2026-05

Dasinya Journal for Engineering and Informatics (Issue : 2) (Volume : 2)

Developing countries are increasingly transitioning to continuous 24/7 electricity supply through multi-tier tariff systems. While improving reliability, this transition imposes a financial strain on consumers struggling to manage real-time consumption. This paper proposes a novel IoT-based Demand Side Management (DSM) and Short-Term Load Forecasting (STLF) framework utilizing an inexpensive Raspberry Pi 3 Model B edge gateway. The system was validated using high-resolution field measurements from the Runaki project in Zakho, Kurdistan, collected between December 1, 2025, and April 1, 2026. Among the four regression algorithms evaluated—Linear Regression, SVR, XGBoost, and Random Forest—the Random Forest model demonstrated superior performance, achieving a Root Mean Square Error (RMSE) of 0.1152 kW and an R-squared (R2) of 0.9858.

2023-06

QALAAI ZANIST JOURNAL (Issue : 8) (Volume : 3)

Due to the difficulty and rarity of the data tab in blood banks, information is often lost or forgotten. Because of this, it is important to have a number of different types of blood and know the total number of blood donors, as well as the number of units of blood imported and exported daily, monthly, or yearly. A web-based intelligent system is a web site that was created to manage the Zakho blood bank by using middleware languages like PHP that were used as transformer languages between MySQL databases as well as back-end web languages like HTML, CSS, and JavaScript that power up the front end, thus accelerating the management of blood import and export and accurately tabulating the blood donors' profiles so hospitals can get facilities that reduce the emergency cases that occur. Machine learning, deep learning, and an algorithm based on long short-term memory (LSTM) and based on time series have been used to predict an estimation of red blood cells (RBC) for periods (weekly, monthly, and yearly) that facilitate attracting the donors and importing and exporting the amount of blood that is needed in the future, and the algorithm produced these results (MAPE 3.259, MSE is 0.001, MAE is 0.031, RMSE is 0.042, and R-Squared is 0.913)

2021-07

Qubahan Academic Journal (Issue : 3) (Volume : 1)

Service composition is gaining popularity because a composite service can perform functions that an individual service cannot. There are multiple web services available on the web for different tasks. The semantic web is an advanced form of the current web in which all contents have well-defined meanings due to nature, allowing machines to process web contents automatically. A web service composition is a collection of web services that collaborate to achieve a common goal. They reveal the established methods for web service composition in both syntactic and semantic environments. In this study Initially, we identify the existing techniques used for the composition. We classified these approaches according to the processing of the service descriptions, which can be syntactic or semantic-based service processes. We have reviewed more than 14 articles in this domain and concluded the merits of the methodologies applied for the implementation of web service composition.