Virtual Power Plant (VPP) Demonstrations
AEMO is collaborating with the Australian Renewable Energy Agency (ARENA), the Australian Energy Market Commission (AEMC), the Australian Energy Regulator (AER), and members of the Distributed Energy Integration Program (DEIP), to establish the VPP Demonstrations.
These demonstrations are the first step in a broad program of work designed to inform changes to regulatory frameworks and operational processes so Distributed Energy Resources (DER) can be effectively integrated into the National Electricity Market (NEM), maximising value to consumers while also supporting power system security.
A virtual power plant (VPP) broadly refers to an aggregation of resources (such as decentralised generation, storage and controllable loads) coordinated to deliver services for power system operations and electricity markets.
November 2020 Update – VPP extension confirmed
On 30 July 2020, AEMO confirmed the extension of the VPP Demonstration to 30 June 2021 conditional on obtaining three (3) additional participants, bringing the total number of VPPs participating in the demonstration to five.
As at 1 November 2020, AEMO can confirm that this condition has been met and that the VPP Demonstration will continue to 30 June 2021
With these additional participants, the VPP Demonstration has obtained diversity in technology being utilised, and location of the VPPs which will provide valuable insights and learnings into the technical, market and consumer impacts of DER participation in the contingency Frequency Control Ancillary Service (FCAS) markets. This information will inform the DER MASS Consultation, with consultation planned for early 2021, and transitioning the demonstration to a fully operationalised end state.
Throughout FY21, AEMO will continue to work with industry and consumer groups to explore and consult on the ongoing arrangements required for DER to participate in the contingency FCAS markets
VPP Demonstrations Insights
Since the first VPP participant joined in September 2019, AEMO has had the opportunity to observe its behaviour during a number of events. AEMO is committed to knowledge sharing to distribute the key learnings from the VPP Demonstrations to a broad audience.
The first such knowledge sharing document published in conjunction with ARENA, which includes early lessons learnt and next steps, is below.
Since the first knowledge sharing report, AGL has also joined the VPP Demonstrations. The second knowledge sharing report below includes a range of insights including information from both Energy Locals and AGL:
In addition to data insights, AEMO in conjunction with ARENA are undertaking a consumer experience study. Key findings from this study, and further insights from the data available will be published in the third knowledge sharing report due to be published in Q1 2021.
VPP Demonstrations objectives
AEMO has established a framework to allow VPPs to demonstrate their capability to deliver services in contingency FCAS and energy markets. The VPP Demonstrations will allow participating VPPs to trial a new specification to deliver contingency FCAS, and AEMO will observe how VPPs respond to energy market price signals as non-scheduled resources. By trialling VPP operations while their aggregated fleets remain of a small scale, the VPP Demonstrations aim to inform the effective integration of VPPs into the NEM before they reach large scale.
The VPP Demonstrations aim to:
- Allow VPPs to demonstrate their capability to deliver multiple value streams across FCAS, energy and potential network support services.
- Provide AEMO with operational visibility to help AEMO consider how to integrate VPPs effectively into the NEM.
- Assess current regulatory arrangements affecting participation of VPPs in energy and FCAS markets, and inform new or amended arrangements where appropriate.
- Provide insights on how to improve consumers’ experience of VPPs in future.
- Understand what cyber security measures VPPs currently implement, and whether their cyber security capabilities should be augmented in future.
Important information for VPP participants
Onboarding update – 12 November 2020:
From 1 December 2020 no further applications to join the VPP Demonstration will be accepted. The VPP Demonstration now has technology, geographical and participant diversity and is positioned well to meet the learning objectives outlined in the NEM VPP Demonstrations Final Design document. Furthermore, AEMO re-iterates that some requirements as part of the Demonstration may not continue beyond the conclusion of the Demonstration on 30 June 2021
Participants currently undergoing enrolment, or who intend to apply prior to 1 December 2020 must be fully enrolled by 18 December 2020, this means;
- VPP has integrated all APIs and is sharing data with AEMO in production
- loads classified as ancillary service loads through the PRC process
- are able to bid in the contingency FCAS market and
- have had the consumer insight survey sent to 75% of the customers that form their VPP.
Information to support the above activities with the allotted time may be requested by AEMO during the enrolment process. From 18 December 2020 any applications submitted to AEMO and still in progress/not complete (i.e. not in production and fully onboarded), will be withdrawn.
AEMO continues to maintain discretion around the VPP Demonstration terms, conditions and above dates.
Future opportunities to participate:
AEMO will be commencing the DER MASS Review Consultation in Q1, 2021. AEMO invite participants and those interested in the ongoing DER MASS requirements to participate in this consultation.
Further information on the operationalisation of VPP participation in contingency FCAS markets will be made available prior to 30 June 2021.
The following information is important to all prospective VPP participants and should be utilised prior to and during enrolment.
Enrolment and onboarding information:
VPP Demonstrations participant onboarding guide621.8 KB
VPP Demonstrations onboarding checklist319.38 KB
VPP Demonstration onboarding timeline - example226.13 KB
Guide to AEMO VPP Demonstration APIs1.21 MB
FCAS verification data template55.66 KB
VPP Demonstrations Enrolment Guide386.76 KB
VPP Demonstration Enrolment Form469.96 KB
VPP Demonstrations Cyber Security Questionnaire47.35 KB
VPP Demonstrations FAQ
Q. What generation types are considered suitable for the VPP Demonstration?
A. VPPs is a technology-neutral concept and can accommodate a variety of technologies within or across a VPP portfolio. For example, batteries, controllable loads and diesel generators could all form part of a VPP if they are classified as ancillary service loads.
Q. What is the maximum capacity of each ancillary service unit that can be enrolled in the VPPs?
A. The maximum capacity of each ancillary service unit at a single NMI is 5MW. A VPP must include an aggregate of at least 2 NMIs under a single DUID, with at least 1 MW combined capacity.
Q. Is it compulsory to engage with the Customer Research component of the VPP Demonstration?
A. Yes. Participation in the customer research component of the VPPs is integral to enhancing ours, and your understanding of the experience customers have had during the VPP demonstrations. It is important that you understand the data sharing components that need to be incorporated into the customer contracts in order to enable this customer research and get in touch early if you have any questions about this. Please also note that while it is important that we have the opportunity to reach out to the VPP customers, customers are not obligated to respond to the surveys.
Category : FCAS Delivery
Q. Can a single VPP include a variety of technologies, vendors and/or hardware types?
A. Yes, a VPP can include a variety of technologies provided the FCAS droop response is the same under a single VPPID and they all utilise the same type of controller (variable/proportional or switched). Careful consideration should be given for utilising a variable/proportional vs switched controller. A mixture of technologies can also be used within a DUID however, the same criteria applies – that is, they must use the same type of controller and must have the same droop settings. Most importantly, AEMO must be able to verify the FCAS delivery of the VPP.
Q. Can FCAS delivery be provided by more than just batteries?
A. Yes, FCAS can also be delivered by other decentralised generation exporting to the grid, or via increasing/decreasing load to meet the requirements. Please note that the VPP must ensure they have available at least 1 MW of FCAS response to meet the VPP enrolment requirements and will need to be able to deliver on the enabled FCAS.
Q. Does the enrolled MW capacity of a VPP need to be available at all times, or is it only required to be available when the VPP ’bids' or is enabled for FCAS?
A. The MW capacity only needs to be available during times where the VPP is enabled for any one (or all) of the six contingency FCAS markets. They must only enable the FCAS capacity that can be delivered at that point in time.
Q. How is the VPP capacity determined from the installed battery capacity?
A. The maximum battery capacity is determined by the frequency injection test. The participant must conduct a VPP wide test which provides the aggregated response of the batteries compared with the expected FCAS response. The VPP wide test must reflect the amount of capacity the VPP is looking to enable in the market, for example, if the VPP enrolls 2 MW of capacity, the VPP wide test must indicate that 2 MW of capacity is available.
When conducting a single battery test, this can be done either at a live site where all technical requirements of the frequency injection test are met, or as a bench test. Details of the Frequency Injection Test can be found in the VPP demonstration FCAS Specification Document.
Q. How is the FCAS service enablement signal/file made available for VPPs and what is the timing of this signal/file?
A. The Contingency FCAS enablement file (dispatch instructions) are made available electronically to VPPs via AEMO Electricity Market Management System (EMMS) interfaces. This file is sent from the NEMDE and will take up to 60 seconds (maximum) to process.
Q. Do droop calculations apply to resistive load control for FCAS?
A. A droop setting is applied to proportional/variable FCAS controllers only. The FCAS delivery from resistive loads can be initiated by switching controllers. If this is managed by switching controllers the droop calculation does not apply (i.e. the response does not need to be proportional to the frequency deviation).
Q. Regarding Fast FCAS, mandating 60 second discharge after 6 seconds, how many times is a response expected within a 30 minute period?
A. The minimum requirement for fast FCAS is to sustain the FCAS response for 60 seconds after the frequency disturbance time (FDR). FDR starts when the frequency leaves the Normal Operating Frequency Band (NOFB).
The mainland frequency data is published on the AEMO website and it can be used to find out the number of times when the frequency leaves the NOFB. Proportional controllers must respond when the frequency leaves the NOFB whereas switching controllers must trigger when the frequency goes above the lower frequency deviation setting or below the raise frequency deviation setting.
Q. Do I need to complete a frequency injection test on a single Battery Energy Storage System (BESS) to capture less than or equal to 50ms measurements of power and frequency if the sampling rate for each site is 100ms?
A. The high speed measurement of power and frequency at intervals up to 50ms is only required for the frequency injection/lab test on a single BESS if the VPP is registered for the fast contingency services (R6, L6). If the VPP is registered for the slow and delayed contingency services (R60, R5, L60, L5), then the measurement of power and frequency must be at intervals of up to 4s.
Q. Do we need to supply data before the time of the frequency disturbance? If so, how many seconds or minutes before the disturbance?
A. This information is available in the VPP demonstrations FCAS specification document. Refer to Clause 4 under Appendix B for details.
For the fast contingency services, the measurements must be made at intervals of 50 milliseconds or less for a period of at least 5 seconds before the frequency disturbance time and at least 60 seconds after the frequency disturbance time.
For the slow contingency services, the measurements must be made at intervals of 4 seconds or less for a period of at least 20 seconds before the frequency disturbance time and 5 minutes after the frequency disturbance time.
For the delayed contingency services, the measurements must be made at intervals of 4 seconds or less for a period of at least 20 seconds before the frequency disturbance time and 10 minutes after the frequency disturbance time.
The frequency disturbance time means the time at which Local Frequency falls or rises outside the normal operating frequency band during a Frequency Disturbance.
Category: APIs and Data
Q. When do participants need to start providing data?
A. VPPs must provide enrolment and FCAS assessment data at time of enrolment. This information is fundamental to the enrolment and FCAS verification process.
The operational data APIs should also be developed and integrated by the time enrolment is completed, however, the following allowances can be made to support the industry’s maturation:
- Aggregated data must be successfully running less than one (1) month after registration is complete.
- Device level telemetry data must be successfully running less than two (2) months after registration is complete.
Both sets of operational data must be back dated to the registration date if not provided from time of registration and a participant is not considered ‘fully enrolled’ until this data is provided.
Enrolment and Operational data must be provided via APIs
FCAS Assessment data (for enrolment and for compliance) must be provided via email in excel templates published on AEMO website.
Q. How often does the password for the AEMO’s energy market systems (which is required for the VPP API function) need to be updated?
A. Passwords must be updated every 90 days. This can be completed by your organisation’s participant administrators (PAs) via their user rights management function.
It might be helpful for your organisation to set up a reminder internally to ensure passwords do not expire.
Note - AEMO do not currently have an automated password expiry reminder email.
Q. When can intending participants access API information, test APIs and integrate with the APIs?
A. Participants can access API related information available on the API developer portal at any time. There is no need to create an account and sign-in to the API portal in order to access this information.
Testing and integrating with the APIs can only occur once the VPP enrolment process has commenced (i.e. an enrolment form has been submitted by the participant).
This is because participants must have the following information prior to commencing testing and integration:
- VPP ID and DUID. After successful submission of Enrolment application, the Registration team will provide participants with VPP ID & DUID.
- Client Certificates issued by AEMO.
- User Accounts created by the VPP’s Participant Administrator (PA).
- User Accounts assigned correct “Rights” by the VPP’s Participant Administrator (PA)
These requirements are in place to support system security.
Please Note: Separate Certificates and User Accounts are required for both Production and Pre-production environment.
Q. Will there be a penalty policy associated with not providing Data in time?
A. Not during the trial – we expect every effort is made by the participant to backdate data that has not come through automatically. If, however, a non-compliance pattern emerges AEMO will question if the participant is operating in accordance with the Terms and Conditions of the Demonstration.
Q. If NMIs are not included in subsequent enrolment APIs, will they be automatically removed?
A. No - submitting a new list of NMIs via the enrolment API does not automatically overwrite the recorded (existing) NMIs. To remove the NMI(s) from the enrolment data, the VPP must submit the NMI or list of NMIs via ‘removeNMIdevice’ API.
Removing a NMI also removes all the devices attached to the NMI from the VPP portfolio.
Q. Regarding API Data Submission, how will AEMO servers guarantee up-time? Do clients need to be able to re-submit failed transfers if the AEMO server is non-responsive?
A. AEMO intends to meet SLA requirements for the VPP application of 99% uptime. All other systems used are existing platforms and meet their specified SLA uptime requirements.
In the event that data could not be submitted, AEMO recommends implementing a re-try policy on client end to cater for scenarios which result in failed data submission due to Network or server issue.
It is expected that clients will re-submit failed data transfers.
Further to this, it is recommended to set a maximum number of re-tries (e.g. 3 to 5) and use a combination of exponential back-off or incremental intervals both with randomization until either a successful response is received or the maximum number of re-tries is reached in which case, the error should be logged and report the operation as failed. The idea behind combining these different re-try policies is to ensure the client application is quite unlikely to send more than one re-try request to the server at the same instant of time to avoid further stress on the backend server (if any) as well as to overcome periods of increased network activity between the two parties.
Q. The Data Specification document outlines that the availability forecast is required in 30 minute resolution and actual performance and operational forecasting data is required in 5 minute resolution. Are there any further specifications about the boundaries of this data?
A. While the specification document does not explicitly state this, the 30-min availability forecast timestamps need to be on the 30min boundary or hourly boundary (i.e. 15:30, 16:00, 16:30, 17:00 etc.). This also applies to the Actual Performance and Operational Forecasting data where the 5-minute timestamps are on the 5-min boundary (i.e. 15:30, 15:35, 15:40, 15:45 etc.).
The published Guide to VPP Demonstrations API document outlines these requirements clearly.
Q. The Operational Data API mentions a ‘controlled load’. What is meant by 'controlled load'?
A. A controlled load is any load (including batteries) that can be explicitly increased or decreased through remote means. For example, an air conditioner, pool pump or hot water system may be able to be ‘turned down’ or on at a particular time and therefore decrease/increase load according to an external signal.
Category: FCAS measurement
Q. FCAS response is measured at the connection point (e.g. site of solar/battery). How is this done and are there any requirements that must be met?
A. The FCAS response can be calculated using measured values from the site and does not need to be directly measured via a dedicated meter or VPP hardware. There is no requirement to install NMI approved meters to measure an FCAS response. This does not negate requirements outlined in the NER for NMI approved meters.
The telemetry requirements that must be met by VPP participants in the FCAS market is set out in the VPP demonstrations FCAS specification. It is important to note that the measurements of power and frequency must meet the accuracy and sampling rate specified in the FCAS specification document.
Category: MASS Review
Q. What is the likelihood that allowing both export and import will continue (outside the trial) and eventually be incorporated into the MASS?
A. AEMO is currently analysing the data obtained via the VPP Demonstration and will use these findings to inform the MASS review. It is important to note that changes to the MASS will undertake a Rules Consultation Procedure and that AEMO must give careful consideration to all relevant submissions from stakeholders. AEMO encourage stakeholders to participate in this consultation when commenced.
Q. What is the likelihood that the relaxations of the measurement speeds will survive past the VPP demonstration, perhaps as interim arrangements and then formalised in the MASS?
A. As per the previous question, the MASS will undergo a consultation procedure in-line with the National Electricity Rules Consultation requirements to make any changes to the measurement requirements in the MASS. Stakeholder participation in this consultation is encouraged by AEMO. The outcome of the consultation will consider the information acquired from the demonstration, the MASS review and the interim arrangements. AEMO will provide clarity and examples in the MASS or a specific VPP FCAS specification document on how the FCAS response can be calculated using direct measurements from every site, rather than being directly measured.
Q. When will these two decisions regarding MASS be made?
A. AEMO proposes to address these two points in an upcoming MASS consultation. While Q1 2021 is being targeted, this will be subject to operational priorities, particularly around managing change throughout the period of the COVID-19 pandemic.
Q. What is the impact on droop setting applied to the VPP Demonstrations on low inertia energy systems?
A. In the event where there is an aggressive droop setting (like the 0.7 droop setting in the VPP demonstrations), the response to events may result in an exasperation of the issue, rather than a correction. Currently, VPPs in the VPP Demonstrations are still small enough to allow for this droop setting, but further data and modelling is required to understand the impacts of a larger set of VPPs with this droop setting. These findings will feed into the MASS review.