Adaptive Computing Can boost Industrial Efficiency

As electric costs increase and the world moves away from fossil fuels, today’s manufacturers are challenged with how to build industrial applications that operate both responsibly and efficiently. Adaptive computing can play an important role by delivering a competitive  balance of cost, power, and performance for a variety of industrial applications.

Adaptive computing technology includes programmable solutions, such as FPGAs and adaptive SoCs, that allow manufacturers to change hardware functionality even after deployment, so systems can adapt to new requirements without new hardware. AMD is a leader in adaptive computing technology and offers a broad and scalable portfolio of FPGAs and adaptive SoCs.

Two applications where adaptive computing solutions can play a pivotal role in driving efficient performance include motor control and EV chargers.

Adaptive Computing Elevates Motor Performance

Inside traditional electric motors, FPGA-based adaptive SoCs help control the magnetic field generated in the stator so that interference between it and the rotor is minimised at any time to help produce motion, instead of just heat.

In an electric motor, there are three phases of currents that are injected to a power stage and they need to be controlled precisely, in microseconds. When you apply a specific algorithm, it reads the currents you are injecting into the stator, and it tries to figure out the currents that are being used in the motor so that it can estimate the torque. This control mechanism is called “field orientation.” It means you are orienting the mathematical field in an optimal way.

Adaptive SoCs help to ensure that the force to the rotor is applied in the optimal direction, helping to create the proper amount of current for every phase of the motor, and enabling it to operate at maximum efficiency. Without this information, the motor would spin inefficiently, consuming a lot of current, and eventually overheating, or it would spin and not efficiently produce the maximum amount of torque possible.

Another thing that’s important to motors is precise voltage control. Adaptive SoCs are particularly good at modulating voltage and electromagnetic interference (EMI) with built-in pulse-width modulation. Using this feature, you can switch the motor on pulses to control motor speed. You can also use it to mitigate EMI by spreading the noise over a wider bandwidth. With FPGA-based devices, you can create more sophisticated modulators with greater scalability and architectural flexibility to minimise EMI noise.

Adaptive Computing Can Accelerate EV Charging

Beyond motor control, another industrial challenge addressed by adaptive computing is EV Charging. One of the biggest challenges in the electric vehicle (EV) industry is the time it takes to recharge the batteries. Depending on where you go, it can take anywhere from 30 minutes to several hours to achieve a full charge.

It turns out the same technology that is used to manage and control motors can also be used to build fast EV Charging stations.

Adaptive computing can improve EV Charging in four key areas:

  • Power and Control – Adaptive SoCs and SOMs from AMD offer complete control of the power system, independent from the underlying software updates in the main controller. They  also offer fast control loops that rapidly deliver power to a vehicle’s batteries.
  • Simulation – Adaptive computing can enable simulations to help minimise the risk of additional necessary hardware iterations and optimise the control system and converter  performance.
  • Safety and Security – Adaptive computing solutions from AMD use a hypervisor and multicores to bound different criticalities, and are safety-certifiable to automotive and industrial standards, including ASIL D and SiL3.
  • Cloud and Web – AMD adaptive computing solutions enable integration with cloud services, Python support, JavaScript offloading and acceleration, and Ethernet connectivity.  Beyond EV Charging, AMD motor control methods can also be used for vehicle-to-grid (V2G) charging.

The AMD Kria™ KD240 Drives Starter Kit is an example of an adaptive computing solution that is ideal for motor control and EV Charging applications. This solution features scalable hardware that allows customers to fine-tune power, cost, and performance levels without changing their PCB. For more information, please visit: https://www.amd.com/kd240

For more than 50 years AMD has driven innovation in high-performance computing, graphics and  visualisation technologies. For more information about how AMD is enabling today and inspiring  tomorrow, visit the AMD (NASDAQ: AMD) website, blog, LinkedIn and  X pages.

Related Posts