Zhang Fuyue, Li Dongjie, Rong Weibin, Yang Liu, Zhang Yu
School of Measurement and Communication, Harbin University of Science and Technology, Harbin 150080, China.
Heilongjiang Key Laboratory of Complex Intelligent System and Integration, Harbin University of Science and Technology, Harbin 150069, China.
Micromachines (Basel). 2021 Dec 20;12(12):1591. doi: 10.3390/mi12121591.
The rate and quality of microscale meniscus confined electrodeposition represent the key to micromanipulation based on electrochemistry and are extremely susceptible to the ambient relative humidity, electrolyte concentration, and applied voltage. To solve this problem, based on a neural network and genetic algorithm approach, this paper optimizes the process parameters of the microscale meniscus confined electrodeposition to achieve high-efficiency and -quality deposition. First, with the COMSOL Multiphysics, the influence factors of electrodeposition were analyzed and the range of high efficiency and quality electrodeposition parameters were discovered. Second, based on the back propagation (BP) neural network, the relationships between influence factors and the rate of microscale meniscus confined electrodeposition were established. Then, in order to achieve effective electrodeposition, the determined electrodeposition rate of 5 × 10 m/s was set as the target value, and the genetic algorithm was used to optimize each parameter. Finally, based on the optimization parameters obtained, we proceeded with simulations and experiments. The results indicate that the deposition rate maximum error is only 2.0% in experiments. The feasibility and accuracy of the method proposed in this paper were verified.
微尺度弯月面受限电沉积的速率和质量是基于电化学的微操纵的关键，并且极易受到环境相对湿度、电解质浓度和外加电压的影响。为了解决这个问题，本文基于神经网络和遗传算法方法，对微尺度弯月面受限电沉积的工艺参数进行优化，以实现高效和高质量的沉积。首先，利用COMSOL Multiphysics分析了电沉积的影响因素，并发现了高效和高质量电沉积参数的范围。其次，基于反向传播（BP）神经网络，建立了影响因素与微尺度弯月面受限电沉积速率之间的关系。然后，为了实现有效的电沉积，将确定的5×10 m/s的电沉积速率设置为目标值，并使用遗传算法对每个参数进行优化。最后，基于获得的优化参数，进行了模拟和实验。结果表明，实验中沉积速率的最大误差仅为2.0%。验证了本文提出方法的可行性和准确性。
