Search

Despite significant progress has been made towards crop yield prediction with remote sensing, there exist knowledge gaps on (1) the impacts of temporal resolution of imaging frequencies on yield prediction, (2) transferability of the models among different genotypes and test sites, and (3) translation of these research developments to crop breeding that benefit farmers. Existing research predominantly provides an on-site perspective, frequently missing the complexities of real-world applications. The objectives of this paper are to investigate the transferability and generalization capabilities of yield prediction models for crop breeding across test sites located in North and South Americas. Toward that goal, we tested different machine learning techniques including Random Forest Regressor (RF), Gradient Boosting Regression (GB), and Deep Neural Networks (DNN) for soybean yield prediction with experiments conducted in different climate and growth conditions. A novel transfer learning approach was proposed for genotype selection and categorizing soybean yield for screening high-yield varieties. Furthermore, we studied the effect of temporal resolution on yield prediction, focusing on the critical development stages and optimal aerial survey frequencies for precise yield prediction using large 31,404 sample data. Results demonstrated that the combined dataset of Argentina and United States representing different climate regimes provided the highest performance with an R2 of 0.76 using RF and GB algorithms. The classification approach was proven to be most useful for crop breeding as demonstrated by accurately identifying the high-yielding genotypes. Increasing temporal sampling of key phenological stages significantly improved yield prediction. Although transfer learning yielded promising outcomes across trials within Argentina the efficacy of transferring models from Argentina to the United States was limited, attributed to significant seasonal and climate variations. This study pioneered the use of transfer learning for model adaptability in real-world breeding scenarios, training and transferring models within the South and North Americas, providing actionable insights and strategies for the breeding community, aiming to facilitate improved decision-making for agricultural productivity. [ABSTRACT FROM AUTHOR]

• For the first time, dual cooled annular fuel in CAREM nuclear reactor is studied. • Neutronics and natural circulation parameters are optimized. • Developed Artificial Neural Network predicts CAREM nuclear reactor parameters. • Optimal geometry of fuel is determined using genetic algorithms. The small modular nuclear reactor (SMR) with annular fuel which is cooled internally and externally has the potential to maintain high power density with a sufficient safety margin. The benefit of the dual cooled annular fuel design is that heat can be transferred to the coolant from both the outer and inner surfaces. In this paper, design of the Argentina small modular reactor (CAREM) core with hexagonal assemblies using dual-cooled annular fuels with internal and external cooling is presented. Neutronics and thermal-hydraulic parameters are evaluated and analyzed by changing internal fuel radius. For this purpose, at the first, the dual-cooled annular fuel under clean and cold conditions is modeled and the effective multiplication factor has been calculated for different inner clad diameter. Indeed, the internal and external radius have been changed in such a way to maintain the reactor under moderated.Then, these annular fuels under full power conditions are modeled and power peaking factor has been calculated. Finally, natural circulation parameters are performed for a simulated fuel rod in the hot channel using computational fluid dynamics (CFD) simulation codes. These results are compared with the conventional CAREM reactor. One of the most prominent advantages of the annular fuels is the ability to make softer neutrons in which reduce maximum power peaking factor and improves fuel management and safety parameters in the reactor core. For data fitting, an artificial neural network is trained using the observed data. The input consists of different internal and external radiuses and output consists of fuel rods' pitch, thermal-hydraulic and neutronic parameters. Finally, the optimal geometry of fuels is determined using the neural network by implementing the genetic algorithms.Indeed, developed Artificial Neural Network (ANN) utilizing the obtained data, predicts the thermal-hydraulic and neutronic parameters of the CAREM reactor core with dual-cooled annular fuels. Presented optimization algorithm, which has a significant ability to attain the best solutions, also determines the optimal values of natural circulation parameters (V max /V avg , V out -V in , and pressure drops), heat transfer coefficients, MDNBR, RPPF, and excess reactivity of the reactor. Also, the designed artificial neural network and genetic algorithm has been validated using neutronic and thermal hydraulic calculations. Results indicate that by using annular fuels, better reactor safety and thermal efficiency are achieved. [ABSTRACT FROM AUTHOR]