Sensitivity analysis of energy inputs in crop production using artificial neural networks

Khoshroo, Alireza; Emrouznejad, Ali; Ghaffarizadeh, Ahmadreza; Kasraei, M; Omid, Mahmoud

doi:10.1016/j.jclepro.2018.05.249

Cited by 71 publications

(33 citation statements)

References 56 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Moreover, for ratoon farms, the 7-9-6-11 structure had the best performance. The developed various multilayer perception ANN models were applied to predict grape yield with respect to inputs and was observed that the 7-6-1 architecture was the best model with the highest mean correlation coefficient and the least standard deviation (Khoshroo et al 2018). In another study several ANNs models with different topologies and distinct learning algorithms was used to modeling energy consumption in greenhouse tomato cultivation.…”

Section: Evaluation Of Annsmentioning

confidence: 99%

Joint of Modeling Techniques and Life Cycle Assessment for Prediction of Yield, Economic Profit, and Global Warming Potential for Wheat Farms

Mobtaker

Kaab

Rafiee

2021

Preprint

View full text Add to dashboard Cite

Forecasting crop yield and its environmental effects can help increase agricultural energy efficiency and reduce environmental impacts. This study provides mathematical, adaptive network-based fuzzy inference system (ANFIS) and neural networks (ANNs) techniques for forecast yield, economic profit, and global warming of wheat production. For this purpose, 75 wheat farms located in the central area of Hamadan province were selected randomly and data were gathered through oral interview. Then, computed the input and output energy, life cycle assessment (LCA) was utilized to specify the environmental effects of wheat cultivation. The calculations displayed that the averages of inputs and outputs energy were about 43055 MJ ha−1 and 117407 MJ ha−1, respectively. The LCA results demonstrated that wheat cultivation cause to the emissions of 624.29 kg CO2 eq. ton−1. ANN structures for predicting yield, economic profit and global warming in wheat production with two hidden layers were the best topologies. ANFIS model results indicated that in the 3-level ANFIS model, the highest R2 is found for net return (0.962). The results comparison showed that ANN and ANFIS models outperform linear models to predict yield, economic profit, and global warming of wheat production.

show abstract

Section: Evaluation Of Annsmentioning

confidence: 99%

Joint of Modeling Techniques and Life Cycle Assessment for Prediction of Yield, Economic Profit, and Global Warming Potential for Wheat Farms

Mobtaker

Kaab

Rafiee

2021

Preprint

View full text Add to dashboard Cite

show abstract

“…To identify the effect of important factors on the value of response variables, sensitivity analysis based on AI numerical models have been applied and proved to be an efficient method in many researches. 54,55 In this study, such method is utilized to identify which input (heat generation per cell, V, and t) has a greater influence on the modification of outputs (T1, T2, and parasitic energy) values, and which one have no effect on dependent variables. The analysis result is shown in Table 3.…”

Section: Sensitivity Analysis Based On Selected Models Obtained Fromentioning

confidence: 99%

A novel procedure combining computational fluid dynamics and evolutionary approach to minimize parasitic power loss in air cooling of Li‐ion battery for thermal management system design

Jishnu

Garg

et al. 2020

Energy Storage

View full text Add to dashboard Cite

Air‐cooling‐based battery thermal management system (BTMS) is a research hotspot for electric vehicles because of lower cost and simpler design. Past research works have immensely concentrated on the enhancement of heat removal from Li‐ion battery, but the minimum consideration has been given to minimize the parasitic power. In this paper, a novel procedure is proposed to predict the operating parameters (inlet velocity, working time of fan, and range of heat generation rates) of air‐cooling design for minimizing parasitic power and ensuring the battery temperature do not exceed the upper threshold limit simultaneously. Based on findings via computational fluid dynamics analysis, an empirical model of air‐cooling BTMS is further developed using an evolutionary approach of model selection criteria approximated genetic programming (MSC‐GP). The model is then optimized to determine operating parameters of air cooling which causes minimum parasitic power while keeping the average temperature of battery cells within the limit. Further, sensitivity analysis and parameter interaction (2D and 3D) analysis is also performed to study the effect and identify the contribution of various operating parameters on parasitic power. Operating time of fan has 49% influence on final temperature while inlet velocity has 36% influence only. However parasitic power is more sensitive to inlet velocity (77%) while operating time has 23% influence only. Finally, the optimal values of operating parameters for various heat generation rate per cell (function of discharge rate) are obtained. The optimized parasitic power was observed to be nonlinearly increasing with heat generation rate. Operating time of fan has 49% influence on final temperature while inlet velocity has 36% influence only. However, parasitic power is more sensitive to inlet velocity (77%) while operating time has 23% influence only.

show abstract

“…The artificial neural networks ANN are inspired from the morphology of biological nervous systems to emulate the functioning of the brain with high capability to learn and to predict based on learning stage. The artificial neural networks consist of multilayer: input layer, output layer and the layer between the input layer and the output layer is named the hidden layer [16]. Generally, there is currently no theoretical criteria adequate for choosing an appropriate number of hidden layer.…”

Section: Artificial Neural Network (Ann)mentioning

confidence: 99%

Application and Comparison of Machine Learning Algorithms for Predicting Mass Loss of Cement Raw Materials due to Decarbonation Process

Boukhari¹

2020

RIA

View full text Add to dashboard Cite

The purpose of this work is to predict the mass loss of cement raw materials during the decarbonation process. The mass loss is influenced by the interaction of several parameters such as chemical composition of raw material, particle size, temperature range of decarbonation and time exposed. Therefore, predicting mass loss based on experimental parameters data is often challenging. For this reason, various machine learning algorithms such as deep networks using autoencoder DN-AE, artificial neural networks optimized by particle swarm optimization PSO-ANN, ANN optimized by ant colony optimization ACO-ANN and ANN are proposed to predict the mass loss. In this research, all models have been applied successfully to predict the mass loss with high accuracy. The results obtained have shown the superiority of DN-AE compared to PSO-ANN, ACO-ANN and ANN. In addition, PSO-ANN and ACO-ANN have a better performance than the individual use of ANN. The values of adjusted R2 indicate that 99.11%, 98.66%, 98.27% and 97.03% of data are explained by DN-AE, ACO-ANN, PSO-ANN and ANN respectively with scatter index (SI) less than 0.1 and maximum error less than 3.32%. Finally, the results justify that all models proposed can be employed to predict the mass loss as alternative tools.

show abstract

Sensitivity analysis of energy inputs in crop production using artificial neural networks

Cited by 71 publications

References 56 publications

Joint of Modeling Techniques and Life Cycle Assessment for Prediction of Yield, Economic Profit, and Global Warming Potential for Wheat Farms

Joint of Modeling Techniques and Life Cycle Assessment for Prediction of Yield, Economic Profit, and Global Warming Potential for Wheat Farms

A novel procedure combining computational fluid dynamics and evolutionary approach to minimize parasitic power loss in air cooling of Li‐ion battery for thermal management system design

Application and Comparison of Machine Learning Algorithms for Predicting Mass Loss of Cement Raw Materials due to Decarbonation Process

Contact Info

Product

Resources

About