本文選題：濕法制粒 + BP神經(jīng)網(wǎng)絡(luò)　； 參考：《蘭州理工大學(xué)》2017年碩士論文

【摘要】：濕法制粒片劑生產(chǎn)過程主要包括濕法制粒、干燥、壓片、包衣四個工段,其各個生產(chǎn)工段之間環(huán)環(huán)相扣,上一工段的好壞嚴重制約下一工段的生產(chǎn)并最終影響片劑產(chǎn)品的質(zhì)量。伴隨著新技術(shù)的層出不窮以及GMP(Good Manufacturing Practice)規(guī)范在濕法制粒片劑生產(chǎn)管理中的實施,傳統(tǒng)的控制要求、控制方法和片劑生產(chǎn)落后的管理模式已不能滿足濕法制粒片劑生產(chǎn)的發(fā)展需求。因此需要研究設(shè)計濕法制粒片劑生產(chǎn)的綜合控制系統(tǒng),以適應(yīng)現(xiàn)代化生產(chǎn)的要求。本文以浙江溫州某制藥設(shè)備生產(chǎn)公司的濕法制粒片劑生產(chǎn)線為研究對象,在深入了解片劑生產(chǎn)工藝、方法、生產(chǎn)需求的基礎(chǔ)上,針對生產(chǎn)過程中存在的一系列問題系統(tǒng)地研究了濕法制粒階段的粒徑大小優(yōu)化、沸騰干燥機流化倉內(nèi)部溫度控制以及濕法制粒片劑生產(chǎn)過程信息化、網(wǎng)絡(luò)化、集成化技術(shù)設(shè)計。論文的主要工作如下:1)為了提高濕法制粒所得藥物片劑的質(zhì)量,必須嚴格控制濕法制粒機出料口顆粒的平均粒度。文中根據(jù)濕法制粒機輸入輸出變量關(guān)系構(gòu)建合理的BP神經(jīng)網(wǎng)絡(luò),用來擬合平均粒度大小與攪拌槳轉(zhuǎn)速、切割刀轉(zhuǎn)速、攪拌時間、切割時間非線性函數(shù)關(guān)系。此外,為避免BP神經(jīng)網(wǎng)絡(luò)由于初始權(quán)值、閾值的隨機選取而導(dǎo)致的易陷入局部極值點而得不到全局最優(yōu)解的缺點,利用遺傳算法優(yōu)化構(gòu)建好的BP神經(jīng)網(wǎng)絡(luò),從而實現(xiàn)對平均粒徑大小與其影響因素的非線性擬合,實現(xiàn)對粒徑的良好控制,提高片劑生產(chǎn)的質(zhì)量。2)針對沸騰干燥機的流化倉內(nèi)溫度控制精度問題,本文以沸騰干燥機的蒸汽熱交換器為研究對象,對該工段的工藝和傳熱過程進行機理分析,建立其簡化模型。針對沸騰干燥機溫度控制的非線性、純滯后、時變性的特點,設(shè)計自適應(yīng)模糊PID控制器并進行Matlab/Simulink系統(tǒng)仿真。最終結(jié)果表明在自適應(yīng)模糊PID控制下,不僅響應(yīng)速度快,無穩(wěn)態(tài)誤差,而且超調(diào)量較少。3)研究設(shè)計片劑生產(chǎn)過程的綜合自動化控制系統(tǒng),實現(xiàn)對濕法制粒片劑生產(chǎn)各個工段的集中監(jiān)控和分散控制。既解決濕法制粒片劑生產(chǎn)過程中對單臺設(shè)備的特殊控制要求,又實現(xiàn)系統(tǒng)的分布式控制和遠程監(jiān)控功能,提高生產(chǎn)過程的自動化程度。實現(xiàn)濕法制粒片劑生產(chǎn)過程的網(wǎng)絡(luò)化、集散化和智能化控制。
[Abstract]:The production process of wet granulation tablet mainly includes four sections: wet granulation, drying, pressing and coating. The production of each section is interlocked with each other. The quality of the next section is seriously restricted by the quality of the previous section and ultimately affects the quality of the tablet product. With the emergence of new technologies and the implementation of good Manufacturing practice (GMP) in the production and management of wet granulation tablets, the traditional control requirements, control methods and the backward management mode of tablet production can no longer meet the development needs of the production of wet granulation tablets. Therefore, it is necessary to design a comprehensive control system for the production of wet granulation tablets to meet the requirements of modern production. This paper takes the wet granulation tablet production line of a pharmaceutical equipment production company in Wenzhou, Zhejiang Province as the research object, on the basis of deeply understanding the production technology, method and production demand of the tablet. Aiming at a series of problems existing in the production process, the optimization of particle size in wet granulation stage, the internal temperature control of fluidized chamber of boiling dryer and the information, network and integrated technology design of wet granulation tablet production process were studied systematically. The main work of this paper is as follows: (1) in order to improve the quality of the tablets obtained by wet granulation, it is necessary to strictly control the average particle size at the outlet of the wet granulator. According to the relationship between input and output variables of wet granulator, a reasonable BP neural network is constructed, which is used to fit the nonlinear function relationship between the average particle size and the rotating speed of the agitator, the cutting cutter speed, the stirring time and the cutting time. In addition, in order to avoid the disadvantage that BP neural network is easy to fall into local extremum and get global optimal solution due to the initial weight and random selection of threshold, genetic algorithm is used to optimize the BP neural network. Thus the nonlinear fitting of the average particle size and its influencing factors, the good control of the particle size and the improvement of the quality of the tablet production are realized.) the temperature control accuracy in the fluidized chamber of the boiling dryer is solved. In this paper, the steam heat exchanger of boiling dryer is studied. The mechanism of process and heat transfer in this section is analyzed, and the simplified model is established. In view of the nonlinear, pure lag and time-varying characteristics of boiling dryer temperature control, an adaptive fuzzy pid controller is designed and simulated by Matlab / Simulink system. The results show that under the adaptive fuzzy pid control, not only the response speed is fast, no steady error, but also the overshoot is less. 3) the integrated automatic control system for tablet production process is designed and designed. To realize the centralized monitoring and decentralized control of each stage of wet granulation tablet production. It not only solves the special control requirements of single equipment in the process of wet granulation tablet production, but also realizes the distributed control and remote monitoring function of the system, and improves the automation degree of the production process. To realize the networked, distributed and intelligent control of the production process of wet granulation tablets.
【學(xué)位授予單位】：蘭州理工大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TQ460.5;TP273

【參考文獻】

相關(guān)期刊論文 前10條

1 劉潔;劉輝;霍震;;沸騰干燥機的自動進料工藝設(shè)計[J];機電信息;2016年23期

2 張鈞勃;;電廠DCS控制系統(tǒng)與現(xiàn)場總線的應(yīng)用問題[J];黑龍江科技信息;2016年08期

3 趙勝杰;;改進型模糊PID控制器在熱工控制系統(tǒng)中的仿真應(yīng)用[J];科技與企業(yè);2016年05期

4 李芳;趙天洋;;遺傳算法理論及其應(yīng)用進展探析[J];技術(shù)與市場;2016年01期

5 趙宏云;;影響片劑濕法制粒工藝穩(wěn)定性的常見因素及控制措施[J];中國高新技術(shù)企業(yè);2015年24期

6 武廣萍;何巍;倪斌;;基于OPC的MATLAB與ABB的數(shù)據(jù)通信[J];甘肅科技;2015年07期

7 LI Bao Jian;CHENG Chun Tian;;Monthly discharge forecasting using wavelet neural networks with extreme learning machine[J];Science China(Technological Sciences);2014年12期

8 霍彥生;;現(xiàn)場總線技術(shù)及其在電力自動化中的應(yīng)用[J];電子世界;2014年12期

9 孟棟;樊重俊;王家楨;;混沌遺傳算法對BP神經(jīng)網(wǎng)絡(luò)的改進研究[J];數(shù)學(xué)理論與應(yīng)用;2014年01期

10 蘭勝坤;;遺傳算法優(yōu)化BP神經(jīng)網(wǎng)絡(luò)的說話人識別系統(tǒng)[J];重慶理工大學(xué)學(xué)報(自然科學(xué));2013年10期

相關(guān)博士學(xué)位論文 前2條

1 李江榮;模糊系統(tǒng)的廣義H_2控制[D];西安電子科技大學(xué);2012年

2 龍祖強;變論域模糊控制器的若干重要問題研究[D];中南大學(xué);2011年

相關(guān)碩士學(xué)位論文 前10條

1 胡瑞卿;基于人工神經(jīng)網(wǎng)絡(luò)的煤礦安全評價研究[D];安徽理工大學(xué);2015年

2 韓春春;換熱器試驗過程溫度控制系統(tǒng)應(yīng)用研究[D];蘭州理工大學(xué);2015年

3 王亞壘;電加熱鍋爐DCS系統(tǒng)的設(shè)計與實現(xiàn)[D];西安科技大學(xué);2014年

4 姜萬強;基于現(xiàn)場總線的加藥機控制系統(tǒng)設(shè)計與實現(xiàn)[D];東北大學(xué);2012年

5 侯艷飛;基于改進模糊粒子群算法優(yōu)化神經(jīng)網(wǎng)絡(luò)的熱軋帶鋼寬度組合智能模型研究[D];太原理工大學(xué);2012年

6 周斌;基于改進遺傳算法的電動公交車行車調(diào)度研究[D];長沙理工大學(xué);2012年

7 郭曉原;遺傳算法中適應(yīng)度尺度變換與操作算子的比較研究[D];華北電力大學(xué);2012年

8 袁振華;現(xiàn)場總線技術(shù)及工業(yè)以太網(wǎng)在多節(jié)點控制系統(tǒng)中的研究與應(yīng)用[D];上海交通大學(xué);2011年

9 洪棟良;以太網(wǎng)和電臺網(wǎng)測量系統(tǒng)中測試協(xié)議的研究與實現(xiàn)[D];西安電子科技大學(xué);2011年

10 周凌翱;改進BP神經(jīng)網(wǎng)絡(luò)在模式識別中的應(yīng)用及研究[D];南京理工大學(xué);2010年

，

本文編號：2064030