5_Conference ITF-Automation 2014

Download Report

Transcript 5_Conference ITF-Automation 2014

“Collaborative automation: water network and the virtual market of energy”,

an example of Operational Efficiency improvement through Analytics

Stockholm, ITF Conference, 6 th February 2014 Analytics for solution team, V. Boutin

Schneider Electric at a glance Customers are looking for integrated solutions that make their lives easier while optimizing costs. Innovation is essential to satisfying those requirements. The convergence of automation, information, and communication technology has created dramatic new opportunities for advancing energy efficiency. Innovation is about combining these opportunities with smart services to deliver high-value yet easy-to-deploy solutions. Pascal Brosset, SVP Innovation, Schneider Electric

24 billion € sales in 2012

41% of sales in new economies

140 000+ people in 100+ countries

4-5% of sales devoted to R&D

Digitization and Analytics bring new opportunities to improve Operational Efficiency

X 2

Increase of the volume of data every two years

1 Billion

Collective volume of data points being generated by Smart meters in the US every day

17 b$

Estimated total revenue for big data by 2015 (IDC)

Beyond basic KPIs

Opportunity to extract value out of collected data

Cloud

Big data storage and analysis across various information inputs

Analytics 3.0

In the new era, big data will power consumer products and services

.

by Thomas H. Davenport

What are Analytics ?

………..……..What will happen next?.............................

Modelling …..Why is this happening?......................

Analysis Notification Alerts ..………What action is needed?..................................... Query Drilldown …………..What is the cause of the problem? …………………….

Ad Hoc ……………..How many? How often? Where?.............................................

Reports Degree of Intelligence 2

7 Analytic features for Operational Efficiency to create new information such as prevision, patterns, early detection of problems to take better actions regarding organization, planning and control to provide rationale for building an optimized design and development strategy for the future

Data correlation & prediction Data Disagreggation & information discovery Performance evaluation & benchmarking Context dependent control Resources & activities planning and scheduling Condition monitoring, diagnostic, maintenance Decision support through simulation

Few concrete examples

Virtual or smart sensors

Get advanced information (such as fermentation for beer micro-filtration, or milk powder hulidity …) by collecting and mixing several correlated data items

Early detection of abnormalities

Extract early signals that would detect abnormal behaviours and possibly link to performance degradations

Demand response for water distribution

Determine the best srategy for pumping, while ensuring that the water demand will be entirely met, and leveraging variable energy prices (modulation market)

Technologies to make it happen 2

Analytics technologies Better control, supervision, operation management, design and continuous improvement

Analytics to OPTIMIZE Analytics to SIMULATE Physical models Analytics to MODEL Dynamic system modeling Pattern learning Pattern discovery Visual analytics

Data from

Pervasive sensors  Low cost  Self powered  Communicating  Easy to install Comfort sensor Energy sensor

Infrastructure for data collection and integration with heterogeneous applications and legacy systems Enable collaborative automation by networked embedded devices

An example in more details: Collaborative automation between water networks and virtual energy market 4

Water is easier to store than electricity and water utilities can turn it into cash

Energy cost is a challenge for water distribution companies Water networks offer good opportunities for virtual energy market Technical enablers are necessary

 Decision making tool ensuring that the water demand will be entirely fulfilled, evaluating the economic equation, and providing the best strategy to maximize benefits  Control system

A typical use case example Automatic calculation of modulation capabilities for 24 coming hours Based on:  Previsional pumping plan  Water demand and operational constraints  Energy prices dynamic context What-if scenarios and decision For each potential modulation, the water network manager can:  Preview the pumping scheduling, tanks storage and pressure levels  Select the modulation offers to be sent to aggregator Transaction with aggegator When the energy demand resource will be required, the updated pumping plan will be sent to operation system

Technical point of view

Main technical bricks

On the water network side  Water hydraulic simulation (Aquis simulation)   Coming from aggregator  Modulation capabilities calculation (Artelys optimization) Transaction module  Water demand forecast Energy prices

Arrowhead technology for bricks interoperability

Results and Take away

Water demonstration was based on a simulated environment

 Extracted from the distribution network of Birkerod (small town in Denmark)

10 to 15% cost savings expectations for the demo case

 Hypothesis: intraday capacity market contract  For other cases, benefits will greatly depend on water network characteristics and energy market

More generally, some key success factors for new features based on analytics:

 Technical infrastructures for easy data sharing  Services for interoperability between heterogeneous   bricks Good interfaces, understanding and interaction with people And an evidence not to forget: the final added value!

Thank you for your attention To contact us [email protected]

[email protected]

[email protected]