SIA OpenIR  > 信息服务与智能控制技术研究室
面向喷印工艺的液滴状态优化模型与智能控制方法研究
其他题名Research of Droplet Property Optimization Model and Intelligent Control Method for Ink-jet Printing Technology
何茂伟1,2
导师胡琨元
分类号TP373
关键词喷墨打印技术 压电式喷头 集总参数模型 液滴状态优化 液滴图像分割
索取号TP373/H32/2015
页数117页
学位专业机械电子工程
学位名称博士
2015-05-28
学位授予单位中国科学院沈阳自动化研究所
学位授予地点沈阳
作者部门信息服务与智能控制技术研究室
摘要喷墨打印技术的基本工作原理是根据计算机指令将油墨从微小的喷孔直接喷射至承载物的指定位置上,从而形成预制图形。随着打印喷头性能的逐步提升以及新型油墨材料的使用,其应用范围不仅仅局限于办公领域,已成功扩展到电子电路加工、智能纺织品、微电子机械系统制备等制造领域,呈现出加工精度高、速度快、低污染、低能耗等优势。 喷头作为打印系统的核心部件,其工作状态直接影响印刷的质量和打印效率。而喷头的工作状态具体反映在其喷射出的液滴体积、液滴速度、卫星液滴等方面,其中:液滴体积直接影响打印分辨率,还决定墨点的沉积厚度;液滴速度影响墨点位置偏差以及撞击基板的效果;卫星液滴不仅会影响打印的分辨率,还会造成墨点误连等现象。由此可见,实现喷头喷射液滴状态优化是保障打印系统稳定、可靠、精确运行的基础,这其中涉及到机理建模、参数优化与控制、图像检测等一系列问题,引起了专家和工程技术人员的广泛关注。 本文以压电式按需打印喷头为背景,在查阅和分析国内外相关领域研究的基础上,提出了改进型的集总参数模型,用于实现喷头工作状态模拟和参数优化;充分考虑基于模型的喷头状态优化控制中存在的不足,设计了基于检测的液滴喷射状态调整系统,并在此基础上实现了液滴状态闭环反馈控制。本文的具体研究工作概括如下: 1.在深入分析压电式按需打印喷头组成结构及工作原理的基础上,研究了适于模拟压电式按需打印喷头工作情况的集总参数模型(Lumped Element Modeling, LEM);针对原始集总参数模型仿真精度不高、动态跟踪性不强等缺点,引入多种压电陶瓷等效模型,分别对原始模型进行改进;并采用模型仿真与物理实验相结合的方式进行验证,结果证明了改进模型在液滴状态模拟精度上的优势。 2.在改进的压电式按需打印喷头集总参数模型基础上,采用遍历方式对驱动波形参数进行了低维度搜索;同时,针对多维驱动波形参数优化问题,引入人工蜂群算法,以克服遍历搜索效率低、适用性差等局限性。通过仿真实例证明上述方法的有效性。 3.针对液滴状态检测问题,采用OTSU多阈值分割技术提取液滴的图像信息,为克服基于遍历搜索的多阈值分割运行时间长这一缺陷,设计了具有分层结构的人工蜂群算法实现阈值搜索,通过仿真试验证明上述方法在求解液滴轮廓提取问题中具有收敛精度高,速度快的特点;并可在基础上,较为精确地计算出液滴速度、体积和卫星液滴。 4.针对模型方法中喷头内部参数存在误差、基于离线模拟难以实现实时调整的不足,设计了一种具有反馈结构的喷头驱动参数在线优化系统;在液滴喷射状态检测的基础上,分别采用人工蜂群算法、PSO算法、遗传算法等方法实现驱动波形参数优化,通过对多个液滴状态优化目标进行试验分析,结果表明系统的有效性。
其他摘要Ink-jet printing technology is a kind of advanced digital manufacture technology. Its basic working principle is that, according to the computer instruction, the droplets fired from tiny nozzles form a prefabricated graphics on the substrate. With the improvement of the work performance of ink-jet printing head, the application area of this technology is not limited in printing paper and has been successfully extended to industrial machining processing, electronic circuits, smart textiles, artificial skin preparation, microelectronics mechanical systems and other fields. This technology has so many advantages, such as high machining accuracy, high machining speed, no contact with the substrate, low pollution and low energy consumption. Printhead is the core of printing system. Its working status directly affects the manufacture quality of products and production efficiency. The working status of printhead reflects in the properties of ejected droplets (droplet size, droplet velocity, satellite droplets, etc.). Droplets with controllable volume affect not only the printing resolution but also the depositional thickness. Meanwhile, droplets with a high speed can remarkably improve the printing efficiency and the quality of impact with substrate. Moreover, the existence of the satellite droplets will not only influence the printing resolution, also can cause short circuit of conductive lines. Therefore, optimization of droplet property ejected by piezoelectric printhead guarantees the stability, reliability and accuracy of the printing system which are essential to the improvement of printing quality. This paper will give a full analyse and research on the structure and working principle of piezoelectric drop-on-demand (DoD) printheads. Based on it, several modified lumped element models (LEM) is constructed, which are employed to simulate and analyse the working status of piezoelectric printheads. Then, a research on the model-based optimization for nozzle’s jet state is made. Moreover, to overcome the shortcoming of model-based optimization for jet state, a droplet properties adjusting system based on detection is built, which is helpful for achieving droplet rapid real-time adjustment with a feedback loop. The main contents of this paper are generalized as follows: 1. With a full study of the structure and working principle of the piezoelectric drop-on-demand printheads, the lumped parameter model (LEM) is set up for simulating the working status of printhead. Due to the weakness of original LEM (low simulation accuracy and poor dynamic tracking performance), this paper brings in a variety of equivalent circuit models of piezoelectric ceramic to improve the performance of LEM. With the comparison of the results obtained by different modified LEMs, the improved model, which is suitable for this study, is determined. Combining with the experiment, the modification is helpful for improving the accuracy of simulating droplet properties. 2. On the basis of the modified LEM, an ergodic low-dimensional search for the parameters of high-frequency driving waveform is adopted. To overcome the limitations (low efficiency and low applicability), combining with the intelligent optimization method, this study makes an intelligent prediction for droplet properties. Through the test against the instance, the feasibility and effectiveness of this method is verified. 3. Although the method combined with LEM and the intelligent optimization algorithm can achieve the prediction of droplets properties, the disadvantages, such as long computation time, slow adjustment speed, and the dependency relationship between prediction effect and simulation accuracy, limit the adjustment effect. To overcome these shortcomings, a droplet properties adjusting system based on detection is built, which is helpful for achieving droplet rapid real-time adjustment with a feedback loop. Through experiments, the applicability of the various intelligent optimization methods is verified, and many tests against optimization targets are made. 4. To extract the droplet properties of images, multilevel thresholding image segmentation based on OTSU method is applied. However, when the number of thresholds increases, the consumption of CPU time grows exponentially. The evolution algorithms are helpful to solve this problem. For the high-dimensional problems, the Otsu methods based on the classical evolution algorithms may get trapped into local optimal or be instability due to the inefficiency of local search. To overcome it, in this paper, we present a new hierarchical artificial bee colony (HABC) Optimizer. Through tests against benchmark functions, HABC has a high convergence rate and good global search ability, which is suitable for multilevel thresholding image segmentation.
语种中文
产权排序1
文献类型学位论文
条目标识符http://ir.sia.cn/handle/173321/16757
专题信息服务与智能控制技术研究室
作者单位1.中国科学院沈阳自动化研究所
2.中国科学院大学
推荐引用方式
GB/T 7714
何茂伟. 面向喷印工艺的液滴状态优化模型与智能控制方法研究[D]. 沈阳. 中国科学院沈阳自动化研究所,2015.
条目包含的文件
文件名称/大小 文献类型 版本类型 开放类型 使用许可
面向喷印工艺的液滴状态优化模型与智能控制(3818KB) 开放获取CC BY请求全文
个性服务
推荐该条目
保存到收藏夹
查看访问统计
导出为Endnote文件
谷歌学术
谷歌学术中相似的文章
[何茂伟]的文章
百度学术
百度学术中相似的文章
[何茂伟]的文章
必应学术
必应学术中相似的文章
[何茂伟]的文章
相关权益政策
暂无数据
收藏/分享
所有评论 (0)
暂无评论
 

除非特别说明,本系统中所有内容都受版权保护,并保留所有权利。