Development On Fire Burnout Point Detector Of Waste Incinerator Based On Embedded

At present, waste incineration has become the main form of waste disposal. The system of combustion control is the core of the whole control of waste incineration in waste incineration. And in the combustion control system, the focus is on the use of flame burning ember point position information to control the incinerator grate speed. At this stage, the flame burning ember point detection system is mainly composed of an industrial camera and image acquisition card and IPC, but this system still has some disadvantages such as poor real-time, big dissipation of power, and poor flexibility. In view of the above shortcomings, the paper design the flame burn ember point detector based on embedded technology. The mainly job are as the following:Firstly, this article studies the process of waste incinerator and puts forward the design requirements of flame burning ember point detector which based on the existing flame burning ember point detection scheme. According to the design requirements, the paper discusses hardware design scheme of the detector and proposes a hardware system architecture based on ARM+DSP, the hardware system scheme which is based on OMAPL138 processor is determined. Then the circuit design of each module is carried out based on hardware scheme, including main processor module, image acquisition module, memory modules, communication modules and power module. In addition, according to the characteristics of the hardware platform, a software system scheme of the detector is proposed. Linux is selected as the operating system of ARM processor, equipped with device driver and application software; DSP/BIOS is selected as the operating system of DSP processor, controlling image acquisition and processing.Secondly, the paper puts forward the method to obtain the position information of flame burning ember point based on flame burning image. In order to improve the adaptability of the algorithm, the flame images of different combustion states are selected. Since image is usually corrupted by the noise in its acquisition and transmission, the mean filter and median filter are respectively used to deal with denoising. Through the analysis of the results of the filter, the median filter is determined as an image denoising method. The bimodal method, the minimum error method and the Ostu method is respectively used for image segmentation. Through the comparison of segmentation results, the Ostu method is selected as image segmentation method. Because the image after segmentation carries thorns and other defects, the mathematical morphology is used to process segmented image. Then on basis of this, the paper proposes the method of detecting flame burning ember point and the method of getting the position of flame burning ember point.Finally, the paper introduces realization method of software system. By using QT software, the graphical user interface for embedded operating system is developed. Through this interface the user can view the information of flame burning ember point, the detector’s operating status and set the related parameters.