Ingenium is attending the International Conference on Integrated Emergin Methods of Artificial Intelligence & Cloud Computing (IEMAICLOUD), placed in London during April 26-29th. This conference is organizated by the Institute of Engineering & Management. The significant characteristic of IEMAICLOUD is the promotion of dialogue between scientists. researchers, engineers and corporates.
Pedro Bernalte has presented the article entitled “A Supervisory Control and Data Acquisition System Filtering Approach for Alarm Management with Deep Learning”. Supervisory control and data acquisition system acquires large volumes of data from different condition monitoring systems. Artificial intelligence algorithms are employed to obtain reliable information, although redundant information is usually employed decreasing the validity of the results. It is proposed an approach based on data reduction employing data filtering, correlation, principal component analysis to reduce redundant information introduced in neural networks. A real case study is proposed with data from real win turbine is used to develop a case study analyzing one critical alarm. The results obtained prove the validity of the methodology, reducing the initial dataset and increasing the reliability of the neural network.