Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Abderrahmane Mendyl has a total of 22 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A Novel Application of Choquet Integral for Multi-Model Fusion in Urban PM10 Forecasting

    by Houria Bouzghiba, Amine Ajdour, Najiya Omar, Abderrahmane Mendyl, Gábor Géczi

    Air pollution forecasting remains a critical challenge for urban public health management, with traditional approaches struggling to balance accuracy and interpretability. This study introduces a novel PM10 forecasting framework combining physics-informed feature engineering with interpretable ensemble fusion using the Choquet integral, the first application of this non-linear aggregation operator for air quality forecasting. Using hourly data from 11 monitoring stations in Budapest (2021–2023), we developed four specialized feature sets capturing distinct atmospheric processes: short-term dynamics, long-term patterns, meteorological drivers, and anomaly detection. We evaluated machine learning models including Random Forest variants (RF), Gradient Boosting (GBR), Support Vector Regression (SVR), K-Nearest Neighbors (KNN), and Long Short-Term Memory (LSTM) architectures across six identified pollution regimes. Results revealed the critical importance of feature engineering over architectural complexity. While sophisticated models failed when trained on raw data, the KNN model with 5-dimensional anomaly features achieved exceptional performance, representing an 86.7% improvement over direct meteorological input models. Regime-specific modeling proved essential, with GBR-Regime outperforming GBR-Stable by a remarkable effect size. For ensemble fusion, we compared the novel Choquet integral approach against conventional methods (mean, median, Bayesian Model Averaging, stacking). The Choquet integral achieved near-equivalent performance to state-of-the-art stacking while providing complete mathematical interpretability through interaction coefficients. Analysis revealed predominantly redundant interactions among models, demonstrating that sophisticated fusion must prevent information over-counting rather than merely combining predictions. Station-specific interaction patterns showed selective synergy exploitation at complex urban locations while maintaining redundancy management at simpler sites. This work establishes that combining domain-informed feature engineering with interpretable Choquet integral aggregation can match black-box ensemble performance while maintaining the transparency essential for operational deployment and regulatory compliance in air quality management systems.


  • Open Access

    Evaluating Two-Dimensional Horizontal Grids in a Deterministic Air Pollution Model: Estimating the Impact on Outputs, Inputs, and ANN-CHIMERE

    by Houria Bouzghiba, Amine Ajdour, Mendyl Abderrahmane, Gábor Géczi

    Air pollution, especially from ozone and particulate matter, significantly contributes to environmental and health problems. Monitoring complexity and high costs have made modeling a parallel approach to surveillance. Models such as CHIMERE are essential in combining weather conditions, emissions, boundaries, and various physical processes, including transport and spatial resolution. This study introduces three methodologies to assess the influence of a two-dimensional horizontal grid in the Eulerian atmospheric transport model. It explores the impact on outputs, inputs, and model fitting. The first approach analyzes the resolution effect on CHIMERE outputs, focusing on O3. The second approach investigates various effects on inputs, including temperature (T), wind speed (WS), Planetary Boundary Layer Height (PBLH), Land Use and Land Cover (LULC), and Emissions (E). The third approach created a straightforward, strong, and precise ANN-CHIMERE adjustment model to evaluate the impact of spatial resolution in this context. The outcomes are verified through ozone measurement data collected in Agadir and Casablanca, Morocco, during various periods in 2016 and 2021.The results show that a higher horizontal grid improves the probability of good predictions. The impact of input data accuracy on high resolution can be concluded, providing insights into the limitations of CHIMERE in predicting output data. The new ANN-CHIMERE adjustment model delivers superior outcomes at high resolution, exhibiting an enhanced correlation coefficient and a significant reduction in the RMSE. In the future, the proposed approaches can be applied for spatial resolution optimization, maintaining the accuracy of the results and the computation time. It can also be adopted as an adjustment process of the inputs and outputs of deterministic air pollution models.

  • Open Access

    Energy, Economic and Environmental (3E) Analysis for an Optimal CSP Technology Integration in Morocco

    by Nabil Ammari, Ahmed Alami Merrouni, Abderrahmane Mendyl, Elmiloud Chaabelasri, Tamás Weidinger

    Among the existing solar technologies, Concentrating Solar Power (CSP) stands out as the most efficient and adaptable option for base load applications, primarily due to its thermal storage capabilities. However, despite its potential, the implementation of this technology still lacks competitiveness compared to Photovoltaic (PV) systems. Therefore, optimizing the plant components and operational factors becomes crucial for its cost-effective utilization, particularly in the desert regions of Morocco. Hence, the objective of this study comprised two main aspects: first, to conduct a parametric analysis aimed at selecting the optimal configuration for a parabolic trough collector (PTC)-based power plant suitable for the Moroccan context. Subsequently, an environmental analysis was performed to assess the impact of soiling on the plant operation. This step aimed to refine the precision of the techno-economic analysis and enhance the project’s bankability. High-quality in situ meteorological data and soiling measurements were utilized for these analyses. Furthermore, to ensure the reliability of the results, the results from the employed simulation tool were validated against real data obtained from an operational power plant. The results indicate that Morocco holds significant potential for the integration of large-scale CSP plants. A capacity of 1 MW utilizing PTC technology could yield an annual electricity production of up to 33 GWhe, with a levelized cost of electricity (LCOE) estimated at 0.1465 EUR/kWh. However, accounting for soiling effects in the yield analysis, which is recommended for precise yield calculations, revealed a decrease in the annual production to 28 GWhe for the same 1 MW capacity. This reduction represented a 20% loss from the nominal conditions, resulting in a corresponding increase in electricity cost by 30.6 €/MWh.


  • Open Access

    Short‐term predictions of PM10 and NO2 concentrations in urban environments based on ARIMA search grid modeling

    by Bouzghiba Houria, Mendyl Abderrahmane, Khomsi Kenza, Géczi Gábor

    Air pollution poses a persistent challenge for urban management departments and policymakers due to its significant health and economic impacts. Various cities worldwide have implemented diverse strategies and initiatives to enhance air quality monitoring and modeling standards. However, the outcomes of these efforts often manifest over the long term, leading to a preference for short‐term statistical methods. The autoregressive integrated moving average (ARIMA) search grid modeling approach has gained widespread use for forecasting air quality. This paper presents a comprehensive time series analysis conducted to predict air quality in urban areas of Budapest, Hungary, with a focus on nitrogen dioxide (NO2) and particulate matter (PM10), using air quality data spanning from 2018 to 2022 for four monitoring categories: Urban traffic, industrial background, urban background, and suburban background. The study employs the ARIMA search grid method to forecast concentrations of these pollutants at multiple air quality monitoring stations based on Akaike information criteria (AIC) and the Bayesian information criteria (BIC) criteria along with the results of augmented Dickey–Fuller (ADF) test. The results demonstrate varying levels of forecast accuracy across different stations, indicating the model's effectiveness in short‐term predicting of air quality. These findings are essential for assessing the reliability of air quality forecasts in Budapest and can inform decisions regarding air quality management and the development of strategies to address air pollution and particulate matter concerns in the region.


  • Open Access

    Enhancing Solar Radiation Forecasting in Diverse Moroccan Climate Zones: A Comparative Study of Machine Learning Models with Sugeno Integral Aggregation

    by Abderrahmane Mendyl, Vahdettin Demir, Najiya Omar, Osman Orhan, Tamás Weidinger

    Hourly solar radiation (SR) forecasting is a vital stage in the efficient deployment of solar energy management systems. Single and hybrid machine learning (ML) models have been predominantly applied for precise hourly SR predictions based on the pattern recognition of historical heterogeneous weather data. However, the integration of ML models has not been fully investigated in terms of overcoming irregularities in weather data that may degrade the forecasting accuracy. This study investigated a strategy that highlights interactions that may exist between aggregated prediction values. In the first investigation stage, a comparative analysis was conducted utilizing three different ML models including support vector machine (SVM) regression, long short-term memory (LSTM), and multilayer artificial neural networks (MLANN) to provide insights into their relative strengths and weaknesses for SR forecasting. The comparison showed the proposed LSTM model had the greatest contribution to the overall prediction of six different SR profiles from numerous sites in Morocco. To validate the stability of the proposed LSTM, Taylor diagrams, violin plots, and Kruskal–Wallis (KW) tests were also utilized to determine the robustness of the model’s performance. Secondly, the analysis found coupling the models outputs with aggregation techniques can significantly improve the forecasting accuracy. Accordingly, a novel aggerated model that integrates the forecasting outputs of LSTM, SVM, MLANN with Sugeno λ-measure and Sugeno integral named (SLSM) was proposed. The proposed SLSM provides spatially and temporary interactions of information that are characterized by uncertainty, emphasizing the importance of the aggregation function in mitigating irregularities associated with SR data and achieving an hourly time scale forecasting accuracy with improvement of 11.7 W/m2.


  • Open Access

    Thermal performance assessment of an indirect solar dryer: A case study of Bananas

    by Abderrahmane Mendyl, Houria Bouzghiba, Rachid Tadili, Tamás Weidinger

    This study presents a design for an absorber used in a solar air collector for an indirect solar dryer. The absorber comprises two aluminium plates corrugated and joined together to form parallel cylinders, enabling airflow within the collector. This research aims to experimentally examine the drying process of two types of bananas, one from Morocco and the other from abroad, using the designed solar air collector. Additionally, the study aims to investigate the peculiarities of the drying process and the performance of the solar dryer employed. The experiments were conducted by subjecting the bananas to the designed solar air collector, and the evolution of drying was monitored. The initial mass of the bananas used was 631.6 g for the Moroccan banana and 713.6 g for the Export banana. After the drying process, the mass of the Moroccan banana reduced to 77.5 g, while the Export banana reduced to 137.3 g, indicating significant moisture removal. The percentage of the amount of water extracted (Q) from the bananas was found to be 87.7% for the Moroccan banana and 80.8% for the Export banana. These results demonstrate the effectiveness of the corrugated aluminium plate absorber in facilitating the drying process in the solar air collector. The significant reduction in the mass of the bananas and the high percentage of water extraction highlight the efficiency of the solar dryer in removing moisture from the agricultural produce. The findings of this study contribute to the understanding of the drying process of bananas and offer valuable insights for the design and optimization of solar drying systems for agricultural applications.


  • Open Access

    Analysis of Wind Data and Assessment of Wind Energy Potential in North-West Hungary

    by Abderrahmane Mendyl, Peter Peter K. Musyimi, Gyöngyösi Adrás Zénó, Tamás Weidinger

    Recently, the world is facing an increasing use of renewable sources of energy that will require a clear understanding of wind energy as an affordable source for sustainability. North-West part of Transdanubia is the windiest region of Hungary. Energy production from two Energon type wind generators are investigated in a wind farm near Mosonmagyaróvár, 10 km from the Austrian border. Wind speed, direction and energy production were recorded with 10 minutes time resolution for a period of 6 years (2010-2015). The hourly wind speed and direction dataset from ERA5 are also investigated. The collection of historical wind data serves the same objective as any other statistical information to improve weather forecasting by knowing what the wind was like on a specific period of time at a specific location and identifying a wind pattern allows you to compare this information to the prediction. This was done using E40 wind-turbine on 65 m height (Max power 600 kW), with mean wind speed of 5.26 ±2.8 m/s and an average power of 103 kW and E70 wind-turbine on 113 m height (Max power 2000 kW), measured wind speed of 5.87±3.12 m/s with an average power of 457 kW. The study adopted Weibull distribution to model wind speeds and applied Power density method to estimate the scale and shape parameters. This is a relatively new, simple formulation and requires less computation. The values of k and c factors were retrieved as functions of the mean wind speed, wind sectors and energy pattern factor respectively.Wind energy production forecasts were provided by WRF model system for 48 hours with site specific model output statistics. The model was run twice a day (00 and 12 UTC) with 10 km horizontal resolution and 28 vertical level based on GFS model outputs. Uncertainties of energy forecasts were also investigated. In most cases, the model underestimated the wind speed. Further application of the developed model system will be carried out in semi-arid region of Morocco.

  • Open Access

    Hourly reference evapotranspiration analysis using synoptic meteorological measurements and ERA5 reanalysis data from Kenyan Counties

    by Peter K. Musyimi, Abderrahmane Mendyl, Agustiyara Agustiyara, Balázs Székely, Tamás Weidinger

    Kenya has had five failed rain seasons for the last three years. In this context, there was a mass recurrent crop failure, death of livestock and wildlife, persistent water scarcity, and droughts of varying intensities. There have been a lot of challenges in assessing climate change and variability impacts in Kenya due to limited data sources. Further, assessing the local and regional effects on the hydrological cycle, food security, and available water resources remains a great regional threat. Reference evapotranspiration, is the evaporative power climatic parameter of the atmosphere, vital for water budgets on the land surface. The study’s main goal was to analyze hourly reference evapotranspiration, from two climatic regions using single levels ERA5 hourly dataset from 2000 to 2022. The dataset was sought from three stations from, arid, and semi-arid savannah tropical conditions regions (Voi Garissa, and Mombasa) with elevations between 57 m to 579 m, and three (Trans-Nzoia, Nyeri, and Embu) sought from humid Kenya highlands (>1350 m). Reference Evapotranspiration was calculated using Penman-Monteith (FAO56), the standard methodology developed by Food and Agriculture Organization. Results from 5 years (2018 to 2022) in Taita-Taveta County indicated that ranged from 0.17±0.2 mm/hour in 2020 to 0.22±0.2 mm/hour in 2022. Daily averages were 4.17±1.2 mm./day to 5.2±1.1 mm/day in 2020 and 2022 respectively. The mean monthly and was highest in March with an estimated value of 159.7±53.7 mm/month while the lowest was 120±15 mm/month in December. This is because March falls at the onset of the long rainy season in Kenya where precipitation is high while December is the last month of the short rainy season when precipitation reduces significantly. These results are vital because they enhance comparisons of the spatial climatological patterns and variability of seasonal precipitation about the evaporative power and demand variation across regions. Further, it will necessitate investigations of uncertainties from the datasets for better decision-making after comparisons with analysis from field meteorological datasets and soil moisture data measurements currently being carried out in Kenya. Further comparison of the results with reference evapotranspiration from the original station and the Global Land Evaporation Amsterdam Model dataset will also be investigated.

  • Open Access

    Calibration and Validation of Global Horizontal Irradiance Clear Sky Models against McClear Clear Sky Model in Morocco

    by Abderrahmane Mendyl, Brighton Mabasa, Houria Bouzghiba, Tamás Weidinger

    This study calibrated and compared the capabilities of hourly global horizontal irradiance (GHI) clear sky models for six Moroccan locations, using the McClear clear sky model as a reference. Complex clear sky models, namely Bird, Simplified Solis, Ineichen and Perez, and simple clear sky models, namely Adnot–Bourges–Campana–Gicquel (ABCG), Berger–Duffie, and Haurwitz were tested. The SOLCAST satellite-based dataset estimates were validated against the McClear clear sky model. pvlib python was used to configure the models, and ERA5 hourly fractional cloud cover was used to identify clear-sky days. The study period was from 2014 to 2021, and the study sites were in different climatic regions in Morocco. Bar graphs, tables, and quantitative statistical metrics, namely relative mean bias error (rMBE), relative root mean square error (rRMSE), relative mean absolute error (rMAE), and the coefficient of determination (R2), were used to quantify the skill of the clear sky model at different sites. The overall rMBE was negative in 5/6 sites, indicating consistent overestimation of GHI, and positive in Tantan (14.4%), indicating frequent underestimation of GHI. The overall rRMSE varied from 6 to 22%, suggesting strong agreement between clear sky models and the McClear clear sky model. The overall correlation was greater than 0.96, indicating a very strong relationship. Overall, the Bird clear sky model proved to be the most feasible. Complex clear sky models outperformed simple clear sky models. The SOLCAST satellite-based dataset and ERA5 cloud fraction information could well be used with quantifiable certainty as an accurate clear sky model in the study region and in other areas where complex clear sky models’ inputs are not available.