Zoekresultaten

Explainable artificial intelligence for exhaust gas temperature of turbofan engines

Data-driven modeling is an imperative tool in various industrial applications, including many applications in the sectors of aeronautics and commercial aviation. These models are in charge of providing key insights, such as which parameters are important on a specific measured outcome or which parameter values we should expect to observe given a set of input parameters. At the same time, however, these models rely heavily on assumptions (e.g., stationarity) or are “black box” (e.g., deep neural networks), meaning that they lack interpretability of their internal working and can be viewed only in terms of their inputs and outputs. An interpretable alternative to the black box models and with considerably less assumptions is symbolic regression (SR). SR searches for the optimal model structure while simultaneously optimizing the model’s parameters without relying on an a priori model structure. In this work, we apply SR on real-life exhaust gas temperature (EGT) data, collected at high frequencies through the entire flight, in order to uncover meaningful algebraic relationships between the EGT and other measurable engine parameters. The experimental results exhibit promising model accuracy, as well as explainability returning an absolute difference of 3°C compared to the ground truth and demonstrating consistency from an engineering perspective.

The robot is not a collegue

Older technologies, such as violins and computers, differ from newer technologies (Internet search engines, robots, chatbots, etc.). Older technologies are primarily tools for achieving our own stated goals, while new technologies are often used by third parties, whereby we ourselves are sometimes used as “instruments”, often without us being aware of it. While we search for something on the Internet, our search behaviour is analysed and we are assigned to an increasingly suitable profile. In other cases, our movement patterns in public space are registered (type of shopper, walker, etc.). Unsolicited we get a digital twin, which can be easily followed by third parties to warn us in time, or, in the wrong hands, to manipulate or even disable us. Anthony Kenny states that “technological anthropomorphism” (approaching an instrument or machine as a person with human qualities) must be avoided. It is argued why and that a robot is not a colleague.

Searching, Selecting, and Verifying

This study utilises a quantitative observation study on student journalists (n=47), as well as reconstruction interviews with experienced editors and reporters in newsrooms (n=12), to understand how Dutch journalists search, select, and verify sources online. Through the recording of screen activity, we show that search strategies are heavily influenced by how the search engine sorts and ranks potential sources. Eventual selection of sources remains relatively traditional, focused on legacy media and their websites. Moreover, online news production clearly challenges the verification process. Results suggest that journalists use no explicit but only so-called hybrid methods of verifications, such as background checks of websites and social media accounts, and cross-checking of sources.

Website awesomization: search engine engineering

An End-to-End Pipeline for Uncertainty Quantification and Remaining Useful Life Estimation: An Application on Aircraft Engines

Estimating the remaining useful life (RUL) of an asset lies at the heart of prognostics and health management (PHM) of many operations-critical industries such as aviation. Mod- ern methods of RUL estimation adopt techniques from deep learning (DL). However, most of these contemporary tech- niques deliver only single-point estimates for the RUL without reporting on the confidence of the prediction. This practice usually provides overly confident predictions that can have severe consequences in operational disruptions or even safety. To address this issue, we propose a technique for uncertainty quantification (UQ) based on Bayesian deep learning (BDL). The hyperparameters of the framework are tuned using a novel bi-objective Bayesian optimization method with objectives the predictive performance and predictive uncertainty. The method also integrates the data pre-processing steps into the hyperparameter optimization (HPO) stage, models the RUL as a Weibull distribution, and returns the survival curves of the monitored assets to allow informed decision-making. We vali- date this method on the widely used C-MAPSS dataset against a single-objective HPO baseline that aggregates the two ob- jectives through the harmonic mean (HM). We demonstrate the existence of trade-offs between the predictive performance and the predictive uncertainty and observe that the bi-objective HPO returns a larger number of hyperparameter configurations compared to the single-objective baseline. Furthermore, we see that with the proposed approach, it is possible to configure models for RUL estimation that exhibit better or comparable performance to the single-objective baseline when validated on the test sets.

Improving Book Search with AI

We present our ongoing work on upgrading the Amsterdam Public Library's book database search capabilities. So far, users have had to input the exact book title and/or author name without any typos or misspellings in order to retrieve any results. This is in sharp contrast with the manner in which users typically use the interface: they frequently search for books on a particular topic, input the names of the characters, or even ask fully-fledged questions. The aim of this project is therefore to enable smart search in natural language based on book content. The initial focus is on the Dutch language, with the possibility of including English and other languages later. In the first phase of the project, we built a proof-of-concept knowledge graph from a sample of the existing tabular database and enriched the data with named entities extracted from book summaries. Based on this first step, a user query like "Heeft u boeken over de Tweede Wereldoorlog in Amsterdam?" would yield all books that mention both WW2 and Amsterdam. We are currently working on augmenting the knowledge graph with embeddings, which will enable us to retrieve semantically similar results. The final step of the research involves integrating our knowledge graph with a pre-trained large language model.