1,721,003 research outputs found
Stochastic Model Predictive Control with Dynamic Chance Constraints
No full textThis work introduces a stochastic model predictive control scheme for dynamic chance constraints. We consider linear discrete-time systems affected by unbounded additive stochastic disturbance. To synthesize an optimal controller, we solve two subsequent stochastic optimization problems. The first problem concerns finding the maximal feasible probabilities of the dynamic chance constraints. After obtaining the probabilities, the second problem concerns finding an optimal controller using stochastic model predictive control. We solve both stochastic optimization problems by reformulating them into deterministic ones using probabilistic reachable tubes and constraint tightening. We prove that the developed algorithm is recursively feasible and yields closed-loop satisfaction of the dynamic chance constraints. In addition, we will introduce a novel implementation using zonotopes to describe the tightening analytically. Finally, we will end with an example illustrating the method's benefits.</p
Data-driven neural feedforward controller design for industrial linear motors
No full text\u3cp\u3eIn this paper we consider the problem of feedforward controller design for industrial linear motors. These motors are safety-critical high-precision mechatronics systems that pose stringent requirements on the feedforward design: safe and predictable behavior for the desired motion profiles, tracking performance within the 10μ m range in the presence of nonlinear friction and real-time implementation within the 1ms range. We investigate and compare several possibilities to design data-driven feedforward controllers using neural networks (NN) and we show that a two-step inverse estimation method is the most suitable approach, due to robustness to noisy data. We also show that basic knowledge about the system dynamics and the friction behavior can be exploited to design neural feedforward controllers with a simple structure, suitable for real-time implementation in industrial linear motors. The developed data-driven neural feedforward controllers are tested and compared with standard mass-acceleration feedforward and iterative learning controllers in realistic simulations.\u3c/p\u3
Efficient Implementation of Piecewise Quadratic Lyapunov Function Computations for Switched Linear Systems
Full text linkWe describe a linear programming (LP) problem to parameterize continuous and piecewise quadratic (CPQ) Lyapunov functions for switched linear systems. We discuss some algorithms and data-structures for its implementation in C++ and compare the computational efficiency of our implementation to an analogous implementation in MATLAB
Artificial intelligence for photovoltaic systems
Full text linkPhotovoltaic systems have gained an extraordinary popularity in the energy generation industry. Despite the benefits, photovoltaic systems still suffer from four main drawbacks, which include low conversion efficiency, intermittent power supply, high fabrication costs and the nonlinearity of the PV system output power. To overcome these issues, various optimization and control techniques have been proposed. However, many authors relied on classical techniques, which were based on intuitive, numerical or analytical methods. More efficient optimization strategies would enhance the performance of the PV systems and decrease the cost of the energy generated. In this chapter, we provide an overview of how Artificial Intelligence (AI) techniques can provide value to photovoltaic systems. Particular attention is devoted to three main areas: (1) Forecasting and modelling of meteorological data, (2) Basic modelling of solar cells and (3) Sizing of photovoltaic systems. This chapter will aim to provide a comparison between conventional techniques and the added benefits of using machine learning methods
Going Beyond Counting First Authors in Author Co-citation Analysis
Full text linkThe present study examines one of the fundamental aspects of author co-citation analysis (ACA) - the way co-citation
counts are defined. Co-citation counting provides the data on which all subsequent statistical analyses and mappings
are based, and we compare ACA results based on two different types of co-citation counting - the traditional type that
only counts the first one among a cited work's authors on the one hand and a non-traditional type that takes into
account the first 5 authors of a cited work on the other hand. Results indicate that the picture produced through this non-traditional author co-citation counting contains more coherent author groups and is therefore considerably clearer. However, this picture represents fewer specialties in the research field being studied than that produced through the traditional first-author co-citation counting when the same number of top-ranked authors is selected and analyzed. Reasons for these effects are discussed
Applied Computational Intelligence in Engineering and Information Technology: Revised and Selected Papers from the 6th IEEE International Symposium on Applied Computational Intelligence and Informatics SACI 2011
No full textThis book highlights the potential of getting benefits from various applications of computational intelligence techniques. The present book is structured such that to include a set of selected and extended papers from the 6th IEEE International Symposium on Applied Computational Intelligence and Informatics SACI 2011, held in Timisoara, Romania, from 19 to 21 May 2011. After a serious paper review performed by the Technical Program Committee only 116 submissions were accepted, leading to a paper acceptance ratio of 65 %. A further refinement was made after the symposium, based also on the assessment of the presentation quality. Concluding, this book includes the extended and revised versions of the very best papers of SACI 2011 and few invited papers authored by prominent specialists. The readers will benefit from gaining knowledge of the computational intelligence and on what problems can be solved in several areas; they will learn what kind of approaches is advised to use in order to solve these problems. A very important benefit for the readers is an understanding of what the major difficulties are and the cost-effective solutions to deal with them. This book will offer a convenient entry for researchers and engineers who intend to work in the important fields of computational intelligence.
Dynamic Prediction of Reblowing Necessity in BOF Steelmaking
No full textThe basic oxygen furnace (BOF) steelmaking process is the primary route for steel production but faces challenges in efficiency, emissions, and precise endpoint control. This study introduces a novel approach to dynamically predict the neces-sity of reblowing using time-series data from sensors, additive patterns, and static variables collected over a year of industrial production. By integrating time-series data and extracting statistical and frequency-domain features, we developed machine learning models, incorporating various sampling techniques to address class imbalance. We applied window encoding to process time-series information, enabling predictions from the start of the process through to the first sampling stage. The XGBoost model, coupled with synthetic sampling, demonstrated the highest performance. Our approach classifies heats as normal or abnormal based on the need for reblowing, providing valuable insights for BOF process control. The continuous improvement in prediction accuracy throughout the process highlights the effectiveness of this method, marking the first study to dynamically predict reblowing necessity in the BOF process and emphasizing the importance of process data in optimizing BOF operations.</p
- …
