by July 19, 2013 0 comments

Reliance Infrastructure provides power to 2 out of 3 households in its Mumbai distribution. Its power distribution is spread across 384 kms. Reliance Infrastructure’s total power requirement stands at about 1500-1700 MW, while its total installed capacity at its Dahanu Generating Power Station is 500 MW. The balance power deficit of about 1000-1200 MW is fulfilled by getting into long term and short term Power Purchase Agreement (PPA) with power generation companies and the national grid. The Govt. of India has introduced the National Electricity Policy and advised all states to introduce Availability Based Tariff for intrastate power supplies. This encourages grid discipline, economic load dispatch, helps promote trading of energy and capacity, and encourages higher availability.

[image_library_tag 631/81631, alt=”reliance-infrastructure” ,default]

As a part of this policy, the Maharashtra Electricity Regularity Commission has to declare the power requirements by every distribution company 24 hours in advance. If the power distribution companies over draw power without advance intimation, heavy penalties are levied to the tune of Rs. 10 per unit. They’re also penalized for under drawing power.

Hence, it was necessary for Reliance Infrastructure to forecast its load requirements as accurately as possible. This was easier said than done.

Unfortunately, Reliance Infrastructure followed a manual process of capturing consumption data at the primary distribution points. Moreover, this process did not consider factors like weather conditions, holidays, humidity and how they would affect consumption.

This manual capturing of meter readings led to inaccurate meter readings from its 182 substations, which resulted in heavy penalties for the company.

Company Scenario

Before deployment: Reliance Infrastructure followed a manual way of predicting the power requirement it needed in order to draw electricity from the national grid for its 2.7 million customer base. This led to incorrect predictions, which attracted heavy penalties from the govt.

What was deployed: The company implemented automated meter reading for its 182 metering nodes, which collected data every 15 minutes. It also implemented a weather conditioning monitoring tool, a real time monitoring system to analyse energy data in real time, and a strong communication architecture to feed all this information into a SAS based load forecasting system.

After deployment: Accurate load forecasting with 98% accuracy helped the company improve financial efficiency by about Rs. 30 Cr. It achieved an estimated increase in revenue by Rs. 38.8 Lakhs due to improvement in reliability indices. The company also saved manpower cost and operational expenses by Rs. 10.4 Lakhs.

Implementation partner: In-House

Project Head’s Perspective: Govind Murlidhar Samant, GM

How did you convince the key stakeholders (management, seniors, etc) for rolling out this project?

Reliance Infrastructure Ltd has a vision to be amongst the most admired and trusted integrated utility companies in the world, delivering reliable and quality products and services to all customers, with international standards of customer care. It is R-Infra’s mission to be a technology driven company and promoting innovation for customer satisfaction. In order to fulfill its vision and mission, R-Infra has been setting newer benchmarks for other utilities by deploying the state-of-the-art technologies and using them in innovative ways to satisfy the changing needs of its customers and maintaining a sustainable competitive edge over its competitors.

During review meetings the pain points highlighted by SCADA and business were discussed and accordingly a core team was formed to develop a comprehensive solution with representatives from Business, SCADA, O&M and ABT team and IT. The core team formations helped in better understanding of the pain areas and joint development of a solution addressing all concerns of critical stakeholders from conception to actual deployment.

How did you overcome user resistance for using this deployment?

R-Infra’s IT department has a PDCA (Plan-Do-Check-Act) process based approach while initiating any major IT deployment. The critical stakeholders were taken on board through core team formation and the detail findings of the group are then internally brainstormed through various quality tools like why-why analysis, Fish-bone analysis. The process has helped the organization immensely as all concerns of stakeholders are addressed during the conceptualization phase and accordingly module-wise IT deployment was initiated. After each module’s deployment, training was initiated so that solution was used by the concerned users and the teething issues are resolved on priority. The PDCA approach for solution deployment helped in continual improvement.

What is the next big IT project that you’re working on?

R-Infra has recently bagged 2 most prestigious IT deployment projects under R-APDRP initiatives of Ministry of Power (MoP)

– Chhattisgarh State Power Distribution Company Ltd (CSPDCL)
– Bihar State Power Holding Company Ltd (BSPHCL)
R-Infra shall be deploying the following 17 modules across the 2 states comprising of 91 towns
1. Meter Data Acquisition, 2. Energy Audit, 3. GIS based customer indexing and asset mapping, 4. GIS based integrated network analysis module, 5. New Connection, 6. Disconnection and Dismantling, 7. Customer Care Services, 8. Management Information System & Business Intelligence, 9. Web Self Service, 10. Identity and Access Management, 11. System Security, 12. Metering, 13. Billing, 14. Collection, 15. Asset Management, 16. Maintenance Management, 17. Document Management System
R-Infra’s entire IT team is arduously working for successful IT deployment across 2 states comprising of 91 towns.


Automated Meter Reading System: The company setup an automated meter reading system for its 182 sub-stations located across suburban Mumbai, which took meter readings every 15 minutes. The company has developed a software to capture all this data and collate it into a central location from where it’s fed into various tools for further analysis.

Weather Condition Monitoring Tool: The company has also commissioned two Automated Weather Stations (AWS) in its supply area. These communicate weather data to a FTP server by using a modem. These stations send several weather parameters, which include temperature, humidity, rainfall, wind speed, and wind direction to a weather vendor. This vendor then forecasts temperature, humidity, and rainfall by email back to R-Infra. A Business Application Programming Interface has been developed to capture this data and feed it into a database server.

SAS Solution: This weather data along with energy parameters captured by the AMR Server are fed into a SAS based solution for further analysis. Apart from data from these two systems, the SAS solution also takes other data as input. This includes allocated demand, customer forecast, contract data, zone details, and special events like holidays, festivals, strike, etc.

This solution provides an accurate load forecast, which is then sent to the State Load Dispatch Center (SLDC), which then commits the power requirement. Any deviation from this load requirement attracts severe penalties.


Real Time Monitoring Solution: The company has also developed a real time monitoring system that analyses energy data on a real time basis and analyses it against the forecast provided to SLDC. The system studies deviations against the forecast so that the company can take corrective measures in order to avoid paying heavy penalties.

Communication Architecture: All data from the primary distribution points is send to the AMR server located at Dhirubhai Ambani Knowledge City. There’s a secondary redundant link in case the primary link fails.

Challenges Faced

There were quite a few challenges that R-Infra faced while deploying this solution. First challenge was to integrate such complex and disparate data into a single solution. There was no historical consumption data to analyse power consumption trends for every 15 minutes. There was no historical weather data available either. Lastly, there was the challenge of integrating this solution with other technologies like GIS, SCADA, and SAP.

[image_library_tag 632/81632, alt=”project-architecture” border=”0″ border=”0″ border=”0″ border=”0″ border=”0″ ,default]

The Impact

Due to this improved forecasting methodologies, the System average Interruption Duration Index has improved by 83%, while the System Average Interruption Frequency Index has improved by 11%. The Average System Availability Index has improved by 25%. Financially, the company would incur a penalty of 10 Cr for 1% deviation in load forecasting. The system has helped the company improve its forecasting from 95% to 98%, which is a 3% improvement, or about Rs. 30 Cr of annual financial efficiency. Further, the company has increased its revenue by about Rs. 38.8 Lakhs per year due to improvement in reliability indices. It has also saved about Rs. 10.4 Lakhs per year in manpower and operational expenses.

Due to elimination of manual processes, the manpower is being optimally utilized for services to the customer. According to Reliance infrastructure’s statistics, optimizing power purchase due to better forecasting has saved them approximately Rs.30 Crores.

This is a fine example of how analytics should actually be used. The company has successfully managed to collect huge volumes of data from disparate systems and put it into a central analytics server automatically and show the result in real time.



No Comments so far

Jump into a conversation

No Comments Yet!

You can be the one to start a conversation.