ISIra Devi Sara

Ira Devi Sara

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Ira Devi Sara has a total of 31 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    A Novel Type-2 Fuzzy Framework Integrating MLP Inference and Savitzky-Golay Filtering for Enhanced Sun Tracker Performance

    by Muhammad Basyir, Yuwaldi Away, Tarmizi Tarmizi, Ira Sara, Andri Novandri

    This study proposes a novel model to enhance the tracking performance of a dual-axis sun tracker system, namely the Savitzky-Golay Filtered Type-2 Fuzzy Multilayer Perceptron (SGF-T2FMLP). The primary novelty of this study lies in integrating the Savitzky-Golay Filter (SGF), Type-2 Fuzzy (T2F), and Multilayer Perceptron (MLP) into a unified framework. In the proposed framework, the MLP serves as a data-driven inference mechanism to replace the Karnik-Mendel (KM) algorithm. Through this approach, the T2F model simplifies the computational process, enabling direct output generation without the iterative procedures required by the KM algorithm. Furthermore, the SGF is employed to reduce noise and improve sensor data quality, while T2F is used to represent data uncertainty via the Upper Membership Function (UMF) and the Lower Membership Function (LMF). These membership degree values are subsequently used as input features for the MLP inference model to generate azimuth and altitude angle outputs. Model evaluation was conducted by testing various combinations of SGF and MLP parameters, including different filter orders, activation functions, and optimizers. The results demonstrate that the proposed framework, particularly the SGF-2 LBFGS-Sigmoid configuration, effectively improves data quality by reducing noise, produces accurate and stable predictions, and maintains strong generalization capability. Monte Carlo-based uncertainty analysis further indicates that the model is highly robust, with 95% confidence intervals of ±5.4723º for azimuth and ±0.0134º for altitude.

  • Open Access

    Systematic Literature Review on AI-Enhanced Dual-Axis Solar Tracking Systems: Techniques and Performance Analysis

    by Muhammad Basyir, Yuwaldi Away, Tarmizi Tarmizi, Ira Devi Sara

    Recent advancements in artificial intelligence have significantly improved the performance of intelligent dual-axis solar-tracking systems, enabling more efficient photovoltaic (PV) energy harvesting under variable irradiance conditions, transient cloud cover, and mechanical uncertainties. This systematic literature review synthesizes 25 peer-reviewed studies (2018–2024) identified from Scopus and Web of Science using PRISMA 2020 procedures. We examine controller families (fuzzy logic, adaptive neuro-fuzzy interface system, deep reinforcement learning, and hybrid designs), sensing and actuation stacks (ephemeris, light sensors, inertial measurement, and computer-vision-based pose), and reported outcomes for tracking accuracy and energy gain. Across comparable conditions, AI-enabled controllers consistently reduce tracking error by ~10–35% and increase annual energy capture by ~8–25% relative to fixed-tilt or conventional rule-based/PID baselines, with the largest gains observed under partial shading and rapidly varying sky conditions. Validation is dominated by simulation, while prototype and hybrid (simulation plus field) evaluations—though fewer—provide stronger evidence of real-world robustness. Persistent challenges include computational cost on embedded hardware, sample-efficient learning and safety for field deployment, inconsistent reporting of metrics, and limited long-horizon testing. To address these gaps, we recommend (i) standardized benchmarking that reports tracking error, normalized energy yield, control latency and controller power, (ii) low-cost edge-AI implementations (model compression, quantization, and microcontroller-class deployment), and (iii) multi-season field trials with reproducible protocols across climates. The findings indicate a clear shift from static, hand-tuned control toward intelligent, adaptive methods. Hybrid designs emerge as a practical choice for deployment due to their combined stability and adaptability, whereas deep reinforcement learning shows state-of-the-art performance primarily in simulation and calls for careful simulation-to-real transfer. Overall, this review clarifies the evidence base and outlines priorities for fault-tolerant, low-cost, and real-time adaptive dual-axis solar tracking.

  • Open Access

    A New Damped Double-Tuned Filter to Improve Power Quality and System Performance for Nonlinear Household Loads

    by Faisal Irsan Pasaribu, Ira Devi Sara, Tarmizi Tarmizi, Nasaruddin Nasaruddin

    The growing use of nonlinear household appliances, such as LED lighting and inverter-based devices, has led to significant power quality problems. This is mainly due to harmonic currents altering the shape of voltage waveforms. Such distortions can lead to increased system losses, transformer overheating, and reduced equipment lifespan. Therefore, this paper proposes an optimized model of a new damped double-tuned filter (DDTF) designed to accommodate dynamic variations in household loads. The particle swarm optimization (PSO) algorithm is used to enhance the design by determining the optimal values for the filter’s constituent parts. Additionally, an artificial neural network (ANN) model is developed to validate and predict filter performance based on experimental data. The DDTF is specifically designed to mitigate dominant harmonics at the 3rd, 5th, and 7th orders. Both simulation and experimental validation were conducted using MATLAB Simulink under realistic household load scenarios. At peak load (2100 W), the unfiltered system exhibited a total harmonic distortion of voltage (THDv) of 155.1%, a total harmonic distortion of current (THDi) of 204.41%, and a power factor of 0.55. After using the new six-stage DDTF at various load levels (from 350 W to 2100 W), the THDv dropped to 7.98%, the THDi fell to 3.57%, and the power factor increased to 0.8089. The ANN-based performance evaluation achieved 94% prediction accuracy, with an error margin of 2% to 6%. These results demonstrate that the designed DDTF is a viable, efficient, and cost-effective approach to mitigating harmonics and enhancing power quality in residential electrical systems.


  • Open Access

    SVR-Optimized ANN Model for Predicting Earthquake Risk in Electrical Substations Based on Disaster Datasets in the Aceh Region, Indonesia

    by Elvy Sahnur Nasution, Yuwaldi Away, Syahrial, Ira Devi Sara, Andri Novandri

    This paper proposes a predictive model based on an Artificial Neural Network (ANN) optimized with Support Vector Regression (SVR) to predict seismic parameters as part of earthquake risk mitigation for substations. The proposed model aims to improve the accuracy of seismic parameter predictions, including seismic energy, Peak Ground Acceleration (PGA), and Modified Mercalli Intensity (MMI). A case study was conducted in the Aceh region, Indonesia, which is located near the subduction zone between the Indo-Australian Plate and the Eurasian Plate, making it highly seismically active. Therefore, this model is developed to identify substations most at risk of earthquake impacts while supporting disaster mitigation planning through early warning systems. The dataset used in the modeling includes information on the epicenter and substation coordinates, earthquake magnitude, depth, and the distance between the two points. The ANN training process uses backpropagation with sigmoid and tanh activation functions. The ANN outputs are then used as input features for the SVR model, applying a Radial Basis Function (RBF) kernel to capture the non-linear relationships between variables. Evaluation results show that the hybrid SVR-ANN model significantly improves prediction accuracy compared to conventional ANN. This improvement is indicated by a reduction in error and an increase in the coefficient of determination (R2). The findings of this paper can serve as a foundation for developing more effective mitigation systems to reduce earthquake impacts on electrical infrastructure.


  • Open Access

    Enhancing Power Density in Switched Reluctance Generators Through Split Axial Segmentation

    by Parlin Siagian, Ira Devi Sara, Tarmizi Tarmizi, Nasaruddin Nasaruddin

    The main contribution of this paper is to propose a methodology for the electromagnetic design of axial split segmentation switched reluctance generators (SRGs) to optimize power and torque. This paper investigates the impact of generator geometry segmentation on performance, addressing the longstanding challenges of torque ripple and power density. The hypothesis suggests that dividing the stator magnetization at each rotor angle can counteract the opposing machine torque, thereby improving torque ripple and enhancing power density. A mixed-methods approach combining literature review, simulation, and experimental validation was used to evaluate torque retarding effects and machine power density. The results demonstrate a significant reduction in torque ripple (59.3%) and an increase in power density 603.3 kW/m3 across all four segments. Specifically, axial separation segmentation markedly enhanced power density and design efficiency compared to conventional SRG designs. This paper provides novel insights into improving torque ripple and power output while maintaining the same overall machine dimensions at 1206.6 watts. Additionally, the simpler fabrication process of axial segmentation—where each segment maintains the same diameter—offers distinct advantages over radial segmentation, making it more practical for manufacturing. These findings highlight the potential for harvesting low-power driving energy from natural sources, contributing to advancing renewable energy technologies. So, the improvements in torque ripple reduction and power density provide valuable directions for optimizing SRGs, with further potential for enhancing driving turbine performance and utilizing better materials for greater energy harvesting efficiency.


  • Open Access

    ANFIS Parallelization Control on Triple-Axis Sun Tracker to Minimize Solar Rays Incidence Angle on Photovoltaic

    by Hari Anna Lastya, Yuwaldi Away, Tarmizi, Ira Devi Sara, Roslidar, Andri Novandri

    This study proposes an Adaptive Neuro-Fuzzy Inference System (ANFIS) based on model with Parallelization Control Technique (PCT) implemented in a triple-axis sun tracker system. The system aims to maximize solar energy absorption by minimizing the angle of incidence of sunlight on Photovoltaic (PV) panels. It is designed with three manipulator arms, each controlled in parallel by a distinct ANFIS model. The primary goal is to adjust the PV panel orientation to optimal azimuth and altitude angles. The tetrahedron sensor is employed to detect the direction of sunlight. This study tests various types of consequent rules to find the optimal configuration for each ANFIS model, including first-order, second-order, and third-order polynomial functions. The analysis results show that the second-order consequent rule provides the highest accuracy across the three ANFIS models. The proposed system can cover azimuth range from 0° to 360° and altitude angle from 60° to 90°. The experiment also demonstrated that the system consistently moves toward a stable condition, even in the presence of minor disturbances. These findings indicate that ANFIS with PCT can minimize the angle of sunlight incidence, enhance system flexibility, and accelerate response time, making it a viable solution for increasing energy harvesting by PV panels.


  • Open Access

    Harnessing Light-Emitting Diode technology for cutting-edge Photovoltaic module characterisation

    by Ira Devi Sara, Fikrel Jasman, Hafidh Hafidh, Amir Zaki Mubarak

    Significant advancements in Light-Emitting Diode (LED) technologies, including the availability of high-power LEDs at reduced costs, have rendered them a highly desirable and dependable solar simulator. However, there are still challenges in designing LEDs as solar simulators to evaluate the characteristics of specific solar modules across different technologies and sizes, particularly regarding spectral composition and spatial uniformity of irradiation. This work seeks to develop a prototype LED solar simulator capable of generating uniform light with stability of at least 90% (class C) throughout an illumination area of 54 cm × 67cm for laboratory instructional applications. The employed procedures encompass the selection of high-power LED bulbs, thermal management, electrical design, prototype fabrication, and performance assessment. The International Electrotechnical Commission (IEC) 60904-9:20 standard delineates the performance criteria for solar simulators and serves as a framework for constructing LED solar simulators. The study's results demonstrate that the LED solar simulator prototype has been successfully constructed and is categorized as class C, exhibiting a light non-uniformity and stability level of 10%. The capacity to delineate the current and voltage of a 50 cm × 30 cm small solar module under specified irradiance and temperature circumstances may be attained with a mean absolute error percentage of 0.01 and a root mean square error of 0.21. The hexagonal configuration of the LED array offers advantages for light design, including modularity and ease of expansion to achieve a broader illumination area. This arrangement can serve as an invaluable instrument to enhance solar energy research and maximize the performance of photovoltaic devices under regulated conditions.

  • Open Access

    Heat Reduction on Light Emitting Diode Solar Simulator

    by Cut Nurmuthaharah, Ira Devi Sara, Tarmizi Tarmizi

    LED solar simulator converts electrical energy into light and heat. The resulting heat effect apply damage to the LED and affect performance and accuracy test. Despite being equipped with a heatsink and fan cooling system, the temperature of the LED solar simulator is still considerably quite high though it is necessary to lowered the temperature even further to increase its service life. One major challenge faced is the management of the heat generated by the LEDs. This article aims to investigate the effect of placing a copper pipe-based cooling system containing refrigerant R32 and Thermoelectric to reduce the temperature of the LED solar simulator by at least 10%. The use of thermoelectric cooler (type 12706) as an alternative cooling technology is safer for surrounding environment compared to vapor compression. Copper pipes were chosen for better conductivity, flexibility and better heat absorbance, meanwhile refrigerant R32 were chosen for it is environmentally friendly and possess a lower GWP. The test results illustrate the application of a copper pipe-based cooling system containing refrigerant R32 and thermoelectric succeeded in reducing the temperature of the LED solar simulator by 11,41% and increased the level of radiation uniformity from an average of 422 W/m² to 429 W/m² with a depreciation of non-uniformity of 0,84%. Thus, this very combination cooling system is proven effectively reducing the temperature of the LED solar simulator by at least 10%.

  • Open Access

    The Effect of Partial Shadings on the Output Power of the Photovoltaic Modules Connected with Different Current and Voltage Characteristics

    by Ira Devi Sara, Aidil Aqsa Sh, Tarmizi Tarmizi

    A mismatch in the output power of photovoltaic (PV) modules in a PV array can occur due to partial shading or a module replacement. Substitution of a module in a PV array with a new one might lead to different current and voltage characteristics between the new and existing modules and result in power losses. The amount of power loss might be increased more if the PV array experiences partial shading. For this reason, this study aims to investigate the effect of partial shadings on the power output of photovoltaic arrays with different current and voltage characteristics. The PV array under test consists of 25 units of solar modules with a total cross-tied (TCT) configuration. There are five shading conditions applied to the test PV array, i.e., the short narrow (ShN), the short wide (ShW), the long narrow (LnN), the long wide (LnW), and the diagonal. A magic square method is applied to reduce the power loss when the PV modules experience partial shading conditions. The results show that the power loss due to partial shadings, either on all identical modules or partially identical, is the same. The most significant power loss occurs in the long comprehensive shading scenario, where 80% of the modules experience shading, which is 41.30%.

  • Open Access

    Development and design of the effect of adding a heatsink as a coolant on improvement output power on solar panels

    by Asri Muhammad Jalil, Yuwaldi Away, Nasaruddin Nasaruddin, Ira Devi Sara, Saifuddin Muhammad Jalil

    This study discusses the use of solar energy as a clean and widely available alternative. The performance of solar panels is influenced by various factors including solar radiation, module, temperature, and distribution of the sunlight spectrum. Temperature issues in solar panels can be addressed through passive cooling, such as installing a heatsink. A passive cooling system is implemented by installing a heatsink with four layers of fins on the body panel of the solar module. The advantage of passive cooling with constellations is that it does not require additional energy from outside, while a large number of fins can achieve the best passive cooling constellation. Each heatsink utilizes eight fins with a total of three heatsinks, manufactured from aluminum. Vector values of wind speed, humidity, and temperature are studied through direct measurements. A 100 W polycrystalline solar module at a standard temperature of 25 ℃ produces an open circuit voltage of 22 +V.