Celso De La Cruz

Tetranychus spp. is a prominent pest that affects strawberry cultivation, causing a significant decrease in economic yields. Manual inspection is the most common method used for identifying this pest, but it requires arduous labor, is slow, expensive, and inaccurate. This situation limits the ability to make an adequate early diagnosis of the damage caused by these spider mites. Thus, this paper presents a non-invasive method for real-time analysis of strawberry crops to determine the severity of Tetranychus spp. infestations. A field experiment was conducted to validate this systematic method through the biological characterization of spider mites, and a lightweight UNet-based convolutional neural network is proposed for the recognition of visual symptoms manifested on strawberry leaves containing Tetranychus spp. This proposed model uses a Residual bottleneck block to reduce the number of layers and network complexity, and an atrous spatial pyramid pooling algorithm with a custom kernel that improves accuracy for this application. A total of 2273 images with a resolution of 512 × 512 pixels were collected from strawberry crops containing visible adaxial symptoms in complex backgrounds and were split into training (1772), validation (197), and test (304) datasets. The proposed model was evaluated alongside other state of the art semantic segmentation networks under the exact same training, validation, and testing conditions for a fair comparison. The results show that the proposed model achieves accuracy indicators comparable to the most recent semantic segmentation models, such as an IoU of 0.4791, but with twice the inference speed of the most accurate methods. Modifying the kernels for atrous convolutions demonstrated a substantial improvement in symptom recognition accuracy compared to standard atrous convolutions. Measurements of Lin’s Concordance Correlation Coefficient evidence that our model produces accurate severity estimations for a proper understanding of the actual degree of infestation present in crops. These outcomes evidence the value of the proposed algorithm to detect Tetranychus spp. in strawberry crop fields through real-time automated inspections.

The aim of this work is to provide new experimental data on laminar burning velocities for a new synthetic mixture composed of Camisea natural gas and CO₂. It was found that the relevant published experimental background data are limited to mixtures composed of methane and CO₂; considering the fact that Camisea natural gas is widely used in Peru, this experimental research will serve as a supportive resource for further experimental and industrial implementations in this country, such as the design and modeling of new engines or industrial burners that are designed to be fueled by this mixture. An experimental setup for analyzing three types of flame geometry, which is feasible to implement for a wide range of conditions, was built in PUCP PI0735 laboratory and all the measurements were obtained for a range of mixtures (0%, 21.2%, 28.5%, 38.9%, 50% CO₂) and ratios from around 0.55 to 0.95 at atmospheric conditions. The laminar burning velocities results obtained were analyzed in groups based on %CO₂. In addition, the experimental margin error was determined by considering all the sources. The following conclusions were reached: (1) The laminar burning velocity decreases with the increase in CO₂ percentage in the mixture due to the CO₂ decreasing the flame temperature effect. (2) The flat flame type provided the highest value of burning velocity for each group of CO₂ percentage in which it appears. (3) The highest obtained laminar burning velocity value was 22.64 ± 0.15 cm/s, for a flat flame with a ratio of 0.72 and 29.98% of CO₂, while the lowest obtained value was 6.78 ± 0.15 cm/s for a conical trunk flame with a ratio of 0.59 and 49.83% of CO₂. (4) The highest evaluated CO₂ percentage was 50.97% for a conical trunk flame with a ratio of 0.69 and a burning velocity value of 11.04.

Pisco is an alcoholic beverage obtained from grape juice distillation. Considered the flagship drink of Peru, it is produced following strict and specific quality standards. In this work, sensing results for volatile compounds in pisco, obtained with an electronic nose, were analyzed through the application of machine learning algorithms for the differentiation of pisco varieties. This differentiation aids in verifying beverage quality, considering the parameters established in its Designation of Origin”. For signal processing, neural networks, multiclass support vector machines and random forest machine learning algorithms were implemented in MATLAB. In addition, data augmentation was performed using a proposed procedure based on interpolation–extrapolation. All algorithms trained with augmented data showed an increase in performance and more reliable predictions compared to those trained with raw data. From the comparison of these results, it was found that the best performance was achieved with neural networks.

En este trabajo se realiza un estudio numérico del flujo de aire de un soplador para la selección de la papa amarilla Tumbay. Se aplica la técnica de volúmenes finitos para resolver las ecuaciones de continuidad, momento y energía. Se consideran el fluido con comportamiento de gas ideal con densidad variable y el modelo de turbulencia SST. La simulación tiene un flujo de entrada, a diferentes presiones, el cual impacta sobre la papa Tumbay cuando se identifica alguna anomalía. Los resultados muestran que un chorro de aire a través de un soplador tiene la fuerza de impacto suficiente para desviar la trayectoria deseada del tubérculo, separándolo del grupo de otra clase. Se concluye que la combinación de herramientas computacionales de método de elementos discretos y la dinámica de fluidos computacional (DEM-CFD) permite describir la iteración del flujo de aire impactando con la papa provocando un desvío en su trayectoria.

There is great interest in the development of a simple system that could identify adulteration or counterfeiting of Peruvian Pisco (a grape-based alcoholic drink). In this study, sensors based on SnO₂-TiO₂ and SnO₂-MoO₃ composites with different weight composition ratios were synthesized and characterized. These sensors were tested with aqueous solutions of EtOH/MeOH and Pisco samples of Italia and Quebranta varieties in order to explore their capacity to identify variations in these beverages. The response profile of the most sensitive sensors showed an enhanced response to alcoholic samples with greater content of ethanol up to a concentration of 45%v/v, while the increased content of methanol in the range of 0.1 to 0.3% v/v diminished the intensity of the sensor response. Differences in the composition of methanol and ethanol in the Pisco varieties studied (Italia and Quebranta) were correlated to the capacity of the composite-based sensors to differentiate them with greater performance. Sensors based on SnO₂-TiO₂-1/2 composites showed greater reproducibility in their response profile over time in comparison to SnO₂-TiO₂-1/1 and SnO₂-MoO₃ composites. The PCA showed that composite sensors were able to differentiate Pisco samples according to the varieties, Italia and Quebranta.

Introduction Electronic noses allow a rapid and economical evaluation of freshness, maturity or decomposition of food and beverages for classification according to their based on their volatile components. The most widespread gas sensors are based on semiconductor metal oxides (such as SnO₂, TiO₂, or ZnO) due to their low cost and high sensibility. However, their main disadvantage is its low selectivity. A doping treatment with noble metals is required to generate an improvement in the response to specific gases and decrease the working temperature [1]. Pisco is a grape-based alcoholic beverage of historical and commercial importance in Peru, with a growing presence in the national and international market. Although its production and commercialization are protected by a designation of origin, the industry faces a severe adulteration issue. This study evaluates the application of SnO₂ gas sensors doped with silver nanoparticles obtained by green synthesis on an electronic nose as a fast and low-cost method to differentiate Pisco varieties. Green Synthesized Silver Nanoparticles Silver nanoparticles were prepared via a conventional synthesis and two green synthesis routes. Silver nitrate salt (J.A. Elmer, >99.8%) was treated with citric acid, Aloe vera, or Allium sativum extracts at 80 °C and pH 8. The nanoparticles obtained were characterized by UV-Visible and ATR spectroscopy. Tin oxide (Merck, >99.9%) was mixed with the nanoparticle suspension and calcinated at 450 °C for 2 h. Method The sensing of Peruvian piscos (Quebranta and Italia varieties) was performed using an electronic nose made at our university with SnO₂-based sensors doped with silver nanoparticles. The working temperature was 240 °C and the sample and purge times were 80 and 220 s. The sensing results obtained were analyzed using multivariate statistical methods that were unsupervised (PCA and HCA) and supervised (SVM and KNN). The parameters of the models were optimized using k-folding cross-validation based on their accuracy, and subsequently used to predict the pisco variety of another set of measurements. Results and Conclusions Tin oxides doped with silver nanoparticles allowed a greater differentiation of Pisco varieties compared to those without doping. The Figure shows that plotting the two principal components (PC1 and PC2) for the Pisco varieties analysis led to a clear differentiation of the samples. Optimized supervised methods achieved prediction rates greater than 90%. It was concluded that it is feasible to use plant extracts from Aloe vera and Allium sativum to prepare gas sensors that can be used in the beverage industry. Keywords: Electronic nose, Aloe vera, Allium sativum, PCA, KNN algorithm, SVM [1] Wang, L.; Wang, Y.; Yu, K.; Wang, S.; Zhang, Y. & Wei, C. Sensors and Actuators B: Chemical 2016 , 232, 91-101. Figure 1

Introduction The electronic nose (e-nose) is an emerging technology with the potential to provide a rapid and affordable assessment of production quality in the food industry. Its operation is based on the detection of volatile compounds that build the aroma of beverages and foods. This data provides information about the level of freshness of food and can be correlated to the quality of beverages. Indeed, the most widely used sensors are made from semiconductor metal oxides(SnO₂, TiO₂, MoO₃, among others) because of their low cost and great sensitivity. Therefore, this variety of gas sensor allows an economically viable alternative to sophisticated techniques: HPLC and GC-MS [1].It should be noted that the low selectivity of these oxides towards some volatile organic compounds (VOCs) has limited their application. A method to solve this problem is the preparation of composites based on two different metal oxides. In fact, the enhanced performance of semiconductor metal oxides as gas sensors originates from the synergistic effects of putting in contact grain boundaries of different metal oxides. In particular, the Schottky barrier at the interface between the oxides grains can be tailored by mixing oxides with different energy gaps [2]. Peruvian Pisco is an artisanal grape brandy with an alcohol content between 38% and 48%. It is considered an emblematic drink of Peru due to its unique characteristics and traditional procedures used for its preparation. Even though this beverage has a protected designation of origin, its artisanal production presented an increasing adulteration's incidence. To find a solution to this problem, this study implemented an electronic nose designed at PUCP and made up of composite oxide sensors based on SnO₂-TiO₂ and SnO₂-MoO₃. Synthesis of Metal Oxide Composites The oxide composites were synthesized by chemical precipitation of SnO₂ over the grains of both metal oxides: TiO₂ and MoO₃. A typical synthesis procedure consisted of dispersing 0.2 g of SnCl₂.2H₂O in 5 mL of 0.1M HNO₃ under ultrasound for 30 minutes. Subsequently, 100 mg of MoO₃ or TiO₂ was added to the mixture, which was left under stirring for 15 minutes. Later a few drops of concentrated ammonium hydroxide were slowly added until reaching a pH of 10. Finally, after stirring for 30 minutes, the solid was washed by centrifugation with Mill Q water and dried at 80°C. Finally, the composite was calcined at 500°C for 2 hours. This e-nose was developed to differentiate two varieties of Pisco: Quebranta and Italia. Method The crystal structures in the metal oxide composites were determined by x-ray diffraction and the crystallite size was estimated from the Sherrer’s equation. The metal oxide composites were deposited over alumina substrates using an organic binder (α-terpineol). Interdigit platinum electrodes were previously fabricated over the substrates. The prepared sensors were calcinated at 500°C and later aged at 150°C for a night. The e-nose sensing response to VOCs in the aroma of the Pisco samples was recorded at different temperatures: 200˚C, 220˚C, and 240˚C. The data obtained were processed using supervised (SVN and KNN) and unsupervised (PCA and HCA) statistical methods. The modeling parameters were optimized by cross-validation of k-folds. Results and Conclusions The two principal components (PC1 and PC2) of the data obtained using the electronic nose confirmed its capacity to differentiate the varieties studied at certain working conditions. The best results were obtained from the application of the sensor based on SnO₂-TiO₂ at 240°C. In the Figure, the PCA plots showed a significant improvement in this sensor's differentiation capacity of the studied Pisco varieties when the sensing temperature was increased from 200°C to 240°C. The supervised classification analyzes allowed the development of predictive models with a success rate greater than 90% using the sensors based on SnO₂-TiO₂. Keywords: Peruvian Pisco, electronic nose, composite oxides, PCA, KNN algorithm and SVM. [1] Quicazán, M. C., Díaz, A. C. y Zuluaga, C. M. La nariz electrónica, una novedosa herramienta para el control de procesos y calidad en la industria agroalimentaria. VITAE, Rev. la Fac. Química Farm. 209–217 (2011). [2] Xing, L. L., Yuan, S., Chen, Z. H., Chen, Y. J. y Xue, X. Y. Enhanced gas sensing performance of SnO2/α-MoO3 heterostructure nanobelts. Nanotechnology 22, (2011). Figure 1

This work presents the development of a robotic wheelchair that can be commanded by users in a supervised way or by a fully automatic unsupervised navigation system. It provides flexibility to choose different modalities to command the wheelchair, in addition to be suitable for people with different levels of disabilities. Users can command the wheelchair based on their eye blinks, eye movements, head movements, by sip-and-puff and through brain signals. The wheelchair can also operate like an auto-guided vehicle, following metallic tapes, or in an autonomous way. The system is provided with an easy to use and flexible graphical user interface onboard a personal digital assistant, which is used to allow users to choose commands to be sent to the robotic wheelchair. Several experiments were carried out with people with disabilities, and the results validate the developed system as an assistive tool for people with distinct levels of disability.