This page contains a Flash digital edition of a book.

USING DATA EFFECTIVELY for energy efficiency

different way. Automatically reading these varying electronic tags, and sorting them into correct groups like “pumps” or “boilers” is an ideal task for artificial intelligence algorithms. This system gathers all of its data from

Mike Darby CTO and controls guru at Demand Logic, a UK start- up with a mission to use data analytics to help building managers cut wastage, explains how the company is taking the initiative to address the wasteful energy oversight currently presenting a major concern for building management personnel


t is staggering that so much energy is still wasted in commercial buildings. Despite

all the modern technological innovations, the buildings we work in every day are still riddled with energy inefficiencies that are really quite easily to solve. For example many of the rooms we work in are heated up and cooled down at the same time for weeks on end. Chillers the size of lorries (and which consume the same amount of electricity as hundreds of homes) are often left on at the weekend by mistake, all of which contribute to the huge amount of energy being continuously wasted. It is understandable to look around for a culprit. But the truth is that modern buildings are actually quite complicated and getting them to run properly is non- trivial. At Demand Logic, we install a small device that streams out all the energy- related data from a building in order to mine this information for insights. In a typical office block, there can be as

many as 25,000 individual “data points” - each relating to, perhaps, a hot water valve, a pump motor or a boiler-enable signal. When an issue arises somewhere in such a complex system it is all too easy to miss the needle in the haystack (especially if the skilled teams we rely on to manage


and maintain our buildings are suffering constant budget cuts). This is where data analytics and simple

visualisations can really help. One of the more popular web-widgets we’ve provided is the 'major plant watchdog' which simply shows red when a big item of a plant is running when it is not expected to be. Another example, is the 'rogue asset finder' which scatters many items of smaller plants across the screen, visually singling out those more likely to be faulty and wasteful.

AUTOMATED CONTROL The ability to sort many plant items such as ceiling-mounted air-conditioning units, bringing to the top those that are doing their job less well, can lead to significant savings on maintenance as well as to energy and carbon reductions. Rather than visiting each unit in turn – which can mean waiting until the evening to gain access to the ceiling voids – maintenance staff can focus only on the ones that are failing to adequately control a room. Over the years, most buildings have been

modified by a host of different engineers and controls experts, each of whom may “label” the data points distributed throughout the building in a slightly

Joe Short, CEO of Demand Logic by the chiller unit at Kings College which uses as much as 20% as a small village when running

the Building Management System (BMS) already present in the building. Many ask why the BMS does not itself provide enough insight. The reason for this is that a BMS is an interconnected network of small micro-controllers, very well suited to reliable 24-7 control of a plant, but entirely unsuited to the storage and manipulation of millions of numbers. It is true that many BMS provide excellent real-time views of specific systems, showing live temperatures and pressures, but it is unreasonable to expect these systems to help with the changing needs of big data analytics. It is better to leave the BMS to do what it's good at, and do the number crunching separately on modern general-purpose servers. However, improving the energy

performance of buildings is not all about providing data, however intuitively presented. The challenge is to help people to act on the findings and make lasting energy-saving and performance enhancing improvements. This is essentially a problem of "behaviour change" - not the commonly-talked-about kind that attempts to convince the mass public to adopt a particular prescribed low-carbon behaviour (very hard to achieve and possibly of short- lived benefit), but rather that which facilitates action amongst a particular expert group of people who hold the keys to a large amount of high-energy-using equipment. Demand Logic is now teaming up with London South Bank University, in a project funded by the Technology Strategy Board, in order to explore more potential for mathematical techniques, of which the findings will be published. There is so much to gain from applying the

latest data analytics to the field of energy efficiency but many more organisations are need to come together on the issue. Perhaps it is even time for a new discipline – maybe to be launched by an inaugural conference: ‘big data for energy efficiency’?

Demand Logic


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32