Short-term load forecasting method based on somatosensory temperature

Abstract

The invention discloses a short-term load forecasting method based on the somatosensory temperature. The method is for forecasting the load in summer and winter, and includes the steps of reading historical data; quantifying the historical data; calculating the somatosensory temperature; collecting information of a day to be forecast; forecasting the maximum daily load; selecting a load trend curve: selecting a historical day most approaching the day to be forecast as a trend similar day through a similar day method and taking the load curve line type of the day as the line type of the day to be forecast; and forecasting the daily load: selecting maximum and minimum load values of the trend similar day, subtracting load data of 96 points of the day from the minimum load value, then dividing the result by a difference between the maximum load and the minimum load, next, multiplying normalized data by a difference between the maximum forecast load and the minimum forecast load of the day to be forecast, and adding a minimum forecast load value to obtain a load forecast value of the whole day. According to the invention, the speed of load forecasting is greatly improved and the accuracy of load forecasting is ensured.

Description

technical field [0001] The invention relates to the field of electric power technology, in particular to a short-term load forecasting method based on sensible temperature. Background technique [0002] Power load forecasting is the main basis for guiding power grid planning, construction, operation, and maintenance. Short-term load forecasting is an important part of load forecasting. Accurate short-term load forecasting can guide the power transmission and distribution system to adjust the operation mode in time, and reasonably arrange the shutdown and transmission maintenance plan. Therefore, high-quality load forecasting can guide power grid companies to make the most reasonable use of human, financial, material and other resources while meeting the requirements of power supply quality to obtain optimal social and economic benefits. In the electricity market environment, the importance of short-term load forecasting has become increasingly prominent. [0003] Over the...

AI Technical Summary

Problems solved by technology

[0006] The calculation speed of the time series method is fast, but its model cannot specifically reflect the factors affecting the load, and the construction of the model requires extensive experience

[0007] The neural network method has high prediction accuracy, but has the disadvantages of long learning time and easy local minimum points

[0008] The fuzzy forecasting method can handle the uncertainty of load change well, but the forecasting accuracy of the fuzzy forecasting method alone is not high

[0009] If the trend analysis method can select an appropriate model and better fit the actual load curve, the prediction result will be better, but the prediction results of different models are quite different, so it is very difficult to choose a suitable model

[0010] To sum up, there is still no effective solution to the short-term load forecasting problem in the prior art

Images