About Me

Hello! I’m Yujie Chen (陈雨节), an undergraduate student majoring in Economics and Statistics at Dongbei University of Finance and Economics, expected to graduate in 2026.

My research interests lie in AI-based methods for power systems forecasting, with a vision to develop innovative and reliable forecasting models that can enhance downstream applications such as optimization, control, and electricity market operations in modern power systems.

🎓 Education

• 2022.09 - 2026.07, B.Sc. in Economics and Statistics, Dongbei University of Finance and Economics (DUFE), Dalian.
  • GPA: 4.42/5.0, Rank: 1/60
  • Core Courses: Time Series Analysis (99), Multimodal Data Analysis (100), Python Programming (100), Probability Theory (98), Probability and Mathematical Statistics (95), Linear Algebra (94)

💼 Research Experience

• 2023.09 - 2026.07, Research Team Member, Key Laboratory of Big Data Management, Optimization and Decision, Dongbei University of Finance and Economics, Dalian, Liaoning, China.
  • Developed the bionic robot real-time dialogue system.
• 2023.09 - 2026.07, Vice President, Yixing Association, Dongbei University of Finance and Economics, Dalian, Liaoning, China.
  • Organized and managed innovation and entrepreneurship projects.

📜 Publication

Note: The asterisk (*) after author names indicates the corresponding author.

2025

• Zhirui Tian, Yujie Chen* (co-first author), and Guangyu Wang. Enhancing PV power forecasting accuracy through nonlinear weather correction based on multi-task learning. Applied Energy, 386:125525, 2025. DOI: https://doi.org/10.1016/j.apenergy.2025.125525. (JCR: Q1, IF: 10.1)

• Yujie Chen and Zhirui Tian*. A crude oil price forecasting framework based on constraint guarantee and Pareto front shrinking strategy. Applied Soft Computing, 112996, 2025. DOI: https://doi.org/10.1016/j.asoc.2025.112996. (JCR: Q1, IF: 7.2)

• Yujie Chen, Mingyao Jin, Zheyu Zhou, and Zhirui Tian*. A novel ensemble learning framework based on news sentiment enhancement and multi-objective optimizer for carbon price forecasting. Computational Economics, pages 1–25, 2025. DOI: https://doi.org/10.1007/s10614-024-10828-6. (JCR: Q2, IF: 1.9)

🔬 Research Projects

1. Enhancing Photovoltaic Power Forecasting Accuracy through Nonlinear Weather Correction Based on Multi-Task Learning

• Challenge: PV forecasting is highly sensitive to weather variables, and simple features fail to capture the complex nonlinear relationship between PV power and weather.
• Highlight:
  • Proposed a multi-angle framework using weather variables to improve PV forecasting accuracy.
  • Introduced a customized multi-task learning model to capture complex weather-PV interactions.
  • Developed an MLP-based weather correction model to refine PV predictions with nonlinear interactions.
  • Validated the model using real-world PV data from Australia, showing improved accuracy.
• Published as corresponding author and co-first author in Applied Energy.

2. A Crude Oil Price Forecasting Framework Based on Constraint Guarantee and Pareto Front Shrinking Strategy

• Challenge: Multi-objective optimization methods have shown success in ensemble learning for crude oil price forecasting, but selecting solutions from the Pareto front to improve prediction accuracy remains a challenge.
• Highlight:
  • Proposed a multi-strategy ensemble learning framework to enhance crude oil price forecasting accuracy.
  • Introduced the Constraint Guarantee Strategy for selecting solutions from the Pareto front suitable for engineering applications.
  • Developed a customized Pareto Front Shrinking Strategy for objectively selecting a unique solution on the Pareto front.
• Published as first author in Applied Soft Computing.

3. A Novel Ensemble Learning Framework Based on News Sentiment Enhancement and Multi-objective Optimizer for Carbon Price Forecasting

• Challenge: Carbon price forecasting is influenced by news sentiment, and how to quantify sentiment and improve model robustness and accuracy using ensemble learning remains a challenge.
• Highlight:
  • Proposed a hybrid forecasting framework combining news sentiment enhancement with multi-objective optimization to improve carbon price forecasting accuracy.
  • Integrated NLP techniques to extract news sentiment features and incorporate them into time series prediction models.
  • Developed a Pareto front selection mechanism with knee point strategies to enhance model robustness.
• Published as first author in Computational Economics.

🌟 Others

🥇 Awards🥈🥉🏅🎖️

• National First Prize🥇, National Mathematical Modeling Contest for College Students, Team Leader
• M Award🏅, Mathematical Contest in Modeling (MCM), Team Leader
• National First Prize🥇, National Statistical Modeling Contest for College Students, Team Leader
• National First Prize🥇, Dongfang Caifu Cup National Financial Contest for College Students
• National Third Prize🥉, China International College Students Innovation and Entrepreneurship Competition
• National Project Grant🏅, College Student Innovation and Entrepreneurship Project
• Third Prize🥉, Challenge Cup Liaoning Province College Student Entrepreneurship Plan Contest

💸 Scholarships

• National Scholarship (2024)
• National Inspirational Scholarship (2023)
• First-Class Comprehensive Scholarship from University

🔐 Patents and Software Copyrights

• Software Copyright for Multi-modal Speech Evaluation System (Registration No.: 2022SR0827021)
• Software Copyright for Real-time Speech Dialogue System with Multi-source Emotion Perception Enhancement (Registration No.: 2025SR0440280)

💻 Skills

• Programming Languages: Python, PyTorch, MATLAB, LaTeX
• English: CET-4: 579, CET-6, Preparing for IELTS