Smart management and monitoring of utilities
The Internet of Things and ‘smart sensors’ are industry buzzwords at the moment, but what do they mean for the monitoring and management of utilities use Rosa Richards reports
The Internet of Things
Industries and businesses could benefit significantly from the intelligence gained from use of the Internet of Things (IoT) by 2025 ($4trn -$11trn ) according to analysis of more than 150 uses (McKinsey Global Institute, 2015).
Factories (1.2trn-3.7trn), work sites (0.2trn-0.9trn) and offices (0.1trn-0.2trn) are all predicted to benefit from a range of improvements in operations management, predictive maintenance, health and safety, organisational redesign, monitoring, augmented reality training and more.
Environmental managers will have the potential to achieve significant improvements in monitoring and control of raw materials, energy, water and heat use as a result of the IoT, generating not only savings but also potential income streams.
Sensors are the critical starting point or end point for IoT communications, as they generate the data that allows control decisions to be made. It is worth noting that utility companies have already been using the IoT, previously known as ‘radio telemetry’, for 20 years. There are some system issues such as interoperability and cyber security that must be overcome for the full potential of the IoT to be realised.
There are multiple disruptive and augmentative factors affecting the utilities market, from smart grids to cognitive computing, and low-cost and smart sensors are playing a role in facilitating these.
Businesses may already be saving money by using strategies such as day-ahead optimisation to use power when it is cheap. Many businesses already generate their own renewable power (solar PV, wind turbines, hydro-electric or combined heat and power), which at present can only be used instantly or sold to the grid. Add in battery storage, and then a site will become a virtual power plant, with the ability to use energy onsite when it is needed and the choice to sell energy to the grid when prices are optimal at peak times.
In addition, there is the potential for electric vehicles to be used as backup power storage for the national grid if energy customers are willing to be flexible. These components make up a more distributed energy system as part of a smart grid with a more complex model of balancing and settlement for the energy companies to manage.
The University of Bristol has been increasing its fleet of internal electricity meters over the past five years, and is moving from a system that collated and provided the data overnight to live readings. It also receives data via its suppliers for both gas and electricity meters.
“There is no real substitute for having half-hourly data for energy management,” says John Brenton, sustainability manager at the University of Bristol.
“Using our meters, we have been able identify and investigate areas with high baseload, and look at where 24/7 services can be controlled better. Being able to discuss a consumption profile with a building user often leads to innovative solutions to longstanding problems. Also we have been able to check that supplies that are exposed to high time-of-day charges, such as the Distribution Use of System, are adequately controlled to reflect that.”
All businesses (and households) must be ‘offered’ a smart (electricity) meter by 2020 as set out in the Smart Meter Bill, 2017. However, energy firms have indicated that completing the rollout of installation of smart meters by 2020 is unlikely, as it has been beset by challenges.
The Institute of Directors (IoD) has highlighted the spiralling costs of the rollout programme, which will be added to business and household bills, and called for the government to put the programme on hold. Other criticisms by the IoD are that the programme has failed to deliver interoperable meters for switching, is behind schedule, over budget and dependent upon out-of-date technology.
In the water sector, the non-domestic water retail market opened in April 2017. Not only will environmental or water managers be able to choose their water and sewerage services retailer but there are potential savings to be made for large water users from ‘self supplying’ – buying water directly from the wholesaler.
There are well-known benefits for large, multi-meter or multi-site users to have a consolidated bill (with a single pricing structure) to be able to compare water use over time and across sites directly and identify water consumption issues to take steps to improve water efficiency.
For example, non-domestic water consumption may be expected to be zero over weekends and bank holidays, so a higher than expected baseline indicates a leak on site caused by taps or showers for example, or from the plumbing. Leakage sensor products that measure consumption on every water outlet – for example, to monitor water usage on each floor of a building – are available to help identify leakage issues (LeakNet and LeakBot, for example).
Sometimes additional services such as leak detection and efficiency audits are provided by the water retailer. Other savings could be made by sites becoming more self-sufficient or autonomous in terms of undertaking their own greywater recycling, rainwater harvesting, local water abstraction or wastewater treatment onsite.
Savings from energy or water management would not be possible without the use of meters to monitor utility use. The design of water meters for example, has evolved significantly over time and is still advancing at a rapid pace, so much so that smart meters have become outdated before a rollout programme has been completed.
Water meters have improved from the days of ‘dumb meters’ visually read by a meter reader, to ‘automated meter reading’ meters read by a meter reader walking past, to the latest ‘advanced metering infrastructure’ smart meters currently being rolled out across London, which transmit data autonomously.
“Smart meters have really revolutionised data collection and analysis of usage data for the benefit of our customers – for example, to aid leakage detection,” explains Yvonne Ryan, head of smart technology development at Thames Water. “Because smart water meters proactively transmit data to a secure database at regular intervals, customers can now view an accurate timeline of their water usage online. The next generation of smart meters will improve upon this with a built-in analytical ability to identify issues in water consumption and flag them to customers.”
Thames Water is currently working with a supplier designing the next generation of smart water meters.
Cognitive computing technologies such as machine learning could lead to advancements that we can barely imagine today. By combining technologies and services, smart technologies can be developed that use artificial intelligence to predict what customers will need based on their consumption patterns before they are even aware themselves.
Artificial intelligence for pattern recognition can be already be built into smart sensors, with environmental compliance thresholds entered by the operator so that an alarm is raised when a set threshold is reached. This enables the operator to take remedial action if needed. The next step, which not everyone is willing to take, is to give the sensor the ability to directly issue restorative actions.
The IoT has enabled new business models to flourish. Building on the predictive functions of smart devices, an ‘ecosystem’ of service providers might in the future provide value to customers (‘prosumers’ ) based on their shared data and chosen priorities. Insurance products could be coupled with utility providers. This will enable business customers to save resources and time.
Future developments in sensors for the IoT will be driven by the need for lower power and longer battery-life, improved networking capability through embedded transmitters or receivers and the ability to process or store large amounts of short bursts or long streams of real-time data – that is, ‘big data’. Such developments will be aided by advancements in miniaturisation (like lab on a chip technology) and hybridisation.
For example, a microfluidic droplet nitrate sensor is under development at the University of Southampton, using low-cost components. This sensor miniaturises the process of detecting the presence and amount of nitrate (NO3) and nitrite (NO2) in a water sample, self-calibrates and can be used for real-time monitoring in the field.
It could be used by industry to measure background levels of pollution in waterbodies adjacent to sites. Adrian Nightingale, research fellow at the University of Southampton, says: “The microfluidic sensor platform we have designed can, in principle, be adapted to detect and measure any chemical analyte for which an assay exists in scientific literature.”
The sensor has low power consumption and uses small amounts of reagent, and can take measurements at high frequencies, even 30 times a minute if needed. Power use is 0.75W and 3.6 µl reagent used per measurement. The sensors are being further developed for commercialisation in collaboration with university spin-out company SouthWestSensor. The technology has so far been applied to measuring the dynamic variation of nitrate in a tidal river over the tidal cycles, and the university is now looking for partners to help them validate the technology in other environments and with other target analytes.
Rosa Richards is an independent environmental consultant specialising in water policy. She is a freelance science writer, programme manager of the Sensors for Water Interest Group (SWIG), and distance learning tutor of integrated environmental management at the University of Bath.