
“In God We Trust, All Others Must Bring Data.”
- W. Edwards Deming
Energy infrastructure is the backbone of an economy, and it is no wonder that lack of timely, actionable insights through powerful energy analytics can be crippling. However, energy data comes in myriad forms and the marriage of energy and data science is no simple affair - in fact, traditional analytics is simply not powerful enough predict nation-wide energy behaviour at both macro and granular level.
FORMCEPT’s open cognitive application MECBOT solves energy problems of the 21st century with highly efficient and reliable big data analytics, seamless, automated pipeline of insights and timely analysis of energy planning outcomes. The following are a few exciting applications of MECBOT in the Energy and Power domain that we have been working upon recently:
Energy Forecasting for Petroleum - The Liquid that Boils the World
The shift towards uniform tax regime in India took a bold turn in July, 2017 when the Goods and Services Tax (GST) became effective in the country. But in a much-debated move, petroleum commodities like petrol and diesel oil, ATF (aviation turbine fuel), motor spirit and natural gas have been kept out of its purview.
Recently, a Global Fortune 500 company that is also a leading manufacturer and supplier of petroleum products in India, wanted to analyze the potential impact of GST and consequent price increase of petroleum products across different cities in India, if and when the same is implemented by the Government. The company partnered with FORMCEPT to carry out energy prediction and energy forecasting for different scenarios across different zones in the country. FORMCEPT deployed its flagship product MECBOT - a unified data analysis platform - to conduct predictive modelling on past energy consumption data of these regions using a combination of AI and cognitive application for big data analysis.
By training MECBOT to ingest and digest large chunks of energy data and by mapping it to the universal knowledge base specific to energy sector, FORMCEPT analyzed the following aspects and visualized the results for better strategy formulation:
GST and price elasticity of petroleum products across select regions
The foremost consideration in devising a strategy for price inflation and GST is to understand how it affects the consumption of fuels across different regions in a given time period. This is important for two reasons - 1) oil and gas being an essential commodity, could cause shock-waves across the economy if its consumption changes drastically. 2) a considerable chunk of the Government’s tax revenue comes from petroleum products:
Thus, with advanced predictive modelling and time-series analysis, MECBOT answers questions such as - by how much percentage would the consumption of diesel in Kolkata increase or decrease, if 5% GST is applied? How does it change if, let’s say, 10% GST is levied? And hence, what is the optimum GST rate that can be applied to cause minimum possible disruption in consumption?
Cross-price elasticity of petroleum products relative to each other
While price elasticity indicates the expected change in demand for a given petroleum product over a period of time, cross-price elasticity pits different petroleum products against each other to identify relative effects on demand, such as - if the price of petrol shoots up, how would the demand of diesel be affected?
According to a recent report, for example, the demand for piped natural gas (PNG) in industrial and commercial sectors is expected to fall as fuel, since the effective tax rate on competing liquid fuels gets reduced from 26-28% to 18% and consumers paying VAT on PNG are not eligible for input tax credit as most of the finished goods their consume come under the purview of GST.
Social and economic impact of price rise and GST across select regions
FORMCEPT applied advanced cause-effect algorithms to multiple energy consumption variables across sectors to determine the possible impact of a strategic levy of GST and subsequent price hike. A case in point is this recent article by Economic Times which highlights the effect of price hike of crude oil and a few other raw materials on consumer prices in November, 2017. It is projected that that prices of consumer electronics such as ACs, fridge, washing machine, could go up 3-5% starting November, while air travellers might have to pay up to 15% more for tickets. Airlines expect a 10-15% hike in prices due to steep rise in aviation turbine fuel (ATF) prices.
Smart Profiling of Street Lamps for Dynamic Energy Saving
India is reported to have 35 million street lights culminating into an aggregate demand of 3,400 MW. The current one-size-fits-all approach to street lighting has led to inefficient utilization of electricity as a resource. Further, lack of automation has caused vulnerability to human errors leading to electricity wastage. FORMCEPT has recently demonstrated an innovative, smart, street-light profiling solution that adjusts luminosity of street lamps based on the incidence of sunlight in a given area at different points of time in a day, also accounting for seasonal and other climatic variations.
The system is capable of interacting dynamically with light-dependent resistors which can automatically switch off lights or reduce the luminosity based on sunlight conditions. Further, MECBOT can be also be tuned to event-based control - such as a citizen reporting lack of street light in an area, or unnecessary usage of street lamps when adequate sunlight is available.
Energy Mapping with Intelligent Solar Water Pumps in Telengana and Chattisgarh
The solar water pumps in Telengana and Chattisgarh is an interesting case in point. These pumps, mostly used by farmers, communicate with FORMCEPT cloud through MECBOT, providing data such as on-off status, volume of water pumped, temperature of water, health of pump, location of pump and anomalies in usage. This data can be accessed on a real-time basis by manufacturers and service providers through dedicated user portals.
The data provided by MECBOT is not only intelligent, but also highly actionable - for example, earlier engineers had to do routine checkups on pumps, while now an engineer does servicing only when a pump is out of order which automatically generates system alerts, leading to enormous cost savings. Inherent to the system is machine learning which learns from repeated usage patterns and links them to possibilities thereby providing predictive management. For example, if a farmer pumps out several times more quantity of water than he usually does, the system will prompt the viewer to check for malpractices such as a farmer selling water to other farmers, or common errors such as a farmer forgetting to turn off the motor. Life is also made easier for farmers as they can turn off and turn on the pump through sms-based remote-sensing tools.
Intelligent Asset Monitoring for Energy Management - Telecom Tower Batteries
It is estimated that on an average 70% of the approximately 0.4 mn telecom towers in India have to be buffered against electrical grid outages for over 8 hours per day. Telecom tower operators currently rely on diesel generators, batteries, and an assortment of power management equipment to bridge this demand-supply gap. The consequent energy costs contribute to nearly 25 percent of the total network operating costs, thus adversely affecting profitability of the operators.
A typical telecom tower has a rack of batteries assigned to operators. In the case of a power-cut, the standard practice is to run on battery for a few hours and then switch to diesel generators. However, fraudulent behaviour on part of the operators often results into overuse of the backup battery. The power used is then falsely depicted to have been generated from diesel generators, while the equivalent quantity of diesel is sold off illegally. The batteries, which come with 2-year replaceable warranty, are then reported to be ‘faulty’ and are requisitioned by the operators to be replaced. For a long time, inability to monitor the situation intelligently on a real-time basis has led to reduced battery life and increased cost to battery suppliers.
FORMCEPT addresses the crisis using their intelligent asset monitoring system. An IoT based device attached to the batteries shares real-time data with MECBOT in the cloud. The data includes status of battery (use status - on / off, charge status - overcharged / undercharged), battery health (dipping / optimal), etc. Whenever a battery dips and needs attention, an sms alert is sent to the field operator, and a system notification to the battery manufacturer as well as the R&D team. The device is equipped with machine learning ability which enables prediction of lifecycle of batteries, reduces pilferage and enables better purchase and servicing decisions, thus saving costs and enabling optimum asset utilization.
At FORMCEPT, we are constantly innovating cutting-edge solutions with state-of-the-art-technology to solve the pressing problems of today and tomorrow. The above use-cases are just a few examples of how our solutions are changing lives and influencing mission-critical decisions. To know more about what we do and how we can help you, visit www.formcept.com or drop a line at contactus@formcept.com.