scholarly journals A Data-Mining Approach for Wind Turbine Fault Detection Based on SCADA Data Analysis Using Artificial Neural Networks

Energies ◽  
2021 ◽  
Vol 14 (7) ◽  
pp. 1845
Author(s):  
Annalisa Santolamazza ◽  
Daniele Dadi ◽  
Vito Introna
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Wind Turbine ◽  
Wind Turbines ◽  
Wind Farm ◽  
Machine Learning Techniques ◽  
Real Time Control ◽  
Time Control ◽  
Artificial Neural ◽  
Using Data

Wind energy has shown significant growth in terms of installed power in the last decade. However, one of the most critical problems for a wind farm is represented by Operation and Maintenance (O&M) costs, which can represent 20–30% of the total costs related to power generation. Various monitoring methodologies targeted to the identification of faults, such as vibration analysis or analysis of oils, are often used. However, they have the main disadvantage of involving additional costs as they usually entail the installation of other sensors to provide real-time control of the system. In this paper, we propose a methodology based on machine learning techniques using data from SCADA systems (Supervisory Control and Data Acquisition). Since these systems are generally already implemented on most wind turbines, they provide a large amount of data without requiring extra sensors. In particular, we developed models using Artificial Neural Networks (ANN) to characterize the behavior of some of the main components of the wind turbine, such as gearbox and generator, and predict operating anomalies. The proposed method is tested on real wind turbines in Italy to verify its effectiveness and applicability, and it was demonstrated to be able to provide significant help for the maintenance of a wind farm.

Control of a robot manipulator and pendubot system using artificial neural networks

Robotica ◽  
2005 ◽  
Vol 23 (6) ◽  
pp. 781-784 ◽  
Author(s):  
Joseph Constantin ◽  
Chaïban Nasr ◽  
Denis Hamad
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Robot Manipulator ◽  
Recursive Algorithm ◽  
Dynamic Control ◽  
Real Time Control ◽  
Time Control ◽  
The Neural Network ◽  
Artificial Neural ◽  
Pendubot System

The paper introduces artificial neural networks for the conventional control of robotic systems for better tracking performance. Different advanced dynamic control techniques are explained and a new second order recursive algorithm has been developed to tune the weights of the neural network. The problem of real-time control of a Pendubot system in difficult situations has been addressed. Examples, such as positioning and balancing structures, are presented and performances are compared to a conventional PD controller.

Simple feedback logic, genetic algorithms and artificial neural networks for real-time control of a collection system

1998 ◽  
Vol 38 (3) ◽  
pp. 187-195
Author(s):  
Pavel Hajda ◽  
Vladimir Novotny ◽  
Xin Feng ◽  
Ruoli Yang
Keyword(s):  
Neural Networks ◽  
Genetic Algorithms ◽  
Artificial Neural Networks ◽  
Real Time ◽  
Real Time Control ◽  
Flow Routing ◽  
Time Control ◽  
On Line ◽  
Artificial Neural ◽  
Simple Feedback

This paper describes a pilot-scale implementation of a simple, real-time control (RTC) algorithm based on feedback and also outlines the development and simulation testing of a new RTC methodology that combines genetic algorithms (GAs) and artificial neural networks (ANNs). Computer simulations indicated that the simple feedback logic could reduce pumping by 50 to 80 percent if used to replace the existing RTC system in the test area. Experience with the algorithm after its implementation has confirmed the potential of the algorithm to reduce pumping. Additional simulations with an emerging approach to control (based on GAs) indicated possibilities of reducing pumping still further. Although relatively simple flow routing was used in the GAs, these algorithms do not restrict flow routing to any particular method. If highly accurate flow routing is incorporated, GAs are likely to be rendered too slow for on-line applications. Nevertheless, GAs can still be used, because they can be combined with fast executing on-line algorithms, such as ANNs. This possibility was demonstrated by training a multi-layer ANN to approximate one of the GAs developed. In verification runs the trained ANN provided virtually the same control decisions as did the GA used as the source of the training data.

scholarly journals Downscaling and Verification of Maximum Wind Speeds by Using Artificial Neural Networks for New European Wind Atlas and Wind Farm Data 

2021 ◽  
Author(s):  
Melek Akın ◽  
Ahmet Öztopal ◽  
Ahmet Duran Şahin
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Wind Speed ◽  
Wind Turbines ◽  
Electricity Generation ◽  
Wind Farm ◽  
Wind Farms ◽  
Maximum Wind ◽  
Wind Speeds ◽  
Artificial Neural

<p>As is known, wind is a renewable and non-polluting energy resource. In addition, there is no transportation problem in wind energy and it does not require very high technology for electricity generation. Wind turbines are used for electricity generation from kinetic energy of wind. In the point of power curves of these turbines, wind speed must be a certain band. Generally, they do not generate electricity cut-in wind speed that is between 0 and 4 m/s and cut out wind speed that is over 20-25 m/s. Over cut-out values cause breaking down of wind turbines, because high wind speeds create extra mechanical loads on them. Therefore, maximum/extreme winds and their estimation and prediction carry weight in terms of energy generation.</p><p>New European Wind Atlas (NEWA) is the project, within the scope of ERANET+ Program, and the attendants are Belgium, Denmark, Germany, Latvia, Portugal, Spain, Sweden, and Turkey. The aim of NEWA is to present a new wind atlas to the wind industry. In this project, the physical model used for obtaining wind speeds is a numerical weather prediction model named Weather Research and Forecasting (WRF).</p><p>One of the methods, which are developed by imitating of biological properties of living forms in a virtual environment, is Artificial Neural Networks (ANNs). Stimulations taken from the environment by using sense organs are transmitted to brain whereby neurons in a body and brain makes a decision towards these stimulations. That is the working form of ANNs. Moreover, ANNs can be thought as a black box, which processes given data and produces outputs against inputs. Furthermore, they are a method of Artificial Intelligence.</p><p>In this study, maximum wind speeds of 4 different wind farms in Turkey were estimated by using a downscaling method based on ANNs and wind data which were produced in grid points of NEWA Project. Besides that, 8 different levels (10, 50, 75, 100, 150, 200, 250, and 500 m) for each wind farm were considered. As a result of determining the best ANN architectures with sensitivity analysis, it was seen that Levenberg-Marquardt Backpropagation (trainlm) approach as a training algorithm and 9 neurons in each layer are common traits of best ANN architectures. In addition, 50 m for 2 wind farms and 10 m with 75 m for others were determined as an optimum downscaling levels. Moreover, according to downscaling results, correlation values were calculated around 0.80.</p><p><strong>Key Words: </strong>ANN, Downscaling, Maximum wind, NEWA, Turkey, Wind farm.</p>

scholarly journals Modelling and Prediction of Monthly Global Irradiation Using Different Prediction Models

Energies ◽  
2021 ◽  
Vol 14 (8) ◽  
pp. 2332
Author(s):  
Cecilia Martinez-Castillo ◽  
Gonzalo Astray ◽  
Juan Carlos Mejuto
Keyword(s):  
Neural Networks ◽  
Artificial Neural Networks ◽  
Random Forests ◽  
Prediction Models ◽  
Meteorological Station ◽  
Ann Model ◽  
Global Irradiation ◽  
Artificial Neural ◽  
Input Variables ◽  
Using Data

Different prediction models (multiple linear regression, vector support machines, artificial neural networks and random forests) are applied to model the monthly global irradiation (MGI) from different input variables (latitude, longitude and altitude of meteorological station, month, average temperatures, among others) of different areas of Galicia (Spain). The models were trained, validated and queried using data from three stations, and each best model was checked in two independent stations. The results obtained confirmed that the best methodology is the ANN model which presents the lowest RMSE value in the validation and querying phases 1226 kJ/(m2∙day) and 1136 kJ/(m2∙day), respectively, and predict conveniently for independent stations, 2013 kJ/(m2∙day) and 2094 kJ/(m2∙day), respectively. Given the good results obtained, it is convenient to continue with the design of artificial neural networks applied to the analysis of monthly global irradiation.

Sign in / Sign up

Export Citation Format

Share Document