International Journal of Engineering Systems Modelling and Simulation (12 papers in press)

Regular Issues

• A novel keyframe extraction technique systems using deep reinforcement learning simulation  
  by M. Dhanushree, R. Priya, P. Aruna, R. Bhavani 
  Abstract: Smartphones and wearables make video capture easy, increasing video generation. The abundance of videos led to the development of a video summarisation study. The most important moments of a longer input video are selected to minimise its length while maintaining context. Video summary using keyframe extraction outputs key frames of key events. Its uses include anomaly detection, efficient video storage, indexing, and retrieval. Extracting semantically meaningful frames is difficult, and a big research gap exists. The keyframe extraction using deep reinforcement learning (KE_DRL) approach extracts representative and distinct semantically relevant keyframes. Frame-level and video-level characteristics are extracted. Frame-level characteristics are extracted using modified ResNet50 and I3Dnet. They are aggregated to generate a feature vector and global average pooled to get video-level features. attention-based video summariser network (AVSumnet) uses semantic video attributes as input. It is trained via reinforcement learning. A new summariser network training reward mechanism is proposed. Experimental findings show that the KE_DRL method creates better video keyframes than existing methods.
  Keywords: keyframe extraction; video summarisation; bi-directional gated recurrent unit; BDGRU; deep reinforcement learning; attention mechanism; squeeze and excitation block; policy gradient; reward function; representativeness.
  
  
• Research on steady speed control strategy of direct-drive pump-controlled hydraulic motor based on improved extended state observer  
  by Guoshuai Li, Yong Sang, Qingli Qi 
  Abstract: This paper introduces the direct-drive volumetric control (DDVC), focusing on the direct-drive pump-controlled hydraulic motor electro-hydraulic system as the primary subject of research. Aiming to address the problem of parameter uncertainties and unknown nonlinear disturbances in the system. A robust backstepping controller based on the extended state observer (ESO) is designed to improve the systems dynamic performance and robustness. In the controller, the ESO is used to estimate the total uncertainties in the system. Then, the robust backstepping controller is designed to enhance both the dynamic and robust performance. Moreover, Lyapunov theory is utilised to rigorously demonstrate the stability of both the ESO and the closed-loop control system. Finally, the results indicate the designed controller has excellent robustness and dynamic performance.
  Keywords: direct drive; electro-hydraulic system; nonlinear disturbances; robust control; extended state observer.
  
  
• Impact analysis and control of EV charging on grid connected to optimal structure hybrid wind-solar PV system  
  by Abdul Zeeshan, Swapnil Srivastava 
  Abstract: A detailed review of research work has been done for wind-PV hybrid generation system (WPVHGS) to identify suitable plant structure. There is scope of optimisation of WPVHGS when feeding to grid at unbalanced voltage conditions. In addition to this, optimal sizing of battery energy storage (BES) and supercapacitor (SC) is required for the same. To support the review, simulations are carried out on MATLAB Simulink for comparative analysis of impact on grid due to EV charging, where a centralised charger is charging battery equivalent to 10 EVs. Using simulations, severity of grid current imbalance is observed, when battery charger is connected to 575-V bus. When doubly-fed induction generator (DFIG)-based wind energy conversion system (WECS) is introduced then currents and voltages are balanced but severe third harmonics of current is introduced. By replacing proportional-integral (PI) controller with adaptive neuro-fuzzy inference system (ANFIS), its observed that voltage and current total harmonic distortion (THD) gets mitigated.
  Keywords: hybrid; wind; solar PV; battery; supercapacitor; doubly-fed induction generator; DFIG; electric vehicle; battery energy storage; BES; adaptive neuro-fuzzy inference system; ANFIS; total harmonic distortion; THD; wind energy conversion system; WECS.
  
  
• Earthquake prediction using deep learning based-recurrent neural network technique  
  by J. Sahaya Ruben, M. Adams Joe, M. Anand, M.Prem Anand 
  Abstract: This article explores the application of deep learning techniques to enhance earthquake prediction and detection, addressing the critical need for improved disaster preparedness and risk mitigation. With natural disasters like earthquakes causing widespread harm to people, and ecosystems, the study focuses on leveraging machine learning methodologies, including recurrent neural networks and reinforcement learning, to develop an innovative earthquake prediction framework. The proposed approach begins with feature extraction using CNN to capture essential seismic data patterns. Subsequently, the RNN model is trained to analyse time series seismic data, allowing for the prediction of earthquake events with enhanced precision. In addition, Q-learning is integrated into the process to make informed decisions based on the current state, potentially reducing false alarms and improving overall prediction accuracy. The promising results and potential impact of this research underscore the importance of ongoing efforts to harness technology for more effective earthquake prediction and mitigation strategies.
  Keywords: convolutional neural network; CNN; recurrent neural network; RNN; Q-learning; deep learning.
  
  
• Analysis of grid-tied inverter based distributed generation with focus on control of active power and voltage compensation  
  by Siddharthsingh K. Chauhan, Vaidehi D. Sathwara, P.N. Tekwani 
  Abstract: Nowadays, renewable energy sources (RES), especially photovoltaic (PV) systems, are popularly used as distributed generators. This is due to availability of high-power grid-tied inverters (GTIs) used for interconnecting PV systems with grid. The PV based distributed generation (DG), conventionally used for exchange of active power with grid, further can be controlled for added functionalities of reactive power exchange, voltage compensation, and harmonic mitigation. This multi-functional operation is achieved by effectively controlling the grid-tied inverter of DG. Such multi-functional control of grid-tied inverter is analysed in this paper. For control of multi-functional grid-tied inverter (MFGTI), comprehensive power quality evaluation (CPQE) index technique is used. In this paper, effective control of the space vector modulation (SVM)-based hysteresis controlled CPQE-based GTI is presented through simulation studies. Effective performance of the proposed GTI for active power exchange as well as voltage compensation during fault condition is presented.
  Keywords: comprehensive power quality evaluation; CPQE; distributed generation; multi-functional grid-tied inverter; MFGTI; voltage compensation; renewable energy sources; RES; grid-tied inverters; GTIs; space vector modulation; SVM.
  
  
• Excitation effects of arc grounding fault on PT ferroresonance of distribution networks connected to photovoltaic in high-altitude areas  
  by Jinpeng Yuan, Jingwen Sun, ZiBin Li, Wande Lin, Zile Wang, Jianwu Li 
  Abstract: Power distribution network systems in high-altitude areas with low air pressure are prone to arc grounding fault, and abundant solar resources lead to the access of numerous distributed new energy sources, which further increases the risk of potential transformer (PT) ferroresonance excited by arc grounding fault. This study designed a device simulating the arcing characteristics in high altitude environment, established a Mayr arc model for simulating arc grounding fault, analysed the fault characteristics of PT ferroresonance overvoltage caused by arc grounding, metallic grounding and accessing to new energy in high altitude areas, and compared the effects of connecting damping resistor at the open delta of PT secondary winding and connecting single-phase PT at the neutral of primary winding to suppress ferroresonance under different conditions, which may provide a reference for suppressing ferroresonance caused by new energy access in high altitude areas.
  Keywords: ferroresonance; overvoltage; potential transformer; arc grounding fault.
  DOI: 10.1504/IJESMS.2025.10071011
   
  
• Spectrum sharing using deep learning: multi-agent reinforcement learning  
  by B.V. Santhosh Krishna, A. Bharathidhasan, N. Ashokkumar, K. Periyar Selvam 
  Abstract: The number of people using cell phones and the requirement for the radio band has increased over the last few years. The fast rise of 5G networks for wireless communication and wireless communication has met this need. There is reason to believe that the issue of improper use of the wireless spectrum could be resolved with the progress of cognitive radio and its spectrum-sensing technology. Deep learning technology is known for being able to learn and change amazingly quickly. The purpose of this research is to provide a brief summary of the approach used in cognitive radio spectrum-sensing technology and deep learning technology. The first part of this study talks about the common spectrum-sensing methods to give a big picture of the benefits of deep learning-based spectrum-sensing algorithms. We find that our method can increase the accuracy of previous work and conventional learning strategies by as much as 83%.
  Keywords: cognitive radio; spectrum sensing; wireless communication; cooperative spectrum sensing.
  DOI: 10.1504/IJESMS.2025.10071409
   
  
• Comparative analysis of marine debris simulation using ensemble learning with XGBoost and deep convolutional neural networks  
  by S. Belina V.J. Sara, Gnaneswari Gnanaguru, S.Silvia Priscila 
  Abstract: Marine ecosystems, wildlife, and human activities are seriously threatened by marine garbage. Deep learning-based systems for categorisation can automate classifying distinct types of marine debris from photos or video recordings, allowing for more effective and precise monitoring and assessment of debris levels in different maritime circumstances. DL is a useful tool that can help with environmental conservation efforts by categorising marine waste. To improve classification accuracy, sensitivity, and specificity for different types of marine debris, we investigate the use of ensemble learning approaches in this work and used for execution in Python. We compare three distinct implementations of the powerful gradient boosting method XGBoost with innovative deep convolutional neural networks: XGBoost with Adam and GoogleNet optimiser, XGBoost with VGG19 and Adam optimiser, and XGBoost with ResNet and Adam optimiser. The XGBoost algorithm and feature extraction from three different pre-trained CNN architectures, GoogLeNet, VGG19, and ResNet, are used in this study to examine the effectiveness of classifying maritime debris. We highlight the outstanding results obtained by combining ResNet + Adam with XGBoost, a reliable and effective method for classifying maritime trash and producing an accuracy of 91%, specificity of 0.88, and sensitivity of 0.91, respectively.
  Keywords: marine debris simulation; XGBoost ensemble; convolution neural network; CNN; Adam optimiser; image classification; environmental monitoring; optimisation techniques; sensitivity enhancement; deep learning; DL.
  DOI: 10.1504/IJESMS.2025.10071659
   
  
• Improving cooling load prediction in residential buildings with multi-layer perceptron models  
  by Yang Wu, Lanlan You 
  Abstract: Today, building energy efficiency is prioritised since it affects operational costs. Buildings take a lot of energy to maintain pleasant temperatures. Combining this researchs cooling load (CL) forecasting method may optimise building energy use. MLPs forecast household cooling demands. MLP models and regressions generally have linear input-output relationships. This research uses two innovative optimisers, cheetah optimiser (CHO) and adaptive opposition slime mould algorithm, to improve MLP model performance. The data used to train the approaches describes each samples unique traits. These methods will be tested on a simulated dataset using CLs as neural network output variables and building technical attributes as input factors. During the process testing phase, the MLCO (2) (MLP+CHO in layer 2) gets the lowest RMSE value of 0.672 and the greatest R2 value of 0.995. The results demonstrate that the proposed hybrid models MLCO and MLAO significantly outperform the standalone MLP and conventional optimisation methods, achieving a minimum error rate. These findings confirm the proposed models superior predictive accuracy and reliability, underscoring their potential for practical application in enhancing energy efficiency in residential buildings.
  Keywords: cooling load; multi-layer perceptron; MLP; cheetah optimiser; CHO; adaptive opposition slime mould algorithm; artificial intelligence; support vector machines; SVM; artificial neural networks; ANNs.
  DOI: 10.1504/IJESMS.2025.10071781
   
  
• Heat transfer and fluid flow analysis in concentric tube heat exchangers  
  by B. Konda Reddy, G. Bhanu Kiran, K. Jayadeep 
  Abstract: This study conducts a detailed parametric analysis of heat transfer and fluid flow in concentric tube heat exchangers using CFD simulations in ANSYS Fluent. Four configurations bare, continuous finned, slotted, and combined finned-slotted are evaluated for their thermal performance. The impact of mass flow rates and inlet temperatures on heat transfer, temperature distribution, and pressure drop is examined. Results show improved heat transfer in modified designs compared to the bare tube. Continuous finned tubes enhance performance by 12.68%-20.89%, slotted by 0.52%-3.33%, and the combined configuration by 14.47%24.63%. Pressure drop changes are minimal for finned and combined designs, with a slight decrease for slotted tubes. Validation against experimental data yields a maximum relative error of just 0.19%, confirming model accuracy. The study highlights the effectiveness of advanced configurations in enhancing thermal performance while maintaining pressure stability, aiding the development of more efficient heat exchangers.
  Keywords: heat transfer; fluid flow; concentric tube heat exchanger; bare; continuous finned; slotted; combined finned-slotted; ANSYS fluent; heat transfer performance; pressure drop performance.
  
  
• Advancing breast cancer detection: a comprehensive investigation of advanced classification techniques  
  by Shilpa Choudhary, Sivaneasan Bala Krishnan, Prasun Chakrabarti 
  Abstract: As the most prevalent cancer in women worldwide, breast cancer requires early detection to have the best possible treatment results. However, conventional screening techniques, like mammography and clinical examinations, can take time and effort. In this paper, we proposed a predictive model for the identification of breast cancer by combining state-of-the-art model BERT with machine learning approaches. Several machine learning algorithms, such as K-nearest neighbours, decision tree, random forest, neural network, and BERT, were tested on the Breast Cancer Wisconsin (Diagnostic) Dataset for early prediction of the diseases. The BERT models accuracy was improved by using hyperparameter optimisation techniques. For the proposed works evaluation, we used metrics like accuracy, F1-score, precision, and recall on the standard publically available datasets. With an accuracy of 0.98 across various splits and an area under the curve (AUC) of 0.98 in receiver operating characteristic (ROC) curves, our results show that BERT consistently works better than other models. These findings highlight the value of early and reliable identification in improving patient outcomes, highlighting the promise of machine learning algorithms, notably BERT, inaccurate breast cancer prediction.
  Keywords: breast cancer prediction; K-nearest neighbours; KNNs; decision tree; neural networks; random forest; BERT; classification.
  DOI: 10.1504/IJESMS.2025.10072165
   
  
• Cost effective biomass supply chain optimisation for the bio-energy industry  
  by Prajwal Panwar, Anand Chauhan, Anubhav Pratap Singh, Ritu Arora 
  Abstract: Increasing energy needs and environmental concerns require sustainable solutions. A promising alternative to fossil fuels is biodiesel, generated from easily accessible macro-algae such as Ulva fasciata, Cystoseira indica and Gracilaria corticata. A cost-effective macroalgal biodiesel supply chain is proposed using optimisation methodology. Utilising an advanced algorithm, the model optimises biodiesel production from macroalgae procurement to bio-refinery and depot placement. The framework incorporates expenses, biorefinery site, biodiesel storage and strategic macroalgae extraction centres. The model is optimised using a genetic algorithm, factoring in the expenses of installing biodiesel manufacturing plants. The sensitivity analysis demonstrates that these initial expenditures considerably impact the supply chains economic burden. Sensitivity analysis confirms the frameworks usefulness, making it valuable for stakeholders such as traders and policymakers promoting the biofuel business. This study sets forth the foundation for the manufacture of biodiesel from macroalgae on a massive scale, ensuring its sustainability.
  Keywords: macro-algae; sustainable energy; supply chain; bio-diesel; genetic algorithm; optimisation.
  DOI: 10.1504/IJESMS.2025.10073145