本文選題：篦冷機(jī)　切入點(diǎn)：自適應(yīng)粒子群算法　出處：《長春工業(yè)大學(xué)》2017年碩士論文　論文類型：學(xué)位論文

【摘要】：篦冷機(jī)是新型干法水泥生產(chǎn)過程中的重要設(shè)備,在整個(gè)水泥熟料的生產(chǎn)過程中承擔(dān)對高溫熟料的冷卻任務(wù)的同時(shí),還有改善水泥熟料的易磨性、回收熟料熱量等作用。目前,已經(jīng)有很多預(yù)測和控制理論在篦冷機(jī)環(huán)節(jié)進(jìn)行了仿真與應(yīng)用,例如BP神經(jīng)網(wǎng)絡(luò)、支持向量機(jī)、最小二乘支持向量機(jī)和各種形式的PID等。這些預(yù)測和控制理論需要設(shè)定一些參數(shù),設(shè)定的參數(shù)決定了在篦冷機(jī)環(huán)節(jié)的預(yù)測和控制效果。本研究的主要工作:1.深入分析一下新型干法水泥生產(chǎn)的流程,對篦冷機(jī)的發(fā)展和工作原理有一個(gè)深入的認(rèn)識。篦冷機(jī)篦壓的控制方式有兩種,最為常用的是通過篦冷機(jī)的篦床速度控制篦下壓力。制定篦冷機(jī)的篦壓的控制目標(biāo),根據(jù)控制目標(biāo)建立篦冷機(jī)熱量轉(zhuǎn)換控制系統(tǒng)的方案,即篦下壓力設(shè)定系統(tǒng)和篦下壓力控制系統(tǒng)。2.為了建立篦下壓力設(shè)定系統(tǒng),首先對粒子群算法和LS-SVM算法進(jìn)行了深入研究,針對粒子群算法的缺點(diǎn),提出一種新的粒子群算法(自適應(yīng)粒子群算法),通過仿真實(shí)驗(yàn)證明自適應(yīng)粒子群算法的優(yōu)越性。篦下壓力設(shè)定系統(tǒng)是基于自適應(yīng)PSO的LS-SVM算法建立的,基于自適應(yīng)粒子群的LS-SVM算法實(shí)質(zhì)就是自適應(yīng)粒子群算法對LS-SVM的兩個(gè)參數(shù)進(jìn)行優(yōu)化,得到最佳的參數(shù),并建立預(yù)測模型,預(yù)測模型的輸入端是提升機(jī)的電流、三次風(fēng)溫、二次風(fēng)溫,輸出端是篦下壓力,并通過仿真對比實(shí)驗(yàn)證明本文建立的壓力預(yù)測模型的實(shí)用性。3.為了建立篦冷機(jī)篦下壓力控制系統(tǒng),首先分析篦冷機(jī)中的比例閥原理,其次查閱文獻(xiàn)得出篦冷機(jī)的電液伺服控制系統(tǒng)的數(shù)學(xué)模型,本文采用的篦下壓力控制方法是基于改進(jìn)PSO(自適應(yīng)粒子群算法)的PID控制方法,其方法的實(shí)質(zhì)就是用自適應(yīng)粒子群算法對傳統(tǒng)PID控制方法進(jìn)行參數(shù)整定優(yōu)化,得到最佳的優(yōu)化參數(shù),并通過仿真實(shí)驗(yàn)證明本文提出的方法比傳統(tǒng)的方法控制效果更好,即調(diào)速過程更平穩(wěn)和調(diào)節(jié)時(shí)間短。4.設(shè)計(jì)了篦冷機(jī)三級監(jiān)控系統(tǒng),即現(xiàn)場級、控制級、企業(yè)管理級。
[Abstract]:Grate cooler is an important equipment in the process of production of new type dry cement. It undertakes the task of cooling high temperature clinker in the whole production process of cement clinker, and at the same time, it also improves the wearability of cement clinker and reclaims the heat of clinker, etc. At present, Many prediction and control theories have been simulated and applied in grate cooler, such as BP neural network, support vector machine, least square support vector machine and various forms of PID. The parameters determined the prediction and control effect in grate cooler. The main work of this study is: 1. Deeply analyze the process of new dry cement production. There are two ways to control the grate pressure of the grate cooler, the most commonly used is to control the down pressure of the grate through the grate cooler's velocity, and to establish the control target of the grate pressure of the grate cooler. According to the control goal, the scheme of establishing the heat transfer control system of grate cooler, I. E. the system of setting down pressure of grate and the control system of pressure under grate, is established. In order to set up the system of setting down pressure of grate, the particle swarm optimization algorithm and LS-SVM algorithm are studied deeply. In view of the shortcomings of particle swarm optimization (PSO), a new PSO (Adaptive PSO) algorithm is proposed. The simulation results show the superiority of APSO. The pressure setting system of grate is based on the LS-SVM algorithm of adaptive PSO. The essence of the LS-SVM algorithm based on adaptive particle swarm optimization is to optimize the two parameters of LS-SVM, to obtain the best parameters, and to establish a prediction model. The input of the prediction model is the hoist current, the third air temperature. The secondary air temperature and the output end are grate down pressure. The simulation results show that the pressure prediction model established in this paper is practical. 3. In order to establish the pressure control system of grate cooler, the principle of proportional valve in grate cooler is analyzed. Secondly, the mathematical model of electro-hydraulic servo control system of grate cooler is obtained by referring to the literature. The method used in this paper is PID control method based on improved PSO (Adaptive Particle Swarm Optimization algorithm). The essence of the method is to optimize the parameters of the traditional PID control method by using the adaptive particle swarm optimization algorithm. The simulation results show that the proposed method is more effective than the traditional control method. That is, the speed regulation process is more stable and the adjustment time is short. 4. The three-stage monitoring system of grate cooler is designed, that is, the field level, the control level and the enterprise management level.
【學(xué)位授予單位】：長春工業(yè)大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2017
【分類號】：TQ172.622.4;TP273

【參考文獻(xiàn)】

相關(guān)期刊論文 前10條

1 楊華芬;董德春;楊麗華;李麗;;一種改進(jìn)的粒子群優(yōu)化算法[J];重慶師范大學(xué)學(xué)報(bào)(自然科學(xué)版);2015年05期

2 余勝威;曹中清;;基于人群搜索算法的PID控制器參數(shù)優(yōu)化[J];計(jì)算機(jī)仿真;2014年09期

3 劉愛軍;楊育;李斐;邢青松;陸惠;張煜東;;混沌模擬退火粒子群優(yōu)化算法研究及應(yīng)用[J];浙江大學(xué)學(xué)報(bào)(工學(xué)版);2013年10期

4 趙志剛;黃樹運(yùn);王偉倩;;基于隨機(jī)慣性權(quán)重的簡化粒子群優(yōu)化算法[J];計(jì)算機(jī)應(yīng)用研究;2014年02期

5 孟麗;韓璞;任燕燕;王東風(fēng);;基于多目標(biāo)粒子群算法的PID控制器設(shè)計(jì)[J];計(jì)算機(jī)仿真;2013年07期

6 孟杰;陳慶樟;張凱;;基于粒子群算法的汽車懸架PID控制仿真[J];計(jì)算機(jī)仿真;2013年04期

7 唐軍;李文星;;基于遺傳算法的數(shù)控機(jī)床轉(zhuǎn)臺交流伺服系統(tǒng)PID參數(shù)優(yōu)化[J];現(xiàn)代制造工程;2013年02期

8 蘇姍姍;蘇小光;;基于PLC的信號采集系統(tǒng)[J];國外電子測量技術(shù);2012年09期

9 聶瑞;章衛(wèi)國;李廣文;劉小雄;;一種自適應(yīng)混合多目標(biāo)粒子群優(yōu)化算法[J];西北工業(yè)大學(xué)學(xué)報(bào);2011年05期

10 馮燁;熊瑞平;張志會;金成毅;;第四代篦冷機(jī)刮板速度控制器設(shè)計(jì)[J];機(jī)床與液壓;2011年19期

相關(guān)碩士學(xué)位論文 前6條

1 康旭;水泥篦冷機(jī)控制模式的研究[D];河北科技大學(xué);2014年

2 郭言;5000t/d篦冷機(jī)熟料厚度電液控制系統(tǒng)研究與實(shí)踐[D];燕山大學(xué);2013年

3 胡國文;水泥生產(chǎn)過程篦冷機(jī)環(huán)節(jié)的優(yōu)化控制研究[D];濟(jì)南大學(xué);2013年

4 吳姝芹;預(yù)測控制理論在水泥熟料篦式冷卻過程中的應(yīng)用研究[D];濟(jì)南大學(xué);2007年

5 傅建青;模糊控制與PID控制在涂層生產(chǎn)線上的應(yīng)用[D];蘭州理工大學(xué);2006年

6 馮紹航;篦式冷卻機(jī)的換熱理論研究[D];西安建筑科技大學(xué);2004年

，

本文編號：1635070