Research On Load Characteristics And Forecasting Of Heze Power Grid Based On Big Data

Power system load forecasting is one of the key contents of the power company’s work.High-quality load forecasting has a key guiding role for the start-stop,power-off plan and operation mode adjustment of the generator set.At the same time,the safe and stable operation of the power system and the operating efficiency of the power grid enter:prises also put forward higher requirements for load forecasting.How to improve the accuracy of load forecasting has become an important and arduous task.With the continuous improvement of the level of science and technology,the complexity of the basic data that can be used for load forecasting is increasing,and the data scale is growing rapidly.This also puts more stringent standards on load forecasters.The traditional load forecasting method can not meet the processing requirements of big data in terms of the coverage of the influencing factors and the prediction speed and accuracy.It has a certain gap compared with the requirements of the actual production and the superior unit.Based on the big data research,the load forecasting applied to Heze Power Grid It is extremely urgent for the platform to carry out high-quality load forecasting.This paper first introduces various indicators of load characteristics,and statistically analyzes the characteristics of the Heze grid load according to the time dimension on the year,month,week and typical day,and then the temperature,air pressure,relative humidity,etc.that affect the load fluctuation.The factors were studied on a case-by-case basis,and the quantitative calculation of the factors affecting the high temperature load in summer was carried out.The multi-dimensional influencing factors are summarized,and the data training is carried out according to various factors.The reasonable kernel function is selected and the load forecasting under multi-dimensional load influencing factors is carried out,and the prediction accuracy has been improved.However,for load data with complicated feature quantities,it is difficult to meet the actual work requirements in terms of speed using a single machine for load forecasting.In order to further improve the processing capacity of load big data,a Hadoop-based Heze grid load forecasting platform was built.The load factor of Heze grid was used as the characteristic quantity of each time load to construct big data samples,and the processing flow of abnormal data was introduced in detail.Parallel optimization calculation is completed,and functions such as data integration,load prediction,and result visibility are realized,and the application standard is achieved in both prediction accuracy and prediction speed.Furthermore,the Heze grid load forecasting platform is used to predict the load during the normal period and the rest period.The prediction accuracy rate is 97.88%and 97.63%respectively,which is greatly improved compared with the existing forecast level and released at a certain level.The work pressure of the load forecaster.Due to the underdeveloped economic attributes of Heze,the summer load of Heze Power Grid shows that the baseline load and air-conditioning usage increase with the year,but the growth rate varies with the macroeconomic situation every year.This is the load based on historical big data samples.The prediction accuracy caused a certain amount of interference.In order to further improve the prediction accuracy under high temperature load in summer,the load forecasting big data analysis was used to find a specific sample interval under high temperature weather,and the prediction result was adjusted based on this.Better application results.The research results can be applied to the load forecasting work of Heze Power System,which has a strong guiding role in improving the accuracy of load forecasting.