Improvement of bootup time using Power Management - Project Update -
Sustainable Energy Management - BOOTUP · 2017-12-04 · 43% increased productivity among staff Top...
Transcript of Sustainable Energy Management - BOOTUP · 2017-12-04 · 43% increased productivity among staff Top...
SustainableEnergy ManagementWith Artificial Intelligence
Bringing intelligence to energy management 03
AI in energy – Use cases – Power generation 04
AI in energy – Use cases – BEMS 05
AI in energy – Use cases – Electric vehicles 06
Technologies enabling intelligent energy management 07
Effective use of AI – Case study 08
Energy 2025: What to look forward to 09
Key players 10
Key players 11
Key investments 12
Why BootUP? 13
02
Bringing intelligence to energy managementIn combination with other technologies, AI has the potential to deliver the active management that will be
required for the grid of the future. Powerful intelligence will be able to balance grids, manage demand,
negotiate actions, enable self-healing and facilitate a host of new products and services. Indeed, AI, will not
just lend itself to the energy transition, it will also enable more efficient and effective utility operations by
helping to analyze unstructured data which typically makes up to 80 percent of data in an organization.
ROI:
46% processes and tasks automated
45% usage of prescriptive/predictive analytics
43% increased productivity among staff
Top 3 motivations for implementing AI:
62% Automating IT processes
61% Automating business processes
60% Increasing innovation
Average investment of the energy sector in AI in 2016 - $5.3 million
29% of organizations in the energy industry have fully
deployed AI technologies and they are working as expected.
Noteworthy numbers
03Source: Infosys report ‘Amplifying human potential: Towards purposeful Artificial Intelligence’
04
AI helps enhance short-term renewable
forecasting and improving equipment
maintenance. AI has been used thus far in
pilots for analyzing wind turbine data solar
panel sensor data to understand sunlight
intensity. This analysis is combined with
atmospheric data gathered by radar,
satellites and ground weather stations to
estimate power requirements and equipment
performance. AI is also being applied in the
areas of energy storage and estimating
battery pack lifetime, among other use
cases.
AI can ensure steady energy availability by
balancing the supply of power from
conventional sources like fossil fuels with
that generated by renewable sources of
energy like solar and wind energy. In this
case, individuals who own equipment related
to renewable energy, such as solar panels,
can be asked by utilities companies to allow
their batteries to be available for the grid in
exchange for reduced repayment costs on
the equipment. This is an attractive
alternative to deploying large scale batteries,
and promises to be made workable on a
large scale.
Using machine learning algorithms,
intelligent systems can collate, compare,
analyze and highlight risks and opportunities.
AI is also used for modeling scenarios and
suggest actions and impacts. Network
operators are using algorithms to inform
decision making or have better situational
awareness of what is going on at the grid at
any given time. By using AI tools and
concepts, power producers can sustain the
longevity and performance of their grid
equipment. Grid management companies
like Siemens and GE are deploying AI
applications to increase the output of
traditional assets.
AI in energy – Use cases – Power generationArtificial intelligence will power the mechanisms that are needed to control an integrated power
network, also known as a ‘smart grid’. This intelligent system can balance grids, manage demand,
negotiate actions, enable self-healing and facilitate a host of new products and services. By making
sense of unstructured data, the smart grid can enable more efficient and effective utility operations.
Renewables Management Demand Management Infrastructure Management
05
AI in energy – Use cases – BEMSBuilding owners today face many
challenges, as they search for ways to:
• Reduce energy consumption and carbon
footprint
• Increase tenant satisfaction and loyalty
• Protect tenants with non-intrusive security
methods
• Lower operating costs
• Maintain buildings and comply with
regulations
• Manage modern buildings’ technology and
systems, and
• Reap maximum ROI
How a Building Energy Management System
(BEMS) works
40%energy savings by using building automation and control systems
06
AI in energy – Use cases – Electric vehiclesThe electric vehicle (EV) market has now evolved to accommodate an environmental-friendly option: Plugged-in Electric
Vehicles (PEVs). Companies like General Motors, Ford, Nissan and Tesla are producing PEVs, with the Tesla Roadster
leading the way in the all-electric PEV category. While the prevalence of PEVs can result in a drastic reduction in
pollution due to not needing fossil fuels or emitting pollutants, they still need charging outlets to power them on the go.
Enter the Smart Grid.
Electricity is still being generated to a large extent through conventional fossil fuel-based energy sources, so to reduce
the potential of pollution, the Smart Grid can interact with the PEV to charge it at the most optimal time, which happens
to be the early morning hours, when power demand is at its lowest and wind at its peak. A Smart Grid can also take
advantage of the multitude of PEV batteries plugged into it at any given time to engineer a reverse transfer of energy
from the batteries to the Smart Grid, a process called ‘vehicle to grid’. This can prevent blackouts or brownouts during
critical peak times, and provide an alternate source of energy to the grid apart from wind and solar. PEV owners can be
incentivized to allow the usage of their vehicle batteries in this way.
Source: NCSU report : Grid Integration of Plug-in Electric Vehicle in Smart Grid
07
Technologies enabling intelligent energy managementArtificial Intelligence• Ensures optimum efficiency by Implementing changes based on energy usage trends analysis• Responds immediately to demand response opportunities without human intervention• Uses distributed energy storage systems effectively and ensures optimum storage at all times• Determines peak times for energy generation using renewable sources based on analysis of weather patterns• Studies complex market trends and analyzes the data to create energy purchase plans that override market volatilities
Internet of Things• Realizes efficiencies by enabling smart lighting based on presence detection or meeting confirmation• Can control heating and cooling in a building based on presence detection• Can communicate with external devices like a smart grid to control energy usage and minimize system strain• Supports the scalable integration of clean and efficient technologies such as PV and EV chargers• Integrates and coordinates connected equipment (load/generator/storage) for energy efficiency and financial benefits
Big Data• Using real time and batch processing tools, big data helps evaluate current green strategies and assess if those strategies are actually working• Monitoring the energy-consumption patterns of devices and benchmarking them against similar devices and locations helps avoid unexpected device, equipment, and system failures• Big data solutions help monitor the energy print of equipment in real time. For e.g., if the malfunctioning of a specific part is causing a machine to use excess power, such patterns can be identified through big data• Providing granular energy data for every device and system in real time as part of advocating changes in human operations patterns and rewarding conformation to these can bring lasting behavioral changes • Enables deep analysis of energy demand and loads and the impact of distributed generation resources
Cloud• Helps reduce business’ carbon footprint by allowing them to remotely access their computing resources• Improves performance for greater operational agility with lower costs to deal with competition, regulations and market conditions• Helps host data and augment supercomputing capabilities, besides supporting analytics-heavy activities, such as geospatial and 4D seismic modeling• Helps build deeper customer relationships and brand value using cloud-enabled customer relationship management and social media tools• Helps transition to lower-carbon alternatives and renewables
Blockchain• Decentralized storage of transaction data increases security and ensures greater independence from a central authority• Such decentralized business models no longer require third-party intermediaries• Supply and demand are balanced by smart contracts (balancing market, microgrids, virtual power plants, storage)• Transaction data is stored on the blockchain using a decentralized mechanism, with parties identifying themselves through their digital identities• Enables secure storage of ownership records, e.g., emission allowances, renewable energy certificates, and asset management
08
OVERVIEW
One of Siemens’ objectives in implementing artificial intelligence to gas turbines was to reduce the emission of nitrogen oxides
during operations. Siemens Power Generation Services and Siemens Corporate Technology have developed a system that
continuously optimizes the operation and control of combustion in gas turbines, using the company’s proprietary AI application Gas
Turbine Autonomous Control Optimizer (GT-ACO).
The technology was first applied to its flagship H class gas turbine installed at the premises of a top Asian client. After the initial
setting for minimum emission of nitrogen oxides was done manually by a turbine expert, the AI system took over control.
PROCESS
GT-ACO’s neural model keeps modulating the distribution of fuel in a turbine’s burners. However, the settings for each burner vary by
location, gas composition and local weather conditions. So GT-ACO needs a few weeks of learning on each turbine before it can
autonomously make beneficial changes to the controls.
RESULTS
In the very first trial itself, just two minutes after GT-ACO took over control of the combustion unit, nitrogen oxide levels dropped by
20 percent. Turbine experts also believe that the technology will also help reduce wear and tear by reducing the amount of
vibrations that happen during operations. Finally, the technology can also help reduce equipment ageing using the collective
knowledge of gas turbine thermodynamics through physical models along with machine learning.
FUTURE APPLICATIONS
Siemens’ turbine experts believe that GT-ACO has vast potential in power distribution, production automation and process industry
applications. Siemens is in negotiations with power generation businesses in this regard and is also developing additional
optimization applications.
Applying neural networks to gas turbine optimization
Just two minutes after artificial intelligence took over control of the combustion unit, nitrogen oxide levels dropped by 20 percent.
Case study
09
Energy 2025: What to look forward toSunshine-based electrification• 430 quintillion Joules of energy received from the sun every hour, enough to power the earth for a whole year.• Because of economies of scale from greater deployment and improving technology, unsubsidized solar photovoltaics can now compete with natural gas power plants and will soon compete with coal plants.• Due to countries like the US, China and India recognizing that solar is a better bet, there have been a spate of cancellations of new proposed coal-based power plants in these countries.• Issues related to intermittency of availability are being countered through improving battery technology for utilities and electric vehicles, which is also driving down costs.• Battery farms connected with solar farms are enabling entire communities in Hawaii and Australia to become self-sufficient.
Environmental and economic imperatives• Powering the world through clean technology is not only desirable for the environment but is also proving to be the economically smarter option, as clean energy is cheaper than fossil fuel energy.• A recent report by RethinkX suggests thatby 2030, the transition to electric transportation will be dominant, which will drive oil prices down to US$25 dollars per barrel.• Electric vehicles will cost the same as conventional cars by 2018, and over time become much cheaper than gasoline or diesel vehicles due to their simplicity and lower O&M costs.• Energy Storage as a Service (ESaaS), which incentivizes energy producers at both individual and enterprise level to conserve energy from non-carbon emitting sources, is a rising trend.• Green energy generation provides a host of benefits, including little to no global warming emissions, improved public health and environmental quality, and a vast and inexhaustible energy supply.
Systems intelligence• The dominant trend will be the electrification of energy. Thus, more transportation, including cars and trucks, will be electric.• Innovations such as Germany’s Ubitricity replacing conventional bulbs in street lamps with LED ones, and then fitting these poles with sockets for charging electric cars have ushered in new energy models.• Secure grids use network design and configuration, control strategies and grid collaboration to allow automatic reconfiguration in response to threats, improved power quality and increased efficiency.• Optimal air-conditioning control uses weather data, energy market pricing and feedback from building occupants to intelligently alter the operation of a building’s air-conditioning system.• Virtual Power Stations intelligently aggregate renewable energy generators – such as solar panels – and energy storage into a cost-effective and reliable electricity supply network.
10
Key playersCOMPANY COUNTRY WHAT IT DOES SOLUTIONSAppOrchid
http://www.apporchid.com/
Deploys deep learning and an NLP-based interface
to understand grid behavior under variable weather
conditions.
• Smart meter and customer- facing apps
• Distribution intelligence apps
• Renewable apps
• Tariff systems, billing, POS, customer service and scheduling systems
• Cross-functional ‘tribal knowledge’ apps
• Industrial IoT apps
• Asset management apps
Alpiq
http://www.alpiq.com/
Uses AI to understand, analyze and predict user
behavior. This helps it understand energy
consumption patterns within a building.
Develops future-proof solutions in the fields of,
• Power lines
• Switchgear
• Transformer stations for energy transmission and energy distribution.
With control systems, light signal systems and tunnel infrastructure solutions, Alpiq
makes rail and road transport safer and more secure for applications in the field of
overhead catenary lines, rail safety and security, and tunnel projects.
Siemens
https://www.siemens.com
Develops artificial intelligence solutions for
improving the performance and longevity of its
products.
Gas Turbine Autonomous Control Optimizer (GT-ACO), an AI application that can
optimize the performance of its gas turbines to increase their longevity.
Arria
https://www.arria.com/
Created a virtual senior engineer in a control center
to diagnose issues and create work orders,
outage-related press releases and customer usage
reports.
• Articulator Pro, a sophisticated NLG toolkit that allows you to build complex
platform applications, from fine-grained, customized NLG integration to large-scale
data-processing and BI.
• A-Lite, a web-based NLG toolkit that allows users who might not be specialist
developers to build their own NLG applications that can convert data into rich
content like stock reports, product descriptions, crime trend reports and sales
updates.
Hazama Ando
http://www.ad-hzm.co.jp/
Developed a new smart energy system using an
AI-based energy management system (EMS).
Adjusting to Human Smart Energy System (AHSES), a system that enables a
business to make full use of electricity generated by a solar power plant, resulting in
a Net Zero Energy Building (ZEB).
Upside
https://upsideenergy.co.uk/
Developed an Advanced Algorithmic Platform (AAP)
that manages the demand response of different
devices to be run in parallel.
Virtual Energy Store, a cloud service that aggregates the energy stored in systems
people and businesses already own – uninterruptible power supplies (UPS), solar PV
systems (solar panels), electric vehicles (EV), domestic heating systems, etc. This
excess energy can then be sold to the grid to help it balance supply and demand. The
revenue generated from these services is shared with device owners and
manufacturers.
AppOrchid
http://www.apporchid.com/
SmartCloud’s AI-driven solutions capture and apply
knowledge in real time to improve the management
of complex situations.
• Situational awareness for smarter bulk power grids.
• Optimized Combined Heat and Power (CHP) across a wide range of operating
conditions to cut costs and improve overall performance.
• Industrial AI to model, simulate, and deploy Transactive Energy solutions for
improved system design and better monitoring.
• Monitor curtailments and predictively alert for non-compliance situations to avoid
costly penalties and gain full financial value from fast demand response programs.
• Model, simulate, forecast and optimize water grid in real time to better achieve
operational performance goals and utilization of assets such as wells.
11
AppOrchid
http://www.apporchid.com/
In talks with National Grid to apply AI to energy use.
Already helped parent company Google to reduce its
total data center power consumption by 40% and its
Power Usage Effectiveness (PUE) overhead by 15%.
Machine learning solutions for energy management of large scale commercial and
industrial systems.
Nextracker
https://www.nextracker.com/
Uses software developed by BrightBox to increase
energy production of solar farms, thereby enabling
faster operations and easier maintenance.
• NX Horizon, a single-axis solar tracker that brings self-contained motor power to
each row of solar panels using less steel and rotating a full 120 degrees.
• TrueCapture, an intelligent, self-adjusting tracker control system that increases
typical PV power plant energy by 2 – 6%.
• NX Fusion & NX Fusion Plus, to increase the energy output and duration of solar
power plants.
GE Energy
https://www.gepower.com/
Using AI to enhance wind turbine efficiency in
Japan, raising power output by around 5% and lower
maintenance costs by 20%.
A wind turbine-related AI technology that correlates past weather and turbine
operation data to predict output up to a week in advance, in increments ranging from
five minutes to an hour. It makes individual forecasts by turbine, as propellers at
different heights face different wind speeds. With these fine-grained predictions, it
can stop turbines for maintenance checks on low-output days, raising overall output.
COMPANY COUNTRY WHAT IT DOES SOLUTIONS
Open Energi
http://www.openenergi.com/
Explores how AI and machine learning can be
leveraged to orchestrate demand-side flexibility for
industrial equipment, co-generation and battery
storage systems.
Dynamic Demand is Open Energi’s unique, patented technology platform which
invisibly unlocks demand side flexibility to help businesses make money and cut
costs. The technology connects to a wide range of equipment – from fridges to
furnaces – and works by intelligently shifting electricity consumption in response to
second-by-second fluctuations in electricity supply and demand UK-wide. The
system is governed by equipment control parameters which determine if and for how
long equipment may respond, ensuring operational performance is never affected.
Verdigris
https://verdigris.co/
Applies AI to building management to listen to
electrical signals and identify the type of equipment
in the building.
Verdigris combines proprietary, high-speed sensors with its powerful artificial
intelligence platform to make informed energy efficiency investment decisions. Using
the Verdigris dashboard, you can see where you’re winning on energy spend and find
improvements.
12
Key investments
Source: Venture Scanner data as of June 2017
Activity by selected investors
Venture investing by category
Average funding by category
13
Why BootUP?Serial EntrepreneursA majority of the founders working at BootUP are serial entrepreneurs who have built multiple companies and achieved no less than a few exits. The BootUP experience is personal and high-touch, providing the opportunity to interact with seasoned entrepreneurs who have built numerous high-impact startups.
Global ReachA majority of the founders working at BootUP are serial entrepreneurs who have built multiple companies and achieved no less than a few exits. The BootUP experience is personal and high-touch, providing the opportunity to interact with seasoned entrepreneurs who have built numerous high-impact startups.
Traction-FocusUnlike many Silicon Valley accelerators, BootUP is not a standard 3-month mass production pitch training program followed by a Demo Day to investors. Instead, BootUP has a boutique approach focused on helping more mature startups gain and sustain traction over 6-24 months to generate tangible results.
NetworkA majority of the founders working at BootUP are serial entrepreneurs who have built multiple companies and achieved no less than a few exits. The BootUP experience is personal and high-touch, providing the opportunity to interact with seasoned entrepreneurs who have built numerous high-impact startups.
Corporate InnovationBootUP drives innovation for leading corporations by scouting for emerging technologies and by running vertical acceleration programs. This provides opportunities for BootUP startups to do pilots, drive sales, access corporate venture capital funds and potentially become acquired.
High-Power EventsA majority of the founders working at BootUP are serial entrepreneurs who have built multiple companies and achieved no less than a few exits. The BootUP experience is personal and high-touch, providing the opportunity to interact with seasoned entrepreneurs who have built numerous high-impact startups.
Clients
68 Willow Rd, Menlo Park, CA 94025, USAPh : +1 800.493.1945