Scientists from China's Hubei University of Technology have proposed a novel deep-learning model for ultra-short-term PV power prediction.
The new technique combines self-attention temporal convolutional networks (SATCN), bidirectional long short-term memory networks (BiLSTM), and dung beetle optimization (DBO) and is therefore called DBO-SATCN-BiLSTM.
“Existing photovoltaic power generation prediction methods have several shortcomings. Meta-heuristic learning methods, such as neural networks, are complex and time-consuming to train, and often rely on empirical parameter adjustment, which cannot obtain the optimal results,” the group explained. “Swarm intelligence algorithms are expected to improve the optimal parameters, but the optimization algorithms currently used still have certain limitations. For example, some optimization algorithms are relatively simple and easy to fall into local optimality. To address the above issues, this paper proposes a DBO-SATCN-BiLSTM.”
SATCN incorporates a convolutional neural network customized to analyze time series data to mitigate gradient vanishing or explosion. The proposed model also uses its self-attention mechanism to capture the dependencies between different time steps in time series data. The BiLSTM layer, however, uses its bidirectional data processing capabilities to obtain the forward and backward data change characteristics in the sequence data.
The DBO section of the model works on the external level by identifying and optimizing the hyperparameters of the hybrid SATCN-BiLSTM model. Unlike parameters that are learned during training, hyperparameters are the external configurations that guide how a model works. The model is named after dung beetles, as it imitates the way they roll dung balls to their nests, using smell and the position of the sun or moon to navigate. In that way, the optimization looks for optimal solutions based on some external guidance.
The model consists of an input layer, three layers of SATCN residual blocks, two layers of BiLSTM, and a fully connected layer. The SATCN structure enables the extraction of the temporal features from photovoltaic power data, while BiLSTM further captures the temporal correlation between forward and backward features.
The model was trained and tested on year-long information from a 30 MW PV power plant in Shaanxi Province, China. Using parameters such as temperature, humidity, barometric pressure, and solar radiation every 15 minutes during this year, the model was asked to predict the power in the following step of 15 minutes. It was also asked to perform a multi-step prediction of 45 minutes, meaning three points in the future.
The results were compared to those of seven reference models: Convolutional neural network (CNN); BiLSTM alone; temporal convolutional network (TCN); hybrid of CNN-BiLSTM; hybrid of TCN-BiLSTM, SATCN-BiLSTM with particle swarm optimization (PSO-SATCN-BiLSTM); and SATCN-BiLSTM with salp swarm algorithm (SSA-SATCN-BiLSTM).
In single-step prediction, the DBO-SATCN-BiLSTM was found to achieve a root mean square error (RMSE ) of 0.357 when predicting one year's PV power.
“This demonstrates a significant reduction of 33.1%, 23.4%, and 18.1%compared to the CNN, TCN, and BiLSTM,” the academics stressed. “It outperforms CNN-BiLSTM, TCN-BiLSTM, PSOSATCN- BiLSTM, and SSA-SATCN-BiLSTM by margins of 17.5%, 10%, 2.9% and 2.4%,” the group said. “In terms of multi-step (3-step) prediction, as the prediction range increases, errors escalate across all models. The DBO-SATCN-BiLSTM stands out with an RMSE of 0.437, indicating substantial improvements by 52.3%, 32.4%, 32.9%, 31.5%, 31.1%, 9.5%, and 4.7% when compared with the other seven models.”
Their findings were presented in “Dung beetle optimization algorithm-based hybrid deep learning model for ultra-short-term PV power prediction,” published in iScience.
This content is protected by copyright and may not be reused. If you want to cooperate with us and would like to reuse some of our content, please contact: editors@pv-magazine.com.
By submitting this form you agree to pv magazine using your data for the purposes of publishing your comment.
Your personal data will only be disclosed or otherwise transmitted to third parties for the purposes of spam filtering or if this is necessary for technical maintenance of the website. Any other transfer to third parties will not take place unless this is justified on the basis of applicable data protection regulations or if pv magazine is legally obliged to do so.
You may revoke this consent at any time with effect for the future, in which case your personal data will be deleted immediately. Otherwise, your data will be deleted if pv magazine has processed your request or the purpose of data storage is fulfilled.
Further information on data privacy can be found in our Data Protection Policy.