3D-Printed Liquid Cell Resonator with Piezoelectric Actuation for In-Line Density-Viscosity Measurements

The in-line monitoring of liquid properties, such as density and viscosity, is a key process in many industrial areas such as agro-food, automotive or biotechnology, requiring real-time automation, low-cost and miniaturization, while maintaining a level of accuracy and resolution comparable to benchtop instruments. In this paper, 3D-printed cuboid-shaped liquid cells featuring a rectangular vibrating plate in one of the sides, actuated by PZT piezoelectric layers, were designed, fabricated and tested. The device was resonantly excited in the 3rd-order roof tile-shaped vibration mode of the plate and validated as a density-viscosity sensor. Furthermore, conditioning circuits were designed to adapt the impedance of the resonator and to cancel parasitic effects. This allowed us to implement a phase-locked loop-based oscillator circuit whose oscillation frequency and voltage amplitude could be calibrated against density and viscosity of the liquid flowing through the cell. To demonstrate the performance, the sensor was calibrated with a set of artificial model solutions of grape must, representing stages of a wine fermentation process. Our results demonstrate the high potential of the low-cost sensor to detect the decrease in sugar and the increase in ethanol concentrations during a grape must fermentation, with a resolution of 10 µg/mL and 3 µPa·s as upper limits for the density and viscosity, respectively.

PREDIKSI SUHU PERMUKAAN MENGGUNAKAN ARTIFICIAL NEURAL NETWORK-CELLULAR AUTOMATA DI WILAYAH CIREBON DAN SEKITARNYA

Ekspansi lahan terbangun melalui kehadiran pemukiman dan kawasan industri berimplikasi pada berkurang lahan bervegetasi di wilayah Cirebon dan sekitarnya. Kondisi menyebabkan peningkatan suhu yang berpotensi memunculkan urban heat island. Perubahan suhu dapat diprediksi menggunakan pemodelan spasial dinamis sebagai bagian dari proses perubahan lansekap. Penelitian ini bertujuan untuk memprediksi dinamika suhu permukaan di wilayah Cirebon dan sekitarnya menggunakan algoritma Artificial Neural Network-Cellular Automata (ANN-CA) dengan melibatkan variabel spasial seperti kepadatan bangunan, kerapatan vegetasi, dan kepadatan jaringan jalan. Sumber data yang digunakan dalam penelitian ini berasal dari citra Landsat-5 TM dan Landsat-8 OLI pada tahun 1999, 2009, serta 2019, sedangkan data jaringan jalan berasal dari OpenStreetMaps. Suhu permukaan diperoleh dari kanal termal yang diolah menggunakan Radiative Transfer Equation. Sementara variabel lainnya diperoleh dari Normalized Difference Built-up Index (NDBI), Normalized Difference Vegetation Index (NDVI) dan line density. Penelitian ini menunjukkan suhu permukaan hasil pemodelan ANN-CA pada tahun 2019 memiliki rerata sebesar 22,61 °C. Model ini memiliki overall accuracy 0,63 dan overall kappa sebesar 0,52. Bila dibandingkan dengan nilai aktual, model ini memiliki nilai r-square mencapai 0,80 dengan selisih sebesar 0,54 °C yang layak untuk prediksi suhu permukaan di masa mendatang. Model ANN-CA menunjukkan sebaran suhu permukaan yang lebih tinggi berpusat pada wilayah kota dan peri-urban. Kajian terkait prediksi suhu permukaan diharapkan dapat menjadi perhatian utama dalam mewujudan resiliensi perkotaan.

Water resistance of heat-treated welded Iroko, ash, tulip, and ayous wood

The friction welding method has been an effective criterion in determining the mechanical performance of wood joints in wood industry applications compared to traditional methods. Although it is used in structural applications, joints from linear vibration are quite sensitive to water. In this study, the water resistance of the heat-treated woods, iroko (Chlorophora excelsa), ash (Fraxinus excelsior L.), tulip wood (Liriodendron tulipifera) and ayous (Triplochiton scleroxylon), were investigated by friction linear welding. The weld line density profiles were examined. The resistance of heat-treated welded wood joints to water remarkably decreased compared to the control sample, depending on water immersion time. The highest shear strength loss was found in tulip wood (60% to 65%) and the lowest shear strength loss was found in ash wood (3%) for the heat-treated group and in Iroko wood (17%) for the control. The heat-treated samples increased in density with welding but had a slightly lower density than the control group. According to the TGA results, it was found that the thermal degradation of untreated welded woods was lower than that of heat-treated welded woods. This difference could be due to the chemical constituents of hardwood and tropical wood. X-ray computed tomography (CT-scanning) is feasible and usable for welding line density change.

Detecting multiple DLAs per spectrum in SDSS DR12 with Gaussian processes

ABSTRACT We present a revised version of our automated technique using Gaussian processes (GPs) to detect damped Lyman α absorbers (DLAs) along quasar (QSO) sightlines. The main improvement is to allow our GP pipeline to detect multiple DLAs along a single sightline. Our DLA detections are regularized by an improved model for the absorption from the Lyman α forest that improves performance at high redshift. We also introduce a model for unresolved sub-DLAs that reduces misclassifications of absorbers without detectable damping wings. We compare our results to those of two different large-scale DLA catalogues and provide a catalogue of the processed results of our GP pipeline using 158 825 Lyman α spectra from SDSS data release 12. We present updated estimates for the statistical properties of DLAs, including the column density distribution function, line density (dN/dX), and neutral hydrogen density (ΩDLA).

License Plate Detection using OCR with KNN

This paper presents associate in nursing economical methodology for the detection of vehicle plate by accurately localizing number plates from pictures. A straight forward downscaling of image methodology is projected initial to boost the speed of vehicle plate localization while not reducing the detection performance compared thereupon victimization the first image. Moreover, a completely unique Line Density Filter (LDF) methodology is projected to identify desired regions and eventually reducing the realm to be analysed for vehicle plate detection. Also, a vehicle plate classifier supported SVMs victimization color prominence options employed to spot the vehicle plate from the image. For performance check, a dataset having thirty pictures taken from completely different scenes under different conditions is additionally used. The data set demonstrates that the projected methodology performs well in terms of accuracy and run-time potency.