一种智能家居照明系统的控制方法及装置研究

发布时间：2018-02-20 21:05

本文关键词： 智能家居照明系统数据融合模糊控制神经网络　出处：《河北工业大学》2014年硕士论文　论文类型：学位论文

【摘要】：智能家居作为一个迅速崛起的新型行业，，它拥有巨大的市场前景和令人惊讶的发展潜力，是21世纪最具投资价值的新锐项目。现有的智能家居控制系统的控制器的控制模式只能按预设的控制规则和方式单一“机械式”的以自动化方式反复运行，周围环境变化时，系统就会失控，人性化服务较差。照明系统作为智能家居组成部分，对智能家居的整体研究有重大意义，笔者研究学习了模糊控制和神经网络控制算法后，提出了一个拥有自我学习、自我分析判断、自我预测功能的智能家居照明控制方式，基于模糊数学理论对控制方式进行建模研制设计研究，将照明系统人性化智能化不再是遐想。以智能家居照明系统为研究对象，引入多传感器数据融合技术，对传感器的特征层进行融合，利用实时数据，在一定规则下进行数据的分析处理，获取被控制对象一致性的特征数据，来完成控制任务。对照明控制系统的整体框架进行了构造，根据智能家居室内特点，对光环境进行了分析和场景设置。将模糊控制和神经网络控制技术嵌入到照明系统的控制算法进行研究，对各自算法进行了公式分析推导，利用神经网络的自学习能力对光环境样本数据进行训练，来调整模糊控制推理系统的模糊规则和隶属度函数的参数，最终确定了基于Mamdani推理的BP神经网络控制方式，并运用模糊神经网络在Matlab中进行建模与仿真训练，使控制系统具有了自我学习、自我判断、自我预测的功能，实现智能家居照明系统的智能化、人性化服务功能。研制设计完成了基于STM32为智能家居照明系统的智能控制中心研制设计，为了降低设备功耗和研制设计成本，无线信息采集和传输模块以2.4G技术为背景，采用CC2530作为主控芯片对智能控制节点研制设计，通过MDK进行程序代码的编程与仿真。对调光电路、人体感应电路及测光强弱电路进行了硬件研制设计。软件设计引用了IEEE.802.15提出的Z-stack协议栈无线通讯程序设计，完成了无线灯光节点与主控制器之间的通信，实现了多任务的实时性。
[Abstract]:Smart home as a rapidly rising new industry, it has a huge market prospects and surprising development potential, is the most valuable new investment project in 21th century. The control mode of the controller of the existing intelligent home control system can only be operated repeatedly by a single "mechanical" mode according to the preset control rules and modes. When the surrounding environment changes, the system will lose control. As a part of smart home, lighting system is of great significance to the whole research of smart home. After studying fuzzy control and neural network control algorithm, the author puts forward a self-learning system. Self-analysis and judgment, self-prediction function of intelligent home lighting control mode, based on fuzzy mathematics theory to model the design and design of the control mode, lighting system is no longer artificial intelligent daydream. Taking the intelligent home lighting system as the research object, the multi-sensor data fusion technology is introduced to fuse the characteristic layer of the sensor, and the real-time data is used to analyze and process the data under certain rules. In order to complete the control task, the whole frame of the lighting control system is constructed, and according to the indoor characteristics of the intelligent home, the characteristic data of the controlled object are obtained to complete the control task. The fuzzy control and neural network control technology are embedded into the control algorithm of lighting system, and the formulas of their respective algorithms are analyzed and deduced. The self-learning ability of neural network is used to train the light environment sample data to adjust the fuzzy rules of fuzzy control inference system and the parameters of membership function. Finally, the control mode of BP neural network based on Mamdani reasoning is determined. Using fuzzy neural network to model and simulate in Matlab, the control system has the functions of self-learning, self-judgment and self-prediction, so that the intelligent lighting system of intelligent home can be realized. The design of intelligent control center based on STM32 is completed. In order to reduce device power consumption and design cost, wireless information acquisition and transmission module is based on 2.4G technology. The intelligent control node is developed with CC2530 as the main control chip, and the program code is programmed and simulated by MDK. The hardware design of human induction circuit and the circuit of measuring intensity and intensity are carried out. The software design uses the design of wireless communication program of Z-stack protocol stack proposed by IEEE.802.15 to complete the communication between the wireless light node and the main controller. Multi-task real-time is realized.
【学位授予单位】：河北工业大学
【学位级别】：硕士
【学位授予年份】：2014
【分类号】：TU855;TM923.0

【参考文献】