Project Info

AI-based Sensorless Control of DC Wind Power System for Reduced Failure

Omid Beik
beik@mines.edu

Project Goals and Description:

Wind power with a capacity of 135 MW contributes to 10% of electrical energy in the United States, while 28% of Colorado’s electricity is supplied by wind turbines (WTs) scattered across the state. The existing wind farms use alternative current (AC) for generation, transmission, and distribution of electrical power, see Fig. 1. Direct current (DC) technologies have gained much attention because of their high performance, controllability, resilience, efficiency, and adaptability to the variable nature of wind power. A schematic of an all-DC wind power system that uses medium voltage DC (MVDC) collector grid and high voltage DC (HVDC) transmission grid is shown in Fig. 2. This system uses a model-based approach with sensors to monitor and control the wind power system. The model-based methods rely on the system detailed parameters, therefore, in the event of a fault or sudden load changes their effectiveness is reduced. To enhance resiliency of power systems non-model-based methods such as model predictive, fuzzy logic and artificial intelligence (AI) controls have been proposed. These approaches use information from sensors for monitoring and control. A fundamental issue with sensors is that they require maintenance and regular calibrations to maintain accuracy of collected data. Inaccurate sensor information results in incorrect controls exerted on the power system, which leads to instability and failures of components and systems. To overcome these challenges, this project proposes an AI-based sensorless monitoring and control approach. Feedforward artificial neural network (ANN) will be used as a subset of AI, and the proposed approach will be applied to the PI’s proposed all-DC wind power system via two controllers, (i) a local controller for the WT, and (ii) a supervisory controller for the MVDC grid.

More Information:

Grand Challenge: Engineer the tools of scientific discovery.
Please visit our website: https://www.epowerhubs.com/publications

Primary Contacts:

Dr. Omid Beik, beik@mines.edu

Student Preparation

Qualifications

  1. Student should have a sound background in math and physics.
  2. Familiarity with electric circuits, analyses of components and elements in a circuit.
  3. Familiarity with basics of power systems, components, and analyses.
  4. Passion for renewables, and power systems.

TIME COMMITMENT (HRS/WK)

4

SKILLS/TECHNIQUES GAINED

  1. The student will learn about wind generation systems, AI, environment, conditions, and challenges.
  2. Student will gain knowledge of high-voltage systems.
  3. student will gain knowledge of power electronics.
  4. Student will learn new simulation software.
  5. Student will gain hands-on experience working in the lab.

MENTORING PLAN

Student will be provided with a space in the PI's lab where he can work and interact with other lab members. Student will have access to the resources in the PI's lab. Here are some elements in my mentorship:
  1. Clear Communication: To ensure that students feel comfortable reaching out with questions, concerns, or ideas at any stage of the project.
  2. Setting Expectations: I will establish clear expectations regarding goals, deadlines, milestones, and the overall scope of the project.
  3. Regular Check-ins: I will schedule regular check-in meetings to assess progress, provide feedback, and address any challenges or obstacles that students may be facing.
  4. Providing Guidance and Resources: I offer guidance and support to students by sharing relevant resources, suggesting strategies for problem-solving, and offering insights based on my own experiences.
  5. Encouraging Independence and Critical Thinking: I encourage students to take ownership of their projects and develop their problem-solving skills. I challenge them to think critically, explore alternative approaches, and consider the implications of their decisions.
  6. Feedback and Reflection: I provide constructive feedback on students' work to help them identify areas for improvement and build upon their strengths. Additionally, I encourage students to reflect on their progress, identify lessons learned, and consider how they can apply their newfound knowledge and skills in future projects.

Preferred Student Status

Sophomore
Junior
Senior
Return to All Project Page
Share This