Application of artificial neural networks and genetic algorithm to predict and optimize greenhouse banana fruit yield through nitrogen, potassium and magnesium

Abstract

The excess of the chemical fertilizers not only causes the environmental pollution but also has many deteriorating effects including global warming and alteration of soil microbial diversity. In conventional researches, chemical fertilizers and their concentrations are selected based on the knowledge of experts involved in the projects, which this kind of models are usually subjective. Therefore, the present study aimed to introduce the optimal concentrations of three macro elements including nitrogen (0, 100, and 200 g), potassium (0, 100, 200, and 300 g), and magnesium (0, 50, and 100 g) on fruit yield (FY), fruit length (FL), and number of rows per spike (NRPS) of greenhouse banana using analysis of variance (ANOVA) followed by post hoc LSD test and two well-known artificial neural networks (ANNs) including multilayer perceptron (MLP) and generalized regression neural network (GRNN). According to the results of ANOVA, the highest mean value of the FY was obtained with 200 g of N, 300 g of K, and 50 g of Mg. Based on the results of the present study, the both ANNs models had high predictive accuracy (R2 = 0.66–0.99) in the both training and testing data for the FY, FL, and NRPS. However, the GRNN model had better performance than MLP model for modeling and predicting the three characters of greenhouse banana. Therefore, genetic algorithm (GA) was subjected to the GRNN model in order to find the optimal amounts of N, K, and Mg for achieving the high amounts of the FY, FL, and NRPS. The GRNN-GA hybrid model confirmed that high yield of the plant could be achieved by reducing chemical fertilizers including nitrogen, potassium, and magnesium by 65, 44, and 62%, respectively, in compared to traditional method.

Comparative analysis of different artificial neural networks for predicting and optimizing in vitro seed germination and sterilization of petunia

“…But several studies in other fields of biology have evidently confirmed that GRNN had better performance than MLP and RBF [ 29 , 63 , 97 ]. Also, according to a comparative analysis of the MLP and GRNN to optimize greenhouse banana fruit yield by Ramezanpour and Farajpour [ 98 ], their results have strongly stated that the GRNN was a more accurate technique than MLP in the prediction of evaluated parameters. In another study, the good performance of the GRNN technique for modeling and predicting in vitro seed germination in cannabis has been highlighted by Pepe et al ., [ 28 ].…”

Enhancing petunia tissue culture efficiency with machine learning: A pathway to improved callogenesis

“…Firstly, the GRNN model is known for its ability to capture complex patterns and relationships in the data due to its non-linear nature [65]. It uses a radial basis function activation function, which allows it to approximate any continuous function with high accuracy [66]. This flexibility enables the GRNN model to effectively capture the intricate relationships between the concentrations of phytohormones and callogenesis [21] in petunia tissue culture.…”

“…This characteristic is particularly beneficial in the context of callogenesis prediction in petunia tissue culture, where historical data and patterns play a crucial role [20]. In contrast, the RBF model and MLP model lack this explicit memory capability, which may limit their ability to effectively capture long-term dependencies and generalize well to new data [66].…”

Demand-Side Management Optimization Using Genetic Algorithms: A Case Study

“…Computational intelligence approaches have also been applied in [19][20][21][22][23] with a focus on cost reduction, peak energy demand, energy saving, and DSM. Aiming at solving optimization problems, several studies have proposed genetic algorithms, including [24][25][26][27][28][29][30]. The results justify the applicability of the genetic algorithms for power utilization reduction, minimizing the electricity cost and peak-to-average ratio considering shifting and nonshifting loads via appliance scheduling.…”