论文部分内容阅读
工业过程中的控制系统经常处于性能不佳的状态,若没有得到定期的维护会造成巨大的经济损失。带遗忘因子的线性回归(ILR)算法是一种基于过程数据的性能评估方法,它在线性回归方法的基础上引入遗忘因子,使评估结果更加具有实时性,为控制系统的维护提供了可靠的依据。该算法具有建模简单、计算快捷和适用于时变扰动模型的优点。本文运用该方法对乙烯裂解炉几个重要回路进行了性能评估,以此来表征系统整体的性能,并验证了算法的时变有效性。
Control systems in industrial processes are often in poor performance conditions, which can result in significant financial losses if not regularly maintained. The linear regression (ILR) algorithm with forgetting factor is a performance evaluation method based on process data. It introduces the forgetting factor based on the linear regression method to make the evaluation result more real-time and provides a reliable in accordance with. The algorithm has the advantages of simple modeling, fast calculation and suitable for the time-varying perturbation model. This method is used to evaluate the performance of several important circuits in ethylene cracker to characterize the overall performance of the system and to verify the time-varying effectiveness of the algorithm.