2771-1463ISSN (Online)
2771-1471ISSN (Print)
The abstract of this work is to design an alternative control scheme – RTD-A, that combines the simplicity of PID controller with technical – brilliance of MPC controller by avoiding the tuning problems associated with both, for a highly used industrial process, Dual CSTR. Continuous stirred-tank reactor (CSTR), is a standard process used in chemical industries/engineering and environmental engineering. Cascading two CSTRs will lead to decrease in cost and volume when compared to single CSTR. In the proposed work, the temperature control of coupled CSTR is attempted by implementing PID, adaptive control, MPC, and the new generation RTD-A controllers. The performance of the proposed control schemes is compared and it is proved that the RTDA controller outperforms the other control schemes in terms of settling time and ISE.
Read MoreDoi: https://doi.org/10.54216/JCHCI.050101
Vol. Volume 5 Issue. Issue 1 PP. 08-19, (2023)
The leading cause of death, which affects millions of individuals globally is the cardiovascular disease. Heart problems are a major issue in health care, particularly in the field of cardiology. Due to a number of risk factors, including diabetes, high blood pressure, high cholesterol, an irregular pulse rate, obesity, and smoking, cardiac illness is difficult to detect. Due to these limitations, researchers are now using Data Mining and Deep Learning Algorithms to predict heart related disorders. The Cardio Vascular Disease (CVD) is as complicated as it sounds if left untreated. So, the early prediction of this could save millions of people from silent attacks, myocardial infarction etc. Many machine learning algorithms like Naïve Bayes, K-Nearest Neighbor Algorithm (KNN), Decision Trees (DT), Genetic algorithm (GA) are used for cardiovascular disease prediction using text datasets and their efficiencies tend to differ. Generally, convolutional neural network (CNN) algorithm is mostly used for prediction using images. But our concept is to switch over this and predict heart disease using the CNN algorithm for Cleveland dataset which consists of numerical. In this dataset we consider 14 attributes and used K Nearest Neighbor and CNN algorithm. In terms of accuracy, CNN beats KNN, proving that deep learning algorithms may support decision-making and prediction-making based on vast volumes of data supplied by the healthcare sector.
Read MoreDoi: https://doi.org/10.54216/JCHCI.050102
Vol. Volume 5 Issue. Issue 1 PP. 20-31, (2023)
Agricultural use of alternative energy has become more prevalent. Utilizing the alternative source when it is widely accessible is economical and prudent. Drip irrigation may be even more effective when alleviated with renewable energy via a power grid link. Fog computing is a cutting-edge method for extending cloud services to the network's edge. With compute and storage capabilities, it offers a widely dispersed, virtualized platform. Fog could analyze vast volumes of data before sending it to the cloud. This work proposes an innovative agricultural system with integrated hydropower management and its functional blocks. For processing and decision- making in this system, the fog router received field data from the aggregator. To use the data for analysis in the future, it will be stored in the cloud. We have constructed an intelligent irrigation and power management system based on the IoT in our suggested design with IIPMS. This prototype model detects heat and light using temperature and light sensors. If this dual parameter is discovered to be sufficient, the intelligent switch automates the switchover to solar power. The gadget and motor operate on a regular power supply from the power plant. Through GSM technology, the cloud informs the farmer about the type of electricity being used and information linked to power, such as voltage. To inform the farmer of the availability of solar power, a built-in prediction module was also proposed with the Time Series Analysis based forecasting to carry out forecasting duties (TSA). Based on the simulation study, we claimed that the proposed approach performs better in various real-world agricultural scenarios. We also compared our energy consumption model with the existing models and claimed the efficacies of the proposed approach.
Read MoreDoi: https://doi.org/10.54216/JCHCI.050103
Vol. Volume 5 Issue. Issue 1 PP. 32-41, (2023)
Earthquake is one of the most threatening natural disasters which is caused due to the shaking of the earth’s surface. Common cause of earthquake is due to ground shaking, underground volcanic eruption. Here, XGBoost Algorithm is used to predict the location of the earthquake. In this paper, a earthquake location prediction method is proposed, which is based on the composition of a known system whose behaviour is administered according to the evaluation of more than two decades of seismic events and is designed as a time series using Machine learning. By analyzing the parameters such as Latitude, Magnitude, Depth, Longitude, Depth error, Gap, Time etc.
Read MoreDoi: https://doi.org/10.54216/JCHCI.050104
Vol. Volume 5 Issue. Issue 1 PP. 42-45, (2023)
This paper presents advancement in lip print perception and a advancement of biometric system by scanning the lips and getting the lip prints of the individual for security purposes to safeguard confidential data and information. The method used here to identify the lip prints is Cheiloscopy and which scans the lips with the help of camera with a micro lens to get lip prints. This security system was developed using machine learning in python and IoT system. This security system is an alternative for fingerprint and footprint. The security system here differentiates lips based on lip pigment, texture, and prints presented on the lips and check the database for a match. The security systems are fully developed with IoT systems and machine learning in python. The R-CNN in machine learning is used for lip analysis and supervised learning in machine learning is used to find a perfect match.
Read MoreDoi: https://doi.org/10.54216/JCHCI.050105
Vol. Volume 5 Issue. Issue 1 PP. 46-52, (2023)