Studies on integration of different capacity offshore wind farms in power grid

Abstract

This work is in continuation of our earlier pre-feasibility study that was carried out to investigate the prospects of constructing an offshore wind farm in the coastal region of south India. After selection of the suitable site and optimal wind turbine, another challenge was to penetrate wind energy in the grid. The present study aims to select an economical and technically feasible connection point in the test grid for wind farm integration. For each of the seven available connection points, load flow analysis was done using MATLAB, to assess how grid reacts to the power produced by different capacity wind farms. Results of the present study suggest that the offshore wind farms consisting of 10 to 23 turbines (RE power 6.2M), may be connected at grid point C, where the voltage level after penetration of wind farm was found to be within the acceptable range.

Keywords

Offshore wind farm grid integration load flow analysis short circuit capacity voltage regulation

Introduction

Escalated depletion of conventional energy resources along with the excessive increase in consumption and emission rates have shifted the focus on utilisation of renewable sources of energy and search for feasible and economic solutions. Wind energy being an abundant, clean, environment friendly, cost effective and inexhaustible energy source has gained wide acceptance across the globe for providing adequate energy to the developing world. Weibull frequency distribution has been documented as an important statistical tool to predict the wind energy output at various locations (Rehman et al., 2018; Sumair et al., 2020; Tizgui et al., 2019). In a recent study, we have done wind resource assessment, at the two locations in the coastal region of south India, by weibull frequency distribution using the four different techniques (Riaz and Khan, 2019). Based on the wind characteristics results, a suitable wind regime and a wind turbine that gave the highest energy yield were selected (Figures 1 and 2 and Table 1). Another challenging task in the development of offshore wind farm is the penetration of wind energy into the power grid, so that the voltage stays within an acceptable range. Voltage variation is a major problem associated with wind energy integration and is the limiting factor for the amount of wind power that can be integrated into the power system (Ayodele, 2012; Chen, 2001; Matevosyan, 2006). At the connection point there may be consumers or other generating sources, thus the voltage has to be kept within certain limits, at least ±10%. This could be done keeping in view the strength of the grid and the possibility of voltage control in the wind power plant. The strength of AC grids can be taken as a standard to define the acceptable limit of integrated wind power and needs to be considered when connecting wind power to AC grids. The short circuit capacity is an indicator of the strength of AC grid to which the energy source is to be connected. The effect of connecting a wind farm on the grid voltage is correlated to the short circuit power level, and that in turn depends on the rated voltage and the absolute value of AC grid impedance (Chen, 2005; Golieva, 2015; Söder, 2007). A high impedance of AC grids leads to a low short-circuit level at a constant rated voltage of a connection point. Thus, usually a low short-circuit level corresponds to a weak AC grid, whereas, a high short-circuit level corresponds to a strong AC grid (Ackermann and Söder, 2005; Chen, 2005; Söder, 2017). When the short circuit level is low or the grid impedance is high and loading is heavy, then the voltages at the buses along the transmission lines become highly sensitive to changes in the power generation (Ahmed et al., 2020; Feilat et al., 2018; Li et al., 2018).

Table 1.

Specification of wind turbines considered for previous study (Riaz and Khan, 2019).

Commercial turbines		Hub height (m)	Rated power (MW)	Rotor diameter (m)
WT-1	Gamesa G128-5 MW Offshore	80	5	128
WT-2	Gamesa G128-5.0 MW Offshore	80	5	128
WT-3	Gamesa G132-5 MW Offshore	95	5	132
WT-4	GE 2.75-100	85	2.75	100
WT-5	GE 2.75-120	85	2.75	120
WT-6	REpower 6.2M126 Offshore	85	6.15	126
WT-7 *	REpower 6.2M152 Offshore	95	6.15	152
WT-8	REpower 5M offshore	85	5.075	126
WT-9	REpower 6M Offshore	85	6.15	126
WT-10	Siemens SWT-3.6-120	90	3.6	120
WT-11	Sinovel SL6000/128 Offshore	110	6	128

selected turbine for grid integration.

Figure 1.

Location of the offshore wind farms considered in the previous study (Research Center O.I.E. and Transvalor S.A., 2019).

Figure 2.

Annual energy output of the above turbines (Riaz and Khan, 2019).

Several earlier studies have addressed the problems related to integration of wind energy into power systems (Ahmed et al., 2020; Chen, 2005; Chi et al., 2007; Tande JOG, 2003; Ibrahim et al., 2011; Jigoria-Oprea et al., 2011). In a recent study the penetration limit of wind power in the regional power grid was assessed and it was found that the strength of wind power grid is the decisive factor for the stable operation of the grid (Wen et al., 2018). Another study analysed the problems of connecting wind power plant into a weak grid and identified that the voltage stability is the most critical technical challenge for the stable operation of wind power plant within a weak grid (Zhou et al., 2013). Furthermore, the voltage stability of wind power grid has been studied by evaluating the relationship between voltage stability and wind farm capacity (Zou and Zhou, 2011). In a previous study, the impact of wind power on power quality and system stability was assessed, the grid requirements for integration of wind turbines were analysed and the control methods to meet these challenges were suggested (Ahmed et al., 2020; Chen, 2005). Analysis of a serious tripping-offline contingency of wind turbines, that occurred at Guyuan of China on May 14, 2012, has shown that the excessive output of wind farm caused voltage instability, although no equipment failure was reported during the operation (Xu et al., 2014). Investigation of the effect of large scale grid-connected wind generators on the power system network has shown that the voltage stability of the system depends largely on the technology used for wind generation. The system is more vulnerable to voltage instability when fixed speed wind generators are used, whereas the variable speed wind generators tend to perform better when connected to the grid (Mathe and Folly, 2017). In another study, the impact of wind power integration both in the strong and weak grids was analysed and the voltage stability limit of the AC grid at different short circuit levels was evaluated by increasing the grid load or active power injection from energy source (Nawir, 2017).

Thus, when planning grid connection of a wind farm, two major aspects have to be considered, firstly, the grid capacity has to be analysed for each proposed connection point and secondly, when a suitable point has been chosen, transformers and connection lines have to be analysed (Söder, 2007, 2017). As wind turbines are usually designed to operate within a specific voltage range (i.e. nominal voltage ± 10%) hence, wind farms require a certain voltage level at the connection point. The installed capacity of a wind farm ideally should not be more than 10% of the short circuit capacity. However, in some cases ratios up to 30% are acceptable (Ackermann and Söder, 2005; Söder, 2007).

In view of this, the present study analyses the voltage stability and capacity issues of the wind farms from the aspect of static voltage using load flow analysis and aims to select an economical and technically feasible connection point in a test grid, so that the voltage should remain within an acceptable range upon connecting different capacity wind farms.

Methodology

Grid integration of wind farms

Figure 3, shows the grid connection of a wind farm, where grid can be considered as a voltage source, U_s, next to the impedance, Z_k, that represents the impedance in all transmission lines, cables and transformers in the grid. The point at which wind farm is to be connected may have a local load. The short circuit power, S_k, at the connection point for wind power is calculated as $S_{k} = \frac{U_{s}^{2}}{Z_{k}}$ .

Figure 3.

An illustration of grid connected wind farm.

Any change in wind power production will cause change in the current through the impedance Z_k and this current change will lead to change in the voltage U_w. If the impedance Z_k is small then the voltage variations will be small resulting in a strong grid. In contrast, if Z_k is large, then the voltage variations will be large and hence it will be a weak grid.

The network voltage at the assumed infinite busbar is U_s and the voltage at the point of common coupling (PCC) U_w. The output power and reactive power of the generation unit are P_w and Q_w, respectively, which correspond to current I_w (Chuong, 2007), as shown below,

I_{w} = {(\frac{S_{w}}{U_{w}})}^{2} = \frac{P_{w} - j Q_{w}}{U_{w}}

(1)

The voltage difference between the system and the connection point, ΔU, is given by

\begin{matrix} U_{w} - U_{s} = Δ U = Z_{k} I_{w} = (R_{k} + j X_{k}) (\frac{P_{w} - j Q_{w}}{U_{w}}) \\ = \frac{R_{k} P_{w} + X_{k} Q_{w}}{U_{w}} + (j \frac{P_{w} X_{k} - Q_{w} R_{k}}{U_{w}}) = Δ U_{p} + j Δ U_{q} \end{matrix}

(2)

The voltage difference, ΔU, is related to the short circuit impedance and also to the real and reactive power output of the wind power generation unit. It is now apparent that any variation in the generated power will result in the variation of the voltage at PCC.

Wind power production and load may be represented as P+jQ, where P is the active power and Q is the reactive power that depends on the phase shift between voltage and the current, such as

$Q = (\frac{Sin ϕ}{Cos ϕ}) \times P$ where, cos Φ denotes the power factor.

From the Figure 3 the voltage U_w can be calculated as done earlier (Ackermann and Söder, 2005; Söder, 2007, 2017):

U_{w}^{2} = - \frac{a_{4}}{2 a_{3}} \pm \sqrt{{(\frac{a_{4}}{2 a_{3}})}^{2} - \frac{1}{a_{3}} (a_{1}^{2} + a_{2}^{2})}

(3)

a_{1} = R (P_{w} - P_{LD}) - X (Q_{w} - Q_{LD})

(4)

a_{2} = X (P_{w} - P_{LD}) - R (Q_{w} - Q_{LD})

(5)

a_{3} = {(1 - X_{k} B)}^{2} + {R_{k}}^{2} B^{2}

(6)

a_{4} = 2 \cdot a_{1} (1 - X_{k} B) - U_{s}^{2} + 2 a_{2} R_{k} B

(7)

B = 0, (line is modelled as short line and hence line capacitance will be neglected)

P_LD=active load power, P_W=active wind power, Q_LD=reactive load power, Q_W = reactive wind power

Parameters of the test grid used in the study (Figure 4)

in station M11, 11 kV with voltage control ± 8 × 1.67%

station T900 has short circuit power 18 MVA (cos(φ) = 0.7)

all voltage values are line-to-line.

for voltages 66 kV and above, the voltage limits are ±10%

for voltages below 66 kV, voltage limits are −10%, +6%

Figure 4.

Test Grid used for analysis of wind farm integration.

Assumptions

the grid is assumed to be purely inductive.

wind farm produces only active power

load power factor is assumed to be 0.9 everywhere.

the wind farm is controlled to preserve a power factor equal to 1

Preliminary selection of connection point:

RE Power 6.2M 126 offshore wind turbine of 6.15 MW was chosen, as it gave maximum production in the selected wind regime. Integration of ten such turbines would lead to the development of 61.5 MW offshore wind farm. The connection of the wind farm introduces a voltage variation at the connection node. The voltage variation caused by the integration of the wind farm can be estimated according to the equation

\frac{P_{w}}{S_{k}} % \approx \frac{U - U_{T}}{U_{T}} % = Δ U % = Δ U %

(8)

Table 2 presents the approximation of voltage variation; the values are calculated using the above-mentioned equation (8).

Table 2.

Comparison of the voltage variations for each of the potential connection points.

No. of Windturbinesin the windfarm(6.15 MW)	PotentialConnectionpoint	Short Circuitcapacity (Sk)	Voltage variationdue towind powerconnection Pw/Sk ≈ ΔU%	Preliminary acceptancefor connectionpoint	Economical	Load flow analysis results(max voltage variation)
No. of Windturbinesin the windfarm(6.15 MW)	PotentialConnectionpoint	Short Circuitcapacity (Sk)	Voltage variationdue towind powerconnection Pw/Sk ≈ ΔU%	Preliminary acceptancefor connectionpoint	Economical	Max Voltage (kV)	% variation
5	A	1155	2.66	Yes	No
	B	735	4.18	Yes	No
	C	480	6.41	Yes	No	65.0477	1.44
	D	256	12.01	Yes	Yes	32.3637*	1.93
	E	87	35.34	No	———
	F	57	53.95	No	———
	G	8.4	366.07	No	———

10	A	1155	5.32	Yes	no
	B	735	8.37	Yes	no
	C	480	12.81	Yes	yes	64.7469	1.90
	D	256	24.02	Yes	yes
	E	87	70.69	No	———
	F	57	107.89	No	———
	G	8.4	732.14	No	———
15	A	1155	7.99	Yes	no
	B	735	12.55	Yes	no
	C	480	19.22	Yes	yes	64.1334	2.83
	D	256	36.04	No	———
	E	87	106.03	No	———
	F	57	161.84	No	———
	G	8.4	1098.21	No	———
20	A	1155	10.65	Yes	no
	B	735	16.73	Yes	no
	C	480	25.63	Yes	Yes	63.1577	4.31
	D	256	48.05	No	———
	E	87	141.38	No	———
	F	57	215.79	No	———
	G	8.4	1464.29	no	———
23	A	1155	12.25	Yes	no
	B	735	19.24	Yes	no
	C	480	29.47	Yes	Yes	62.3898	5.47
	D	256	55.25	No	———
	E	87	162.59	No	———
	F	57	248.16	No	———
	G	8.4	1683.93	No	———
25	A	1155	13.31	Yes	no
	B	735	20.92	Yes	Yes	64.1241	2.84
	C	480	32.03	No	———
	D	256	60.06	No	———
	E	87	176.72	No	———
	F	57	269.74	No	———
	G	8.4	1830.36	No	———
30	A	1155	15.97	Yes	no
	B	735	25.10	Yes	Yes	65.3905	0.92
	C	480	38.44	No	———
	D	256	72.07	No	———
	E	87	212.07	No	———
	F	57	323.68	No	———
	G	8.4	2196.43	No	———
35	A	1155	18.64	Yes	Yes	64.5599	2.18
	B	735	29.29	Yes	Yes	56.4537	14.46
	C	480	44.84	No	———
	D	256	84.08	No	———
	E	87	247.41	No	———
	F	57	377.63	No	———
	G	8.4	2562.50	No	———

voltage variation at 33kV line.

Results

Different capacity wind farms with varying number (5–35) of wind turbines, were selected for the study. Comparison of the voltage variation at seven potential connection points (A–G) was done for each of the wind farm (Table 2). For 10 turbine wind farm, voltage variation at points A, B, C and D were found to be within the acceptable range, that is, showed less than 30% voltage variation (Table 2). Voltage variation at point C is shown in Figure 5. However, at points E, F and G much higher voltage variation was observed and hence, these points will not be suitable for wind power integration. Further suitability of the connection points (A–D) for wind integration was determined on the basis of economical and load flow analysis. For economic analysis cost of network components was taken into consideration. As we go further close to a stronger grid (towards A), then the higher rating network components would be needed and therefore point D would be economically more feasible as the connection point in the test grid. However, further load flow analysis showed lesser voltage variation at point C, and therefore, point C would be the most suitable point for integration of a wind farm with 10 turbines of 6.15 MW capacity.

Figure 5.

Voltage variation at connection point C (short circuit capacity 480 MVA).

Load flow analysis

The selected points were then tested by load flow analysis in the sequence D, C, B and A that is, from weak to the strong part of the grid.

At point D

The equivalent load current per line is 140 A–380 A, for a phase voltage of 33 kV. Since there are three lines, the total equivalent load capacity at point D would be in the range.

3 \times 140 A \times \frac{33 kV}{\sqrt{3}} = 8.002 MW to 3 \times 380 A \times \frac{33 kV}{\sqrt{3}} = 21.72 MW

The worst situation that could cause a maximum voltage variation at this point would be a small load, that is 8.002 MW and a maximum wind power that is 61.5 MW and hence the net power transmitted to transformer Tr2 would be 53.498 MW, but as the capacity of this transformer is 40 MW, consequently this transformer would be overloaded. Thus, due to the overloading risk, point D is not chosen as the connection point.

At point C

The capacity of the 66 kV transformer is 220 MW, hence there would be no risk of overloading of this transformer. The analysis was done by, first calculating the equivalent Thevenin impedance at the connection point (Figure 6), and then using the load flow analysis, the maximum voltage variation at the connection point was calculated using the above equations (3)–(7) as described earlier (Ackermann and Söder, 2005; Söder, 2007, 2017).

Figure 6.

Thevenin equivalent circuit at point C.

R = Sum of the resistive component of Z_thC

R = 0 Ω (pure inductive)

X = sum of the inductive components of Z_thC

X = $j \frac{{(66 kV)}^{2}}{480 MVA} =$ j9.075 (purely inductive)

B = 0, because the line capacitances are neglected.

U_{j} = {\bar{U}}_{T} = [\frac{66}{\sqrt{3}}] kV

Considering bus j as slack bus and bus k as PQ bus

The load would vary as

P_{LD} = 15 x Cos (φ) = 15 x 0.9 = 13.5 MW Q_{LD} = 15 x Sin (φ) = 15 x 0.435 = 6.54 MVAr

P_{LD} = 40 x Cos (φ) = 40 x 0.9 = 36.0 MW Q_{LD} = 40 x Sin (φ) = 40 x 0.435 = 17.44 MVAr

Considering the worst situation when load is small and wind power production is maximum, the voltage variation at the connection point was calculated when the power flows from bus k to bus j;

P_{kj} = P_{w} - P_{LD} and Q_{kj} = 0 - Q_{LD}

Therefore, worst case is, when P_LD = 13.5 MW and Q_LD = 6.54 MVAr

A simulation was made using MATLAB to see how the grid reacts to the power produced from the wind farm. At the point C, the maximum voltage variation was found to be 66 kV to 64.7469 kV. This value is acceptable as it fulfils the voltage variation constraints at point C that is, ±10% (Figure 7), which is of the range; 66 kV + 6.6 kV = 72.6kV to 66 kV−6.6 kV = 59.4 kV. If we go further close to a stronger grid (towards A), then the required network components would be of higher rating and hence would be more expensive. Thus, point C would serve as an economically and technically feasible connection point in the test grid for integration of 61.5 MW (10 turbines) wind farm.

Figure 7.

Voltage variation as determined by load flow analysis for grid integration of 10–23 turbines.

Conclusion

Using the mathematical model of grid-connected wind farm, the relationship between the output power of wind farm and voltage at the connection point was analysed. Results of the present study show that offshore wind farms consisting of varying number, that is 10, 15, 20 and 23 of RE Power 6.2M 126 wind turbines (WT-7), at the hub height of 95 m may be connected at point C in the grid, where the voltage level after penetration of wind farms was found to be within the acceptable range and also fulfilled the voltage stability requirement of the wind farm operator. In contrast, opting for more strong grid connection points, with high short circuit capacities, would not be economical as the highly rated network components would be required, further adding to the cost of energy. However, if the wind farm consists of 24 or more turbines then point C would not be suitable and instead points A and B with higher short circuit capacities may be used. While, for the wind farm consisting of 35 turbines the load flow analysis showed that at point B the voltage variation is 14.46% which exceeds the limit of ±10%, therefore in that case, point A would be more suitable for grid connection.

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Mohammad Mushir Riaz

References

Ackermann

Söder

(2005) Wind power in power systems: An introduction. In: Ackermann

(ed.) Wind Power in Power Systems. Chichester, UK: John Wiley & Sons, Ltd, pp.25–49.

Ahmed

Al-Ismail

FSM

Shafiullah

, et al. (2020) Grid integration challenges of wind energy: A review. IEEE Access 8(type 1): 10857–10878.

Ayodele

(2012) Challenges of grid integration of wind power on power system grid integrity: A review. International Journal of Renewable Energy Research 2(4). Available at: https://www.ijrer.org/ijrer/index.php/ijrer/article/view/317/pdf (accessed on 4 April, 2020).

Chen

(2001) Grid power quality with variable speed wind energy conversion. IEEE Transactions on Energy Conversion 16(2): 148–154.

Chen

(2005) Issues of connecting wind farms into power systems. In: Proceedings of the IEEE power engineering society transmission and distribution conference, Dalian, China, 18 August 2005, pp.1–6. New York: IEEE.

Chi

Liu

Wang

, et al. (2007) Voltage stability analysis of wind farm integration into transmission network. In: 2006 international conference on power system technology, POWERCON2006, Chongqing, China, 22–26 October 2006, pp.1–7. New York: IEEE.

Feilat

Azzam

Al-Salaymeh

(2018) Impact of large PV and wind power plants on voltage and frequency stability of Jordan’s national grid. Sustainable Cities and Society 36: 257–271.

Golieva

(2015) Low short circuit ratio connection of wind power plants. M.Sc. Thesis, TU Delft, The Netherlands.

Ibrahim

Ghandour

Dimitrova

, et al. (2011) Integration of wind energy into electricity systems: Technical challenges and actual solutions. Energy Procedia 6: 815–824.

10.

Jigoria-Oprea

Barbulescu

Kilyeni

, et al. (2011) Large wind farms integration in Romanian power system. Case study: Moldavia region. In: SACI 2011 – 6th IEEE international symposium on applied computational intelligence and informatics, proceedings, Timisoara, Romania, 19–21 May 2011, pp.565–570. New York: IEEE.

11.

Zhang

, et al. (2018) Volt/var control for power grids with connections of large-scale wind farms: A review. In: IEEE Access, 1 May 2018, vol. 6, pp.26675–26692. New York: IEEE.

12.

Matevosyan

(2006) Wind power integration in power systems with transmission bottlenecks. PhD Thesis, Royal Institute of Technology, Sweden.

13.

Mathe

Folly

(2017) Impact of large scale grid-connected wind generators on the power system network. In: 2017 IEEE PES PowerAfrica, Accra, Ghana, 2017, pp. 328–333. Ghana: Institute of Electrical and Electronics Engineers Inc. (IEEE).

14.

Nawir

(2017) Integration of wind farms into weak AC grids. PhD Thesis, Cardiff University, UK. Available at: https://orca.cf.ac.uk/107597/1/2017NawirMPhD.pdf

15.

Rehman

Mohandes

Alhems

(2018) Wind speed and power characteristics using LiDAR anemometer based measurements. Sustainable Energy Technologies and Assessments 27: 46–62.

16.

Research Center O.I.E. and Transvalor S.A . (2019) MERRA-2 meteorological re-aalysis wind data. Available at: http://www.soda-pro.com/web-services/meteo-data/merra (accessed 17 May 2019).

17.

Riaz

Khan

(2019) Estimation of Weibull parameters and selection of optimal wind turbine for the development of large offshore wind farm. In: International conference on electrical, electronics and computer engineering, UPCON 2019, 8–10 November 2019. Aligarh: Institute of Electrical and Electronics Engineers Inc. (IEEE). DOI: 10.1109/UPCON47278.2019.8980167.

18.

Söder

(2007) Compendium 2C117, Wind Power System. Stockholm, Sweden: Royal Institute of Technology.

19.

Söder

(2017) Lectures on integration of high penetration of solar and wind power in power systems. Available at: https://www.kth.se/profile/lsod/page/lectures-on-integration-of-high-penetration-of-solar-and-wind-power-in-power-systems (accessed 1 January 2020).

20.

Sumair

Aized

Gardezi

SAR

, et al. (2020) Wind potential estimation and proposed energy production in Southern Punjab using Weibull probability density function and surface measured data. Energy Exploration & Exploitation 166: 0144598720920748.

21.

Tande JOG. Grid Integration of Wind Farms. Wind Energy 2003; 6(3): 281–295. Available at: https://doi-wiley-com.web.bisu.edu.cn/10.1002/we.91

22.

Tizgui

El Guezar

Bouzahir

, et al. (2019) Wind speed distribution modeling for wind power estimation: Case of Agadir in Morocco. Wind Engineering 43(2): 190–200.

23.

Wen

Wang

(2018) Study on wind power penetration limit of regional power system based on grid constraints. In: Chinese Control Conference, CCC, Wuhan, China, 25–27 July 2018, pp.8979–8981. Technical Committee on Control Theory, Chinese Association of Automation. New York: IEEE.

24.

Jia

, et al. (2014) Practical criteria of static voltage stability in power systems with high wind penetration. Dianli Xitong Zidonghua/Automation of Electric Power Systems 38(9): 15–19.

25.

Zhou

Nguyen

Kjær

, et al. (2013) Connecting wind power plant with weak grid – challenges and solutions. In: IEEE Power & Energy Society General Meeting, 21–25 July 2013, pp.1–224. Canada: Institute of Electrical and Electronics Engineers Inc. (IEEE). DOI: 10.1017/CBO9781107415324.004.

26.

Zou

Zhou

(2011) Voltage stability of wind power grid integration. In: International conference on electrical machines and systems, ICEMS 2011, Beijing, China, 20–23 August 2011.