Abstract
To make full use of the effective information of spatiotemporal data for curve prediction, a parallel spatiotemporal fusion network is proposed to predict curves. A differential feature fusion module is designed for spatiotemporal feature fusion, which makes the fused features have richer feature information. The designed adaptive adjusted Huber loss can find adjustable parameters suitable for the curve according to the input data and reduce the influence of outliers on the model prediction. The experiment used the logging data of Huabei Oilfield as the data set. The adaptive adjusted Huber loss outperforms the MAE, MSE, and Huber losses by 9%, 4%, and 2% on the correlation coefficient metric. The parallel spatiotemporal fusion network has 94.11%, 96.81%, 99.90%, 97.93%, and 99.19% accuracy in AC, CAL, SP, GR, and R4. Compared with the LSTM network, the accuracies are improved by 2.23%, 1.18%, 0.01%, 1.42%, and 1.45%, respectively. The experimental results verify that the parallel spatiotemporal fusion network proposed in this paper has good robustness and can combine spatiotemporal data to achieve higher curve prediction accuracy.








Similar content being viewed by others
References
Ciulla, A.G.: Modelling and analysis of real-world wind turbine power curves: assessing deviations from nominal curve by neural networks. Renew. Energy 140, 477–492 (2019). https://doi.org/10.1016/j.renene.2019.03.075
Ghatak, M.D., Ghatak, A.: Artificial neural network model to predict behavior of biogas production curve from mixed lignocellulosic co-substrates. Fuel 232, 178–189 (2018). https://doi.org/10.1016/j.fuel.2018.05.051
Zheng, L.: Short-term parking demand prediction method based on variable prediction interval. IEEE Access 8, 58594–58602 (2020). https://doi.org/10.1109/ACCESS.2020.2976433
Zhang, R., Zou, Q.: Time series prediction and anomaly detection of light curve using LSTM neural network. J. Phys Conf. Ser. 1061, 012012 (2018)
Li, J.: Robust face recognition using the deep C2D-CNN model based on decision-level fusion. Sensors 18(7), 2080 (2018). https://doi.org/10.3390/s18072080
Yang, Y.X.: Face recognition using the SR-CNN model. Sensors 18(12), 4237 (2018). https://doi.org/10.3390/s18124237
Zhao, Y.F.: Intelligent Recognition of fatigue and sleepiness based on inceptionv3-LSTM via multi-feature fusion. IEEE Access 8, 144205–144217 (2020)
Yang, L.X., Sam, Z.D., Ji, L.L.: S-wave velocity prediction for complex reservoirs using a deep learning method. SEG Tech. Program Expand. Abstr. (2019). https://doi.org/10.1190/segam2019-3216200.1
Li, J.: Prediction of I-V characteristic curve for photovoltaic modules based on convolutional neural network. Sensors 20(7), 2119 (2020). https://doi.org/10.3390/s20072119
Erofeev, A.: Prediction of porosity and permeability alteration based on machine learning algorithms. Transp. Porous Med. 128, 677–700 (2019). https://doi.org/10.1007/s11242-019-01265-3
Job, G.U.: A new approach for porosity and permeability prediction from well logs using artificial neural network and curve fitting techniques: a case study of Niger Delta Nigeria. J. Appl. Geophys. 183, 104207 (2020). https://doi.org/10.1016/j.jappgeo.2020.104207
Zhang, D., Chen, Y., Meng, J.: Synthetic well logs generation via recurrent neural networks. Pet. Explor. Dev. 45(4), 629–639 (2018)
Yang, B.: Time series analysis and long short-term memory neural network to predict landslide displacement. Landslides 16, 677–694 (2019). https://doi.org/10.1007/s10346-018-01127-x
Rocha, L.G.: Short-term electric load forecasting using neural networks: a comparative study. In: 2020 IEEE PES Transmission & Distribution Conference and Exhibition - Latin America (T&D LA), 1–6 (2020) https://doi.org/10.1109/TDLA47668.2020.9326196
Song, L., Pin, L.: Short-term Load Forecasting Model Based on Attention Mechanism and Gated Recurrent Unit. In: IEEE 8th International Conference on Advanced Power System Automation and Protection (APAP), 1784–1788, (2019)
Wang, J., Cao, J., Yuan, S.: Shear wave velocity prediction based on adaptive particle swarm optimization optimized recurrent neural network. J. Pet. Sci. Eng. 194, 107466 (2020)
Geng, Z., Wang, Y.F.: Physics-guided deep learning for predicting geological drilling risk of wellbore instability using seismic attributes data. Eng. Geol. 279, 105857 (2020). https://doi.org/10.1016/j.enggeo.2020.105857
Zhang, D.: Integrating feature selection and feature extraction methods with deep learning to predict clinical outcome of breast cancer. IEEE Access 6, 28936–28944 (2018). https://doi.org/10.1109/ACCESS.2018.2837654
Zhang, M.: The methodology to run CBM reservoir prediction by maximum utilizing seismic and well data in Australia. In: IFEDC 2019. Springer Series in Geomechanics and Geoengineering. Springer, Singapore. https://doi.org/10.1007/978-981-15-2485-1_73
Deng, T.Q.: Diagenetic facies classification in the arbuckle formation using deep neural networks. Math. Geosci. 53(7), 1491–1512 (2021). https://doi.org/10.1007/s11004-021-09918-0
Li, T., Hua, M., Wu, X.: A Hybrid CNN-LSTM model for forecasting particulate matter (PM2.5). IEEE Access 8, 26933–26940 (2020)
Qin, D.: A novel combined prediction scheme based on CNN and LSTM for urban PM25 concentration. IEEE Access 7, 20050–20059 (2019)
Li, P., Mohamed, A.A., Yuan, J.H.: Real-time crash risk prediction on arterials based on LSTM-CNN. Accid. Anal. Prev. 135, 105371 (2020). https://doi.org/10.1016/j.aap.2019.105371
Xiao, Q.: Multi-information spatial-temporal LSTM fusion continuous sign language neural machine translation. In IEEE Access (2020). https://doi.org/10.1109/ACCESS.2020.3039539
Eapen, J., Bein, D., Verma, A.: Novel Deep Learning Model with CNN and Bi-Directional LSTM for Improved Stock Market Index Prediction. In: IEEE, 0264–0270 9th Annual Computing and Communication Workshop and Conference (CCWC), (2019) https://doi.org/10.1109/CCWC.2019.8666592
Kong, Z.: Multimodal feature extraction and fusion deep neural networks for short-term load forecasting. IEEE Access 8, 185373–185383 (2020). https://doi.org/10.1109/ACCESS.2020.3029828
Yu, Q.: High-performance SAR automatic target recognition under limited data condition based on a deep feature fusion network. IEEE Access 7, 165646–165658 (2019). https://doi.org/10.1109/ACCESS.2019.2952928
Lu, J.: Multi-feature fusion for enhancing image similarity learning. IEEE Access 7, 167547–167556 (2019). https://doi.org/10.1109/ACCESS.2019.2953078
Xu, Z.Y., Cao, Y., Kang, Y.: Deep spatiotemporal residual early-late fusion network for city region vehicle emission pollutionprediction. Neurocomputing 355, 183–199 (2019). https://doi.org/10.1016/j.neucom.2019.04.040
Zhou, T.: Multi-feature fusion based recognition and relevance analysis of propagation scenes for high-speed railway channels. IEEE Trans. Veh. Technol. 69(8), 8107–8118 (2020). https://doi.org/10.1109/TVT.2020.2999313
Liu, Z.G.: A lithological sequence classification method with well log via SVM-assisted bi-directional GRU-CRF neural network. J. Pet. Sci. Eng. 205, 108913 (2021). https://doi.org/10.1016/j.petrol.2021.108913
Elwardy, M.: Objective perceptual video quality prediction using spatial and temporal information differences. In: 2019 19th International Symposium on Communications and Information Technologies (ISCIT), 436–441(2019) https://doi.org/10.1109/ISCIT.2019.8905186
Acknowledgements
This work was supported in part by the Natural Science Foundation of Xinjiang Uygur Autonomous Region (2020D01A131), the Fund of Hubei Ministry of Education (B2019039), the Graduate Teaching and Research Fund of Yangtze University (YJY201909), the Teaching and Research Fund of Yangtze University (JY2019011), the Undergraduate Training Programs for Innovation and Entrepreneurship of Yangtze University under Grant Yz2020057、Yz2020059、Yz2020156, and the National College Student Innovation and Entrepreneurship Training Program (202110489003).
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
Springer Nature or its licensor holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.
About this article
Cite this article
He, ST., Wen, C., Xie, K. et al. Curve generation method of deep parallel spatiotemporal fusion network. SIViP 17, 1305–1313 (2023). https://doi.org/10.1007/s11760-022-02338-5
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11760-022-02338-5