Data mining and machine learning in building energy analysis / Frédéric Magoulès, Hai-Xiang Zhao.

Format:

Book

Author/Creator:

Magoulès, F. (Frédéric), author.

Zhao, Hai-Xiang, author.

Series:

Computer engineering series.

Computer Engineering Series.

THEi Wiley ebooks.

Language:

English

Subjects (All):

Buildings--Energy conservation--Research.

Buildings.

Data mining.

Physical Description:

1 online resource (187 pages)

Edition:

1st ed.

Place of Publication:

London, England ; Hoboken, New Jersey : ISTE : Wiley, 2016.

System Details:

Access using campus network via VPN at home (THEi Users Only).

Summary:

The energy consumption of a building has, in recent years, become a determining factor during its design and construction. With carbon footprints being a growing issue, it is important that buildings be optimized for energy conservation and CO2 reduction. This book therefore presents AI models and optimization techniques related to this application. The authors start with a review of recent models for the prediction of building energy consumption: engineering methods, statistical methods, artificial intelligence methods, ANNs and SVMs in particular. The book then focuses on SVMs, by first applying them to building energy consumption, then presenting the principles and various extensions, and SVR. The authors then move on to RDP, which they use to determine building energy faults through simulation experiments before presenting SVR model reduction methods and the benefits of parallel computing. The book then closes by presenting some of the current research and advancements in the field.

Contents:

Cover

Title Page

Copyright

Contents

Preface

Introduction

Chapter 1: Overview of Building Energy Analysis

1.1. Introduction

1.2. Physical models

1.3. Gray models

1.4. Statistical models

1.5. Artificial intelligence models

1.5.1. Neural networks

1.5.2. Support vector machines

1.6. Comparison of existing models

1.7. Concluding remarks

Chapter 2: Data Acquisition for Building Energy Analysis

2.1. Introduction

2.2. Surveys or questionnaires

2.3. Measurements

2.4. Simulation

2.4.1. Simulation software

2.4.2. Simulation process

2.4.2.1. Simulation details

2.4.2.2. Simulation of one single building

2.4.2.3. Simulation of multiple buildings

2.5. Data uncertainty

2.6. Calibration

2.7. Concluding remarks

Chapter 3: Artificial Intelligence Models

3.1. Introduction

3.2. Artificial neural networks

3.2.1. Single-layer perceptron

3.2.2. Feed forward neural network

3.2.3. Radial basis functions network

3.2.4. Recurrent neural network

3.2.5. Recursive deterministic perceptron

3.2.6. Applications of neural networks

3.3. Support vector machines

3.3.1. Support vector classification

3.3.2. ε-support vector regression

3.3.3. One-class support vector machines

3.3.4. Multiclass support vector machines

3.3.5. υ-support vector machines

3.3.6. Transductive support vector machines

3.3.7. Quadratic problem solvers

3.3.7.1. Interior point method

3.3.8. Applications of support vector machines

3.4. Concluding remarks

Chapter 4: Artificial Intelligence for Building Energy Analysis

4.1. Introduction

4.2. Support vector machines for building energy prediction

4.2.1. Energy prediction definition

4.2.2. Practical issues

4.2.2.1. Operation flow

4.2.2.2. Experimental environment

4.2.2.3. Data preprocessing.

4.2.2.4. Model selection

4.2.3. Support vector machines for prediction

4.2.3.1. Prediction of single building energy

4.2.3.2. Extensive model evaluation

4.2.3.3. Prediction of multiple buildings energy

4.3. Neural networks for fault detection and diagnosis

4.3.1. Description of faults

4.3.2. RDP in fault detection

4.3.2.1. Introduce faults to the simulated building

4.3.2.2. Experiments and results

4.3.3. RDP in fault diagnosis

4.4. Concluding remarks

Chapter 5: Model Reduction for Support Vector Machines

5.1. Introduction

5.2. Overview of model reduction

5.2.1. Wrapper methods

5.2.2. Filter methods

5.2.3. Embedded methods

5.3. Model reduction for energy consumption

5.3.1. Introduction

5.3.2. Algorithm

5.3.3. Feature set description

5.4. Model reduction for single building energy

5.4.1. Feature set selection

5.4.2. Evaluation in experiments

5.5. Model reduction for multiple buildings energy

5.6. Concluding remarks

Chapter 6: Parallel Computing for Support Vector Machines

6.1. Introduction

6.2. Overview of parallel support vector machines

6.3. Parallel quadratic problem solver

6.4. MPI-based parallel support vector machines

6.4.1. Message passing interface programming model

6.4.2. Pisvm

6.4.3. Psvm

6.5. MapReduce-based parallel support vector machines

6.5.1. MapReduce programming model

6.5.2. Caching technique

6.5.3. Sparse data representation

6.5.4. Comparison of MRPsvm with Pisvm

6.6. MapReduce-based parallel ε-support vector regression

6.6.1. Implementation aspects

6.6.2. Energy consumption datasets

6.6.3. Evaluation for building energy prediction

6.7. Concluding remarks

Summary and Future of Building Energy Analysis

Building energy consumption

Predicting building energy consumption.

Detection and diagnosis of building energy faults

Feature selection and model reduction

Parallel computing

Future work

Bibliography

Index.

Notes:

Description based upon print version of record.

Includes bibliographical references and index.

Description based on print version record.

ISBN:

9781118577592

1118577590

9781118577486

1118577485

OCLC:

934770383