Search

Anomaly detection in commercial aviation is an extremely challenging yet crucial task. Accurately detecting operationally significant anomalies can save civilian lives and/or result in significant savings in maintenance cost. The current practice uses manually tuned rule-based mechanisms to flag exceedances from predefined safety boundaries. However, this system cannot identify unknown risks and emerging vulnerabilities. Recently, innovative approaches based on machine learning have been used to automate anomaly detection. However, there are limits to their applicability in the field of aviation due to several challenges: 1) Properly reviewed data are scarce in aviation and, as a result, supervised learning cannot reach optimal performance. 2) Operationally significant anomalies do not coincide with statistically significant ones and, as a result, unsupervised learning fails to provide reliable and robust performance. In this paper, we propose a semi-supervised active learning framework for anomaly detection (SALAD), which detects operationally significant anomalies in flight operational quality assurance data. The developed framework works with vast amounts of unlabeled data as well as a small quantity of labeled data reviewed by subject matter experts to reliably identify safety anomalies in flight operations. Moreover, the model’s active learning strategy allows it to detect unknown anomalies that might emerge in the system. We validate the performance of the SALAD with a real-world case study of anomaly detection during the approach to landing of commercial aircraft. We specifically show that the proposed framework reaches reliable performance when only 1% of the data is labeled and can identify unknown anomalies effectively.

The development of renewable fuels has achieved enormous progress motivated by the rapid decline in petroleum resources and their oscillating price. After a wave of interest in bioethanol and biodiesel, biobutanol has been intensely investigated over the last decade and is considered a suitable renewable fuel. Biobutanol is most often produced through the Acetone-Butanol-Ethanol (ABE) bacterial fermentation process. This process is plagued, however, with high recovery costs because of the low final butanol concentration (∼0.01 g/g, 1 wt %) due to product inhibition on the microorganisms. To partly alleviate this problem and increase productivity, in situ butanol recovery techniques from the fermentation broth were proposed. These integrated methods increase the final concentration of butanol, and as a result improve the separation efficiency of the process. This paper compares the economic feasibility of a large-scale continuous ABE fermentation process with and without integrating a vacuum separation unit. It is illustrated that based on the current market price of butanol, only the integrated fermentation process is economically feasible for discount rates up to ∼45 %. Adding an in situ butanol recovery technique made this process profitable compared to the conventional fermentation process analyzed under two different scenarios: constant fermenter volume and constant butanol production rate. The net present values (NPV) of the conventional process were found to be negative for both scenarios, whereas the integrated vacuum process had a NPV of US$ 87 000 000 at the end of 10 years of operation based on a discount rate of 10 %.

Acetone, butanol, and ethanol (ABE) continuous fermentation for biobutanol production was simulated and optimized for three case studies where the ABE fermentation process was integrated in turn with gas stripping, pervaporation, and vacuum separation methods in an attempt to partly mitigate the product inhibition effect. Following the multi-objective optimization of the three integrated processes, the Pareto-optimal solutions were ranked using the net flow method to determine the best operating conditions. Results were compared to a standard continuous fermentation process without an integrated recovery method. The integrated butanol recovery methods improved butanol productivity by a factor of 6–10, and sugar conversion by a factor of 3, while butanol yield remained essentially unchanged. In addition, the butanol average concentration based on all exit streams for the fermentation process integrating one of the separation methods is approximately 2.25 times higher when compared to the non-integrated fermentation process, with values in the vicinity of 30 g/L. Results of the gas stripping and vacuum fermentation processes were very similar to each other but superior to pervaporation in terms of butanol productivity and average concentration.