Search

In this paper, hybrid feed-forward deep neural network and ANFIS framework is developed to control an active suspension system of half vehicle. The dataset were generated from previous literature that studies driver comfort on different road profiles. The deep neural network aims to learn intricate relationships within dataset between features and output. The deep neural network was trained using back propagation algorithm and automated search method was implemented to obtain optimum network structure. The paper starts generating various road roughness profiles according to ISO 8608. Then, through comprehensive examination of rear and front body displacements and pitch angle accelerations, the study highlights system's significant contributions to ride comfort and vehicle dynamics. The proposed framework outperforms other controllers like proportional-integral-derivative (PID), demonstrating its robustness across different road profiles. The results demonstrated effectiveness of proposed control to minimize peak overshooting and settling times which improves ride comfort and stability significantly. Also, the proposed model has small root mean square error (RMSE) values which indicate smoother and less energetic responses, which are typically preferred for passenger comfort. Furthermore, the adaptive neural fuzzy inference system-deep neural network (ANFIS-DNN) has minimum crest factor (CF) which indicates that the signal has fewer peaks relative to its average. [ABSTRACT FROM AUTHOR]

Since the beginning of the 21st century, it has been noted that the rate of energy consumption has increased globally due to the technological progress that has taken place. The industrial sector occupies one of the first sectors in the consumption of electrical energy. Therefore, this paper aims to use the method of artificial neural networks (ANN) to model, analyze, and forecast electricity consumption in Jordan's industrial sector. In the present analysis, several factors affecting energy consumption in this sector are studied and verified, namely: the number of the industrial establishments (ES), the number of employees (EM), the electricity price (E$), the price of fuel (F$), gross output (G), structural effect (GI/GN) and the capacity utilization (CU). Several networks are executed and tested in terms of the root mean square error (RMSE), mean absolute percentage error (MAPE), and coefficient of determination (R2). The results of this study implicated that the electricity consumption levels in the Jordanian industrial segment are highly driven by the number of industrial establishments and employees followed by the gross output, among others. Additionally, the present ANN-predicted electricity consumption results are compared with literature and showed superior accuracy. Finally, this ANN model is combined with the time series analysis approach to forecast the electricity needs of the Jordanian industrial sector for the next decade. [ABSTRACT FROM AUTHOR]

Fuel consumption analysis has become crucial for the local government of Jordan and decision makers for securing economic stability and environmental protection. The current study employs an Artificial Neural Network (ANN) to analyze and estimate fuel use and demand in Jordan's residential sector. Throughout the investigation, five criteria are assessed: population (P), income level (IL), electricity unit price (E$), fuel unit price (F$), and ambient temperature (AT). From 1985 to 2020, data on fuel use and independent variables are gathered from official government reports and literature sources. Therefore, ANN is used to model, to analyze and estimate fuel use and demand in Jordan's residential sector. The results of this study showed that in the Jordanian residential sector fuel demands are mainly driven by the population and fuel price. Additionally, the present ANN model findings accuracy outperforms the literature models. Thus, the current ANN model is combined with time series analysis to forecast the fuel demands in Jordan's household sector for the next decade. The findings of this study are expected to be helpful for the decision makers in the country for developing reliable and productive planning for future energy policies in the household sector.. [ABSTRACT FROM AUTHOR]

Accurate heat transfer predictions during CO2 condensation in porous media are needed to improve two-phase flow devices like refrigeration and air-conditioning systems. However, phase-change condensation system behavior makes predicting of the heat transfer coefficient very difficult. A reference dataset is used by gathering information from previous experimental investigations on heat transfer during condensation of CO2 as refrigerant in porous media at subcritical pressures. This database comprises four key factors: porosities of 39.8–44.5%, mass flow rates of CO2 8 × 10–5–16 × 10–5 kg⋅s−1, input pressures 3450–4300 kPa, and volume flow rates of 3–6 L⋅min−1. In the current study, four machine learning methods—two neural network models and two enhanced adaptive neural fuzzy inference systems—are developed for the prediction of internal heat transfer coefficient. The models are applied as machine learning regression models. For validating proposed models, the internal heat transfer coefficient is predicted and then examined against the corresponding actual values. The output data showed acceptable results for predicting and the proposed models successfully decreased the level of uncertainty in predicting the porous condenser's internal heat transfer coefficient. The error analysis for the proposed methods is reported and discussed by using four statistical criteria. The BPNN, ANFIS-ANN, ANFIS, and ANN-3 models had RMSEs of 2.2036, 2.4413, 3.5532, and 2.7471, respectively. [ABSTRACT FROM AUTHOR]