PV Power Forecasting Based on Relevance Vector Machine with Sparrow Search Algorithm Considering Seasonal Distribution and Weather Type

Abstract

Accurate photovoltaic (PV) power forecasting is indispensable to enhancing the stability of the power grid and expanding the absorptive photoelectric capacity of the power grid. As an excellent nonlinear regression model, the relevance vector machine (RVM) can be employed to forecast PV power. However, the optimization of the free parameters is still a key problem for improving the performance of the RVM. Taking advantage of the strong global search capability, good stability, and fast convergence rate of the sparrow search algorithm (SSA), this paper optimizes the parameters of the RVM by using the SSA to develop an excellent RVM (called SSA-RVM). Consequently, a novel hybrid PV power forecasting model via the SSA-RVM is proposed to perform ultra-short-term (4 h ahead) prediction. In addition, the effects of seasonal distribution and weather type on PV power are fully considered, and different seasonal prediction models are established separately to improve the prediction capability. The benchmark is used to verify the accuracy of the SSA-RVM-based forecasting model under various conditions, and the experiment results demonstrate that the proposed SSA-RMV method outperforms the traditional RVM and support vector machine models, and it even shows a better prediction effect than the RVM models with other optimization approaches.

Fault diagnosis of rolling bearing based on SEMD and ISSA-KELMC

“…To enhance optimization algorithms, Xue introduced a novel swarm intelligence method called the sparse search algorithm (SSA) [27]. Owing to its simplicity, rapid convergence, and stability, SSA has found applications in microgrid operational planning [28], power forecasting [29], and route planning [30].…”

Bayesian Optimization - LSTM Modeling and Time Frequency Correlation Mapping Based Probabilistic Forecasting of Ultra-short-term Photovoltaic Power Outputs

“…Through the Gaussian process, the mean absolute error between the real PV power and the model forecasting power can be obtained [17]. The optimal local solution will be found by the sample calculation.…”

“…Deep learning is widely used for both deterministic and probabilistic photovoltaic power forecasting [17]. In this paper, an improved BO-LSTM deep learning forecasting algorithm based on output power data time-frequency analysis and feature extraction algorithms is proposed.…”

Ultra-short-term PV power prediction using optimal ELM and improved variational mode decomposition

“…The technique was introduced to address the issue of global optimization by imitating the feeding habits of a population of sparrows (Xue and Shen, 2020). Compared with the traditional meta-heuristic learning algorithm, SSA has certain advantages in convergence speed and stability (Ma et al, 2022). SSA assumes that there are three kinds of sparrows: discoverers, followers, and guards, each sparrow's position correspond to a solution.…”

“…In order to evaluate effects of various decomposition techniques on prediction outcomes, three different decomposition methods: ISSA-VMD, SSA-VMD, and VMD, are used in combination with the proposed ISSA-ELM model for prediction comparison. Due to the different fluctuations of PV power under different weather conditions, in order to reduce the impact of weather on the prediction results, the paper forecasts the three different weather conditions (Ma et al, 2022). Figures 12-14 demonstrates that the decomposition effect of VMD will impact prediction accuracy.…”