Cycle 5 (2015 Deadline)
Towards Smart Microgrids: Renewable Energy Integration into Smart Buildings
PI: Mohamed Riduan Abid (R.Abid@aui.ma), Alakhawayn University, with co-PIs Mohamed Bakhouya, International University of Rabat and Khalid Zinedine, University Chouaib Doukkali
U.S. Partner: Driss Benhaddou, University of Houston
Project Dates: December 2016 - November 2019
Besides being a global concern, energy efficiency is growing as a potential market with very promising development and environmental impacts. Smart Grids (SGs) promote energy efficiency in electrical grids, mainly via the integration of renewable energy (thus minimizing greenhouse gas emissions) and via the leveraging of Information and Communication Technology (ICT). ICT is a key element in the optimization of the Demand/Response (DR) variance, which stipulates a real-time dissemination of data between SG components, namely, smart meters at the production site (i.e., renewable energy sources), sensors measuring electricity consumption at the consumer site, and actuators. The interconnection of these components needs a reliable network: the Advanced Metering Infrastructure (AMI).
This project will leverage energy efficiency in smart buildings by promoting “context awareness” whereby the switching on and off of electrical appliances will be based on the context, i.e., temperature, number of people in rooms, humidity, light, and so forth.
To this end, these researchers plan to deploy a holistic platform that implements a real-world microgrid testbed at a building on the Alakhawayn University campus. The deployed smart microgrid model will be promoted for deployment by other organizations at the national level, especially since Morocco is adopting a promising policy for renewable energy integration. In the medium term, the team hopes to promote this technology in sub-Saharan countries as well, given Morocco's geographical location. Supporting research in renewable energy can foster the growth of the green economy in Morocco and in the longer term create job opportunities for Moroccan youth.
| The first prototype of the USAID-NI Lab at Alakhawayn University. Photo courtesy of Dr. Abid|
The project intends to have a positive impact in reducing greenhouse gas emissions, in line with USAID's Global Climate Change and Development Strategy. It is also in line with a Moroccan national initiative to promote renewable energy development.
Summary of Recent Activities
In this reporting period, the PEER team worked on promoting final results and building capacity towards attracting further funds in related venues. In this context, they recruited 2 PhD students (Anas Oujja and Bouali Ettaibi) to work on USAID/AUI Lab respectively on 1. HPC and Big Data Processing (which is one pillar component in our current PEER project) for Genomics Data and 2. Renewable Energy integration in Smart Agriculture. The tird PhD student (Safae Bourhnane) is currently working towards finalizing a Journal Paper, and the 4th PhD student (Naji Najem) is working on finalizing his thesis (Working on a Journal Paper).
The following is a summary of the work done, and the expected work, by the 4 PhD students:
1. "Context-Awareness for Renewable Energy Integration into Smart Buildings" (PhD student Naji Najem). They investigated the energy aware routing protocols in WSN and looked for different ways to optimize packets and transmissions in order to extend the lifetime of the WSN and avoid frequent battery change. They investigated the way to combine the AODV routing protocol with EACRA in order to optimize sampling rate, packet size, and transmissions. This step was based on a theoretical study of the impact of EACRA on the routing process. As a result of this period, they drafted a journal paper that discusses all the previous work and results. The next step, will be to modify the AODV routing protocol in order to fit with the context awareness of data acquisition.
2. "Towards Smart Grids: Renewable Energy Integration into Smart Buildings" (PhD student: Safae Bourhnane)
The work accomplished in the last quarter can be divided into two main parts:
a. Data Prediction: they have published a work that takes a closer look at the prediction of the energy consumption and the scheduling of different smart appliances using ML (Machine Learning). They tried to investigate other solutions and techniques that would deliver a better accuracy for the prediction. They found some statistical methods predicting time series data through less complicated and easier models. Thus, they are opting for the ARIMA (Autoregressive Integrated Moving Average).
b. Raspberry Pi (RP) Based Green Data Center
They are in the final phase of testing our RP based Green Datacenter. At this level of the study, they are looking at the gain/loss in performance and energy consumption of our approach compared to a single performant server. They conducted an experiment where we compared our solution to the Dell Precision Tower server, performance and energy consumption wise. Besides, they made use of the Amdahl’s Law to find the maximum speedup of the both setups. The results have shown that the Raspberry Pi consumes less energy than a performant server. However,
3. "Renewable Energy Integration into Smart Agriculture" (PhD student: Bouali Ettaibi)
After researching WSN technologies in smart irrigation, we used open platform hardware to implement WSN data acquisition using two different communication technologies GPRS and ZigBee. In order to control the irrigation system, they have developed a control algorithm based on fuzzy logic which is responsible for calculating the duration of irrigation and keeping the basin filled by water. The Raspberry Pi control unit is linked to two actuators (relays) that are responsible for the translation from digital to electrical signals, and hence controlling (ON / OFF) the water pumps. In the next three months, they will optimize their model by establishing connection with weather station for precipitation information and investigating the use ML techniques to predict the soil moisture which is the most influencing factor in the irrigation process.
4. "HPC and Big Data processing for Genomics"
During the last quarter, they installed Apache Hadoop framework that provides a distributed storage for Big Data using the MapReduce programming model in a cluster of 3 computers, linked in a network working together to solve a specified task. Afterwards, they started researching for DNA genomic sequences available on public online databanks. They investigated the Longest Common Subsequence (LCS) algorithm and using a standalone computer, they have used dynamic programming (DP) to compute the LCS length with two different approaches: Bottom-Up (Tabulation) and Top-Down (Memoization). Even if the Tabulation approach showed better results compared to the Memoization approach, both approaches are still computationally intensive and led them to try to adapt the LCS problem to the parallel processing using Hadoop.
One more task they did using Hadoop framework is retrieving putative gene regions from DNA sequences.
In the next three months, we will start investigating the use of machine learning algorithms to analyze and understand some of the Coronavirus properties.
Back to PEER Cycle 5 Grant Recipients