{"course":{"id":"115112800N01","name":"人工智慧與智慧電網","enroll_type":"選修","general_type":"","dept":"電機系碩士班","grade":1,"class_group":"甲","teacher":"李浩賢","credit":3,"hour":3,"time":[{"day":4,"section":[5,6,7]}],"classroom":"","max":30,"min":5,"remark":"","alt_for":"","courses":""},"detail":[["指定教材","作者：Liu Yanli\n書名：Energy Big Data and Artificial Intelligence\n版別：1\n出版商：Science Press\n發行地：China\n出版年份：2024"],["參考教材","Instructor-prepared supplementary materials"],["成績評量標準","Midterm Exam: 20%\nFinal Exam: 20%\nProject Report: 30%\nAttendance +Homework/Assignments: 30%"],["教學方法","理論講授,實務講授,個案分析"],["教學目標","This course, based on the textbook Energy Big Data and Artificial Intelligence (edited by Liu Yanli, aims to equip graduate students with a solid theoretical foundation and practical skills at the intersection of artificial intelligence (AI) and smart grid technologies.\n\nKnowledge goals: Students will understand the strategic importance of smart grids in addressing climate change and achieving “dual carbon” goals. They will master the hierarchical and clustered architecture of smart grids and the key technology systems; become familiar with the composition, sources, characteristics, and multi‑dimensional value of energy big data; and systematically learn the history and core methods of AI, including convolutional neural networks (CNNs), recurrent neural networks (RNNs/LSTMs), the Transformer architecture, and fine‑tuning techniques for large language models (LLMs).\n\nSkill goals: Students will develop the ability to apply big‑data analytics and AI methods to solve practical smart‑grid problems, covering supply‑side and demand‑side applications such as wind/PV power forecasting, customer load profiling, electric vehicle behaviour analysis, and battery state estimation; as well as system‑level applications including situational awareness, operational control, and electricity market trading.\n\nCompetency goals: The course fosters interdisciplinary thinking and research innovation, preparing students for scientific research or engineering practice in the fields of smart grids and AI."]],"office_time":[["李浩賢","Wed: 13:00-15:00"]],"schedule":[["2026/09/10","5","7","3","李浩賢","Introduction and Course Overview","","General introduction to the course; explanation of grading policy and learning requirements. Discussion of the significance of smart grids in addressing climate change and the need for digital and intelligent transformation of power systems under the “dual carbon” targets. Overview of the course structure and learning pathway.",""],["2026/09/17","5","7","3","李浩賢","Fundamentals of Electric Power Systems","","Review of basic concepts in electric power systems, including generation, transmission, transformation, distribution, and consumption. Introduction to fundamental physical laws and mathematical models for system operation, laying the foundation for subsequent study of smart grids and AI applications.",""],["2026/09/24","5","7","3","李浩賢","Smart Grid Concepts and Key Features","","Introduction to the main characteristics and overall vision of smart grids; comparison of functions between smart grids and conventional grids. Discussion on how information technologies enable grid modernisation and intelligence.",""],["2026/10/01","5","7","3","李浩賢","Hierarchical and Clustered Architecture of Power Grids","","In‑depth explanation of the hierarchical and clustered structure of power grids, including the architecture of interconnected (transmission) systems, an overview of the hierarchical clustering, and its mathematical foundations. Exploration of how to build a grid as intelligent as the Internet.",""],["2026/10/08","5","7","3","李浩賢","Key Technology Systems for Smart Grids","","Presentation of operational technologies, layered description of smart‑grid components, and related technical connotations. Coverage of communication, sensing, control, and information integration technologies used in smart grids.",""],["2026/10/15","5","7","3","李浩賢","Composition and Characteristics of Energy Big Data","","Introduction to the composition of energy system data, sources of electric power big data, and their characteristics. Discussion on the 4V properties (Volume, Velocity, Variety, Value) and their specific manifestations in power systems.",""],["2026/10/22","5","7","3","李浩賢","Multi‑dimensional Value and Analytical Methods of Power Big Data","","xploration of the value of energy big data from multiple perspectives. Definition of data analysis methods, classification of power big‑data analytics, with emphasis on AMI (Advanced Metering Infrastructure) data analysis techniques.",""],["2026/10/29","5","7","3","李浩賢","AI History and Convolutional Neural Networks (CNNs)","","Overview of the definition and historical development of AI. Principles and architecture of CNNs; application workflows and case studies to illustrate CNN usage in power system scenarios.",""],["2026/11/05","5","7","3","李浩賢","Midterm Review and Examination","Midterm Exam","Midterm Exam, covers topics from Weeks 1–8, including power system fundamentals, smart‑grid concepts and architecture, energy big‑data basics, AI overview, and CNNs.",""],["2026/11/12","5","7","3","李浩賢","Recurrent Neural Networks (RNNs) and Long Short‑Term Memory (LSTM)","","Principles and architecture of RNNs, with emphasis on LSTM networks. Application procedures and case analyses to demonstrate the advantages of RNN/LSTM in processing time‑series data.",""],["2026/11/19","5","7","3","李浩賢","Transformer Architecture and Self‑Attention Mechanism","","Introduction to the overall Transformer architecture, positional encoding, and the self‑attention mechanism. Examples and application cases to illustrate the potential of Transformers in the energy domain.",""],["2026/11/26","5","7","3","李浩賢","Large Language Models and Fine‑Tuning Techniques","","Development history of large language models; fine‑tuning methodologies. Hands‑on demonstration of fine‑tuning using the Swift library to give students practical experience.",""],["2026/12/03","5","7","3","李浩賢","Supply‑Side Applications (I) – Wind and PV Power Forecasting","","Methods for wind power forecasting based on high‑frequency data generation; Bayesian optimisation for PV forecasting; spatio‑temporal feature enhancement for renewable forecasting; and reinforcement learning‑based probabilistic wind forecasting under extreme scenarios.",""],["2026/12/10","5","7","3","李浩賢","Demand‑Side Applications (II) – Customer Load Profiling and EV Analysis","","Techniques for customer load profile imputation, super‑resolution load perception, and non‑intrusive load monitoring. Analysis of electric vehicle behaviour using big data and AI.",""],["2026/12/17","5","7","3","李浩賢","Demand‑Side Applications (III) – Battery State Estimation","","SOC (state of charge) estimation using ensemble models; SOH (state of health) estimation based on GRU networks; and remaining useful life prediction using TCN (temporal convolutional networks) for lithium‑ion batteries.",""],["2026/12/24","5","7","3","李浩賢","System‑Level Applications – Situational Awareness and Operational Control","","Hybrid data‑model driven topology identification and line parameter estimation; model‑guided deep extreme learning machine for power flow calculation; GAN‑based generation of security boundaries for large power grids. Introduction to problem formulation, method classification, application workflow, and case studies for system operation and control.",""],["2026/12/31","5","7","3","李浩賢","Final Review and Examination","Final Exam","Final Exam, covers Weeks 10–16, focusing on RNN/LSTM, Transformer, LLM fine‑tuning, and various applications on both supply/demand sides and system level.",""],["2027/01/07","5","7","3","李浩賢","Self-directed learning","Project Report submission","Project Report submission",""]]}