Climate change demands clean energy solutions, and renewable sources such as solar and wind are prime candidates. However, their variability poses challenges for their integration into large-scale power systems. This paper addresses this issue by proposing a novel hybrid mathematical model. The proposal integrates both fossil and renewable sources, considering real-world constraints such as system demand, reserves, and transmission dynamics. The model combines several approaches. By using a novel block composition technique, the computational complexity is reduced, making the model applicable to large-scale systems. A neural network is also developed to improve the forecasting of renewable energy production, which is crucial for managing its intermittency. The effectiveness of the proposed model is tested by considering the large Argentinean electricity system, demonstrating its practical applicability. The results show that acceptable forecasts can be obtained for the generation and transmission scheduling of the whole system.

Over time, the number of smart grids installed worldwide is gradually increasing. However, the major portion of the required electricity is still being produced by traditional large-scale and centralized power systems. The main requirement, then, is to study and develop mathematical methods that attend the integration between the two systems previously announced. In this paper, a novel model that addresses this issue is presented. The model minimizes the total operating cost of the large-scale system considering the participation of the smart grid as a dynamic entity, entailing a close relationship between both systems. This approach distinguishes the novel proposal from others that solve similar situations by taking into account the two systems in isolation. Besides, the models that represent the most common organizational structures of the smart grids are also presented in this paper. They are needed to develop the integrated model. Many similar problems in the literature are solved by implementing decomposition techniques, which might obtain a local optimum different from the global one. By contrast, problems with this proposal are solved by using mixed-integer linear programming models that ensure the reaching of a global optimum. The real test case is the integrated Argentine large-scale system and the Armstrong smart grid. Results indicate that the novel model can reach solutions that are 5% lower in comparison with the traditional techniques of considering in isolation. Efficient CPU times enable the possibility of promptly obtaining solutions if there is any change in the parameters. In addition, other benefits, apart from the economical reductions, are also achieved. Operating information closer to the reality of both systems is obtained because it considers the effects of the smart grid in large-scale system solving.

Electricity systems are currently expanding towards more efficient forms of production. Several expansionary strategies are being developed to cover increases in future electricity demand. Goals such as reducing greenhouse gas emissions, increasing the efficiency of operations, and achieving more equitable participation of the actors in charge of the investments are set. Following this premise, this paper presents a multi-objective model that helps in decision-making on the problem of expanding electricity generation. The model considers more realistic views than other works in the literature. The vast majority of the stakeholders in the studied field are satisfied with the present proposal. Investment costs, greenhouse emissions, and investment contribution rates are considered. Also, the actual procedures of the generation and transmission stages are rigorously studied. This means obtaining solutions that are closer to reality. The case study is the electricity system of Argentina. The results obtained indicate that the recommended solutions are the most convenient from all points of view. They constitute a mix of the generation with renewable and non-renewable technologies. The case study reveals emission reductions of up to 25% and it can be achieved that the most vulnerable social groups do not have to finance future system expansions.

Effects of climate change can already be observed in many regions of the world. The basin of the Paraná River, in South America, has been suffering an important drought since 2019 in the whole region. The extension of the crisis has increased the risks in the flora and fauna, losses in logistics of navigation, besides the problems of urban water cleansing. In this regard, the novel proposal presents a new mathematical model to study the impact of this crisis. Besides the traditional constraints of the literature for hydraulic systems, this paper enhances inventory constraints, connections with electric systems, and other considerations as the head effects in electricity generation. Because several equations related to electricity generation are nonlinear (which the subsequent computational effort impacts), this proposal applies linearization techniques to reduce CPU times. The core is related to hydropower production, and the consequences of the water crisis in the regional markets. The mathematical model analyzes the interrelationships between reservoirs and water flows of the basin. To study the effectiveness of the novel proposal, the reported situation of the basin of the Paraná River is studied by considering two scenarios (normal conditions of the river flow and the conditions related to the drought). Results show that the crisis implies daily net economic losses of about 7 million USD for the operators of the power plants. Other problems (different from the ones related to the energy field) are also mentioned and analyzed.