气动调节阀的故障诊断及其粘滞故障补偿

发布时间：2018-06-13 01:43

本文选题：气动调节阀 + 故障诊断　；参考：《华东理工大学》2015年硕士论文

【摘要】：气动调节阀是过程工业的控制回路中使用最为广泛的一类执行器件,它的完好运行是回路良好控制性能的保证。然而随着控制回路运行时间的增长,阀门就会慢慢的出现各种故障。针对弹簧膜片式气动阀门故障的检测和分类问题,提出了一种基于累计残差设置阈值的阀门故障检测方法,在故障检测后,提出利用粒子群优化BP神经网络的算法来实现故障的分类。利用所提方法在DAMADICS仿真平台上对阀门故障情况和故障种类进行了应用研究,最后通过某工厂的实际阀门故障数据进行验证。仿真分析和工厂实际数据的验证结果都表明该检测算法能够有效地检测出阀门是否出现故障,优化后的分类算法能达到较高的分类准确率,较传统的BP神经网络分类算法准确率有提高。针对阀门的粘滞故障,本文提出了利用T-S型模糊控制器代替传统的PI控制器来消除阀门粘滞所引起的振荡。该控制器利用阀门粘滞时被控对象的状态信息与控制器输出变化率之间的关系来构建模糊控制的规则,通过对传统PI控制器的积分系数进行修正使得阀门快速移出粘滞区,最终实现对粘滞的补偿。将该算法应用在实际的液位控制回路中,实验结果表明该控制器的补偿效果明显,且能够适应不同的设定值,具有一定的鲁棒性。
[Abstract]:The pneumatic regulating valve is the most widely used kind of actuator in the control loop of the process industry. Its perfect operation is the guarantee of the good control performance of the loop. However, with the increase of the running time of the control loop, the valve will slowly appear all kinds of faults. A method of valve fault detection based on accumulative residual setting threshold is proposed. After the fault detection, the algorithm of particle swarm optimization (PSO) optimization BP neural network is proposed to classify the faults. The application of the proposed method to the valve failure and the type of fault on the DAMADICS simulation platform is carried out, and the actual valve of a factory is finally passed. The simulation analysis and the actual data of the factory verify that the detection algorithm can effectively detect the valve failure, and the optimized classification algorithm can achieve a higher classification accuracy. Compared with the traditional BP neural network classification algorithm, the accuracy rate is improved. In this paper, the valve malfunction is proposed in this paper. The T-S type fuzzy controller is used to replace the traditional PI controller to eliminate the oscillation caused by the valve viscosity. The controller uses the relationship between the state information of the controlled object and the output change rate of the controller to construct the rules of the fuzzy control by using the valve's viscosity. The valve is modified to make the valve through the correction of the integral coefficient of the traditional PI controller. The algorithm is applied to the actual liquid level control loop. The experimental results show that the compensation effect of the controller is obvious, and it can adapt to different set values and has a certain robustness.
【学位授予单位】：华东理工大学
【学位级别】：硕士
【学位授予年份】：2015
【分类号】：TH138.52

【参考文献】