International Journal of Integrated Engineering
Not a member yet
2309 research outputs found
Sort by
Enhancing Short-Term Solar Generation Forecasting: The Superiority of LSTM Over SVR
Get PDFThe global surge in solar power systems and AI-driven technology has spurred the development of precise solar generation forecasts. These are particularly essential in Malaysia\u27s tropical climate with abundant sunlight. Traditional forecasting methods like ARIMA, SARIMA, and ANN are inadequate in modelling solar energy systems\u27 inherent complexity and non-linearity. This study addresses the challenge of forecasting solar generation in a tropical region. The objective is to develop a 30-minute-ahead solar generation forecasting model using advanced techniques, specifically Long Short-Term Memory (LSTM), and compare its performance with Support Vector Regression (SVR). Utilising data from February 2022 to April 2023 collected from a rooftop solar system in Pulau Pinang, Malaysia, this approach effectively handles sudden changes in solar output, known as "ramping events," caused by cloud movement and unpredictable weather. The results reveal LSTM\u27s superiority, with an nRMSE of 6.75%, outperforming SVR (nRMSE of 7.28%). This pattern recognition capability of LSTM holds promise for larger datasets, offering precise forecasts beneficial for weather prediction and power management. Implementing this technique in more solar PV systems can enhance power reliability and promote sustainable energy practices, showcasing LSTM\u27s potential for optimising solar generation forecasting
Reliability Simulation Mechanism Model of Big Data Mileage Prediction for Automotive Warranty
Get PDFStiff competition among automotive manufacturers to secure market share results in a short period of product development until production. Thus, there is a gap of limited attention concerning effective early detection tools leveraging information technology to identify product quality by optimising warranty data towards expediting market action. A big data simulation analysis model of warranty prediction is proposed based on parameter mileage using the Weibull statistic platform. Input of warranty prediction analysis based on warranty historical data, continued with data cleaning and selection. The algorithm model is applied to support big data analysis based on the application of Weibull statistics. The product with the highest failure rate in terms of warranty amount and quantity, part number PN312, was selected. The shape ? value is 2.308, which matches the Rayleigh distribution with the shape ? of product failure at 403,948km when the incidence is 63%. The model orchestrated the future warranty outcome and consequences. Relatively, the warranty prediction system simulates the evaluation costs of poor quality. The development of a new prediction simulation model will enhance the application of QC tools, expedite the selection of poor-quality products, eliminate wasteful resources such as time and manpower, and simplify the investigation proces
Thermodynamic Analysis of Flammability and Ecological Safety Levels of Liquefied Petroleum Gas and R134a Refrigerant Mixtures for Refrigeration Applications
Get PDFThe flammability of hydrocarbons such as propane, butane, and liquefied petroleum gas (LPG) affects their wide acceptance as working fluids in air-conditioning, refrigeration, and heat pump systems. Mixing LPG with non-flammable R134a will produce a mixture with lower flammability. Therefore, this paper investigated the physical properties, flammability levels, and global warming potentials of LPG and R134a mixtures. The results obtained revealed that the mixtures of LPG and R134a exhibit low critical pressures and higher critical temperatures that are reasonably higher than the ambient temperature, which are beneficial to the cooling effect per unit volume of the refrigeration system. The mixture’s boiling point temperature and global warming potential (GWP) reduce as its flammability increases, which will require a diligent trade-off among these three properties of the mixture. The percentages of LPG in the mixture that produced satisfactory GWPs with respect to the EU’s F-gas regulation and those that falls into the safety class of mild flammable range from 45 to 100% and 5 to 90%, respectively. Conclusively, this study reveals that novel ecologically acceptable refrigerants with reduced flammability could be obtained from the mixtures of LPG and R134a refrigerants within the range of 45 to 90% content of LPG
Compressive Strength and Bonding Performance of Concrete Beams Strengthened with GFRP Plates
Get PDFThis study aims to test the bond strength and the structural response of concrete beams reinforced with Glass Fiber Reinforced Polymer (GFRP) plates. The experimental program included a total of 41 prisms, of which 12 were tested for compressive strength, and 25 were configured into beams with four-point loading systems, referred to as the GFRP beam set. In total, six beams attached GFRP plates of 50 mm x 100mm and 200 mm (GFRPP-50 and GFRPP-100), while three beams served as control specimens without externally bonded plates. The findings verified that when GFRPP-100 was used as reinforcement, its ultimate load-bearing capacity increased by 137% compared to the controls, being substantially stronger than the other beams. The research emphasizes the role of the epoxy adhesives described in the methods section, including the bonding surface preparation, enabling a cohesive bond connection. GFRP confinement may be used as an innovative strengthening method instead of traditional methods which results in improved structural performance. Further investigations should look into the effects of differing environmental exposures and durations of loads applied on the durability of GFRP-reinforced beams over time. This study provides evidence to facilitate the acceptance of GFRP as an efficient reinforcement tool that improves the serviceability lifespan
Enhancing Energy Consumption Prediction by Integrating Occupant Activity with Machine Learning Models
Get PDFThe precision of the forecast of the power consumption of buildings isessential for big constructions in the present day. However, many ofthe models in use fail to consider the effect of people’s activitieswithin the building on energy consumption. To overcome thislimitation, this paper uses a synchronized data collection approach tocollect data from different sensors about occupancy activity andpower consumption. Several machine learning models are employedwith this coordinated data, and the effects of occupant behaviour onpower usage are explored. By analyzing the results of the modelsgenerated by the two algorithms, the best ways of reachingbehaviour-sensitive power consumption prediction are determined.Therefore, the findings establish that the additional data concerningoccupant activity provides more accurate assessments of energyusage that can be quite beneficial for enhancing the furtherdevelopment of better adaptive and more efficient buildingmanagement systems. This work also helps to fill the existing gap inenergy prediction literature wherein, unlike other fields, the humanfactor is considered in machine learning models that can lead to moreaccurate and more immune to distortion energy forecasting
LSTM: Anomaly Activity Type Classification Using Distributed Acoustic Sensing Based on MFCC Features
Get PDFThe intergrity and connectivity of the fiber optical network are important in preserving the quality of services and internet realibility between providers and end users. However, these network are vulnerable to disruptions due to unitentional break and damage caused by physical disturbances such as construction activity. An accurate classification of anomaly activty at surounding area plays a crucial role in monitoring the buried fiber optical network from harm which can lead to denial of services. Distributed acoustic sensing (DAS) with combination of deep learning-based technique have potential in targeting this issue, by leveraging the unique pattern of vibration signal measured by the DAS to classify and identify anomaly activities. This work demonstrated utilization of dark fiber buried along the road until the server room, then connected to the DAS interrogator unit(IU). The vibration signals induced by construction hand tools, including hoe, shovel, and sledgehammer, which are used to mimic anamoly activity, are measured by DAS IU and underwent pre processing before exract the mel frequency cepstral coefficient (MFCC) features for long short- term memory (LSTM) model training. The average accuracy score using 25 MFCC resulted up to 87% indicating that the proposed method hasa great potential for anamoly activity monitoring for fiber break prevention
Inverse-Dash Rectangular Slot 2x4 Array 5.8 GHz Antenna for Multiple Access Integration of IoT Systems and Terrestrial Satellite Communication Networks
Get PDFLEO satellites provide a phenomenal solution for the provision of internet networks around the world, especially in remote areas. Along with this, the increasing demand for fast and stable information and communication network access technology, especially in applications such as the Internet of Things (IoT) and satellite communications, demands innovative solutions to overcome infrastructure limitations in remote areas. From 2022 to 2026, this need is expected to increase by around 25%. IoT technology, which is key in the industrial revolution 4.0 and the development of smart cities, relies heavily on the support of communication networks, especially in remote areas that are underserved by mobile and fiber optic infrastructure. This study proposes an optimal antenna design for point-to-point applications on WLAN-IoT networks integrated with LEO satellite communication systems. The antenna design uses a rectangular inverted-dash patch-shaped array configuration with an operating frequency of 5.8 GHz, designed to improve communication range and efficiency. The antenna has a high gain of 5 dBi and a directional radiation pattern that is suitable for difficult environments in remote areas. The prototype antenna test showed positive results with a return loss of -15.78 dB, a VSWR of 1.39, and a radiation efficiency of 96%. The test results also showed a significant increase in transmission signal strength compared to the reference antenna, providing an innovative solution to extend the range of the satellite communication system with high efficiency. This study emphasizes the importance of antenna design that takes into account technical specifications and environmental conditions to optimize the performance of communication networks in remote areas
Automatic Bamboo Fed Type Cutting Mechanism Prototype
Get PDFCurrently, most of the bamboo cutting process is done manually. However, the growing demand for bamboo products and the high cost of labor in some countries requires automatic bamboo cutting machine to increase the production rate. The aim of this project was to develop a bamboo cross cutting mechanism prototype which is controlled by Programmable Logic Controller (PLC). Structural design work was carried out using Finite Element Analysis (FEA) to determine the stability and reliability of this prototype. The cutting mechanism was designed to cut bamboo stem at specific length uniformly. It consists of a rotary disc cutter which is powered by a 240VAC motor, bamboo stem positioner and chain-type feeding system. The feeding mechanism feeds the bamboo stem to the rotary cutter in a proportional manner. Once assembled, the prototype underwent for system verification and production rate was calculated to determine its processing capacity. The result shows this prototype has cut length offset at 2.3 cm in average and the production rate was approximately 15.3 cuts per minute or 943.3 cuts per hour in average. The cut length offset is quite wide and can be improved by redesigning the bamboo stem positioner mechanism
Comparison of Polyaniline Performance as A Nitrate EGFET Sensor Layer with Sonication Temperature
Get PDFThe performance of the nitrate Extended Gate Field Effect Transistor (EGFET) sensor has been compared using polyaniline (PANI) with different sonication temperatures, as presented in this paper. PANI was prepared with two sonication temperatures, 28°C and 38°C then employed as the nitrate sensor layers of the EGFET that fabricated on Zinc Oxide (ZnO) and Indium Titanium oxide (ITO) substrate using dip coating method with 15 seconds dipping times. The sensing electrode (SE) was characterized to observe the performance of the nitrate EGFET sensor by comparing PANI performance with different sonication temperatures when interacting with nitrate ions, resulting in sensitivity and linearity of the sensor toward nitrate concentrations. PANI/ZnO exhibits excellent nitrate detection capabilities when prepared at room temperature of 28°C, with a sensitivity of 58.1 mV/dec and linearity of 0.9956. To assess the long-term response of SE, drift measurements were conducted by tested sensor with a 20-ppm nitrate concentration for durations ranging from 1 to 10 minutes where ZnO/PANI SE at 28°C show response gradually decreased over time, exhibiting a drift rate of 83.4 mV/hour over 10 minutes. Consequently, a sonication temperature of 28°C emerges as the ideal parameter for preparing PANI solutions to yield high-performance nitrate sensors
Gain Enhancement of E-Shaped Microstrip Patch Antenna for Underground RFID Reader Application
Get PDFRadio Frequency Identification or (RFID) is a wireless communication technology that uses radio waves and consists of a reader, tag, and antenna. This paper introduces an innovative concept: an E-shaped microstrip patch antenna utilizing an air gap substrate. The proposed design is tailored for the specific application of underground RFID readers, aiming to optimize the antenna\u27s performance within this unique context. The study encompassed the execution of a parameter analysis to thoroughly explore and optimize the impact of alterations in the antenna dimension on the performance of S11. The antenna design uses coaxial probe feeding technique with aluminum patch. Then, the performance of the final antenna design is analyzed and evaluated. The gain obtained for the proposed design is 9.121 dBi, while the S11 value at the desired resonant frequency 921 MHz is -11.99 dB. The simulation is performed with CST suite software