Solar

Deep networks dominate photovoltaic (PV) thermographic fault detection but need large annotated datasets and resist interpretation. We present a lightweight, interpretable infrared workflow combining support-vector-machine (SVM) module/background segmentation from four handcrafted features with an adaptive grid analysis labelling regions as nominal-intensity, high-intensity anomaly or low-intensity anomaly relative to a module-internal reference; the anomaly classes are inspection candidates, not confirmed faults. Evaluation used 21 close-range images of one 20 W module—recorded with the camera’s visible-light edge fusion active, so they are fused infrared/visible frames—and all 596 of a public five-sector UAV dataset. Segmentation against manual masks reached a mean intersection-over-union of 0.64; a feature ablation shows intensity statistics dominate, and an end-to-end Otsu pipeline gives almost the same high-intensity share (4.54% versus 4.50%): the SVM contributes reproducibility—removing the manual segmentation threshold, though not the empirical +48/−60 offsets—not accuracy. High-intensity regions concentrated in the module’s lower half, co-locating with a bus-bar defect known from hardware inspection—suggestive, not validated. The single-module, image-level close-range evaluation is optimistic, and the UAV shares, from a separately trained SVM, illustrate cross-domain application only. Segmentation runs at about 15 images per second on CPU. The method is a relative-intensity thermal screening workflow, not a validated defect-diagnosis or plant-health assessment method, and applies only where acquisition is controlled and the offsets are recalibrated for the target camera and palette. Full article

We report a two-layer computational workflow for designed heterocyclic polymer dimers as candidates for organic solar cell (OSC) materials. The workflow integrates geometry-optimized B3LYP/6-31G(d) quantum-chemical descriptors (HOMO, LUMO, gap, dipole moment) computed for five fully disclosed monomer–dimer pairs (1m–5m; 1d–5d), with a verified, literature-curated 17-entry OSC dataset (PCE 3.6–19.9%, years 2016–2024) modeled by a non-tautological ridge regression baseline (Model A; predictors Year + source_block + log10 hole mobility). All five dimers were computed under uniform neutral closed-shell conditions. Pareto-front analysis in the gap–dipole descriptor space identifies dimer 2d (difluorinated thiophene–diazine D-A dimer; gap 1.74 eV, dipole 16.59 D) as the Tier I lead candidate, with 3d (bis(thiophene–triazine) dimer) and 1d (bis-thiophene–thiazole dimer) as additional Tier I candidates. Model A yields R²(LOOCV) = 0.660, MAE = 2.36%, and RMSE = 3.67%, surviving a 500-shuffle permutation null at empirical p < 0.001. A descriptor-augmented Model B (Eg + HOMO added) demonstrates that the present literature dataset cannot support a deployable molecular-descriptor regression without expansion. The combined DFT–regression workflow provides a transparent screening framework that identifies 2d as the priority synthesis target. Full article

Energy system optimization seeks to utilize multiple energy sources efficiently under technical, economic, and environmental constraints. The increasing integration of renewable energy and the need for sustainable operation have made the optimal planning and management of hybrid energy systems crucial. Classical optimization methods, including Linear Programming, Nonlinear Programming, and simulation-based models, often face limitations when addressing high-dimensional and nonlinear problems. This study introduces a deep learning–based surrogate modeling framework for sizing the components of hybrid renewable energy systems. Initially, Particle Swarm Optimization (PSO) is employed to determine the optimal component sizes for a large number of synthetically generated hourly solar irradiance and load profiles. These optimal solutions are then used as target labels. The associated annual time-series data are transformed into multi-channel Data Map (DMAP) images, which serve as inputs for convolutional neural networks (CNNs). After training, the CNN models are capable of directly estimating the required number of photovoltaic (PV) panels, inverter capacity, and battery units from the DMAP images, eliminating the need to perform the iterative PSO optimization during the prediction stage. Various convolutional neural network architectures, including ResNet, DenseNet121, RegNet, ConvNeXt, EfficientNet, SqueezeNet, MobileNet, and InceptionV3, were evaluated for this multi-output regression task. The results indicate that ResNet and DenseNet121 achieve the best performance, while ConvNeXt provides strong results with a modern architectural design. Among the evaluated models, DenseNet121 achieved coefficients of determination (R²) of 0.934, 0.988, and 0.947 for predicting the sizes of the PV array, inverter, and battery bank, respectively. These results correspond to an average prediction accuracy of approximately 90.6%. ResNet produced similar performance, with its highest R² value reaching 0.983 for inverter sizing. Lightweight networks such as SqueezeNet and MobileNet demonstrate notable effectiveness for resource-constrained systems, whereas InceptionV3 underperforms in leveraging its multi-scale architecture. These results demonstrate that, once the models have been trained, deep learning–based surrogate models can generate sizing decisions comparable to those obtained using PSO with only a fraction of the computational effort. As a result, they provide a fast and practical alternative to conventional iterative optimization methods for component sizing in smart grid and sustainable energy planning applications. Full article

Night-time low temperature remains a major constraint on thermal stability, crop safety and energy-efficient operation in winter solar greenhouses, especially when heat release and auxiliary heating are triggered only after the indoor temperature has approached a low temperature threshold. This study developed a temperature prediction-based active–passive heat storage and release system integrating Internet of Things monitoring, liquid neural network (LNN)-based multi-horizon temperature forecasting, heat storage and release circulation, and decision-making control. The LNN achieved the best forecasting performance among the tested models, with MAE/RMSE values of 0.620/0.775, 0.683/0.854 and 0.758/0.948 °C for 12 h, 24 h and 48 h forecasts, respectively, and was embedded into the system for prediction-assisted operation. A continuous 30-day winter test was conducted in two consecutive stages: heat storage and release without predictive control (HS-NPC, days 1–15) and with prediction-assisted operation (HS-PC, days 16–30). During the consecutive-stage winter test, HS-PC showed higher daily minimum indoor temperature and night-time mean temperature than HS-NPC by 1.92 °C and 4.29 °C, respectively, while reducing daily exposure below 13 °C and 10 °C by 55.0% and 98.2%. Stage-based equivalent input-energy evaluation indicated reductions of 17.1% and 34.3% for HS-NPC and HS-PC relative to the corresponding TG reference periods, respectively. Because HS-NPC and HS-PC were tested in consecutive weather windows rather than in fully synchronized parallel experiments, these improvements should be interpreted as stage-based operational benefits supported by the TG reference and outdoor environmental statistics, rather than as completely weather-independent causal effects. These results indicate that integrating temperature forecasting with heat storage and release regulation can improve low-temperature buffering and energy-saving operation in winter solar greenhouses, while further synchronized or weather-normalized validation is still needed. Full article

The integration of renewable energy into smart grids is beneficial for a sustainable future and the environment. Still, it has challenges such as energy optimisation, environmental conditions and power quality degradation. Existing Maximum Power Point Tracking (MPPT) techniques often focus on tracking efficiency under steady-state conditions, ignoring the impact of real-time variation in environmental conditions and load. The predictive power flow control (PPFC) algorithm is available with one or more fixed MPPT algorithms. No studies have reported on how the choice of MPPT affects PPFC harmonic mitigation. This paper addresses both concerns through a systematic comparative analysis of MPPT techniques integrated with a PPFC method to mitigate harmonics in renewable-integrated smart grid systems. To address this research gap, a comprehensive comparative analysis of various MPPT techniques, such as Perturb and Observe (P&O), Incremental Conductance (INC), Fuzzy Logic Control (FLC), and hybrid Machine Learning (ML) techniques, integrated with PPFC to achieve effective harmonic mitigation in a smart grid environment is conducted. A 3 MW solar farm integrated with a battery storage system is modelled in MTALB/Simulink 2025b under real-time varying conditions, such as environmental and load variations over time in Auckland, New Zealand. The study focuses on key performance parameters such as total harmonic distortion (THD), power loss, stability and efficiency. The Adaptive Neuro-Fuzzy Inference System (ANFIS)-based MPPT controller, integrated with forecast-based power flow control, achieved overall performance by providing higher efficiency (97.5%), effective harmonic mitigation, and enhanced system stability under the nonlinear behaviour of the photovoltaic system. The proposed ANFIS-based system ensured a stable and smooth power output under varying environmental conditions, outperforming conventional and other intelligent MPPT techniques. Full article

The sizing of rooftop solar energy systems is commonly based on the most visible load or on generic end-use allocations, leading to an inadequate distribution of the limited rooftop area between heat and electricity. This study formalizes the energy audit within a three-level deterministic framework that selects and sizes solar technologies by energy vector: demand is first decomposed by vector; the technology for the thermal vector is then selected through a levelized cost of heat selection ratio $\psi$, while the photovoltaic system of the electrical vector is sized for self-consumption; and the rooftop area is finally allocated among vectors according to marginal value per unit area. In a 75-bed university residence in Cienfuegos, Cuba, air conditioning is the dominant energy end-use in terms of installed power (accounting for $77\%$ of the connected load), whereas the thermal vector dominates annual energy consumption (domestic hot water: 127,440 versus 76,818 kWh/year for electricity; thermal-to-electric ratio $1.66$). Solar thermal technology has been selected for the thermal vector ($0.018$ versus $0.088$ USD/kWh_th; $\psi =0.21$, a robust value according to the sensitivity analysis), and the marginal value (≈111 versus ≈32 USD/(m²·year)) allocates 104 m² to solar thermal collectors and 134 m² to photovoltaic energy, thereby reversing the original design that prioritized photovoltaic energy. The resulting portfolio achieves an annual solar fraction close to $100\%$ in both vectors on an energy balance basis, avoids $86.5$ t of operational CO₂ emissions per year, and combines a simple payback of $1.1$ years (solar thermal) with a net present value of 55,327 USD and an internal rate of return of $28\%$ (photovoltaics). The sizing decision is shown to be robust to the choice of statistical design criterion (median, mean, P90, maximum), and none of the three framework decisions is reversed under $\pm 30\%$ parameter variations. By replacing the subjective weightings of multi-criteria methods with observable economic criteria, the framework provides a replicable and auditable design protocol. Full article

Grid-connected photovoltaic (PV) systems paired with battery storage are becoming a core element of low-carbon distribution networks. This paper develops a complete closed-form formulation together with an independent, simulation-based verification of a single-phase grid-connected PV-battery microgrid sized for high-irradiance residential conditions in Saudi Arabia, using measured solar-resource and tariff data for Riyadh. A 6.25 kW monocrystalline array feeds a 400 V DC link through a perturb-and-observe boost stage; a bidirectional converter couples a 13.5 kWh LiFePO₄ battery; and an IEEE 1547 smart inverter interfaces a 230 V grid through an LCL filter. Governing equations for every subsystem are derived and evaluated numerically, and a Python re-implementation of the phasor power-flow model verifies the analysis over a 24 h cycle run to periodic steady state, reproducing the reference design values with a mean absolute error of 0.5%. Using measured monthly solar-resource and temperature data for Riyadh, a full twelve-month analysis gives an annual self-sufficiency of 51.8% and a PV self-consumption of 72.9% for the optimised energy-management scheme. A dedicated time-domain switching simulation with FFT analysis shows that the LCL filter limits grid-current total harmonic distortion to 0.8%, far below the L-filter value of 6.2% and below the 5% current-distortion reference of IEEE 519 (full compliance additionally requires the PCC short-circuit ratio). Twelve-month, battery-size and load-sensitivity studies confirm robustness, and a techno-economic assessment based on the Saudi Electricity Company residential tariff quantifies levelized cost, payback and battery degradation, showing that economic viability hinges on tariff reform. Full article

Studying photovoltaics in engineering and science curricula is a time-consuming and expensive activity. This is why simulations are used, which, however, do not allow direct observation of the physical phenomena occurring in the devices. Based on an array of LEDs (light-emitting diodes) in photodetection mode, a low-cost educational platform for simulating photovoltaic systems including bypass and blocking diodes was developed. This allowed for the experimental characterization of the system’s I-V and P-V characteristics, obtained with a variable-load method under controlled lighting, as well as the qualitative reproduction of key photovoltaic phenomena such as mismatch and bypass diode activation. Additionally, the system allows for quantitative analyses starting from a reference value of 25 μW, obtained under full illumination conditions. This value will inevitably decrease as the platform’s operating conditions worsen, intentionally generated to study the behavior of the platform. Although the method does not provide a representation of the real photovoltaic field, it provides a simple and low-cost tool for the experimental study of photovoltaic behavior. The paper has been conceived for educational purposes, oriented towards laboratory teaching activities. Full article

This study experimentally compares the thermal performance of two vertically oriented passive solar water heating systems under arid outdoor conditions in Gabes, Tunisia: a thermosiphon solar water heater (TSWH) and an integrated solar water heater (ISWH). Both prototypes were installed side by side, facing south, with identical collector areas and the same climatic exposure. Experiments were conducted over two consecutive clear-sky days under no-load and load conditions. The systems were evaluated in terms of water-temperature evolution, thermal stratification, thermal efficiency, heat-retention behaviour, overall heat-loss coefficient, and useful hot-water production. The results showed that the ISWH achieved higher peak water temperatures, reaching 59.8 °C and 57.5 °C during the two test days, compared with 48.55 °C and 53.65 °C for the TSWH. The ISWH also showed higher peak thermal efficiencies of approximately 49% and 48%, while the TSWH reached approximately 41% and 40%. Under hot-water extraction conditions, the ISWH delivered about 25 L of usable hot water at 45 °C, compared with 19 L for the TSWH. However, the TSWH exhibited better thermal retention, with a lower overall heat-loss coefficient of 1.763 W/m²K compared with 2.38 W/m²K for the ISWH. These findings demonstrate a clear trade-off between rapid daytime heat capture and non-solar heat preservation. The ISWH is more suitable for applications requiring higher daytime hot-water production, whereas the TSWH is preferable when improved heat retention after solar input decreases is required. Full article

Effective solar receiver design is pivotal to improving the efficiency of concentrated solar collectors, thereby advancing sustainable energy development. This study investigates the impact of receiver configuration on the thermal and exergy performance of parabolic trough collectors and proposes adopting a commercially available copper-finned tube as the receiver. Notably, thermal efficiency improved when the flat corrugated fin was redesigned into an arc-shaped thin-profile configuration. Specifically, the arc-shaped fin receiver outperformed its flat-fin counterpart, achieving thermal efficiencies of 75.8% and 68.5% at a mass flow rate of 0.1 kg/s, respectively. At a lower flow rate of 0.033 kg/s, peak water temperatures of 80 °C and 72 °C were recorded for the arc-shaped and flat-fin receivers, respectively. Comparative analysis further revealed that the arc-shaped fin yielded an average heat transfer coefficient 19–33.9% higher than that of the flat-fin receiver. Furthermore, at 0.033 kg/s, the arc-shaped fin demonstrated superior exergy efficiency, with peak and average values of 7.2% and 4.2%, respectively, compared with 5.2% and 3.0% for the flat-fin receiver. These findings highlight the potential of arc-shaped copper-finned tube receivers to optimize parabolic trough collector performance, offering a promising avenue for advancing sustainable energy development. Full article

This work reports the results of a study on the optical, morphological, and structural properties of cesium lead bromide iodide mixed perovskite thin films (CsPbBr_3−xI_x), synthesized by sequential evaporation of precursors (CsBr, PbBr₂, PbI₂). First, the deposition conditions were optimized to obtain thin films predominantly composed of the pure CsPbBr₃ phase. Subsequently, the influence of partial substitution of Br⁻ by I⁻ on the film properties was investigated. Particular emphasis was placed on evaluating the effect of partial Pb²⁺ substitution by Zn²⁺ on the optical, morphological, electronic, and structural properties using optical transmittance, photoluminescence, scanning electron microscopy (SEM), X-ray diffraction (XRD), Urbach energy analysis, and density functional theory (DFT) calculations. Zn²⁺-doped CsPbBr_3−xI_x films were prepared by evaporating a ZnBr₂ layer onto the pre-deposited PbBr₂/PbI₂ precursor layers. It was found that Zn²⁺-doped inorganic CsPbBr_3−xI_x perovskite films exhibit enhanced crystallinity and improved surface morphology. Additionally, photoluminescence characterization confirms that non-radiative recombination decreases significantly, apparently due to a reduction in intrinsic defect density. The effect of Zn²⁺ doping on the power conversion efficiency of carbon-based planar solar cells was also evaluated. Collectively, Urbach energy, photoluminescence, and SEM analyses revealed that the optimal Zn²⁺ doping range for CsPbBr_3−xI_x perovskite films is ≤5%. Full article

By the end of 2024, global photovoltaic (PV) capacity exceeded 2.2 TW, shifting planning from feasibility demonstration toward site–technology co-selection under energy, technical, economic, environmental, territorial, and socio-regulatory constraints. The existing multicriteria literature treats site and technology selection as independent problems under an implicit infinite-grid assumption, which is untenable in markets such as Chile and Peru. This study develops and validates an integrated Delphi–AHP framework with six criteria and eighteen subcriteria calibrated by twenty-eight experts from six Latin American countries. The framework underwent Delphi binary validation, AHP consistency control ($C{R}_{\mathrm{agg}}$ between 0.0013 and 0.0247; discard rate 2.6%), geometric-mean aggregation, deterministic sensitivity analysis, Monte Carlo simulation (10,000 iterations), rank-reversal testing, and nonparametric subgroup analysis. The dominant pair $\{I_2,E{c}_2\}$, consisting of grid hosting capacity and LCOE, appears as Top-2 in 84.77% of Monte Carlo iterations and is preserved across 15 of 16 leave-one-out scenarios. Grid hosting capacity surpasses useful solar resource by a factor of 3.41. A demonstrative application to 18 site–technology alternatives confirms the ranking, with an objective-weighting benchmark (entropy, CRITIC) yielding concordant results (Spearman $\rho \ge 0.89$). The findings formalize a shift in the PV planning bottleneck from solar resource to grid capacity. Full article

Space missions working under harsh heliocentric conditions demand more efficient photovoltaics operating under high solar concentration, high temperatures, and harsh radiation conditions. Although most simulation work has been conducted using the terrestrial AM1.5 spectrum, AM0 high concentrators are of great importance to realistic satellite missions. Though III–V multijunction solar cells are currently the norm in space applications, their efficiency under extremely high solar concentration ratios is not yet optimized to support future space missions. This work designs and numerically optimizes a GaAs VTJ solar cell using SILVACO ATLAS software (5.40.0.R). In the optimization, the thickness of the front and back layers, as well as the doping profile within the emitter, base, and tunnel junction regions, were adjusted. The important PV semiconductor attributes, including the short-circuit current density (Jsc), open-circuit voltage (Voc), fill factor (FF), and efficiency (η), were examined over a concentration factor ranging between 1 and 10,000 suns. The efficiency of the optimized VTJ solar cell increased from 20.4% at 1 sun to 26.0% at 10,000 suns. This is mainly due to the near-linear increase in Jsc and the stable FF, which remains between 87% and 89%. In addition, the solar cell shows a steady increase in Voc between 1.85 V and 2.33 V. An optimized GaAs VTJ solar cell design is a promising component in future space missions, which require high power density and are suited to operating under high heliocentric orbits, such as in the Parker Solar Probe and solar-electric propulsion systems. Full article

Water is an indispensable resource for the survival of all living organisms on Earth. However, many coastal villages continue to face challenges in accessing potable water, particularly during extended droughts. This comprehensive study evaluates the implementation and performance of a solar desalination system that employs photovoltaic (PV) panels and a parabolic solar concentrator to meet clean water demand in a drought-prone area of Indonesia. The system harnesses both solar-generated electricity and thermal energy to power an advanced desalination apparatus, effectively converting seawater into safe drinking water. Over a rigorous 4-month testing period, the device maintained an average steam outlet temperature of 105.9 °C, enabling a direct single-stage evaporation and condensation desalination process. Under optimal sunlight conditions, the system produced 1500 mL of purified water every 30 min, resulting in a total daily output of approximately 12 L (1500 mL × 8 cycles over 4 h). Laboratory analysis revealed a decrease in pH from 8.0 in raw seawater to 6.8 in treated water after post-treatment pH adjustment, meeting established safety standards for human consumption. Electrical conductivity measurements fell from 40–50 mS/cm to 480–500 µS/cm, confirming substantial salt removal. These results demonstrate the system’s capacity to generate potable water using sustainable energy sources and support circular economy principles by repurposing renewable resources for water desalination in water-scarce environments. Full article

The rapid growth of solar photovoltaic (PV) capacity is increasingly reshaping the operation of electricity systems, particularly in countries where renewable energy already represents a large share of generation. In Albania, where electricity production is strongly dominated by hydropower, increasing solar penetration is expected to affect short-term system behaviour, especially in terms of variability, surplus generation, and ramping dynamics. This study investigates PV integration at the system level using hourly electricity demand data for 2024 together with PV generation profiles scaled to different capacity scenarios. PV scenarios representing installed capacities of 150, 300, and 450 MWp, based on real PV deployment data, are analysed under varying levels of hydropower dominance. The analysis combines net-load modeling, ramping assessment, and a simplified flexibility-oriented mitigation approach to evaluate operational impacts under different hydropower conditions. The results indicate that increasing PV capacity significantly modifies the net-load profile. During summer periods, high solar generation substantially reduces midday net load, creating pronounced net-load valleys, whereas winter conditions remain more strongly influenced by electricity demand. As PV penetration increases, ramping intensity also increases. For example, extreme ramp values (Q99) rise from 80.87 MW/h at 300 MWp to 111.45 MW/h at 450 MWp, while the share of hours with ramp events exceeding 100 MW/h increases from 0.05% to 2.55%. The results of a conceptual flexibility approach that limits ramps to 60 MW/h show that extreme ramp events can be effectively mitigated, while moderate variability is largely unaffected. In summary, the results show that increasing solar PV penetration shifts the main operational challenge in Albania from energy balancing toward flexibility and variability management. The findings are particularly relevant for long-term system planning in hydropower-dominated systems and highlight the growing importance of flexibility measures and surplus management under high PV penetration. Full article

Renewable energy is a central component of global sustainable energy development, with solar energy experiencing substantial growth over recent decades. Solar power is widely regarded as one of the most accessible routes to clean energy generation. However, high upfront costs remain a major barrier to adoption. Many potential users are reluctant to invest in solar photovoltaic (PV) systems because of the longer payback period. To address this financial constraint, a range of business models has been developed. This study used a systematic literature review to examine existing and emerging business models for promoting Solar PV solutions. The review included peer-reviewed journal articles published in English from 2020 to 2026. In total, 39 articles were critically evaluated considering their characteristics. Nine potential business models were identified, several of which are commonly used internationally and have shown positive results that could also be applied in the UK. Importantly, Community Energy Models have shown success in Europe, Sub-Saharan and Asian regions. This has been widely supported by the government due to sustainability and climate change targets. The UK has set their target to achieve net-zero in greenhouse gas emissions by 2050. Beyond financial barriers, reliance on weather conditions and the mismatch between energy demand and supply remain substantial barriers to wider solar PV deployment. Full article

Hydropower-dependent electricity systems, such as Ecuador’s, face critical supply disruptions during droughts: a vulnerability exemplified by the 2024 power outages. This study assesses the technical, economic and environmental feasibility of a 200.84 MWp grid-connected solar photovoltaic (PV) plant proposed for the Pacific Refinery site in Manabi, Ecuador, as a strategy to diversify the energy matrix and reduce hydrological risk. Using site-specific solar resource data (4.65 kWh/m²/day) and PVSyst simulations, the plant achieves an annual energy production of 295 GWh with a performance ratio (PR) of 85.3%. A discounted cash flow analysis over 25 years, assuming a 7% discount rate and an electricity price of 60 USD/MWh, yields a net present value (NPV) of 104.9 MUSD, an internal rate of return (IRR) of 62.2%, and a levelized cost of energy (LCOE) of 14.5 USD/MWh, well below current industrial tariffs in Ecuador. Sensitivity analysis confirms project viability under ±15% variations in investment cost, energy price, and solar resource. Over its lifetime, the plant avoids 1.83 Mt of CO₂ emissions, supporting national decarbonization goals. The results demonstrate that large-scale PV deployment in high-radiation, low-latitude regions can be highly profitable and contribute to energy sovereignty in hydropower-dependent systems. Furthermore, this study provides a replicable model for repurposing unused industrial land for renewable energy generation, offering actionable insights for policymakers and investors in developing economies. Full article

The increased deployment of photovoltaic (PV) technology has led to an increased demand for grid-balancing capacity owing to growing short-term variability and forecast uncertainty. Simultaneously, higher PV penetration can lead to daytime energy market oversupply, pushing day-ahead prices toward zero and undermining PV revenues. Against this backdrop, this study investigated a market participation paradigm in which PV power plants supply reserve power themselves while actively absorbing their own uncertainty, rather than merely relying on balancing the services provided by external resources. We propose a risk-aware framework that classifies solar irradiance prediction errors into four risk categories using GPV-GSM numerical weather forecast data, translating the inferred risk level into practical bidding rules for balancing market participation. We adopted a hierarchical classification pipeline consisting of sign determination (stage 1, under- vs. overprediction), followed by degree determination (Stages 2 and 3), implemented with a multi-layer perceptron. To enhance class separability and reduce features, we introduced a stage-wise area under the curve (AUC)-based feature selection and compared AUC-selected and all-features settings under identical training conditions. The proposed strategies substantially reduce shortage events compared with directly using the original predictions as bids, although they increase surplus energy. The AUC-based model achieves comparable imbalance evaluation results, indicating that the selected features are sufficient for practical bidding support. Full article

Grid-connected photovoltaic–energy storage systems (PV-ESSs) enhance electricity reliability and lower energy costs in emerging markets. However, their commercial viability under multi-objective optimization remains under-quantified. This study offers a techno-economic and financial analysis of PV-ESS setups optimized with the Non-Dominated Sorting Genetic Algorithm II across Nigeria, South Africa, and India. The best systems feature 1.3–1.5 MW of solar capacity and 2.5–2.9 MWh of lithium-ion batteries. Results show unsubsidized levelized energy costs of USD 0.061–USD 0.064/kWh, achieving 27–35% savings compared to grid tariffs. Battery storage accounts for 67–76% of total capital costs, making battery expenses the key economic factor. Financial analysis reports net present values of USD 238,000–USD 522,000, internal rates of return of 13.7–15.8%, and discounted payback periods of 7.9–9.2 years. Monte Carlo simulations indicate an 83.4–100% probability of a positive net present value. Sensitivity analysis highlights grid tariffs and battery costs as major influences. Revenue diversification through grid services, capacity credits, and demand response can boost net present value by up to 35%. Overall, optimized PV-ESS projects can be commercially viable in emerging markets with suitable tariffs, financing, and revenue strategies. Full article