Crop and weed discrimination in natural field environments is still challenging for implementing automatic agricultural practices, such as weed control. Some weed control methods have been proposed. However, these methods are still restricted as they are implemented under controlled conditions. The development of a sound weed control system begins by recognizing the crop and the different weed plants presented in the field. In this work, a classification approach of Zea mays L. (Crop), narrow-leaf weeds (NLW), and broadleaf weeds (BLW) from multi-plant images are presented. Moreover, a large image dataset was generated. Images were captured in natural field conditions, in different locations, and growing stages of the plants. The extraction of regions of interest (ROI) is carried out employing connected component analysis (CCA), whereas the classification of ROIs is based on Convolutional Neural Networks (CNN) and compared with a shallow learning approach. To measure the classification performance of both methods, accuracy, precision, recall, and F1-score metrics were used. The best alternative for the weed classification task at early stages of growth and in natural corn field environments was the CNN-based approach, as indicated by the 97% accuracy value obtained.

To alleviate soil compaction, subsoiling practices using subsoiler implements are commonly implemented. However, subsoiler bodies are subjected to great draft forces because they work deep in the soil. Therefore, to contribute to draft force reduction, in this work, a bio-inspired attack edge for a subsoiler body based on the internal and external contour lines of the claws of the Mexican ground squirrel (Spermophilus mexicanus) is proposed. As a first step, computational fluid dynamic (CFD) modeling was used to select the best bionic subsoiler (BS) according to the draft force requirements. Then, the BS was fabricated and field-evaluated, and its real draft force during tillage was contrasted with those of a curve subsoiler (CS) and a straight subsoiler (SS). The field evaluation demonstrated that the BS demands, on average, $12.37\%$ and $22.25\%$ less draft force than the CS and SS, respectively. Additionally, the BS was better at entering the soil since its mean tillage depths were $24.86\%$ and $5.73\%$ higher than those of the SS and CS geometries, respectively. Therefore, it was found that modeling the attacking edge of a subsoiler body after the Mexican ground squirrel clearly reduced the draft force during tillage.

The fatigue behaviour of a 90° long radius elbow, which is adjacent to the feedwater nozzle in a BWR, was analyzed. The start-up and shutdown transients were considered. A thermo-mechanical finite element analysis was carried out to determine the stresses induced by thermal transients, considering the environmental conditions in the reactor feedwater pipe. In addition, the Palmgren–Miner methodology and the ASME B&PVC code fatigue curve were applied to evaluate the accumulated damage and service life of the component. Environmental correction factors were considered to estimate environmentally assisted fatigue. Reductions in fatigue life were observed. In the second part of this paper, a part-through thickness semielliptical crack was also postulated in the internal surface of the elbow. It was aligned along the axial direction at the crown zone. Its growth was modelled using the Paris equation, evaluating the risk of failure using fracture parameters. It was found that the vulnerable area is located on the inner surface of the elbow, due to the concentration of stress caused by the curved geometry. Failure assessment diagrams (FADs) were plotted. It was found that the crack depth is the main factor governing crack behaviour under the conditions studied. The results provide a methodology for assessing the integrity of pipes subjected to specific environmental and operating conditions.