User information Fronius Argeno - Advanced Settings

Reactive power can be used in electrical energy supply networks to bolster the level of voltage. As such, feed-in inverters can contribute to static voltage stability. Reactive power brings about a voltage drop at the inductive and capacitive components of the equipment which can either bolster or reduce the level of voltage depending on what is indicated. If the generating system draws inductive reactive power while effective power is being fed in, part of the voltage swing caused by the effective power feed can be compensated for by drawing reactive power.

The reactive power mode and the respective control process are specified by the utility. If no control process has been specified, the system should be operated with a fixed reactive power specification of 0%.

Reactive power can be used in electrical energy supply networks to bolster the level of voltage. As such, feed-in inverters can contribute to static voltage stability. Reactive power brings about a voltage drop at the inductive and capacitive components of the equipment which can either bolster or reduce the level of voltage depending on what is indicated. If the generating system draws inductive reactive power while effective power is being fed in, part of the voltage swing caused by the effective power feed can be compensated for by drawing reactive power.

The reactive power mode and the respective control process are specified by the utility. If no control process has been specified, the system should be operated with a fixed reactive power specification of 0%.

The device can be operated within the respective fixed voltage range stated. The maximum apparent power in the event of undervoltage depends on the mains voltage due to the maximum continuous current, as shown in the table below.
The following figures show the reactive power operating range as a function of the effective power and the apparent power operating range as a function of the mains voltage for various devices.

P-Q operating range

Apparent power subject to mains voltage

For all of the control methods, the specified set value at the inverter's connection terminals is adjusted using a stationary deviation of the reactive power of maximum 2% S_N. This maximum deviation always relates to the value specified for the reactive power. If the cos j power factor is specified in the control method, the deviation relates to the reactive power value resulting from the current power level.

The transient response of the control methods is determined by a PT-1 filter. The settling time is 5 tau, which corresponds to achieving approx. 99% of the final value of a PT-1 filter. Depending on the control method selected, there are also other parameters that determine dynamic behavior.

Reactive power is prioritized by default for all methods but can be disabled. The priority can be selected when using the Q constant and Q(U) mode. The maximum possible effective power that can be fed in is reduced in line with the P-Q operating range when the maximum useful power is reached.

cos constantj
In cos j constant mode, the specified power factor is set permanently by the inverter. The reactive power level is set in line with Q = P * tan j as a function of the power so the specified power factor is continually maintained. If the setting value changes, the new value is adopted by way of a filter in a muted manner. The settling time can be parameterized and is 1 second (corresponds to 5 tau, while the ideal time according to SunSpec is 3 tau) with the transient response of a PT-1 filter with a time constant of tau = 200 ms. The specified power factor can be configured on the display or via communication using the RS485 protocol and MODBUS/SunSpec.

If the applicable grid code calls for the j to respond to the set value with a defined gradient or a defined settling time that is shorter than the configured tau = 200 ms, this gradient or settling time must be implemented in the system controls

Q constant
In Q constant mode, the specified reactive power value is set permanently by the inverter. If the specification is changed, the new value is adopted by way of a filter in a muted manner. The settling time and gradient limit can be configured via the web interface. The settling time is 1 second and follows the transient response of a PT-1 filter with a time constant of tau = 200 ms. The specified reactive power can be configured on the display or via communication using the RS485 protocol and MODBUS/SunSpec.

If the applicable grid code calls for the reactive power to be adjusted to the set value with a defined gradient or a defined settling time that is shorter than the configured tau = 200 ms, this gradient or settling time must be implemented in the system controls.

cos j (P)
In the cos j (P) operating mode, the cos j set value and the reactive power set value derived from this are calculated continuously as a function of the actual power level. This function ensures that grid support is provided by the reactive power when a significant voltage boost is anticipated due to a high feed-in level. A characteristic is specified which can be used to configure up to 10 nodes, i.e., value pairs for effective power and j. The effective power is entered as a percentage in relation to the maximum apparent power Slim. Other parameters can be used to limit functionality and to limit activation to certain voltage ranges.

cos j (P) standard characteristic with three nodes

Q(U) 10 nodes
In Q(U) mode, the reactive power set value is calculated continuously as a function of the mains voltage. This function ensures that grid support is provided by the reactive power as soon as the voltage actually deviates from the target voltage. In this case, a characteristic is specified which can be used to configure up to 10 nodes consisting of value pairs for voltage and reactive power. Other parameters allow you to limit functionality and to limit activation to certain power levels as well as to parameterize the transient response.
The displacement voltage of the neutral point is used to calculate the reactive power set value for three-phase devices.

Q(U) standard characteristic with five nodes

Q(P) 10 nodes
In Q(U) mode, the reactive power set value is calculated continuously as a function of the effective power. In this mode, a characteristic is specified which can be used to configure up to 10 nodes consisting of value pairs for power and reactive power. The function allows you to parameterize the transient response. The displacement power of the neutral point is used to calculate the reactive power set value for three-phase devices.

Q(P) standard characteristic with five nodes

In all of the control methods described below the specified set value at the inverter's terminals is adjusted using a stationary deviation of the effective power of maximum 2% S_N.
The transient response of the control methods is determined by a PT-1 filter. The settling time is 5 tau, which corresponds to achieving approx. 99% of the final value with a PT-1 filter. Depending on the control method selected, there are also other parameters that specify dynamic behavior.

In all of the control methods described below the specified set value at the inverter's terminals is adjusted using a stationary deviation of the effective power of maximum 2% S_N.
The transient response of the control methods is determined by a PT-1 filter. The settling time is 5 tau, which corresponds to achieving approx. 99% of the final value with a PT-1 filter. Depending on the control method selected, there are also other parameters that specify dynamic behavior.

On all inverters, the function "P set value" is integrated in the MPP tracking of the inverter. The P set value is continually recalculated based on the MPP tracking algorithm.
Model 704 (DER AC controls) thus also corresponds to the current market specifications.

The "P limit" function is available for limiting the maximum power of feeding in. If necessary, this can be used to reduce the maximum possible feed-in of an inverter, e.g., for managing bottlenecks for the operator of the distribution grid.

P limit is only available via the MODBUS/SunSpec inverter model 123 Immediate Controls and via RS485 communication.
When a set value is received for P limit, the output power of the inverter is limited to the specified power value. If the limit value is changed, the new value is adopted by way of a filter and a gradient limit. The actual power may be below the specified limit value because the available power (PV) and the set power value (storage) may be below the specified limit value. Depending on the inverter series, the settling time and gradient limit may be adjustable.

If the applicable grid connection guidelines call for the effective power to be adjusted to the set value with a defined gradient or a defined settling time, the device can be configured in such a way that this gradient is adhered to. The gradient can also be implemented on the system controller. This second solution is to be used for all other inverters.

If the voltage increases too sharply in the upstream distribution network and not enough reactive power can be absorbed, it may be necessary to reduce the effective power. In this case, P(U) control is available for making optimum use of the grid.
P(U) control reduces the effective power that is fed in as a function of the mains voltage using a prescribed characteristic as a basis. P(U) control is implemented as an absolute power limit. The actual power of the inverter may vary freely below this limit due to a possible fluctuation in the available power or the set value, but at no time increases above the absolute power limit.

Adjusting the effective power P(f) in the event of overfrequency
Feed-in inverters must assist with frequency stability in the interconnected grid. If the grid frequency leaves the normal tolerance range (e.g., ±200 MHz), the grid is in a critical state. In the event of overfrequency, there is a generation surplus; in the event of underfrequency, there is a generation deficit. Photovoltaic systems must adapt their power of feeding in relative to the frequency deviation. In the event of overfrequency, the power adjustment is determined by a maximum feed-in limit. The actual power of the inverter may vary freely below this limit due to a possible fluctuation in the available power or the set value, but at no time increases above the absolute power limit.

Equation 1

Equation 2

Equation 1 defines the maximum limit with DP as per equation 2, P_M the actual power at the time of activation, and P_ref the reference power. P_ref is defined as P_M, the actual power at the time of activation. f is the actual frequency and f₁ is the specified activation threshold.

Equation 3

Equation 4

In some standards, the power adjustment is specified by a drop (s) instead of a gradient (g), as shown in equation 3. The drop s can be transformed into a gradient g in accordance with equation 4.
The frequency f remains above the activation threshold f₁ during an overfrequency event. Consequently, the expression (f1 – f) is negative and DP corresponds to a reduction in the power of feeding in.
The measurement accuracy of the frequency is greater than 10 MHz.
The specific mode of operation of the function is specified by the utility, the relevant standards, or the grid connection guidelines. The option to configure the function makes it possible to satisfy a wide variety of standards and guidelines. Certain configuration options are not available in some country settings because the relevant standards or grid connection guidelines prohibit adjustments.

Adjusting the effective power P(f) in the event of underfrequency
Some grid connection rules also require that the effective power P(f) be adjusted in the event of underfrequency. Because photovoltaic systems are typically operated in the maximum power point, a power reserve is not available to increase the power in the event of underfrequency.
However, if the system is curtailed based on the market regulations, the effective power can be increased up to the available power. Because the inverter is unable to distinguish set values for the P constant between the utility's obligatory bottleneck management and market regulations, this must be implemented as part of the site-specific infrastructure for controlling the system.

Example behavior with hysteresis

Example behavior without hysteresis

The capacity of a generating system to remain immune to voltage dips and voltage spikes in the supply system is a key element in establishing a reliable energy supply. Interference immunity ensures that brief disruptions do not result in a loss of relevant generation capacity in a larger area of an interconnected grid. Grid support via fast feed-in of residual current also limits the spatial extent of the incident.
The device offers this characteristic with its dynamic grid support by way of interference immunity. The ability to remain on the grid is the relevant factor. The protection settings also determine the device's ability to remain on the grid or not. Protection settings prevail over the capacity of interference immunity.

The capacity of a generating system to remain immune to voltage dips and voltage spikes in the supply system is a key element in establishing a reliable energy supply. Interference immunity ensures that brief disruptions do not result in a loss of relevant generation capacity in a larger area of an interconnected grid. Grid support via fast feed-in of residual current also limits the spatial extent of the incident.
The device offers this characteristic with its dynamic grid support by way of interference immunity. The ability to remain on the grid is the relevant factor. The protection settings also determine the device's ability to remain on the grid or not. Protection settings prevail over the capacity of interference immunity.

Interference immunity against undervoltage
Voltage dips above the limit curve (see figure below) can be overcome without the need for shutdown from the grid. The power of feeding in is constantly maintained within the limits of the maximum continuous current of the inverter.
If a reduction in power occurs, the power is brought back up to the pre-fault level within 100 ms of the voltage returning.
The inverter can overcome voltage fluctuations provided that the voltage level does not stay above the continuous operating voltage range for more than 100 s and does not rise above the short-term maximum operating voltage range (up to 100 s). The specific values for each inverter can be found here.
The interface protection integrated in the inverter (voltage, frequency, anti-islanding) can be configured in a range that permits the above behavior. If, however, the setting of the interface protection limits the voltage/time characteristic, the interface protection triggers and interrupts the run-through as configured.

Interference immunity characteristic in relation to the nominal voltage p(u)

When dynamic grid support via fast feed-in of residual current is activated, residual current is fed-in in addition to the interference immunity properties against dips and spikes described above.
The inverter adapts its current feed-in as soon as a dip or spike event occurs in order to bolster the mains voltage. The support takes place in the event of voltage dip with overexcited reactive current (corresponds to a capacitive load), and in the event of a voltage spike with underexcited reactive current (corresponds to an inductive load). In the reactive current priority mode, the active current is reduced to the extent necessary to comply with the limits of the maximum continuous current of the inverter.

A dip or spike is detected if either the normal operating voltage range setting is exceeded by at least one phase-phase or phase-neutral voltage, or if a jump in the positive or negative sequence component of the voltage greater than the deadband setting occurs. The magnitude of the voltage jump of the positive and negative phase-sequence system equates to the difference between the pre-fault voltage and the actual voltage based on the reference voltage. The pre-fault voltage is calculated as the mean value over 50 periods.

Formula 1

The reactive current is adapted using a response time of < 20 ms and a settling time of < 60 ms after the event has occurred. Responses to changes in the voltage during the event or to the voltage recovery at the end of the event take place with the same dynamic.
The formula for calculating the dynamic reactive current that is fed in for the positive and negative phase-sequence system is as follows:

Formula 2, depending on the nominal current I_N of the inverter

DFor the positive and negative phase-sequence system, u equates to the difference between the pre-fault voltage and the current voltage in relation to the reference voltage. The pre-fault voltage is calculated as a 1-min mean value.

Formula 3

On account of the definition of a voltage jump in pre-norm EN50549-2 and in VDE-AR-N 4120 and VDE-AR-N 4110, it is typically the case that another voltage jump is detected when the event is at an end, when the fault is rectified, and when the voltage returns to a fault-free state. The result of this is that in an active operating mode, the dynamic grid support via a fast feed-in of residual current remains active even after the event has passed, and that reactive current is fed in according to formulas (2) and (3). Dynamic grid support using fast feed-in of residual current is then deactivated after a configured minimum support time, usually 5 s.

The inverter stops feeding in current if the zero current threshold is exceeded. If the inverter is feeding in reactive power before the fault, the reactive power is reset to the value before the fault Q with the settling time configured in the activated reactive power control mode after the fault.

Formula 4

The maximum effective and apparent power to be installed for a generation system is agreed between the utility and system operator. The device capacity of a system can be set to the exact agreed value using the S_lim and P_lim settings. To ensure that the load on the devices in a system is uniform, we recommend distributing the power reduction evenly across all devices.

Some grid connection rules require that the agreed reactive power be supplied from every operating point of the system without a reduction in the actual effective power. Given that the inverter has the full P-Q operating range, a reduction in the effective power is, however, required during operation at maximum effective power because an apparent power reserve is not available.
By adjusting P_lim, the maximum effective power can be limited in order to establish an apparent power reserve and ensure that the agreed reactive power can be delivered from every effective power operating point.
The diagram below shows the appropriate P-Q operating range with a required example effective power of 48% of the maximum apparent power of the system or 43% of the maximum effective power of the system.

P-Q operating range with limited effective power for PV inverters

The power limitation parameters can be adjusted using SunSpec model DID123. During this process, you should also check whether the internal and/or external power limitation is active.

If the ramp time "WMaxLimPct_RvrtTms" in the SunSpec model is specified as 0 s, the internal output gradient is used. Otherwise, the set value will be used.
Irrespective of the communication protocol used, the settling time "WMaxLim_Ena" is used to transfer the new power value. Otherwise, the internally configured value is used. The additional ramp time "WMaxLimPct_RmpTms" specifies the jump time from one power value to the new power value.

The following formulas are used to calculate the gradient S_lim/min:

Power gradient according to sample parameters and calculation

The following formulas are used to calculate the Q filter parameter and cos j gradient:

Formula for calculating the Q filter parameter

Formula for calculating the cos j gradient (internal power gradient)

The maximum effective and apparent power to be installed for a generation system is agreed between the utility and system operator. The device capacity of a system can be set to the exact agreed value using the S_lim and P_lim settings. To ensure that the load on the devices in a system is uniform, we recommend distributing the power reduction evenly across all devices.

Some grid connection rules require that the agreed reactive power be supplied from every operating point of the system without a reduction in the actual effective power. Given that the inverter has the full P-Q operating range, a reduction in the effective power is, however, required during operation at maximum effective power because an apparent power reserve is not available.
By adjusting P_lim, the maximum effective power can be limited in order to establish an apparent power reserve and ensure that the agreed reactive power can be delivered from every effective power operating point.
The diagram below shows the appropriate P-Q operating range with a required example effective power of 48% of the maximum apparent power of the system or 43% of the maximum effective power of the system.

P-Q operating range with limited effective power for PV inverters

The power limitation parameters can be adjusted using SunSpec model DID123. During this process, you should also check whether the internal and/or external power limitation is active.

If the ramp time "WMaxLimPct_RvrtTms" in the SunSpec model is specified as 0 s, the internal output gradient is used. Otherwise, the set value will be used.
Irrespective of the communication protocol used, the settling time "WMaxLim_Ena" is used to transfer the new power value. Otherwise, the internally configured value is used. The additional ramp time "WMaxLimPct_RmpTms" specifies the jump time from one power value to the new power value.

The following formulas are used to calculate the gradient S_lim/min:

Power gradient according to sample parameters and calculation

The following formulas are used to calculate the Q filter parameter and cos j gradient:

Formula for calculating the Q filter parameter

Formula for calculating the cos j gradient (internal power gradient)

A soft start up function is available to prevent the grid from being negatively impacted by a sudden increase in feed-in power from the inverters.
When the inverter is activated or switched on, the increase in power is limited by the set gradient.

Primarily because there is a risk that many systems could increase their power levels simultaneously after they have been switched off by the grid protection, a soft start-up is usually only required for start-up after a device has been switched off by the grid protection.

The soft start up is implemented by an absolute power limitation that increases with a continuous gradient up to the maximum power. The actual power of the inverter may vary freely below this limit due to a possible fluctuation in the available power or the set value, but at no time increases above the absolute power limit.

In the case of very large systems, it may be necessary to limit the maximum change in power during normal operation. The grid power feed-in is increased or decreased according to the configured gradients when the specified set value (for increasing and decreasing power) is changed and when the solar irradiation (for increasing power) is changed. The change in power cannot be limited if solar irradiation is reduced.

The function is not active if there are changes in power that are defined by a different grid support function, such as power recovery after fault ride through, P(f), P(U).

Due to decentralized generation, there is the possibility that a deactivated part of the grid will remain live in an unintended island due to a local balance between load and generation in this part of the grid. The detection of unintended island formation is an important function of decentralized generating units and plays a role in preventing damage to equipment as well as ensuring the safety of personnel.

In the case of the last example in particular, the generating units need to be disconnected very quickly to cause the forming island to collapse. At the same time, any island formation detection method may cause false tripping. The industry is therefore working continually to develop methods that are fast and reliable and can also be depended on to prevent false tripping.

Due to decentralized generation, there is the possibility that a deactivated part of the grid will remain live in an unintended island due to a local balance between load and generation in this part of the grid. The detection of unintended island formation is an important function of decentralized generating units and plays a role in preventing damage to equipment as well as ensuring the safety of personnel.

In the case of the last example in particular, the generating units need to be disconnected very quickly to cause the forming island to collapse. At the same time, any island formation detection method may cause false tripping. The industry is therefore working continually to develop methods that are fast and reliable and can also be depended on to prevent false tripping.

Enhanced island detection uses a reliable island detection strategy based on the characteristic differences between an interconnected grid and an islanded grid, thus ensuring rapid and reliable detection and prevention of false tripping.

An interconnected grid is dominated by rotating machinery, so the frequency is proportional to the effective power balance and the voltage is proportional to the reactive power balance. By contrast, an islanded grid behaves like an oscillating circuit, so the frequency is proportional to the reactive power balance and the voltage is proportional to the effective power balance. The active enhanced anti-islanding method detects this difference by monitoring the behavior of the grid. The enhanced anti-islanding method monitors the natural fluctuation in the grid frequency and injects a minimal reactive power that is proportional to the rate of change of the frequency. In the moment an island is formed, the connected grid closes a positive feedback loop which allows the inverter to detect the change in the situation and to disconnect. If an island forms, the inverter disconnects within several 100 ms, and well below 1,000 ms.

This detection method is combined with a two-stage passive rate of change of frequency (ROCOF) observation. If the ROCOF of the grid exceeds the configured shutdown threshold (stage 1) for the configured shutdown time, the device switches to zero current mode. If the ROCOF of the grid exceeds the configured shutdown threshold (stage 2) for the shutdown disconnection time, the device switches off. In case of an island, this will shut down the island instantaneously. If the grid stabilizes, which might be the case if the ROCOF event was due to a brief disturbance in the grid, the inverter will resume normal operation. If stage 1 is active, the device has switched to zero current mode and will recommence feed-in after a few 100 ms. If stage 2 is active, the device has switched off and the configured reconnection conditions apply.

The "Q on Demand" function can also provide a reactive power Q outside of grid power feed operation to stabilize the grid (e.g., at night).

The specifications that the inverter receives from the utility through the system controller via Ethernet or RS485 take first priority. The parameters for Q constant and Q(U) stored in the inverter take second priority.
If the AC supply is disconnected during "Q on Demand" operation outside of grid power feed operation, the "Q on Demand" function can only be used again following proper grid power feed operation (if there is an adequate DC supply). The existing "Night Shutdown" that has been deactivated also remains active.
The following figures show normal operation in the P-Q operating range during the day (grid power feed operation) (1) and "Q on Demand" operation at night (2).

Only reactive power is produced at night. A small amount of effective power will inevitably be required for the internal power supply in order to keep the pre-configured reactive power functions in "Q on Demand" mode (see item 2 in the negative P range).

(1) Normal operation: Effective power and reactive power supply at different voltages.

(2) "Q on Demand" operation: Reactive power supply with nominal grid voltage outside grid power feed operation.

The "Q on Demand" function can also provide a reactive power Q outside of grid power feed operation to stabilize the grid (e.g., at night).

The specifications that the inverter receives from the utility through the system controller via Ethernet or RS485 take first priority. The parameters for Q constant and Q(U) stored in the inverter take second priority.
If the AC supply is disconnected during "Q on Demand" operation outside of grid power feed operation, the "Q on Demand" function can only be used again following proper grid power feed operation (if there is an adequate DC supply). The existing "Night Shutdown" that has been deactivated also remains active.
The following figures show normal operation in the P-Q operating range during the day (grid power feed operation) (1) and "Q on Demand" operation at night (2).

Only reactive power is produced at night. A small amount of effective power will inevitably be required for the internal power supply in order to keep the pre-configured reactive power functions in "Q on Demand" mode (see item 2 in the negative P range).

(1) Normal operation: Effective power and reactive power supply at different voltages.

(2) "Q on Demand" operation: Reactive power supply with nominal grid voltage outside grid power feed operation.