With the growth of distributed resources at the edge, such as solar photovoltaics (PV), battery storage and electric vehicles (EVs), intelligence is key if impacts on the grid are to be avoided.
While there is a lot of activity on vehicle-to-grid technology and smart charging for EVs, there has been little so far on PV interconnection given its lower potential impact. However, this is now changing as the penetration is increasing. In California, for example, since September 2017 the state has required all new systems to be interconnected with smart inverters with intelligence that can be utilised to address grid concerns alongside the traditional function of converting the power from the panels from the native direct current (DC) into alternating current (AC) for use either onsite or for grid feed-in.
The extent of the issue is illustrated by Pacific Gas & Electric (PG&E). With over 350,000 solar customers, the utility is adding another approximately 5,000 per month and expects its installed retail solar base to reach 4,800MW in 2024, up by a factor of 10 from 488MW in 2015.
According to PG&E, a study commissioned from Navigant found that a high penetration of solar PV on feeders could cause challenges including reverse power flow exceeding the loading limits, feeder voltages exceeding acceptable limits and power quality issues.
Smart inverters are still an emerging technology and in a new report, PG&E has documented findings in their use in mitigating potential local grid issues on two residential distribution feeders. Overall the project, which is still ongoing, is aiming to develop understanding of the functionalities, requirements and investments needed by utilities to utilise smart inverter capabilities.
PG&E has forecast that by 2020, roughly half of its interconnected PV will be equipped with smart inverters and that by 2025 the number on its network will reach 1m.
In this first phase, 15 PV systems totalling 65kW (DC) of smart inverter-enabled installed capacity were deployed and subject to six field tests of active/reactive power control.
Several key learnings emerged in this project, not least that smart inverters can indeed enable customer PV to help with local secondary voltage regulation through autonomous active or reactive power support. The extent of impacts on the secondary voltage depends on the amount of smart inverter active or reactive power output and the electrical properties of the secondary system. On average, 1kVAR of reactive power absorption resulted in a 0.25V change at the smart inverter terminal, while 1kW of active power output resulted in a 0.6V change at the smart inverter terminal.
Smart inverters were found to be able to execute programmed volt-var and volt-watt functions, the latter when no other active power control commands were executed. In addition, remote change of autonomous smart inverter voltage curves and schedules using a vendor-specific aggregation platform was demonstrated possible. However, while autonomous settings and remote change of autonomous settings may address some grid constraints, there is likely still a need for active management in some instances, e.g. sending real or reactive power set points.
Another finding was that reliable communication links are critical for success. In the project residential Wi-Fi internet was used in combination with Zigbee to communicate with the PV assets, but communication infrastructure performance must be improved for use cases that require real-time active control at scale.
In addition, current utility operational systems are not yet capable of fully integrating large numbers of smart inverters effectively and using the technology to its fullest extent. Further utility investment is required, especially if the smart inverters are to be controlled at scale and in real-time across the electrical distribution system.
Commenting on the findings, PG&E Vice President, Grid Integration and Innovation, Roy Kuga said: “The smart inverters being installed on our customers' solar and energy storage systems, paired with our investment in grid operations systems and technology, show promise to facilitate distribution system reliability and power quality in the increasingly complex grid.”
Secondary and primary voltage support
The ongoing phase of the project is focussed on evaluating the potential of smart inverter enabled PV to provide not only secondary but also primary voltage support. This phase involves larger commercial and agricultural customers and retrofitting of the smart inverters.
The feeder for this part of the project is more prone to voltage disturbances than the residential feeders and has previously experienced high voltage conditions, possibly tied to its overall high PV penetration. Using this feeder to quantify voltage response to changes in smart inverter reactive and active power operation should increase understanding of autonomous functionality, according to PG&E.
Other aspects of this phase, which is expected to be completed in late 2018, include evaluating a vendor-agnostic aggregation platform and evaluating customer energy generation curtailment as a function of smart inverter settings. A series of smart inverter test cases also are being conducted in the laboratory and the potential cost savings and benefits of smart inverters across PG&E’s system will be explored.
With the project, PG&E anticipates better understanding for utilities and other stakeholders such as regulators of the potential of smart inverters as their use becomes more widespread.