本文選題：水泥生產(chǎn)　切入點：電耗預測　出處：《天津理工大學》2015年碩士論文　論文類型：學位論文

【摘要】：能源消耗問題是制約企業(yè)發(fā)展的關(guān)鍵所在,更是關(guān)乎國民經(jīng)濟與社會可持續(xù)發(fā)展的重大問題。建材工業(yè)能耗約占全國工業(yè)總能耗的13%,而其中水泥制造業(yè)的能耗量高達建材工業(yè)能耗總量的76%左右�？梢�,對于水泥企業(yè)能耗管理問題的研究刻不容緩。水泥生產(chǎn)的主要消耗能源是煤炭和電力,隨著生產(chǎn)工藝以及自動化程度的不斷提高,水泥企業(yè)的煤耗指標已呈逐年下降的趨勢,但水泥綜合電耗仍呈遞增趨勢,企業(yè)用電費用開支居高不下。針對上述情況,本文以山東某水泥廠為研究對象,通過有效的數(shù)據(jù)分析和預測技術(shù)手段,建立電耗預測模型,達到能源的合理配置和利用,以此為基礎(chǔ),建立用電負荷調(diào)度模型,進行峰谷平用電負荷調(diào)度,在保證產(chǎn)量前提下,達到日用電成本最低。通過本課題的實施,實現(xiàn)安全、優(yōu)良供能,最終達到提高整體電能利用效率、降低消耗、節(jié)約成本的目的。首先,本文從實際出發(fā),在分析了我國水泥行業(yè)發(fā)展現(xiàn)狀后,深入研究了水泥生產(chǎn)的工業(yè)流程,總結(jié)出電耗分布及單位指標。此外,針對水泥企業(yè)能耗數(shù)據(jù)過多且復雜等特點,研究了水泥企業(yè)中的能源管理系統(tǒng),尤其是能源數(shù)據(jù)采集的原理與實現(xiàn),為下文的建模做鋪墊。其次,運用主成分分析法解得幾個影響水泥生產(chǎn)電耗的關(guān)鍵因素,有效表征和替代了原始的眾多影響因素,降低了模型的復雜度。同時提出了基于改進多元非線性算法的水泥電耗預測模型,提高了模型的預測精度。之后,建立基于分時電價的水泥企業(yè)用電負荷優(yōu)化調(diào)度模型,利用Matlab平臺的模式搜索法、單純形法和遺傳算法進行了線性規(guī)劃問題的求解,并對比了三種方法的優(yōu)劣。最后,根據(jù)之前所得的電耗預測值對未來全廠各工段及主要耗電設(shè)備進行峰谷平用電調(diào)度,制定出各個主要耗電環(huán)節(jié)設(shè)備的運行時刻表。本文通過算例分析驗證了電耗預測模型及改進算法的有效性,使電耗預測值與實際值有較高的吻合度,對于水泥廠的電耗預測管理具有重要的參考意義。此外還驗證了用電負荷優(yōu)化調(diào)度模型的正確性和Matlab求解線性規(guī)劃算法的優(yōu)越性。從而將線性規(guī)劃應用于輔助高耗能企業(yè)優(yōu)化分配資源方面,降低了用電成本,具有較強的實用性。
[Abstract]:The problem of energy consumption is the key to restricting the development of enterprises. The energy consumption of the building materials industry accounts for about 13 percent of the total energy consumption in the national industry, and the energy consumption of the cement manufacturing industry is about 76% of the total energy consumption of the building materials industry. It is urgent to study the problem of energy consumption management in cement enterprises. The main energy consumed in cement production is coal and electricity. With the continuous improvement of production technology and automation, the coal consumption index of cement enterprises has been decreasing year by year. However, the comprehensive power consumption of cement is still increasing, and the expense of electricity consumption of enterprises is high. In view of the above situation, this paper takes a cement plant in Shandong province as the research object, and establishes the forecasting model of electricity consumption through effective technical means of data analysis and prediction. On the basis of the reasonable allocation and utilization of energy, the model of power load dispatching is established, and the peak and valley power load dispatching is carried out. Under the premise of ensuring the output, the daily electricity cost is the lowest. Through the implementation of this subject, the security is realized. In order to improve the efficiency of electric energy utilization, reduce consumption and save cost, this paper analyzes the present situation of cement industry in China, and deeply studies the industrial process of cement production, based on the analysis of the present situation of cement industry in China. In addition, the energy management system in cement enterprises, especially the principle and realization of energy data acquisition, is studied in view of the characteristics of excessive and complex energy consumption data in cement enterprises. Secondly, the principal component analysis method is used to solve several key factors that affect the power consumption of cement production, which effectively represent and replace many original factors. The complexity of the model is reduced. At the same time, the prediction model of cement power consumption based on improved multivariate nonlinear algorithm is proposed, which improves the prediction accuracy of the model. After that, the optimal dispatching model of electric load of cement enterprise based on time-sharing price is established. The linear programming problem is solved by using the pattern search method of Matlab platform, simplex method and genetic algorithm, and the advantages and disadvantages of the three methods are compared. According to the predicted value of electricity consumption obtained before, all sections and main power consuming equipment of the whole plant in the future will be scheduled for peak-valley flat power consumption. In this paper, the validity of the power consumption prediction model and the improved algorithm are verified by the example analysis, which makes the predicted value of electricity consumption have a high consistency with the actual value. It also verifies the correctness of power load optimal dispatching model and the superiority of Matlab in solving linear programming algorithm. Thus, linear programming is applied to auxiliary high level. With regard to the optimal allocation of resources by energy-consuming enterprises, The utility model reduces the cost of electricity consumption and has strong practicability.
【學位授予單位】：天津理工大學
【學位級別】：碩士
【學位授予年份】：2015
【分類號】：TQ172.6;TM73

【參考文獻】

相關(guān)期刊論文 前10條

1 王惠文;孟潔;;多元線性回歸的預測建模方法[J];北京航空航天大學學報;2007年04期

2 周仲禮,馮文新,龔灝,雍自權(quán),劉樹根;基于小波神經(jīng)網(wǎng)絡(luò)模型的中國能耗預測[J];成都理工大學學報(自然科學版);2005年05期

3 牛東曉;賈建榮;;改進GM(1,1)模型在電力負荷預測中的應用[J];電力科學與工程;2008年04期

4 鄭靜 ,杜秀華 ,史新祈;大型鋼鐵企業(yè)電力負荷的短期預測研究[J];電力需求側(cè)管理;2004年01期

5 阮文駿;王蓓蓓;李揚;楊勝春;;峰谷分時電價下的用戶響應行為研究[J];電網(wǎng)技術(shù);2012年07期

6 郭宇;茹海鵬;;鋼鐵企業(yè)能源分配動態(tài)優(yōu)化數(shù)學模型[J];硅谷;2008年13期

7 鄒健,諸靜;模糊預測函數(shù)控制在水泥回轉(zhuǎn)窯分解爐溫控系統(tǒng)中的應用研究[J];硅酸鹽學報;2001年04期

8 朱曉東;魯鐵定;陳西江;;正交多項式曲線擬合[J];東華理工大學學報(自然科學版);2010年04期

9 程瑜;翟娜娜;;面向智能電網(wǎng)的峰谷分時電價評估[J];電力系統(tǒng)保護與控制;2010年21期

10 劉暢;張少華;王f[;;多時段可中斷負荷調(diào)度的智能優(yōu)化算法[J];電力系統(tǒng)保護與控制;2010年24期

相關(guān)博士學位論文 前1條

1 徐俊剛;流程企業(yè)生產(chǎn)調(diào)度方法與應用研究[D];中國科學院研究生院（軟件研究所）;2003年

，

本文編號：1630486