Optimizing of phase plan, sequence and signal timing based on flower pollination algorithm for signalized intersections
- PDF / 1,751,607 Bytes
- 17 Pages / 595.276 x 790.866 pts Page_size
- 37 Downloads / 217 Views
(0123456789().,-volV)(0123456789().,-volV)
METHODOLOGIES AND APPLICATION
Optimizing of phase plan, sequence and signal timing based on flower pollination algorithm for signalized intersections Ersin Korkmaz1
•
Ali Payıdar Akgu¨ngo¨r1
Springer-Verlag GmbH Germany, part of Springer Nature 2020
Abstract The purpose of this study is to develop a control system that optimizes the phase plan, sequence and signal timing using the flower pollination algorithm (FPA). At the same time, it is aimed to improve the fixed-time control system with the optimum cycle length search approach based on the differential evolution algorithm. The applicability and performances of these two control systems were examined in 15 different traffic situations according to 4 different intersection geometries. Fixed-time and optimized fuzzy logic traffic controller (FLC) developed by Dogan were used as the reference control systems in performance comparison. The optimum cycle length search system can achieve approximately 18% improvement over the fixed-time system, but showed lower performance than the FPA and FLC control systems. The FPA system has proven its applicability by achieving the best performance with about a 30% improvement compared to the fixed-time system and about 3% improvement compared to the FLC system. The FPA approach, which has a fast and effective performance, has been found to be an alternative method for intersection control, and it is foreseen that it can increase the intersection capacity and reduce the negative effects such as delay and fuel consumption. Keywords Traffic signal controller Flower pollination algorithm Fuzzy logic Differential evolution algorithm Phase and signal optimization
1 Introduction Signal control is an important concept in traffic engineering and allows to increase capacity without changing intersection geometry. For this reason, ensuring minimum delay and maximum traffic safety are the basic principle of optimum intersection control. The increase in travel demand and in the number of vehicles leads to the problem of traffic congestion especially in large cities, and traditional intersection control techniques such as fixed-time applications are insufficient to overcome this problem. Thus, many researchers have adopted different control approaches and have carried out studies on the application of techniques based on artificial intelligence in intersection control.
Communicated by V. Loia. & Ersin Korkmaz [email protected] 1
Department of Civil Engineering, Engineering Faculty, Kirikkale University, 71451 Kirikkale, Turkey
The artificial intelligence approach is one of the important issues that have been studied and developed for a long time to overcome deficiencies of traditional techniques. Many algorithms such as Krill Herd, Swarm optimization, Artificial Bee Colony, etc., especially inspired by nature and based on the flawless properties of biological systems, have been developed (Yang 2010). These algorithms have been successfully applied in many different engineeri
Data Loading...