(1. 中南大學(xué) 物理與電子學(xué)院，長沙410083；
2. 湖南力得爾智能科技有限公司，長沙410006)
摘 要: 針對(duì)目前鋁電解行業(yè)對(duì)于槽似在電阻的采集不夠準(zhǔn)確并且延時(shí)較高的問題，本文提出一種基于卡爾曼濾波的區(qū)間式槽似在電阻采集算法。該算法以卡爾曼濾波為基礎(chǔ)，用預(yù)測值與采樣值的均方差表征它們的高斯白噪聲功率，使其能夠在電阻平穩(wěn)的狀態(tài)下有著較強(qiáng)的跟蹤性能；再結(jié)合一階慣性濾波的強(qiáng)濾波特性和卡爾曼濾波的強(qiáng)跟蹤優(yōu)勢，設(shè)置適用的濾波區(qū)間，確保組合算法在槽似在電阻波動(dòng)較大的情況下能夠?yàn)V除掉噪聲的影響，并在電解槽穩(wěn)定后能對(duì)槽似在電阻進(jìn)行快速收斂跟蹤。結(jié)果表明：與一階慣性濾波相比，改進(jìn)后的卡爾曼濾波在電阻平穩(wěn)的狀態(tài)下其均方根誤差減少50%，在電解槽反生針振和擺動(dòng)情況之后的收斂時(shí)間減少90%。
關(guān)鍵字: 鋁電解；電阻采集；卡爾曼濾波算法；一階慣性濾波
（1. School of Physics and Electronics, Central South University, Changsha 410083, China;
2. Hunan Leader Intelligent Technology Co., Ltd., Changsha 410083, China）
Abstract:Aiming at the current problems of the aluminum electrolytic cell''''s inaccurate collection of slot-like resistance and high delay, this paper proposed an interval slot-like resistance acquisition algorithm based on Kalman filtering. This algorithm is based on Kalman filtering and uses the mean squared error of predicted and sampled values to characterize their Gaussian white noise power, enabling it to have strong tracking performance in the state of stable cell resistance. Combined with the strength of first-order inertial filtering characteristics and the strong tracking advantage of Kalman filtering, an applicable filtering interval is set to ensure that the combined algorithm can filter out the influence of noise when the slot-like resistance fluctuates greatly, and can quickly converge the slot-like resistance after the electrolytic cell is on stable track. The simulation results show that compared with the first-order inertial filter, the improved Kalman filter reduces the root mean square error by 50% when the cell resistance is stable, and reduces the convergence time by 90% after the electrolytic cell regenerates needle vibration and swing.
Key words: aluminum electrolysis; resistance acquisition; Kalman filter algorithm; first-order inertial filtering
