Embedded Software IP & Technology Transfer in Power Electronics Applications

FPGAs and Plug-in Hybrid Vehicles

During the summer, I have had the chance to drive during one week one of the five Toyota Plug-in Hybrid currently under test in Canada. Those vehicles are part of a experimental project involving Toyota and canadian universities, among them Université Laval and its power electronics lab LEEPCI (my former lab). Click here to see the public announcement made with Toyota during the summer.

The vehicle is actually working very fine and it is a real pleasure to drive. For those interested, I did use only 3L of gas for normal use during the week (around 80km) which is very energy efficient! Click on the figure below to see more pictures of the car :

Where’s the link with this blog ? EV are obviously heavy power electronics applications by being used to convert energy stored in the batteries to the motor and vice-versa (motor to batteries while braking). Where’s the link with FPGAs? You may be interested to read this SAE Report written by Delphi people in 2006 and titled “FPGA considerations for Automotive applications”. According to its authors,

The complexity of automotive products will continue to increase, even as the pressure to decrease the development cycle, decrease cost, and increase quality and reliability mounts. FPGA usage to meet application needs will continue to grow as a means of reducing cycle time and development costs. Understanding and developing all aspects of FPGA manufacturing, design, implementation, application usage and performance can address the quality and reliability aspects of using FPGAs as product solutions. A low unit cost should not be the only major driving factor in choosing an FPGA. It has been shown that there are other items that can significantly add to unit cost based on the design methodology used for the implementation and verification of an FPGA. The complexity and challenge of implementing an FPGA device will erode any advantages of a traditional design flow. FPGA development requires discipline in assuring adherence to robust practices.

Briefly, they mean to have a look to the total cost of ownership (TCO) against other IC solutions.

Smart Grid : An opportunity for FPGAs in Home Appliance space?

According to this EEtimes article, the Association of Home Appliance Manufacturers (AHAM) took the opportunity of being at the United Nations Climate Change Conference in Copenhagen to release a very well written 25 pages document titled “The Home Appliance Industry’s Principles & Requirements for Achieving a Widely Accepted Smart Grid“. In this document, the AHAM – based on its unique perspective to the Smart Grid Vision – is intended to provide three essential requirements for the Smart Grid’s interaction with consumers in order for the Smart Grid to be successful. Among those three requirements, the second one is the most interesting from a technological (embedded systems) perspective :

Communication Standards must be open, flexible, secure, and limited in number

This requirements then splits in four requirements : open, flexible, secure and limited in number.

From a FPGA perspective, flexible sounds very familiar because its embedded in the name of the technology itself : Field-Programmable. But is this flexibility may solve problems and help the development of Smart Grid enabled homes ? According to the authors,

Smart Grid enabled homes will have varying levels of sophistication, depending on the type of appliances, devices, and networks that are installed. There are many configurations, combinations, and options for energy management inside the home. Some possibilities could include a simple email notice for a manual demand response by the consumer, a smart meter directly communicating with a specific appliance to ask it to turn on and off, or a meter communicating with a Programmable Communicating Thermostat allowing for temperature adjustment.”

From now to the moment that every appliance is going to talk the same language – even with such standardization, that is limited to the US only – one can think that this is going to be long and costly. This process has been started since a long time on industrial side (with many types of protocols) and there is still no single communication standard. Altera and Xilinx are actually taking advantage of this massive willingness to connect but protocol-segmented environment. Their programmable chip solutions enables them to sell a platform on which industrial equipment manufacturers can then use to build their own platform which is going to be finally customized with a regional/market-specific set of IP blocks. This approach enables flexibility while also reducing costs and time to market.

Is the same idea is going to happen in Home Appliance space ? As we all know, this high-volume market is very focused on costs. Not considering smart grid, a chip to chip price analysis would probably give only small chances to FPGAs. But considering that :

– according to a recent Whirlpool survey, 84% of consumers choose energy – not water or time – as most important when it comes to home appliance efficiency, and that

– according to Electric Power Research Institute, the implementation of Smart Grid technologies could reduce electricity use by more than 4 percent by 2030 providing a mean savings of $20.4 billion for businesses and consumers,

… there may have an opportunity there for FPGA chip manufacturers. Among the most important ones, Altera is already there.

Acromag goes green with FPGA solutions in Wind Turbine Control applications

Here’s a very small article from Acromag positionning its FPGA products toward smart-grid and Wind Turbine Control applications.

The basic components of a control system to maximize energy capture from a wind turbine include: rotor pointing, blade speed regulation, minimization of pointing and pitch control, and the mitigation of disturbances (i.e. excessive rotor speed, wind gusts, etc.). Additionally, the environment inside the rotor head, or nacelle, is very hostile. Any control device used in this environment must be self-diagnostic, rebootable, extreme temperature tolerant, vibration tolerant, and of course, affordable.”

OPAL-RT and Real-Time Simulation of Electric Motor Drives

FPGA technology in power electronics application is not only a matter of glue logic or system-on-a-chip platform but is also a matter of high-performance computing for real-time simulators of electric motor drives and any dynamic systems.

The use of FPGA in high-performance computing is not new: many applications have benefitted of the accelerated computing capabilities of those devices enabled by their parallelism capabilities, such as DNA sequencing and financial services. Since power electronics applications are typical fast-dynamic, complex and non-linear application, there might have a natural interest there to exploit accelerated computing platforms to simulate those systems in real-time.

This is exactly what OPAL-RT, a Montreal-based company, is doing : FPGA-based real-time simulators for power electronics applications. Their products enable system designers to quickly, safely and cheaply test their designs on a virtual plant. A typical example would be a PMSM-based system motor control designers that want to test his algorithms on many size of motors (2kW-5kW-15 kW) without having to plug its controller on a “real” motor : this designer can then plug its controller on a OPAL-RT “virtual” motor and test its motor control algorithm at a very early stage in its project (before having the real system protype available for example). The model of those “virtual” motors can be based on standard analytical equations or based on JMAG finite element analysis.

This type of technology opens a very broad range of new possibilities in the design of power electronics systems.

Actel Mixed-signal FPGA introduction for Motor Control

Here’s an interesting online tutorial presented by the famous Clive ‘Max’ Maxfield introducing Actel Fusion mixed-signal FPGA and how they can useful for Motor Control applications. A general introduction on FPGAs is presented for people that aren’t familiar with this type of technology (which typically the case for motor control system designers).

Hurry up : this tutorial is only available until August 15th 2009 !

Horses and FPGAs

Kevin Morris’ most recent article of FPGA and Structured ASIC Journal makes a very humourous but interesting analogy between FPGA vs ASICs design battle and the battle that occured not so long ago between car cars vs horses as a mean of transportation :

The new BMW 5-series sedan outperforms the horse and buggy in every important way. Your family will travel farther in a day and arrive less fatigued thanks to our superior cruising speed, climate-controlled cabin, and luxurious upholstery. It’s so much easier to use as well – no more hitching up the team before you start, and no more watering, feeding, and grooming at the end of the day. You just turn the key and drive away. Simple as that. So, before you snap up that new stallion you’ve been eyeing – consider a car instead.”

Morris’ article gets interesting at the end where he points out that applications that can benefit from hardware programmability are subject to profound change :

FPGA companies are defending against this attack, of course, by equipping their devices with both hard- and soft-core processors so that they can reap the advantages of software programmability as well. The outcome of that game, however, will probably be determined by the existence of design requirements that mandate hardware programmability – features where software cannot deliver the performance or power efficiency required. Designs with these sorts of requirements will remain in the sweet spot of FPGA, while general-purpose embedded platforms have a better-than-even chance of winning where software alone can do the job.”

This is obviously the case with power electronics applications.

FPGA-based Motor Control Design – A new era has come in power electronics

The movement has been going on for a few years and it accelerates. From specialized digital signal processing chips mainly used in communication applications, FPGA manufacturers have entered in the embedded system market to compete DSP and MCU chips.

Regarding raw computing performances (and not considering cost and design methodology), it is generally accepted that FPGAs, with their inherent parallelism capabilities, outperforms DSPs by many order of magnitude. This has a positive impact on performance on computional intensive applications such as digital signal processing and real-time control.

Knowing that more than 60% of electric power in the industrialized world is used to power electric motors, motor real-time control is an important application that can potentially take advantage of the high computing capabilities of FPGA-based embedded systems. Variable-speed motor drives are power electronics applications that have time-variant parameters and non-linear dynamics in which complex calculations must be achieved in real-time at high frequency to get optimal performance. Crunching numbers at high speed, this is exactly where FPGA can perform.

How motor control applications can benefit from FPGA technology ? What can a low-cost programmable hardware and software chip make better than a DSP or a MCU chip ?

There’s is no obvious answer to those questions and it must be looked not only on the technical perspective (performance gains), but also on a engineering perspective (better reliability) and a business perspective (cost reductions). Each of those perspective are going to the subject of subsequent posts.