盾构机密封舱土压预测及智能优化控制策略

发布时间：2018-01-24 10:08

本文关键词： 盾构土压平衡 LS-SVM 预测 PSO　出处：《大连理工大学》2015年硕士论文　论文类型：学位论文

【摘要】：世界经济的快速发展促进了城市化建设的大规模进行,然而随着建筑的日益密集以及地面空间的愈加有限,人们不得不将目光转向地下空间,因此,如今对地下空间的开发进入了一个前所未有的阶段。土压平衡式盾构机是一种挖掘地下隧道的专用大型施工设备,现已广泛应用于城市地铁隧道、资源开发、市政建设、海底隧道等地下工程建设,它以安全可靠、掘进速度快、劳动强度低以及对地层影响小等显著优点,逐渐成为大型地下工程施工的重要手段。盾构施工时开挖面不稳定易引起地表塌陷或隆起的灾难性事故。由于密封舱土压的变化情况与开挖面压力密切相关,因此精确预测及控制密封舱土压是有效预防开挖面压力失衡的关键技术之一。由于盾构施工是一个极其复杂的过程,工况突变和地质状况异常时有发生,而各掘进参数之间存在着强耦合性,难以建立精确的数学模型,目前的盾构掘进施工主要依靠人工经验操作完成,土压控制精度难以保证,灾难性事故很难从根本上避免。所以采用智能建模方法建立土压预测控制模型,对于实现掘进过程的智能控制,保证掘进施工安全,有着重要的理论和现实意义。为了建立土压预测模型,首先从机理上对土压与掘进参数的关系,特别是刀盘扭矩与土压的关系,做了详细分析,以此为基础确定了密封舱土压模型新的输入输出参数。然后,以密封舱承压隔板上有四个土压传感器的典型土压平衡盾构机为例,提出了基于最小二乘径向基核函数支持向量机的密封舱土压预测模型的智能建模方法,并采用粒子群优化算法对最小二乘径向基核函数支持向量机的两个参数寻优。最后,利用所提出的预测模型,以密封舱内土压预测值与设定值总体绝对偏差最小为优化指标,采用粒子群算法对影响密封舱土压变化的两个控制参数——螺旋输送机转速和推进速度进行在线优化,实现密封舱土压平衡的最优控制。结合现场施工数据对所提出的建模和优化控制方法进行了仿真对比分析,验证了模型的有效性和准确性,为实现密封舱土压的准确预测和自动控制,保证掘进施工过程安全提供了技术支持。
[Abstract]:The rapid development of the world economy has promoted the large-scale construction of urbanization. However, with the increasing density of buildings and the increasingly limited space on the ground, people have to focus on underground space, so. At present, the development of underground space has entered an unprecedented stage. Earth pressure balance shield machine is a kind of special underground tunnel construction equipment, which has been widely used in urban subway tunnel, resource development. Municipal construction, subsea tunnel and other underground engineering construction, it is safe and reliable, fast tunneling speed, low labor intensity and small impact on the formation and other prominent advantages. It has gradually become an important means of large underground engineering construction. The unstable excavation surface during shield construction is liable to cause catastrophic accidents of surface collapse or uplift. The variation of soil pressure in sealed chamber is closely related to excavating surface pressure. Therefore, accurate prediction and control of sealed chamber earth pressure is one of the key technologies to effectively prevent excavating surface pressure imbalance. Because shield construction is an extremely complex process, sudden changes in working conditions and abnormal geological conditions occur from time to time. However, there is strong coupling among the tunneling parameters, so it is difficult to establish accurate mathematical model. The current shield tunneling construction mainly depends on manual experience operation, and the precision of soil pressure control is difficult to guarantee. It is difficult to avoid the catastrophic accident fundamentally, so the intelligent modeling method is used to establish the predictive control model of earth pressure, which can realize the intelligent control of the tunneling process and ensure the safety of the tunneling construction. It has important theoretical and practical significance. In order to establish the prediction model of soil pressure, the relationship between soil pressure and tunneling parameters, especially the relationship between cutter head torque and soil pressure, is analyzed in detail. On this basis, the new input and output parameters of the earth pressure model of the sealed tank are determined. Then, the typical earth pressure balance shield machine with four earth pressure sensors on the pressure partition board of the sealed chamber is taken as an example. An intelligent modeling method for soil pressure prediction model of sealed cabin based on least square radial basis function kernel support vector machine is presented. Particle swarm optimization algorithm is used to optimize the two parameters of least squares radial basis function support vector machine. Finally, the proposed prediction model is used. The minimum absolute deviation between the soil pressure prediction value and the set value in the sealed cabin is taken as the optimization index. The particle swarm optimization (PSO) algorithm is used to optimize the speed and propulsion speed of the spiral conveyor, which affects the change of soil pressure of the sealed cabin. The optimal control of soil pressure balance in sealed cabin is realized. The simulation and contrast analysis of the proposed modeling and optimization control methods are carried out in combination with the field construction data, and the validity and accuracy of the model are verified. It provides technical support for accurate prediction and automatic control of soil pressure in sealed cabin and for the safety of tunneling construction process.
【学位授予单位】：大连理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：U455.39

【参考文献】