摘要
金氰化浸出过程中,动力学反应速度的不可测量给模型辨识带来了困难,而由浓度等可测变量估计动力学反应速度是一个不适定逆问题,极易将浓度测量数据中的噪声放大,造成估计结果不准确.采用Tikhonov正则化方法,去除了复杂的动力学反应模型,利用浓度、流量等测量数据和物质守恒方程来辨识金和氰根离子的动力学反应速度,有效地减少了测量数据中误差的传递.实验结果表明,该方法不但能很好地跟踪动力学反应速度的变化趋势,抑制噪声的放大,而且易于实现.与有限差分、平滑滤波估计方法相比,其估计效果明显得到了改善.
In the gold cyanidation leaching process, the immeasurability of kinetic reaction rate brings difficulties of the model identification, and estimating rate by concentration data is an ill- posed inverse problem. The noise in the measured concentration data will be amplified arbitrarily leading to unreliable results. The Tikhonov regularization method was used to estimate the kinetics reaction rate of gold and cyanide ion only using concentration, flux data and mass conservation equations without establishing any kinetics reaction model, which reduces the propagation of errors in the measured data effectively. The experimental results showed that the proposed method is easier to implement and can better track the kinetics reaction rate trends, and the method can also restrain the amplification of noise. By comparison with the methods of finite difference and smooth filter, it has better estimate performance than other common methods.
出处
《东北大学学报(自然科学版)》
EI
CAS
CSCD
北大核心
2012年第11期1538-1541,共4页
Journal of Northeastern University(Natural Science)
基金
国家高技术研究发展计划项目(2011AA060204)
中央高校基本科研业务费专项资金资助项目(N110304006)