论文部分内容阅读
针对复杂工业过程数据分布复杂的情况,提出了一种基于IC-SVDD(independent component-support vector data description)的工业过程故障监测方法。由于实际工业过程数据存在非线性和非高斯性问题,为了解决这两个同时存在的问题,采用IC-SVDD算法对数据进行处理。首先,利用独立成分分析算法对工业过程数据进行ICA分解,寻找一个分离矩阵W,实现分离原始数据,通过W的线性变化,可以将独立主元从混合信号中分离出来。然后,把提取出来的数据利用SVDD算法进行数据重构,进而构建新的统计量和统计限。最后,对Tennessee Eastman(TE)过程进行仿真,实验结果验证了该方法的可行性和有效性。
Aiming at the complicated distribution of complex industrial process data, a method of industrial process fault monitoring based on IC-SVDD (independent component-support vector data description) is proposed. Due to the nonlinear and non-Gaussian nature of the actual industrial process data, IC-SVDD algorithm is used to process the data in order to solve these two simultaneous problems. First, ICA decomposition of industrial process data is carried out by using independent component analysis algorithm to find a separation matrix W, which can separate the original data. Through the linear change of W, the independent principal component can be separated from the mixed signal. Then, the extracted data using SVDD algorithm for data reconstruction, and then build new statistics and statistical limits. Finally, the Tennessee Eastman (TE) process is simulated and the experimental results verify the feasibility and effectiveness of the proposed method.