Testing the emerging Smart Grid

Ken Christensen, Keysight Technologies


Staying ahead of a rapidly-growing critical infrastructure

The implementation of the “Smart Grid” is accelerating rapidly.  Once just a vision in eyes of power engineers and utilities, the Smart Grid is swiftly taking shape.  A major driver in realizing a truly smart electrical grid is the adoption of distributed energy resources (DERs).  Grid-connected distributed power production is an increasingly important component in the power production portfolio of utilities around the world. 

As more renewable energy production systems and grid-connected storage devices are installed and their power-producing capacity realized, not only has their generation has proven to be a viable source of electricity, in addition these DERs assist in power grid stabilization.  Intelligently coordinating and controlling the power from these DERs is essentially the Smart Grid.

When DERs such as wind and solar inverters, battery storage, diesel generation, and bi-directional Electric Vehicle (EV) chargers are intelligently coordinated and controlled, both grid operators and consumers benefit.   These controllable combined power generators a common called microgrids.   Microgrids can operate with both connected to the grid and when disconnected from the grid in an “island” mode.   Microgrids come in sizes from a few kWs in a residential application to tens of MW in a community application such as the San Diego Gas & Electric Berrego Springs microgrid.

Remote control

For the grid operator, these intelligent DERs, with modern power electronics, can be remotely controlled to regulate both their active and reactive power.  This allow the operator to change active power to mitigate abnormal frequency instances and change the reactive power to assist the grid during abnormal voltage instances. 

For utilities and government agencies also see huge benefits from microgrids.  Microgrids can be disconnected from the grid and provide power during emergencies and other power outages.  In the case of natural disasters, microgrids on schools and grocery stores can provide emergency power for lighting, communications, and food safety.  One such case of this is the microgrid designed by Advanced Solar Products for the Midtown Community School in Bayonne, NJ.   During Superstorm Sandy, this microgrid provided power for communication and lighting for days.  

Consumers can also benefit from microgrids.  By integrating battery storage and EVs with solar power, intelligent microgrids can be used to reduce or eliminate peak demand spikes and to store excess energy from mid-day peak production to later in the day when the Time of Use (TOU) charges from utilities can be nearly double that of mid-day.  

In a recent announcement from Tesla’s Elon Musk, he painted a vision of solar power generation from his SolarCity company combined with Tesla vehicles charging (and discharging) in the garage, and Tesla PowerWall battery storage all working together to optimize energy production.  He cleverly stated their symbiotic relationship when he said, “Solar and battery go together like peanut butter and jelly.”  In the same week that Musk spoke about microgrids, ConEd, the electricity provider for New York City announced that it will be deferring $1.2 billion on subsystem upgrades by deploying DERs to offset peaks which would have required major infrastructure enhancements.

Dropping cost

The cost of DERs has dropped dramatically over the past 5 years to a point where renewable energy production (Levelized Cost of Energy – LCOE) is on par with conventional power generation.  In fact, in most recent articles and discussion of DERs, the conversation is entirely on the positive financial aspects of integrating renewable energy with the Smart Grid, leaving the side benefit of saving the planet unmentioned!

The challenge now is for the producers of DERs to test and verify their products to be used on the electrical grid.   Many of the suppliers have never created a product that must comply with the safety standards and grid-interaction requirements of utilities and government entities.   The time and resource challenges of testing and qualifying DERs are an ever increasing burden on suppliers of DERs.  State-of-the-art laboratories are able to test a comprehensive variety of environmental conditions, grid electrical characteristics, and system components. 

This complex set of varying parameters is impossible to test in the real-world in a timely manner.  For example, testing the performance solar inverters during morning start-up conditions can only be done once per day, with one set of weather conditions and one type of solar module.   And if the conditions are not correct of if there is problem in the test setup, the test have to wait for a better day.  Controlled lab testing is the only reasonable way to meet the ever increasing time-to-market pressures on suppliers of DERs.

There is a gauntlet of testing required for any grid-connected power generator (see Table 1).   In many cases, each country (and sometimes each utility) has unique requirements.  These tests can be segmented in four areas (examples of each for solar inverters in North America are in parenthesis):

•          Product design verification

•          Safety / Electrical codes / Government Regulations  (NFPA70/NEC, FCC)

•          Industry Standards (IEEE 1547 / UL1741, CEC Inverter Efficiency)

•          Grid Connection Requirements (CAISO Rule 21, ERCOT, WECC)

Click image to enlarge

Table 1:  Typical DER standards and grid requirements

In addition, suppliers of distributed power generators are asked to perform complete characterization of the power generator’s performance in wide variety of operating conditions that can experienced when the device is in service for the life of the project which is typically over 20 years.  In all these operating conditions, engineers need to measure power quality, harmonics, energy production and power conversion efficiency.

Solar inverters are excellent example of a DER.  DERs are sometimes referred to as distributed generation (DG).  Testing to all of the areas listed previously typically consumes 30-40% of the timeline to develop a solar inverter.   Solar inverters must reliably produce power is the harshest environmental conditions; from blowing snow on a roof-top in northern Canada in January to blowing sand in a desert in the southwest US in August.  The inverter must prove to detect abnormal voltage and frequency conditions, and “ride-through” or disconnect at the discretion of the grid-operator. 

Efficiency is key

One of the most important as aspects to the financial viability a solar power generation station is how well solar inverter maximizes the conversion efficiency from DC to AC power.   Over the lifetime of the power generator, fractions of a percent of increased energy production can lead to $100s or $1000s of increased energy to $1Ms on utility scale projects.   With the thin margins in today’s energy economy, these financial gains often make the difference in a projects viability.

The overall conversion efficiency of solar inverters is dominated by two key components: Maximum Power Point Tracking (MPPT) efficiency, and DC-to-AC Power conversion efficiency.    The power generated by a solar panel depends on where it is operating location on its characteristic I-V curve.   An example of solar panel I-V curve is shown in Figure 1.   The current produced by a panel increases as the irradiation increases.   The voltage decreases as the temperature increases.   Throughout the day, temperature and irradiance are constantly change.  Accurately and efficiently tracking the Maximum Power Point (MPP) is the job of the control firmware embedded in the inverter.   To some degree, these control algorithms are the “secret sauce” in every inverter. 

Click image to enlarge

Figure 1: Typical I-V and Power Curves of a Solar Panel

To test the inverters ability to track the maximum power point of the solar array, a photovoltaic (PV) array simulator is used to emulate the behavior of a selected array technology.   A PV array simulator produces a power curve that matches a particular solar panel’s power production characteristics.   Each solar panel model has its unique I-V curve depending on its technology (thin-film, poly-silicon, …)  and solar cell design characteristics.    You can imagine the complexity and cost trying to test every possible solar panel with an inverter without a PV array simulator.  

Advanced PV Array simulators also automate the testing and reporting to EN50530 which is the industry standard for measuring MPP tracking performance.   The EN50530 protocol exercises the inverter’s MPP tracker in both static and dynamic modes to consistently measure the inverter’s MPPT efficiency.  The EN50530 takes nearly 7 hours to fully complete.  Automation of the EN50530 protocol and automatic generation of the reports specified in the standard increase the effectiveness of test engineers. 

DC-to-AC power conversion testing typically follows the Sandia Inverter Test protocol and/or IEC 61683, the international standard specifying the procedure measuring the efficiency of power conditioners.  These are complex procedures that requires an environmental test chamber and precision measurement instrumentation. 

As with all measurements, the test instrument must have better precision than the precision of the measurement you would like to make.  The peak conversion efficiency of most contemporary solar inverters is above 98%.  In the case of conversion efficiency measurements, engineers are trying to see 0.1% improvements in their designs.   A power analyzer with at least 0.05% accuracy is required make the DC input and AC output measurements.

To add to the complexity of all this testing, the solar inverter technology is continually changing.  To reduce component costs, installation costs and i2T losses, the DC voltage of solar arrays continues to increase.   Next generation solar power generators, including the solar panels and inverters, are operating at 1500Vdc.  Therefore, test laboratories must upgrade their power analyzers, PV array simulators, and power supplies to operate at this higher DC voltage.  

Keysight Technologies has created a test solution capable of testing next generation 1500Vdc inverters as well as 600Vdc and 1000Vdc inverters. This solution includes a Keysight N8937APV PV Array Simulator (see Figure 2) and a Keysight IntegraVision PA2203A Power Analyzer (see Figure 3) connected to a 12kW, 3-phase solar inverter. With this solution, engineers can simulate real-world environmental conditions and evaluate their effect on all aspects of their inverter. Combined with an AC grid simulator, this inverter test solution enables engineers to test for power quality issues, the quality of the generator’s power, and efficiency.  

Click image to enlarge

Figure 2: The Keysight N8900APV (60kW Configuration)

Click image to enlarge

Figure 3: Keysight PA2203A 4-Channel Power Analyzer

The test challenges of testing DERs individually can be complex and daunting.  Studying how all of the components in microgrids safely interact with each other and continue to abide by strict grid interaction requirements exponentially raises the complexity for test engineers.  As microgrids proliferate on the electrical grid, the qualification testing required by grid operators will no doubt increase.  

One of the primary responsibilities of electrical grid operators is to supply clean, reliable electricity to their constituents.   Without proper testing before connecting products to the grid, the grid becomes less reliable.   Look for test and measurement leaders to collaborate with suppliers of these devices to stay in front of the Smart Grid evolution in the power generation industry.

Keysight Technologies