China
Africa
Explore chapters and articles related to this topic
Wind Power Forecasting via Deep Learning Methods
Published in Jacqueline A. Stagner, David S-K. Ting, Green Energy and Infrastructure, 2020
Accurate wind power forecasting involves numerous challenges. With the development of machine learning, forecasting models started to receive attention. Intelligence approaches gain dominance on wind power forecasting. Deep learning is a type of machine learning, which has many hidden layers. Besides the other field, deep learning has recently been applied to wind power forecasting problems. Wind power forecasting provides cost savings for utility operations, facilitates plant dispatch scheduling, and optimizes plant maintenance.
A Hybrid bVAR-NARX Wind Power Forecasting Model Based on Wind and Load Demand Correlation: A Case Study of ERCOT’s System from an ISO’s Perspective
Published in Electric Power Components and Systems, 2018
Leena Heistrene, Poonam Mishra, Makarand Lokhande
Artificial intelligence (AI) and fuzzy-based model have also been developed for wind power forecasting. Neural network, wavelet decomposition and other such AI-based techniques have been used in [15]–[19]. Hybrid models try to overcome the disadvantage of using a single approach and in fact, bring out the good in all the approaches involved in the model. Many of the hybrid models are made with neural network which is apt for both stationary as well as nonstationary ensemble prediction [20]–[35]. Most of these models are univariate in nature. A handful of them are multivariate models such as the work in [21]. But none of these models have tried to use the underlying relation between the total demand of the power system and the total wind power generated in that system for prediction. This article proposes a model that exploits these underlying relations to get better prediction results.
A Survey of Machine Learning Applications in Renewable Energy Sources
Published in IETE Journal of Research, 2022
Pulavarthi Satya Venkata Kishore, Jami Rajesh, Nakka Jayaram, Sukanta Halder
Wind energy is also the exciting green energy sources because it is clean, safe, and unlimited. Over the past decade, wind power has been growing very rapidly. As wind power is an intermittent renewable energy source, integrating it into the system has a negative impact on the grid’s reliability, protection, and power quality. Metrological factors precipitation, air density, and wind speed affect the power generation from wind energy systems. For proper scheduling and planning of the load for the grid, wind power forecasting is important. This gives a great advantage in ensuring successful grid management, thus reducing losses and thus reducing the cost of power generation. Machine learning algorithms help in forecasting wind power accurately.
Wind turbine output power forecasting based on temporal convolutional neural network and complete ensemble empirical mode decomposition with adaptive noise
Published in International Journal of Green Energy, 2023
Huajian Yang, Wangqiang Niu, Xiaotong Wang, Wei Gu
There are four main methods for wind power forecasting: statistical methods, physical methods, intelligent methods, and hybrid methods. Among them, statistical methods and physical methods have great difficulty in dealing with nonlinear problems. With the rapid development of artificial intelligence and big data technology, intelligent methods are widely used, and good prediction effects have been achieved on some non-linear problems. Hybrid methods, that is, multi-model fusion, are being used increasingly (Zhao and Jia 2020).