Fahad Javed
Department of Computer Science
LUMS School of Science and Engineering

Lab 9-126, LUMS,
Opposite Sector U, DHA, Lahore,
Pakistan. 54792

SMaRTI: Self-Managing Renewable Technology Integration

Tools Deployment

SMaRTI is the exciting new cyber-physical system designed to integrate renewable technologies with energy management systems for optimal and maximal usage of renewable energy. Specifically it integrates the solar panel generating units with smart office infrastructure we have been working on. The goal is to reduce the overall cost of the system and increase the throughput of energy.

How do we do this? The main problem in sizing an electric system is peak power - the maximum demand of the system. Systems are designed to provide for this peak power. But for most part of the day the demand is way below the peak power be it a city or an office.

When an office sets up a renewable system then they plan for the worst scenario, minimum generation of electricity through renewable and maximum demand of the system. Although it is very likely that such an event can occur, the majority of system life does not require such extreme analysis.

Our contention is that we can mitigate the worst scenarios through software components and instead of planning for the worst case scenario plan for a drastically less costing average scenario. What we add to this system are software components to mitigate peak power and supplement low renewable generation. Through this supplementation we can drastically reduce the initial cost of investment which by the way is the biggest hindrance in widespread acceptance of renewable energy.

What we do is that we manage the demand. We set a limit to the maximum power based on generation forecasts. We limit the consumption of energy to this set demand by moving elastic loads around. So for example the current system limit is 600 Watts. We have 2 laptops, 1 desktop and 2 fans in this system. The total average demand is less than 600 Watts but if say the desktop is being used for heavy computation requiring so much power that the system demand goes over 600 Watts then we will move the laptop off the solar system onto its own batteries. Since the heavy computation will be done for a limited time, we will bring the laptop back on when the demand of the desktop stabilizes to its normal loads. Since the batteries of the laptop would have been drained, it will draw power above its normal demand. If the demand crosses 600 then we switch off the second laptop. The two laptops will eventually reach the steady state where the demand will be stabilized. If the demand is not stable by managing the laptops only then either or both the fans can also be switched off. In all this time if a desktop or laptop user steps out then the system stability will be reached earlier.

For setting the limit of the power we will observe and predict the solar generation for the next day. What we do on the supply side to reduce the cost is that instead of using batteries as an active component of system to provide energy, we restrict their use as stabilizing power sources. The energy drained from the system is limited to the average generation of energy from the solar panels so the net energy in the system remains constant. During partial shading due to clouds etc., the batteries will supplement the energy but only to the magnitude which can be recuperated over the period of a day. This reduces the depth of discharge, the critical value which determines the life of the battery. Since the depth of discharge will remain low, the life of the battery will be extended exponentially compared to deeper discharge of the battery by setting a system limit which does not consider net energy generation in a day.

Our hypothesis is that through this combination of hardware and software system, or a cyber-physical system, we can have a sustainable system with a smaller renewable and storage component reducing the cost and footprint of the system. SMaRTI has been just setup. We are planning a range of experiments, some of which are given below. If you have any ideas for evaluation or improvement of the system then drop me an email at fahadedupk@gmail.com

Validations and Verifications:

Total savings: Total savings in 1 year through this system, also payback period, ROI etc., will be calculated.

System Stability: How small a system can sustain the energy requirements of the system. Currently if our system can not sustain the energy requirements then we switch to utility power. However, in an islanded system this will not be pos0sible.

Maximum load/minimum generation/maximum spread: Statistics of generation and consumption and how are they distributed over 24 hour period for maximal utilization of resources.

Safe margins: Safe margins for offloading loads, battery charging etc. for optimal planning. This will also give us insight on optimal sizing of the batteries and panels.

Image Gallery

600Watt Panels for SMaRTI at SBA-SSE LUMS building
Screen shot of the visualization of solar usage and utility usage through our tool PCAT.
System Architecture of SMaRTI.
Communication Architecture of SMaRTI.