Frances Cleveland fcleve@xanthus-consulting

1. SIWG Phase 3Details on Advanced Functions for Distributed Energy Resources (DER) SystemsMarch 31, 2017 – Final Frances Cleveland fcleve@xanthus-consulting.com

2. Contents • Decisions on SIWG Phase 3 Functions • For historical purposes, detailed slides of each function that were used to help the discussions

3. SIWG Phase 3 Discussions and Recommendations of 8 Functions (1)

4. SIWG Phase 3 Discussions and Recommendations of 8 Functions (2)

5. Functions Discussed and Finalized for Phase 3

6. Monitor Key DER Data • Monitor Key DER Data: All DER systems shall have the capability to provide key DER data at the DER’s ECP and/or at the PCC. • Utilities shall define in the Utility Handbooks when and under what conditions the data exchange requirements shall be provided, including what types of data, whether and how it may be aggregated, frequency of monitoring, time latency, etc. • Key data requirements include as a minimum the data items listed in the SIWG Phase 2 document. These cover: • Administrative Data: DER system identification, facility identification, updates to nameplate information, updates to DER ratings, indications of which functions are supported, and other essentially static data. • Monitored Data: Individual and/or aggregated DER state of readiness – define this more clearly (on/off, changes from nameplate, major alarm that would take it off line), real-time measurements, metered data, and any future states that deviate from planned or scheduled states. • Error conditions: If the mutually agreed upon exchanges of data are not taking place within the agreed upon time latency and completeness, these conditions shall be reported. • Discussion: • Joint IOUs has prepared a list of alarms for inclusion within IEEE 1547. These should be included in Rule 21 either by itself or through pointing to IEEE 1547. Or would the protocol identify the list of alarms? Or in a requirements document such as Rule 21, IEEE 1547 or a document pointed to by one of those documents, such as the CSIP document. Must ensure that the DER must be able to provide those alarms. Or a separate document so that other protocols will also be required to support them. • What about State of Charge? Or kWh available plus max discharge rate? SoC is percent of Usable Capacity. Or both? IEEE 1547 looks like it will go with kWh • Decisions: • Use list of data from IEEE 1547 • Include the list of alarms in an Annex or separate document • State of Charge will be represented as available kWh • Rates of charging/discharging may change depending upon state of charge. • List of alarms, status, and measurement data for forecasting (per IOU action item)

7. IOU Alarm List Alarms, status, and measurement data can also be used for forecasting. This data, if and where available, could affect and improve future operations (something is failing, planned maintenance, short term planning) Action Item: IOUs to determine what data should be included

8. Dynamic Reactive Current Support Mode • The Dynamic Reactive Current Support mode shall provide reactive current support in response to dynamic variations in voltage (rate of voltage change) rather than changes in voltage. Details of the function will be provided by IEEE 1547. • IEEE 1547 Draft 6.1 includes requirements for 6.3.2.5 Dynamic voltage support – These are mostly descriptive and not quantitative – Do these requirements meet our Phase 3 requirements? • Key requirements include: • Enable and Disable settings for the dynamic reactive current support mode shall be provided. When enabled, the DER shall respond to voltage variations at the Referenced Point by modifying reactive current according to the mode settings. When disabled, the DER shall revert to a previously defined state at the established ramp rate. • Acknowledge and/or monitor the data at the Referenced Point: Receipt of the mode parameters and the enable/disable commands shall be acknowledged or the power measurements at the Referenced Point shall be monitored. • Error conditions: If the dynamic reactive current support mode requirements cannot be met or are not being met, this condition shall be reported.

9. Jens: Dynamic Reactive Current Support Mode This function is optional in P1547. It was intentionally not specified in detail in order to give latitude for grid operators and DER vendors to implement reasonable and effective controls, considering the potential benefits and caveats of this function. It is important to discuss the objective of this function. I can see three objectives: • 1.       Mitigate fault-induced delayed voltage recovery (FIDVR). • 2.       Maintain legacy DER during low voltage conditions online that would otherwise trip. • 3.       Prevent 1-phase induction motors from stalling in the initial fault period. Furthermore, the potentially adverse impact of this function on distribution system planning and operation should be discussed. Possible caveats may include: • 1.       Complication of coordination of distribution protection with DER current contribution during faults on the feeder. • 2.       Potential creation of overvoltage at fault clearing. • 3.       Interaction with inverter’s anti-islanding detection. • 4.       Potential creation of overvoltage in “healthy” phases during unbalanced fault conditions if this function is implemented in the positive sequence and without negative sequence current control. Due to the complexity of this topic, the latest draft of P1547 includes this function as optional. A future revision of P1547 may include specifications of this function as mandatory. EPRI is currently undertaking detailed EMT-type simulation and analysis to inform this discussion. Results are expected for summer 2017.

10. Dynamic Reactive Current Support from IEC 61850-90-7 • Additional key requirements include the following basic requirements: • The minimum voltage deviation relative to the average voltage, expressed in terms of % of VRef • The maximum voltage deviation relative to the average voltage, expressed in terms of % of VRef • The gradient, expressed in unit-less terms of %/%, to establish the ratio by which capacitive % Var production is increased as %delta-voltage decreases below DbVMin • The gradient, expressed in unit-less terms of %/%, to establish the ratio by which Inductive % Var production is increased as %Delta-Voltage increases above DbVMax • The time, expressed in seconds, over which the moving linear average of voltage is calculated to determine the Delta-Voltage • Additional possible settings include: • The selection setting that identifies whether the dynamic reactive current support acts according to the basic method or the alternative method • The voltage limit, expressed in terms of % of VRef, used to define a lower voltage boundary, below which dynamic reactive current support is not active. • The hysteresis added to BlkZnV in order to create a hysteresis range, expressed in terms of % of VRef. • The time (in milliseconds), before which reactive current support remains active regardless of how deep the voltage sag. • Enable/Disable Event- Based Behavior, the selection of whether or not the event-based behavior is enabled. • Dynamic Reactive Current Mode, the selection of whether or not watts should be curtailed in order to produce the reactive current required by this mode. • The time (in milliseconds) that the delta-voltage must return into or across the dead-band before the dynamic reactive current support ends, frozen parameters are unfrozen, and a new event can begin.

11. Dynamic Reactive Current Support Mode Diagrams from IEC 61850-90-7

12. Alternative of Dynamic Reactive Current Support Diagrams from IEC 61850-90-7

13. Scheduling Power Values and Modes • Schedules shall be capable of setting active and reactive power values as well as enabling and disabling DER modes for specific time periods. • Key requirements include: • Schedule consisting of an array of time periods of arbitrary length that define the offset from a starting date and time. • Scheduled value or mode: Each time period shall be associated with a real or reactive value or shall indicate which mode, which set of parameters for the mode, and whether to enable or disable the mode. • Starting date and time: The start date and time shall be provided before the schedule is enabled. • Referenced Point identifier: The identity of the Referenced Point shall be provided where the relevant measurements or calculations are provided for the PCC or other Referenced Point. • Time Window within which the value or mode shall be achieved or a Ramp Rate shall be settable. A time window of 0 seconds or a ramp rate of 100% shall indicate immediate action. • Schedule repeat interval: Schedules shall be able to be repeated periodically. • Schedule event trigger: Schedules shall be able to be initiated by an event • Multiple schedules which may be active at the same time shall be supported • Schedule priority to determine which schedules take precedence if they overlap with mutually exclusive requirements. • Schedule ending process: When a schedule ends, the default state of the DER shall be reverted to, with any ramping or other settings to arrive at that default state. • Enable and Disable settings for the schedules. When a schedule is enabled, the schedule shall take effect at the first scheduled time. The DER shall then modify its output to achieve the scheduled value at the established ramp rate. When a schedule ends or is disabled, the DER shall revert to a previously defined state at the established ramp rate. • Acknowledge and/or monitor the data at the Referenced Point: Receipt of the mode parameters and the enable/disable commands shall be acknowledged or the power measurements at the Referenced Point shall be monitored. • Error conditions: If the schedule requirements cannot be met or are not being met, this condition shall be reported. • Additional scheduling capabilities may optionally be supported, such as providing pricing signals for different scheduled times.

14. Scheduling Constructs and Priorities • IEEE 1547 does not address scheduling, although Phase 2 communications support scheduling • Should Scheduling be included in Rule 21? • How should the SIWG address this? • Minimal scheduling requirements, such as: • schedules must be capable of scheduling modes • support # schedules • x minute granularity • able to execute x number of schedules at one time … • Priority handling of conflicting modes • How can schedules be overridden and how restarted? Example of Scheduling from IEC 61850-90-10

15. Command DER to Cease to Energize or Return to Service • DER Cease to Energize: The cease to energize request shall cause a “cease to energize” state at the ECP or optionally shall allow the opening of a switch at a Referenced Point. The cease to energize shall cause the DER to output zero active current flow and (close to zero) reactive power flow. Key requirements include: • Cease to energize request shall cause the DER to enter the cease to energize state. • Referenced Point identifier: The identity of the Referenced Point shall be provided where the cease to energize state shall be applied. If none is provided, the default is the DER’s ECP. • A ramp rate or time window shall be settable. A time window of 0 seconds or a ramp rate of 100% shall indicate immediate action. • Reversion time shall be included determining when the DER can return to service if communications are not available. • Acknowledge and/or monitor the data (export of power or switch status) at the Referenced Point: These requests shall either be directly acknowledged or the switch status at the Referenced Point shall be monitored. • Error conditions: If DER did not cease to energize at the Referenced Point, this condition shall be reported. • DER Return to Service: The return to service request shall end the “cease to energize” state or shall initiate the closing of the DER switch at the Referenced Point. Additional key requirements include: • Ramp rate or a time window for random return to service shall be settable. • “Permission to return to service” shall be supported to allow actual connection to take place at some later time. • Acknowledge and/or monitor the data (export of power or switch status) at the Referenced Point: These requests shall either be directly acknowledged or the switch status at the Referenced Point shall be monitored. • Error conditions: If DER is not ready or capable of returning to service at the Referenced Point, this condition shall be reported. • IEEE 1547 D6

16. Disconnect versus Cease to Energize • The Disconnect command causes galvanic isolation, typically via a switch. • Cease to energize implies that no power is exported. • What still needs to be resolved on cease to energize is: • Just follow IEEE 1547’s current definition? • Whether each DER must cease to export active power or whether just the net export at the PCC must be zero • Whether vars may be exported • Whether energy storage systems may still charge

17. Limit Active Power Mode Diagram • Discussions: • IEEE 1547 Draft 6.1 requirements below • Phase 3 recommendations are for a percent of the maximum active power capability with no discussion on supplying loads in the Local EPS • Concern that utility may not be able to validate the compliance to this requirement. • In Hawaii, it is combined with load following measured at the PCC or the DER terminals. If at the PCC, need to measure the active power at the PCC. • Utility needs to know which method will be used – or specify at interconnection time or through communications. A utility command for reliability reasons is different from demand response situations. It would also be useful to have real-time response as to which is used 4.5.2 Response to active power limit set points The DER shall be capable of responding to an active power limit set point as a percentage of the maximum power setting and shall limit its active power output to the active power limit set point in no more than 30 seconds or the primary energy source response time whichever is greater. In cases where the DER is supplying loads in the Local EPS, the active power limit set point may be implemented as a maximum active power export limit set point. The DER shall not be required to reduce active power below the level needed to support local loads.

18. Limit Maximum Active Power • The Limit Maximum Active PowerPercent mode shall limit the active power level at the Referenced Point as a percent of the maximum active power capability, • The Limit Maximum Active PowerLevelmode shall limit the active power level at the Referenced Point to a specific active power value. • Key communication requirements to include??: • Referenced Point identifier: The identity of the Referenced Point shall be provided where the active power is measured or calculated for the PCC or other Referenced Point. • Accuracy: Delta active power allowed to exceed the limit and time allowed to exceed the limitshall be settable, indicating the precision required for the functional requirements to be met. • A Ramp Rate or Time Window within which the active power limit shall be met shall be settable. A time window of 0 seconds or a ramp rate of 100% shall indicate immediate action. • Reversion Timeout in seconds shall be settable, after which the active power limit is removed. A reversion timeout = 0 means that there is no timeout. • Enable and Disable settings for the Limit Maximum Active Power mode shall be provided. When enabled, the active power at the Referenced Point shall be limited to be within the percent or level established. When disabled, the DER shall revert to a previously defined state at the established ramp rate. • Acknowledge and/or monitor the data at the Referenced Point: Receipt of the mode parameters and the enable/disable commands shall be acknowledged or the active power at the Referenced Point shall be monitored. • Error conditions: If the commanded limit at the Referenced Point cannot be met or is not being met, this condition shall be reported.

19. Set Active Power Mode • Set Active Power Mode is not exactly the same as Limit Active Power mode for DER systems that can control their active power output (such as energy storage, synchronous generators, etc.), • Question: Treat this function like Limit Active Power, or include it as a separate function? It is already used for Reg Up and Reg Down. It can be used to offset high loads at substations. Do we want it in Rule 21? We do want to have a standard way of doing this function even if not explicitly for reliability or safety reasons. • the Set Active PowerPercentmode shall set the active power level at the Referenced Point as a percent of the maximum active power capability, and/or • the Set Active PowerLevelmode shall set the active power level at the Referenced Point to a specific active power value. • Key communication requirements include: • Referenced Point identifier: The identity of the Referenced Point shall be provided where the active power is measured or calculated for the PCC or other Referenced Point. • Accuracy: Maximum delta active power allowed to deviate from the required setting and the time allowed to deviate from the setting shall be settable, indicating the precision required for the functional requirements to be met. • A Ramp Rate or Time Window within which the active power level shall be met shall be settable. A time window of 0 seconds or a ramp rate of 100% shall indicate immediate action. • Reversion Timeout in seconds shall be settable, after which the active power limit is removed. A reversion timeout = 0 means that there is no timeout. • Enable and Disable settings for the Set Active Power mode shall be provided. When enabled, the active power at the Referenced Point shall be set to the percent or level established. When disabled, the DER shall revert to a previously defined state at the established ramp rate. • Acknowledge and/or monitor the data at the Referenced Point: Receipt of the mode parameters and the enable/disable commands shall be acknowledged or the active power at the Referenced Point shall be monitored. • Error conditions: If the commanded active power level at the Referenced Point cannot be met or is not being met, this condition shall be reported.

20. Issues Related to Set Active Power Mode Set Maximum Generation Rate:The maximum power rate at which the DER may be generating energy in Watts. Set Maximum Consumption Rate:The maximum power rate at which the DER may consume energy in Watts [implicit by using negative number]. Definition of State of Charge (SoC): Amount of energy that is stored for use, typically a percentage from 0% to 100% of usable capacity Definition of usable energy capacity: Maximum % of Actual Capacity minus Minimum % Actual Capacity Discussions: Available kWh, rather than SoC

21. Frequency-Watt Abnormal and/or Smoothing Mode • The Frequency-Watt Emergency mode shall counteract frequency excursions during H/LFRT events by decreasing or increasing active power. The change in active power may be provided by changing generation, changing load, or a combination of the two. Details of the function will be provided by IEEE 1547. • Key communication requirements include: • High and low frequency threshold to initiate changing active power: This mode applies to both decreasing active power output on high frequency and increasing active power output on low frequency for units that can provide that capability at that point in time. • Rate of active power change shall be settable. • High and low frequency stop settings at which to stop changing active power, including a ramp rate. • Hysteresis: If hysteresis is enabled, then the rate of change is also set for returning from the hysteresis level to the normal active power level. • Enable and Disable settings of the Frequency-Watt Emergency mode shall be provided. When enabled, the DER shall counteract frequency excursions during H/LFRT events by decreasing or increasing active power. • Acknowledge and/or monitor the data at the Referenced Point: Receipt of the mode parameters and the enable/disable commands shall be acknowledged or the active power at the Referenced Point shall be monitored. • Error conditions: If the frequency-watt emergency mode requirements cannot be met or is not being met, this condition shall be reported. • Use of this Frequency-Watt function for frequency smoothing during normal operations shall be permitted but is not mandatory. • Jens: The parameters of this function in P1547 were chosen to operate during normal operation. Other countries, like Germany, still use this function in emergency conditions only, e.g., when frequency deviates more than 100 mHz from nominal. • Discussions: • Should this function be used only during normal operation or also during emergency conditions? • Are the IEEE 1547 values acceptable? • Are other capabilities of interest?

22. IEEE 1547: 6.4.2.6 Frequency Droop (frequency/power)

23. European Frequency-Watt Mode Diagrams Reducing power with increasing frequency (and with hysteresis as an option? – utilities to discuss further) ENTSO-E: Limits on power reduction with decreasing frequency ENTSO-E: Increasing power with decreasing frequency Figure 19: Active power frequency response capability of power-generating modules in LFSM-U Figure 18: Maximum power capability reduction with falling frequency

24. Volt-Watt Mode • The Volt-Watt mode shall respond to changes in the voltage at the Referenced Point by decreasing or increasing active power. The change in active power follows a piece-wise-linear characteristic and may be provided by changing generation, changing load, or a combination of the two. Details of the function are provided by IEEE 1547.

25. Volt-Watt Mode Diagrams IEEE 1547 includes absorbing active power

26. Concept of Electrical Connection Points (ECP) and the “Referenced (ECP) Point” needed for many Modes

27. Slides Used in Earlier Calls

28. Current Status of SIWG Phase 3 • Workshop November 17, 2016 • Main debate: Given the significant overlap between the eight Phase 3 functions and the proposed revision to the international IEEE 1547 standard, should California move forward with their own requirements or adopt IEEE 1547 once the revision is complete? It was estimated that the revision would be complete somewhere between Q4 2017 and Q4 2018; however, there is no guarantee of this timeline. • Energy Division staff offered the perspective that California can and should remain in a leadership role in the development and implementation of advanced smart inverter standards. Energy Division acknowledges that coordination with IEEE 1547 is valuable; however, such collaboration is contingent on a reasonable time period so that California can continue to move forward • CPUC DER Action Plan Continuing element 2.8 as a guiding vision for the SIWG to help the CPUC achieve: • 2.8. By 2020, fully operationalize advanced smart inverter functionalities to enhance the integration of DERs into the grid.

29. The Eight (8) SIWG Phase 3 Functions • Monitor Key DER Data: DER systems identified by utilities during the interconnection process shall have the capability to provide key DER data at the DER’s electrical connection point (ECP) and at the point of common coupling (PCC) (through the meter), including key administrative, status and measurements on current energy and ancillary services; • DER Disconnect and Reconnect Demand: The disconnect command shall either cause a “cease to energize” state or shall initiate the opening of the DER switch referenced in the ECP in order to galvanically isolate the DER system from the Local or Area EPS, while the reconnect command shall initiate the closing of the DER switch at the referenced ECP or shall end the cease to energize state; • Limit Maximum Active Power Mode: This mode shall limit the maximum active power level at the referenced ECP either as a percent of the maximum active power capability or to a specific active power value; • Set Active Power Mode: This mode shall set the active power level at the referenced ECP as a percent of the maximum active power capability or to a specific active power value; • Frequency-Watt Emergency Mode: This mode shall provide settings to counteract frequency excursions during high or low frequency ride-through events by decreasing or increasing active power; • Volt-Watt Mode: This mode shall set the volt-watt curve parameters necessary to respond to changes in the voltage at the referenced ECP by decreasing or increasing active power; • Dynamic Reactive Current Support: This mode shall provide reactive current support in response to dynamic variations in voltage (i.e., rate of voltage change) rather than changes in voltage; and, • Scheduling Power Values and Modes: Schedules shall be capable of setting real and reactive power values as well as enabling and disabling DER modes for specific time periods.

30. Stakeholder Analysis of Functions, Late 2016

31. Additional Modes Not Selected for Phase 3

32. Active Power Following Mode • The Active Power Following mode shall follow load and/or generation at the Referenced ECP (PCC) • Parameters include the Active Power Threshold and the Active Power Following Percentage for the referenced ECP. There could active power following by a percentage. • A Time Window or a Ramp Rate shall be settable within which the active power level shall be met. A time window of 0 seconds or a ramp rate of 100% shall indicate immediate action. • A Reversion Timeout in seconds shall be settable, after which the active power output limit is removed. A reversion timeout = 0 means that there is no timeout. • Allowed Time and Active Power Amount to exceed settings. • The capability to Enable and Disable the ECP Active Power Following mode shall be provided. When enabled, the DER shall counteract the active power at the referenced ECP if it exceeds the mode’s Active Power Threshold, using the mode’s Following Ratio. • Acknowledge and/or allow data at the referenced ECP to be monitored: Receipt of the mode parameters and the enable/disable commands shall be acknowledged or the active power at the referenced ECP shall be monitored. If the active power following mode requirements cannot be met, this condition shall be reported.

33. Active Power Following Mode Diagrams: Load Following and Generation Following

34. Dynamic Volt-Watt Function (Not Voltage Droop) • Dynamic Volt-Watt function permits DER systems to respond to large voltage deviations from the nominal voltage level by reducing or consuming active power on overvoltage, or by producing additional active power on undervoltage. It is expected to be used primarily as an alternative to emergency tripping and may be used in conjunction with the volt-var function. (Chase) Mitigate overvoltage in small conductors due to over generation which is causing reverse power flow. • Or is it for transient events? • Focus on Energy Storage as most effective DER type? • Volt-Watt curve is expressed as Watts (%WMax) / Volts (%VRef), • Voltage calculated as a moving average over a preselected filtering time period. A deadband, probably wider than ANSI Range A, should be included in the curve to minimize unnecessary changes in active power. Need a delay to avoid responding too fast to motor starts, etc. • Need “Use Cases” for the uses of the Volt-Watt function • See EPRI Report, “Common Functions for Smart Inverters, Version 3”, Section s 2 & 18, Report No. 3002002233, 2014