基于并行处理的熟料强度预测的研究与实现-计算机应用技术专业论文.docxVIP

基于并行处理的熟料强度预测的研究与实现 基于并行处理的熟料强度预测的研究与实现 Abstract We analyze the X-ray images ofthe cement clinker by the high performance computer DAWNING 3000 in this paper,and then distill the proportion of the cement clinker composition，and use neural network and intelligent technology to establish relational model for cement microscopic structure parameter and mechanics performance(28 days compressive strength)．Then we can achieve the compressive strength of cement clinker real-time and exact． It is the springboard to analyze the X-ray images of the cement clinker,the parallel image processing，the research ofthe parallel intelligent algorithm is the key technology in this topic．Therefore，in this topic，we should first make the image processing algorithm parallel，and make Sure the content of four major mineral phases from the X-ray images of the cement clinker，in this foundation，we can achieves the compressive s仃ength ofcement clinker by parallel intelligent method． The contents are listed as follow： 1．The important meaning and developing state of analyzing the cement material， as well aS the necessity of analysis for high performance of cement material． and points out the development ofparallel processing at present． 2． We analyze several kind of quite popular parallel computation models——the LogP model and the BSP model，the parallel algorithm is designed on the parallel computation model，in fact,the parallel algorithm is the bridge between the hardware and software． 3． We carries on the parallel processing to several kind of X-ray images which includes several principal element．and select appropriate parallel image processing algorithm to process them．we discuss the filter algorithm，and make the median filtering parallel，and analyze the parallel algorithm deeply． We make unidimensional Otsu algorithm two·dimensional Otsu algorithm by using the union histogram of primitive image and its smooth image，and we make this algorithm parallel． 1I 济南大学硕士论文 济南大学硕士论文 According as what we