基于粒子滤波的自航耙吸挖泥船.doc
约72页DOC格式手机打开展开
基于粒子滤波的自航耙吸挖泥船,摘要近年来，中国的疏浚项目不断增加，疏浚工程也越来越大。疏浚作业的优化对提高效率、精度，以及节省劳动力都是非常重要的。其中，耙吸挖泥船在疏浚作业中的作用越来越重要。鉴于耙吸挖泥船的巨大作用，必须加强对其进行研究。尽管现代自航耙吸挖泥船上都安装了自动化控制系统，但是在挖泥船上至今还没有作业条件不确定的情形下（如土壤类型、...
内容介绍
此文档由会员 违规屏蔽12 发布
摘要
近年来，中国的疏浚项目不断增加，疏浚工程也越来越大。疏浚作业的优化对提高效率、精度，以及节省劳动力都是非常重要的。其中，耙吸挖泥船在疏浚作业中的作用越来越重要。鉴于耙吸挖泥船的巨大作用，必须加强对其进行研究。尽管现代自航耙吸挖泥船上都安装了自动化控制系统，但是在挖泥船上至今还没有作业条件不确定的情形下（如土壤类型、流速等）对疏浚性能进行优化的决策支持系统。因此，如何提高疏浚船舶设备水平，使疏浚产量最优化，达到提高挖泥船性能和效率的目的，也是国外疏浚行业积极研究的热点与内容。
本文即针对疏浚作业的效率受土壤类型和操作人员技术影响的情况，利用粒子滤波算法来估计自航耙吸挖泥船的溢流损失，采用实测工程数据进行了仿真，以此来提高挖泥船的生产效率、施工质量，为操作人员的施工提供了决策支持，达到降低生产成本，获取更高社会效益和经济效益的目的。
首先，本文从三个方面介绍了粒子滤波的基本理论，包括常用的预测滤波算法、递推贝叶斯估计以及蒙特卡洛分析等；在此基础上论述了粒子滤波的基本算法，讨论了滤波中的常见问题以及解决的方法。并将粒子群优化算法引入到粒子滤波中，作了进一步的研究。
其次，介绍了自航耙吸挖泥船的系统模型，重点分析了泥舱模型。针对耙头模型，通过动态建模给出了耙头吸入密度公式。之后详细论述了泥舱模型中的沉积过程，给出了溢流损失估计模型，并给出了相应的评估指标进行性能评估。考虑到泥沙颗粒大小对实际沉降速度的影响，对疏浚现场的泥沙粒径及实际沉降速度进行估计，以提高溢流损失估计的准确性。
最后，将粒子滤波算法应用在对溢流损失的估计上，通过MATLAB仿真说明了估计的正确性。在控制系统中，通过模糊控制器将粒子滤波器估计的溢流损失经被控对象反馈到系统的输入，通过改变系统的输入达到控制溢流损失的目的。结果表明，通过粒子滤波器估计的溢流损失能很好的为挖泥船操作人员提供控制决策，具有一定的实际意义。
关键词 自航耙吸挖泥船；粒子滤波；溢流损失估计；模糊控制；
Abstract
In recent years, China's dredging projects is escalation, and dredging project is bigger and bigger. The optimization of dredging operations is not only important to improve efficiency and precision, but also important to save labor. And Trailing Suction Hopper Dredger (TSHD) is playing a more and more important role in dredging operations. As the great effect of TSHD, we must strengthen the research on it. Although modern self-propelled TSHD has installed automatic control system, there is still no decision support system to optimize the performance of TSHD where the conditions are not sure, such as soil types and velocity. Therefore, how to improve the dredging ship equipment level to make the dredging output optimized and to enhance the performance and efficiency, is also the hot research spot and content of abroad.
As the dredging efficiency is affected by soil types and the technology of operating personnel, this paper uses particle filtering algorithm to estimate the overflow loss of TSHD, and take actual engineering data to complete the simulation. The purpose is to improve the efficiency and production, reduce the production cost and obtain higher social efficiency and the economic efficiency. Also, it provides the decision-making support for operator's construction.
From three aspects, this paper introduces the basic theory of particle filter, including the popular predictive filtering algorithm, the recursive Bayesian estimation and Monte Carlo analysis. On the basis of the above, the common filter problems and solving methods are discussed. Then, the particle swarm optimization algorithm is introduced to the particle filter, and made further research.
Secondly, the system models of TSHD are introduced, and emphatically analyses the hopper model. As for the head model, it gives the first inhaled formula of density through dynamic model. After that, the sediment process in hopper is detailed discussed. Then, the estimation model of overflow loss is given, as well as the corresponding eva luation indexes to make performance eva luation. Considering the influence of particle size on the actual settling velocity, the sediment particle size and actual settlement speed are estimated, to improve the accuracy of overflow loss that has been estimated.
Finally, the particle filtering algorithm is applied on the estimation of overflow loss. The correctness of the estimation was illustrated by MATLAB simulation results. In the control system, the overflow losses estimated by particle filterer were feedback to the system input through the fuzzy controller. By changing the system input to control the overflow loss. The results showed that the overflow loss estimated by particle filters could provide a very good decision-making for ship operators. And it has practical significance.
Key words： Trailing Suction Hopper Dredger; Particle Filter; the Estimation of Overflow Loss；Fuzzy Control;
目录
摘要 1
Abstract III
第1章 绪论 1
1.1 课题的研究背景、目的及意义 1
1.1.1 课题研究的背景 1
1.1.2 课题研究的目的和意义 4
1.2 国内外的研究现状及发展趋势 4
1.2.1 挖泥船溢流损失的研究现状 4
1.2.2 粒子滤波研究现状 5
1.2.3 粒子滤波的应用领域 6
1.3 本课题研究内容及安排 8
第2章 ..
近年来，中国的疏浚项目不断增加，疏浚工程也越来越大。疏浚作业的优化对提高效率、精度，以及节省劳动力都是非常重要的。其中，耙吸挖泥船在疏浚作业中的作用越来越重要。鉴于耙吸挖泥船的巨大作用，必须加强对其进行研究。尽管现代自航耙吸挖泥船上都安装了自动化控制系统，但是在挖泥船上至今还没有作业条件不确定的情形下（如土壤类型、流速等）对疏浚性能进行优化的决策支持系统。因此，如何提高疏浚船舶设备水平，使疏浚产量最优化，达到提高挖泥船性能和效率的目的，也是国外疏浚行业积极研究的热点与内容。
本文即针对疏浚作业的效率受土壤类型和操作人员技术影响的情况，利用粒子滤波算法来估计自航耙吸挖泥船的溢流损失，采用实测工程数据进行了仿真，以此来提高挖泥船的生产效率、施工质量，为操作人员的施工提供了决策支持，达到降低生产成本，获取更高社会效益和经济效益的目的。
首先，本文从三个方面介绍了粒子滤波的基本理论，包括常用的预测滤波算法、递推贝叶斯估计以及蒙特卡洛分析等；在此基础上论述了粒子滤波的基本算法，讨论了滤波中的常见问题以及解决的方法。并将粒子群优化算法引入到粒子滤波中，作了进一步的研究。
其次，介绍了自航耙吸挖泥船的系统模型，重点分析了泥舱模型。针对耙头模型，通过动态建模给出了耙头吸入密度公式。之后详细论述了泥舱模型中的沉积过程，给出了溢流损失估计模型，并给出了相应的评估指标进行性能评估。考虑到泥沙颗粒大小对实际沉降速度的影响，对疏浚现场的泥沙粒径及实际沉降速度进行估计，以提高溢流损失估计的准确性。
最后，将粒子滤波算法应用在对溢流损失的估计上，通过MATLAB仿真说明了估计的正确性。在控制系统中，通过模糊控制器将粒子滤波器估计的溢流损失经被控对象反馈到系统的输入，通过改变系统的输入达到控制溢流损失的目的。结果表明，通过粒子滤波器估计的溢流损失能很好的为挖泥船操作人员提供控制决策，具有一定的实际意义。
关键词 自航耙吸挖泥船；粒子滤波；溢流损失估计；模糊控制；
Abstract
In recent years, China's dredging projects is escalation, and dredging project is bigger and bigger. The optimization of dredging operations is not only important to improve efficiency and precision, but also important to save labor. And Trailing Suction Hopper Dredger (TSHD) is playing a more and more important role in dredging operations. As the great effect of TSHD, we must strengthen the research on it. Although modern self-propelled TSHD has installed automatic control system, there is still no decision support system to optimize the performance of TSHD where the conditions are not sure, such as soil types and velocity. Therefore, how to improve the dredging ship equipment level to make the dredging output optimized and to enhance the performance and efficiency, is also the hot research spot and content of abroad.
As the dredging efficiency is affected by soil types and the technology of operating personnel, this paper uses particle filtering algorithm to estimate the overflow loss of TSHD, and take actual engineering data to complete the simulation. The purpose is to improve the efficiency and production, reduce the production cost and obtain higher social efficiency and the economic efficiency. Also, it provides the decision-making support for operator's construction.
From three aspects, this paper introduces the basic theory of particle filter, including the popular predictive filtering algorithm, the recursive Bayesian estimation and Monte Carlo analysis. On the basis of the above, the common filter problems and solving methods are discussed. Then, the particle swarm optimization algorithm is introduced to the particle filter, and made further research.
Secondly, the system models of TSHD are introduced, and emphatically analyses the hopper model. As for the head model, it gives the first inhaled formula of density through dynamic model. After that, the sediment process in hopper is detailed discussed. Then, the estimation model of overflow loss is given, as well as the corresponding eva luation indexes to make performance eva luation. Considering the influence of particle size on the actual settling velocity, the sediment particle size and actual settlement speed are estimated, to improve the accuracy of overflow loss that has been estimated.
Finally, the particle filtering algorithm is applied on the estimation of overflow loss. The correctness of the estimation was illustrated by MATLAB simulation results. In the control system, the overflow losses estimated by particle filterer were feedback to the system input through the fuzzy controller. By changing the system input to control the overflow loss. The results showed that the overflow loss estimated by particle filters could provide a very good decision-making for ship operators. And it has practical significance.
Key words： Trailing Suction Hopper Dredger; Particle Filter; the Estimation of Overflow Loss；Fuzzy Control;
目录
摘要 1
Abstract III
第1章 绪论 1
1.1 课题的研究背景、目的及意义 1
1.1.1 课题研究的背景 1
1.1.2 课题研究的目的和意义 4
1.2 国内外的研究现状及发展趋势 4
1.2.1 挖泥船溢流损失的研究现状 4
1.2.2 粒子滤波研究现状 5
1.2.3 粒子滤波的应用领域 6
1.3 本课题研究内容及安排 8
第2章 ..
