WindSound – Ingeniot

Despite the advances in maintenance management in the wind energy industry, it remains challenging to combine the new technologies and information systems with advanced analytics. The next generation of this industry is facing new opportunities employing the emergent advancement in information and communication technologies to maintain their competency and market needs. These challenges for transforming the industry into the next level require the integration of advanced analytics, called Cyber-Physical Systems (CPS). CMS and SCADA in WT generate a worldwide data volume potential of 25 TB per day, 4 bigger times than Twitter (8 TB/day). The massive data is transformed into useful information by CPS, e.g., analyzing patterns of degradations and inefficiencies to optimize decision-making based on a correct maintenance policy. It will support the actions required to maximize the uptime, productivity, and efficiency of the industrial systems. Nowadays, researchers are focused on the development of new sensors, CMS, algorithms, approaches, etc., to detect and diagnose faults, i.e., alarms. It is also generating a large number of false alarms and, therefore, downtimes and lost production in this industry.

The acoustic signals are not being considered yet in this industry due to the signal analysis and pattern recognition complexity regarding conventional CMS. Any fault in WTs in rotatory components or blades generates noises. Noise has been demonstrated to be useful for the early detection of some faults or abnormal conditions in other industries, but it can be also employed for false alarms detection and diagnosis (FADD). We have found that it has not been done yet. We have been employing unmanned aerial vehicles (UAVs) for CMS of similar devices, creating patents (ES 2768778 & ES2580302).

WindSound will develop a new False Alarm Detection and Diagnosis (FADD) methodology for improving offshore WTs maintenance, by using acoustics sensors embedded in UAVs. The acoustic dataset will be analysed with the SCADA and CMS datasets from the industry. SCADA and CMS in WTs generate a large volume and variety of data, i.e., Big Data. Artificial Neural Network will be employed for the multivariable analysis. There are not enough studies in FADD in this industry and no FADD using acoustics signals collected by CMS embedded in UAV due to the complexity of the signal processing and pattern recognition. Acoustic sensors will collect complementary information to CMS and SCADA data, with the objective of validating False Alarms. We have been focused on this topic for the last 5 years: We have demonstrated in references the need of this research, and we have begun to solve it by Artificial Neural Network and Fuzzy Logic. The reduction of false alarms will lead to reducing O&M costs while improving availability, safety, and reliability, and it will increase the safety of the workers. A new challenge ahead is to harness all the information obtained from WindSound that has to be integrated and analysed with current data to get optimal decision-making. It will require creating new importance measures to study the effect of the components over the system in a certain time and to discriminate the components. A novel decision-making analysis will be done considering the endogenous variables, such as economics, resources and reliability, and exogenous variables, such as weather, market opportunities, and legal issues, using new importance indexes, qualitatively solved by logical decision trees and quantitatively by binary decision diagrams, where we have worked for the last 10 years.

WindSound proposes, for the first time, the use of acoustic sensors in drones for CMS of WTs. The information will be analysed with the data from the SCADA and CMS of the offshore WTs to analyse the alarms, i.e., it will replace human resources to access the WTs to verify the alarms in favour of their safety. This information will be finally studied with endogenous (resources, budgets for maintenance tasks) and exogenous variables (electricity price, weather conditions, legal issues). WindSound presents a new holistic multidisciplinary approach to this paradigm in favour of the competitiveness and reliability of this industry that has not been previously studied. We have experience accredited by papers in most reference journals, leading projects, international awards, collaborations with other universities, etc.

Dronecampo22

Imagen0

Imagen1

Imagen2

Imagen4

Imagen5

Imagen6

Imagenx

IMG_2996

If you wish to collaborate with us, please fill the following form:

References

101 entries « ‹ 1 of 7 › »

2024

Babu, N. R.; Chiranjeevi, T.; Devarapalli, R.; Márquez, F. P. García: Performance analysis on an integrated three-area thermal AGC system with various wind velocities considering HCSA-optimized PI-TIDN Controller. In: vol. 34, no. 4, pp. 755-776, 2024. (Type: Journal Article | Links | BibTeX)

Ramirez, I. S.; Gonzalo, A. P.; Sánchez, P. J. B.; Márquez, F. P. G.: Wind Turbine Engineering Management: A Real Case Study with Acoustic Monitoring. In: vol. 215, pp. 647-656, 2024. (Type: Journal Article | Links | BibTeX)

Sánchez, P. Bernalte; Ramírez, I. Segovia; Márquez, F. P. García; Marugán, A. Pliego: Acoustic signals analysis from an innovative UAV inspection system for wind turbines. In: 2024. (Type: Journal Article | Links | BibTeX)

Karad, S. G.; Thakur, R.; Alotaibi, M. A.; Khan, M. J.; Malik, H.; Marquez, F. P. G.; Hossaini, M. A.: Optimal Design of Fractional Order Vector Controller Using Hardware-in-Loop (HIL) and Opal RT for Wind Energy System. In: vol. 12, pp. 35033-35047, 2024. (Type: Journal Article | Links | BibTeX)

Malik, H.; Ayob, S. M.; Idris, N. R. N.; Jusoh, A.; Márquez, F. P. G.; Almutairi, A.: Data Reliability Analysis for Early Fault Diagnosis of Air Handling Unit (AHU). In: vol. 1086, pp. 649-674, 2024. (Type: Journal Article | Links | BibTeX)

2023

Chacón, A. M. Peco; Ramírez, I. Segovia; Márquez, F. P. García: K-nearest neighbour and K-fold cross-validation used in wind turbines for false alarm detection. In: vol. 6, 2023. (Type: Journal Article | Links | BibTeX)

Verma, N.; Kumar, N.; Gupta, S.; Malik, H.; Márquez, F. P. García: Review of sub-synchronous interaction in wind integrated power systems: classification, challenges, and mitigation techniques. In: vol. 8, no. 1, 2023. (Type: Journal Article | Links | BibTeX)

Chacón, A. M. Peco; Ramirez, I. Segovia; Márquez, F. P. García: False alarm detection in wind turbine by classification models. In: vol. 177, 2023. (Type: Journal Article | Links | BibTeX)

Chacón, A. M. Peco; Ramírez, I. Segovia; Márquez, F. P. García: SCADA Data-Based Support Vector Machine for False Alarm Identification for Wind Turbine Management. In: vol. 37, no. 3, pp. 2595-2608, 2023. (Type: Journal Article | Links | BibTeX)

Chacón, A. M. Peco; Márquez, F. P. García; Ramírez, I. Segovia: Alarms and false-alarm analysis by support vector machine in wind turbines. 2023. (Type: Book | Links | BibTeX)

Chacón, A. M. Peco; Márquez, F. P. García; Ramírez, I. Segovia: Prediction and classification of different wind turbine alarms using K-nearest neighbors. 2023. (Type: Book | Links | BibTeX)

Sánchez, P. J. Bernalte; Márquez, F. P. García: Artificial neural networks applied for wind turbines maintenance management in unmanned aerial vehicle acoustic inspection case. 2023. (Type: Book | Links | BibTeX)

Chacón, A. M. Peco; Ramírez, I. Segovia; Márquez, F. P. García: False Alarm Detection with K-NN Algorithm for Wind Turbine Maintenance Management. In: vol. 160, pp. 488-493, 2023. (Type: Journal Article | Links | BibTeX)

Chacon, A. M. Peco; Marquez, F. P. G.: Classification of SCADA Alarms and False Alarm Identification Using Support Vector Machine for Wind Turbine Management. In: vol. 643 LNNS, pp. 535-547, 2023. (Type: Journal Article | Links | BibTeX)

Chacon, A. M. Peco; Márquez, F. P. García: Support Vector Machine and K-fold Cross-validation to Detect False Alarms in Wind Turbines. In: vol. 333, pp. 81-97, 2023. (Type: Journal Article | Links | BibTeX)

101 entries « ‹ 1 of 7 › »