College of Engineering

A PH.D dissertation was discussed in Department of Computer Engineering/College of Engineering at University of Mosul entitled ” Intelligent Power Management and Forecasting System for Institutes Using Wireless Sensor Network ” submitted by the student (Marwa Mushtaq Talib) on Tuesday, March 18, 2025.

The dissertation introduces an Intelligent Power Management system (IPMS) for the optimization of energy distribution by use of Wireless Sensor Networks (WSNs). It incorporates a machine learning (ML) model and a cloud-based monitoring and control capability. The proposed IPMS comprises a Central Unit (CU) including a ML model responsible of data processing and real-time decision-making activities as well as distributed Local Units (LUs) have sensors for collecting occupancy and environmental data. LUs interact with CU via the MQTT protocol, a dependable and effective method best fit for WSN interactions. MQTT protocol depends on PUBLISH/SUBSCRIBE mechanism that allows flawless communication across distributed components depending on topics.
The dissertation evaluates the performance of four different ML models namely: Random Forest (RF), Support Vector Machine (SVM), K-Nearest Neighbor (KNN), and Naïve Bayes (NB) besides the Deep Neural Network (DNN) model to determine the optimal model to be deployed in CU for real-time classification. Each model is trained on a measured data set taken from a university building over one-year duration. The main objective of this process to classify the operating modes in the buildings into three
categories: Shutdown, Full, and Select
. RF model attained the best results in terms of accuracy with 100% accurate classification results and optimal results .