Smart Cities and Smart Grid

Indian government has announced an ambitious plan of setting up hundreds of smart cities across the country. A critical and important part of a smart city is to have an efficient electrical infrastructure, often termed as a smart grid.

Little motivation is required for converting our electrical grid to a smart grid as a smart grid can provide significant improvement in electricity transmission, distribution and consumption. Currently, the T&D losses in India stands at approx. 30% as compared to an average of less than 10% in the developed countries. In order to reduce the high losses, it is imperative to do real time monitoring at different levels from distribution transformer, to substations and eventually at the consumer premises.

A major challenge in electricity distribution is the management of peak load. Variations in the electricity consumption at different times of the day result in demand peaks. The distribution companies need to have enough supply to cater to these peak demands. However, this means a lot of electricity gets wasted during the off peak hours. Therefore, efficient demand side management and accurate peak demand prediction is crucial to ensure 24×7 electricity supply without much wastage during off-peak hours. Further, as renewables (primarily solar in the context of smart cities) are added on the supply side, there will be much higher fluctuations on the supply side as compared to the current situation where primary dependence is on thermal and hydro power plants that have little daily variability.

User (consumer) engagement will become a critical piece in the smart grid. Currently, the user is passive; getting a monthly bill that gives little information about inefficiencies in their daily operations and how they may save on their bill. Several studies show that effective feedback on consumption can itself result in more than 15% reduction in consumption.


The above figure illustrates different components of a smart grid for a smart city. Some of these components are explained below:

1. Real Time Monitoring and control: The monitoring devices (monitoring energy and other related parameters) need to be installed all the way from end consumer, through different distribution networks to the eventual source point where the electricity is fed in. These devices should be able to communicate the collected data in real time (often faster than once every minute) to a cloud service for monitoring and analysis. Monitoring by itself is not useful unless accompanied by efficient control. Devices spread through the distribution network will also provide capabilities for remote (web/mobile based) control which will be useful for demand management.

2. Cloud based data management and energy analytics: Cloud will play a big role in data aggregation from large number of sensors and efficient management of collected data such that several applications can be built using the collected real time data

3. Building Energy Optimization: Typically buildings (both residential and commercial) consume more than 40% of overall energy consumption in a city. Within the buildings, Air Conditioning consumes about 60% of the total energy consumption. Real time monitoring that can provide feedback to users on optimal usage of air conditioning (e.g. when to turn the AC on and off and at what temperature with their corresponding impact on energy consumption) have a huge potential towards energy savings.

4. On-site generation and optimal energy management: Providing grid power for 100% and 24×7 consumption will never be feasible through large power plants. Many building owners are already moving towards solar installations. These buildings also have a Diesel generator backup and UPS systems with battery storage. A system that can optimally schedule between the grid power, diesel generator, solar power and battery backup depending on different environmental conditions, grid availability and demand can help manage the peak load effectively and reduce the overall burden of electricity cost on the end consumer.

5. Demand Side Management: Consumption patterns are typically very dynamic across different times of the day. In order to achieve 24×7 electricity availability, one has to cater to the peak demand. Correspondingly, during non-peak hours, significant amount of electricity gets wasted. If the peak can be flattened, it can result in easier and cost-effective path towards 24×7 availability. Real time monitoring and control, together with analytics in the cloud can help utilities perform several demand side management programs such as Demand Response System and work towards peak flattening.

6. Supply side management: As renewable sources of energy, both from on-site generation and smart micro-generators, gets added to the grid, the variability on the supply side also increases, depending on weather conditions. Similar to demand side management, supply side management is also required. Real time monitoring and cloud analytics can again play a big role here.

7. Involve residential consumers: Often the biggest consumers in a city are the aggregate residential homes. These are the consumers that have little insights into their consumption pattern but biggest motivation to reduce (as the prices pinch them the most). Real time monitoring and cloud analytics can again play a major role in providing such users with effective feedback on their consumption e.g. disaggregating their bill to provide electricity consumption for all major appliances separately, comparing their consumption with their neighbors (or people of the same apartment size). Such feedback can result in faster replacement of energy inefficient appliances and optimal usage, both resulting in significant reduction on the consumption side. There is also an opportunity to use the home inverters for demand side management without incurring much infrastructure cost.

Data privacy and security: As large volumes of data gets collected, it is important to put in place policies around data privacy i.e. who owns this data and data security i.e. who gets access to the data, at what time and at what resolution.