With the continuous maturity of China's power grid as well as the advancement of electricity market reform in China, accurate and efficient investment decision has become an inevitable requirement of power grid enterprises. However, China's Power grid investment demand has complicated nonlinear and non-stationary characteristics due to it's complex causes of formation, thus make it hard to be forecasted. Aiming at this problem, this paper puts forward a novel hybrid VMD-RELMLOO-PSOSVM forecasting model based on variational mode decomposition (VMD), leave-one-out cross validation error based optimal regularized extreme learning machine (RELM-LOO) and support vector machines optimized by particle swarm optimization algorithm (PSO-SVM). Firstly, the VMD method is employed to decompose the original power grid investment data sequence into several modes which have specific sparsity properties while producing main signal. Then, according to the different characteristics of each subsequence, the RELM-LOO and PSO-SVM model will be used to forecast different modes, respectively; Next, the prediction results of all modes are aggregated to obtain the final prediction results of China's power grid investment demand. Finally, this paper predicts China's power grid investment demand from 2018 to 2020 under 5 different scenarios based on the proposed VMD-RELMLOO-PSOSVM hybrid forecasting model.
[1]
A. Selakov,et al.
Hybrid PSO-SVM method for short-term load forecasting during periods with significant temperature variations in city of Burbank
,
2014,
Appl. Soft Comput..
[2]
Dominique Zosso,et al.
Variational Mode Decomposition
,
2014,
IEEE Transactions on Signal Processing.
[3]
Qianqian Wu,et al.
A Hybrid Forecasting Model Based on EMD-GASVM-RBFNN for Power Grid Investment Demand
,
2018,
Mathematical Problems in Engineering.
[4]
Yaru Han,et al.
Forecasting of Power Grid Investment in China Based on Support Vector Machine Optimized by Differential Evolution Algorithm and Grey Wolf Optimization Algorithm
,
2018
.
[5]
Chee Kheong Siew,et al.
Extreme learning machine: Theory and applications
,
2006,
Neurocomputing.
[6]
Kai Zhang,et al.
Extreme learning machine and adaptive sparse representation for image classification
,
2016,
Neural Networks.
[7]
Rui Zhang,et al.
Short-term load forecasting of Australian National Electricity Market by an ensemble model of extreme learning machine
,
2013
.