Summary of NASPI Technical Report on CPOW Data#

This post provides an overview of Continuous Point on Wave Data (CPOW) from the NASPI Technical Report (2020). The original report can be found here: https://www.naspi.org/node/819

A Summary of the 2020 North American SynchroPhasor Initiative (NASPI) Technical Report

Continuous Point on Wave Data (CPOW) capture the highest resolution of time series data for power systems monitoring when compared to PMU (synchrophasor) or SCADA data. Current practices of power systems monitoring often use SCADA at a sample rate of 4-6 second intervals or else PMUs which samples at rates of 30-60 measurements per second. This NASPI paper advocates for the deployment of continuous point on wave monitors, (CPOW) taking measurements at a rate of 256 samples per second or faster. SCADA monitoring can generate 21,600 records per day, while a PMU over 5 million and a CPOW 124 million (pg. 31)! Similarly to how PMUs did not replace SCADA immediately but instead gradually have become deployed on the grid, it is expected that CPOW sensors will complement PMUs rather than replace them. Current data protocols used to stream real-time PMU data may also have quality issues due to data losses from using the User Datagram Protocol protocol (UDP) else latency issues using the Transmission Control Protocol (TCP). While the former provides faster transferring for a large volume of real-time data, TCP is capable of retrieving lost data packets while UDP is not. An excited advancement, The Department of Energy (DoE) has funded the Grid Protection Alliance to develop the Streaming Telemetry Transport Protocol (STTP) designed to transmit PMU or POW data with lossless compression.

This NASPI report advocates for continuous recording of grid conditions with CPOW rather than event-triggered continuous recording of grid conditions. This data is then retained data for post event analysis, real time streaming ability, and archiving of streaming data.

Furthermore, continuous point on wave data must meet three characteristics:

  • Waveform sampling– sampling of the actual, analogue waveform as opposed to fitting the measurement to a sinusoid wave form (sine wave).

  • High data accessibility– even when systems may go down, data must be readily accessible, likely through cloud storage.

  • Time Synchronization- time-stamped data should be uniformly synchronized across sensors via Coordinated Universal Time (UTC). When locally collected time stamps aren’t verified across wide geographic areas, this can be problematic for modeling if samples are not synchronized.

In short, long-duration, high-resolution data allow us to understand the complexities of the grid at a holistic level previously unseen. According to NASPI, “The lack of high-resolution, longer-duration, archived CPOW data is limiting our ability to understand and diagnose high-speed grid conditions and events” (NASPI, pg.5). Below we can see the difference in resolution between CPOW and PMU data. As we can see, the POW measurement is a significantly higher resolution, capturing details that phasor approximation misses.

(NASPI, pg. 11)

Problems from Multiple End-Uses and DERs

The energy mix is a group of different primary energy sources that are converted into the secondary energy which we use. Historically, this system consisted of a limited number of generation sources, such as coal-fired power plants supplying the primary source of energy for a given municipality. When alternative energy generation sources are deployed across the grid like solar, hydroelectric, and wind, this significantly complicates system dynamics. These distributed energy resources (DERs) are essential for propelling the green energy transition forward. Within the United States the DER contributions to the energy mix now approaches 10% (pg. 19). Furthermore, DERs are growing faster than any other energy generation source.

With the adoption of DERs, both industrial sized generation operations and even individual consumers seek to sell energy back to the grid. Problematically, monitoring devices are not capturing needed data on fast events, such as inverter-related resource behaviors that can cause faults. Understanding how inverter health relates to system health is an area ripe for further study. Furthermore, there are interesting use cases beyond monitoring inverters which convert direct current (DC) generated by DERs to the alternating current (AC) required to transmit electricity across the grid! Geomagnetic disturbances and wind turbine oscillations are examples of issues that can lead to unexpected power plant faults.

Notedly, Digital Fault Recorders (DFRs) and Dynamic Disturbance Recorders (DDRs) do recording high resolution data at 960 samples per second and >30 samples per second respectively, however, these measurements are event triggered, meaning that they collect data in windows rather than continuously. Since they are triggered in the case of predetermined event-conditions, they are not suitable to investigate with algorithms the very events in which they may need to be triggered (pg.19). Continuous data is therefore necessary to understand transient stability failures that can occur within just a few cycles in areas with high penetration of inverted-based generation sources.

Effects of inverted based resources can include:

  • Voltage fluctuations

  • Reverse power flows

  • Low-fault currents

    Shown below, applications for monitoring are required due to the changing character of end-use loads. As we transition from things end uses like incandescent lighting the LEDs and battery storage, the dynamics of a wide-area grid can change drastically.

Geomagnetically induced currents (GIC) are also an interesting use case for CPOW data analysis. Hydro Quebec found that “fine-grained, localized analysis of geomagnetic activity within a three hour time window can be used to predict potential impact at nearby locations on the grid” and can alert operators in advance of large-scale geomagnetic disturbance (GMD) events.

Finally, the report recommends that this higher-resolution data are combined with presently used monitors (SCADA, PMU) to provide a holistic understanding of the grid. Additionally, the NASPI paper argues for streaming CPOW data to a central source, since in the event of a fault, a local system may not be capable of operating, analyzing the event, and intervening. More uses for combining data types are outlined below:

In conclusion, CPOW data are a much higher resolution than PMU data. As DER’s are rapidly integrated into the grid, inverter-specific use cases may become increasingly necessary for research. Moreover, exciting use cases such as cybersecurity, DER integration, wildfire prevention, solar forecasting and others are quickly becoming of interest to power system engineers. These modern solutions will require joining many types of data together to analyze system dynamics, and CPOW data provides the clearest resolution to make this analysis possible.