Smart Cities: Big Data Prediction Methods and Applications

✍ Scribed by Hui Liu

• Publisher: Springer
• Year: 2020
• Tongue: English
• Leaves: 338
• Category: Library

✦ Synopsis

Smart Cities: Big Data Prediction Methods and Applications is the first reference to provide a comprehensive overview of smart cities with the latest big data predicting techniques.

This timely book discusses big data forecasting for smart cities. It introduces big data forecasting techniques for the key aspects (e.g., traffic, environment, building energy, green grid, etc.) of smart cities, and explores three key areas that can be improved using big data prediction: grid energy, road traffic networks and environmental health in smart cities. The big data prediction methods proposed in this book are highly significant in terms of the planning, construction, management, control and development of green and smart cities.

Including numerous case studies to explain each method and model, this easy-to-understand book appeals to scientists, engineers, college students, postgraduates, teachers and managers from various fields of artificial intelligence, smart cities, smart grid, intelligent traffic systems, intelligent environments and big data computing.

✦ Table of Contents

Preface
Acknowledgements
About the Book
Contents
List of Figures
List of Tables
Abbreviations
Part I
Chapter 1: Key Issues of Smart Cities
1.1 Smart Grid and Buildings
1.1.1 Overview of Smart Grid and Building
1.1.2 The Importance of Smart Grid and Buildings in Smart City
1.1.3 Framework of Smart Grid and Buildings
1.2 Smart Traffic Systems
1.2.1 Overview of Smart Traffic Systems
1.2.2 The Importance of Smart Traffic Systems for Smart City
1.2.3 Framework of Smart Traffic Systems
1.3 Smart Environment
1.3.1 Overview of Smart Environment for Smart City
1.3.2 The Importance of Smart Environment for Smart City
1.3.3 Framework of Smart Environment
1.4 Framework of Smart Cities
1.4.1 Key Points of Smart City in the Era of Big Data
1.4.2 Big Data Time-series Forecasting Methods in Smart Cities
1.4.3 Overall Framework of Big Data Forecasting in Smart Cities
1.5 The Importance Analysis of Big Data Forecasting Architecture for Smart Cities
1.5.1 Overview and Necessity of Research
1.5.2 Review on Big Data Forecasting in Smart Cities
1.5.3 Review on Big Data Forecasting in Smart Gird and Buildings
1.5.4 Review on Big Data Forecasting in Smart Traffic Systems
1.5.5 Review on Big Data Forecasting in Smart Environment
References
Part II
Chapter 2: Electrical Characteristics and Correlation Analysis in Smart Grid
2.1 Introduction
2.2 Extraction of Building Electrical Features
2.2.1 Analysis of Meteorological Elements
2.2.2 Analysis of System Load
2.2.3 Analysis of Thermal Perturbation
2.3 Cross-Correlation Analysis of Electrical Characteristics
2.3.1 Cross-Correlation Analysis Based on MI
2.3.1.1 The Theoretical Basis of MI
2.3.1.2 Cross-Correlation Result of Electrical Characteristics
2.3.2 Cross-Correlation Analysis Based on Pearson Coefficient
2.3.2.1 The Theoretical Basis of Pearson Coefficient
2.3.2.2 Cross-Correlation Result of Electrical Characteristics
2.3.3 Cross-Correlation Analysis Based on Kendall Coefficient
2.3.3.1 The Theoretical Basis of Kendall Coefficient
2.3.3.2 Cross-Correlation Result of Electrical Characteristics
2.4 Selection of Electrical Characteristics
2.4.1 Electrical Characteristics of Construction Power Grid
2.4.2 Feature Selection Based on Spearman Correlation Coefficient
2.4.2.1 The Theoretical Basis of Spearman Coefficient
2.4.2.2 Result of Feature Selection
2.4.3 Feature Selection Based on CFS
2.4.3.1 The Theoretical Basis of CFS
2.4.3.2 Result of Feature Selection
2.4.4 Feature Selection Based on Global Search-ELM
2.4.4.1 The Theoretical Basis of Global Search-ELM
2.4.4.2 Result of Feature Selection
2.5 Conclusion
References
Chapter 3: Prediction Model of City Electricity Consumption
3.1 Introduction
3.2 Original Electricity Consumption Series
3.2.1 Regional Correlation Analysis of Electricity Consumption Series
3.2.2 Original Sequences for Modeling
3.2.3 Separation of Sample
3.3 Short-Term Deterministic Prediction of Electricity Consumption Based on ARIMA Model
3.3.1 Model Framework of ARIMA
3.3.2 Theoretical Basis of ARIMA
3.3.3 Modeling Steps of ARIMA Predictive Model
3.3.4 Forecasting Results
3.4 Power Consumption Interval Prediction Based on ARIMA-ARCH Model
3.4.1 Model Framework of ARCH
3.4.2 The Theoretical Basis of the ARCH
3.4.3 Modeling Steps of ARIMA-ARCH Interval Predictive Model
3.4.4 Forecasting Results
3.5 Long-Term Electricity Consumption Prediction Based on the SARIMA Model
3.5.1 Model Framework of the SARIMA
3.5.2 The Theoretical Basis of the SARIMA
3.5.3 Modeling Steps of the SARIMA Predictive Model
3.5.4 Forecasting Results
3.6 Big Data Prediction Architecture of Household Electric Power
3.7 Comparative Analysis of Forecasting Performance
3.8 Conclusion
References
Chapter 4: Prediction Models of Energy Consumption in Smart Urban Buildings
4.1 Introduction
4.2 Establishment of Building Simulating Model
4.2.1 Description and Analysis of the BEMPs
4.2.2 Main Characters of DeST Software
4.2.2.1 Combine Buildings and Environmental Control Systems with Base Temperature
4.2.2.2 Design and Simulation in Stages
4.2.2.3 Graphical Interface
4.2.3 Process of DeST Modeling
4.2.3.1 Start Establishing Architectural Models
4.2.3.2 Set the Floors of the Building
4.2.3.3 Set the Windows and Doors
4.2.3.4 Pre-processing of Construction
4.2.3.5 Set Wind Pressure
4.2.3.6 Set Indoor Ventilation
4.2.3.7 Condition of Building Simulation
4.2.3.8 Simulation Process
4.2.3.9 Global Building Settings
4.2.3.10 Calculation of Shadow and Lighting
4.2.3.11 Output of Simulation
4.3 Analysis and Comparison of Different Parameters
4.3.1 Introduction of the Research
4.3.2 Meteorological Parameters
4.3.3 Indoor Thermal Perturbation
4.3.4 Enclosure Structure and Material Performance
4.3.5 Indoor Design Parameters
4.4 Data Acquisition of Building Model
4.4.1 Data After Modeling
4.4.2 Calculation of Room Temperature and Load
4.4.3 Calculation of Shadow and Light
4.4.4 Calculation of Natural Ventilation
4.4.5 Simulation of the Air-Conditioning System
4.5 SVM Prediction Model for Urban Building Energy Consumption
4.5.1 The Theoretical Basis of the SVM
4.5.2 Steps of Modeling
4.5.2.1 Data Selection
4.5.2.2 Samples Setting
4.5.2.3 Initialization and Training of the SVM Model
4.5.2.4 Testing the Trained SVM Model
4.5.3 Forecasting Results
4.6 Big Data Prediction of Energy Consumption in Urban Building
4.6.1 Big Data Framework for Energy Consumption
4.6.2 Big Data Storage and Analysis for Energy Consumption
4.6.3 Big Data Mining for Energy Consumption
4.7 Conclusion
References
Part III
Chapter 5: Characteristics and Analysis of Urban Traffic Flow in Smart Traffic Systems
5.1 Introduction
5.1.1 Overview of Trajectory Prediction of Smart Vehicle
5.1.2 The Significance of Trajectory Prediction for Smart City
5.1.3 Overall Framework of Model
5.2 Traffic Flow Time Distribution Characteristics and Analysis
5.2.1 Original Vehicle Trajectory Series
5.2.2 Separation of Sample
5.3 The Spatial Distribution Characteristics and Analysis of Traffic Flow
5.3.1 Trajectory Prediction of Urban Vehicles Based on Single Data
5.3.1.1 Theoretical Basis of the ELM and BPNN
The Theoretical Basis of the ELM
The Theoretical Basis of the BPNN
5.3.1.2 Framework of Model
5.3.1.3 Steps of Modeling
5.3.1.4 Forecasting Results
Forecasting Results of the ELM Based on Single Data
Forecasting Results of the BPNN Based on Single Data
5.3.2 Trajectory Prediction of Urban Vehicles Based on Multiple Data
5.3.2.1 Framework of Model
5.3.2.2 Steps of Modeling
5.3.2.3 Forecasting Results
Forecasting Results of the ELM Based on Multiple Data
Forecasting Results of BPNN Based on Multiple Data
5.3.3 Trajectory Prediction of Urban Vehicles Under EWT Decomposition Framework
5.3.3.1 The Theoretical Basis of the EWT
5.3.3.2 Framework of Model
5.3.3.3 Steps of Modeling
5.3.3.4 Forecasting Results
Forecasting Results of EWT-ELM Based on Single Data
Forecasting Results of the EWT-BPNN Based on Single Data
5.3.4 Comparative Analysis of Forecasting Performance
5.4 Conclusion
References
Chapter 6: Prediction Model of Traffic Flow Driven Based on Single Data in Smart Traffic Systems
6.1 Introduction
6.2 Original Traffic Flow Series for Prediction
6.3 Traffic Flow Deterministic Prediction Driven by Single Data
6.3.1 Modeling Process
6.3.2 The Prediction Results
6.4 Traffic Flow Interval Prediction Model Driven by Single Data
6.4.1 The Framework of the Interval Prediction Model
6.4.2 Modeling Process
6.4.3 The Prediction Results
6.5 Traffic Flow Interval Prediction Under Decomposition Framework
6.5.1 The Framework of the WD-BP-GARCH Prediction Model
6.5.2 Modeling Process
6.5.3 The Prediction Results
6.6 Big Data Prediction Architecture of Traffic Flow
6.7 Comparative Analysis of Forecasting Performance
6.8 Conclusion
References
Chapter 7: Prediction Models of Traffic Flow Driven Based on Multi-Dimensional Data in Smart Traffic Systems
7.1 Introduction
7.2 Analysis of Traffic Flow and Its Influencing Factors
7.3 Elman Prediction Model of Traffic Flow Based on Multiple Data
7.3.1 The Framework of the Elman Prediction Model
7.3.2 Modeling Process
7.3.3 The Prediction Results
7.4 LSTM Prediction Model of Traffic Flow Based on Multiple Data
7.4.1 The Framework of the LSTM Prediction Model
7.4.2 Modeling Process
7.4.3 The Prediction Results
7.5 Traffic Flow Prediction Under Wavelet Packet Decomposition
7.5.1 The Framework of the WPD-Prediction Model
7.5.2 Modeling Process
7.5.3 The Prediction Results
7.6 Comparative Analysis of Forecasting Performance
7.7 Conclusion
References
Part IV
Chapter 8: Prediction Models of Urban Air Quality in Smart Environment
8.1 Introduction
8.2 Original Air Pollutant Concentrations Series for Prediction
8.2.1 Original Sequence for Modeling
8.2.2 Separation of Sample
8.3 Air Quality Prediction Model Driven by Single Data
8.3.1 Model Framework
8.3.2 Theoretical Basis of ELM
8.3.3 Steps of Modeling
8.3.4 Forecasting Results
8.4 Air Quality Mixture Prediction Model Driven by Multiple Data
8.4.1 Model Framework
8.4.2 Steps of Modeling
8.4.3 Forecasting Results
8.5 Air Quality Prediction Under Feature Extraction Framework
8.5.1 Model Framework
8.5.2 Theoretical Basis of Feature Extraction Method
8.5.2.1 Principal Component Analysis
Algorithm Principle
The Identification Process
Data Standardization
Principal Components
Information Contribution Rate and Cumulative Information Contribution Rate
Identification Result
8.5.2.2 Kernel Principal Components Analysis
Algorithm Principle
The Identification Process
Nonlinear Mapping Based on the Gaussian Kernel
Information Contribution Rate and Cumulative Information Contribution Rate
Identification Result
8.5.2.3 Factor Analysis
Algorithm Principle
Factor Analysis Model
Contribution Degree
Heywood Case
The Identification Process
Data Standardization
The Applicability of FA
Information Contribution Rate and Cumulative Information Contribution Rate
Identification Result
8.5.3 Steps of Modeling
8.5.4 Forecasting Results
8.6 Big Data Prediction Architecture of Urban Air Quality
8.6.1 The Idea of Urban Air Quality Prediction Based on Hadoop
8.6.2 Parallelization Framework of the ELM
8.6.3 The Parallelized ELM Under the MapReduce Framework
8.7 Comparative Analysis of Forecasting Performance
8.8 Conclusion
References
Chapter 9: Prediction Models of Urban Hydrological Status in Smart Environment
9.1 Introduction
9.2 Original Hydrological State Data for Prediction
9.2.1 Original Sequence for Modeling
9.2.2 Separation of Sample
9.3 Bayesian Classifier Prediction of Water Level Fluctuation
9.3.1 Model Framework
9.3.2 Theoretical Basis of the Bayesian Classifier
9.3.3 Steps of Modeling
9.3.3.1 Dataset Preparation
9.3.3.2 Classifier Training
9.3.3.3 Self-Predicting Water Level Fluctuation Trend
9.3.4 Forecasting Results
9.4 The Elman Prediction of Urban Water Level
9.4.1 Model Framework
9.4.2 The Theoretical Basis of the Elman
9.4.3 Steps of Modeling
9.4.4 Forecasting Results
9.5 Urban River Water Level Decomposition Hybrid Prediction Model
9.5.1 Model Framework
9.5.2 The Theoretical Basis
9.5.2.1 Maximal Overlap Discrete Wavelet Transform
9.5.2.2 Empirical Mode Decomposition
9.5.2.3 Singular Spectrum Analysis
9.5.3 Steps of Modeling
9.5.4 Forecasting Results
9.5.5 Influence and Analysis of Decomposition Parameters on Forecasting Performance of Hybrid Models
9.5.5.1 Effect Analysis of the Decomposition Layers on the Performance of the MODWT
9.5.5.2 Effect Analysis of the Mother Wavelet on the Performance of the MODWT
9.5.5.3 Effect Analysis of the Window Length on the Performance of the SSA
9.6 Comparative Analysis of Forecasting Performance
9.7 Conclusion
References
Chapter 10: Prediction Model of Urban Environmental Noise in Smart Environment
10.1 Introduction
10.1.1 Hazard of Noise
10.1.2 The Significance of Noise Prediction for Smart City
10.1.3 Overall Framework of Model
10.2 Original Urban Environmental Noise Series
10.2.1 Original Sequence for Modeling
10.2.2 Separation of Sample
10.3 The RF Prediction Model for Urban Environmental Noise
10.3.1 The Theoretical Basis of the RF
10.3.2 Steps of Modeling
10.3.3 Forecasting Results
10.4 The BFGS Prediction Model for Urban Environmental Noise
10.4.1 The Theoretical Basis of the BFGS
10.4.2 Steps of Modeling
10.4.3 Forecasting Results
10.5 The GRU Prediction Model for Urban Environmental Noise
10.5.1 The Theoretical Basis of the GRU
10.5.2 Steps of Modeling
10.5.3 Forecasting Results
10.6 Big Data Prediction Architecture of Urban Environmental Noise
10.6.1 Big Data Framework for Urban Environmental Noise Prediction
10.6.2 Big Data Storage for Urban Environmental Noise Prediction
10.6.3 Big Data Processing of Urban Environmental Noise Prediction
10.7 Comparative Analysis of Forecasting Performance
10.8 Conclusion
References