【摘要】：隨著傳統(tǒng)行業(yè)競爭的日益加劇，為了獲得更多的利潤，企業(yè)不得不加強(qiáng)成本的管理，作為成本支出中第三大項的電力成本支出越來越受到管理者的重視。特別是隨著能源危機(jī)的暴發(fā)，國家加強(qiáng)宏觀調(diào)控，對電力進(jìn)行能效管理已經(jīng)成為企業(yè)不容忽視的問題。目前，大部分的能效管理研究都是針對電力系統(tǒng)供電段，少數(shù)一些學(xué)者從建筑設(shè)計的角度闡述了商業(yè)建筑的節(jié)能方法，本文在總結(jié)和學(xué)習(xí)已有科研成果的基礎(chǔ)上，，對商業(yè)負(fù)荷的能效管理進(jìn)行了研究，提出建立一個以負(fù)荷預(yù)測模型為核心，采用統(tǒng)計過程控制方法設(shè)置報警模式的能耗監(jiān)控平臺。 詳盡、準(zhǔn)確的負(fù)荷特性分析是提高負(fù)荷預(yù)測精度的必要前提，也是能耗監(jiān)測平臺可以正常運(yùn)行的先決條件。因此，本文首先對商業(yè)負(fù)荷進(jìn)行了特性分析，指出相對于電力系統(tǒng)負(fù)荷，商業(yè)負(fù)荷具有明顯的商業(yè)周期性、基負(fù)荷小、波動性大的特點(diǎn)，并根據(jù)不同的周期性和波動性，將各子負(fù)荷劃分為無規(guī)律大波動負(fù)荷、規(guī)律大波動負(fù)荷以及規(guī)律小波動負(fù)荷等三類�？紤]到數(shù)據(jù)中存在噪聲且噪聲易淹沒于負(fù)荷曲線的毛刺中，本文基于噪聲和信號具有不同的傳播頻率假設(shè)，利用小波軟閥值降噪法對歷史數(shù)據(jù)進(jìn)行降噪預(yù)處理。為提高大波動負(fù)荷的預(yù)測精度，本文采用了組合模型對負(fù)荷進(jìn)行預(yù)測。首先，利用小波多分辨率分析法將降噪后的負(fù)荷分解為不同頻率的分量；然后，利用支持向量回歸模型對各分量分別進(jìn)行回歸預(yù)測，最后，各分量的預(yù)測值經(jīng)小波重構(gòu)后得到負(fù)荷的預(yù)測值。本文利用實(shí)例，驗證了該模型的可行性和有效性。 本文最后將應(yīng)用于企業(yè)生產(chǎn)質(zhì)量管控的SPC方法引入到商業(yè)負(fù)荷的能耗管控領(lǐng)域，與前文建立的組合模型相結(jié)合，構(gòu)建了一個能耗監(jiān)控平臺。該監(jiān)控平臺可以實(shí)時對異常負(fù)荷進(jìn)行預(yù)警，使能耗處于合理管控區(qū)域，為商業(yè)負(fù)荷的能效管理提供了一個切實(shí)可行的方法。
[Abstract]:With the increasing competition in traditional industries, in order to obtain more profits, enterprises have to strengthen the cost management. As the third largest item of the cost expenditure, the electric power cost expenditure is paid more and more attention by the managers. Especially with the outbreak of energy crisis, it has become an important problem for enterprises to strengthen macro-control and energy efficiency management. At present, most of the research on energy efficiency management is aimed at the power supply section of power system. A few scholars expound the energy saving methods of commercial buildings from the point of view of architectural design. In this paper, the energy efficiency management of commercial load is studied, and a monitoring platform of energy consumption based on load forecasting model and alarm mode is established by means of statistical process control method. Detailed and accurate analysis of load characteristics is a necessary prerequisite for improving the accuracy of load forecasting and is also a prerequisite for the normal operation of the energy consumption monitoring platform. Therefore, this paper firstly analyzes the characteristics of commercial load, and points out that compared with power system load, commercial load has obvious commercial periodicity, small base load and large volatility, and according to different periodicity and volatility, The subloads are divided into three categories: irregular large fluctuating load, regular large fluctuating load and regular small fluctuating load. Considering that there is noise in the data and the noise is easily submerged in the burr of the load curve, based on the assumption of different propagation frequency of the noise and the signal, the wavelet soft threshold de-noising method is used to pre-process the noise reduction of the historical data. In order to improve the forecasting accuracy of large fluctuating load, the combined model is used to forecast the load. Firstly, wavelet multi-resolution analysis method is used to decompose the noise reduction load into components with different frequencies. Then, the support vector regression model is used to predict each component separately. The predicted value of each component is obtained by wavelet reconstruction. The feasibility and validity of the model are verified by an example. In the end of this paper, the SPC method which is applied to enterprise production quality control is introduced into the field of energy consumption control of commercial load, and a platform of energy consumption monitoring is constructed by combining with the combination model established above. The platform can warn the abnormal load in real time, make the energy consumption in the reasonable control area, and provide a feasible method for the energy efficiency management of the commercial load.
【學(xué)位授予單位】：華僑大學(xué)
【學(xué)位級別】：碩士
【學(xué)位授予年份】：2013
【分類號】：TM73

【參考文獻(xiàn)】

相關(guān)期刊論文 前1條

1 胡雪梅;劉鋒;;半?yún)?shù)時變系數(shù)模型的序列相關(guān)檢驗[J];應(yīng)用數(shù)學(xué)學(xué)報;2011年06期

本文編號：2235591