Scholarlink

<p>The plasma in the sun causes various magnetic activities on the surface of the sun, for example, the appearance of dark regions on the sun’s surface, known as sunspots. These dark regions are temporary and are cooler than their surroundings. The sunspot number is a variable that follows a periodic function having a period of 9 to 13 years. The sunspot phenomena are closely related to the solar flares and coronal mass ejection phenomena. Mathematical modeling and artificial neural networks have been used in this study to predict the number of sunspots. The sunspot cycles vary according to the magnetic activities, and the variation in profile affects shape and scale parameters. Weibull distribution with two parameters (shape and scale) has been used to model the profile of sunspot cycles. The shape parameters are modeled using the sine function, and the scale parameters are predicted using regression and Artificial Neural Network (ANN). The amplitude of cycle 25 is predicted using the precursor method applied via deep learning and found to be 166&#xa0;± 28. The expected occurrence time of the amplitude of cycle 25 is April 2025. The amplitude of cycle 26 is also determined.</p>

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Prediction of sunspot numbers via Weibull distribution and deep learning

  • 期刊论文

  • 发表日期：

    2025年7月8日

查看全文
  • Waqas Qamar,
  • Majid Hussain,
  • M. Basit Zaheer,
  • Jawaid Akram,
  • Naeem Sadiq,
  • Zaheer Uddin

摘要

The plasma in the sun causes various magnetic activities on the surface of the sun, for example, the appearance of dark regions on the sun’s surface, known as sunspots. These dark regions are temporary and are cooler than their surroundings. The sunspot number is a variable that follows a periodic function having a period of 9 to 13 years. The sunspot phenomena are closely related to the solar flares and coronal mass ejection phenomena. Mathematical modeling and artificial neural networks have been used in this study to predict the number of sunspots. The sunspot cycles vary according to the magnetic activities, and the variation in profile affects shape and scale parameters. Weibull distribution with two parameters (shape and scale) has been used to model the profile of sunspot cycles. The shape parameters are modeled using the sine function, and the scale parameters are predicted using regression and Artificial Neural Network (ANN). The amplitude of cycle 25 is predicted using the precursor method applied via deep learning and found to be 166 ± 28. The expected occurrence time of the amplitude of cycle 25 is April 2025. The amplitude of cycle 26 is also determined.