ERCOT model quality tests

PG 6.2(5)(a) and DWG 3.1.1 to 3.1.4 · Models and platforms

What it means. Under the Planning Guide (PG) and the DWG Procedure Manual, each dynamic device needs a model whose parameters accurately represent it over the entire range of operating conditions, compatible with the current planning and operations software: PSS/E (version 35, moving to version 36) and PSCAD 4.5 or higher for planning, and TSAT for operations. Standard library models are preferred where they are accurate. A user-written model needs a model manual and the ERCOT model guideline checksheet; PSS/E user-written models are dynamic link libraries (DLLs), TSAT user-written models are in the TSAT user-defined model (UDM) format, and all PSCAD models count as user-written. Where several platforms are provided, responses must be consistent to the extent of each platform's capability. A model may not prevent the DWG from using an integration time step of a quarter cycle or less in positive sequence tools, and no model on the DWG's Unacceptable Model List may be used. Models go to both ERCOT and the interconnecting Transmission Service Provider (TSP) with populated model templates, and ERCOT or the TSP may reject data that is incomplete, inaccurate or inconsistent across platforms.

Evidence ERCOT expects. The model package with manuals, checksheets and templates, a statement of supported software versions, and dated transmittals to ERCOT and the TSP.

PG 6.2(5)(c) and DWG 3.1.5 · Model quality tests

What it means. MQT results and simulation files accompany every new or updated model, cover all site-specific models of the Facility, and can be run on a simple test system with a controllable infinite bus. For IBRs, WGRs and inverter-based transmission elements: a flat start of at least 20 seconds; a 3 percent POI voltage step up and down; a 0.3 Hz frequency step up and down from 80 percent dispatch, with Intermittent Renewable Resources run both curtailed with headroom and with no headroom, and ESRs also at 80 percent charging; low voltage ride-through using both the legacy profile (Nodal Operating Guide 2.9.1.2) and a voltage dip profile based on Section 2.9.1.1, and high voltage ride-through using the preferred curve (plus the legacy curve if the preferred is not met), each at 0.95 leading and 0.95 lagging power factor at the POI; a system strength test at short circuit ratios of 5, 3, 1.5 and 1.2, with acceptable responses required at 3 and higher and a technical reason if not acceptable at 1.5; and, for PSCAD only, a phase angle jump test to find the largest jump the model rides through. In the ride-through tests the model shall not exhibit momentary cessation; after a low voltage event, real power should recover to full output within 1.0 second of POI voltage recovery, and during high voltage it should be sustained. PSS/E, TSAT and PSCAD results are overlaid on the same axes; TSAT testing is not required for a model using a TSAT standard library model. The owner also assesses sensitivities, including POI voltage set point and real and reactive output, and explains any mismatch between platforms.

Evidence ERCOT expects. The MQT report with overlaid plots and simulation files for each platform, leading and lagging cases for the ride-through tests, the maximized low and high voltage ride-through (LVRT and HVRT) capability documented in the format of the Nodal Operating Guide tables, and a written explanation, with any exempt status, for each criterion a model does not meet.

PG 6.2(5)(c)(iv) and DWG 3.1.5.10 to 3.1.5.15 · Advanced Grid Support ESRs

What it means. For inverter-based ESRs required to comply with Nodal Operating Guide Section 2.14, the test set is flat start, an AGS small voltage disturbance test, a frequency change and inertia response test, low and high voltage ride-through, an AGS system strength test with multiple short circuit ratios, an AGS phase angle jump test and a loss of synchronous machine test. Section 2.14 asks the ESR to show, within its inverter current limit, that it can maintain an internal voltage phasor in the sub-transient-to-transient timeframe and keep synchronism with the ERCOT Transmission Grid. The loss of synchronous machine test is not intended to require the Facility to operate without connection to the ERCOT Transmission Grid.

Evidence ERCOT expects. The AGS MQT report with the required plots for each test, and confirmation in the unit model validation report that the AGS control settings are enabled.

PG 6.2(5)(d) and DWG 3.1.6 · Unit model validation

What it means. IBRs include, with any PSCAD model submitted, results showing that the PSCAD model accurately represents all inverter-based dynamic devices in the Facility. This is a hardware type test, not a site-specific study: the model is compared with actual inverter hardware programmed with typical default settings, and results for one inverter model can be reused wherever that inverter is used. The tests are a voltage step, voltage ride-through, system strength, phase angle jump, and a subsynchronous frequency scan from 5 to 55 Hz in 1 Hz steps at a short circuit ratio of 10 at unity power factor and at 1.5 with 0.95 lagging and 0.95 leading power factor. Energy storage is tested charging and discharging, and auxiliary inverter-based equipment such as static synchronous compensators (STATCOMs) is also tested.

Evidence ERCOT expects. A validation report per inverter type with the test setup, overlaid hardware and model plots, and the subsynchronous resistance and reactance as both plot and table on the inverter MVA base.

PG 5.3.2 and 5.3.5 · Interconnection milestones

What it means. To start the Full Interconnection Study (FIS) through ERCOT's Resource Integration and Ongoing Operations (RIOO) system, the Interconnecting Entity submits complete Resource Registration data, including the dynamic model and MQT results and simulation files, subject to verification by the lead TSP with approval from ERCOT. For the quarterly stability assessment, the final dynamic model is due at least 45 days before the prerequisite deadline; if ERCOT cannot finish its review by that deadline, the resource is not included, and a large generator left out is not eligible for Initial Synchronization in that three-month period. Typical manufacturer data may be used before detailed design, but the model is updated once design is complete and again at commissioning.

Evidence ERCOT expects. RIOO submission records, the dated final model submittal, and ERCOT's notifications.

PG 5.5(4) to (6) and 6.2(5)(b) · Commissioning and continuing operations

What it means. Before the Resource Commissioning Date of an IBR, the Interconnecting Entity sends ERCOT the as-built models and the models used in the quarterly stability assessment, a list of differences, as-built MQT results overlaid on the assessment results, and simulation files. ERCOT responds within ten Business Days, extendable by 20. Within 30 days after the Resource Commissioning Date, the Resource Entity submits as-built model updates with MQTs through RIOO and a plant verification report showing that commonly tuned parameters match field settings. Further verification reports are due within 30 days after a settings change, 12 to 24 months after the later of the Resource Commissioning Date or March 1, 2021, and at least every ten years. Before changing any IBR control setting or equipment that affects dynamic response at the POI, the Resource Entity submits the new model with before-and-after MQT overlays; ERCOT may find the change acceptable, unacceptable, in need of a TSP stability study, or subject to a new Generator Interconnection or Modification (GIM) request. Model updates are also due within 30 days of any facility change or test result that requires one.

Evidence ERCOT expects. Plant verification reports supported by delivery and testing reports, screenshots or photos of actual settings, or OEM attestations; dated submissions; and ERCOT's responses and determinations.

For inverter-based plants

These requirements apply to all generator Facilities and dynamic devices, but the ride-through, system strength, phase angle jump and unit model validation tests and the PSCAD requirement apply to inverter-based resources, WGRs and inverter-based transmission elements. Synchronous machines run a shorter test set with a four-cycle fault instead of the ride-through profiles.