Long-term Power Research Articles (Page 1)

Aiming to solve the challenges of the weak spatial and temporal correlation of medium- and long-term photovoltaic (PV) power data, as well as data redundancy and low forecasting efficiency brought about by long-time forecasting, this paper proposes a medium- and long-term PV power forecasting method based on the Transformer, SP-Transformer (spatiotemporal probsparse transformer), which aims to effectively capture the spatiotemporal correlation between meteorological and geographical elements and PV power. The method embeds the geographic location information of PV sites into the model through spatiotemporal positional encoding and designs a spatiotemporal probsparse self-attention mechanism, which reduces model complexity while allowing the model to better capture the spatiotemporal correlation between input data. To further enhance the model’s ability to capture and generalize potential patterns in complex PV power data, this paper proposes a feature pyramid self-attention distillation module to ensure the accuracy and robustness of the model in long-term forecasting tasks. The SP-Transformer model performs well in the PV power forecasting task, with a medium-term (48 h) forecasting accuracy of 93.8% and a long-term (336 h) forecasting accuracy of 90.4%, both of which are better than all the comparative algorithms involved in the experiment.

Abstract Integrated power systems (IPS) aboard electrified ships require energy management strategies that ensure safe, autonomous operation. Next-generation platforms are expected to make such decisions with minimal human oversight. However, the complex, multidomain, multitimescale dynamics of IPS—combined with high ramp rate loads like electronic warfare systems—pose significant challenges. Additionally, these systems often face uncertain, time-varying, mission-specific constraints that create nonconvex feasible regions, limiting the effectiveness of conventional energy management approaches. This work presents a hierarchical, two-stage framework for safe and adaptive energy management in shipboard IPS. At the upper level, a sampling-based rapidly exploring random tree (RRT) algorithm identifies feasible long-term power and energy trajectories within nonconvex constraint spaces. At the lower level, a robust model predictive control (MPC) scheme ensures accurate trajectory tracking with bounded error, accommodating the dynamics of major components while maintaining constraint satisfaction. The framework is demonstrated on a two-zone IPS model with a high ramp rate load. Simulation results show the proposed planner efficiently generates feasible mission plans that adapt to evolving constraints, while the MPC controller ensures reliable tracking and constraint adherence. By bridging long-term planning with short-term control, this architecture enables safe, flexible, and autonomous operation of complex shipboard power systems. It addresses key limitations of existing strategies in managing nonconvex constraints and dynamic mission contexts, making it well-suited for resilient autonomy in future maritime platforms.

To mitigate renewable energy curtailment and maintain long-term power balance, both planning and operational strategies must be addressed. However, most existing studies on power system capacity optimization focus on a single objective, such as economic efficiency or carbon reduction. To overcome this limitation, this paper proposes a two-stage robust capacity optimization and decision-making framework for power systems that incorporates multi-objective optimization. In the first stage, a bi-level robust capacity optimization model is developed, where the upper-level problem targets capacity expansion planning and the lower-level problem addresses chronological production simulation and operational optimization. The upper-level objectives include minimizing investment and operating costs, maximizing supply reliability, and maximizing renewable energy integration. Secondly, the NSGA-II algorithm is employed to solve the constructed bi-level multi-objective optimization model. Finally, a decision-making model based on the Best–Worst Method (BWM), entropy weighting, and the Technique for Order Preference by Similarity to Ideal Solution (TOPSIS) is constructed to further evaluate and select among multiple Pareto-optimal solutions obtained in the first stage, thereby determining the final capacity configuration scheme. The case study demonstrates that the proposed two-stage framework maintains good stability under scenarios such as extreme weather, ensuring a power supply reliability of 98.78% and a new energy utilization rate of 98.5% under various conditions.

Photovoltaic power generation prediction is crucial for the effective integration of renewable energy into the grid, real-time grid balancing, and the optimization of energy storage systems. However, PV power generation is highly dependent on environmental factors such as weather conditions. Photovoltaic power generation prediction is crucial for the effective integration of renewable energy into the grid, real-time grid balancing, and the optimization of energy storage systems. However, PV power generation is highly dependent on environmental factors such as weather conditions. Effectively integrating environmental information remains a major challenge for photovoltaic power forecasting. This study proposes a hybrid deep learning model that incorporates an adaptive neural network to capture the latent relationships between PV power generation and environmental variables, thereby enhancing forecasting accuracy. The adaptive graph neural network employs a data-driven directed graph structure, where TCN and variable interaction layers are alternately stacked to better model the spatiotemporal coupling among variables for long-term PV output forecasting. The proposed model was evaluated on three sites located in different regions, with a fixed input length of 96 and output horizons ranging from 96 to 768 steps. Compared with state-of-the-art baselines, the model achieved average improvements of 2.19% and 1.57% in MSE and MAE at a 384-step horizon, and 2.81% and 2.47% at a 768-step horizon, respectively, demonstrating superior performance in long-term PV output forecasting tasks.

As global Marine resource development continues to expand into deep-sea and ultra-deep-sea domains, the intelligent and green transformation of deep-sea aquaculture equipment has become a key direction for high-quality development of the Marine economy. Large deep-sea cages are considered essential equipment for deep-sea aquaculture. However, there are significant challenges associated with ensuring their structural integrity and long-term monitoring capabilities in the complex Marine environments characteristic of deep-sea aquaculture. The present study focuses on large deep-sea cages, addressing their dynamic response challenges and long-term monitoring power supply needs in complex Marine environments. The present study investigates the nonlinear vibration characteristics of flexible net structures under complex fluid loads. To this end, a multi-field coupled dynamic model is constructed to reveal vibration response patterns and instability mechanisms. A self-powered sensing system based on triboelectric nanogenerator (TENG) technology has been developed, featuring a curved surface adaptive TENG array for the real-time monitoring of net vibration states. This review aims to focus on the research of optimizing the design of curved surface adaptive TENG arrays and deep-sea cage monitoring. The present study will investigate the mechanisms of energy transfer and cooperative capture within multi-body coupled cage systems. In addition, the biomechanics of fish–cage flow field interactions and micro-energy capture technologies will be examined. By integrating different disciplinary perspectives and adopting innovative approaches, this work aims to break through key technical bottlenecks, thereby laying the necessary theoretical and technical foundations for optimizing the design and safe operation of large deep-sea cages.

Abstract The optimization of energy consumption is an emerging topic in the manufacturing sector because it is the first and most likely step for the transition toward a greener manufacturing strategy. The present study focuses on monitoring and optimizing the energy consumption of milling machines, which are essential tools in modern manufacturing and are used by many manufacturing companies but which also consume a large amount of energy and generate significant environmental impacts. This study presents a step-by-step methodology for energy profiling of milling machines using vector-quantization–based unsupervised machine learning. The process includes long-term power monitoring, preprocessing with peak shaving, clustering into machine states using K-means, and subsystem-level analysis. Data were collected from a PAMA Speedram 2000 milling machine, and the approach demonstrated its ability to differentiate between operational states and identify energy optimization opportunities. Results show that adjusting auxiliary system duty cycles based on machine states can reduce total energy use by more than 50 % in some scenarios. Our findings indicate that specific operational modes exhibit distinct energy-consumption characteristics, which can be leveraged to enhance the efficiency of milling operations. A scenario that implements some solutions to develop a greener milling process is presented based on the partial use of the most energy-demanding auxiliary systems.

Distributionally robust multi-stage stochastic programming for mid- and long-term cross-regional power markets

Accurate prediction of coal demand is essential for optimizing energy resources in long-term power system planning. This paper examines the coal demand in North China from 2007 to 2022 using econometric methods to identify key influencing factors as input variables. Then, the Sparrow Search Algorithm (SSA) is used to optimize the key parameters of the Least Squares Support Vector Machine (LSSVM) algorithm to enhance the prediction accuracy of coal demand. Case studies are conducted on actual data in North China, and the results show that the proposed hybrid SSA and LSSVM method outperforms traditional approaches in small-sample, multivariable forecasting, making it suitable for predictions in long-term power system planning.

Accurate short-term and long-term power load forecasting plays a pivotal role in ensuring the reliability and economic efficiency of modern smart grids. To address the challenges of complex temporal patterns, multi-scale periodicity, and dynamic variability in load data, we propose a novel deep forecasting framework named TPA-Net (Temporal Pyramid CNN–BiLSTM–Attention). The proposed architecture consists of three key components: a temporal pyramid multi-scale convolutional module to extract hierarchical periodic features across hourly, daily, and weekly levels; a bidirectional LSTM (BiLSTM) module to capture global temporal de-pendencies in both forward and backward directions; and an attention mechanism to dynamically emphasize critical time steps. Extensive experiments conducted on real-world power consumption datasets demonstrate that TPA-Net consistently outperforms state-of-the-art baselines across multiple forecasting horizons (1-h, 6-h, and 24-h), achieving significant improvements in RMSE, MAPE, and R 2 metrics. These results highlight the effectiveness and generalizability of TPA-Net in complex load forecasting scenarios.

With the rapid growth of renewable energy integration, power systems are facing increasing uncertainty and variability in operation. The intermittent and uncontrollable nature of wind and solar generation requires operational decisions to anticipate future fluctuations, creating strong temporal coupling across days. This leads to large-scale mixed-integer linear programming (MILP) with a large number of binary variables, which is computationally intensive—especially in year-long simulations. As a result, there is a growing need for efficient modeling approaches that can reduce complexity while preserving key temporal features. This paper proposes a temporal–spatial acceleration framework for long-term power system operation simulation. In the temporal dimension, a monthly K-means clustering algorithm is applied to reconstruct typical scenario days from 8760 h time series, preserving the characteristics of seasonal and intraday variability. In the spatial dimension, thermal units with similar characteristics are aggregated, and binary decision variables are relaxed into continuous variables, transforming the MILP into a tractable LP model, and thereby reducing computational burden. Case studies are performed based on the six-bus and the IEEE RTS-79 systems to validate the framework, being able to provide a practical solution for renewable-integrated power system planning and dispatch applications.

Multi-Granularity Autoformer for long-term deterministic and probabilistic power load forecasting.

In heating-intensive areas, low-to-intermediate temperature hydrogeothermal energy (depth >1500 m below ground level, temperature <200 °C) has the potential to replace fossil fuels in the heating sector. One of the biggest obstacles to the wide-spread implementation of hydrogeothermal energy is the exploration risk, the probability of finding geological conditions, which do not yield long-term economic thermal power. Here, we develop and investigate different assessment criteria of potential hydrogeothermal projects to minimize this exploration risk and the associated economic consequences in an inventory-portfolio approach. To do so, we combine a simplified inventory-portfolio approach with uncertain and spatially varying subsurface parameters and a cost model in a Monte Carlo simulation framework. We use an established hydrogeothermal energy play in SE Germany as an example and evaluate the performance of the tested assessment criteria vs. average exploration risk, total produced energy, total cost and cost of failure due to non-discoveries. Our results demonstrate that careful selection of formalized assessment criteria is key to mitigate exploration risk. We conclude that a holistic top-down planning approach, which combines the comprehensive and standardized characterization of geological and economic conditions on geothermal play-scale, is necessary to effectively employ hydrogeothermal energy as a replacement of fossil fuelled heating.

Abstract. Models used in wind resource assessment (WRA) range from engineering wake models and computational fluid dynamics models to mesoscale weather models with wind farm parameterizations and, more recently, large-eddy simulation (LES). The latter two produce time series of wind farm power of a certain period. This simulation period is, in the case of LES, mostly limited to ≤ 1 year due to the computational costs. However, estimates of long-term (O(10 years)) power production are of high value to many parties involved in WRA. To address the need to calculate long-term annual energy production from ≤ 1-year model runs, therefore, this paper presents methods to estimate the long-term (O(10 years)) power production of a wind farm using a ≤ 1-year simulation. To validate the methods, a 10-year LES of a hypothetical large offshore wind farm is performed. The methods work by estimating the conditional probability densities between wind farm power from the LES and wind speed from reanalysis data (ERA5) from a short (≤ 1 year) LES run. The conditional probability densities are then integrated over 10 years of ERA5 wind speed, yielding an estimate of the long-term mean power production. This “long-term correction” method is validated on varying simulation periods, selected with four different day-selection techniques. When applied to a simulation period of 365 consecutive days, the methods can estimate the 10-year mean power production with a mean absolute error of around 0.35 % of the long-term mean. When choosing the simulation period with day-selection techniques that represent the long-term climate, only roughly 200 simulation days are needed to achieve the same accuracy. Finally, a method to also include wind observations in the long-term correction is presented and tested. This requires an additional “free stream” LES run without active turbines and gives estimates of long-term power and wind that are corrected for a potential LES bias. Although validation of this final approach is difficult in the employed modeling strategy, it gives valuable insights and fits within the common WRA practice of combining models and observations. The presented techniques are based on physical arguments, computationally cheap, and simple to implement. Furthermore, they are not limited to LES but can be applied to other time-series-based models. As such, they could be a useful extension for the diverse set of modeling, observational, and statistical techniques used in WRA.

The rapid advancement of ultrafast science necessitates reliable and efficient generation of few-cycle pulse sources. However, achieving this presents several challenges, including maintaining high transmission efficiency, managing nonlinear effects, and ensuring spectral coherence. In this work, we demonstrate a compact approach to generate 0.74-mJ, 8-fs pulses using a two-stage hybrid MPC and filamentation system with a high overall transmission of 74%. This few-cycle source features uniform spatial spectral distribution and excellent long-term power stability.

Ultrasound-based mechanical energy harvesting materials and thereof devices emerge as a promising technology for powering implantable electronic devices within the human body. Although this approach takes advantage of ultrasound's noninvasive nature, the compact design inherent in frictional electricity-generating components, and the biocompatibility of materials engaged in friction, the behavior of frictional materials under ultrasound necessitates a systematic and controllable design strategies.Here, we present the development of an ultrasound energy harvesting device engineered for long-term stability. Our results found that the incidence and reflection characteristics of ultrasound interestingly vary upon pairs of materials exhibiting disparate moduli in the ultrasound environment. We not only experimentally but also theoretically establish that a pronounced difference in modulus values between materials facilitates the generation of high-amplitude oscillations. Importantly, we validate that the manifestation of oscillations during ultrasound propagation is governed by the high and low modulus boundaries encountered by the material. This design strategy may benefit long-term performance of the ultrasound-based energy harvesting device. We, through this design strategy, experimentally validated that the ultrasound-based energy harvesting device that has the potential for long-term stability (>6 weeks) and continuous power generation.

Abstract This article examines the suspension of the Yucatán Solar PV park, a project awarded through Mexico’s first long-term power auction, situating it within the broader context of the country’s energy reform and its implications for Indigenous territories. Drawing on ethnographic fieldwork and informed by political ecology and critical agrarian studies, the study analyzes how legal instruments, spatial tools, and discursive practices are deployed to render Maya territory investable for solar development. Central to the analysis are the dynamics of land acquisition, which reveal the land's profound social, cultural, and ecological significance for Maya communities, including the location of a sacred site. These dynamics exemplify green grabbing as an initial phase in the contested processes of land control. The article outlines the mechanisms of legitimation through which actors seek to consolidate authority over land, as well as the counter-processes of resistance, legal action, and competing territorial claims that ultimately led to the project’s suspension. The study considers the wider implications of the suspension for the social fabric and ecological integrity of surrounding communities, offering critical insights into the contested politics of land control and territorial transformation under green capitalism, and how grassroots resistance can disrupt these trajectories.

Efficient long-term electric load forecasting is vital for power system stability, yet traditional time series models often fall short in addressing complex trends and seasonal variations due to their reliance on fixed patterns and single-dimensional feature extraction. To overcome these limitations, this paper introduces the Learnable-DTimesNet-Linear model for enhanced load forecasting accuracy. The model leverages learnable decomposition to adaptively separate time series into seasonal and trend components. Seasonal sequences are processed with an enhanced TimesNet to capture periodicity, while trend sequences are modeled via a weighted summation using a linear model. This approach enables the model to adaptively capture subtle temporal fluctuations, improving predictive precision. Validation against six baseline models, including the original TimesNet, demonstrate the superiority of the proposed method, with a reduction in mean squared error by 10%–22%. These results underscore the Learnable-DTimesNet-Linear model's efficacy in handling complex time series data for accurate long-term electric load forecasting.

The article addresses the problem of determining the permissible long-term load for power transformers while considering their thermal regime. The relevance of this work is due to the high level of physical wear and tear on transformer equipment and the need to improve its operational reliability. Exceeding the temperature of windings and oil accelerates insulation aging, necessitating the development of methods to predict the thermal state of transformers. The purpose of this work is to develop a methodology for determining the permissible long-term load of a transformer depending on ambient temperature. The scientific novelty lies in the development of an approximate model of the thermal operating mode of a power transformer, allowing for the prediction of element temperatures and permissible loads. Materials and methods. Research methods include mathematical modeling using Newton – Rikhman equations to describe heat exchange between thermally homogeneous elements of the transformer. The fourth-order Runge – Kutta method with a time step of 30 seconds was used for numerical solution of the system of differential equations. Calculations were performed using author-developed programs in Python with Numpy and Matplotlib libraries. A comparative analysis of simulation results with experimental data from open sources and GOST 14209-85 requirements was conducted. Results. A mathematical model is proposed, treating the transformer as a system of three thermally homogeneous elements – the core, windings, and oil – with heat exchange described by Newton – Rikhman equations. Equations are obtained that make it possible to determine the steady-state temperature of transformer elements and the maximum allowable load factor, taking into account the limitation of the temperature of the winding and oil, the set ambient temperature and the load factor. Approximate formulas were obtained for determining the volumes and contact surface areas of transformer elements based on a calculation scheme. Ratios are presented that allow these values to be estimated approximately depending on the rated power for geometrically similar transformers. It was shown that the heat transfer coefficient between the oil and the environment has the greatest impact on the thermal regime. A condition for the permissible operation of the transformer was formulated, taking into account the required load level and ambient temperature. Simulation results were validated by comparison with experimental data. Conclusions. A mathematical model of the transformer’s thermal regime has been proposed, taking into account the inertia of temperature changes in its thermally homogeneous elements. The influence of the load factor and ambient temperature on the steady-state temperature values of the elements has been determined. A condition for the permissible operation of the transformer has been obtained, ensuring the limitation of winding and oil temperatures.

The rapid growth of the private space industry has intensified the demand for lightweight, efficient, and cost-effective photovoltaic technologies. Metal halide perovskite solar cells (PSCs) offer high power conversion efficiency (PCE), mechanical flexibility, and low-temperature solution processability, making them strong candidates for next-generation space power systems. However, exposure to extreme thermal cycling, high-energy radiation, vacuum, and ultraviolet light in space leads to severe degradation. This study addresses these challenges by introducing three key design strategies: self-healing perovskite compositions that recover from radiation-induced damage, gradient buffer layers that mitigate mechanical stress caused by thermal expansion mismatch, and advanced encapsulation that serves as a multifunctional barrier against space environmental stressors. These approaches enhance device resilience and operational stability in space. The design strategies discussed in this review are expected to support long-term power generation for low-cost satellites, high-altitude platforms, and deep-space missions. Additionally, insights gained from this research are applicable to terrestrial environments with high radiation or temperature extremes. Perovskite solar cells represent a transformative solution for space photovoltaics, offering a pathway toward scalable, flexible, and radiation-tolerant energy systems.

Abstract: Renewable energy sources are increasingly critical in addressing global energy needs while reducing carbon emissions and energy costs. Accurate forecasting of power generation in solar power plants is essential for efficient energy management and planning. This study introduces a novel hybrid prediction model that combines several prevalent machine learning algorithms to improve the accuracy of solar power generation forecasting. Using real meteorological and production data, the proposed model significantly outperforms individual prediction models. The hybrid model's integration of meteorological data ensures more reliable short-term and long-term power predictions, contributing to improved decision-making in solar plant operations. The results demonstrate the advantages of this approach, providing valuable insights into enhancing the predictability and operational efficiency of solar power plants.Cite this article as: N. Aksoy and V. M. I. Genc, “Improving accuracy in solar power plant power generation prediction: A hybrid model proposal,” Turk J Electr Power Energy Syst., 2025; 5(1), 10-18.