Optimization Of Tunnel Illumination System Based On Intelligent Control

For the plan of the "7918" national highway network and the policy thatencouraging the development for the center and west regions, road construction hasbeing in a period of rapid development, the number of high-speed highway tunnel alsomaintained an increasing trend gesture. Compared with the railway tunnel, highwaytunnel has its own characteristics, the design of more complex, higher constructiondifficult. The speed of the vehicle when crossing the tunnel is fast, which leading to astrong brightness changes within the field of vision. In addition, the vehicle density,coupled with the effects of vehicle exhaust fumes inside the tunnel and other issues, aserious threat to road safety.Improve the tunnel lighting levels can give driver more discomfort, but it’llincrease the tunnel’s operating costs if increased the illumination blindly. Now, theresearchers focus on how to balance the relationship between illumination and thetunnel’s operating costs.In this thesis, comparing the present situation of tunnel lighting control researchand analyzing various types of tunnel lighting control method, summed up thedevelopment trend of intelligent lighting control. Then, studying the main factorswhich affecting tunnel lighting.After that, the brightness outside the tunnel, vehiclespeed through the tunnel and traffic tunnel lighting are the three parameters of thesystem to decide the brightness. Then, exploring two methods, which are used widelyin the field of intelligent control, fuzzy control and neural network application. Theintelligent dimming control not only meet the needs of the tunnel lighting, savingelectrical energy, but also by creating a more comfortable environment, greatlyenhanced the safety for the vehicle crossing the high tunnel.After analyzing and comparing the various control algorithms of fuzzy control,and a variety of neural network inference algorithm suitable for automatic control field, the paper using TS fuzzy RBF neural network control and joint control oftunnel lighting that fuzzy neural network control method, and its structure isequivalent to using type of binding mode.Simulating the system on the effect between the brightness outside the tunnel,traffic volume, speed and tunnel lighting by software MATLAB which are decided onthe norm named Specifications for Design of Ventilation and Lighting of HighwayTunnel. The simulation compares the performance of BP and RBF Network andneural networks and fuzzy neural network performance, verify this thesis, the controlalgorithm superiority. Finally, the simulation RBF Fuzzy Neural Network in thelighting control system, testing its performance, error rate, and network generalizationapproach.At the last of the paper is the summary of the full text, saying what had beingdone, you may say. Firstly, is the summary of the inadequate in the paper.Subsequently elaborated late outlook.The system lied on fuzzy logic dimming method, based on a small amount offeature points, trained by RBF neural network, generalization and approximation toobtain the optimum degree of tunnel lighting control system.