Maximising Synergies between Power Electronics and ICT for improving Energy Efficiency
Around 40% of the world’s energy usage is in buildings and 50% is used in motors and drives. The power electronics “DNA footprint” is considerable in these domains. Savings of 10-30% are possible. A significant portion of these savings can be derived from embedding ICT into power electronics solutions.
Around 40% of the world’s energy usage is in buildings and 50% is used in motors and drives. The power electronics “DNA footprint” is considerable in these domains. Savings of 10-30% are possible. A significant portion of these savings can be derived from embedding ICT into power electronics solutions.
The purpose of this article is to help readers recognize this potential for synergy by bringing a broader perspective of “energy efficiency” to the table so that at early design stages the potential impact of a given power electronics design can be maximized for a given application; and by giving some tangible example of ICT’s role in enabling energy efficiency, particularly the use of wireless sensor networks (WSN) for retrofit applications.
Holistic definition of efficiency
Traditionally many power electronic (device and circuit) engineers primarily think of the graph of efficiency versus load (& perhaps line and temperature) in order to define design targets but to get maximum impact in the application one needs to consider a much broader range of criteria.
The efficiency of components and devices is relatively straightforward and design and integration choices/trade-offs can be made, e.g. conduction and switching losses in a semiconductor device for a given topology, copper and core losses in a transformer. Of course other application constraints such as size, airflow, thermals, EMC also need to be taken into account. However major opportunities are missed, particularly where the application for the power electronics solution is known. To give a couple of simple examples:
- If an application operates for most of the time at very light loads it might be worthwhile trading off efficiency at max load (if the operation can manage it thermally and electrically) in exchange for improved efficiency at light loads. In other words shape the efficiency curve based on the application, not the max load. Implications may vary simply from component selection to complete changes in topology and magnetics. Similar arguments can be applied for the typical versus maximum specification operational voltage, temperature of the application.
- Taking this one step further there may be additional savings possible at a system level if the operation mode of the power solution is understood. For a HVAC application the addition of VSD (variable speed drive) control may enable a steadier and more incremental distribution of air, rather than the “too hot”/“too cold” extremities endured by people sitting directly under HVAC unit as a heater/fan clicks on and off. A higher degree of comfort to the end user improves their working environment and their productivity.
- We also need to be more efficient in our (ICT enabled) integration of renewable energies into our overall energy usage. Energy from a local wind turbine can directly supply an electrical load in a building and any excess energy used to heat water for later usage. In some cases (e.g. supermarkets, cold storage units) excess energy can be used to power freezers that are set to a slightly lower temperature for the period of excess generation. The internal freezer temperature can be programmed to “coast” back to its nominal setting, thereby using less electricity, at a time when it is more expensive. A variant of this is to avail of renewable energy from the grid based on a concept known as “demand response”. In this interaction the consumer is incentivized to avail of renewable/cheap electricity offering cost savings to the consumer and CO2 and peak load reductions to the supplier. So whilst purely from an “output to input power” perspective it is obviously more efficient to use energy as needed, one needs to think about the input power that is a variable that changes dramatically in terms of its cost and CO2 footprint. The efficiency in creating the input power itself at that given time must also be considered. This is an excellent example as to why micro-generation and storage are becoming increasingly important for successful operation of the smart grid.
For me an intriguing question is what role can digital control play for these scenarios?
ICT role in accordance with this holistic definition
While some decent guesses can be made to orientate the power electronics design efficiency somewhat towards the application, the big opportunity lies in the provision of solutions aware of their operational environment that can adjust their own behavior as well as the inter-connected system elements. To do this we need ICT enabled systems that can sense or receive information supported by an intelligent infrastructure to convert the information into decisions and energy saving actions.
One of the key technology platforms in this space is the use of Wireless Sensor Network (WSN) modules, “wireless” in that they use wireless communications and are not mains powered. Sensors, meters interfaces and actuators can measure the operating conditions in a given environment (temperature, light, humidity, CO2 levels, pressure, etc.) & feed this back to an intelligent system that combines algorithms and live data to adjust the operating environment accordingly. A few simple examples:
- On a sunny day it may be possible to dim lighting in some parts of an office area
- If the weather forecast for a winter’s day is warm defer the turn on of the heating system by an hour or two
- Defer activating the heating/cooling in a room until shortly before the next scheduled meeting and use occupancy sensors to activate the lights
All of these systems have some kind of ‘power electronics DNA’ embedded.
One of the most compelling reasons for the use of WSN technology is the cost and ease with which it can be installed. The biggest market potential is for retrofit in into existing building and systems where it was previously not cost effective or to override older inefficient systems.
An additional attraction of WSN technology is its re-configurability over a building’s life, for example if there is a substantial re-layout in an office area. The integration and optimization of renewable energy elements (generation and storage) offers yet another reason for using WSN.
WSN technology also offers an ability to perform ‘conditional monitoring’, i.e. use the sensory data to diagnose any anomalies in behavior in machines, devices and systems. Think about your car engine when it sounds “a little sick” or that fridge when it seems to get noisy and does not seem to cool your orange juice as effectively as it used to. In both cases they are both probably guzzling power and likely to break down sooner rather than later. In an industrial context the benefit is twofold, eliminating excessive running costs as well as minimizing risk of machine downtime. Algorithms can compare sensory data with benchmark data and automatically flag preventative maintenance actions.
The key message is to get all of these people from the power electronics and ICT communities together and get them to appreciate the benefits of working together. The power electronics elements themselves will become more efficient in their operating environment as well as enabling system energy efficiency improvements to be undertaken.
