Cyber-Physical Systems: Intelligent Models and Algorithms (Studies in Systems, Decision and Control, 417)

✍ Scribed by Alla G. Kravets (editor), Alexander A. Bolshakov (editor), Maxim Shcherbakov (editor)

• Publisher: Springer
• Year: 2022
• Tongue: English
• Leaves: 277
• Category: Library

✦ Synopsis

This book is devoted to intelligent models and algorithms as the core components of cyber-physical systems. The complexity of cyber-physical systems developing and deploying requires new approaches to its modelling and design. Presents results in the field of modelling technologies that leverage the exploitation of artificial intelligence, including artificial general intelligence (AGI) and weak artificial intelligence. Provides scientific, practical, and methodological approaches based on bio-inspired methods, fuzzy models and algorithms, predictive modelling, computer vision and image processing. The target audience of the book are practitioners, enterprises representatives, scientists, PhD and Master students who perform scientific research or applications of intelligent models and algorithms in cyber-physical systems for various domains.

✦ Table of Contents

Preface
Contents
Bio-inspired Modelling
An Extension of the Neural Additive Model for Uncertainty Explanation of Machine Learning Survival Models
1 Introduction
2 Survival Models
3 Uncertainty and Certainty Measures of Survival Models
4 NAM and Its Use for Uncertainty Explanation
5 Numerical Experiments with Real Data
6 Conclusion
References
Biocomputation Using Molecular Agents Moving in Microfluidic Channel Networks: An Alternative Platform for Information Technology
1 Introduction
1.1 Alternative Computing Approaches
1.2 Computational Complexity and Energy Considerations
2 Fabrication of Network-Based Biocomputation
2.1 Manufacturing Paradigma
2.2 Architectural Elements and Fabrication Tools
2.3 Integrated Manufacturing Scheme for a Functional NBC Network Including Microfluidic Channels, Junctions and Detection of Agents
2.4 Integrated Electronic Detectors in Microfluidic Channels Actuated by Varying Ionicity
3 Summary and Outlook
References
Artificial Immune Systems—Models and Applications
1 Introduction
2 Basic Models
3 Clonal Selection Algorithm
4 Negative Selection Algorithm
5 Positive Selection
6 Immune Network Models
7 The Danger Theory
8 Comparison of the Main Models
9 Applications
10 Conclusion
References
The Analysis of Technology Development Trends Based on the Network Semantic Structure “Subject-Action-Object”
1 Introduction
2 Algorithm for Parsing English-Language Patents
3 The Subject-Action-Object Extraction Algorithm (SAO)
4 Conclusion
References
Machine Learning of Diagnostic Neural Network for Railway Track Monitoring
1 Background
2 Rail Track Monitoring System
3 Convolutional Neural Network
4 Balancing Training Dataset
5 Training Dataset Quality Analysis
6 Diagnostic System Tests and Neural Network Accuracy Assessment
7 Conclusion
References
Neural Network Analysis of Electromagnetic Field Effect on the Thermofluctuation Characteristics Measurement in the Power Cable Insulating Materials
1 Introduction
2 Review of Literary Sources
3 Permissible Rated Cables Current Load
4 Development of the Thermal Model of the XLPE Cable Line
5 The Signal Noise Removal Procedure
6 Description of the Architecture, Parameters and Choice of ANN
7 Experimental Research Results
8 Conclusion
References
Fuzzy Models and Algorithms
Algorithm for Configuring Sugeno-Type Fuzzy Inference Systems Based on the Nearest Neighbor Method for Use in Cyber-Physical Systems
1 Introduction
2 Problem Statement
3 Formal Description of the Algorithm
4 Research of the Effectiveness of the Developed Algorithm
5 Conclusion
References
Construction of Membership Functions for Fuzzy Management of Security Information and Events
1 Introduction
2 Related Work
3 Theoretical Part
4 Methodological Part and Discussion
5 Conclusion
References
Building a Fuzzy Model to Automate the Search for Product Names in Procurement Documentation for a Shipbuilding Plant
1 Introduction
2 Problem Statement
3 Method for Solving the Problem
4 Preparation of Initial Data
5 The Results of Tsukamoto’s Fuzzy Inference Algorithm
6 Conclusion
References
Synthesis of a Neuro-Fuzzy Model in the Structure of an Automatic Rolling Stock Control System
1 Introduction
2 Structure of the Automatic Train Operation System (ATO)
3 Purpose of the Research
4 Building a Fuzzy Neural Network
5 Results of Computational Experiments
6 Conclusions
References
Predictive Modelling
Predicting the Equipment Useful Lifetime Based on the Deep Neural Networks
1 Introduction
2 Materials and Methods
3 Interpretation and Discussion of Research Results
4 Conclusion
References
Operational Model Predictive Control on the Example of the Stabilization Process for Hydro Treatment of Oil Fractions
1 Introduction
2 Formulation of the Problem
3 Regression Model
4 Creating a Simulation Model
5 Solution of the Optimal MPC Problem
6 Modelling
7 Conclusion
References
The Models for Predictive Maintenance of Robotic Agricultural Vehicles
1 Introduction
2 Related Work
3 The Diagnostic and Technical Condition Prediction System
4 Simulation Model on the Timed Colored Petri Nets
5 Experiments and Discussion
6 Development of the Hierarchy of the Maintenance and Repair Simulation Model
7 Conclusion
References
Development of a Cyber-Physical System for Neurofuzzy Prediction of the Concentration of the Contained Prime During Transportation of Oil Wells Emulsion
1 Introduction
2 Statement of the Problem of Obtaining Information Method About an Object and Technological Process
3 The Concept of Development and Integration of New Neural Fuzzy System Models for Parameters Prediction and Identification
4 Conclusion
References
Non-life Insurance Reserve Prediction Using LightGBM Classification and Regression Models Ensemble
1 Introduction
2 Literature Review
3 Research Methods
3.1 Data
3.2 Algorithm
4 Results and Discussion
References
Machine Learning-Based Cyber-Physical Systems for Forecasting Short-Term State of Unstable Systems
1 Introduction
2 Problem Formalization and Precedent Forecasting Algorithm
3 Specifics of Algorithm and Precedent Forecasting Implementation
4 Numerical Studies and an Example Implementation of Precedent Forecasting
5 Comparison of Precedent Forecast Implementations
6 Conclusion
References
Computer Vision and Image Processing
Application of Computer Vision Tools to Create a System for Monitoring the Work of Ground Equipment in Open Pits of Gold Mining Enterprises
1 Introduction
2 Methods and Materials
2.1 Research Technology of Mining
2.2 Computer Vision Tasks
2.3 Materials
3 Results
3.1 Learning Outcomes of the Neural Network
3.2 Accuracy Metric Results
3.3 Results of Model Operation on the Test Site
4 Discussion
References
Framework for Biometric User Authentication Based on a Dynamic Handwritten Signature
1 Introduction
2 Background
2.1 Characteristics of a Dynamic Handwritten Signature
2.2 Basic Challenges
3 The Architecture of the Developed Framework
4 Experimental Research
4.1 Selection of the Parameters
4.2 Comparison with Other Algorithms
5 Conclusion
References
Image Compression Based on the Significance Analysis of the Wavelet Transform Coefficients Using the Energy Feature Model
1 Introduction
2 Wavelet Transform of Images
3 Model of Energy Features of the Image
4 Image Compression Based on the Energy Feature Model
5 Conclusion
References
Construction of a Fuzzy Model for Contour Selection
1 Introduction
2 Building a Fuzzy Inference Model
3 Reduction of the Knowledge Base of the Fuzzy Model of Contour Selection
4 Determination of the Parameters of the Membership Functions of a Fuzzy Model
5 Conclusion
References
Identification of Vehicle Trajectory Anomalies on Streaming Video
1 Introduction
2 Background
3 Suggested Approach
3.1 Trajectories Extraction
3.2 Trajectories Thinning
3.3 Trajectories Comparison
3.4 Trajectories Clustering
3.5 Trajectory Classification
4 The Framework for Identification of vehicle’s Trajectory Anomalies on Streaming Video
5 Evaluation
6 Conclusion
References
Analyzing the Content of Business Documents Recognized with a Large Number of Errors Using Modified Levenshtein Distance
1 Introduction
2 Relevance
3 Problem Statement
4 Modified Levenshtein Distance
5 Results of Experiments
6 Conclusion
References