# Title Team Members TA Documents Sponsor
15 Automated Self-cleaning Solar Panel
Area Award: Solar Energy
Terry Green
Yann-Tyng Lin
Yousaf Abdul Salam
Our project is to design an automated solar panel cleaning mechanism that detects obstructions (bird droppings, leaves, etc.) on the panel and automatically cleans them away. Since solar panels are outdoors, they are prone to shading due to natural debris. For large solar installations in remote areas it could become difficult and/or expensive to monitor and clean hundreds of large solar panels. Because of the build-up of dirt on a cell in a solar module, hot spots are created and heat is produced instead of electrical power. This reduces efficiency and life-time of the solar panel. Our project would greatly reduce the amount of service calls for cleaning, resulting in lower operating costs in the long run. The project will be implemented using current and voltage sensing on individual solar cells in an array to provide performance feedback that determines if a cell has become shaded. Cloud shading will be taken in to account by programming and observing whether a whole group of solar cells have become shaded or just isolated cells. We will be adapting the car windshield cleaning mechanism for optimal cleaning (direction, speed, frequency). A soap/water mixture will be sprinkled during the cleaning cycle. Periodic cleaning will be programmed to occur in intervals based on the tradeoff between power consumption and cleaning effectiveness. For example, solar panels located in areas with high bird populations would require more frequent periodic cleaning. Our project will incorporate power electronics, microcontroller programming and control systems for motors.

Control System and User Interface for Hydraulic Bike

Iain Brearton

Featured Project

Parker-Hannifin, a fluid power systems company, hosts an annual competition for the design of a chainless bicycle. A MechSE senior design team of mechanical engineers have created a hydraulic circuit with electromechanical valves, but need a control system, user interface, and electrical power for their system. The user would be able to choose between several operating modes (fluid paths), listed at the end.

My solution to this problem is a custom-designed control system and user interface. Based on sensor feedback and user inputs, the system would change operating modes (fluid paths). Additionally, the system could be improved to suggest the best operating mode by implementing a PI or PID controller. The system would not change modes without user interaction due to safety - previous years' bicycles have gone faster than 20mph.

Previous approaches to this problem have usually not included an electrical engineer. As a result, several teams have historically used commercially-available systems such as Parker's IQAN system (link below) or discrete logic due to a lack of technical knowledge (link below). Apart from these two examples, very little public documentation exists on the electrical control systems used by previous competitors, but I believe that designing a control system and user interface from scratch will be a unique and new approach to controlling the hydraulic system.

I am aiming for a 1-person team as there are 6 MechSE counterparts. I emailed Professor Carney on 10/3/14 and he thought the general concept was acceptable.

Operating modes, simplified:

Direct drive (rider's pedaling power goes directly to hydraulic motor)

Coasting (no power input, motor input and output "shorted")

Charge accumulators (store energy in expanding rubber balloons)

Discharge accumulators (use stored energy to supply power to motor)

Regenerative braking (use motor energy to charge accumulators)

Download Competition Specs:

Team using IQAN system (top right corner):

Team using discrete logic (page 19):