This brief report describes the forum with a focus on AI and K–12, which was part of the larger AI for Industries conference. Several contemporary learning resources and methods were presented and discussed.
Many production processes are characterized by numerous and complex cause-and-effect relationships. Since they are only partially known they pose a challenge to effective process control. In this work we present how Structural Equation Models can be used for deriving cause-and-effect relationships from the combination of prior knowledge and process data in the manufacturing domain. Compared to existing applications, we do not assume linear relationships leading to more informative results.
Industrial users can be justifiably hesitant in adopting Condition Monitoring Systems (CMSs) unless evidence indicates benefits from their use. Measuring a CMS’s ability to prevent losses is difficult and lacks standard procedures. The increasing availability of closed-box Artificial Intelligence (AI)- driven CMSs exacerbates the hesitancy as predicting their impacts is more challenging. This paper details three key elements critical to evaluating CMS impact:(1) the Application Area, (2) the Risk Management Processes, and (3) the Monitoring Mechanism. This paper discusses these elements in their capacity to contextualize a CMS’s role within an asset’s risk management processes, which can lead to justifying CMS use.
We propose a one-dimensional convolutional neural network (1D-CNN) with spatial coupling for post-fault power grid voltage prediction. Our proposed deep learning framework was inspired by the celebrated Prony’s method in classical signal processing. Our spatio-temporal model significantly outperforms existing benchmarks, including long short-term memory model, and is applicable to other strong transients in power industries.
EEG classification is a promising approach to facilitate the life of handicapped people and to generate future human-computer-interfaces. In this paper we want to compare the effectiveness of current state of the art deep learning techniques for EEG classification. Therefore, we applied different approaches on various datasets and did a crosscomparison of the results in order to get more knowledge on the generalization capabilities. Additionally, we created a new EEG dataset and made it available for further research.
