Abstract:Proportional-integral-derivative (PID) controllers are undoubtedly the most employed controllers in industry, and they have significantly contributed to the impact of control systems in society. This book is a very welcome addition to the PID control literature as it aims to reduce the gap between theory and practice and foster technology transfer from academia to industry. Indeed, the topics included in the book are useful for students and practitioners in different fields to better understand how to design a PID controller and address the many issues that are faced in practice. The only prerequisites that are assumed of the reader are a basic knowledge of feedback control systems and dynamic systems (for example, transfer functions, frequency response, and state space models). The main feature of this book is the presence of many detailed worked examples that help the reader understand the underlying concepts. The reader is guided in developing Matlab scripts (also available from the book's associated website) that help to clearly outline the sequence of steps associated with each design technique. There are also study questions, provided after each section, designed to evaluate whether the reader fully understands the chapter topics. Numerical and simulation problems are also presented at the end of each chapter. Overall, the pedagogical approach followed in this text is appreciated and commendable. This book is suitable for those students and practitioners who have been exposed to a single control course and want to learn more about PID controllers or are asked to work with them in an industrial setting. It is worth emphasizing that, due to the inclusion of resonant controllers and a chapter dedicated to unmanned aerial vehicles, this book addresses general areas of application and is not only limited to process control problems. In summary, this book is a very good reference for those who want to deepen their knowledge about PID control without requiring-a strong background in control theory. Further advancements in the field can then be learned from other more advanced books once the basic concepts have been mastered.
What problem does this paper attempt to address?
This paper aims to solve the problem of detecting communities from social networks with positive and negative links (i.e., signed social networks, SNs). Specifically, most of the existing community detection algorithms are mainly designed for networks that only contain positive links and cannot effectively handle large - scale social networks that contain both positive and negative links simultaneously. Therefore, this paper proposes a similarity - based multi - objective evolutionary algorithm (Multiobjective Evolutionary Algorithm, MEA s - SN) to solve the problem of detecting communities in signed social networks.
The main contributions of the paper include:
1. **Extension of similarity measure**: Based on the social balance theory, the original similarity measure is extended to a signed similarity measure in order to be able to handle the influence of negative links.
2. **Design of two objective functions**: These two objective functions respectively reflect the natural contradiction between positive and negative links. One objective is to place all positive links inside the community, and the other objective is to place all negative links between the communities.
3. **Proposal of a direct and indirect combination representation method**: This representation method allows the algorithm to switch between different representations during the evolution process, so that it can detect separated and overlapping communities.
4. **Implementation of an algorithm based on the MOEA/D framework**: Using the Tchebycheff decomposition method, the multi - objective optimization problem is decomposed into multiple scalar optimization sub - problems, and these sub - problems are optimized simultaneously.
Through experimental verification on benchmark networks and large - scale synthetic networks, the results show that MEA s - SN has high effectiveness and efficiency in detecting community structures and performs well in various noise situations. Compared with the three existing algorithms (FEC, Louvain method and CSA HC - SN), MEA s - SN performs excellently. In addition, this algorithm does not need to know the information of the community structure in advance, such as the number of communities or a good initial partition, so it is more convenient to use.