+ All Categories
Home > Software > 7 Ways to unlock value from Smartmeter Big Data

7 Ways to unlock value from Smartmeter Big Data

Date post: 18-Oct-2014
Category:
View: 510 times
Download: 4 times
Share this document with a friend
Description:
The utility industry is undergoing a fundamental transformation with increased digitation and tighter coupling between IT and OT. Flutura outlines 7 ways by which utilities can monetize smartmeter data
11
WAYS TO UNLOCK VALUE FROM SMART METER BIG DATA 7 Jobil Louis, Allen & Swapnil Flutura Decision Sciences & Analycs
Transcript
Page 1: 7 Ways to unlock value from Smartmeter Big Data

WAYS TO UNLOCK VALUE FROM SMART METER BIG DATA7 Jobil Louis, Allen & Swapnil Flutura Decision Sciences & Analytics

Page 2: 7 Ways to unlock value from Smartmeter Big Data

Technology shifts occur in history periodically and change the rules of the game. It is Flutura's belief that

Machine 2 Machine (M2M) & Big Data Analytics are two such phenomenons which are profoundly disrupting

business models globally. M2M + Big Data Analytics offer revolutionary opportunities by harvesting behavioural

patterns which were previously not visible and provide breakthrough answers to powerful questions.

The Utility Sector for example, is ripe for unlocking energy efficiencies. This can be done by understanding the

energy consumption habit patterns at a level of granularity which was previously not possible – neighbourhood

& consumer level. Furthermore, it also enables you to reduce technical and commercial losses along the

complete grid value chain. Flutura would like to present an outline of a 7 point framework to unlock value

embedded within ‘Smart Grid’ data.

Outage Predictor

Distribution Transformer

Interventions

Consumer Energy Habit Gami�cation

TOU (Time of Use)Dynamic Pricing Models

GuzzlerMicro Segmentation

Extract DeviceSignatures

Bottom up EnergyDemand Forecast

1

2

3

45

6

7

Page 3: 7 Ways to unlock value from Smartmeter Big Data

The last three years have seen a paradigm shift in the increase of data points (by using more instruments) resulting in a sudden data avalanche for the Utility sector. This has been driven by two waves and Utility companies need to make sense of it. In the first wave, as Smart meters proliferate, Utilities have to process data at 15 minute intervals which is a 3000 fold increase in daily data processing. In the second wave, as the number of SCADA devices which are metering energy flow throughout the grid (like substations, transformers and other elements of the distribution systems) increases there will be a next level of data explosion. The massive release of data from Utility grids has profound implications for the industry as it opens up a huge set of possibilities to monetize these massive grid data pools.

ROI on money already investedxxx dollars are spent on smart grid investments. One of the challenges is to demonstrate the value unlocked from smart meter data.

Questions Utility Companies are asking• How can we monetize the massive smart grid data which has been collected from millions of customers spanning billions of events at intra hour level?

• How can we see energy consumption patterns not seen before and drive last mile changes?

Smart grid analytics is a graceful blend of art and science and success is possible if one harmonizes these two dimensions.How can one unlock value from massive Smart meter investments using analytics?

1

Energy economics from Guzzler segmentation

DATA

1 Energy Habits data

2 Consumer pro�les

3 Location data

TARGETED ACTIONS31 Targeted Energy Audit

2 RT SMS behaviour alerts

3 Neighborhood Gami�cation

4 Bill “what if” energy calculators

4

2 GUZZLER MICRO SEGMENTATION CEREBRA APP

MONETIZATION – IMPACT QUANTIFICATION

Page 4: 7 Ways to unlock value from Smartmeter Big Data

Context:Peak power demand is a frequent source of concern for utility. Before smart meters, the meter reading frequency was once a month. As a result one could not specifically pinpoint consumers who were responsible for ‘guzzling’ power. Now with Smart grid data one can have granular energy consumption patterns in an hourly on 15 min interval time frame. Now, it’s even possible to micro segment consumers based on the amount of power they consume, their deviation from baseline consumption, consumer type and location.

Unanswered questions• Are the numbers of ‘guzzlers’ increasing or decreasing with time?• Which segment type experienced maximum surge in ‘guzzler’ migration?• What is the change in habit between peak time ‘guzzlers’ and night time ‘guzzlers’?

Actions triggeredEnergy audits can be offered to ‘guzzlers’ to encourage them to optimize their equipment. For example, in the industrial segment, factories may have a large number of inefficient motor or pumps which have been habitually ‘guzzling’ a lot of power.

Smart meter value-1Pinpoint grassroots level neighborhood guzzlers

Blind spots during peak power when grid gets taxed

Page 5: 7 Ways to unlock value from Smartmeter Big Data

Price is an untapped lever in utility

Today, two large sectors like the Airlines and Retail have dynamic pricing. Why not consider the same in the Utility sector? What’s more, in order to sensitize people towards energy consumption, it imay be relevant to have ‘time-of-use’ pricing. For example, peak power tariff for industrial units would be different from peak power tariff for hospitals and government entities. Another opportunity is to increase the tariff for individual households who have two-sigma-variance compared to neighborhood baselines.

• Which households are responding to peak power price?• Do we need to have more pricing slabs?• How much should we recalibrate pricing to optimize energy usage?• What is the tipping price at which consumers become sensitive to energy usage?

Smart meter value-2Time of use pricing

Page 6: 7 Ways to unlock value from Smartmeter Big Data

Smart meter value-3‘Gamification’ of consumer energy habits

‘Gamification’ is about integrating gaming frameworks to alter energy habits of specific high value guzzlers and engaging them. For example, if in a neighborhood there are 2000 individual households and in the bill if one puts a big bold statement saying, "You are in the top 100 energy consumers in your neighborhood" or "Congrats you have altered your habits to climb-down from the top 100 list" or "Your change in energy habits has earned you 100 points which you can redeem at the local store". Consumers are creatures of habit and if their change of habit is benchmarked in the neighborhood and rewarded, their sensitivity towards peak power usage is highly likely to change. In order to put this into action Utility companies have to delineate target energy behaviors – say households not consuming above a certain average threshold during peak power and create activity loops when that pattern is detected.

Human beings are creatures of habit

Page 7: 7 Ways to unlock value from Smartmeter Big Data

Smart meter value-4Signature extraction & Habit design

There are individual devices within a commercial organisation or household which typically consume more energy than others – for example heaters, dish washers, etc. These energy intensive appliances can be put on a watch list and their consumption signatures detected. This consists of analyzing changes in the voltage and current going into a house from the smart meter time series data and inferring the specific individual energy consumption of appliances. The appliance signatures can be decoded by identifying patterns in the variation in measured power change each time an appliance is switched ON or OFF. Once these appliance signatures are detected, utility companies can provide tailored energy feedback in their bills to influence their habits.

• What are the devices in the appliance watch list?• What are the energy consumption signatures of these appliances in the watch list?• How do we direct the consumer’s attention to appliances or actions that have high energy saving potential?

Page 8: 7 Ways to unlock value from Smartmeter Big Data

Smart meter value-5Predictive models for preventative outage hotspots

Smart meter value-6Next best Distribution transformers interventionsThe harmonic distortion of current is increasing with the enhanced use of nonlinear loads from solid state devices. Examples of nonlinear loads are personal computer, laptop, laser printer, fax machine, television set (TV), fluorescent tube with electronic ballast, compact fluorescent lamp, battery charger, adjustable speed drives, uninterrupted power supply (UPS) and any other equipment powered by switched-mode power supply (SMPS) unit.

Strategic customers like hospitals, military establishments and political establishments are affected to a greater extent when an outage event happens. Now with machine learning algorithms one can decode patterns leading up to an outage event – brownout frequencies, transient voltage, step change in energyconsumption. Earlier when an outage occurred one could not learn from the patterns leading to it. For example:

CUSTOMER DATA

DTR DATA

OUTAGE EVENTS

METERING DATA

GRID NETWORKSTRUCTURE

TAMPER EVENTS

8 0 6 5 4 7 3

E L E C T R I C M E T E R

SMART METER INFORMATION LANDSCAPE

Page 9: 7 Ways to unlock value from Smartmeter Big Data

Smart meter value-7Reduce spot buying through bottom up forecasting

Load forecasting is currently a top down process which looks at historical patterns to predict future demand. It's a complex problem to be solved since energy is a "perishable" item. As a result over forecasted excess power procured cannot be "stored". At the same time under forecasting would result in last minute procurement of power which is extremely expensive. With the availability of granular data, neighborhood level energy profiles can be created based on individual smart meter data and then used to triangulate on the amount of power to be procured resulting in enhanced value.

These nonlinear loads draw more current than the fundamental current and cause overloading of the Distribution Transformers (DTR). This leads to higher losses, reduces the strength of insulation and subsequently leads to reduction of useful life of the transformer. Aging of transformer increases due to overheating caused by overloading. Current harmonics from Smart meter data can be used to identify aging of transformer caused by harmonics due to non linear loads. It's also compounded by the fact that many of the distribution transformers in the grid responsible for the last mile distribution of power have not been changed for many years. One can look at power harmonics data, brownouts, blackouts and transient event data to rank order and prioritize DTRs in specific neighborhoods where it needs to be replaced. In order to do this DTR master and event data must be collected and stored in a central smart grid event repository DTR 360:

• Which are the neighborhoods where the last mile DTR performance needs to be analyzed?

• Which segment of customers – industrial, household, strategic needs to be prioritized?

• Which events must be brought into Central Smart Grid Event Repository (CSGER)?

• What is the DTR scoring process we must deploy?

Nonlinear Loads Impact Aging Distribution Transformers

Page 10: 7 Ways to unlock value from Smartmeter Big Data

So can Utilities get started?Build Foundational Smart Grid Data Model

Bottom up demand response

Strategic Outage Hotspots

Outage FrequentSequence Analyzer

Time of Use (TOU) Dynamic Pricing

Distrb Transformer DTR 360

Device Signatures

Outage Events

Customer Data

DTR Data

Tamper Events

Grid Network Structure

Metering Data

100+ Energy Vectors

Signal Detectors

12 Core Energy Markers

Scoring Models

VEEDA

Apriori

Advanced Visualisation

Machine Learning Algorithms

Tame Big data ref arch

Guzzler Micro Segmentation

Cerebra Smart Meter Nano Apps

Cerebra Signal Studio

ConclusionThe Utility industry is facing an inflection point where technology is shifting to Machine 2 Machine (M2M) & Big Data Analytics and profoundly disrupting business models. Utilities must act and move quickly to respond to changes and leverage the advantages. M2M and Big data analytics offer immense opportunities for monetizing from investments in Smart meter infrastructure.

Page 11: 7 Ways to unlock value from Smartmeter Big Data

About FluturaFlutura is a niche big data analytics solutions company with a vision to help contain massive risk exposures for organizations and radically unlock operational efficiencies. It does this by extracting meaningful signals from data using Big Data Analytics. The name Flutura stands for butterfly; inspired by nature's greatest transformation from a caterpillar to a butterfly. We are obsessed with Trust and Transformation and align our daily lives to these core principles. Flutura is funded by Silicon Valley’s leading venture capital firm The Hive which primarily invests in big data companies worldwide. Flutura at a very early stage has been identified as among the Top 10 most promising big data companies by CIO Review, a leading analyst magazine.


Recommended