The Real-Time Recurrent Learning Gradient (RTRL) algorithm is characterized by being an online learning method for training dynamic recurrent neural networks, which makes it ideal for working with non-linear control systems. For this reason, this paper presents the design of a novel Maximum Power Point Tracking (MPPT) controller with an artificial neural network type Adaptive Linear Neuron (ADALINE), with Finite Impulse Response (FIR) architecture, trained with the RTRL algorithm. With this same network architecture, the Least Mean Square (LMS) algorithm was developed to evaluate the results obtained with the RTRL controller and then make comparisons with the Perturb and Observe (P&O) algorithm. This control method receives as input signals the current and voltage of a photovoltaic module under sudden changes in operating conditions. Additionally, the efficiency of the controllers was appraised with a fuzzy controller and a Nonlinear Autoregressive Network with Exogenous Inputs (NARX) controller, which were developed in previous investigations. It was concluded that the RTRL controller with adaptive training has better results, a faster response, and fewer bifurcations due to sudden changes in the input signals, being the ideal control method for systems that require a real-time response.

In several ways, the AEC has increased connectivity between the businesses, merging business activities, and funneling them to end customers. Moreover, it increased energy consumption and increased CO 2 emissions in ASEAN countries. This study analyzed the driving factors of carbon emissions in ASEAN and identified the differences between member countries based on decomposing the extended Kaya identity via the logarithmic mean divisia index (LMDI) method. Since the energy intensity effect "EI-effect," gross domestic effect "GDP-effect," population effect "POP-effect" and CO 2 emission effect "CO 2 -effect" were a mixture of I(0) and I(1), Johansen cointegration test cannot be applied. Hence the study deployed an autoregressive distributed lag (ARDL). This study's ARDL model captured a long-run and short-run relation of the whole cointegrated variables in ASEAN countries. Based on a panel of cross-country and time-series observations, the study analyses that the ARDL model was used to cover a model of short-and long-run implications. Based on the result, we identified the root cause of significantly increasing CO 2 emission in the past 36 years. This study's result was that a positive long-run relationship interacted with a mostly negative short-run relationship between the energy intensity' EI-effect,' gross domestic effect' GDP-effect,' population effect' POP-effect "and CO2 emission effect' CO2-effect."

This paper presents a detailed description of the data obtained as a result of the computational simulations and experimental tests of an MPPT controller based on an ADALINE artificial neural network with FIR architecture, trained with the RTRL and LMS algorithms that were used as mechanisms of control in an off-grid photovoltaic system. In addition to the data obtained with the neural control method, the data for the MPPT controller based on the traditional Perturb and Observe (P&O) algorithm are presented. The simulations were performed in MATLAB/Simulink software without using the Neural Network Toolbox for controller training. The experimental tests were performed in an open space without shaded areas, exposing the neurocontroller to varying environmental conditions. Additionally, the scripts developed in MATLAB for the neural training algorithms used in the simulations are presented. These computational simulations were structured in five test cases to represent the behavior of each controller under varying environmental conditions. The codes developed in C are part of the implementation of the MPPT neurocontroller in the PIC18F2550, from which the experimental data were obtained. The data and codes presented in this research are available in the Mendeley Data repository, which allows evaluating the performance and optimizing the training algorithms with the purpose of improving the control methods applied to photovoltaic systems.

No abstract

No abstract