ELECTRIMACS 2022

Table of Contents

1. Optimisation in Complex Electrical Systems

   1. Frontmatter

   2. User Experience Inquiry to Specify COFFEE: A Collaborative Open Framework For Energy Engineering

      Sacha Hodencq, Fabrice Forest, Théo Carrano, Benoit Delinchant, Frédéric Wurtz

      The chapter delves into the challenges and opportunities of the energy transition, emphasizing the importance of open energy modelling. It introduces COFFEE, a collaborative platform that aims to make energy research accessible to various stakeholders. The platform's design is informed by a user experience inquiry, resulting in 12 recommendations to enhance its usability and effectiveness. The chapter also discusses the implementation strategy and future perspectives of COFFEE, highlighting its potential to foster innovation and virtuous energy behaviours.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

   3. Optimal Sizing of Tramway Electrical Infrastructures Using Genetic Algorithms

      Anass Boukir, Vincent Reinbold, Florence Ossart, Jean Bigeon, Paul-Louis Levy

      The chapter 'Optimal Sizing of Tramway Electrical Infrastructures Using Genetic Algorithms' delves into the critical challenge of sizing electrical infrastructure for tramways. It begins by highlighting the environmental and economic importance of electrifying urban public transport systems. The current manual approach to sizing electrical infrastructure, which involves trial-and-error and extensive simulation, is inefficient and often leads to oversized infrastructures. The chapter introduces a novel method using genetic algorithms to optimize the sizing process, considering both overall costs and voltage security margins. The studied system includes feeding substations, overhead transmission lines, rails, and trains, with a detailed electrical model simulated using an in-house developed tool. The optimization problem is formulated as a bi-objective function, balancing investment and energy costs with power supply quality. The chapter concludes with a test case demonstrating the effectiveness of the proposed method, offering a significant advancement over existing linear power flow models.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

   4. A Comparative Study of Existing Approaches for Modeling the Incident Irradiance on Bifacial Panels

      Soufiane Ghafiri, Maxime Darnon, Arnaud Davigny, João Pedro F.Trovão, Dhaker Abbes

      The chapter delves into the advantages of bifacial solar panels, which capture both front and rear irradiance, potentially increasing energy yield by up to 30%. It explores the key parameters affecting bifacial module performance, such as the bifaciality factor and bifaciality gain. The study compares three sophisticated modeling tools—Bifacial_radiance, Sandia View Factor Model, and Pvfactors—each with unique methodologies for estimating front and rear irradiance. The chapter concludes with a detailed comparison of these tools, highlighting their accuracy and computational efficiency, and recommends Pvfactors for real-time production prediction. This comparative study offers valuable insights into optimizing the performance of bifacial solar panels, making it a must-read for professionals seeking to enhance solar energy systems.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

   5. Self-Adaptive Construction Algorithm of a Surrogate Model for an Electric Powertrain Optimization

      Marvin Chauwin, Hamid Ben Ahmed, Melaine Desvaux, Damien Birolleau

      The chapter introduces a self-adaptive algorithm for constructing surrogate models to optimize electric powertrain systems. It addresses the complexity of multiphysics modeling, which includes mechanical, electromagnetic, and thermal principles. The algorithm aims to reduce computational time by using Kriging to estimate model outputs accurately. The method involves creating a sampling plan, computing the model on each sample, and optimizing the surrogate model parameters. The chapter also discusses the efficiency of Kriging and the use of Latin HyperCube for distributing samples effectively. Enrichment techniques, both OFF-Line and ON-Line, are explored to improve the accuracy of the surrogate models. The Sub-Latin HyperCube method is highlighted as a tool to add new samples efficiently, reducing the time required to achieve the desired accuracy. The chapter concludes by emphasizing the potential of these tools to create high-speed computing surrogate models, though further local accuracy improvements may be needed.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

   6. Optimization of Neural Network-Based Load Forecasting by Means of Whale Optimization Algorithm

      Pooya Valinataj Bahnemiri, Francesco Grimaccia, Sonia Leva, Marco Mussetta

      This chapter delves into the optimization of neural network-based load forecasting, specifically focusing on the use of the Whale Optimization Algorithm to enhance the performance of Echo State Networks. The Echo State Network, a type of recurrent neural network, is introduced as a powerful tool for handling time-series data in power systems. The Whale Optimization Algorithm, inspired by the hunting behavior of humpback whales, is employed to fine-tune the network parameters, resulting in improved forecasting accuracy. The chapter highlights the advantages of this approach over conventional methods and other computational intelligence techniques, providing a comprehensive overview of the implementation and benefits of this innovative solution for short-term load forecasting.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

2. Modelling and Simulation of Electrical Machines and Electromagnetic Devices

   1. Frontmatter

   2. Estimation of Steady-State Torque of Line Start Permanent Magnet Synchronous Motor Using Reluctance Network Approach

      Hamza Farooq, Nicolas Bracikowski, Patricio La Delfa, Michel Hecquet

      The chapter delves into the estimation of steady-state torque in Line Start Permanent Magnet Synchronous Motors (LSPMSMs) using a Reluctance Network Approach (RNA). It highlights the significant role of air-gap flux density, influenced by permanent magnets, in determining the torque characteristics. The study introduces a novel RNA model that accounts for notable rotor leakage flux components, including flux barriers, bridges, and bars, and compares linear and nonlinear approaches. The proposed method allows for rapid parametric analysis and optimization, making it a valuable tool for motor designers. The chapter also presents a detailed comparison with Finite Element Method (FEM) simulations, validating the accuracy of the RNA model. The findings offer insights into the early design stages of LSPMSMs, enabling the identification of rotor leakage flux and the optimization of motor performance.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text

   3. An Overview of High-Speed Axial Flux Permanent Magnets Synchronous Machines

      Hoda Taha, Georges Barakat, Yacine Amara, Mazen Ghandour

      The chapter delves into the advancements and challenges of high-speed axial flux permanent magnet synchronous machines, emphasizing their superior power density and efficiency compared to low-speed machines. It explores various machine topologies, mechanical stresses, and high-speed losses, while also discussing suitable materials and innovative design solutions. The chapter highlights the growing applications of these machines in industries requiring high precision and reliability, such as aerospace, automotive, and energy storage systems. It concludes by stressing the need for a multidisciplinary approach to optimize the performance of these machines.

      AI Generated

      This summary of the content was generated with the help of AI.

      Show original text