Skip to main content

Agent-Based System for Enhanced Controlling and Monitoring of a Solar Driven DC Microgrid

Time: Tue 2021-10-26 09.00

Location: 1330,, Greta Woxen, Teknikringen 31, KTH, Sweden, Stockholm (English)

Subject area: Information and Communication Technology

Doctoral student: Diana Rwegasira , Elektronik och inbyggda system

Opponent: Professor Josep M Guerrero, Aalborg University, Denmark

Supervisor: Professor Hannu Tenhunen, Elektronik och inbyggda system; Professor Anders Hallén, Elektronik och inbyggda system; Associate professor Masoumeh Ebrahimi, Elektronik och inbyggda system


The early efforts in grid computing started as a project to link supercomputing sites, but have now grown far beyond their original intent. This has led to the introduction of smart systems such as smart grids for power systems, and e-health systems for the health sector. The smart grid system has provided many solutions with regards to energy management, energy efficiency resources, accessibility of power supply, controlling and monitoring of the electric system, etc. An agent-based system has been seen to work successfully in electric sectors which involve the use of artificial intelligence and autonomous actions during the operations and controlling activities. These significant facts of the agent-based system have led to the design and implementation of smart systems to reduce human interventions.

 In this thesis, controlling and monitoring are the significant aspects of the provision of an autonomous system in the upper layer of the microgrid. The proposed approach involves the use of the solar-driven DC microgrid as a source of power with a capacity of 48V DC. An agent-based control is a technology used for deploying the load shedding technique with a demand response scheme, which enhances the sharing of power among individuals. The target usage is for low voltage appliances such as lighting and charging activities of low power consumption devices, which can be applied to the areas without access to electricity.

 Firstly, the analysis of the existing applications, simulation platforms, and algorithms in the agent-based system was done. Secondly, the simulation model based on solar DC microgrid was developed for critical and non-critical loads. It involves three main agents which are: - solar agent, storage agent, and load agent. This was further extended to include the demand response scheme for the pricing algorithm to provide flexibility in the model. The model runs with different scenarios such as static loads, dynamic loads, and also provides the mechanism to integrate with the agent-based hardware platform. Using the simulation model, the requirements for designing and implementation the DC microgrid was deduced. Finally, the deployment of the prototype was done to demonstrate the control and monitoring aspects using an agent-based system. The test-bed is designed with low-cost and high-performance devices which concluded the best usage of the agent-based system. The performance of the proposed solar DC microgrid was analyzed using different aspects, including energy cost, and scalability. The advantages and impact of this research can be applied in any community for the application of various renewable energy systems.