Energy thieves beware

Donald Fisher explains how data from smart meters could equip energy suppliers with a powerful weapon to combat the growing and expensive problem of energy theft.

Amid all the public concern about pricing, switching and smart meter rollout, energy theft remains a major problem for utility companies in the UK. Some estimates put the cost of it as high as £500 million a year. The UK Revenue Protection Association – a trade association for those involved in energy theft detection – believes it costs at least £300 million a year.

Most of this energy crime is conducted when users manipulate, tamper with or bypass the consumption meter. Criminal gangs have reportedly been tampering with 15 meters a day, while landlords have been found stealing energy from several properties in the same road.

Illegal marijuana growers, on the other hand, operate on a different scale altogether, accounting for about one-third of all the electricity stolen in the UK.

However, in the fight against energy theft, utility companies should now be contemplating how the government’s smart meter installation programme will give them a potent weapon with which to combat the energy thieves: data.

Admittedly only about a million-plus smart meters have so far been installed in the UK. Yet when all 53 million are online, saving an estimated £17 billion in energy use, they will supply utilities with a mass of data about consumption patterns. It is a raw material that will not be in short supply, because smart meters automatically take readings every quarter or half-hour and transmit them to head office.

Of course, manually sifting through all this information to look for unexplained dips in individual users’ energy consumption, putting it in context and deciding whether anything suspicious is happening, would be impossible.

What makes it entirely feasible, however, is the combination of streaming analytics and advanced algorithms. Together, they give utility companies the capability to build a predictive model that will quickly detect a customer’s deviation from their normal pattern of use and supply the context.

Building the model starts with the creation of a detailed load profile of consumption at a premises or household over a 24-hour period, showing the increases and declines over the course of the day.

Comparison can then be established between individual accounts and a group that should have similar patterns of use, based on factors such as household size, location and tariff. Much of the basic segmentation of customers by tariff and postal code required for the model has already been conducted by utility companies.

Other factors include the type and age of the home, level of insulation and the type of heating or cooling system installed.

Over time, the accumulation of data allows the predictive model to be created. Customers are placed in homogeneous groups, giving a good indication of what their normal pattern of use should be and quickly revealing any deviations.

Recognising a departure from the norm is crucial, but other important data has to supply context, because there are many legitimate reasons why energy use might drop off, for example, maintenance works, power outages, warm weather, or damage caused by a storm or by a construction digger going through a cable.

The use of streaming analytics and advanced algorithms enables all this information to be pulled in for correlation. So when, for example, meter sensors notify the utility company that the device may have been tampered with and has lost current, the company’s analytics immediately correlate this alert with all available information from its management system.  

This might then indicate that no company maintenance work was scheduled, that the customer had not given any prior notification of electrical or building work at the address and nor were there any outages in the area.

Very quickly the system has established a strong likelihood of suspicious activity and the address is placed on a list of suspect sites that can be passed to the utility’s recovery arm. Investigators can then examine the suspect’s historical patterns of consumption and billing – whether they owe the company any money or have a poor payment record.

Clearly, it is not possible to employ these exact methods when cannabis farms are illegally linked to the mains electricity supply in rural areas, bypassing meters entirely. However, utilities can install meters on the grid or distribution system, so that, for example, a line serving 20 customers can be monitored. Their respective load profiles can be aggregated to provide a single profile against which actual use can be monitored via the meter and compared with that of other similar groups.

Once suspicious cases have been passed for investigation by the utility’s revenue recovery arm, business process management tools allow the entire process to become fully visible to managers, who can see how many investigations are underway, the estimated total of losses associated with them and the length of time it is taking to process them.

From end to end, then, the utility companies’ massive programme of smart meter installation will open up new opportunities for detecting energy theft. The key will be the ability to understand every customer in relation to a similar, but homogeneous group, and to build a predictive model that provides meaningful insights.

It is these insights that utility companies currently lack – even in countries where smart meter rollout is more advanced. Although the volumes of data at their disposal have expanded massively, many still really do not know the real nature of their problems. They remain in the dark about the true extent of energy theft and the only way to find out more about it and take effective action, is to build a predictive solution. UK utilities take note.

Donald Fisher, senior director, energy solutions, Software AG