Agitator tanks are widely used in industrial fields. Improvement in their efficiency is critical to achieving high productivity. That is to say, an agitator tank system should have a short response time to produce a desired reagent with an accurate solution concentration and a moderate liquid level. Therefore, a noise-rejection zeroing dynamics (NRZD) model for the control of the agitator tank based on a neural-dynamics method with anti-noise performance is proposed in this paper. The solution concentration and the liquid level of the agitator tank synthesized by the NRZD model are able to converge to the desired trajectories polluted with different noises. Then, theoretical analyses on the convergence and anti-noise performance of the agitator tank system equipped with the NRZD model are presented. Furthermore, to verify the superiority of the agitator tank system equipped with the NRZD model, we perform tracking trajectories simulations on solution concentration and the liquid level of the agitator tank with different noises. Moreover, the simulation results verify that the NRZD model is more effective than the existing models in the reagent preparation process.

搅拌罐广泛用于工业领域。效率的提高对于实现高生产率至关重要。也就是说，搅拌罐系统应具有较短的响应时间，以产生具有准确溶液浓度和适中液位的所需试剂。因此，本文提出了一种基于神经动力学方法的具有抗噪性能的噪声抑制调零动力学模型（NRZD）。通过NRZD模型合成的搅拌器罐的溶液浓度和液位能够收敛到被不同噪声污染的所需轨迹。然后，对装有NRZD模型的搅拌罐系统的收敛性和抗噪性能进行了理论分析。此外，为了验证配备有NRZD模型的搅拌器槽系统的优越性，我们对具有不同噪声的搅拌器槽的溶液浓度和液位进行了跟踪轨迹模拟。此外，仿真结果证明，在试剂制备过程中，NRZD模型比现有模型更有效。