Using Energy Harvesting for Smart Grid Monitoring

Thermal and vibrational energy harvesting is increasingly being used to power sensors and wireless networks to provide a cost effective way to monitor and control power equipment as part of a smart grid.

There are several challenges surrounding the implementation of the smart grid. As more renewable sources of energy such as wind turbines are added to the grid, the power provision becomes more variable and the need for more accurate information at all stages of the grid becomes vital. More battery systems will be added to the grid to even out the supply to homes and businesses, and these need close monitoring.

Adding monitoring to the power network in the home, sub-station, and renewable sources requires sensors, microcontrollers with power, and a link back to a central monitoring unit, all needing a reliable source of power. (Figure 1) While a system like this can be relatively easy to install, the on-going maintenance of replacing millions of batteries has become a significant headache for power utilities around the world. Being able to take power from the surrounding environment is an increasingly popular way to minimize the maintenance costs and enhance the reliability of the system, and there are several techniques in energy harvesting that can help.

Using energy harvesting to create the smart home by powering sensors and wireless links without having to install power lines or replace batteries is also a key part of reducing power consumption and part of the smart grid ideal.

Energy harvesting covers a wide range of different technologies. Solar is the most obvious one, using photovoltaic cells to provide power, but this is the one technology that cannot easily be used for the smart grid as almost all the equipment is indoors and in cupboards with no natural lighting. While you might be able to use artificial light, this is a highly inefficient approach that increases the overall power bill.

The most obvious power source is electromagnetic induction, but there are other approaches that can work. Using the difference in temperature with thermal energy harvesting can be effective in the outdoor sub-stations and cellular base-stations where there is a significant temperature gradient. For energy sensors dotted around the home, vibrational energy can also be used.

The initial element of an induction design is relatively simple –a coil can be placed around a cable to induce a current that can then be used to power a sensor hub and wireless link. (Figure 2) Often this can be integrated into a ‘smart’ meter in the home that uses the same technique to monitor the total current flow. However, these are not used in sub-stations, and one of the key elements of the smart grid is the ability to monitor the flow of power throughout the network and direct it more efficiently to areas of higher consumption. Base-stations for mobile phones are also significant users of power. While operators are looking at using energy harvesting to power sensors for maintenance, additional sensors can also be used to feed back data into the grid on the power usage and requirements. Being able to install these without additional wiring and without needing maintenance allows the utility companies to overcome some of the main objections from the operators.

Figure 2: A typical energy harvesting system (Source: Cymbet).

Vibration

Vibrational energy harvesting makes use of the properties of the piezoelectric crystal. While traditionally a voltage across such a crystal makes it vibrate, the reverse is also true – when it vibrates, it generates a current. This current is then harnessed to provide the power for sensors and increasingly for wireless links. The power consumption of microcontrollers and wireless transceivers has dropped dramatically in recent years with new low power designs and new process technologies. This is coupled with a low duty cycle for the link that further reduces power consumption. All of this has brought the system power requirements down to a level that piezoelectric crystals can provide and opens this up as a practical design option.

These systems of course work best when there are lots of obvious vibrations, such as an industrial setting with motors and large, noisy equipment. However, the low level of vibration that is needed to produce the power means that the vibrations may be imperceptible and yet still be able to power systems. This is particularly relevant for sub-station and base-station implementations and even for cabinets at the side of the road that contain broadband and telephone equipment. It can also be used very effectively with household equipment that uses large amounts of power such as washing machines and tumble dryers.

Systems can be built from scratch using devices such as the V22BL (Figure 3) from Midé Technology. This is a hermetically sealed analog piezoelectric crystal for energy harvesting that works with vibrations as low as 26 Hz and up to 110 Hz. The first step in successful energy harvesting using such a device is to fully understand the vibration environment, and the best way to do this is to measure the vibration using an accelerometer. Once the data is captured, a Fast Fourier Transform (FFT) extracts the frequency information. Often this is directly associated with the equipment, such as a 120 Hz AC motor or a 60 Hz appliance, but most applications will require some form of vibration characterization. Midé offers a standalone vibration characterization device that can easily be installed into many different vibration environments to capture data in hard-to-reach areas. A built-in timer allows the capture of many different types of vibration and a simple USB interface allows the user to easily characterize any vibration. This can also help make sure that the crystal is not exposed to excessive vibrations that can damage the device and shorten its useful life.

Figure 3: The V22BL vibrational energy harvesting sensor from Midé Technology.

To support these piezoelectric transducers, Linear Technology has integrated a low-loss full-wave bridge rectifier with a high efficiency buck converter that is optimized for such high output impedance energy sources. The LTC3588-2 is an ultra-low quiescent current under-voltage lockout (UVLO) mode with a 16 V rising threshold to provide efficient energy extraction from piezoelectric transducers with high open circuit voltages.

Thermal

Thermal energy harvesting systems use pyroelectric sensors and the Seebeck effect, or the thermoelectric effect that comes from dissimilar metals being heated, to generate power. Pyroelectric sensors, built with thin film ferroelectric materials, can generate around a microamp of power with a 10 degree temperature difference to power devices. Increasingly, researchers are also looking at micro-machined systems (MEMS) with shapes in the order of a few microns that can maximize the thermoelectric effect.

Micropelt in Germany, for example, has developed a configurable thermal energy harvesting module that is aimed at the hundreds of millions of radiators across Europe. The module attaches magnetically and produces electric energy when put in contact with a heat source from about 10 °C above ambient air and provides a 2.4 V fixed output voltage with a power output ranging from 150 microwatts to 10mW, depending on the temperature differential. The thermoelectric chips are based on a patented scalable thin film micro-structuring platform technology.

The EDK312 (Figure 4) development kit from EnOcean supports both solar and thermal energy harvesting. The thermal version uses the ECT 310 low-cost ultra-low-voltage DC/DC converter with a standard low-cost Peltier element that can be soldered on to the back of a standard wireless module. This typically provides 20 mV from a 2 Kelvin temperature difference to power the wireless link and is being used in a variety of applications such as window contacts, temperature and humidity sensors, light sensors, pressure sensors, or gas sensors for battery systems.

The Enerchip EP CBC-EVAL-09 from Cymbet (Figure 5) is an evaluation system for thermal energy harvesting that uses a range of different energy sources. While it ships with a solar cell, it can also use a piezoelectric source or a thermal source. The DV164133 XLP 16-bit Energy Harvesting Development Kit from Microchip uses the Enerchip alongside its ultra-low power microcontroller for energy harvesting applications. The Microchip nanoWatt XLP PIC MCUs are ideal for these low power applications with sleep currents down to 20 nA, active mode currents down to 50 uA/MHz, code execution efficiency, and multiple wake-up sources. Powered only by light, the XLP kit enables rapid prototyping of low power applications such as RF sensors, temperature/environmental sensors, utility meters, remote controls, and security sensors.

Solar power

Solar power is perhaps the most obvious and well-established method of energy harvesting, and for some equipment, particularly outdoors, it is highly effective. The cost of photovoltaic (PV) solar cells has fallen dramatically over the last few years and the efficiency is creeping up to allow smaller cells to power sensors and wireless links.

Texas Instruments has developed a complete Solar Energy Harvesting development kit to help create a perpetually powered wireless sensor network based on the ultra-low-power 16 MHz MSP430 microcontroller. The Solar Energy Harvester module in the EZ430-RF2500-SEH development kit includes a high-efficiency solar 2.25 x 2.25 inch panel optimized for operating indoors under low-intensity fluorescent lights, which provide enough power to run a wireless sensor application with no additional batteries. It also has enough backup power to provide over 400 transmissions in the dark before lights are switched back on again. As with all these development systems, inputs are also available for external energy harvesters such as thermal, piezoelectric, or another solar panel.

NXP has developed the first dedicated IC for performing the Maximum Power Point Tracking (MPPT) function, designed for use in applications that use solar photovoltaic (PV) cells. To simplify development and maximize system efficiency, the MPT612 is supported by a patent-pending MPPT algorithm, an application-specific software library, and easy-to-use application programming interfaces (APIs). Dedicated hardware functions for PV panels, including voltage and current measurement and panel parameter configuration, simplify design and speed development. The controller is based on a low-power ARM7TDMI-S processor that operates at up to 70 MHz and can achieve system efficiency ratings up to 98 percent. It controls the external switching device through a signal derived from a patent-pending MPPT algorithm. The DC source can be connected to the IC through appropriate voltage and current sensors. The IC dynamically extracts the maximum power from the DC source, without user intervention, and can be configured for boundary conditions by setting them in software. There are up to 15 kB of flash memory available for application software.

Home systems

There is a wide range of applications in the home that can be part of the smart grid. The simple model sees a smart meter in the home with communication links back to the power grid to indicate what the power requirements are, and this can be done in the number of ways. A localized Zigbee wireless network, low data rate wide area GSM module, or a low power Bluetooth link to a broadband gateway in the home have all been proposed for this link, and all can be powered by inductive coupling, thermal (for a meter mounted on an outside wall) or vibrational energy source. However, there is much more to the smart grid that opens up opportunities for sensors all around the home. Either the smart meter or a home gateway can be the central controller for a wide network of devices, monitoring and controlling all kinds of equipment.

The main example is the ‘smart’ thermostat that links to sensors and controllers on heaters and radiators to provide the exact heat requirements around the home, and thermal energy harvesting is a natural source for such systems.

Adding sensors and controllers to equipment such as tumble dryers can allow the grid to decide when to switch it on, depending on the power requirements in the local area (with data also coming from other homes and the local sub-station). This kind of flexibility would give lower bills to the householder and spread the overall demand, reducing the peak requirements and the number of power stations required.

The ZigBee Alliance and Energy@home consortium in Italy are now working together on an integrated residential energy platform for Europe to control consumer smart appliances, communication with broadband networks, and communication with Automatic Meter Management systems. This is proposed by Electrolux, Enel, Indesit and Telecom Italia to help European consumers better manage energy use in their homes and is a prime example of the increasing need for control and sensors as part of a smart grid implementation.

Providing ‘free’ power for sensors to switch off lights and other equipment when the room is not being used also helps reduce power bills and is very much a part of the smart grid idea. This becomes even more compelling in public and commercial buildings where there is little or no compulsion for people to manage the energy supply. Over 65 percent of commercial buildings have no management systems and being able to install sensors for temperature and room occupancy, all linked to a central controller, is a key approach to reducing the overall power requirements. Allowing these sensors to be placed in the right location without rewiring and without having to regularly replace batteries saves millions of dollars in installation and maintenance costs and enhances the savings from the power reduction. The central controller then becomes part of the smart grid, feeding back vital usage data to the grid and allowing other sources, such as renewable energy, to be used more effectively.

For such systems, the ENERGY-HARVEST-RD Energy harvesting reference design from Silicon Labs combines the Si1012 Energy Harvesting Wireless Sensor Node with an EZRadioPRO USB Dongle to demonstrate an ultra low power wireless sensor powered from an energy harvesting source. It is designed to have a life expectancy greater than 15 years or 7000 mAh. The wireless node can be designed with a very thin profile, as the height of the rechargeable battery is just 0.17 mm.

The Sensor Node operates at 919.84 MHz and is powered by a solar energy harvesting power supply. When the wireless Sensor Node is not transmitting data, the MCU remains in a low-power state where it only consumes 50 nA. The leakage current of the energy harvesting supply, just 3 µA, is cancelled out by as little as 50 lux shining into the solar cell. This allows the energy harvesting supply to power the system for approximately seven days in a dark closet or indefinitely if there is a periodic light source. This means the system works both in indoor lighting of around 200 lux and in outdoor situations of 10,000 lux.

Industrial smart grid

The requirements for electrical sub-stations, renewable energy sources, and base-stations are quite different from the home environments. Monitoring of the back-up battery systems with temperature, humidity and gas sensing is increasingly important. Being able to add sensors exactly where they are needed without having to wire them in and avoiding the risks of high voltage short circuits dramatically increases the reliability and safety of the systems. Often these can work with thermal harvesting as the electronics in the base-stations or the converters in the electrical sub-stations generate sufficient heat. Vibrational systems also work well for wind turbine sources and also in sub-stations.

Conclusion

With the increasing demand on power around the world, the idea of analyzing power usage and reducing consumption is becoming more important. The smart grid is a key step towards this, adding in variable renewable energy sources, but requires more data from more sensors via wireless links. Process technology and ultra low power design now means smart meters and sensor modules with wireless links can be powered from thermal, vibrational, or light energy. This allows system and building designers and operators to place modules at the right point without having to install additional power lines. Feeding that data back into the grid both improves the analytics and allows more control over the provision of power. It also allows more control in the home and commercial buildings over electrical and electronic devices, allowing the grid to manage the provision of power even further. Being able to do all of this simply and without the maintenance cost of replacing batteries is a compelling benefit for operators and this is driving forward the use of a wide range of energy harvesting technologies in the smart grid.