1. UKM Learning and Research Repository
2. PHD Thesis / Tesis PHD
3. Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina

Title:	A serverless smart grid infrastructure for big data management in a heterogeneous environment with a hybrid load forecasting model
Authors:	Ammar Jameel Hussein Al-Bayati (P98736)
Supervisor:	Nor Fadzilah Abdullah, Assoc. Prof. Dr.
Keywords:	Universiti Kebangsaan Malaysia -- Dissertations Dissertations, Academic -- Malaysia Big data management Heterogeneous environment Database management
Issue Date:	29-Sep-2022
Description:	A smart grid (SG) has the potential to bring significant improvements to the energy sectors for adequate handling of energy demand fluctuations. However, in the absence of effective infrastructures such as next-generation hardware and software, optimized energy production, transmission, and distribution are a challenge for SG implementation at a national level. Likewise, the data resulting from power consumption can be represented in multiple dimensions, with a mixture of structured and unstructured data. Dealing with a multitude of heterogeneous data from both mechanical and smart meters such as data acquisition, data federation, data management and data analytics is an ongoing challenge in the energy sector. Moreover, converting these data into useful information remains an open research area, especially in a transitional environment that may take numerous years to fully transfer mechanical meters to smart meters and may require many phases to take place. This research proposes a scalable serverless SG-advanced metering infrastructure (SG-AMI) based on fog-edge computing, virtualization, and function as a service (FaaS) architecture model that can improve operational flexibility, system performance, and total cost of ownership. Firstly, the serverless SG-AMI design is benchmarked against the Iraqi Ministry of Electricity (MOELC) proposed designs based on current traditional computing (TC) and cloud computing (CC). The results show that the proposed design offers an improvement of 20% to 65% performance on network traffic load, latency, and time to respond, with a reduction of 50% to 67% on the total cost of ownership, lower power, and cooling system consumption compared to the SG design proposed by MOELC. The second part of this research focuses on modelling realistic and efficient power consumption data management in the heterogeneous Iraq energy sector environment. A novel Power Consumption Information and Analytics System (PIAS) is proposed, which can perform various roles such as data acquisition from mechanical and smart meters, data federation, data management, data visualization, data analysis, and load forecasting. The PIAS offers an integrated power consumption data management platform and a hybrid load forecasting model combining Fuzzy C-Means clustering (FCM), Auto-Regressive Integrated Moving Average (ARIMA), and Gradient Boosted Tree Learner (GBTL). The dataset used is based on an hourly basis, for a one-year duration from (1st Jan 2019 to 31st Dec 2019), for the Baghdad governorate. The proposed FCM-ARIMA-GBTL hybrid model was evaluated against standalone ARIMA and GBTL models. The outcomes confirm that the proposed hybrid load forecasting accomplished a high accuracy with MAPE of 0.509, MAE of 0.000865, and RMSE of 0.001151. This research offers a roadmap for a heterogeneous SG-AMI infrastructure toward increasing the scalability and interoperability, automation, and standardization of the energy sector in a developing country.,Ph.D
Pages:	138
Publisher:	UKM, Bangi
Appears in Collections:	Faculty of Engineering and Built Environment / Fakulti Kejuruteraan dan Alam Bina

Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.