Big data shaping energy optimisation

Share

by Megan Houghton, Executive General Manager Energy Solutions, ERM Power

Big data is changing the way industries manage, analyse and leverage information, and is playing an increasingly important role in shaping the way businesses use and optimise their energy.

Data science can reshape the energy sector in the same way it is transforming financial services and health, where data scientists have built models for credit scoring and risk modelling and algorithms in the trading and medical fields for years.

Twenty years ago, no one would have predicted the reality of accessing financial advice from virtual robo-traders via our smartphones or using machine learning algorithms to detect and track health conditions like heart or respiratory disease.

Just as data science has transformed these sectors, it’s opening new doors in energy management in Australia right now.

Why is this so important? 

With continued high energy costs a reality for businesses, along with the transition to renewables and locally generated and managed energy, what’s required is a fundamental shift in the way we think about and use energy, and, importantly, how we realise the potential of an avalanche of data being captured and delivered through multiple sources.

It’s widely acknowledged that the energy sector has been slow to recognise and harness the opportunity and power of data to transform business productivity and enhance customer value.

As a retail energy business, ERM Power has always been data rich and data driven. As the second largest retailer to commercial and industrial customers in Australia, we have a unique perspective on the needs of large energy users.

The increasing complexity of the industry and the proliferation of offerings and suppliers is challenging territory for businesses to navigate. Data can play a pivotal role in energy productivity by helping businesses make evidence-based decisions that can enable them to save money, reduce consumption, improve sustainability or generate revenue from their energy assets.

Data is core to determining the best energy management solutions and the right approach for a business’ sector, size and operations.

There is a clear value in investing in data science and digital capabilities to stay ahead of the game.

Just as the health sector turned to data science models for diagnoses to solve health problems, the energy sector is increasingly using big data sets to diagnose the health of buildings and plants to solve their energy problems, in ways that consider the challenges and opportunities on both the supply-side and the demand-side of the energy equation.

An investment in big data and analytics can help businesses to make accurate, targeted and prioritised decisions around their energy productivity.

Exponential data growth

The amount of big data sets available from businesses own usage and those publicly available has grown exponentially with the Internet of Things. IoT’s ability to turn otherwise ‘dumb’ devices into digital intelligence, combined with faster WiFi and mobile networks and cheaper cloud-based storage options, is making the data science opportunity even bigger. In particular, IoT has the ability to deliver large amounts of real-time data, which has massive implications for the sector.

The utility sector is one of three sectors, alongside manufacturing and transportation, expected to spend the most on IoT worldwide in 2018, and this $73 billion spend will be dominated by smart grids for electricity, gas and water. A further $92 billion is expected to be spent on cross-industry IoT areas like connected vehicles and smart buildings, air conditioning and security systems.

Globally, the digitisation of the utility industry is occurring across the entire supply chain; through the network, via meters and into buildings, plants and equipment.

Seizing data opportunities behind the meter 

Real-time connectivity to what’s happening behind the meter is critical to enabling a data-led approach to energy management.

For example, with increasing deployment of ‘behind-the-meter’ energy resources like renewable power systems, grid-synced generators, battery storage and automated demand management solutions, IoT devices can be used to collect data from these resources and combined with existing data sets from utility electricity meters and weather data and market data.

Data scientists can use these new combined data sets to build very rich and accurate data models that can predict the impact of proposed energy efficiency or demand management projects, both from an energy productivity and financial point of view. These models are particularly well suited to multi-site business environments, as they identify issues across the portfolio, then prioritise actions based on a financial business case.

They find the best prospects for each site, whether it’s solar, battery storage, power factor correction or another energy efficiency solution. They draw on data from historical use, meters, individual equipment and tariffs, as well as using external weather and solar radiance data for each site to forecast future energy consumption and costs.

ERM Power uses models like this to simulate different solutions over multiple time periods to work out the optimal combination from a cost and payback perspective, while also considering the impact of different tariffs and other efficiency measures for businesses.

As an example, demand flexibility, which relies on accurate, real-time data, is increasingly important for organisations driving an agenda of improved energy productivity. It enables organisations to reduce power consumption at different times when wholesale prices spike or there are constraints on the network. As the grid incorporates more renewable energy, modelling demand flexibility and taking action will be even more important.

Making the complex, simple

Modelling and data visualisation are being used to create user friendly products like smart apps and dashboards that turn the otherwise complex energy data into something that’s simple, easily digestible and actionable. At ERM Power, design thinking methodology sits at the heart of new product development to ensure capabilities and tools are in line with user needs.

Using data science and machine learning technology, ERM Power has developed a predictive forecasting and notifications smart app.

Relatively new to market, the ERM Power app can predict a spike in an organisation’s energy usage and advise them in advance so they can take action to reduce or defer consumption. Real-time data feeds the results back to customers via the app – they can immediately see the results of their actions or inaction and understand the cost implications.

A new data-driven frontier

With more connected things than people in the world now and the number still growing exponentially, IoT and big data will undoubtedly lead to far more automation in the way energy is consumed and managed behind the meter.

As more businesses install their own infrastructure behind the meter, whether it’s solar, embedded generation or other new technology, this will create even more meaningful data sets that can be incorporated into modelling. New types of sensor-equipped devices will provide more granular data, with the potential to drive more accurate outcomes for businesses with very little human interaction.

The two-way connectivity that these devices will enable will change the way organisations view and manage their operations in the future. For example, machines may run when prices are favourable, and demand is low, or when businesses are generating their own energy to reduce consumption from the grid. This is a clear example of automation driving energy optimisation in business.

Data is continually and rapidly transforming the future of energy management. The opportunity for Australian business is to harness its power to create a sustainable, competitive advantage.

CONTACT US

We're not around right now. But you can send us an email and we'll get back to you, asap.

Sending

©2018 Energymagazine. All rights reserved

Log in with your credentials

or    

Forgot your details?

Create Account