This book describes describes applications of machine learning techniques to maximum power point tracking (MPPT), PV-cell design and sizing, and hybridmodule design and modeling. PV-cell parameter extraction and estimation are proposed using moth–flame and chaotic optimization or deterministic and metaheuristic global optimization algorithms. Neuro-fuzzy inference systems (ANFIS) and a particle swarm optimization-artificial neural network model are applied to PV power output forecasting, while a domain adaptation of deep neural networks is proposed for multistep solar irradiance forecasting. Machine learning approaches are also presented for PV power prediction based on available environmental parameters. Additionally, PV plant operation failure modes and module deterioration diagnosis are studied using complex network and image analysis. After a preface, the book includes 16 carefully selected articles to cover recent trends in machine learning applications to PV systems. The book is an open access tool for engineers and researchers in applying machine learning methods to PV systems. The book is also suited for students willing to further use machine learning skills on PV applications and is a valuable resource for practicing professionals in need of understanding and pursuing advanced trends in PV systems.

中文翻译：

---

光伏应用中建模、控制、优化、预测和诊断的计算智能 [图书新闻]

本书描述了机器学习技术在最大功率点跟踪 (MPPT)、光伏电池设计和尺寸调整以及混合模块设计和建模方面的应用。PV 电池参数提取和估计是使用飞蛾火焰和混沌优化或确定性和元启发式全局优化算法提出的。神经模糊推理系统（ANFIS）和粒子群优化人工神经网络模型应用于光伏功率输出预测，而深度神经网络的域适应被提出用于多步太阳辐照度预测。还提出了基于可用环境参数的 PV 功率预测机器学习方法。此外，使用复杂网络和图像分析研究了光伏电站运行故障模式和组件劣化诊断。在序言之后，本书包含 16 篇精心挑选的文章，涵盖了光伏系统机器学习应用的最新趋势。这本书是工程师和研究人员将机器学习方法应用于光伏系统的开放获取工具。这本书也适合愿意在光伏应用中进一步使用机器学习技能的学生，对于需要了解和追求光伏系统先进趋势的实践专业人员来说，这是一个宝贵的资源。