A D-Vine Copula-Based Quantile Regression Approach for the Prediction of Heating Energy Consumption. Using Historical Data for German Households

Inhaltsverzeichnis (Table of Contents)

• Abstract
• Acknowledgements
• Table of Contents
• List of Figures
• List of Tables
• Glossary
• 1 Introduction
• 2 Categorizing Energy Models
• 2.1 Building Type
• 2.2 Energy Type
• 2.3 Occupant Behavior
• 2.4 Prediction Model Type for Energy Consumption
• 3 Data-Driven Methods for the Prediction of Heating Energy Consumption
• 3.1 Common Characteristics of Data-Driven Methods
• 3.2 Data-Driven Methods in Academic Literature
• 3.2.1 Least Squares Regression
• 3.2.2 Artificial Neural Network
• 3.2.3 Genetic Algorithm
• 3.2.4 Support Vector Machines
• 3.2.5 Quantile Regression
• 3.3 Summary of Data-Driven Methods for Prediction of Heating Energy Consumption
• 4 An Introduction to Copula-Based Quantile Regression
• 4.1 Pair-Copula Construction
• 4.2 Bivariate Copula Families and Parameter Estimation
• 4.3 D-Vine Copula-Based Quantile Regression
• 4.4 Model Evaluation
• 5 Description of the Underlying Data Set
• 5.1 Data Understanding
• 5.2 Pre-Processing of the Data
• 6 Empirical Results
• 6.1 Performance of Point Estimation Methods
• 6.2 D-Vine Copula Fitting Results
• 7 Practical Implications of the Introduced Quantile Regression
• 7.1 Prediction of Quantiles for Post-Retrofit Energy Consumption
• 7.2 Quantile-Based Analysis of the Rebound Effect and Performance Gap
• 7.3 Rebound Effect Similarity Analysis
• 8 Conclusion
• References
• Appendix

Zielsetzung und Themenschwerpunkte (Objectives and Key Themes)

This master's thesis investigates the prediction of heating energy consumption in residential buildings, focusing on the use of data-driven methods. The primary objective is to evaluate the effectiveness of a D-vine copula-based quantile regression model in predicting heating energy consumption based on historical data from German households. The study explores the potential of this model to overcome the "performance gap" between predicted and actual energy consumption, a persistent issue in traditional building energy models.

• Data-driven energy consumption modeling
• Quantile regression and D-vine copula application
• Analysis of the rebound effect and performance gap
• Evaluation of model accuracy and predictive power
• Practical implications for energy efficiency policies and private investments

Zusammenfassung der Kapitel (Chapter Summaries)

• Chapter 1: Introduction Provides an overview of the research problem, highlighting the significance of accurate energy consumption prediction for achieving climate goals. It introduces the concept of the "performance gap" between predicted and actual consumption and outlines the potential of data-driven methods to address this issue.
• Chapter 2: Categorizing Energy Models Presents a framework for classifying energy models based on building type, energy type, occupant behavior, and prediction model type. This chapter provides context for the study by placing data-driven methods within the broader landscape of energy modeling approaches.
• Chapter 3: Data-Driven Methods for the Prediction of Heating Energy Consumption Reviews common data-driven methods used for predicting heating energy consumption, including least squares regression, artificial neural networks, genetic algorithms, support vector machines, and quantile regression. The chapter discusses the strengths and limitations of each approach and explores their relevance to the research question.
• Chapter 4: An Introduction to Copula-Based Quantile Regression Explains the concept of copula-based quantile regression and its potential for analyzing the entire distribution of energy consumption. This chapter delves into the mathematical principles of pair-copula construction, bivariate copula families, parameter estimation, and model evaluation.
• Chapter 5: Description of the Underlying Data Set Provides a detailed description of the data set used for the study, including its origin, size, and key variables. The chapter explains the process of data understanding and pre-processing, preparing the data for analysis.
• Chapter 6: Empirical Results Presents the results of applying the D-vine copula-based quantile regression model to the data set. This chapter analyzes the model's performance compared to other point estimation methods and examines the fitted D-vine copula results.
• Chapter 7: Practical Implications of the Introduced Quantile Regression Discusses the practical implications of the findings, including the prediction of quantiles for post-retrofit energy consumption and the analysis of the rebound effect and performance gap. This chapter explores the insights gained from the quantile-based analysis and its relevance for energy efficiency policies and private investments.

Schlüsselwörter (Keywords)

The study focuses on data-driven methods for the prediction of heating energy consumption in residential buildings, specifically exploring the use of D-vine copula-based quantile regression. Key themes include the performance gap between predicted and actual energy consumption, the rebound effect, and the application of these methods to support policy-making and investment decisions in the energy sector.