Skip to main content
SpringerLink
Log in

Detecting broken receiver tubes in CSP plants using intelligent sampling and dual loss

  • Published:
Applied Intelligence Aims and scope Submit manuscript

Abstract

Concentrated solar power (CSP) is one of the growing technologies that is leading the process of change from fossil fuels to renewable energies for electricity production. The sophistication and size of the systems require an increase in maintenance tasks to ensure reliability, availability, maintainability, and safety. Currently, automatic detection of broken glass envelopes of receiver tubes in CSP plants using parabolic trough collector systems has two main drawbacks: 1) the devices in use need to be manually placed near the receiver tube, 2) the machine learning–based solutions have only been tested in constrained environments. We address both gaps by combining the data collected by an unmanned aerial vehicle with data provided by sensors placed within 7 real CSP plants. The resulting dataset is the first of this type and can help standardize research activities for the problem of fault detection in this type of power plant. Our work proposes supervised machine-learning algorithms for detecting broken envelopes of the receiver tubes in CSP plants. The proposed solution takes the class imbalance problem into account, boosting the accuracy of the algorithms for the minority class without harming the overall performance of the models. For a deep residual network, we solve an imbalance and a balance problem at the same time, which increases the recall of the minority class by 5% with no harm to the F1 score. Additionally, the random under-sampling technique boosts the performance of traditional machine learning models. The histogram gradient boost classifier was found to be the algorithm with the highest increase (3%) in the F1 score. To the best of our knowledge, this paper is the first to provide an automated solution to this problem using data from operating plants, drones, and highly unbalanced datasets.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2
Fig. 3
Fig. 4
Fig. 5
Fig. 6
Fig. 7
Fig. 8
Fig. 9
Fig. 10

Similar content being viewed by others

Data availability

The data that support the findings of this study will be made available upon reasonable request for academic use. Every request will be reviewed by Virtualmech, and the researcher will need to sign a data access agreement with Virtualmech after approval.

Notes

  1. RTSet can be requested from the corresponding author for academic and research purposes.

  2. https://virtualmech.com/

  3. National Renewable Laboratory US: https://www.nrel.gov/.

  4. The pressure values are usually analysed after applying a base 10 logarithm, which is why the next power of 10 was selected as reference.

  5. Training data is randomly split into a training and validation set

References

  1. Fuqiang W, Jianyu T, Lanxin M, Chengchao W (2015) Effects of glass cover on heat flux distribution for tube receiver with parabolic trough collector system. Energy Convers Manag 90:47–52

    Article  Google Scholar 

  2. Forristall R (2003) Heat transfer analysis and modeling of a parabolic trough solar receiver implemented in engineering equation solver. National Renew Energy Lab., Golden, CO. (US), Tech. Rep

  3. Zhang C, Soda P, Bi J, Fan G, Almpanidis G, García S, Ding W (2023) An empirical study on the joint impact of feature selection and data resampling on imbalance classification. Appl Intell 53(5):5449–5461

    Google Scholar 

  4. Espinosa-Rueda G, Hermoso JLN, Martinez-Sanz N, Gallas-Torreira M (2016) Vacuum evaluation of parabolic trough receiver tubes in a 50 mw concentrated solar power plant. Sol Energy 139:36–46

    Article  Google Scholar 

  5. Kesseli D, Chidurala V, Gooch R, Zhu G (2023) A combined computer vision and deep learning approach for rapid drone-based optical characterization of parabolic troughs. J Sol Energy Eng 145(2):021008

    Article  Google Scholar 

  6. Fernandes M, Corchado JM, Marreiros G (2022) Machine learning techniques applied to mechanical fault diagnosis and fault prognosis in the context of real industrial manufacturing use-cases: a systematic literature review. Appl Intell pp 1–35

  7. Benkercha R, Moulahoum S (2018) Fault detection and diagnosis based on c4.5 decision tree algorithm for grid connected pv system. Sol Energy 173:610–634

    Article  Google Scholar 

  8. Chen Z, Han F, Wu L, Yu J, Cheng S, Lin P, Chen H (2018) Random forest based intelligent fault diagnosis for pv arrays using array voltage and string currents. Energy Convers Manag 178:250–264

    Article  Google Scholar 

  9. Dhibi K, Fezai R, Mansouri M, Trabelsi M, Kouadri A, Bouzara K, Nounou H, Nounou M (2020) Reduced kernel random forest technique for fault detection and classification in grid-tied pv systems. IEEE J Photovoltaics 10(6):1864–1871

    Article  Google Scholar 

  10. Das S, Hazra A, Basu M (2018) Metaheuristic optimization based fault diagnosis strategy for solar photovoltaic systems under non-uniform irradiance. Renew Energy 118:452–467

    Article  Google Scholar 

  11. Niazi KAK, Akhtar W, Khan HA, Yang Y, Athar S (2019) Hotspot diagnosis for solar photovoltaic modules using a naive bayes classifier. Sol Energy 190:34–43

    Article  Google Scholar 

  12. Hafdaoui H, Chahtou A, Bouchakour S, Belhaouas N et al (2022) Analyzing the performance of photovoltaic systems using support vector machine classifier. Sustain Energy Grids Netw 29:100592

    Article  Google Scholar 

  13. Jufri FH, Oh S, Jung J (2019) Development of photovoltaic abnormal condition detection system using combined regression and support vector machine. Energy 176:457–467

    Article  Google Scholar 

  14. Adhya D, Chatterjee S, Chakraborty AK (2022) Performance assessment of selective machine learning techniques for improved pv array fault diagnosis. Sustain Energy Grids Netw 29:100582

    Article  Google Scholar 

  15. Aziz F, Haq AU, Ahmad S, Mahmoud Y, Jalal M, Ali U (2020) A novel convolutional neural network-based approach for fault classification in photovoltaic arrays. IEEE Access 8:41 889–41 904

  16. Chen Z, Chen Y, Wu L, Cheng S, Lin P (2019) Deep residual network based fault detection and diagnosis of photovoltaic arrays using current-voltage curves and ambient conditions. Energy Convers Manag 198:111793

    Article  Google Scholar 

  17. Meligy R, Klaina H, Picallo I, Lopez-Iturri P, Azpilicueta L, Astrain JJ, Rady M, Villadangos J, Alejos AV, Falcone F (2021) Iot enabled low cost distributed angle measurement fault detection system for lfr plants. IEEE Sensors J 21(21):24 855–24 868

  18. Ruiz-Moreno S, Gallego AJ, Sanchez AJ, Camacho EF (2022) Fault detection and isolation based on deep learning for a fresnel collector field. IFAC-PapersOnLine 55(6):563–568

    Article  Google Scholar 

  19. Rodríguez F, Chicaiza WD, Sánchez A, Escaño JM (2023) Updating digital twins: Methodology for data accuracy quality control using machine learning techniques. Comput Ind 151:103958

    Article  Google Scholar 

  20. Ruiz-Moreno S, Gallego AJ, Sanchez AJ, Camacho EF (2023) A cascade neural network methodology for fault detection and diagnosis in solar thermal plants. Renew Energy 211:76–86

    Article  Google Scholar 

  21. Arcos Jiménez A, Gómez CQ, García Márquez FP (2018) Concentrated solar plants management: Big data and neural network. In: Renew Energies, Springer, pp 63–81

  22. Gómez Muñoz CQ, Arcos Jimenez A, García Marquez FP, Kogia M, Cheng L, Mohimi A, Papaelias M (2018) Cracks and welds detection approach in solar receiver tubes employing electromagnetic acoustic transducers. Struct Health Monit 17(5):1046–1055

    Article  Google Scholar 

  23. Herruzo J, Imponenti L, Valverde J, Shininger R, Price H (2023) Evaluation of freeze recovery strategies for molten salt parabolic trough collectors employing solar and impedance heating methods: A thermo-mechanical modeling study. Submitted Sol Energy

  24. Sola J, Sevilla J (1997) Importance of input data normalization for the application of neural networks to complex industrial problems. IEEE Trans Nucl Sci 44(3):1464–1468

    Article  Google Scholar 

  25. Selvam SK, Rajendran C (2021) tofee-tree: automatic feature engineering framework for modeling trend-cycle in time series forecasting. Neural Comput Appl pp 1–20

  26. Zhang X, Han Y, Xu W, Wang Q (2021) Hoba: A novel feature engineering methodology for credit card fraud detection with a deep learning architecture. Inform Sci 557:302–316

    Article  Google Scholar 

  27. Lou R, Lalevic D, Chambers C, Zafar HM, Cook TS (2020) Automated detection of radiology reports that require follow-up imaging using natural language processing feature engineering and machine learning classification. J Digit Imaging 33(1):131–136

    Article  Google Scholar 

  28. Pérez-Cutiño M, Rodríguez F, Pascual L, Díaz-Báñez J (2022) Ornithopter trajectory optimization with neural networks and random forest. J Intell & Robot Syst 105(1):1–16

    Article  Google Scholar 

  29. Zhu P, Tan Y, Zhang L, Wang Y, Mei J, Liu H, Wu M (2020) Deep learning for multilabel remote sensing image annotation with dual-level semantic concepts. IEEE Trans Geosci Remote Sens 58(6):4047–4060

    Article  Google Scholar 

  30. Nagi AS, Kumar D, Sola D, Scott KA (2021) Ruf: Effective sea ice floe segmentation using end-to-end res-unet-crf with dual loss. Remote Sens 13(13):2460

    Article  Google Scholar 

  31. Han S, Pool J, Narang S, Mao H, Gong E, Tang S, Elsen E, Vajda P, Paluri M, Tran J et al (2016) Dsd: Dense-sparse-dense training for deep neural networks. In: International conference on learning representations

  32. Kingma DP, Ba J (2014) Adam: A method for stochastic optimization. arXiv:1412.6980

  33. Breiman L (2001) Random forests. Mach Learn 45(1):5–32

    Article  Google Scholar 

  34. Ke G, Meng Q, Finley T, Wang T, Chen W, Ma W, Ye Q, Liu T-Y (2017) Lightgbm: A highly efficient gradient boosting decision tree. Adv Neural Inform Process Syst vol. 30

  35. Chawla NV, Bowyer KW, Hall LO, Kegelmeyer WP (2002) Smote: synthetic minority over-sampling technique. J Artif Intell Res 16:321–357

    Article  Google Scholar 

  36. Van der Maaten L, Hinton G (2008) Visualizing data using t-sne. J Mach Learn Res 9(11)

Download references

Funding

This work is partially supported by grants PID2020-114154RB-I00, TED2021-129182B-I00 and DIN2020-011317 funded by MCIN/AEI/10.13039/501100011033 and the European Union NextGenerationEU/PRTR.

Author information

Authors and Affiliations

  1. Department of Applied Mathematics II, University of Seville, Seville, 41092, Spain

    M. A. Pérez-Cutiño, J. Valverde & J. M Díaz-Báñez

  2. Virtualmechanics S.L, Seville, 41015, Spain

    M. A. Pérez-Cutiño & J. Valverde

Authors
  1. M. A. Pérez-Cutiño
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. J. Valverde
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. J. M Díaz-Báñez
    View author publications

    You can also search for this author in PubMed Google Scholar

Corresponding author

Correspondence to M. A. Pérez-Cutiño.

Ethics declarations

Ethics approval

All of the authors confirm that there are no potential acts of misconduct in this work, and approve of the journal upholding the integrity of the scientific record.

Consent to participate

The authors consent to participate.

Consent for publication

The authors consent to publish.

Conflict of Interest

The authors have no conflicts of interest to declare that are relevant to the content of this article.

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 (e.g. a society or other partner) 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.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Pérez-Cutiño, M.A., Valverde, J. & Díaz-Báñez, J.M. Detecting broken receiver tubes in CSP plants using intelligent sampling and dual loss. Appl Intell 53, 29902–29917 (2023). https://doi.org/10.1007/s10489-023-05093-3

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10489-023-05093-3

Keywords

Search

Navigation