Enhanced Control of Three-Phase Grid-Connected Renewables with Fault Ride-Through Capability under Voltage Sags

Round 1

Reviewer 1 Report

This paper proposes an enhanced control of three-phase grid connected renewable energy sources under sags. I have following concern about the manuscript:

 

1. The introduction section is not properly synchronized. For reference, please review following paper to write the manuscript to clearly understand the contribution of paper with proper follow in the writing

[Ref1] Active Disturbance Rejection Control Scheme for Reducing Mutual Current and Harmonics in Multi-Parallel Grid-Connected Inverters, https://doi.org/10.3390/en12224363.

2.      Please also comment on how the other dynamics of grid-connected renewable energy sources can be handled by the proposed idea like grid impedance. Please review following paper in this regard

[Ref1] Active Disturbance Rejection Control Scheme for Reducing Mutual Current and Harmonics in Multi-Parallel Grid-Connected Inverters, https://doi.org/10.3390/en12224363.

[Ref2] Resonance damping for an LCL filter type grid-connected inverter with active disturbance rejection control under grid impedance uncertainty, https://doi.org/10.1016/j.ijepes.2019.02.004.

 

[Ref3] Active Disturbance Rejection Control Based Single Current Feedback Resonance Damping Strategy for LCL-Type Grid-Connected Inverter, 10.1109/TEC.2020.3006151. 

3.      How is the coupling effect eliminated with the proposed method? Please comment on this by reading above reference papers.

4.      Due to high uncertainty in the system, how authors see their proposed method better than Active Disturbance Rejection Control.

5.      A separate nomenclature must be there to clearly understand every symbol used in the paper.

6.      Why the inverter gain is selected as 2/3 of Vdc.

7.      Compare the results with any other conventional method.

 

 

Author Response

The authors thank the reviewer for his/her recommendations. Following our answers:

1. The paper is focused on the effect of very short-time duration balanced and unbalanced voltage sags and the possibility to implement the ancillary algorithm named Low-voltage Ride-through (LVRT) to prevent the disconection of the inverter from the mains. The international standard IEC 61400-21, and the proper injection of reactive power according to the spanish normative are the valid normatives or grid-code (described in Figure 2) to implement the LVRT (it can be extrapolated to other normatives or grid-codes of any country if whished). In addition, some amount of constant active power can be delivered to the mains with a limitation in the amplitude of the grid currents to their nominal values. 

2. So, the influence of parallel inverters and the resonance of the LCL filter to voltage harmonics are not studied because the authors consider a stiff grid in this paper rather than weak ones at the point of common coupling (PCC). In this case the effect of the voltage harmonics and also the potential resonance due to the LCL filter can be neglected and the behaviour of the LVRT ancillary algoritm can be studied separately. Of course, for weak grids the outcomes described in the papers the reviewer has suggested can be added to the LVRT algorithm.

3. The coupling effect is not studied in this paper because stiff grids are used.

4. The proposed ancillary algorithm LVRT studied in this paper cannot be compared with the coupling effect described in the suggested papers.

5. The reviewer is right and a Nomenclature is added to the paper.

6. The inverter gain is 2/3 of Vdc because we are dealing with a 2-levels 3-phase inverter with isolated neutral yielding 5 voltage levels in the phase-to-neutral inverter voltage (-2/3 Vdc; -1/3 Vdc; 0; 1/3 Vdc; 2/3 Vdc)

7. The reviewer is right, some additional comments are added in the introduction section to compare conventional synchronization algorithms in dq axes with enhanced synchronization algorithms in alpha,beta axes, as well as the control of the 3-phase grid currents in conventional dq axes using PI regulators and in alpha,beta axes using PR regulators:

"...For example, the synchronization method described in [16] in the synchronous reference frame (SRF) (dq axes) only computes the phase angle and the PS value of the grid voltages, whereas the decoupled double SRF Phase-locked loop (PLL) [17] in two SRFs (dq axes) can be implemented to compute the phase angle together with the PS and the NS values dur-ing the occurrence of unbalanced voltage sags.

On the other hand, two enhanced synchronization algorithms that are also able to es-timate the utility grid frequency and the PS and NS values of the grid voltages during voltage sags with higher dynamics can be implemented in αβ axes (which is known as the stationary reference frame). These are the dual second order generalized integrator fre-quency- locked loop (DSOGI-FLL) [18] and the Multiple second order generalized integra-tor frequency- locked loop (MSOGI-FLL) [19]. The latter is also capable to separate the harmonics components of the mains from its fundamental nominal frequency (w0), but a high computational burden is needed. Eventually a stiff grid (Z -> 0) is to be considered in this paper and the effect of the harmonics can be neglected. Hence, the DSOGI-FLL syn-chronization algorithm, which is easier to implement, will be used." [L100-116]

.
.
.

"...The control of the 3-phase currents can be exerted in the conventional SRF (i.e., dq axes) using PI regulators [20] or in the stationary reference frame (i.e., αβ axes) using Propor-tional Resonant (PR) ones [21]. The latter will be used in this paper to regulate the PS and the NS values of the 3-phase inverter currents simultaneously in αβ axes by using an only one resonant block in each axis." [L121-126]

In Section 2, the expressions of the average active and reactive powers with its PS and the NS components (P0+ + P0- and Q0+ + Q0-), together with the oscillating active and reactive power components (P ̃ and Q ̃ ) at twice the system frequency will allow to cancel the oscillating component of the active power (P ̃ ) if whished. Doing the control of the grid currents using Proportional Resonant (PR) regulators simplifies the calculation of the reference commands of the instantaneous active (P*) and reactive (Q*) powers, as well as the reference commands of the currents in alpha,beta axes, compared with the conventional control in dq axes using PI regulators [20]. Moreover, it can be easily delivered a constant active power to the grid instead of an oscillating one at twice the system frequency if a control in dq axes were done.

"It is worth noting that the application of the PR regulators for the grid currents makes it possible to deliver a constant active power to the utility grid easily by making P ̃=0 in Equation 1, whereas an oscillating active power at twice the fundamental frequency (2w0) would be normally delivered if the control were exerted with conventional PI regulators in dq axes, unless a very complicated strategy and a huge amount of computational burden were used [27]." [L222-227]

Reviewer 2 Report

The proposed dual second order generalized integrator frequency-locked loop synchronization algorithm for frequency estimation is interesting from a scientific point of view. Authors should make more detailed comparison with other control concepts for a grid-connected renewable energy system. This article should be added with comparison graph with the others control methods for reduce voltage sags with grid-connected renewable sources. Figure 4 should be more detailed.

Author Response

The authors thank the reviewer for his/her recommendations. Following our answers:

The reviewer is right, some additional comments are added in the introduction section to compare conventional synchronization algorithms in dq axes with enhanced synchronization algorithms in alpha,beta axes, as well as the control of the 3-phase grid currents in conventional dq axes using PI regulators and in alpha,beta axes using PR regulators:

"...For example, the synchronization method described in [16] in the synchronous reference frame (SRF) (dq axes) only computes the phase angle and the PS value of the grid voltages, whereas the decoupled double SRF Phase-locked loop (PLL) [17] in two SRFs (dq axes) can be implemented to compute the phase angle together with the PS and the NS values dur-ing the occurrence of unbalanced voltage sags.

On the other hand, two enhanced synchronization algorithms that are also able to es-timate the utility grid frequency and the PS and NS values of the grid voltages during voltage sags with higher dynamics can be implemented in αβ axes (which is known as the stationary reference frame). These are the dual second order generalized integrator fre-quency- locked loop (DSOGI-FLL) [18] and the Multiple second order generalized integra-tor frequency- locked loop (MSOGI-FLL) [19]. The latter is also capable to separate the harmonics components of the mains from its fundamental nominal frequency (w0), but a high computational burden is needed. Eventually a stiff grid (Z -> 0) is to be considered in this paper and the effect of the harmonics can be neglected. Hence, the DSOGI-FLL syn-chronization algorithm, which is easier to implement, will be used." [L100-116]

.
.
.

"...The control of the 3-phase currents can be exerted in the conventional SRF (i.e., dq axes) using PI regulators [20] or in the stationary reference frame (i.e., αβ axes) using Propor-tional Resonant (PR) ones [21]. The latter will be used in this paper to regulate the PS and the NS values of the 3-phase inverter currents simultaneously in αβ axes by using an only one resonant block in each axis." [L121-126]

In Section 2, the expressions of the average active and reactive powers with its PS and the NS components (P0+ + P0- and Q0+ + Q0-), together with the oscillating active and reactive power components (P ̃ and Q ̃ ) at twice the system frequency will allow to cancel the oscillating component of the active power (P ̃ ) if whished. Doing the control of the grid currents using Proportional Resonant (PR) regulators simplifies the calculation of the reference commands of the instantaneous active (P*) and reactive (Q*) powers, as well as the reference commands of the currents in alpha,beta axes, compared with the conventional control in dq axes using PI regulators [20]. Moreover, it can be easily delivered a constant active power to the grid instead of an oscillating one at twice the system frequency if a control in dq axes were done.

"It is worth noting that the application of the PR regulators for the grid currents makes it possible to deliver a constant active power to the utility grid easily by making P ̃=0 in Equation 1, whereas an oscillating active power at twice the fundamental frequency (2w0) would be normally delivered if the control were exerted with conventional PI regulators in dq axes, unless a very complicated strategy and a huge amount of computational burden were used [27]." [L222-227]

The main goal of the paper is not to reduce the voltage sags, but to act when they happen as explained before.

Reviewer 3 Report

Review comments on:

Enhanced Control of Three-phase Grid-connected Renewables 2 with Fault Ride-through Capability under Voltage Sags

 

The authors focus on a Controller-HIL validation approach to show the behaviour of a control algorithm of 3-phase voltage source inverter that is a common element when connecting renewable energy sources like photovoltaic arrays or wind turbines to the power grid. They highlight the grid voltage sag and a possible operation of the renewable source supplying the grid with increased reactive power to improve the grid voltage shape.

The initial part is well described and the key highlights of other works are presented. The control scheme is defined, however, we know little about the performance of the simulated PV (or wind) generator, since “a DC source which mimics” the PV source is used in HIL “simulation”. Are there any problems to expect if the PV inverter is to be working at maximum power point (MPP) and a grid voltage sag occurs? How is this handled in “real world”? Can you give more details about this?

Next, some observations are given line by line:

L20: renewable agent; maybe source?

L33: keywords like DSOGI-FLL and similar are not classified as “proper” keywords; a keyword “constant active power control” is probably better just “active power control”?

L62: store > storage

P3 onwards: please correct the line numbering. It is strange looking now.

P3L14 (last paragraph): … is approximately equal to the…

P4L15 (below figure 1): synchronization algorithm: how does this work in case of a deep voltage sag? Or only 1 phase or 3 phase? Any comments on a fact that a PV inverter needs to be synchronized to the grid, so if there is a grid outage or voltage is substantially lower – can you comment on this?

There is a cross reference to Eq. 6, but we do not know nothing about yet. Can this sentence (…Vfault smaller than 0.85…) be modified for clarity?

P5L18: P^* and Q^* should be introduced as reference values.

P9L29: 4.1 Tables 1 and 2need more introduction to better clarify their data.

P10L32 and onwards: numbers and units should not be written italic

P11L34 onwards: Fig. 5 gives “experimental results” for a deep 3-phase voltage sag. Can you comment in more details, how/why this relatively low values of Q are delivered to the grid and why not the full P? Next, there is a “common sense” question – are the renewables capable of actively supporting the grid, meaning what happens in a limiting case when all conventional generators are disconnected and there are only renewables “online”. They need a synchronization to the grid which is not supported by “key players”. Any comments on this?

P12, P13, P14 and P15: maybe a notation “x 1e3” should be used on P(t) and Q(t) graphs to better show the kW or kVAr respectively.

Author Response

The authors thank the reviewer for his/her recommendations. Following our answers:

The reviewer is right. 

The authors have used a DC voltage source that "mimics the output of a two-stage system (eg. wind, PV) because a DC/DC boost converter is used between the Wind Turbine or the PV Generator and the grid-connected inverter through a large link capacitor. So, the output voltage of the DC/DC converter must remain constant to feed the inverter with the nominal DC voltage input. However, if a fault hap-pens, the MPP must not be tracked anymore, but a non-MPPT algorithm should be used instead (that is the reason why P*max is delivered to the mains when there is no fault in Figure 1, whereas P*fault is delivered when the fault happens). In this case the DC/DC will deliver a lower DC current because a lower power is achieved (non-MPPT) during the fault to allow the power balance. A similar situation for a single-stage grid-connected PV Sys-tem can be seen in [20]." [L154-163]

.
.

"...are sent to the block control of the DSOGI-FLL algorithm. This synchronization algorithm will calculate the PS and the NS values of the mains in αβ axes (u_gαβ^+ and u_gαβ^-) as well as the estimated system frequency w0’ for the PS value when unbalanced voltage sags happen. On the contrary, when balanced faults happen, there will be no NS value at all. So, the synchronization between the utility grid and the grid-connected inverter is never lost under normal or faulty modes of operation." [L188-193]

.
.
.

"Tables 1 and 2 provide an overview the power and control subsystem parameters which have been used for the following experiments using the well-known Controller Hardware-in-the-Loop (CHIL) validation approach [10,33]. In this, a digital real-time simulator (DRTS) and the real microcontroller to be used in the final prototype will be run in real time. All the parameters for both the power and the control subsystems are referred to Figure 1, whereas the physical meaning of the parameters for the non-ideal PR regulators are described in detail in [21]. 

The step sizes for both the DRTS and the microcontroller are TS = 5.1196 μs and TREG = 40.9568 μs respectively. Moreover, the inverse of the switching frequency (fSW) has the same value as TREG for the synchronization between the sample time of the 3-phase currents through the inverter and the digital PWM. The resulting controller delay is 3/2 T_REG=61.44 μs for the worst case [34]." [L340-352]

.
.
.

"...It must be said that, in all cases, the limitation in the amplitude of the grid currents is al-ways guaranteed when the appropriate instantaneous active power (Pmax or Pfault) is deliv-ered to the mains [27] according to the algorithm described in the flowchart of Figure 3." [L319-322]

.
.
.

"Finally, it is worth notice that the grid-connected renewable system analyzed in this work cannot support the grid because, in islanding mode, these kinds of systems are una-ble to guarantee a voltage with the proper amplitude and frequency at their outputs. This feature is because of the controlled variables are the 3-phase grid currents (and not the 3-phase voltages), and a disconnection from the mains is mandatory. So, for the case of having only renewables “online” (working in the so-called standalone mode) it is neces-sary to implement at least one renewable system capable to impose the 3-phase utility grid voltages at the output with an amplitude and a frequency in compliance with the corre-sponding grid-code, although no synchronization algorithm is needed in this case. The parallel operation of several renewables in the standalone mode is out of the scope of this paper." [L325-335]

Round 2

Reviewer 1 Report

Authors have addressed all of my comments properly. However, I can see some symbols which are not mentioned in nomenclature. Pls check again. Other than this overall paper has been improved considerably. 

Author Response

The authors would like to thank the reviewer for his/her suggestions to improve the quality of the paper.

The reviewer is right.

Some additional symbols have been added (in red color) to the Nomenclature in the second page of the revised manuscript for the reviewer's consideration.