Estimating parameters of nonlinear dynamic systems in pharmacology using chaos synchronization and grid search

Modeling of energy and emissions from animal manure using machine learning methods: the case of the Western Mediterranean Region, Turkey

In Turkey, facilities for the use of biomass resources in energy production are increasing, and new conversion facilities are commissioned every year to provide environmentally friendly energy production. Therefore, reliable energy potential estimates are needed. In this study, the animal manure-based-biogas potentials of Antalya, Isparta, and Burdur provinces in the Western Mediterranean Region of Turkey were calculated. Here, special information on cattle, small ruminants, and poultry, and animal age, number, and manure amount information were used in detail. In addition, carbon dioxide emissions, coal, electricity, and thermal energy, methane emission values with the Tier 1 and Tier 2 approaches were calculated and predicted by machine learning algorithms. To determine the model with the best results, machine learning algorithms support vector machine (SVM), multi-layer perceptron (MLP), and linear regression (LR) were used, and hyper-parameter optimization was performed. According to the results of biogas potential, CO₂ emission, electricity production, and thermal energy estimations SVM models are seen as the best models with R² = 0.999. When the coal amount estimation is examined, the LR models produce better results than SVM and MLP with R² = 0.997. In the estimation of CH₄ using the Tier 1 approach, the MLP model can perform the best estimation with R² = 0.977. In the CH₄ modeling obtained using the Tier 2 approach, the LR models were superior to the other models with the performance value of R² = 0.962.

Control in the loop for synchronization of nonlinear chaotic systems via adaptive intuitionistic neuro-fuzzy: a comparative study

This paper proposes control in the loop (CIL) for the synchronization between two nonlinear chaotic systems at the existence of uncertainties and disturbances using an adaptive intuitionistic neuro-fuzzy (AINF) control scheme. The chaotic systems have been subedited as one is the master and the other is the slave. They both have different initial conditions and parameters. The variation in the initial conditions leads to the butterfly effect, the concept that is well known in the chaos field and means that both systems diverge over time. Therefore, AINF control scheme has been proposed in this paper as a powerful scheme to get over this problem effectively. The main objective of using the AINF control scheme is that it collects the features of its contents. As the intuitionistic fuzzy gives the system flexibility and helps the controller collecting more information about the problem. The neural networks give the controller the ability to learn over time. The experimental results were obtained to verify the applicability and effectiveness of the proposed control scheme against external disturbance and model uncertainties with comparative study.

Long-Term Groundwater Level Prediction Model Based on Hybrid KNN-RF Technique

Reliable seasonal prediction of groundwater levels is not always possible when the quality and the amount of available on-site groundwater data are limited. In the present work, a hybrid K-Nearest Neighbor-Random Forest (KNN-RF) is used for the prediction of variations in groundwater levels (L) of an aquifer with the groundwater relatively close to the surface (<10 m) is proposed. First, the time-series smoothing methods are applied to improve the quality of groundwater data. Then, the ensemble K-Nearest Neighbor-Random Forest (KNN-RF) model is treated using hydro-climatic data for the prediction of variations in the levels of the groundwater tables up to three months ahead. Climatic and groundwater data collected from eastern Rwanda were used for validation of the model on a rolling window basis. Potential predictors were: the observed daily mean temperature (T), precipitation (P), and daily maximum solar radiation (S). Previous day’s precipitation P (t − 1), solar radiation S (t), temperature T (t), and groundwater level L (t) showed the highest variation in the fluctuations of the groundwater tables. The KNN-RF model presents its results in an intelligible manner. Experimental results have confirmed the high performance of the proposed model in terms of root mean square error (RMSE), mean absolute error (MAE), Nash–Sutcliffe (NSE), and coefficient of determination (R2).

Hyperspectral Inversion of Soil Organic Matter Content Based on a Combined Spectral Index Model

Soil organic matter (SOM) refers to all carbon-containing organic matter in soil and is one of the most important indicators of soil fertility. The hyperspectral inversion analysis of SOM traditionally relies on laboratory chemical testing methods, which have the disadvantages of being inefficient and time-consuming. In this study, 69 soil samples were collected from the Honghu farmland area and a mining area in northwest China. After pretreatment, 10 spectral indicators were obtained. Ridge regression, kernel ridge regression, Bayesian ridge regression, and AdaBoost algorithms were then used to construct the SOM hyperspectral inversion model based on the characteristic bands, and the accuracy of the models was compared. The results showed that the AdaBoost algorithm based on a grid search had the best accuracy in the different regions. For the mining area in northwest China, Rp2 = 0.91, RMSEp = 0.22, and MAEp = 0.2. For the Honghu farmland area, Rp2 = 0.86, RMSEp = 0.72, and MAEp = 0.56. The detection of SOM content using hyperspectral technology has the characteristics of a high detection precision and high speed, which will be of great significance for the rapid development of precision agriculture.