Operating with maximum power point tracking (MPPT) strategy is the key functions of the solar energy management system in solar PV deployment. Therefore, in this paper, the strengths and weaknesses of the MPPT based on an incremental conductance (IC) algorithm are analyzed. Furthermore, assessments in terms of performance of two MPPT-IC techniques are presented. More specifically, a two-stage optimized MPPT-IC algorithm in this study is developed and implemented for isolated solar PV system with the conventional single-stage MPPT-IC algorithm. Both MPPT-IC methods are analyzed and compared based on their tracking efficiencies, steady state error and transient speed responses. For the case of the two-mode MPPT-IC algorithm, significant modification is carried-out by combining the conventional IC and modified constant voltage methods. As a result, a better performance under both sunny and cloudy weather scenarios respectively is achieved with higher tracking efficiencies than the conventional IC algorithm. The two stage-algorithm not only rapidly tracks the maximum power point (MPP) under dynamic weather conditions, but also significantly reduces the fluctuations in power around the MPP when subjected to fast and slow speed’s change of irradiations.
Neural networks have increasingly been utilized in electric drive systems to enhance modeling, control, and optimization. These data-driven techniques enable accurate predictions of complex nonlinear behaviors, including the magnetization characteristics of electric machines. This paper investigates the use of neural networks for predicting magnetization surfaces in switched reluctance motors, a key aspect of their design and operational efficiency. Three neural networks-based methods are studied: classical neural networks, radial basis function neural networks, and physics-informed neural networks. Experimental data from a 7.5 kW switched reluctance motor are used to assess the capabilities of each approach. Moreover, the study evaluates predictive accuracy, computational requirements, and the ability to reflect physical dynamics. Results demonstrate that classical neural networks and radial basis function networks can model magnetization surface, but with inaccuracy due to failure to comply with flux behavior, with radial basis function networks excelling in computational efficiency. Physics-informed neural networks achieve the highest accuracy by integrating physical laws into the learning process. This research highlights the potential of neural networks techniques in advancing switched reluctance motors modeling, paving the way for improved electric drive systems.
During the last two decades, interest in grid integrated renewable energy sources (RERs) into the electrical power grids has significantly grown. The motive being a promising strategy to reduce greenhouse gas emissions generated by conventional power plants (CPPs) and hence environmental consequences, decrease losses in distribution lines to improve power system supply efficiency and reliability to consumers. As a result, several CPPs are being replaced by RERs such as solar PV system, wind energy, micro hydro-turbine, fuel cell, and small gas power plants. These are installed close to customers on low-voltage and medium-voltage sites to reduce losses in the network and improve power quality, overall efficiency and reliability. However, grid-connected RERs via inverter decreases system inertia due to decrease in CPPs with rotating mass serving as an immediate natural response to frequency deviations following network disturbances. Therefore, this paper presents the design of dynamic virtual-inertia control (DVIC) scheme. The developed DVIC incorporates the mechanism to adjust active and reactive demand in response to any frequency and voltage variations to ensure system stability. To validate the suitability of the presented technique, MATLAB/Simulink model is developed and simulated under various operating scenarios. Using the results obtained simulation study, it is evidently the proposed DVIC shows improved performance is achieved than when no inertia control system applied.
The power utility, Tanzania Electric Supply Company Limited (TANESCO) is faced with challenges of supplying adequate power to the ever-increasing demand. To this end, this paper explores the installed Ubungo Gas Power Plant (UGP II) located in Dar es Salaam to find means of utilizing it in co-generation method. The study embarked on detailed modelling of the system and run simulations which indicated that the co-generation can result to power generation efficiency increase between 10.42 to 50.92%. It is recommended to change the UGP II to exploit this co-generation method. Further, the study could be improved by including more parameters of the gas and steam plant and verify them experimentally.
This chapter explores artificial intelligence's transformative impact on wind energy, addressing intermittency challenges to enhance reliability, market integration, and trading efficiency. As wind power expands to meet decarbonization goals, supplying over one-third of global electricity by 2050, its variability strains grids and markets, incurring balancing costs and penalties. The analysis dissects reliability issues, highlighting AI's role in predictive maintenance via machine learning on SCADA data for failure anticipation, deep learning for real-time fault detection using convolutional neural networks, and adaptive controls to minimize downtime and extend asset life. These advancements shift operations from reactive to proactive, leveraging vast datasets to predict and prevent disruptions. It then examines electricity market dynamics, where forecast errors lead to imbalances