基于安徽省统计数据的电力需求预测模型研究

发布时间：2019-02-19 22:12

【摘要】：电力需求预测是电力系统优化调度的基础,其预测精度的提高,对电力行业乃至整个国民经济的发展都具有重要的意义。安徽省目前处于经济转型发展的重要时期,用电结构发生很大变化,而安徽省电力需求除受到经济发展的影响之外,还受到其他外在因素的影响,如当前经济结构的变化、气象因素等,其中,气象因素对短期电力需求的影响较显著。而在以往的中长期电力需求预测中,气象因素的作用并不能有效地体现出来。因此,研究气象因素对电力需求的影响机理,建立合适的预测模型能够进一步提高电力需求预测精度。 本文首先分析了安徽省1990年以来用电结构、产业结构的变化,定性分析了二者之间的关联性,发现用电结构很好地反映了目前安徽省经济发展及产业结构的现状。安徽省目前虽然正处于重工业化时期,但第二产业用电量效率有很大提升空间,相反,第三产业电力投入成本较大却没有相应地增加产出比例,可见安徽省经济结构有待进一步优化调整,通过内部升级以满足低成本高效率的经济增长要求,经济结构的优化升级将会带动电力行业用电效率的提升。其次,通过主成分分析确定了影响安徽省电力需求的主要因素,如宏观经济形势、居民生活水平、电价水平及气候变化,由此确定了气象因素对电力需求的影响作用不可忽视。再后,本文以气温为例重点剖析了安徽省气象因素对电力需求的影响机制,得出结论是气温对用电量的影响是分级而言的,气温基数越高,单位气温升高会导致用电量增加越多。 根据描述性分析的结论,本文参照宏观经济波动的研究方法,利用趋势分解从全社会用电量中分离出气象用电量和趋势用电量。通过时间趋势脱离法、季节调整法、H-P滤波分解法这三种趋势分解方法结果的比较,本文选择了季节调整法作为安徽省月度用电量数据的气象用电分离方法。气象用电量主要为气象因素引起的用电量的短期波动,根据每月不同的气象影响因素建立分月的气象用电预测模型；而趋势用电量是受经济因素影响的用电需求量,具有稳定的增长趋势,通过趋势方程预测,二者合并得到的全社会用电量的预测模型。经过验证,该模型预测精度较高,由此说明气象用电量分离方法在预测短期电力需求方面是有效可行的,可以作为中长期电力需求预测的辅助模型。 本文最后利用传统的长期均衡模型预测方法对安徽省未来三年的电力需求情况进行了预测分析,经数据验证,预测效果较好。 本文结论显示,安徽省目前已经进入重工业时代的加速期,未来一段时期,安徽省电力需求仍会保持上涨趋势,用电增速大于GDP增速的可能性较大,电力消费弹性系数从2013年开始将进入大于1的增长阶段。加之目前“稳增长、转方式、调结构”的宏观经济政策,安徽省电力消费增长将超前于经济增长,这也是促进安徽省加快产业结构调整及技术升级步伐的信号。
[Abstract]:The power demand forecast is the basis of the power system optimization and scheduling, and the prediction accuracy is improved, and it is of great significance to the development of the power industry and the whole national economy. At present, Anhui province is in the important period of economic transformation and development, and the electric structure has changed greatly, and the power demand of Anhui province is affected by other factors, such as the change of the current economic structure, the meteorological factors and so on, in addition to the influence of economic development. The effect of meteorological factors on short-term power demand is significant. In the past long-and medium-term power demand forecast, the role of meteorological factors can not be reflected effectively. Therefore, the influence mechanism of the meteorological factors on the power demand is studied, and a suitable prediction model can be established to further improve the power demand forecasting precision. This paper first analyzes the change of the electric structure and the industrial structure since 1990 in Anhui province, and analyzes the relationship between the two. It is found that the structure of the electric power has well reflected the present economic development and the present industrial structure of Anhui Province. In Anhui province, although it is in the period of heavy industrialization, the electricity consumption efficiency of the second industry is greatly improved. On the contrary, the cost of the third industry's power input is relatively large, and the output ratio is not increased accordingly, and the economic structure of Anhui Province is to be further optimized. In addition, through the internal upgrade to meet the economic growth requirements of low cost and high efficiency, the optimization and upgrading of the economic structure will drive the power utilization efficiency of the power industry Then, the main factors, such as the macro-economic situation, the living standard of the residents, the electricity price level and the climate change, are determined by the principal component analysis, and the effect of the meteorological factors on the power demand is determined. It is concluded that the effect of the temperature on the electricity consumption is the classification, the higher the temperature base, the higher the air temperature, and the higher the electricity consumption. According to the conclusions of the descriptive analysis, this paper, referring to the research method of macro-economic fluctuation, uses the trend decomposition to separate the meteorological power consumption and the trend from the total social electricity consumption. Based on the comparison of the results of the three trend decomposition methods, the time trend separation method, the seasonal adjustment method and the H-P filter decomposition method, the seasonal adjustment method is selected as the meteorological power of the monthly electricity consumption data in Anhui province. The meteorological power consumption is mainly the short-term fluctuation of the electricity consumption caused by the meteorological factors, and the monthly meteorological electricity forecast model is set up according to the different weather influence factors of each month; and the trend electricity consumption is the demand for electricity consumption which is influenced by the economic factors and has a stable increase. The long-term trend is predicted by the trend equation, and the combination of the two is the pre-set of the total social electricity consumption. The model is proved to be effective in predicting short-term power demand, and it can be used as an auxiliary to the long-and medium-term power demand forecast. In this paper, the traditional long-term equilibrium model is used to forecast the power demand of Anhui in the next three years. The results of this paper show that Anhui province has entered the accelerating period of heavy industry, and the demand of electricity in Anhui province is still rising in the coming period, and the growth rate of electricity is greater than that of GDP. The possibility of speed is large, and the elastic coefficient of electric consumption will start in 2013. In addition to the current 鈥淪table growth, rotation mode and modulation structure鈥，

本文编号：2426931