With the continuous progress of people's living standard, electricity has been applied to every aspect of people's daily life, and people's demand for electricity is also growing. Photovoltaic power generation, as a kind of power generation method, has been developing rapidly in recent years both domestically and internationally, and has made great breakthroughs. However, PV power generation is subject to fluctuations due to weather and other factors, and how to accurately forecast PV power generation is an issue that needs to be considered. In this study, BiTCN-DCMA model based on bidirectional temporal convolutional neural network is proposed for the prediction of photovoltaic power generation. BiTCN-DCMA uses dynamic convolution for forward and backward optimization to achieve features that can adapt to different scales; the bi-directional fused feature information enters the attention layer, which consists of a bi-directional long and short-term memory network and a multi-head attention mechanism, and the attention layer is able to carry out the contextual data that captures the complex interactions between different locations in the sequence, and these optimization methods strengthen the linear layer's prediction of power generation Accuracy. The results show that the root mean square error (RMSE) of BiTCN-DCMA is reduced by 75.81% and the correlation coefficient R2 is improved by 12.50% compared to the Transformer model, showing good prediction accuracy and stability.

Yifei Ren. BiTCN-DCMA: Research on Prediction Method of Photovoltaic Power Generation Based on Dynamic Convolution and BiTCN-BiLSTM. Academic Journal of Computing & Information Science (2024), Vol. 7, Issue 7: 23-30. https://doi.org/10.25236/AJCIS.2024.070704.

[1] Chen S, Bai Y, Zhang Y, et al. A Framework of Decentralized Electricity Market Based on The Collaborative Mechanism of Blockchain and Edge Computing[C]//2019 IEEE International Conference on Service Operations and Logistics, and Informatics (SOLI). 2019: 219-223.

[2] Fu J, Shi Z, Zhang Z. An Edge Computing Framework for Digital Grid[C]//2020 IEEE 3rd International Conference on Electronic Information and Communication Technology (ICEICT). 2020: 670-673.

[3] He X, Cai W, He H, et al. Top-level design of electricity consumption information collection system based on edge computing[C]//2021 13th International Conference on Measuring Technology and Mechatronics Automation (ICMTMA). 2021: 522-526.

[4] Yao J, Li Z, Li Y, et al. Cost-Efficient Tasks Scheduling for Smart Grid Communication Network with Edge Computing System[C]//2019 15th International Wireless Communications & Mobile Computing Conference (IWCMC). 2019: 272-277.

[5] Chen S, Wen H, Wu J, et al. Internet of Things Based Smart Grids Supported by Intelligent Edge Computing[J]. IEEE Access, 2019, 7: 74089-74102.

[6] Hu J, Li Y, Zhao G, et al. Deep Reinforcement Learning for Task Offloading in Edge Computing Assisted Power IoT[J]. IEEE Access, 2021, 9: 93892-93901.

[7] Xu Z, Han G, Zhu H, et al. Adaptive DE Algorithm for Novel Energy Control Framework Based on Edge Computing in IIoT Applications[J]. IEEE Transactions on Industrial Informatics, 2021, 17(7): 5118-5127.

[8] Wang Y, Shen Y, Mao S, et al. LASSO and LSTM Integrated Temporal Model for Short-Term Solar Intensity Forecasting[J]. IEEE Internet of Things Journal, 2019, 6(2): 2933-2944.

[9] Wang Y, Shen Y, Mao S, et al. Adaptive Learning Hybrid Model for Solar Intensity Forecasting[J]. IEEE Transactions on Industrial Informatics, 2018, 14(4): 1635-1645.

[10] Wang Y, Zou H, Chen X, et al. Adaptive Solar Power Forecasting based on Machine Learning Methods[J]. Applied Sciences, 2018, 8(11): 2224.

[11] Jan Z, Ahamed F, Mayer W, et al. Artificial intelligence for industry 4.0: Systematic review of applications, challenges, and opportunities[J]. Expert Systems with Applications, 2023, 216: 119456.

[12] Ke S, Dai G, Pan H, et al. Intelligent vineyard blade density measurement method incorporating a lightweight vision transformer[J]. Egyptian Informatics Journal, 2024, 26: 100456.

[13] Dai G, Fan J, Tian Z, et al. PPLC-Net:Neural network-based plant disease identification model supported by weather data augmentation and multi-level attention mechanism[J]. Journal of King Saud University - Computer and Information Sciences, 2023, 35(5): 101555.

[14] Zhu Z, Lei Y, Qi G, et al. A review of the application of deep learning in intelligent fault diagnosis of rotating machinery[J]. Measurement, 2023, 206: 112346.

[15] Xiang X, Li X, Zhang Y, et al. A short-term forecasting method for photovoltaic power generation based on the TCN-ECANet-GRU hybrid model[J]. Scientific Reports, 2024, 14(1): 6744. 

[16] Li S, Ma W, Liu Z, et al. Short-Term Prediction of Wind Power Based on NWP Error Correction with TimeGAN and LSTM-TCN[C]//Yadav S, Arya Y, Muhamad N A, et al. Energy Power and Automation Engineering. Singapore: Springer Nature, 2024: 939-948.

[17] Mo J, Wang R, Cao M, et al. A hybrid temporal convolutional network and Prophet model for power load forecasting[J]. Complex & Intelligent Systems, 2023, 9(4): 4249-4261.

[18] Gao Q, Chen J, Ji Y, et al. Short-term Load Prediction Based on Whale Optimized WTCN-BiGRU Hybrid Neural Network[C]//Jia Y, Zhang W, Fu Y, et al. Proceedings of 2022 Chinese Intelligent Systems Conference. Singapore: Springer Nature, 2022: 806-814.

[19] Chen Y, Dai X, Liu M, et al. Dynamic Convolution: Attention Over Convolution Kernels[C]//Proceedings of the IEEE/CVF Conference on Computer Vision and Pattern Recognition. 2020: 11030-11039.

[20] Zhang D, Chen B, Zhu H, et al. Short-term wind power prediction based on two-layer decomposition and BiTCN-BiLSTM-attention model[J]. Energy, 2023, 285: 128762.

[21] Woźniak M, Wieczorek M, Siłka J. BiLSTM deep neural network model for imbalanced medical data of IoT systems[J]. Future Generation Computer Systems, 2023, 141: 489-499.

[22] Altaheri H, Muhammad G, Alsulaiman M. Dynamic Convolution With Multilevel Attention for EEG-Based Motor Imagery Decoding[J]. IEEE Internet of Things Journal, 2023, 10(21): 18579-18588. 

[23] Ge Y, Ma J, Sun G. A structural pruning method for lithium-ion batteries remaining useful life prediction model with multi-head attention mechanism[J]. Journal of Energy Storage, 2024, 86: 111396.