Graph neural network for integrated water network partitioning and dynamic district metered areas.

Water distribution systems (WDSs) are used to transmit and distribute water resources in cities. Water distribution networks (WDNs) are partitioned into district metered areas (DMAs) by water network partitioning (WNP), which can be used for leak control, pollution monitoring, and pressure optimization in WDS management. In order to overcome the limitations of optimal search range and the decrease of recovery ability caused by two-step WNP and fixed DMAs in previous studies, this study developed a new method combining a graph neural network to realize integrated WNP and dynamic DMAs to optimize WDS management and respond to emergencies. The proposed method was tested in a practical case study; the results showed that good hydraulic performance of the WDN was maintained and that dynamic DMAs demonstrated excellent stability in emergency situations, which proves the effectiveness of the method in WNP.

Gravitational Wave-Signal Recognition Model Based on Fourier Transform and Convolutional Neural Network.

The recent detection of gravitational waves is a remarkable milestone in the history of astrophysics. With the further development of gravitational wave detection technology, traditional filter-matching methods no longer meet the needs of signal recognition. Thus, it is imperative that we develop new methods. In this study, we apply a gravitational wave signal recognition model based on Fourier transformation and a convolutional neural network (CNN). The gravitational wave time-domain signal is transformed into a 2D frequency-domain signal graph for feature recognition using a CNN model. Experimental results reveal that the frequency-domain signal graph provides a better feature description of the gravitational wave signal than that provided by the time-domain signal. Our method takes advantage of the CNN's convolution computation to improve the accuracy of signal recognition. The impact of the training set size and image filtering on the performance of the developed model is also evaluated. Additionally, the Resnet101 model, developed on the Baidu EasyDL platform, is adopted as a comparative model. Our average recognition accuracy performs approximately 4% better than the Resnet101 model. Based on the excellent performance of convolutional neural network in the field of image recognition, this paper studies the characteristics of gravitational wave signals and obtains a more appropriate recognition model after training and tuning, in order to achieve the purpose of automatic recognition of whether the signal data contain real gravitational wave signals.

Structure of microprismatic Fresnel lenses for creating uniform focal images.

Flat Fresnel lenses are known to form a point image in the focal plane. However, several practical applications require transforming lens to concentrate a parallel light beam into a uniformly illuminated light circle. We previously proposed a novel algorithm for simulating such a transforming Fresnel concentrator. In this study, we applied this method to the diamond-cutting technique to create prismatic refractive surfaces of high optical quality. To reduce the discreteness of formed images, each refractive lens zone was fabricated from several small identical microprisms in the simulation. The new fabricated circular light beam concentrators were investigated by computer modelling and experimentally with a collimated laser beam.