Assia Mahrouch

Mrs. Assia Mahrouch

orcid
Ph.D. Student

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
28 June 2024 in IEEE Access
Verified
Journal Article
Open access
Rafiq AsgharRafiq Asghar
Rafiq Asghar
FFFrancesco Riganti Fulginei
Francesco Riganti Fulginei
Michele QuercioMichele Quercio
Michele Quercio
AMAssia Mahrouch
Assia Mahrouch
Solar energy is largely dependent on weather conditions, resulting in unpredictable, fluctuating, and unstable photovoltaic (PV) power outputs. Thus, accurate PV power forecasts are increasingly crucial for managing and controlling integrated energy systems. Over the years, advanced artificial neural network (ANN) models have been proposed to increase the accuracy of PV power forecasts for various geographical regions. Hence, this paper provides a state-of-the-art review of the five most popular and advanced ANN models for PV power forecasting. These include multilayer perceptron (MLP), recurrent neural network (RNN), long short-term memory (LSTM), gated recurrent unit (GRU), and convolutional neural network (CNN). First, the internal structure and operation of these models are studied. It is then followed by a brief discussion of the main factors affecting their forecasting accuracy, including forecasting horizons, meteorological conditions, and evaluation metrics. Next, an in-depth and separate analysis of standalone and hybrid models is provided. It has been determined that bidirectional GRU and LSTM, whether used as a standalone model or in a hybrid configuration, offer greater forecasting accuracy. Furthermore, hybrid and upgraded metaheuristic algorithms have demonstrated exceptional performance when applied to standalone and hybrid ANN models. Finally, this study discusses various limitations and shortcomings that may influence the practical implementation of PV power forecasting.

26 January 2022 in Technology and Economics of Smart Grids and Sustainable Energy
Verified
Journal Article
Closed access
PDF file
AMAssia Mahrouch
Assia Mahrouch
Mohammed OuassaidMohammed Ouassaid
Mohammed Ouassaid

28 March 2025 in IEEE Access
Verified
Journal Article
Open access
Rafiq AsgharRafiq Asghar
Rafiq Asghar
Michele QuercioMichele Quercio
Michele Quercio
LSLorenzo Sabino
Lorenzo Sabino
AMAssia Mahrouch
Assia Mahrouch
FFFrancesco Riganti Fulginei
Francesco Riganti Fulginei
Photovoltaic (PV) power forecasting is essential for providing accurate data on future power production, ensuring secure power grid operations, and reducing solar energy operation expenses. This research introduces a novel dual-steam hybrid model that uses Bidirectional Long-Short Term Memory (BiLSTM) and Convolutional Neural Networks (CNN) to predict PV power production. The proposed model employs a parallel processing technique, with BiLSTM and CNN analyzing input data independently to detect temporal and spatial features. These features are then combined and passed to the multihead attention layer to further identify the most desirable features for PV power forecasts. The proposed model’s performance is thoroughly assessed by a series of experiments that include various window sizes, four seasons, and different weather conditions. Subsequently, the predictive accuracy of the developed model is compared with three single and five hybrid deep learning models. The findings show that the dual-stream attention-based hybrid network can precisely predict future PV production across various meteorological, seasonal, and climatic conditions.

27 December 2022 in Energies
Verified
Journal Article
Open access
PDF file
AMAssia Mahrouch
Assia Mahrouch
Mohammed OuassaidMohammed Ouassaid
Mohammed Ouassaid
ZCZineb Cabrane
Zineb Cabrane
SLSoo Hyoung Lee
Soo Hyoung Lee
Wind turbine generators (WTGs) are highly sensitive to the disturbances of the grid and tend to disconnect quickly during a voltage dip (when the voltage value is less than 80% of the nominal voltage) or when the frequency is greatly changed. As an increasing number of permanent magnet synchronous generators (PMSGs) are incorporated into the modern power grid, system operators expect PMSG-WT to play an active role in low-voltage ride-through (LVRT) and primary frequency regulation (PFR). Consequently, PMSG-WTs must be capable of supplying additional active power in response to changes in system voltage and frequency. In this context, a new de-loaded technique enhanced by a fuzzy-logic controller is suggested to develop the PMSG-pitch angle control (PMSG-PAC). The studied MG consists of a wind farm (WF), variable load, and a battery energy storage system (BESS). The WF contains five PMSG-WTs which are considered to be the principal resource. The proposed DT-FLC ensures maximum aerodynamic reserve power for the plant, enhances its capability to regulate the PAC, adjusts the WTG drop in response to the wind speed, and increases the resilience of the PMSG-WT in the presence of low voltage. Moreover, the PFR is significantly improved in terms of controlling the PAC (−0.0007 Hz) which meets the frequency maximum droop recommended by the IEEE Std 1547-2018 and the Moroccan grid code, −3 Hz and −2.5 Hz, respectively.

1 May 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
AMAssia Mahrouch
Assia Mahrouch
CCCedric Caruana
Cedric Caruana
Reiko RauteReiko Raute
Reiko Raute
Accurate short-term forecasting of photovoltaic generation and load demand is essential for efficient operation of residential PV–battery energy storage systems. This paper presents a regression-based forecasting and control framework for 24-hour and one-week horizons, integrated with a lifetime-enhancing battery energy management strategy. The proposed control method restricts battery charging to the energy required to meet forecasted load demand, thereby utilising excess PV generation more effectively while reducing unnecessary cycling and battery stress. The forecasting models are developed using MATLAB Regression Learner and trained on real operational data, with meteorological inputs such as solar irradiance and temperature obtained from the NASA POWER dataset. The methodology is evaluated using a residential PV system in Malta under summer operating conditions, where high solar availability leads to frequent energy surpluses and grid feed-in peaks. Three forecasting scenarios are compared: persistence forecasting as a worst-case baseline, perfect forecasting as an ideal upper bound, and regression-based machine learning forecasting as the practical implementation. Results show that PV forecasts generally achieve higher accuracy than load forecasts due to their stronger dependence on weather variables, while load demand remains more difficult to predict due to stochastic user behaviour. Regression-based models improve prediction accuracy compared to persistence methods and enable more efficient battery scheduling, although performance varies across seasonal and system conditions. The proposed control strategy significantly reduces average battery state-of-charge, limits grid peak power, and improves self-consumption. Overall, the integrated forecasting and control framework enhances energy management performance, reduces battery degradation, and improves grid interaction, demonstrating strong potential for practical deployment in residential PV–battery systems.