Research On Intelligent Home Lighting Control System Based On Neural Network

As an important part of human daily life,the role of lighting has changed from meeting the needs of visual light and shade at the beginning to creating a rich light environment to bring comfortable experience.In the process of lighting,there is a large amount of waste of resources.How to meet the needs of comfort and achieve energy conservation and environmental protection has become the focus of intelligent lighting research.Light environment and computer control algorithm are deeply studied,the application of neural network in the home lighting control system to realize the intelligent control of home lighting is also proposed in this dissertation.First,illumination and evenness of illumination are taken as the judgment basis of lighting control in the research,and the calculation method of indoor illumination is determined through the study of natural light and lamp light.It is difficult to describe by mathematical model since there are many factors affecting indoor light environment,Therefore,an indoor space model is established by DIALux to analyze the effect of weather,time and space position on indoor lighting,which provides a foundation for the study of lighting control algorithm.The intelligent control algorithm of neural network combined with fuzzy reasoning is determined after analysis.In the process of algorithm construction,BP neural network model is built to complete the setting of input layer,output layer,transfer function and relevant initial parameters.The light data changed with natural light and different scenes are obtained by simulating dimming in the spatial model.Then the neural network is trained in MATLAB to acquire the mean square error(MSE)results and regression line graph,and the network weights and thresholds of each layer are obtained.The constructing fuzzy inference machine is built with the activity time of each region as input and the eigenvalue of the target scene as output to judge the scene mode.The membership function and argument domain are determined,and fuzzy rules are formulated.Then,the intelligent home lighting control system is studied and the overall scheme of the system is determined.The design of environmental monitoring module is completed which achieves the measurement of illumination through the BH1750FVI digital light sensor and detects user activity area through the HC-SR501 human thermal release infrared sensor;the control module is used by STM32F107 processor to complete the control algorithm in the main controller;the wireless transmission module is used by the CC2530 ZigBee to realize the data transmission between controller,sensor and terminal lamp;the modulation module of the terminal lamp which converts the PWM signal into the driving current to drive the LED to realize the modulation of the lighting brightness by PT4115 step-down constant current source.Finally,the feasibility of the control algorithm is verified based on the simulation model,and the function of the system is tested in this dissertation.The experimental results show that the intelligent home lighting control system based on neural network can not only create a comfortable light environment,but also achieves effective energy saving.