Unveiling the data: An analysis of plastic waste with emphasis on the countries of the E³UDRES2 alliance.

This paper offers an examination of the current plastic waste landscape, with emphasis on the nine countries of the European University Alliance E³UDRES2, based on both the literature and official numbers, to verify the alignment of practical waste management practices with scientific tendencies and advancements. The paper includes a bibliometric analysis focusing on the overall plastic waste literature and the plastic waste literature of the E³UDRES2 countries. Additionally, a mass balance was calculated regarding the domestic waste management of each of the alliance countries in 2021. The main goal is to assess how scientific research in the field of plastic waste management is being implemented in practice, particularly in the context of the E³UDRES2 countries. Bibliometric results reveal significant growth in publications since 2006, with China, the USA, and India leading. Key themes reveal evident clusters around behavior and technology, encompassing both the properties of plastics and societal attitudes toward waste management policy measures. Mass balance results reveal that, in the nine countries of the alliance, Latvia and Finland exhibited high plastic recycling rates (85% and 49%, respectively), and Germany, despite its high population, generated less waste per capita and incinerated 64% of its plastic waste. Despite progress, the results highlight ongoing challenges in implementing comprehensive circular economy-focused policies for waste management in Europe yet reveal a growing commitment to improving waste treatment systems, leading to lower environmental impacts of plastic waste.

Hyperparameter Optimization of a Convolutional Neural Network Model for Pipe Burst Location in Water Distribution Networks.

The current paper presents a hyper parameterization optimization process for a convolutional neural network (CNN) applied to pipe burst locations in water distribution networks (WDN). The hyper parameterization process of the CNN includes the early stopping termination criteria, dataset size, dataset normalization, training set batch size, optimizer learning rate regularization, and model structure. The study was applied using a case study of a real WDN. Obtained results indicate that the ideal model parameters consist of a CNN with a convolutional 1D layer (using 32 filters, a kernel size of 3 and strides equal to 1) for a maximum of 5000 epochs using a total of 250 datasets (using data normalization between 0 and 1 and tolerance equal to max noise) and a batch size of 500 samples per epoch step, optimized with Adam using learning rate regularization. This model was evaluated for distinct measurement noise levels and pipe burst locations. Results indicate that the parameterized model can provide a pipe burst search area with more or less dispersion depending on both the proximity of pressure sensors to the burst or the noise measurement level.