智能化大功率高速流程离心泵关键技术基础研究

Because on-line monitoring, fault diagnose and safety prediction is the guarantee of reliable and stable operation for intelligent high-power high-speed process centrifugal pump, it is necessary to carry out systematic and deep research on inner flow simulation, rotor dynamic characteristics and fault diagnose, including three key scientific problems, the first is the simulation of characteristics in the pump at high speed; the second is influence mechanism of fluid exciting force on rotor dynamic characteristics, and the last is rotor vibration fault diagnosis model. However, it is not enough for the basic research on key scientific problems of high-power high-speed process centrifugal pump.. The purpose of the project is to carry out basic research on the key scientific problems of intelligent high-power high-speed process centrifugal pump from theoretical analysis, numerical simulation and experiment. The high precise flow field information in centrifugal pump is obtained under all operating conditions by using the dynamic subgrid stress model with the Helicity constraints which is established according to weak compressible fluid.. The influence mechanisms of fluid exciting force on rotor dynamic characteristics are analyzed by using the calculation methods of rotor dynamic characteristics which is presented by considering the main flow field and the gap flow field. . Therefore, on-line monitoring, operation judgment and safety warning can be achieved by using the rotor vibration fault diagnosis model which is established based on fluid dynamics and rotor dynamics. This will provide theoretical and technical support for the formation of a more perfect design method of intelligent high-power high-speed process centrifugal pump.

对大功率高速流程离心泵进行在线监测、故障诊断和安全预警，即提高智能化水平是实现可靠运行的关键，其前提需要对内部全流场流动、转子动力特性和故障诊断开展系统深入的研究，其中涉及三个关键科学问题：一是高转速下泵内流场的非定常流动特性计算，二是流体激振力对转子系统动力学特性的影响，三是转子振动故障诊断模型。但目前针对这三个关键科学问题的系统研究不够充分。. 本项目拟从理论分析、数值模拟和实验三方面对大功率高速流程离心泵开展关键技术基础研究。建立带螺旋度约束动态亚格子模式的泵内弱可压流体数值模拟方法，用大涡模拟方法获得离心泵全流量工况、高精度的流场信息；建立考虑主流场和间隙流场激振力的转子系统动力特性计算方法，获取转子系统动力特性；建立基于流场和转子动力特性的故障诊断模型，实现泵的在线监测、运行判断和安全预警，为形成较为完善的智能化大功率高速流程离心泵设计方法提供理论支持和技术支撑。

大功率高速多级离心泵在航天、核电、石油化工等领域至关重要，基于内部全流场流动、转子动力特性和故障诊断开展系统的研究对此类泵型进行在线监测、故障诊断和安全预警，即提高智能化水平是实现可靠运行的关键。.项目围绕高转速下泵内流场的非定常流动特性计算、流体激振力对转子系统动力学特性的影响及转子振动故障诊断及在线监测开展了深入研究。在泵内流场的非定常流动特性计算方面，通过引入螺旋度约束构建非线性亚格子模型，并针对RANS湍流模型进行旋转曲率及弱可压缩性修正，最终完成了带螺旋度约束动态亚格子模式的泵内弱可压流体数值模拟方法，用大涡模拟方法获得离心泵全流量工况、高精度的流场信息。其次，在考虑流体激振力的转子动力学特性计算方面，基于全流场计算结果构建了主流场激振力模型并修正了摄动法下的间隙流场速度与压力边界，结合构建了四方程的间隙流场控制方程组及其求解方法，提高了间隙内流体激振力及其等效动力学特性的计算精度及离心泵转子系统动力学特性的计算精度，获得了流体激振力对转子动力学响应特性的影响。最后在离心泵转子振动故障诊断及在线监测方面，建立了基于流场和转子动力特性的自编码及变分自编码的故障诊断模型，实现了4台工程样泵的在线监测、运行判断和安全预警平台的搭建与应用。.研究成果完善了大功率高速多级离心泵设计理论、运行及故障诊断研究，现已用于多款大功率流程离心泵的性能预测和设计开发中，取得良好效果，极大提高了设计效率，降低实验、生产和运行、维护成本。