This study presents a multimodal framework based on machine learning (ML) and explainable AI (XAI) techniques for the compositional analysis of waste biomass towards an efficient waste-to-energy (WTE) system. The framework encompasses principal component analysis (PCA) for explaining the variance within the biomass-dataset, SHapley Additive exPlanations (SHAP) for interpretable ML prediction-outcomes, k-means cluster-analysis of the waste-biomass compositional profile to reveal distinct-group within the dataset, k-nearest neighbour (k-NN) for energy-based classification, and different ML models for predicting the higher heating value (HHV). SHAP identified Carbon (C) and Hydrogen (H) as the most influential positive-predictors of HHV, while moisture-content (MC) and ash had the strongest negative-impacts. PCA captured about 78% of the variance of the dataset in 3 PCs, while k-means clustering revealed 3 distinct biomass-groups, with cluster-3 mostly suitable for direct-combustion. k-NN classifier achieved peak performance at k = 3 for classifying biomass by energy suitability, with 95.1% training accuracy. Particle swarm optimization (PSO)-optimized Artificial neural network (ANN) predicted HHV more accurately than all other models, with RMSE, MAE, and MAPE values of 3.8394, 1.5669, and 7.8502, during testing. This framework provides a scalable and interpretable tool which enables data-driven WTE planning, supports more efficient feedstock selection, optimize thermal conversion.
Hydrogen-integrated renewable microgrids represent a promising pathway for sustainable decarbonisation of remote communities; however, conventional techno-economic optimisation often overlooks the operational behaviour of hydrogen subsystems, leading to optimistic performance estimates and less practically deployable designs. This study proposes an integrated framework combining Hybrid Optimisation of Multiple Energy Resources (HOMER) Pro-based techno-economic optimisation with Python-based post-optimisation operational assessment. The framework evaluates operational indicators, including minimum-load compliance, start-stop cycling, large-ramp events, and hydrogen storage utilisation, using hourly dispatch data and literature-informed criteria. A scenario-based design-space exploration is conducted for a remote off-grid community in Australia, followed by operational feasibility screening and Pareto-based multi-objective analysis balancing economic performance, renewable energy utilisation, and electrolyser operation. The optimised reference configuration achieves zero operational CO2 emissions with a levelised cost of energy (LCOE) of 0.258 AUD/kWh and a levelised cost of hydrogen (LCOH) of 6.77 AUD/kg. Post-optimisation assessment reveals that 5,912 h of reported electrolyser operation reduces to 4,775 h of acceptable operation; the excluded 1,137 h (19.2%) correspond to low-load periods, indicating an overestimation of effective electrolyser utilisation of up to 23.8%. Threshold sensitivity analysis identifies hydrogen tank near-full hours as the most sensitive indicator, while core screening outcomes remain stable under low-load and ramp-event threshold variations. Operational feasibility screening reduces the 22 candidate designs to 15 feasible configurations, demonstrating that techno-economic attractiveness does not necessarily ensure operational viability. The proposed framework provides a practical methodology for operational assessment and design selection of hydrogen-integrated renewable microgrids, supporting informed planning and deployment of reliable, low-carbon energy systems.
The rapid deployment of solar photovoltaic (PV) and battery energy storage systems (BESS) in distribution networks offers significant technical and economic benefits, yet their optimal long-term planning under uncertainty remains a complex challenge. This paper proposes a multi-objective optimal planning framework for the simultaneous integration of solar PV and BESS in distribution systems, evaluated through a multi-horizon comparative analysis across short-term (1-year), mid-term (5-year), and long-term (10-year) planning horizons. Uncertainties in load demand and solar irradiance are modelled using Monte Carlo Simulation with scenario reduction via the Backward Reduction Algorithm. The multi-objective optimization minimizes power losses, voltage deviation, and total system costs including operational, planning, and carbon emission components, with TOPSIS applied for optimal solution selection. The framework is validated on the IEEE 33-bus test system over multiple planning horizons. Results demonstrate that long-term planning consistently outperforms shorter horizons, achieving power loss reductions of 55.19%, voltage deviation improvements of 43.92%, and total cost savings of 15.08% compared to the base case. Furthermore, DER penetration progressively increases from short-term to long-term planning, with an 11.11% improvement observed from Case II to Case IV. This study establishes planning horizon length as a critical determinant of techno-economic performance, providing distribution planners with a structured decision-support framework for DER investment under real-world uncertainty.
Bottom-up modelling approaches of national energy systems with high fractions of variable renewable energy (VRE) are becoming more significant due to factors such as environmental impact, energy security, and cost efficiency towards future years. This modelling approach is broadly applied in energy planning and it enables the simulation of energy systems, including an evaluation of the energy balances, total CO2 emitted, and total costs. In this paper, the development of MOO scenarios for Malaysia towards 2050 based on the 1.5-S RE90 and RE100 targets from the International Renewable Energy Agency (IRENA) Malaysia Energy Transition Outlook is conducted. Multi-objective optimisation (MOO) is applied using EnergyPLAN and EPLANopt to obtain the approximated best trade-off low cost and low emission solution for Malaysia towards 2050. Analysis is carried out in setting the decision variables to be the capacities of energy sources and parameters of lithium-ion (Li-ion) battery energy storage systems (BESS). The outcome of this paper will be obtaining the approximate Pareto optimal solutions with the objective functions set to minimise the total CO2 emissions and total annual costs. The approximated best trade-off of the near-optimal Pareto solutions between the two objective functions was selected to be evaluated in terms of the high temporal resolution electricity balance results, costs, and CO2 emissions. It was found that rooftop solar had a significant role in the variation of Pareto optimal solutions when moving from the high emission low cost scenario to the low emission high cost scenario. The findings from this paper can offer insights to policymakers in analysing the optimal capacity ranges of energy sources and BESS required by Malaysia’s power sector towards 2050 based on the IRENA 1.5-S targets.
High-renewable power systems require planning methodologies that integrate long-term investment optimisation with resource adequacy, storage sizing, and chronological operational validation. This paper presents a novel scalable multi-stage techno-economic planning framework implemented in PLEXOS, linking capacity expansion (CE), resource adequacy assessment (RAA), storage volume optimisation (SVO), and chronological production cost modelling (PCM). The framework is demonstrated across four power systems representing benchmark, wind-dominated, solar-dominated, interconnected, and high-demand-growth renewable transition pathways. Capacity expansion alone produced annual Loss of Load Probability (LOLP) values of up to 50, 49, 245 and 376 h/year across the four case studies. Following the RAA stage, LOLP was reduced to zero in all systems. Chronological production cost simulations confirmed that all principal planning pathways satisfied the adopted 0.05% unserved-energy criterion. For the representative GCC-type system, mean unserved energy remained at 0.015% under fuel-blend operation but increased to 0.183% under hydrogen-only operation. The supporting flexibility portfolio varied substantially across renewable transition pathways, requiring approximately 24–40 MW of 10-h battery energy storage and 12–56 MW of 4-h battery energy storage per 100 MW of installed renewable capacity. Across the validated transition pathways, the resulting levelised cost of energy (LCOE) ranged from 7.97 to 32.48 $/MWh, reflecting differences in renewable-resource mix, storage and flexibility requirements, retained conventional capacity, and interconnection support. The results demonstrate that high-renewable power systems do not converge towards a universal storage or flexibility solution. Instead, storage and balancing requirements are system-specific, with solar-dominated systems requiring substantially greater short- and long-duration storage deployment, while interconnected wind-dominated systems rely more heavily on system flexibility and cross-border balancing. The proposed framework provides a scalable methodology for identifying system-specific storage and flexibility requirements for long-term renewable transition planning.
Renewable energy source (RES)-driven displacement of synchronous generators reduces system inertia, reactive power support capability, and short-circuit current contribution, thereby challenging the secure and stable operation of future RES-dominated power systems. Synchronous condensers (SCs) have emerged as one of the most relevant, well-understood, and effective solutions to mitigate these issues. However, the relatively high investment cost and longer deployment time of SCs necessitate detailed studies for determining their optimal number, size, and location. The paper proposes a comprehensive methodology for optimal SC allocation by simultaneously considering inertia support, reactive power capability, and short-circuit current contribution. The allocation problem minimizes installation, operation, and maintenance costs while ensuring adequate frequency stability, voltage performance, and system strength. Unlike conventional approaches that primarily rely on steady-state short-circuit ratio indices, the proposed methodology incorporates detailed system dynamic behaviour and area-wise inertia requirements into the allocation process, enabling a more comprehensive assessment of SC services. Furthermore, comparative studies of new SC installations and repurposing of retired synchronous generators are carried out to evaluate trade-offs in economic and technical performance. Validation studies are carried out using a modified IEEE 39-bus system and a real-life model of the Gujarat state grid in India with high shares of RES. Results indicated that the proposed methodology could identify cost-effective SC allocations while maintaining the required dynamic performance and enhancing overall system resilience. While repurposed generators provide a lower-cost solution, new SC installations offer enhanced frequency response, voltage stability, and system strength. A further analysis also demonstrates the benefits of optimized SC allocation relative to PV-based emulated inertia support.
Carbon financing policies such as emission trading schemes have been extensively utilized to assist in global emission reduction efforts. However, effectively tackling the substantial carbon emissions associated with prosumers within the electricity sector continues to pose a critical challenge. Distribution System Operators (DSOs) try to alleviate network congestion by harnessing a range of flexibility resources, with prosumers emerging as vital contributors within distribution networks. Prosumers possess the unique capability to adjust their load profiles, thereby offering valuable flexibility through active participation in Demand Response (DR) programs. This paper introduces an innovative market-based framework for carbon-responsive prosumers, aimed at amplifying prosumers’ engagement in incentive-based DR programs. Within this framework, an aggregator adopts a forward-thinking approach, treating the incentive signal as supplementary carbon emission capacity for prosumers willing to contribute flexibility. In order to improve prosumers’ motivation, their participation in a local Day-Ahead (DA) carbon emission market is guaranteed, which enhances their engagement in DR programs. Additionally, a novel bidding strategy is introduced to ensure fairness and competitiveness, which establishes a market mechanism that prioritizes prosumers’ benefits. This framework presents two crucial advantages. Firstly, it renders flexibility provision financially attractive to prosumers, thereby improving the effectiveness and efficiency of congestion management. Secondly, it facilitates the mitigation of prosumers’ carbon emissions. Through the case study, it is demonstrated that the proposed model effectively addresses flexibility requirements while establishing carbon emission trading policies as incentives, which provides prosumers with additional revenue streams.
Optimal Power Flow (OPF) is an important optimization task in power system. It remains difficult to solve efficiently because of strong nonlinearity, nonconvexity, and multiple constraints. Although numerous algorithms have been developed to solve it, they still are faced with some limitations such as early stagnation, limited exploration capabilities and low convergence rates. To overcome these drawbacks, this study introduces an adaptive multi-operator differential evolution algorithm based on binary tree population (AMBTDE) for efficiently resolving OPF with stochastic wind and solar power. Building upon the binary tree population framework, the proposed algorithm introduces a multi-operator pool mechanism and an adaptive operator selection method to dynamically adjust operator usage probabilities. This enables the algorithm to flexibly use proper search strategies across different evolutionary stages to balance global exploration and local exploitation. Results on the IEEE 30-bus, 57-bus, and large-scale 118-bus test systems indicate that AMBTDE significantly outperforms mainstream differential evolution variants. Crucially, the integration of stochastic wind and solar power in the IEEE 30-bus and 57-bus systems demonstrates substantial environmental and technical benefits. Specifically, AMBTDE highlights a strong potential for environmental sustainability, achieving a reduction in harmful greenhouse gas emissions of up to 3.8%. Active power losses and voltage deviations are also improved by up to 29.13% and 38.016%, respectively. Furthermore, validation on the large-scale IEEE 118-bus system confirms the algorithm’s exceptional scalability and economic efficiency, achieving a remarkable reduction of 9.505% in generation cost even in highly complex scenario. These quantitative findings confirm the effectiveness and robustness of AMBTDE, providing a highly capable tool for modern, renewable-integrated power system optimization.
Planning and managing the connection of isolated communities to the electrical grid remains a major challenge in improving energy access and quality of life in remote regions. This paper proposes an integrated EGAC–MILP framework for planning electrical interconnections and optimizing day-ahead energy dispatch in isolated communities. The methodology minimizes the levelized cost of energy (LCOE), reduces operating costs, increases renewable energy utilization, and considers the socioeconomic benefits of electricity access. The framework is applied to seven off-grid localities in Miraflores, Guaviare (Colombia), considering photovoltaic, small hydropower, battery, and diesel generation. The proposed approach achieved 100% demand coverage for approximately 1,073 users, supplying energy requirements from 0.1 to 2.6 MW/day and monthly energy coverage between 3 and 78 MWh. Compared with isolated operation, the optimal multi-community configuration reduced the daily operating cost from USD 9,497.74/day to USD 9,065.94/day while eliminating diesel generation under normal photovoltaic conditions, reducing daily CO2 emissions from 2.15 t/day (Case 1) and 1.73 t/day (Case 2) to 0 t/day (Case 3). The battery management strategy maintained the state of charge within an optimal operating range of approximately 34–50%, improving operational flexibility and renewable energy utilization. The resulting optimal configuration achieved LCOE values of 0.01 USD/kWh for photovoltaic generation and 0.041 USD/kWh for small hydropower, demonstrating the technical, economic, and environmental advantages of coordinated multi-community off-grid energy management.
The increasing demand for clean energy has accelerated photovoltaic (PV) deployment to reduce fossil-fuel dependence and greenhouse gas emissions. Grid-connected PV performance depends on climate, orientation, tracking, thermal effects, soiling, and electrical design. This study develops a multi-scenario optimization framework for a 131 kWp rooftop PV system using PVsyst. The analysis considers fixed-tilt optimization, seasonal and monthly tilt adjustment, unlimited sheds with Ground Coverage Ratio (GCR) optimization, one-axis tracking with backtracking, two-axis tracking, DC/AC ratio optimization, soiling losses, and thermal loss coefficients. An optimum fixed tilt of 22° produced 261.66 MWh annually. Seasonal adjustment increased generation to 271.65 MWh, a 3.8% improvement. Monthly optimization showed that the optimum tilt varies from 0° during May–July to 50° in December, increasing annual yield to 273.91 MWh and providing an upper-bound reference for manually adjustable systems. The optimized one-axis tracker, with a 6.0 m pitch and 50.3% GCR, offered the best balance between performance and practicality, delivering 305.51 MWh annually, a specific yield of 2341 kWh/kWp/year, and a performance ratio of 82.89%. Although two-axis tracking generated the highest annual energy, its structural complexity and space requirements limit rooftop suitability. The final configuration comprised a 131 kWp array, a 100 kW inverter, and a DC/AC ratio of 1.3. A 5% soiling loss and a thermal loss coefficient of 29 W/m2K were identified as suitable local design assumptions. Overall, combining tracking, electrical sizing, and environmental optimization improves rooftop PV energy yield and operational efficiency in desert climates and provides practical guidance for high-performance solar system design.
The rapid expansion of photovoltaics (PV) is intensifying competition for land, particularly in regions dominated by high-value crops such as vineyards. Agrivoltaics offers a potential solution by enabling dual land use for food and energy production, yet many existing configurations rely on elevated or structurally intensive systems that may affect crop performance, increase material demand, or disrupt culturally sensitive agricultural landscapes. This context calls for agrivoltaic designs that structurally prioritize cultivation while enabling energy generation with minimal agronomic impact. Here we assess, through field-scale experimentation, a low-height, trellis-integrated vertical agrivoltaic concept implemented in three pilot installations in southeastern Spain. We provide the first multi-site empirical evidence, based on a single 2024 harvest season, that these integrated PV systems can operate in vineyards without statistically significant differences in several yield and grape-composition indicators under the conditions tested, including °Brix, acidity and phenolic composition at harvest under the conditions tested. We demonstrate that meaningful energy production can be achieved under a cultivation-first design logic, resulting in Land Equivalent Ratios (LER) consistently above unity (1.16-1.53), thus confirming real land-use efficiency gains. We identify a clear potential for photovoltaic densification in low-trained vineyards, indicating that installed capacities comparable to ground-mounted PV systems could be reached without additional land occupation or agronomic interference (with LER values reaching 1.95). We show that the proposed structural solution achieves low material intensity (161-141 kg steel/kWp) while maintaining mechanical adequacy, positioning it at the lower boundary of reported agrivoltaic steel demand. We link technical performance with social acceptance, highlighting the suitability of low-height, trellis-aligned integration for viticultural landscapes with strong identity constraints.
Natural gas, being economical is widely used in Pakistan. However, the depleting reserves and rising LPG imports necessitates transitioning to energy efficient environment friendly technologies (SDG 9 & SDG 11). This study evaluates the impact of natural gas substitution on electricity demand, natural gas cost savings, HRES techno-economic feasibility, and its policy implications. The analysis indicated the total annual natural gas consumption of households to be 312,963-mmcft with Commercial consuming 21,114-mmcft. Results of the proposed study revealed that the total electricity demand of both the residential and commercial sector was projected to be 1,060.431-TWh for year 2070 with 100% substitution of natural gas and complete electrification of the households. Whereas, the environmental analysis demonstrated the cumulative CO2 reduction for the year 2070 due to substitution and HRES to be 883,855.7267-Gg of CO2/year. The economic analysis of substitution equipment indicated the payback period of 6.56 years for electric appliances in Households while 2.04 years for W&R and 0.194 years for H&R with combined natural gas savings potential of Rs. 412,073.86 million/year. HRES were deployed in Balochistan (Solar of 584.15 GW, Wind of 454.81 GW, Initial Capital of $1,474 billion, LCOE of $0.1767/kWh, ROI of 44% and PBP of 4.2 years), Sindh (Solar of 328.77 GW, Wind of 104.26 GW, Initial Capital of $797 billion, LCOE of $0.1922/kWh, ROI = 24% and PBP of 9.1 years) and KPK (Solar of 269.14 GW, Wind of 392.07 GW, Initial Capital of $766 billion, LCOE of $0.183/kWh, ROI of 24% and PBP of 4.6 years).
Renewable-fed islanded microgrids rely on cascaded photovoltaic (PV), battery energy storage system (BESS), DC-DC, and DC-AC conversion stages whose ability to maintain regulated load voltage is challenged by upstream outages, renewable shortfall, and compound power-quality disturbances. This paper presents a unified control framework for a dynamic voltage restorer (DVR) that enhances microgrid resilience by combining a synchronverter-based grid-forming reference generator, multi-resonant proportional-resonant controllers, and a supervisory mode-management layer. The proposed scheme coordinates standby, power-quality compensation, and voltage-interruption ride-through within a single architecture. In compensation mode, it mitigates sag/swell, unbalance, and harmonics at the critical-load bus. In interruption mode, it reconfigures the injection transformer and temporarily supplies the protected feeder from the local DC-link, thereby coupling renewable/storage-side energy conversion to AC-side continuity support. The method is evaluated in a MATLAB/Simulink model of a PV-BESS islanded microgrid with DC-DC converters, mixed grid-forming/grid-following inverters, and disturbance-producing loads, and is further validated through OPAL-RT/DSP hardware-in-the-loop experiments to demonstrate implementation feasibility. Under combined disturbances, load-voltage THD and VUF are reduced to 0.469% and 0.22%, respectively. During complete source interruption, restoration times of 14.2 ms for the instantaneous voltage and 32.2 ms for the RMS voltage are achieved with negligible overshoot. Under partial shading, causing a 38.9% PV-voltage drop and a 55.2% uncompensated DC-bus voltage drop, the protected load remains regulated at 1.0 p.u. with 0.0794% THD and 0.0265% VUF. The results show that the proposed controller extends conventional DVR functionality from power-quality compensation to resilienceoriented support in renewable-fed microgrids.
The rapid deployment of renewable energy in offshore and coastal regions increases the need for storage systems that can manage intermittency, spatial mismatch, and grid integration constraints. Coordinated operation across offshore, coastal, and land-based energy hubs remains insufficiently developed when mobile storage logistics, green hydrogen pathways, multi-carrier coupling, and market mechanisms are considered together. This study proposes a digital twin-enabled optimization framework for mobile and hybrid storage management in offshore and coastal multi-energy communities. The architecture integrates electrical, thermal, cooling, hydrogen, and water subsystems, while offshore energy transport vessels relocate mobile storage units and support dynamic energy exchange among distributed hubs. A mixed-integer optimization model determines storage scheduling, inter-hub exchange, desalination operation, grid interaction, and market participation. A Stackelberg-Bayesian structure captures hierarchical interactions and information asymmetry among hubs, while a Wasserstein distance-based robust optimization framework handles uncertainty in renewable generation, demand, and market prices. Carbon emission trading and green certificate trading are embedded as market-based decarbonization incentives. Numerical results show that coordinated mobile storage operation improves renewable utilization and flexibility, eliminates curtailed electrical loads, and reduces operating costs by up to 21% compared with non-coordinated operation. Under environmental constraints, groundwater extraction is fully eliminated, with the cost increase partly offset by carbon and green certificate revenues. Coordinated operation increases carbon trading revenues by approximately 27% and green certificate revenues by up to 54%. These findings show that digital twin Stackelberg-Bayesian coordination of mobile storage can strengthen uncertainty-resilient renewable integration in offshore and coastal green hydrogen energy hubs.
Inclusion of distributed energy resources (DER) into the power system grid may bring numerous advantages, but at the same time due to the intermittent nature of DER’s, power systems operators have to deal with various challenges to maintain a balance between generation and demand. Need for coordinated and collaborative efforts between operators at transmission and distribution level is becoming increasingly vital due to more deployment of DER’s in the power systems. Transmission system operators (TSO) manage the high voltage transmission network and distribution system operators (DSO) are responsible for handling the low voltage distribution network. This paper provides an overview of important topics of study in a coordinated TSO-DSO system, grouped into a hierarchical taxonomy comprising three levels, namely, (i) operational coordination dealing with ancillary services, flexibility improvement, optimal power flow, and congestion control (ii) planning coordination including expansion planning and integration of DER’s and (iii) enabling infrastructure via information and communication technology (ICT), interoperability, and security. Article also expounds the coordination objectives, challenges, and different optimization techniques to stretch additional advancement for more resilient, authentic, and sustainable power system in the future.
This study highlights the significance of coupling solar, biomass, geothermal, and wind energy power sources with a proton exchange membrane fuel cell (PEMFC) for green and reliable grid-connected power production. By integrating these four renewable power sources, a proton exchange membrane electrolyzer (PEME) efficiently generates hydrogen, which is then stored and later utilized in a PEMFC, ensuring a continuous and reliable supply of clean energy. This approach not only maximizes the use of diverse renewable resources but also enhances energy security and reduces greenhouse gas emissions, contributing to a sustainable energy future. In the present research, a 4E study is done on the proposed configurations. In considered layouts, the four different types of renewable power produce green hydrogen in a PEME, and the hydrogen is injected into a PEMFC for power generation. Furthermore, the waste energy of the PEMFC is recovered by a bottoming organic Rankine cycle (ORC). Results demonstrates that in the optimal output performance mode, the biomass-based system achieves the highest exergy efficiency (6.25%), while the geothermal-based system achieves the lowest values for total cost rate and output unit cost (25.89 $/h and 45.77 $/GJ, respectively). In the optimal case of the geothermal-based system, the topping system produces 3.167 kg/h of green hydrogen with a unit cost of 29.98 $/GJ (4.25 $/kg). The produced hydrogen is then supplied to the bottoming PEMFC–ORC system, which generates 49.2 kW of electrical power. In this configuration, the unit cost of the produced electricity is calculated as 45.77 $/GJ.
