Distribution Strategy Optimization of Standalone Hybrid WT/PV System Based on Different Solar and Wind Resources for Rural Applications

: The characteristics of solar and wind energy determine that the optimization of a standalone hybrid wind turbine (WT)/photovoltaic panel (PV) system depends on the natural resources of the installation location. In order to ensure system reliability and improve the resource utilization, a method for determining the installed capacity ratio of a hybrid renewable energy system is required. This study proposes a calculation method to optimize the installed capacity ratio, considering the system reliability to meet the needs of the hybrid system to adapt to different natural resources. In this paper, a standalone hybrid WT/PV system to provide electricity for rural areas is designed. Taking the power supply guarantee rate and electricity supply continuity as indicators, the system is simulated by using the Transient System Simulator solver. The results show that the recommended installed capacity ratio of the WT and PV is 5:1 when the total solar irradiation is less than 5040 MJ/(m 2 · a) and the annual average wind velocity is in the range of 3.0~3.5 m/s. When the annual average wind velocity is in the range of 2.0~3.0 m/s, the PV plays an increasingly signiﬁcant role in the hybrid system and exceeds the WT if the total solar irradiation is greater than 6300 MJ/(m 2 · a). However, if the total solar irradiation and the annual average wind velocity are less than 5040 MJ/(m 2 · a) and 2.0 m/s, respectively, it is not recommended to use the standalone hybrid system because it cannot meet the power demand. These conclusions provide guidance for the distribution strategies of the standalone hybrid WT/PV system within different natural resources.


Introduction
Renewable resources have attracted more and more attention due to the continuous energy demand and the impact of fossil fuels on the environment. Among all the renewable technologies, the photovoltaic panel (PV) and wind turbine (WT) are the most profitable ones [1]. However, the common disadvantage of wind and solar energy is their intermittency [2]. Both solar and wind resources are susceptible to weather fluctuations [3]. PV power generation in summer is usually higher than that in winter, while WT power generation increases during the winter seasons [4]. Therefore, taking advantage of the complementarity of these two resources and combining them properly to form a hybrid system can partially overcome the unpredictability [5]. In addition, because of a more stable overall annual power production, the hybrid system is extremely attractive [6], especially for remote areas [7]. Nevertheless, in the hybrid system, the power generation will inevitably fluctuate due to the oscillations of solar radiation and wind velocity. Hence, suitable battery banks are needed to mitigate the fluctuation of PV and WT power generation [8].
the classification discussion and system optimization design based on different solar and wind resources.
In the past decade, China's consumption of renewable energy has been increasing, and renewable energy has replaced nonrenewable energy more and more [25]. China has set ambitious targets to control carbon emissions by 2030 or earlier and intends to increase the use ratio of low-carbon energy to 20% [26]. This will lead to more renewable energy being incorporated into the Chinese power system, especially wind and solar energy. According to the latest China 2050 High Renewable Energy Penetration Scenario and Roadmap Study from the Energy Research Institute National Development and Reform Commission of China, the installed capacities of WTs and PVs will reach 64% of Chinese total power generation by 2050 [27]. However, there are still some problems in China's renewable energy development. Particularly, the distributions of solar and wind resources are uneven. This will lead to regional heterogeneity of renewable energy utilization. Many studies have conducted initial assessments of solar and wind resources within China as a necessary precursor to their utilization [28][29][30][31][32].
In this paper, a standalone hybrid WT/PV system is designed for rural electricity consumption as the load, and the optimal distribution strategies suitable for different regions are proposed by taking the different solar and wind resources in 26 regions of China. It is hoped that the research results of this paper could provide systematic design ideas and distribution strategies for areas with similar natural resources.

System Description
In this study, the software used for the modeling and simulation is TRNSYS. The standalone hybrid WT/PV system is established as shown in Figure 1. The main components used in the model of the standalone hybrid WT/PV system are: (1) PV module. TRNSYS Type 94 models the electrical performance of a photovoltaic array. In this type, the current-voltage characteristics of a single module are predicted by an empirical equivalent circuit model. (2) Wind turbine. TRNSYS Type 90 models a wind turbine. It makes use of some readily available manufacturer data as well as analytical solutions. In this type, the energy extracted by a wind turbine results from the change in momentum of the air moving through the rotor. (3) Storage battery. TRNSYS Type 47 operates in conjunction with solar cell array and power conditioning components. In this type, the battery state of charge varies over time and can be obtained given the rate of charge or discharge.
(4) Regulator and inverter. TRNSYS Type 48 models both the regulator and inverter. In this type, the regulator distributes the power from the PV and WT to and from the battery, and the inverter converts the DC power to AC and sends it to the load.
The mathematical descriptions of above types are shown in Section 2.2. The strength of the current source by the PV module and the power produced by the wind turbine are simulated based on the manufacturer data, empirical formulas, and the natural resources. The current-voltage equation for the equivalent circuit is [33]:

System Components
where I L is the module photocurrent, I O is diode reverse saturation current, q is electron charge constant, k is Boltzmann constant, and T c is module temperature. The negative reciprocal of the short-circuit IV slope closely approximates the shunt resistance: Equation (2) (6) where µ Voc is the temperature coefficient of open-circuit voltage, µ Isc is the temperature coefficient of short-circuit current, ε is semiconductor bandgap, and A is the ratio of γ to individual cells number in the module.
To compute the module temperature at each timestep, the temperature data from standard nominal operating cell temperature (NOCT) measurements are used. The module loss coefficient is determined by the NOCT data.
where τα is module transmittance-absorptance product, U L is module loss coefficient, T c,NOCT is module temperature at NOCT conditions, T a,NOCT is ambient temperature at NOCT conditions, and G T,NOCT is incident radiation at NOCT conditions. The module temperature at any timestep can be written as follows, assuming that the ratio of the module transmittance-absorptance product to the module loss coefficient is constant: where η c is the conversion efficiency of the module, which varies with ambient conditions. YL-280-30b is selected in this study, and the technical data is shown in Table 1. The PV modules are placed horizontally towards the south.

Wind Turbine
To simulate the wind turbine, some readily available manufacturer data, as well as analytical solutions included in Type 90, are used. The impact of the air density and hub height is considered in turbine power output. Wind turbines transform the kinetic energy of moving air into useful work. The power output of a wind turbine can be written as: where C p is the power coefficient as a function of the axial induction factor a, ρ is the air density, A WT is area of the rotor, and U is wind velocity in the free stream. The mass flow rate of air is a function of its density. Introducing the ideal gas law mboxemphp = ρRT, the pressure difference from one altitude to another can be written as: For elevations where wind energy applications apply, temperature decreases linearly with altitude [34]: where B = 6.5 K/km of altitude. Then an expression for pressure decrease with altitude can be obtained taking into account the temperature lapse rate: where T = 288 K. The modeling of wind velocity per change in height above the ground is based on the boundary layer theory of atmosphere [35]: where α determines the rate of wind velocity increase as a function of height. Under ideal boundary layer conditions, the value of α is taken to be 1/7. Three kinds of wind turbines are selected to adapt to different distribution strategies of the hybrid system. The technical data of them are shown in Table 2. The power output curves from the manufacturer are used, as shown in Figure 2. These data are gleaned from Guangzhou Infinite Wind Power Generator Manufacture Co., Ltd.

Storage Battery
Currently, lithium-ion and lead-acid are the commonly used batteries in stationary applications [37]. Among them, lead-acid battery is a popular choice for storing uninterruptible power supply due to its low cost, reliability, and efficiency [38]. The university of Navarra had built a microgrid platform that used a lead-acid battery pack consisting of 120 FIAMM SMG300 cells in series as its storage component [39]. A stand-alone hybrid power generation for a remote community in Bangladesh was studied by Das et al. [40]. Due to the participation of lead-acid batteries, the system could satisfy the energy demand of a primary load of 248 kWh/day and a peak load of 44.41 kW. Therefore, this paper chooses the lead-acid battery as the power storage component of the system. To simulate the lead-acid storage battery, the Hyman model which is more realistic at very low currents included in Type 47 is used.
On discharge, the formula is: On charge, it is: where V oc is the open circuit voltage at full charge, V zp is an additional voltage term in Hyman model, g c and g d are small-valued coefficients of H in voltage-current state, FSOC is the fractional state of the charge, Q is the state of the charge, Q m is rated capacity of the cell, r qc and r qd are internal resistances, m c and m d are cell-type parameters which determine the shapes of the I-V-Q characteristics, and Q c and Q d are capacity parameters on the charge and discharge.
To prolong the battery life, the battery should not be charged to too high a voltage or discharged to too low a voltage. The cutoff voltage on charge (V c ) and discharge (V d ) will be used in the regulator and inverter module so that the voltage charging and discharging can be initiated so to ensure not to exceed V c or to drop below V d .
Sail-GFM-200 is selected as the storage battery and the technical data are shown in Table 3. Taking into account the complementarity of solar and wind energy in terms of time, the number of self-contained days for the battery is selected as one day in the system design. Then, the total energy capacity of the battery is: where N 0 is the number of self-contained days, Q L is the daily average of annual electricity consumption, η is the inverter efficiency, and D is the maximum discharge depth.
The number of batteries in series: where V is the input voltage of the inverter and V 0 is the rated voltage of the battery. The number of batteries in parallel: where A c is the energy capacity of a single battery.

Regulator and Inverter
The battery's voltage level and charge/discharge rate as well as its state of charge will be monitored using Type 48. The charge to discharge limit on the fractional state of charge (F B ) is an important parameter in the regulator and inverter. If FSOC < F B and the battery has been charging, then the battery must be on the "total charge" state. On the "total charge," the first priority is given to recharging the battery rather than sending the output to the load until FSOC > F B . The technical data of the regulator and inverter are shown in Table 4.

Load Characteristics
To determine the load profile over a specific period of time, Type 14 h is used. TRNSYS Type 14 h provides a time-dependent forcing function model in transient simulation. The pattern of the forcing function is established by a set of discrete data points indicating its values at various times through one cycle. The hourly electricity consumption of rural residences in China is used as the load data in this study [41], as shown in Figure 3. The referenced hourly electricity consumption is the statistical value of 30 rural residences' test results, which reflects the characteristics of rural residential electricity consumption in different months in China. The annual electricity consumption is divided into three stages, namely the transition stage, the heating stage, and the air conditioning stage. May, June, September, and October are the transition stage. January, February, March, April, November, and December are the heating stage. July and August are the air conditioning stage. It can be seen from the figure that the hourly electricity consumption of rural residences presents the characteristics of three waves, and the occurrence time of the wave in the morning has variation at different stages which is in line with the reality of life. It needs to be explained here that China's rural energy consumption structure is developing towards a clean type, but the heating mode in winter is mainly coal burning [42]. Therefore, the electricity consumption of heating in winter is not considered in this paper.

Meteorological Data
For the sake of optimizing the distribution strategy of the hybrid system in different solar and wind resources, 26 regions of China are taken as samples, as shown in Figure 4.  Table 5. Hourly meteorological data are extracted from the TRNSYS database for a complete year.

Distribution Strategy
Based on the electricity consumption in rural areas, the average generating capacity of the hybrid system is 3.12 kWh/day. Taking this value as the benchmark, the distribution strategies of total power with the different installed capacity ratios of the PV and WT are shown in Table 6. Table 6. Distribution strategies of total power with the different installed capacity ratios of WT and PV. Because the voltage at reference condition with a maximum power point of the PV module is 31.10 V and the input voltage of inverter is 24.00 V, the number of PV modules in the series is one. The number of PV modules in parallel can be calculated by:

Installed Capacity Ratio of WT and
where P PV is the generating capacity of the PV modules, N PV−s is the number of PV modules in series, and P e is the maximum power at the reference condition.

Procedure of Methodology
The procedure of the methodology of this study is shown in Figure 5. Based on TRN-SYS, through three types of input data, the power supply guarantee rate and electricity supply continuity under different distribution strategies are obtained. Then, the recommended installed capacity ratio of the region is proposed after comparing the results. After that, the distribution strategies suitable for different natural resources are summarized by the simulation of different regions.

Model Verification
To verify the modules utilized by TRNSYS in this paper, the meteorological data and the sizes of the PV and WT in the literature are used [13]. The solar irradiation and wind velocity distributions during the year are shown in Figure 6a according to the literature. The comparisons of the power output between the literature data and the simulated data of TRNSYS are shown in Figure 6b. The specifications of the PV module and the WT are shown in Table 7. It can be seen that the power output of the simulation results during the year are in good agreement with the literature data. The maximum relative error is only 7.15%, which verifies the model reliability of the PV module and the WT in TRNSYS.

Case Analysis
In this section, two representative regions are selected to analyze the influence of the different distribution strategies on the hybrid system.

Case One Wind and Solar Resources
Kaba-He is chosen for its abundance of solar and wind energy. Its wind and solar resources are both in the range of the first category. The daily radiation and wind velocity data of different months in a year are shown in Figure 7. As can be seen from the figure, solar energy and wind energy complement each other in a day. Solar radiation is intermittent, and its value is zero at night. However, although there is no fixed trend in wind velocity over the course of the day, the wind velocity is generally higher at night. The monthly average radiation and wind velocity data of Kaba-He are shown in Figure 8. It can be seen that solar energy and wind energy also complement each other over the months. The average solar radiation increases first and then decreases with the increase of month, whereas the trend of average wind velocity is opposite to that of average radiation.

Hourly Characteristics in Typical Days
When the installed capacity ratio of the WT and PV is 5:1, the hourly characteristics for two typical days (15 January and 15 July) are shown in Figure 9. Combined with the radiation and wind velocity data on January 15 (Figure 7), it can be seen that the power output of the WT and PV are consistent with the changes of the wind velocity and radiation. On 15 January 15, the maximum wind velocity is 8.10 m·s −1 at 23:00, and the WT output is 2188.01 W at the same time. The maximum radiation is 795.60 kJ·h −1 ·m −2 at 14:00, and the PV output is 99.94 W at the same time. As shown in Figure 9b, the battery is in a state of neither charging nor discharging for two periods (9:00~11:00 and 17:00~24:00). This is because the FSOC value of the battery is greater than the upper limit of 1.00 during those periods, while the WT and PV output can meet the load consumption and there is excess power. The excess power will be dumped by the system, as shown in Figure 9a. On 15 July, the wind resource is poor, and the WT output is very little as the wind velocity is not up to the cut-in wind velocity. The load electricity consumption mainly comes from the battery and PV output. As shown in Figure 9d, before 11:00, the battery is in the discharge state, and the FSOC goes down from 0.97 to 0.95; between 12:00 and 16:00, the battery is in the charge state, and the FSOC goes up from 0.95 to 0.96; after 17:00, the battery is in the discharge state. It can be seen from Figure 9c that from 8:00 to 22:00, the PV output is greater than zero, while the PV output could only not meet the load electricity consumption from 12:00 to 16:00, and there is a surplus which will be stored in the battery.

Effect of Different Distribution Strategies
To compare the influence of different distribution strategies on the hybrid system, the power supply guarantee rate is used as an evaluation index. The power supply guarantee rate with the different installed capacity ratios of the WT and PV are shown in Figure 10. The power supply guarantee rate does not reach 100% because the battery is in a state of no power at first. The WT and PV output obtained in the early stage need to be charged to the battery until the FSOC reaches the threshold (0.30). The hourly FSOC of the first ten days and the start time of the electricity supply continuity with the different installed capacity ratios of the WT and PV are shown in Figure 11. As shown in Figure 10, the power supply guarantee rate varies little when the installed capacity ratios of the WT and PV are 5:1~1:1. When the ratio is less than 1:1, the power supply guarantee rate decreases obviously. This is because when the ratio is from 5:1 to 1:1, it takes around 5 days for the FSOC to become greater than 0.30. In addition, the system can guarantee the continuity of the electricity supply when the FSOC is greater than 0.30. However, when the ratio is less than 1:1, as an example of 1:2, although the FSOC reaches 0.30 at 136 h, it fluctuates between 0.28 and 0.31 until 191 h, so the start time of the electricity supply continuity is 191 h. Therefore, according to the wind and solar resources of Kaba-He, the recommended installed capacity ratios of the WT and PV for the hybrid system is from 5:1 to 1:1.

Case Two
Andir is chosen because it is rich in solar energy but short on wind energy. Its solar energy resource is in the range of the first category, whereas the wind energy resource is in the range of the fourth category. The monthly average radiation and wind velocity in Andir are shown in Figure 12. It can be seen that the maximum average wind velocity is 1.82 m·s −1 in March and the minimum average wind velocity is 0.90 m·s −1 in December. The power supply guarantee rate with the different installed capacity ratios of the WT and PV in Andir are shown in Figure 13. The hourly FSOC of the first twenty days and the start time of the electricity supply continuity are shown in Figure 14. As shown in Figure 13, the power supply guarantee rate rises significantly when the installed capacity ratio of the WT and PV is from 5:1 to 1:1. When the ratio is less than 1:1, the power supply guarantee rate increases slowly. This is because when the ratio is greater than 1:1, as an example of 2:1, although the FSOC reaches 0.30 at 184 h, it fluctuates between 0.28 and 0.31 until 469 h, so the start time of electricity supply continuity is 469 h. However, when the ratio is less than 1:1, as an example of 1:5, it takes 86 h for the FSOC to become greater than 0.30, and the system can guarantee the continuity of the electricity supply after that time. Therefore, according to the wind and solar energy resources of Andir, the recommended installed capacity ratio of the WT and PV of the hybrid system is 1:5.

Summary
Taking the power supply guarantee rate and the electricity supply continuity as indicators, the standalone hybrid WT/PV system with different solar and wind resources in 26 regions of China are simulated. The recommended installed capacity ratios of the WT and PV are shown in Table 8. It can be seen that the recommended installed capacity ratio of the WT and PV is 5:1 when the total solar irradiation is less than 5040 MJ/(m 2 ·a) and the annual average wind velocity is in the range of 3.0~3.5 m/s. In addition, when the WER is still in this range, the scope of the recommended installed capacity ratio of the WT and PV increases with the increase of the SER. This is because when the WER is in the range of the first category, the WT dominates in the hybrid system. With the improvement of the SER, the PV plays an increasingly apparent role but does not exceed the WT. However, when the WER is in the range of the second and third categories, the PV plays an increasingly significant role in the hybrid system and exceeds the WT if the total solar irradiation is greater than 6300 MJ/(m 2 ·a). When the WER is scarce, the PV dominates in the hybrid system. It is worth noting that when there are several options for the installed capacity ratio of the WT and PV, the capacity with less PV should be selected because the reduction of the PV capacity could reduce the total capital cost [8,20]. In particular, the standalone hybrid WT/PV system is not recommended if the total solar irradiation is less than 5040 MJ/(m 2 ·a) and the annual average wind velocity is less than 2.0 m/s. Taking Chengdu as an example, although the power supply guarantee rate reaches its maximum 95.99% when the installed capacity ratios of the WT and the PV is 1:5, the continuity of the electricity supply cannot be guaranteed under any ratio.

Conclusions
In this paper, a standalone hybrid WT/PV system is designed for the purpose of providing electricity in rural areas. The system is simulated by the TRNSYS solver, and the input parameters are the meteorological data of the selected locations and the specifications of the WT and PV. The installed capacity ratio of the PV and WT is taken as the index of the distribution strategy to optimize the hybrid system. To obtain the optimal distribution strategies suitable for different natural resources, 26 regions of China are taken as samples. Several references for the optimal distribution strategies of the standalone hybrid WT/PV system within different natural resources are as follows: (1) When the annual average wind velocity is in the range of 3.0~3.5 m/s, the recommended installed capacity ratios of the WT and PV are 5:1 for when the total solar irradiation is less than 5040 MJ/(m 2 ·a), 5:1~3:1 for when the total solar irradiation is in the range of 5040~6300 MJ/(m 2 ·a), 5:1~2:1 for when the total solar irradiation is in the range of 6300~6660 MJ/(m 2 ·a), and 5:1~1:1 for when the total solar irradiation is greater than 6300 MJ/(m 2 ·a). (2) When the annual average wind velocity is in the range of 2.5~3.0 m/s, the recommended installed capacity ratios of the WT and PV are 3:1~2:1 for when the total solar irradiation is less than 5040 MJ/(m 2 ·a), 1:1 for when the total solar irradiation is in the range of 5040~6300 MJ/(m 2 ·a), 1:1~1:2 for when the total solar irradiation is in the range of 6300~6660 MJ/(m 2 ·a), and 1:2 for when the total solar irradiation is greater than 6300 MJ/(m 2 ·a). (3) When the annual average wind velocity is in the range of 2.0~2.5 m/s, the recommended installed capacity ratios of the WT and PV are 3:1~2:1 for when the total solar irradiation is less than 5040 MJ/(m 2 ·a), 1:1 for when the total solar irradiation is in the range of 5040~6300 MJ/(m 2 ·a), 1:2 for when the total solar irradiation is in the range of 6300~6660 MJ/(m 2 ·a), and 1:3 for when the total solar irradiation is greater than 6300 MJ/(m 2 ·a). (4) When the annual average wind velocity is less than 2.0 m/s, the recommended installed capacity ratios of the WT and PV are 1:2~1:3 for when the total solar irradiation is in the range of 5040~6300 MJ/(m 2 ·a), 1:3 for when the total solar irradiation is in the range of 6300~6660 MJ/(m 2 ·a), and 1:5 for when the total solar irradiation is greater than 6300 MJ/(m 2 ·a). (5) If the total solar irradiation is less than 5040 MJ/(m 2 ·a) and the annual average wind velocity is less than 2.0 m/s, it is not recommended to use the standalone hybrid system because it cannot meet the power demand.