發(fā)布時間：2019-03-07 22:15

【摘要】：隨著電力市場改革在我國的不斷深入,電力行業(yè)在快速發(fā)展的同時,也遇到了新的困難。電力市場改革之后,電力系統(tǒng)中各個市場成員成為了不同的利益主體,經濟性被提到了和安全性同樣重要的地位。如何盡可能地實現(xiàn)電力系統(tǒng)運行的經濟優(yōu)化?除了在市場規(guī)則方面制定相應的措施外,可以考慮通過向各個市場成員提供一些關于市場狀況的預測信息,以使整個電力市場安全、穩(wěn)定、平穩(wěn)的運行,減少由于對未來市場狀況準備不足所帶來的對系統(tǒng)正常運行的沖擊,同時,也方便各個市場成員合理地決策,謀取利益最大化。本文通過大數(shù)據(jù)在電力市場預測和分析中的應用研究,為電力市場預測提供建模指導,出具預測趨勢計算模型,分析預測期內需電量、負荷及負荷特性和公司售電量等相關指標預測信息,加強規(guī)劃,運行,營銷的信息共享,減少因重復建設造成的成本投入,減少專業(yè)部門工作量,有效提升電力市場預測數(shù)據(jù)處理的及時性、可靠性和準確性,提高數(shù)據(jù)的使用價值,為公司領導層和管理層提供更準確的預測支撐數(shù)據(jù),促進企業(yè)業(yè)務運作效率的提升,提高電力企業(yè)服務社會的工作效率。本文主要從兩個方面進行了相關理論方法的創(chuàng)新研究。首先,文章提出了以電力大數(shù)據(jù)平臺為基礎構建電力市場分析預測系統(tǒng),采用以Hadoop為核心的數(shù)據(jù)采集、分布式存儲、分布式處理等大數(shù)據(jù)生態(tài)系統(tǒng)技術,實現(xiàn)數(shù)據(jù)資源的集中管理、實時監(jiān)測和可視化管理。其次,本文選取了基于溫度變化的居民用電消費習慣主題作為典型數(shù)據(jù)挖掘應用,實現(xiàn)日最高負荷與溫度的關聯(lián)分析、居民日均用電量與溫度的關聯(lián)分析,掌握負荷隨著溫度變化的趨勢以及城市和農村地區(qū)基于溫度變化的用電量差異,為有序用電決策和措施提供輔助分析。
[Abstract]:With the deepening of the reform of power market in China, the electric power industry has encountered new difficulties as well as its rapid development. After the reform of the electricity market, the members of the power market have become different stakeholders, and the economy has been mentioned as important as the security. How to realize the economic optimization of power system operation as far as possible? In addition to establishing appropriate measures with regard to market rules, consideration could be given to providing market members with some forecasting information on market conditions in order to ensure the safe, stable and smooth operation of the entire electricity market, It can reduce the impact on the normal operation of the system caused by the lack of preparation for future market conditions, and at the same time, it is convenient for each member of the market to make reasonable decisions and maximize the profits. Through the research of big data's application in forecasting and analysis of electricity market, this paper provides the guidance of modeling for the forecast of electricity market, presents the calculation model of forecasting trend, and analyzes the quantity of electricity demand in the forecast period. Load and load characteristics and the company's electricity sales and other related indicators forecast information, strengthen planning, operation, marketing information sharing, reduce the cost of repeated construction input, reduce the workload of professional departments. Effectively improve the timeliness, reliability and accuracy of forecast data processing in the electricity market, improve the value of data use, provide more accurate prediction support data for the company's leadership and management, and promote the efficiency of business operations. Improve the efficiency of electric power enterprises to serve the society. This article mainly carries on the innovation research of the related theories and methods from two aspects. First of all, this paper puts forward the construction of electricity market analysis and prediction system based on power big data platform, adopting big data ecosystem technology, such as data collection, distributed storage, distributed processing and so on, which is based on Hadoop. Realize the centralized management of data resources, real-time monitoring and visual management. Secondly, the topic of household consumption habits based on temperature change is selected as a typical data mining application to realize the correlation analysis between daily maximum load and temperature, and the correlation analysis between daily average electricity consumption and temperature. The trend of load changing with temperature and the difference of electricity consumption based on temperature change in urban and rural areas are grasped, which can provide auxiliary analysis for orderly decision-making and measures of power consumption.
【學位授予單位】：華北電力大學(北京)
【學位級別】：碩士
【學位授予年份】：2017
【分類號】：TP311.13;F426.61

【引證文獻】

相關碩士學位論文 前1條

1 龔澤威一;基于機器學習的居民用電行為分析[D];昆明理工大學;2018年

，

本文編號：2436503