scholarly journals Software-in-the-Loop Combined Reinforcement Learning Method for Dynamic Response Analysis of FOWTs

2021 ◽  
Vol 7 ◽  
Author(s):  
Peng Chen ◽  
Jiahao Chen ◽  
Zhiqiang Hu
Keyword(s):  
Machine Learning ◽  
Policy Decision ◽  
Maximum Error ◽  
Machine Learning Algorithms ◽  
Dynamic Responses ◽  
Offshore Wind ◽  
Response Analysis ◽  
Learning Method ◽  
Mean Values ◽  
Floating Offshore Wind Turbines

Floating offshore wind turbines (FOWTs) still face many challenges on how to better predict the dynamic responses. Artificial intelligence (AI) brings a new solution to overcome these challenges with intelligent strategies. A new AI technology-based method, named SADA, is proposed in this paper for the prediction of dynamic responses of FOWTs. Firstly, the methodology of SADA is introduced with the selection of Key Disciplinary Parameters (KDPs). The AI module in SADA was built in a coupled aero-hydro-servo-elastic in-house program DARwind and the policy decision is provided by the machine learning algorithms deep deterministic policy gradient (DDPG). Secondly, a set of basin experimental results of a Hywind Spar-type FOWT were employed to train the AI module. SADA weights KDPs by DDPG algorithms' actor network and changes their values according to the training feedback of 6DOF motions of Hywind platform through comparing the DARwind simulation results and that of experimental data. Many other dynamic responses that cannot be measured in basin experiment could be predicted in higher accuracy with this intelligent DARwind. Finally, the case study of SADA method was conducted and the results demonstrated that the mean values of the platform's motions can be predicted by AI-based DARwind with higher accuracy, for example the maximum error of surge motion is reduced by 21%. This proposed SADA method takes advantage of numerical-experimental method and the machine learning method, which brings a new and promising solution for overcoming the handicap impeding direct use of traditional basin experimental technology in FOWTs design.

Software-in-the-Loop Combined Machine Learning for Dynamic Responses Analysis of Floating Offshore Wind Turbines

2021 ◽  
Author(s):  
Peng Chen ◽  
Changhong Hu ◽  
Zhiqiang Hu
Keyword(s):  
Machine Learning ◽  
Experimental Data ◽  
Wind Turbines ◽  
Dynamic Responses ◽  
Offshore Wind ◽  
Learning Technology ◽  
Mean Values ◽  
Offshore Wind Turbines ◽  
Floating Offshore Wind Turbines ◽  
Brute Force Method

Abstract Artificial intelligence (AI) brings a new solution to overcome the challenges of Floating offshore wind turbines (FOWTs) to better predict the dynamic responses with intelligent strategies. A new AI-based software-in-the-loop method, named SADA is introduced in this paper for the prediction of dynamic responses of FOWTs, which is proposed based on an in-house programme DARwind. DARwind is a coupled aero-hydro-servo-elastic in-house program for FOWTs, and a reinforcement learning method with exhaust algorithm and deep deterministic policy gradient (DDPG) are embedded in DARwind as an AI module. Firstly, the methodology is introduced with the selection of Key Disciplinary Parameters (KDPs). Secondly, Brute-force Method and DDPG algorithms are adopted to changes the KDPs’ values according to the feedback of 6DOF motions of Hywind Spar-type platform through comparing the DARwind simulation results and those of basin experimental data. Therefore, many other dynamic responses that cannot be measured in basin experiment can be predicted in good accuracy with SADA method. Finally, the case study of SADA method was conducted and the results demonstrated that the mean values of the platform’s motions can be predicted with higher accuracy. This proposed SADA method takes advantage of numerical-experimental method, basin experimental data and the machine learning technology, which brings a new and promising solution for overcoming the handicap impeding direct use of conventional basin experimental way to analyze FOWT’s dynamic responses during the design phase.

scholarly journals Digital Smoke Taint Detection in Pinot Grigio Wines Using an E-Nose and Machine Learning Algorithms Following Treatment with Activated Carbon and a Cleaving Enzyme

Fermentation ◽  
2021 ◽  
Vol 7 (3) ◽  
pp. 119
Author(s):  
Vasiliki Summerson ◽  
Claudia Gonzalez Viejo ◽  
Damir D. Torrico ◽  
Alexis Pang ◽  
Sigfredo Fuentes
Keyword(s):  
Machine Learning ◽  
Activated Carbon ◽  
Low Cost ◽  
Aroma Compounds ◽  
Machine Learning Algorithms ◽  
Enzyme Treatment ◽  
Mean Values ◽  
Volatile Aroma ◽  
Carbon Treatment ◽  
Volatile Aroma Compounds

The incidence and intensity of bushfires is increasing due to climate change, resulting in a greater risk of smoke taint development in wine. In this study, smoke-tainted and non-smoke-tainted wines were subjected to treatments using activated carbon with/without the addition of a cleaving enzyme treatment to hydrolyze glycoconjugates. Chemical measurements and volatile aroma compounds were assessed for each treatment, with the two smoke taint amelioration treatments exhibiting lower mean values for volatile aroma compounds exhibiting positive ‘fruit’ aromas. Furthermore, a low-cost electronic nose (e-nose) was used to assess the wines. A machine learning model based on artificial neural networks (ANN) was developed using the e-nose outputs from the unsmoked control wine, unsmoked wine with activated carbon treatment, unsmoked wine with a cleaving enzyme plus activated carbon treatment, and smoke-tainted control wine samples as inputs to classify the wines according to the smoke taint amelioration treatment. The model displayed a high overall accuracy of 98% in classifying the e-nose readings, illustrating it may be a rapid, cost-effective tool for winemakers to assess the effectiveness of smoke taint amelioration treatment by activated carbon with/without the use of a cleaving enzyme. Furthermore, the use of a cleaving enzyme coupled with activated carbon was found to be effective in ameliorating smoke taint in wine and may help delay the resurgence of smoke aromas in wine following the aging and hydrolysis of glycoconjugates.

Dynamic Responses of a Spar Type Floating Offshore Wind Turbine With Failed Moorings

2020 ◽  
Author(s):  
Yajun Ren ◽  
Vengatesan Venugopal
Keyword(s):  
Wind Turbine ◽  
Wind Turbines ◽  
Dynamic Responses ◽  
Offshore Wind ◽  
Offshore Wind Turbine ◽  
Mooring System ◽  
Floating Offshore Wind Turbine ◽  
Offshore Wind Turbines ◽  
Floating Offshore Wind Turbines ◽  
Platform Motions

Abstract The complex dynamic characteristics of Floating Offshore Wind Turbines (FOWTs) have raised wider consideration, as they are likely to experience harsher environments and higher instabilities than the bottom fixed offshore wind turbines. Safer design of a mooring system is critical for floating offshore wind turbine structures for station keeping. Failure of mooring lines may lead to further destruction, such as significant changes to the platform’s location and possible collisions with a neighbouring platform and eventually complete loss of the turbine structure may occur. The present study focuses on the dynamic responses of the National Renewable Energy Laboratory (NREL)’s OC3-Hywind spar type floating platform with a NREL offshore 5-MW baseline wind turbine under failed mooring conditions using the fully coupled numerical simulation tool FAST. The platform motions in surge, heave and pitch under multiple scenarios are calculated in time-domain. The results describing the FOWT motions in the form of response amplitude operators (RAOs) and spectral densities are presented and discussed in detail. The results indicate that the loss of the mooring system firstly leads to longdistance drift and changes in platform motions. The natural frequencies and the energy contents of the platform motion, the RAOs of the floating structures are affected by the mooring failure to different degrees.

scholarly journals A New Conceptual Design and Dynamic Analysis of a Spar-Type Offshore Wind Turbine Combined with a Moonpool

Energies ◽  
2019 ◽  
Vol 12 (19) ◽  
pp. 3737 ◽  
Author(s):  
Thanh Dam Pham ◽  
Hyunkyoung Shin
Keyword(s):  
Wind Turbine ◽  
Bending Moment ◽  
Time Domain Analysis ◽  
Domain Analysis ◽  
Frequency Domain Analysis ◽  
Dynamic Responses ◽  
Offshore Wind ◽  
Offshore Wind Turbine ◽  
Operational Conditions ◽  
Floating Offshore Wind Turbines

Floating offshore wind turbines promise to provide an abundant source of energy. Currently, much attention is being paid to the efficient performance and the economics of floating wind systems. This paper aims to develop a spar-type platform to support a 5-MW reference wind turbine at a water depth of 150 m. The spar-type platform includes a moonpool at the center. The design optimization process is composed of three steps; the first step uses a spreadsheet to calculate the platform dimensions; the second step is a frequency domain analysis of the responses in wave conditions; and the final step is a fully coupled simulation time domain analysis to obtain the dynamic responses in combined wind, wave, and current conditions. By having a water column inside the open moonpool, the system’s dynamic responses to horizontal and rotating motions are significantly reduced. Reduction of these motions leads to a reduction in the nacelle acceleration and tower base bending moment. On the basic of optimization processes, a spar-type platform combined with a moonpool is suggested, which has good performance in both operational conditions and extreme conditions.

scholarly journals Review of Experimental-Numerical Methodologies and Challenges for Floating Offshore Wind Turbines

2020 ◽  
Vol 19 (3) ◽  
pp. 339-361
Author(s):  
Peng Chen ◽  
Jiahao Chen ◽  
Zhiqiang Hu
Keyword(s):  
Control Strategies ◽  
Full Scale ◽  
Dynamic Responses ◽  
Offshore Wind ◽  
Structural Dynamic ◽  
Hybrid Approaches ◽  
Global Dynamic ◽  
Floating Offshore Wind Turbines ◽  
Calibration Methods ◽  
Industrial Projects

Abstract Due to the dissimilar scaling issues, the conventional experimental method of FOWTs can hardly be used directly to validate the full-scale global dynamic responses accurately. Therefore, it is of absolute necessity to find a more accurate, economic and efficient approach, which can be utilized to predict the full-scale global dynamic responses of FOWTs. In this paper, a literature review of experimental-numerical methodologies and challenges for FOWTs is made. Several key challenges in the conventional basin experiment issues are discussed, including scaling issues; coupling effects between aero-hydro and structural dynamic responses; blade pitch control strategies; experimental facilities and calibration methods. Several basin experiments, industrial projects and numerical codes are summarized to demonstrate the progress of hybrid experimental methods. Besides, time delay in hardware-in-the-loop challenges is concluded to emphasize their significant role in real-time hybrid approaches. It is of great use to comprehend these methodologies and challenges, which can help some future researchers to make a footstone for proposing a more efficient and functional hybrid basin experimental and numerical method.

scholarly journals COVID-19 effect on supply and demand of essential commodities using unsupervised learning method

2020 ◽  
Author(s):  
P. Anitha ◽  
Malini M. Patil ◽  
Rekha B Venkatapur
Keyword(s):  
Machine Learning ◽  
Supply Chain ◽  
Clustering Algorithm ◽  
Supply And Demand ◽  
Machine Learning Algorithms ◽  
Learning Method ◽  
The People ◽  
Future Data ◽  
Predicted Values ◽  
Yearly Data

Abstract The affliction caused by the Covid-19 Pandemic is diverse from other disasters seen so far. Supply chain industries are facing unique challenges in fulfilling the essential needs of the people. The objective of the paper is to analyse the supply and demand of essentials during pre-pandemic and post-pandemic lockdowns using machine learning algorithms. This helps for supply chain industries in forecasting and managing the supply and demand of essential stocks for the future. Data is analyzed using prediction algorithms to check the actual and predicted values. The clustering algorithm along with rolling mean is used for half-yearly data of 2019 and 2020 to identify the sales of different categories of essential commodities. This paper aims at applying intelligence in predicting various categories of sales by providing timely information for B2B Industries during the time of disasters.

Response Analysis of a Nonstationary Lowering Operation for an Offshore Wind Turbine Monopile Substructure

2015 ◽  
Vol 137 (5) ◽  
Author(s):  
Lin Li ◽  
Zhen Gao ◽  
Torgeir Moan
Keyword(s):  
Wind Turbine ◽  
Time Domain ◽  
Dynamic Responses ◽  
Offshore Wind ◽  
Irregular Waves ◽  
Offshore Wind Turbine ◽  
Response Analysis ◽  
Steady State Condition ◽  
Strip Theory ◽  
Lifting System

This study addresses numerical modeling and time-domain simulations of the lowering operation for installation of an offshore wind turbine monopile (MP) with a diameter of 5.7 m and examines the nonstationary dynamic responses of the lifting system in irregular waves. Due to the time-varying properties of the system and the resulting nonstationary dynamic responses, numerical simulation of the entire lowering process is challenging to model. For slender structures, strip theory is usually applied to calculate the excitation forces based on Morison's formula with changing draft. However, this method neglects the potential damping of the structure and may overestimate the responses even in relatively long waves. Correct damping is particularly important for the resonance motions of the lifting system. On the other hand, although the traditional panel method takes care of the diffraction and radiation, it is based on steady-state condition and is not valid in the nonstationary situation, as in this case in which the monopile is lowered continuously. Therefore, this paper has two objectives. The first objective is to examine the importance of the diffraction and radiation of the monopile in the current lifting model. The second objective is to develop a new approach to address this behavior more accurately. Based on the strip theory and Morison's formula, the proposed method accounts for the radiation damping of the structure during the lowering process in the time-domain. Comparative studies between different methods are presented, and the differences in response using two types of installation vessel in the numerical model are also investigated.

Operational Data to Maintenance Optimization: Closing the Loop in Offshore Wind O&M

2018 ◽  
Author(s):  
Alexios Koltsidopoulos Papatzimos ◽  
Tariq Dawood ◽  
Philipp R. Thies
Keyword(s):  
Machine Learning ◽  
Wind Farm ◽  
Service Providers ◽  
Learning Algorithms ◽  
Machine Learning Algorithms ◽  
Offshore Wind ◽  
Sensor Data ◽  
Support Vector ◽  
Offshore Wind Turbine ◽  
Offshore Wind Farm

Offshore wind assets have reached multi-GW scale and additional capacity is being installed and developed. To achieve demanding cost of energy targets, awarded by competitive auctions, the operation and maintenance (O&M) of these assets has to become increasingly efficient, whilst ensuring compliance and effectiveness. Existing offshore wind farm assets generate a significant amount of inhomogeneous data related to O&M processes. These data contain rich information about the condition of the assets, which is rarely fully utilized by the operators and service providers. Academic and industrial research and development efforts have led to a suite of tools trying to apply sensor data and build machine learning models to diagnose, trend and predict component failures. This study presents a decision support framework incorporating a range of different supervised and un-supervised learning algorithms. The aim is to provide guidance for asset owners on how to select the most relevant datasets, apply and choose the different machine learning algorithms and how to integrate the data stream with daily maintenance procedures. The presented methodology is tested on a real case example of an offshore wind turbine gearbox replacement at Teesside offshore wind farm. The study uses k-nearest neighbour (kNN) and support vector machine (SVM) algorithms to detect the fault using supervisory control and data acquisition (SCADA) data and an autoregressive model for the vibration data of the condition monitoring system (CMS). The implementation of all the algorithms has resulted in an accuracy higher than 94%. The results of this paper will be of interest to offshore wind farm developers and operators to streamline and optimize their O&M planning activities for their assets and reduce the associated costs.

Dynamic Response Analysis of a Floating Offshore Wind Turbine During Severe Typhoon Event

2013 ◽  
Author(s):  
Tomoaki Utsunomiya ◽  
Iku Sato ◽  
Shigeo Yoshida ◽  
Hiroshi Ookubo ◽  
Shigesuke Ishida
Keyword(s):  
Dynamic Response ◽  
Wind Turbine ◽  
Dynamic Responses ◽  
Offshore Wind ◽  
Offshore Wind Turbine ◽  
Response Analysis ◽  
Dynamic Response Analysis ◽  
Floating Offshore Wind Turbine ◽  
Demonstration Experiment ◽  
Typhoon Event

In this paper, dynamic response analysis of a Floating Offshore Wind Turbine (FOWT) with Spar-type floating foundation is presented. The FOWT mounts a 100kW down-wind turbine, and is grid-connected. It was launched at sea on 9th June 2012, and moored on 11th for the purpose of the demonstration experiment. During the experiment, the FOWT was attacked by severe typhoon events twice. Among them, Sanba (international designation: 1216) was the strongest tropical cyclone worldwide in 2012. The central atmospheric pressure was 940 hPa when it was close to the FOWT, and the maximum significant wave height of 9.5m was recorded at the site. In this paper, the dynamic responses of the platform motion, the stresses at the tower sections and the chain tensions during the typhoon event, Sanba (1216), have been analyzed, and compared with the measured data. Through the comparison, validation of the numerical simulation tool (Adams with SparDyn developed by the authors) has been made.

scholarly journals Application and Comparison of Machine Learning Algorithms for Predicting Mass Loss of Cement Raw Materials due to Decarbonation Process

2020 ◽  
Vol 34 (4) ◽  
pp. 403-411
Author(s):  
Yakoub Boukhari
Keyword(s):  
Machine Learning ◽  
Mass Loss ◽  
Raw Materials ◽  
Learning Algorithms ◽  
Maximum Error ◽  
Machine Learning Algorithms ◽  
Raw Material ◽  
Material Particle ◽  
Individual Use ◽  
The Individual

The purpose of this work is to predict the mass loss of cement raw materials during the decarbonation process. The mass loss is influenced by the interaction of several parameters such as chemical composition of raw material, particle size, temperature range of decarbonation and time exposed. Therefore, predicting mass loss based on experimental parameters data is often challenging. For this reason, various machine learning algorithms such as deep networks using autoencoder DN-AE, artificial neural networks optimized by particle swarm optimization PSO-ANN, ANN optimized by ant colony optimization ACO-ANN and ANN are proposed to predict the mass loss. In this research, all models have been applied successfully to predict the mass loss with high accuracy. The results obtained have shown the superiority of DN-AE compared to PSO-ANN, ACO-ANN and ANN. In addition, PSO-ANN and ACO-ANN have a better performance than the individual use of ANN. The values of adjusted R2 indicate that 99.11%, 98.66%, 98.27% and 97.03% of data are explained by DN-AE, ACO-ANN, PSO-ANN and ANN respectively with scatter index (SI) less than 0.1 and maximum error less than 3.32%. Finally, the results justify that all models proposed can be employed to predict the mass loss as alternative tools.

Sign in / Sign up

Export Citation Format

Share Document