Search

Fossil fuel power plants continue to contribute significantly to carbon emissions, necessitating a transition towards cleaner energy sources. Despite the growing presence of renewables within the power systems, the incorporation of carbon capture technologies into the traditional thermal power plants holds great potential in emissions reduction. In this paper, the integration of renewable energy sources (RES) and coal-fired power generation units outfitted with carbon capture schemes is addressed. Multiple demand response (DR) programs and hydropower plants are strategically utilized to increase the power system flexibility. To effectively plan the day-ahead (DA) operation of the power system, a presumed market-clearing framework is adopted and modelled as a risk-constrained two-objective stochastic mixed-integer linear programming problem. The proposed framework helps to tackle the uncertainties related to RES and demand variations by employing a hidden Markovian process (HMP) technique. To simultaneously minimize the system's operational costs and CO 2 emissions, an enhanced version of the augmented ɛ -constraint method is employed. To prove its value, the proposed framework is devoted to the 24-bus IEEE reliability test system (IEEE-RTS). The system features substantial penetration of RES (exceeding 87% of peak load) and standard DR options capacities (less than 25% of peak load). The results show a 24% reduction in load peaks, an over 63% decrease in emissions, and a 17% reduction in the overall operation costs. • Proposing a comprehensive linearized mathematical framework accelerating the energy transition. • Coordinating the responsive power systems with carbon capture and hybrid ESS (BESS, PHS, HES). • Enhancing an effective augmented ɛ -constraint technique to simultaneously optimize cost and CO 2. • Presenting robust decision-making with two-stage SP, hidden Markovian process, and CVaR strategy. • Integrated approach: 87% of RES, cost & CO 2 reductions, fewer load peaks, enhanced flexibility. [ABSTRACT FROM AUTHOR]

Integrating renewable energy resources (RES) is a challenge for power system operators due to their fluctuations and unpredictability. At the same time, the water shortage problem and the needs to desalinate more freshwater increase the prominence of sufficient energy resources for sustainable operation. Therefore, this paper presents a market-clearing mechanism in a co-optimization model that coordinates the operation of grid-connected reverse osmosis water desalination plants (RO-WDPs) and the operation of renewable-rich power systems. It is assumed that electric demands can participate in the provision of demand response (DR) to the market via multiple DR options. The DR options related to the general loads are demand shifting, load curtailment, distributed generation, and hybrid energy storage systems (HESS). These HESS includes battery storage and hydrogen storage systems. Pertaining to their special characteristics, a new DR option is proposed as a customized option for RO-WDPs. The market clearing mechanism is assumed to be based on the security-constrained unit commitment (SCUC) and formulated as a stochastic mixed-integer linear programming problem (MILP) to optimally schedule the day-ahead operation of the power system. The presented model is applied on 6-bus and IEEE 24-bus reliability test system (RTS) test power systems with significant penetration of RES to demonstrate its merits. The simulation results show that the system efficiency is enhanced by adding the energy flexibility of RO-WDP without endangering the water demand-supply when using the proposed coordinated model. Hence, the total operation cost is minimized, the RES integration is facilitated, and the hourly electricity prices are smoothened. • We propose coordination model of grid-connected RO desalination and responsive power system. • The power system in our model considers five DR options including hybrid ESS. • We avoid the generic DR model to reschedule the RO desalination flexibility. • The day-ahead operation of our energy–water nexus incorporates system uncertainties. • The incorporated approach minimizes cost and CO2 and enhances the system flexibility. [ABSTRACT FROM AUTHOR]

Solar photovoltaic power generation accurate prediction is crucial for optimizing the efficiency and reliability of solar power plants. This research work focuses on predicting photovoltaic power using various machine learning algorithms, including ensemble of regression trees, support vector machine, Gaussian process regression, and artificial neural networks. Performance of these algorithms is further improved through hyperparameter optimization using Bayesian optimization and random search optimizers. Hourly data with a 30-min temporal resolution for an entire year is collected from a 10 MW Masdar solar photovoltaic project based in the United Arab Emirates. Photovoltaic historical power curve is generated using the System Advisor Model software, and to ensure data consistency, the collected dataset is normalized, with the interrelationships among variables computed using the Pearson relation coefficient. The results substantiate that Gaussian process regression demonstrates the best performance (lowest prediction errors) in terms of computing predicted solar photovoltaic generation power, followed by artificial neural networks, ensemble of regression trees, and the support vector machine across both optimizers. Concerning hyperparameter optimization, Bayesian optimization -based model outperformed support vector machine, Gaussian process regression, and artificial neural networks algorithms, except for the ensemble of regression trees. The proposed work contributes to the advancement of solar photovoltaic power prediction by combining the power of machine learning algorithms with hyperparameter optimization techniques. Additionally, the results emphasize the importance of hyperparameter optimization in enhancing machine learning model performance, providing valuable insights into adaptability and accuracy across varying seasonal conditions. Proposed methodology for solar PV power prediction using machine learning algorithms with different optimizers. [Display omitted] • The study employs machine learning techniques to predict solar photovoltaic power. • It utilizes UAE's Masdar PV project data with 30-min resolution for entire year. • Bayesian and random search are used to enhance machine learning models performance. • Performance evaluation of these models is computed by error, speed and size metrics. • Gaussian process regression shows superior predictive accuracy than other models. [ABSTRACT FROM AUTHOR]

The increased penetration levels of renewable energy resources (RESs) transform the power generation schema into one that is more susceptible to changes due to weather conditions, complicated load profiles, and reduced inertia levels. One of the main challenges that power grids can encounter under varying and low system inertia (SI) is poor frequency response (FR). Therefore, this paper proposes a novel online tool for assessing, predicting, and enhancing the frequency stability (FS) of power systems with renewable power generation and energy storage systems (ESS). Firstly, the tool provides accurate tracking of the SI in real-time settings using recursive least square identification and Kalman filtering to provide accurate and computationally efficient results. Secondly, the tool estimates the FR of the whole system in a virtual environment for different contingencies and inertia levels. The accuracy of the FR is improved by classifying the frequency response models (FRMs) based on the SI levels. Lastly, the proposed tool computes the optimal additional estimated reserve power (ERP) required from PV and battery energy storage systems (BESS) to provide inertial and primary frequency support to the power grid at the onset of a contingency. In addition, a reserve power allocation and load-shedding strategies are proposed and implemented to dynamically adjust the reserve power of PV power plants. This dynamic control of the PV power plant reserve capacity guarantees the availability of adequate reserves for addressing unforeseen systems contingencies. Moreover, dedicated FR controllers for PV and BESS are employed to emulate the synchronous machine's inertial response (IR) and primary frequency control (PFC). The performance of the proposed tool and the FR controllers are tested and validated on a generic power grid and the IEEE 39 bus system using MATLAB Simulink and the OPAL-RT real-time simulation software. • Power system inertia estimation in a real-time environment with minimum data. • Frequency nadir estimation using multi-regressed models based on the system inertia. • Estimation of optimal reserve power for PV and BESS for frequency stability. • Dynamic adjustment of PV sub-optimal operation. • Frequency response controllers for PV and BESS, and optimal load shedding strategy. [ABSTRACT FROM AUTHOR]

• The KDE-UCV is introduced for the first time for long-term power system planning. • The nonparametric model is applied to both of wind speed and solar irradiance datasets. • KDE-UCV produces accurate probability densities than conventional approaches. • Assessments reveal robust performance with high R2 and low error metrics. • K-S test results indicate a clear evidence of a good fit for the KDE-UCV estimator. Uncertainties associated with power generation from wind energy systems and Photovoltaic (PV) power systems present a major challenge for power system planners and operators. To account for such uncertainties, probabilistic models and probability density estimations for wind speed and solar irradiance, and their corresponding wind and PV power are highly required for long-term (multi-year) power system planning, expansion, and dispatching tools. In this article, a novel Kernel Density Estimator (KDE) model using unbiased cross-validation method for bandwidth selection is proposed for the estimation of both wind speed and solar irradiance probability densities. The estimation performance of the proposed model is assessed against the traditional parametric models (Weibull and Rayleigh distributions for wind speed, and Beta distribution for solar irradiance), and the traditional nonparametric KDE approach employing a rule-of-thumb method for bandwidth selection. The performance accuracy of all models is tested using the coefficient of determination R 2 , two error metrics (Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE)), in addition to the Kolmogorov–Smirnov (K–S) test that was used to assess the goodness-of-fit. The proposed approach achieved the highest percentage improvements for R 2 (24% and 23%), and the lowest MAE (66% and 36%) and RMSE (63% and 25%) metrics over the popular parametric distributions for wind speed and solar irradiance, respectively, in addition to the K–S test p-values indicating a clear evidence of a good fit. Results confirm the accuracy and robustness of the probability density estimates for wind speed and solar irradiance produced by the proposed model. [ABSTRACT FROM AUTHOR]