[心理學(xué)]置換蒸煮論文：置換蒸煮DDS過(guò)程蒸煮終點(diǎn)軟測(cè)量的研究.doc

置換蒸煮論文：置換蒸煮（DDS）過(guò)程蒸煮終點(diǎn)軟測(cè)量的研究【中文摘要】由于能耗和環(huán)境等問(wèn)題,造紙制漿工業(yè)的發(fā)展已嚴(yán)重受阻。尋找節(jié)能環(huán)保型制漿技術(shù)是擺在我國(guó)漿紙工業(yè)面前的一個(gè)重要戰(zhàn)略課題。置換蒸煮系統(tǒng)(Displacement Digester Systems)是20世紀(jì)80年代發(fā)展起來(lái)的一項(xiàng)高效節(jié)能的間歇式制漿技術(shù),由于其顯著的環(huán)境和經(jīng)濟(jì)效益,該技術(shù)已逐漸成為了制漿蒸煮技術(shù)發(fā)展的主要方向。本文在介紹置換蒸煮工藝機(jī)理的基礎(chǔ)上,針對(duì)其蒸煮終點(diǎn)難以判斷這一問(wèn)題,從以下幾個(gè)方面展開(kāi)了工作。首先,文章提出了置換蒸煮終點(diǎn)神經(jīng)網(wǎng)絡(luò)軟測(cè)量的方法。蒸煮終點(diǎn)是制訂蒸煮時(shí)間的依據(jù),對(duì)紙漿質(zhì)量以及能耗有著直接影響。目前較為科學(xué)可靠的蒸煮終點(diǎn)判定法是依靠卡伯值,以及其它影響因素共同確定。由于蒸煮過(guò)程的復(fù)雜性,蒸煮終點(diǎn)卡伯值的在線檢測(cè)非常困難。文中提出利用神經(jīng)網(wǎng)絡(luò)軟測(cè)量的方法,通過(guò)比較容易測(cè)量的輔助變量來(lái)間接測(cè)量紙漿卡伯值,從而確定置換蒸煮的終點(diǎn)。在該軟測(cè)量模型中,紙漿卡伯值作為模型的主導(dǎo)輸出變量,硫化度、H因子和有效堿濃度作為模型的輸入變量。其次,分別采用RBF和BP神經(jīng)網(wǎng)絡(luò)兩種方法建立置換蒸煮終點(diǎn)卡伯值軟測(cè)量模型,并進(jìn)行對(duì)比分析。在其RBF軟測(cè)量模型的基礎(chǔ)上,介紹了該模型在DCS控制系統(tǒng)中的實(shí)現(xiàn)問(wèn)題。置換蒸煮DCS系統(tǒng)是以SIEMENS S7-400 PLC作為硬件開(kāi)發(fā)平臺(tái),WinCC和Step7為軟件開(kāi)發(fā)平臺(tái)。上位機(jī)不僅要完成WinCC軟件設(shè)計(jì),還要實(shí)現(xiàn)與MATLAB軟件的對(duì)接,將置換蒸煮終點(diǎn)軟測(cè)量技術(shù)嵌入到DCS控制系統(tǒng)中。最后,在對(duì)置換蒸煮DCS控制系統(tǒng)設(shè)計(jì)的基礎(chǔ)上,文章還對(duì)H因子計(jì)算、蒸煮鍋上中下溫度一致性控制、蒸煮鍋壓力控制等一系列控制要點(diǎn)提出了解決方案?！居⑽恼縊wing to energy consumption, pollution and other issues, the development of paper pulp industry has been seriously hampered. Searching for a new environmental and energy-saving pulping technology is an important strategic issue placed in pulp and paper industry of china. Displacement digester systems, which was developed in 1980, is a high-efficiency and energy-conserving pulping technology. Because of its good environmental and economic benefits, displacement digester systems has gradually became an important development direction of pulp technology.Based on mechanism introduction of displacement digester systems, aiming at the difficulty of judging pulp cooking endpoint, the work in this paper was expanded from following aspects.Firstly, soft measurement method on cooking endpoint of displacement digester systems has been proposed in the article. The cooking endpoint, which is the basis for time, has a direct impact on pulp quality and energy consumption. At present, more scientific and reliable method for judging cooking endpoint is to rely on kappa number and other influencing factors. However, online direct detection of kappa number is very difficult as the complexity of the cooking process, a soft measurement scheme of kappa number was proposed based on artificial neural network to indirectly measure kappa number and determine the endpoint of cooking process. Kappa number was selected as the output variable of model and H factor, sulfidity and the effective concentration of alkali were selected as input variables of the model.Secondly, two approaches were used for soft measurement model by RBF neural network and BP neural network in this article. On the basis of the model, its application in the DCS control system was introduced. SIEMENS S7-400 PLC was chosen as hardware development platform and WinCC and Step7 software were selected as software development platform in DCS. In order to monitor and operate the entire production process, the host computer not only need to design WinCC development platform, but also achieve docking with the MATLAB software, which will embed soft sensor technology into DCS control system.Thirdly, based on the design of DCS control system, some control schemes of displacement digester systems were briefly presented, such as, H factor calculation, consistency of the upper and lower temperature of digester, the stability of digester pressure and so on.【關(guān)鍵詞】置換蒸煮 蒸煮終點(diǎn) 軟測(cè)量 RBF模型【英文關(guān)鍵詞】Displacement Digester Systems Cooking Endpoint Soft Sensor RBF Neural Network【目錄】置換蒸煮（DDS）過(guò)程蒸煮終點(diǎn)軟測(cè)量的研究摘要4-5ABSTRACT5-61 緒論9-131.1 選題的目的和意義9-101.2 國(guó)內(nèi)外研究現(xiàn)狀10-111.3 課題研究的主要內(nèi)容及章節(jié)安排11-131.3.1 主要研究?jī)?nèi)容11-121.3.2 章節(jié)安排12-132 置換蒸煮系統(tǒng)(DDS)工藝13-202.1 置換蒸煮系統(tǒng)工藝流程13-172.2 DDS蒸煮機(jī)理分析17-203 置換蒸煮終點(diǎn)軟測(cè)量原理20-303.1 蒸煮終點(diǎn)軟測(cè)量方法20-233.1.1 蒸煮終點(diǎn)的定義203.1.2 軟測(cè)量的定義20-223.1.3 DDS蒸煮終點(diǎn)軟測(cè)量22-233.2 DDS蒸煮終點(diǎn)軟測(cè)量模型23-273.2.1 DDS蒸煮終點(diǎn)軟測(cè)量的影響因素23-243.2.2 典型的間歇蒸煮過(guò)程軟測(cè)量模型24-273.3 現(xiàn)有的蒸煮終點(diǎn)軟測(cè)量模型存在的問(wèn)題27-283.4 本文擬采用的置換蒸煮終點(diǎn)軟測(cè)量方法28-304 RBF神經(jīng)網(wǎng)絡(luò)原理30-354.1 人工神經(jīng)網(wǎng)絡(luò)原理304.2 BP神經(jīng)網(wǎng)絡(luò)與RBF神經(jīng)網(wǎng)絡(luò)的比較30-314.3 徑向基神經(jīng)網(wǎng)絡(luò)原理31-334.3.1 徑向基函數(shù)31-324.3.2 徑向基函數(shù)神經(jīng)網(wǎng)絡(luò)32-334.3.3 RBF神經(jīng)網(wǎng)絡(luò)訓(xùn)練過(guò)程334.4 基于MATLAB軟件的RBF網(wǎng)絡(luò)設(shè)計(jì)33-354.4.1 MATLAB工具箱33-344.4.2 徑向基函數(shù)網(wǎng)絡(luò)的嚴(yán)格設(shè)計(jì)344.4.3 更有效的徑向基函數(shù)網(wǎng)絡(luò)的設(shè)計(jì)34-355 置換蒸煮終點(diǎn)卡伯值RBF模型的建立35-495.1 RBF模型的構(gòu)建35-365.2 RBF網(wǎng)絡(luò)的訓(xùn)練與測(cè)試36-405.3 BP模型的構(gòu)建40-475.4 DDS蒸煮終點(diǎn)RBF模型與BP模型的比較47-496 置換蒸煮終點(diǎn)軟測(cè)量模型在DCS中的實(shí)現(xiàn)49-686.1 置換蒸煮終點(diǎn)軟測(cè)量模型輔助變量的檢測(cè)49-556.2 置換