Electricity Correlation Evaluation Based on Improved Logistic Algorithm Integrating Periodic Characteristics of Load and Temperature

Abstract

The power system (abbreviated as PS for convenience) is one of the indispensable infrastructures in modern society, and its stable operation is crucial for ensuring the normal operation of the national economy and society. With the continuous expansion and complexity of the power grid, power correlation analysis has become increasingly important in the operation, planning, and management of the power system. Temperature is one of the main factors affecting power load (abbreviated as PL for convenience), and how to integrate the periodic characteristics of temperature with load analysis has become a top priority. This article improved the logistic algorithm and applied it to the power correlation analysis of combined load and temperature periodic characteristics and collected four seasonal PL parameters and temperature parameters from January to December 2022. The study analyzed the correlation between PL and temperature periodic characteristics, and also compared the accuracy of PS correlation analysis using the logistic algorithm and improved logistic algorithm. According to the experimental results, it could be concluded that at 1 and 2 o’clock on 1 January 2022, the temperature was at the lowest, both of which were −3 °C, while the PS load was 1000 MW and 1100 MW, respectively. It could be seen that in winter, as the temperature was lower the PL increased. In July 2022, the load and temperature of the PS were continuously increasing, reaching their maximum at 10 o’clock and it could be observed that as the temperature increased, the PL also increased. This was because both low and high temperatures increased the operation of the power equipment, thereby increasing the PL. It was also confirmed that fusing multiple features and adopting an improved logistic algorithm could improve the accuracy of the prediction results. The improved logistic algorithm could be applied to related fields such as PL forecasting and provide a scientific decision-making basis for the power industry. This could also provide a reference for data analysis and prediction in other fields.

Keywords

• power load
• temperature periodicity
• electricity correlation
• improved logistic algorithm
• machine classification algorithm