日照钢铁360m~2烧结机过程自动控制系统的分析与设计
发布时间：2018-11-15 21:49
【摘要】：目前，国内大中型烧结机都具备了过程检测和设备控制能力，当务之急就是研究和开发烧结过程控制方法，开发出我们国家自主知识产权的烧结过程自动控制系统。 本文是基于日照钢铁控股集团有限公司烧结自动控制系统基础上，通过查阅大量的参考文献和对现场烧结工艺的熟悉了解，对烧结系统的仪器仪表选型标准以及PLC以及上位机的软硬件结构、网络通信等有了更深一步的认识。结合配料现场的工艺要求和控制说明，最终完成了配料自动控制系统硬件的选型及安装调试和软件上的编程。 在实际生产过程中对烧结终点的控制，由于受到料层透气性性或者设备缺陷的影响，很难直接得到理想的烧结终点的位置，从而无法实现烧结终点的闭环控制。本文引入了模糊小波神经网络，提出了多尺度小波逼近方法采用了前馈和反馈相结合的模糊控制系统。借鉴国内外模糊控制模型，论文中基于现场所得到的风箱废气温度曲线，，采用模糊小波神经网络的控制算法对烧结终点进行预测，建立了烧结机的简易模型，在假定该模型能够较稳定模拟的烧结机运行状况下，通过Matlab进行建模仿真，并与Elman神经网络预测控制进行对比。仿真结果表明，从理论上证明了对烧结终点的预测控制算法要优于Elman神经网络控制算法。
[Abstract]:At present, large and medium-sized sintering machines in China have the capability of process detection and equipment control. It is urgent to study and develop the sintering process control method and develop the sintering process automatic control system of our country's independent intellectual property rights. This paper is based on the automatic sintering control system of Rizhao Iron and Steel holding Group Co., Ltd., through consulting a large number of references and familiar with the field sintering process. It also has a deeper understanding of the instrument selection standard of sintering system, the hardware and software structure of PLC and host computer, network communication and so on. Combined with the process requirements and control instructions, the hardware selection, installation, debugging and software programming of the automatic proportioning control system were completed. Due to the influence of the permeability of the material layer or the defect of the equipment, it is difficult to obtain the ideal position of the sintering end point directly in the actual production process, so that the closed loop control of the sintering end point can not be realized. In this paper, a fuzzy wavelet neural network is introduced, and a multiscale wavelet approximation method is proposed, which combines feedforward and feedback. Referring to the fuzzy control model at home and abroad, the paper uses fuzzy wavelet neural network control algorithm to predict the sintering end point, and establishes a simple model of sintering machine based on the temperature curve of bellows exhaust gas, which is obtained from the field, and the control algorithm of fuzzy wavelet neural network is used to predict the sintering end point. Under the condition that the model can be simulated stably, the model is modeled and simulated by Matlab and compared with Elman neural network predictive control. The simulation results show that the predictive control algorithm for the sintering end point is superior to the Elman neural network control algorithm in theory.
【学位授予单位】：河北工业大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP273.5
本文编号：2334491
[Abstract]:At present, large and medium-sized sintering machines in China have the capability of process detection and equipment control. It is urgent to study and develop the sintering process control method and develop the sintering process automatic control system of our country's independent intellectual property rights. This paper is based on the automatic sintering control system of Rizhao Iron and Steel holding Group Co., Ltd., through consulting a large number of references and familiar with the field sintering process. It also has a deeper understanding of the instrument selection standard of sintering system, the hardware and software structure of PLC and host computer, network communication and so on. Combined with the process requirements and control instructions, the hardware selection, installation, debugging and software programming of the automatic proportioning control system were completed. Due to the influence of the permeability of the material layer or the defect of the equipment, it is difficult to obtain the ideal position of the sintering end point directly in the actual production process, so that the closed loop control of the sintering end point can not be realized. In this paper, a fuzzy wavelet neural network is introduced, and a multiscale wavelet approximation method is proposed, which combines feedforward and feedback. Referring to the fuzzy control model at home and abroad, the paper uses fuzzy wavelet neural network control algorithm to predict the sintering end point, and establishes a simple model of sintering machine based on the temperature curve of bellows exhaust gas, which is obtained from the field, and the control algorithm of fuzzy wavelet neural network is used to predict the sintering end point. Under the condition that the model can be simulated stably, the model is modeled and simulated by Matlab and compared with Elman neural network predictive control. The simulation results show that the predictive control algorithm for the sintering end point is superior to the Elman neural network control algorithm in theory.
【学位授予单位】：河北工业大学
【学位级别】：硕士
【学位授予年份】：2013
【分类号】：TP273.5
【参考文献】
相关期刊论文 前10条
1 吴胜利;陈东峰;赵成显;薛方;;基于烧结过程各带迁移规律的烧结终点预报方法[J];北京科技大学学报;2009年09期
2 王亚刚;邵惠鹤;;自整定PID控制器的研究综述[J];电气自动化;2000年01期
3 王宏文;蒋君杰;吴玲玲;唐永学;;WinCC监控系统在烧结过程中的应用[J];河北工业大学学报;2011年01期
4 李明河,王群京,孙雁飞;基于神经网络的烧结终点预报的研究[J];合肥工业大学学报(自然科学版);2004年06期
5 李桃,冯其明,范晓慧,姜涛;基于自组织神经网络的烧结终点自适应预报系统的开发[J];计算机工程与应用;2001年06期
6 王海东,邱冠周,黄圣生;烧结过程控制技术的发展[J];矿冶工程;1999年03期
7 袁廷奇,刘文江;基于Haar小波变换的连续时间系统鲁棒参数辨识[J];控制与决策;2000年03期
8 郑军,颜文俊,诸静;基于小波逼近和尺度变换的非参数辨识方法[J];控制与决策;2004年10期
9 吴为民;烧结混合料水分检测及智能控制系统[J];烧结球团;1997年04期
10 胡长庆,张玉柱;烧结过程中的传输理论[J];烧结球团;2000年03期
相关博士学位论文 前2条
1 王春生;铅锌烧结配料过程的智能集成建模与优化控制策略研究[D];中南大学;2008年
2 李洋;小波过程神经网络相关理论及其应用研究[D];哈尔滨工业大学;2008年
本文编号：2334491
本文链接：https://www.wllwen.com/falvlunwen/zhishichanquanfa/2334491.html
