Study the optimization of offshore floating PV power systems and its application

In recent years, the application of solar photovoltaic (PV) has been developing rapidly with a large amount of PV power stations built and operating. However, the land resources for PV power plants are relatively limited. Offshore PV power plants are expected to become an alternative means to solve such problems. In this paper, the PVsyst and AQWA are used to comprehensively simulate the power generation performance of floating PV power stations and thin-film PV power stations under offshore conditions, and the factors affecting the performance of floating PV systems are studied. By establishing the nonlinear regression numerical models, the relationship between the factors is compared by path analysis. The results show that the swing angle has the greatest influence on the system performance. The above work provides certain guidelines for the development and construction of offshore PV systems.


Introduction
Along with the development of renewable energy technologies, offshore PV power systems have potential advantages over onshore PV power systems including 1) the marine environment helps to reduce the temperature of PV modules and improve efficiency, 2) offshore PV power systems reduce the occupation of limited land resources and reduce interference with the land ecosystem, 3) the offshore PV power system may achieve higher energy output under certain conditions due to lower operational temperature [1].The offshore PV power systems show clear competitive advantages in improving energy efficiency, land resource conservation, and reducing ecological and water stress [2].However, at present, most studies focus on the performance optimization of onshore PV power systems.The studies on offshore PV power systems are very limited so far.In this paper, the influence of different factors on the power generation of offshore PV power systems is compared and analyzed via the regression model and path analysis, providing an important guideline for offshore PV power systems and their application.In this work, the studied factors involve the north-south distance of the PV array (D), the surface reflectivity (Refl), the tilt angle of the PV module (ȕ), the distance between the lowest edge of the PV module and the sea surface (H), and the swing angle (Į) under different sea states.It shows that Į has the most important effect on the performance of offshore PV power systems, and the influence degree of other factors from high to low is Refl, ȕ, D, and H.At the same time, the power generation of the thin-film PV power system is lower than that of the floating type.Although the thin-film PV power system is close to the sea surface (i.e. a better cooling effect on the PV module for power generation improvement), its tilt angle of 0° makes the received global solar irradiance much lower than that of the floating type [3].This work comprehensively analyzes the optimization IOP Publishing doi:10.1088/1742-6596/2771/1/012038 2 factors of offshore PV systems and provides an important guideline for future design and optimization of such PV systems.

Methodology
This work simulated the performance of offshore floating PV power stations at N38~40° via PVsyst, where two types of offshore PV systems which are floating and thin film ones are studied.The doubleglass double-sided PV modules (referred to as bi-PV modules) are used in floating PV power stations, and double-glass single-sided PV modules (referred to as mono-PV modules) are used in thin-film PV power stations.For convenience, the nominal power of both offshore PV power systems is set to 5 kWp.Meanwhile, the hydrodynamic characteristic analysis software AQWA was used to simulate the influence of waves on the performance of the two types of offshore PV systems under different sea conditions, where the irradiance received by the two types of offshore PV power stations was calculated.The above simulation results were statistically analyzed by SPSS (Statistical Package for the Social Sciences) software including regression and path analysis, so that the degree of influence of each factor on the system performance can be determined.

Results & discussion
Figure 1 presents a schematic of the floating PV power station in its front and top views.Five PV modules in each row are connected in series, making a total of 10 PV modules for such systems.D is the distance between the front and rear PV modules in the north and south direction.With different D values, the shadow loss caused by the front components to the back row is different, thus affecting the power generation.Table 1 shows the simulated linear loss percentage of D value to the direct radiation amount of PV modules under different tilt angles (ȕ), where D is set to 8 different values between 1 m and 3 m, ȕ for onshore PV system is set to 15°~50°, ȕ for offshore system is set to 5°~ 20°.The H value is set to 0.5 m, too small H will reduce the power generated on the surface of the bi-PV module, and too large H will reduce the stability of the floating body.According to Table 1, the linear loss of direct solar radiation decreases with the increase of D value, which is negatively correlated.When D is 3 m, the linear loss of the amount of direct solar radiation in the floating PV system is minimal, and the optimal state is reached.Since only geometric factors, including D, H, and ȕ are considered, the loss of the offshore PV system is always less than that of the onshore PV system at the same value of D. For bi-PV modules, the power generation is dominated by the front surface and could be further enhanced by the rear surface to a certain degree [4].The power generation from the rear surface is more significantly determined by the temperature (directly related to the H value) and reflectivity (Refl).In this work, a series of reflectivities in corresponding different back surface conditions such as water surface, concrete, and aluminum foil are selected to explore their effect on the performance of the bi-PV module.The Refl value of the water surface will change with the weather, where it becomes higher in cloudy (0.1) rather than sunny days (0.07) [5].
Table 2 shows the simulation of daily power generation of onshore/offshore PV power plants under different reflectivities, with geometric factors D=3 m, ȕę[6°, 48°], H=0.5 m.The results show that since offshore Refl is lower than land, its PV system also generates less power than terrestrial PV system.However, offshore power generation on sunny days is much higher than that on cloudy days (offshore reflectivity on sunny days is lower than on cloudy days) because of differences in solar irradiance under different weather conditions.Figures 2 a) and b) show the variation curves of annual power generation for offshore and onshore PV systems with different ȕ values respectively.Since the tilt angle of the offshore PV system using the bi-PV module is generally small (<20°), the variation trend of energy generation in the range of ȕ=2~30° is investigated.The power generation of offshore PV systems increases with ȕ.For the onshore PV system, the variation trend of power generation in the range of ȕ=28~48° shows that the power generation increases at first and then decreases with increasing ȕ, and a peak value of 7416 kWh is observed with an optimal tilt angle of 33° (ȕoptimal=33°).b) The annual power generation of onshore PV systems change trend with ȕ from 28 to 48°.
Since the specific heat capacity of water (i.e.4.2×10 3 J/kgႏ) is 3 times higher than that of air (i.e.1.4×10 3 J/kgႏ), there is usually a certain temperature difference between the sea surface and air above it under the same solar irradiation.In these locations (N38~40°) around the Bohai Sea of China, the temperature of seawater is typically lower than that of air in summer and autumn and higher in spring and winter.The performance of PV modules is closely related to the operational temperature and can be determined by the power temperature coefficient (i.e.-0.35%/ႏ at STC for both PV modules in this work) [6].To explore the impact of water surface cooling benefits on floating PV systems, a simple linear change trend of water surface temperature in different seasons (winter, spring, summer, and autumn) was constructed.The performance of the floating PV system was investigated by exploring the influence of different seasons and the height of the sea on the operating temperature of the module.The sea temperatures at different heights were collected through the database of the National Environmental Information Center (https://www.ncei.noaa.gov/).The boundary condition assumes that the air temperature is no longer affected by the sea once the altitude is greater than 4 m. Figure 3 shows the typical day temperatures on the sea at different altitudes and seasons.The temperature above the sea surface on typical days in spring and winter (black line) decreases with the increase of altitude, and the sea water warms up, which is not conducive to the power generation of PV modules.The temperature above the sea surface on typical summer and autumn days (red line) rises with the increase in height, and the seawater plays a cooling role, which is conducive to the power generation of PV modules.According to the provisions of T/CPIA 0017-2019 "Design Code for Offshore PV Power System" issued by the China PV Industry Association [7], the working surface of the floating equipment platform should be no less than 300 mm (i.e.H 300 mm) from the water surface and corresponding wave prevention measures should be applied.On the other hand, H should not be too high to minimize the impact of sea wave action on the floating body (i.e. the worst case is the floating body overturned).To explore the optimal H value for two PV power systems, the following parameters are used: (1) thin-film PV power plant (mono-PV module, ȕ=0°, D=1.2 m, H=0.5 m, Refl=7%) and (2) floating PV power plant (bi-PV module, ȕ=12°, D=2 m, H=0.5~4 m, Refl=7%).The selected mono/bi-PV module models have a nominal power of 500 Wp and a power temperature coefficient of -0.35%°C.
Figure 4 shows that the power generation of the floating PV system increases with H and tends to saturate.The annual power generation of the bi-PV module is higher than that of the mono-PV module.This is because with the increase of H, the increased rate of radiation received by the surface of the bi-PV module slows down, and the cooling effect of seawater is weakened.In summary, the power generation of the bi-PV module increased by ~10% and tends to stabilize with the increase of H value.However, too large H will result in the floating body being overturned as mentioned, hence the appropriate H value should be between 1 and 1.5 m.In general, in terms of the power generation improvement of floating PV power stations, the air ventilation effect of the bi-PV modules is better than the water cooling effect of mono-PV modules.It should be noted that the above discussions take place under optimal calm sea conditions [8].In real situations, the solar irradiation of floating PV power stations at sea to generate electricity will inevitably be affected by the floating body movement caused by waves.The motion state of the floating body under the action of sea waves is very complicated, which is related to the sea state and the characteristics of the floating body itself [9].This work uses wave theory to analyze the motion characteristics of floating bodies under different sea conditions.It is assumed that the water depth is infinite, the density of seawater (ȡ) is 1025 kg/m 3 , and the acceleration of gravity (g) is 9.8 N/kg.The cuboid model used in the simulation has a side length of 7 m, a height of 3 m, and a submerged depth of 1.7 m.As for the floating radius of the thin film, the height is 0.5 m and the submerged depth is 0.1 m (Figure 5).The radiant energy received by the floating body under different swing angles is calculated [10].Figure 6 shows that the ratio of total radiant energy to radiant energy Ș at the optimum installation tilt angle of the two floating power generation systems (a is the bi-PV module, b is the mono-PV module) decreases with the increase of the maximum swing angle Įmax.Therefore, improving the stability of the floating body in the wave (that is, reducing the angle of swing of the floating body) helps to increase the amount of solar radiation it receives.Under the same sea condition, the Ș value of the floating body is higher than that of the thin film one, and its system performance is less affected by the floating body's swing.The Ș value of the floating body is the highest under the condition of small waves, reaching 75.66%, that is, the photoelectric performance is the best at this time.The ratio of the solar radiation energy received by the floating body on the sea surface to the radiation energy received at the optimal tilt angle on the land is between 71.29% and 75.66%.The ratio Ș of the solar radiation energy received by the thin-film floating body on the sea surface to that on the land is between 40.99% and 75.52%.Ș values are strongly correlated with sea state conditions (i.e. the maximum swing angle Įmax).performance of PV modules was quantitatively analyzed, and the correlation between PV output power and influencing factors was determined by correlation analysis [11].The equation of correlation coefficient R [12] is: where n is the number of variables; xi is the independent variable (the influence factor x1~x5); yi is the dependent variable (the total annual electricity generation y).
Since there is only one dependent variable and many independent variables, the relationship between the PV output power of the dependent variable and the influencing factors of the independent variable is not all linear, so it is necessary to apply nonlinear regression analysis.The nonlinear regression model is mostly expressed as follows [13]: where y is the dependent variable; xi is the independent variable; f (xi, ș) is a nonlinear regression function; ș is the nonlinear parameter; and İ is the random variable error term.The nonlinear regression function equation is established.Considering the nonlinearity between variables, the objective nonlinear regression equation required in this work is constructed as follows: The R 2 of the nonlinear multiple regression equation is 0.925, indicating a high degree of fit.Through path analysis, the influence degree of different factors on the power generation performance of photovoltaic systems can be known.The reflectivity (Refl) has a great influence on the performance of bi-PV modules compared to other PV plants.Different values of ȕ and D have a significant impact on the solar irradiance received by PV modules.In contrast, the impact of H on performance is relatively limited.In summary, the influence degree of different factors on the performance of offshore floating PV power plants from high to low is Į, Refl, D, ȕ, and H.

Conclusions
In this paper, the factors affecting the performance of offshore PV system was firstly quantitatively analyzed, which include PV array spacing (D), PV module tilt angle (ȕ), the distance between the lowest side of the PV module and the sea surface (H), sea surface reflectivity (Refl), and the swing angle of the floating body affected by waves (Į).The correlation of these factors was explored by multiple regression method, and the results showed that the influence degree of the above factors on the PV system from high to low is Į, Refl, D, ȕ, and H.The swing angle of the floating body affected by waves has the greatest influence on the power generation of the PV system, reaching ~40%.Under harsh ocean conditions, the solar radiation received by the PV module is much less than that under the calm sea.The tilt angle of the PV module affects the system performance accounting for ~14%, and the tilt angle should be increased as much as possible under external conditions.At the same time, it should be noted that too large a tilt angle will make the center of gravity of the PV array unstable and cause overturn.The minimum impact on system performance is the distance between the lowest side of the PV module and the sea, and it is necessary to comprehensively consider the air and water cooling effects to enhance the overall performance of offshore PV systems at certain seasons and altitudes.In conclusion, the optimal offshore PV system should be designed as a floating PV power system using double-glass double-sided PV modules (or bi-PV module) and operating in a calm sea (or regular wave height less than 0.5 m) with parameters as below: Doptimal =2 m, ȕoptimal =12°, Hoptimal = 1~1.5 m.By using such settings, the power generation of the floating PV system is ~10% higher than that of the thin-film PV system (Į=0°).This work is a comprehensive study of offshore PV systems and provides important guidelines for future design and optimization work.

Figure 1 .
Figure 1.A schematic of a simulated PV array with two rows connected in series.

Figure 2 .
Figure 2. a) The annual power generation of offshore PV system change trend with ȕ from 2 to 30°.b)The annual power generation of onshore PV systems change trend with ȕ from 28 to 48°.

Figure 3 .
Figure 3.The environmental temperature on the sea surface at different altitudes and seasons (black for winter and spring, red for summer and autumn).

Figure 4 .
Figure 4. Monthly power generation at different heights for two categories of floating PV plants.

Figure 6 .
Figure 6.Two types of floating bodies (i.e.a) cuboid floats; b) thin-film cylinder float) and the ratio of the solar radiant energy received at the optimal tilt angle of the land Ș the law of change with the maximum swing.

Table 1 .
The linear loss percentage of direct solar irradiation at different D and ȕ values. 3

Table 2 .
Daily power generation simulations of onshore/offshore PV power plants at different Refl and ȕ.