Non Intrusive Load Management System

Project Title

Non Intrusive Load Management System

Project Area of Specialization Internet of ThingsProject Summary

Our project brings the implementation and application of Non-Intrusive Load Monitoring (NILM) into demand response (DR). For designing and implementing an intelligent energy conservation system, it is essential to have insight into the activities and actions of the occupants. Normally it is done by implanting of sensor to every load of interest which makes the system to be costly and time consuming. In contast to that, NILM system is simple to install and does not require an individual sensor for each load. The basic concept behind the project is to distinguish each load type (i.e. identification of each appliance) from a single load on main distribution board. The distinguishing of loads can be done on the basis of their unique signatures (i.e. active power, reactive power and total harmonic distortion etc). The collected will be processed through optimiation technique (machine learning algorithm) and IoT plays its role in transfering data to user devices. Our aim is to present a gadget capable of monitoring as well metering of load of every second. This will help consumers to adjust their habits in order to lower electricity bill. The project can be extended to load forecasting which will enables the utility company to plan well since they have an understanding of the future consumption or load demand.

Project Objectives

• appliances’ disaggregation/identification non-intrusively
• load monitoring 
• a powerful energy management and measurement system
• cost effective
• to determine the energy consumption pattern of consumer
• reliable and less complex in terms of circuitry

Project Implementation Method

With a proper algorithm, the NILM system can identify various loads and the status of loads, including electric power demands, type of appliances, time of use, overload capacities of loads, etc.

We will select one of the best suited metaheuristic algorithm(probably honey bee algorithm). The algorithm will classify each load on basis of their unique signatures.

Benefits of the Project

• The NILM provides the appliances’ disaggregation/identification non-intrusively, which means no need of entry into the building before giving account on the appliances, hence the NILM method takes into consideration the privacy of the building’s owners.
• The single point metering of the NILM method also makes it reliable and less complex in terms of circuitry when compared to ILM where the system has to communicate to every appliance.
• This type of metering also makes the NILM to be cost effective, as only one set of sensors is required for the whole system.
• The non-intrusive monitoring also has the blessing of other professions like fluid/water engineering in which the flow is observed non-intrusively using sensors.
• This method also provides the opportunity to do forecasting of individual appliance of a place.
• The energy consumption pattern of consumer can be determined for further planning and management of load demand and load generation.
• One can also detect fault before its occurrence by detecting abnormal changes in signatures of loads.
• It can be proved as a beneficial gadget for future smart meters

Technical Details of Final Deliverable

Final deliverables of project will be a smart meter able to monitor load(non-intrusively), metring of energy consumed, and forecasting of future load demand and it's cost. The every second data of energy consuption will be send to user devices(smartphones and laptops etc) using IoT. While the gadget will be implanted on main electrical distribution board.

Final Deliverable of the Project HW/SW integrated systemType of Industry Energy Technologies Artificial Intelligence(AI), Internet of Things (IoT), OthersSustainable Development Goals Industry, Innovation and Infrastructure, Responsible Consumption and ProductionRequired Resources

Item Name	Type	No. of Units	Per Unit Cost (in Rs)	Total (in Rs)
			Total in (Rs)	67500
Power Analyzer	Equipment	1	40000	40000
Raspberry pi	Equipment	1	8000	8000
Voltage and current sensors(CTs & PTs)	Equipment	2	3000	6000
Bulbs, wires	Equipment	2	1000	2000
PCBs	Equipment	2	1000	2000
LCD (for display)	Equipment	1	1000	1000
Main DB box	Equipment	1	2000	2000
Stationary	Miscellaneous	1	2000	2000
Poster	Miscellaneous	1	2000	2000
brochures	Miscellaneous	50	50	2500