Projects per year
An accurate power load prediction in smart grid plays an important role in maintaining the balance between power supply and demand and thus ensuring the safe and stable operation of power system. In this paper we develop a hybrid power load prediction method, which involves three main steps: data decomposition with the empirical mode decomposition method, data processes with the minimal redundancy maximal relevance method and the weighted gray relationship projection algorithm, and support vector machine prediction, whose parameters are optimized through the particle swarm optimization algorithm with a second-order oscillation and repulsive force factor. Moreover, we predict the power load with our hybrid forecasting method based on the real dataset from the electricity market in Singapore, and also compare our prediction results with those by using other forecasting methods. Our comparison results show that our novel hybrid method possesses a high accuracy in both the level and directional predictions.
Bibliographical noteFunding Information:
This work was supported by the National Natural Science Foundation of China [No. 72171126 ], Ministry of Education Project of Humanities and Social Science [No. 20YJA630009 ], Faculty Research Grant of Lingnan University under the grant number DB21B1 .
© 2022 Elsevier Inc.
- Empirical mode decomposition
- Minimal redundancy maximal relevance
- Power load forecasting
- Second-order oscillation and repulsion particle swarm optimization
- Weighted gray relation projection algorithm
FingerprintDive into the research topics of 'Forecasting power load: A hybrid forecasting method with intelligent data processing and optimized artificial intelligence'. Together they form a unique fingerprint.
- 1 Finished
1/06/21 → 31/05/23
Project: Grant Research