel experiments are conducted to validate the simulation results.

Towards a generalized energy prediction model for machine tools

Energy prediction of machine tools can deliver many advantages to a manufacturing enterprise, ranging from energy-efficient process planning to machine tool monitoring. Physics-based, energy prediction models have been proposed in the past to understand the energy usage pattern of a machine tool. However, uncertainties in both the machine and the operating environment make it difficult to predict the energy consumption of the target machine reliably. Taking advantage of the opportunity to collect extensive, contextual, energy-consumption data, we discuss a data-driven approach to develop an energy prediction model of a machine tool in this paper. First, we present a methodology that can efficiently and effectively collect and process data extracted from a machine tool and its sensors. We then present a data-driven model that can be used to predict the energy consumption of the machine tool for machining a generic part. Specifically, we use Gaussian Process (GP) Regression, a non-parametric machine-learning technique, to develop the prediction model. The energy prediction model is then generalized over multiple process parameters and operations. Finally, we apply this generalized model with a method to assess uncertainty intervals to predict the energy consumed to machine any part using a Mori Seiki NVD1500 machine tool. Furthermore, the same model can be used during process planning to optimize the energy-efficiency of a machining process.

Bayesian Ascent: A Data-Driven Optimization Scheme for Real-Time Control With Application to Wind Farm Power Maximization

This paper describes a data-driven approach for real-time control of a physical system. Specifically, this paper focuses on the cooperative wind farm control where the objective is to maximize the total wind farm power production by using control actions as an input and measured power as an output. For real time, data-driven wind farm control, it is imperative that the optimization algorithm is able to improve a target wind farm power production by executing as small number of trial actions as possible using the wind farm power monitoring data. To achieve this goal, we develop a Bayesian ascent (BA) algorithm by incorporating into the Bayesian optimization framework a strategy that regulates the search domain, as used in the trust region method. The BA algorithm is composed of two iterative phases, namely, learning and optimization phases. In the learning phase, the BA algorithm approximates the target function using Gaussian process regression to fit the measured input and output of the target system. In the optimization phase, the BA algorithm determines the next sampling point to learn more about the target function (exploration) as well as to improve the target value (exploitation). Specifically, the sampling strategy is designed to ensure that the input is selected within a trust region to improve the target value monotonically by gradually changing the input for a target system. The results from simulation studies using an analytical wind farm power function and experimental studies using scaled wind turbines show that the BA algorithm can achieve an almost monotonic increase in the target value.

A Bayesian optimization approach for wind farm power maximization

The objective of this study is to develop a model-free optimization algorithm to improve the total wind farm power production in a cooperative game framework. Conventionally, for a given wind condition, an individual wind turbine maximizes its own power production without taking into consideration the conditions of other wind turbines. Under this greedy control strategy, the wake formed by the upstream wind turbine, due to the reduced wind speed and the increased turbulence intensity inside the wake, would affect and lower the power productions of the downstream wind turbines. To increase the overall wind farm power production, researchers have proposed cooperative wind turbine control approaches to coordinate the actions that mitigate the wake interference among the wind turbines and thus increase the total wind farm power production. This study explores the use of a data-driven optimization approach to identify the optimum coordinated control actions in real time using limited amount of data. Specifically, we propose the Bayesian Ascent (BA) method that combines the strengths of Bayesian optimization and trust region optimization algorithms. Using Gaussian Process regression, BA requires only a few number of data points to model the complex target system. Furthermore, due to the use of trust region constraint on sampling procedure, BA tends to increase the target value and converge toward near the optimum. Simulation studies using analytical functions show that the BA method can achieve an almost monotone increase in a target value with rapid convergence. BA is also implemented and tested in a laboratory setting to maximize the total power using two scaled wind turbine models.

Real-time energy prediction for a milling machine tool using sparse Gaussian process regression

This paper describes a real-time data collection framework and an adaptive machining learning method for constructing a real-time energy prediction model for a machine tool. To effectively establish the energy consumption pattern of a machine tool over time, the energy prediction model is continuously updated with new measurement data to account for time-varying effects of the machine tool, such as tool wear and machine tool deterioration. In this work, a real-time data collection and processing framework is developed to retrieve raw data from a milling machine tool and its sensors and convert them into relevant input features. The extracted input features are then used to construct the energy prediction model using Gaussian Process (GP) regression. To update the GP regression model with real-time streaming data, we investigate the use of sparse representation of the covariance matrix to reduce the computational and storage demands of the GP regression. We compare computational efficiency of sparse GP to that of full GP regression model and show the effectiveness of the sparse GP regression model for tracking the variation in the energy consumption pattern of the target machine.

Multivariate Analysis and Prediction of Wind Turbine Response to Varying Wind Field Characteristics Based on Machine Learning

Site-specific wind field characteristics have a significant impact on the structural response and the lifespan of wind turbines. This paper presents a machine learning approach towards analyzing and predicting the response of wind turbine structures to varying wind field characteristics. Machine learning algorithms are applied (i) to better understand changes of wind field characteristics due to atmospheric conditions and (ii) to gain new insights into the wind turbine loads being affected by fluctuating wind. Using Gaussian Mixture Models, the variations in wind fields are investigated by comparing the joint probability distribution functions of several wind field features, which are constructed from long-term monitoring data taken from a 500 kW wind turbine in Germany, which is used as a reference system. Furthermore, based on Gaussian Discriminative Analysis, representative daytime and nocturnal wind turbine loads are predicted, compared, and analyzed.