Typhoon Disaster Research on Solar Photovoltaic Equipment in Taiwan

doi:10.21203/rs.3.rs-3301388/v1

In April 2017, the government of Taiwan passed the amendment to "Energy Development Guidelines," officially beginning the race of domestic energy transformation. The purpose was not only to reduce the usage of fossil fuels but also to diversify the risk of energy shortage. Among various renewable energy applications, the solar photovoltaic industry played a crucial role in the market. However, Taiwan's location is prone to typhoon disasters and poses the most significant threat to the solar photovoltaic industry. With an average of four typhoons hitting the island each year, events like Typhoon Soudelor in 2015 and Typhoon Meranti in 2016 brought power winds, causing severe damage to solar panels across Taiwan and increasing investment risks in the industry. Therefore, effectively adapting to the environment with frequent typhoons in Taiwan will be an important goal for disaster prevention in the solar photovoltaic industry. This study conducted a questionnaire survey during three typhoon events: Typhoon Soudelor, Typhoon Nepartak, and Typhoon Meranti, to analyze the disaster situation of solar photovoltaic equipment during typhoon events. The results indicated that the actual loss rates for solar photovoltaic equipment during Typhoon Soudelor, Typhoon Nepartak, and Typhoon Meranti were 5.6%, 2.3%, and 1.4%, respectively. Additionally, the losses of solar photovoltaic equipment showed a significant positive correlation with wind speed. The loss rate of the equipment may be three times more when the wind speed doubles.

Solar photovoltaic panels

Typhoon losses

Questionnaire survey

Disaster probability

Due to limited energy production capacity and environmental carrying restrictions, most of Taiwan's energy supply has relied on imports. In recent years, with the rise of climate change issues prompting the shift in domestic energy policies, Taiwan has gradually focused on expanding renewable energy options. The "Sustainable Energy Policy Guidelines" was officially launched in 2008, whereby the regulation advocates sustainable energy development with an aim for a win-win situation that balances energy production, economic development, and environmental protection by setting the target of alleviating renewable energy to account for more than 8% of the power generation system by 2025. Therefore, the central and local governments are requested to actively implement various projects to achieve a low-carbon and sustainable economy.

According to Bureau of Energy (2022), both solar energy’s capacity and its proportion of renewable energy sector has been growing in Taiwan (Fig. 1). The surge of solar energy generation and supporting government policies undoubtfully set solar as the key driver forces for Taiwan's energy transformation. However, Taiwan's location is prone to natural disasters. It requires a much more sophisticated policy in response to the direct damage the solar industry might encounter from these natural disasters. According to past research (Li et al., 2016), among all natural disasters in Taiwan from 1958 to 2012, typhoon disasters accounted for the highest occurrence rate at 67%, followed by flooding at 21%, earthquakes at 8%, and other disasters at 4%. Among these disasters, typhoons are ranked the top threat to solar photovoltaic facilities in-country. Typhoon Soudelor in 2015 and Typhoon Meranti in 2016 came with powerful winds exceeding 17 on the Beaufort scale, resulting in significant losses for the domestic solar photovoltaic industry.

As solar photovoltaic panels have only become an accessible energy-generating tool in the last decades, there are relatively few research cases on wind-induced damage to solar panels, while many only discuss the general causes of solar panel damage. Official statistics from Japan covering the period from 2012 to 2017 (Japan Ministry of Economy, 2019) showed that solar panel damage incidents increased from 1.1% in 2013 to 7.1% in 2017, with natural disasters being its main contributing factor. Disasters contributing to solar panel damage include flooding, landslides, strong winds, and typhoon storm surges. The disaster scenarios involve incidents such as panels being struck by flying debris due to strong winds, entire lines of solar panels being blown away due to damaged facilities, and circuit boards being damaged or short-circuited due to water immersion.

On the domestic level, the earliest on-record investigation of solar panel damage was done for Typhoon Haitang in 2005 (Li et al., 2005). The loss rate for solar-powered water heating systems in counties and cities across Taiwan ranged from 0.03% to 4.45%, with more losses observed in the eastern counties. In the same year, Typhoon Longwang also hit. Shaw et al. (2007) conducted a face-to-face survey right after the event using a sampling method in Hualien County to calculate wind-induced household losses. The survey results showed that Typhoon Longwang, which brought strong winds up to 17 on the Beaufort scale, caused property losses for approximately 42.3% of the population in the county. The most severe losses tallied were in rooftop additions, main roof structures, and signboards.

Similarly, when Typhoon Soudelor landed in Taiwan in August 2015, the National Science and Technology Center for Disaster Prevention documented the most severe wind damage in Taipei and New Taipei City in the past decade. Roof damages (including rooftop additions) experience the most significant damage, accounting for 40.9% of the total loss, followed by building components (balconies, windows, etc.). These common damaged locations during typhoons are the major installation sites for solar photovoltaic facilities, indicating a relatively high probability of disaster risk when a wind-induced disaster arrives.

Few studies have shown the relationship between solar panel damage rate and wind speed, directly or indirectly. Wind speed directly acts as the primary force contributing to the damage. At the same time, structural fragility or insufficient structural support may also indirectly contribute to the panel's damage when wind speed exceeds a certain level (Fu, 2020). All the literature has demonstrated the major installation sites (rooftops and balconies) being the most commonly damaged part of the housing structure during typhoons in Taiwan, indicating a relatively high probability of solar panel damage when a wind-induced disaster arrives. On top of that, according to Ministry of Science and Technology (2022), under the climate change scenario, the number of typhoons hitting Taiwan will decrease while the intensity will increase. In recent years, however, Taiwan has experienced fewer typhoon events than in the past decade (Fig. 2); the rapid development of solar photovoltaics might neglect the potential danger of wind disasters in the future. Evaluating the threat of typhoon disasters to domestic solar photovoltaic facilities and the potential losses they may cause therefore becomes a critical study that may provide evidence-based decision-making for future disaster reduction strategies for the solar industry.

2.1 The survey data of solar photovoltaic facilities disasters

In 2021, our team conducted surveys in areas of Taiwan where solar panels have experienced rapid development, including:

Town and townships with government-approved special solar panel development zones
Town and townships with the most solar panel damage records in past typhoon events
Town and townships with the highest capacity for solar-generated energy

This investigation aimed to build analytical model to assess the relationship between solar panel loss and wind speed. The collected data will then serve as a reference for formulating disaster prevention strategies.

The survey was conducted at the household level, primarily focusing on the condition of solar panel equipment that reported damages during and after Typhoon Soudelor (2015), Typhoon Nepartak (2016), and Typhoon Meranti (2016) event. A total of 215 valid questionnaires were collected, and the distribution of survey samples based on towns and townships is illustrated in Fig. 3.

2.2 Historical Meteorological Data

In the past, the most significant challenge in researching solar photovoltaic wind-induced losses was the need for more sufficient meteorological data on finer scales to match meteorological conditions based on the location of solar photovoltaic installations. This study utilizes publicly available Taiwan ReAnalysis Downscaling data (TReAD) produced by Taiwan Climate Change Projection Information and Adaptation Knowledge Platform (TCCIP) project.

TReAD data is based on the state-of-the-art ERA5 (Hersbach et al. 2020) historical reanalysis data and further downscaled to a high-resolution dataset for Taiwan from 1980 to 2018. The spatial resolution of this data is 2 x 2 km, with a total of 25,000 grid points across Taiwan (shown in Fig. 4). This dataset helps compensate for the uneven distribution of weather stations. It also provides a diverse range of meteorological variables, making it suitable for conducting a comprehensive analysis of the relationship between the damage rate of solar photovoltaic equipment and wind speeds during typhoon events.

In this study, the wind speed data from TReAD was used to conduct a comprehensive analysis of Typhoon Soudelor, Typhoon Nepartak, and Typhoon Meranti, as shown in Fig. 5 (blue dots represent the locations of solar photovoltaic sample surveys). From Fig. 5(a), it can be observed that Typhoon Soudelor had the highest wind speeds among the three typhoon events, with most of the surveyed solar photovoltaic installations exposed to areas with wind speeds exceeding 15m/s. Fig. 5(b) shows the wind speed spatial distribution across Taiwan during Typhoon Nepartak, with the highest wind speeds occurred in the southeastern region where solar photovoltaic installations are relatively scarce. In the southwestern region there were still some area experiencing exceeding 15m/s. Finally, Fig. 5(c) displays during Typhoon Meranti, higher wind speeds primarily occurred in the southern region, and some of the surveyed sample locations were also located in areas with higher wind speeds.

To further evaluate how the solar photovoltaic equipment may be affected under different wind speeds, this study presents the data of each sample's experienced wind speed and the actual damage or loss situation in Fig. 6. While some samples may experience damage in all three events, the three typhoon events occurred at different times can thus be considered as three independent events. Figure 4 shows all samples divided into three separate events based on the occurrence time of the three typhoons. By matching with the maximum wind speed values from Figure 3, the maximum wind speed for each sample in each event was calculated. Simultaneously, the samples that experienced loss during each event were also marked for comparison, allowing for a comprehensive assessment of the disaster situation for each event.

In Fig. 4, the x-axis represents the sample number of the sample in different typhoon events, and the y-axis shows the maximum wind speed experienced (from TReAD) by all samples in those events. It can be observed that Typhoon Soudelor had significantly higher wind speeds compared to the other two typhoon events. Consequently, Typhoon Soudelor also resulted in more affected samples, and the wind speeds of the disaster samples mostly ranged from 15 m/s to 40 m/s. The overall disaster rate for solar photovoltaic equipment during Typhoon Soudelor was 5.6%. On the other hand, the wind speeds experienced by the affected samples during Typhoon Nepartak and Typhoon Meranti were relatively lower, ranging from approximately 10 m/s to 15 m/s. The overall loss rates for solar photovoltaic equipment during Typhoon Nepartak and Typhoon Meranti were 2.3% and 1.4% respectively.

This study further classified all samples into evenly spaced wind speed intervals and resorted them based on different wind speed levels (as shown in Table 1). From the results in Table 1, it can be observed that when the wind speed of the samples is below 10 m/s, the disaster loss rate is only 1.8%. As the typhoon wind speed reaches 15 m/s and 20 m/s, the loss rates for the samples increase to 2.2% and 2.7%, respectively. However, when the wind speed exceeds 20 m/s, the loss rate rises significantly to 5.5%.

This study employed binomial statistical methods to perform regression analysis on wind speed and loss rates, obtaining the relationship curve between the maximum wind speed and the loss rate for solar photovoltaic installations, as shown in Formula (1):

Y = 0.0013x²+0.1628x (1)

where Y represents the loss rate (%) of solar photovoltaic equipment during typhoon events, and x represents the maximum wind speed (m/s). The regression model has a coefficient of determination (R2) of 0.89, indicating a good explanatory power.

Table 1 Loss Rate of Statistical Samples at Different Wind Speeds

Wind speed (m/s)	Total Sample (N=645)	Loss Sample (N=19)	Loss Rate (%)
0.0-5.0	0	0	0.0
5.0-10.0	56	1	1.8
10.0-15.0	231	5	2.2
15.0-20.0	226	6	2.7
20.0-25.0	91	5	5.5
>25	41	2	4.9

From Fig. 7, it can be observed that the surveyed samples are roughly normally distributed, and the loss rate is clearly proportional to the wind speed. Under low wind speed conditions, the number of affected samples is relatively small, while the loss rate is also low. In contrast, with wind speeds exceeding 20m/s, the number of surveyed samples decreases, but the loss rate increases significantly. In the future, with the impact of climate change, Taiwan may face more severe typhoon events, the resulting losses may be even greater than current observation. Therefore, based on the classification of the Beaufort scale and using the regression formula (1), this study can produce a general estimate on the disaster risks of solar equipment under different wind speed thresholds, offering a scientific basis for emergency responses and protective applications, as shown in Table 2.

Table 2 Estimated Loss Rate of Photovoltaic Equipment Classified by Beaufort Scale Levels

Beaufort Scale		Wind Speed (m/s)	Loss Rate (%)
0	---	0.0 ~0.2	0.0 ~0.0
1		0.3 ~1.5	0.0 ~0.2
2		1.6 ~3.3	0.3 ~0.6
3		3.4 ~5.4	0.6 ~0.9
4		5.5 ~7.9	0.9 ~1.4
5		8.0 ~10.7	1.4 ~1.9
6		10.8 ~13.8	1.9 ~2.5
7		13.9 ~17.1	2.5 ~3.2
8	Mild Typhoon	17.2 ~20.7	3.2 ~3.9
9		20.8 ~24.4	3.9 ~4.7
10		24.5 ~28.4	4.8 ~5.7
11		28.5 ~32.6	5.7 ~6.7
12	Moderate Typhoon	32.7 ~36.9	6.7 ~7.8
13		37.0 ~41.4	7.8 ~9.0
14		41.0 ~46.1	8.9 ~10.3
15		46.2 ~50.9	10.3 ~11.7
16	Strong Typhoon	51.0 ~56.0	11.7 ~13.2
17	Strong Typhoon	56.1 ~61.2	13.2 ~14.8

Based on the estimates from Table 2, when the typhoon wind speed reaches the category of Mild Typhoon, the disaster rate for solar photovoltaic equipment ranges from 3.2% to 6.7%. When the wind speed reaches the category of Moderate Typhoon, the disaster rate is between 6.7% and 11.7%. When the wind speed exceeds the threshold of Strong Typhoon, according to the formula for estimation,, the loss rate for solar photovoltaic equipment can reach up to 14.8%. If the wind speed exceeds level 17, it will cause even more severe damage to solar photovoltaic equipment.

With the lack of record data due to the fewer direct impact of typhoons in Taiwan, it has been challenging to understand the actual impact of typhoon events on solar photovoltaic equipment. Therefore, this study uses different typhoon categories as examples to estimate the potential losses and associated economic costs of installation when typhoons directly affect the main solar equipment sites.

Based on the Energy Statistical Monthly Report from Bureau of Energy (2022), the total installed capacity of solar photovoltaic systems in Taiwan reached 9,724 MW by the end of 2022. According to the loss rates calculated from the minimum wind speed thresholds for each typhoon category in Table 2, the loss of installed capacity caused by a Mild Typhoon would be 310 MW, a Moderate Typhoon would cause a loss of 653 MW, and a Strong Typhoon would cause a loss of 1,110 MW. The detailed calculation results are shown in Table 3. It is essential to emphasize that the total capacity loss caused by a Strong Typhoon would be equivalent to half of the newly established solar panels in 2022, indicating a severe impact on the industry.

In addition, by using a construction cost of 1.13 million USD/MW (Hsiao et al. 2016) to estimate the potential losses, the result ranges from 351 million USD to 1,258 million USD for a single typhoon event. If considering the time and cost required for the recovery of solar photovoltaic equipment, the actual impact could be even more severe.

Table 3 Losses of Photovoltaic (PV) Equipment Caused by Different Levels of Typhoon Events

Typhoon Level	Beaufort Scale	Wind Speed Threshold (m/s)	Loss Ratio (%)	Installed Capacity Loss (MW)	Loss (million USD)
Mild	8~11	17.2	3.2	310	351
Moderate	12~15	32.7	6.7	653	740
Strong	>16	50.1	11.4	1110	1258

Solar generated energy is a booming industry, but there have been very few typhoons hitting Taiwan in recent years, and relatively few losses to photovoltaic panels. With very few studies on the reported losses on solar installations, Taiwan of the possible impact of typhoons on the solar energy industry, and the research on it is also lacking, let alone explore the potential threat of typhoons to the solar photovoltaic industry under future climate change.

This study shows that the typhoon wind speed is positively correlated with the loss of photovoltaic panels, and the damage rate caused by the typhoon ranges from 1.8% to 4.9% according to loss data reported during Typhoon Soudelor (2015), Nepartak (2016), and Meranti (2016). The converted loss ranges from mild typhoon 351 million USD to strong typhoon 1258 million USD. Under the warming scenario, the intensity of typhoons hitting Taiwan will increase, and if the current solar photovoltaic panels set-up remains unchanged, it will experience greater losses in the future.

Solar energy is yet in its early phase of development in Taiwan. It is undoubtedly that more data is needed to design a more complex model to explain the correlation between the wind speed and the loss rate of solar installations. As time progresses, we believe this study can include more factors such as construction materials and height of installation sites, so to reduce the bias for selection of indicators. But this study does serve as a starting point to consider possible adaptative and protective measures in the of current solar energy development plan.

Funding This work was supported by National Science and Technology Center for Disaster Reduction.

Conflict of interest There is no declaration of interest by authors.

Author Contribution The first draft of the manuscript was written by Hsin-Chi Li and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.

Bureau of Energy, Ministry of Economic Affairs, R.O.C. (2022) Energy Statistical Monthly Reports. (In Chinese)Fu ZL (2020) Wind Characteristics and Comparison of Solar Photovoltaic Panel Support Structures in Various Countries. Construction Information, Issue 446. (In Chinese)
Hersbach H, Bell B, Berrisford P, Hirahara S, Horányi A, Muñoz-Sabater J, Nicolas J, Peubey C, Radu R, Schepers D, Simmons A, Soci C, Abdalla S, Abellan X, Balsamo G, Bechtold P, Biavati G, Bidlot J, Bonavita M, Thépaut JN (2020) The ERA5 global reanalysis. Quarterly Journal of the Royal Meteorological Society. 146(730), 1999–2049. https://doi.org/10.1002/QJ.3803
Shaw D, Chen SH, Chiu CT, Lin SY, Li H.C, Kuo YL, Mai WJF, Chen YY, Huang DS, Yang HH, Fan YK (2007) Investigation of the Social, Psychological, and Economic Impacts of Natural Disasters: Floods and Typhoon Longwang. Taipei: National Science and Technology Center for Disaster Reduction (NCDR 95-T12). (In Chinese)
Hsiao KH, Que DH, Hong JY (2016). Cost Analysis of Renewable Energy - Future Solar Photovoltaic and Wind Power Costs Will Continue to Decline. Industrial Technology Research Institute. Green Energy and Environment Research Institute. (In Chinese)
Japan Ministry of Economy, Trade and Industry, Electricity Safety Division (2019) Current Status and Challenges of Electrical Safety Associated with the Expansion of Renewable Energy Power Generation Equipment (In Japanese) https://www.meti.go.jp/shingikai/sankoshin/hoan_shohi/denryoku_anzen/newenergy_hatsuden_wg/018.html
Li HC, Deng CZ, Chang CL, Chen YC (2016) The Loss Surveying and Analysis of Typhoon SOUDELOR. International Journal of digital humanities and creative innovation management (IJDHCIM). 4(4):1–10
Li QA, Zhang KQ, Li CS, Zhong GM (2005) Investigation and Analysis of Typhoon-induced Damage to Solar Water Heating Systems: A Case Study of Typhoon Haitang. Journal of Solar Energy and New Energy. 10(2): 2–5 (In Chinese) https://km.twenergy.org.tw/Document/reference_more?id=142
Ministry of Science and Technology Taiwan (2022) Scientific Highlights of IPCC 2022: Climate Change 2022: Impacts, Adaptation, and Vulnerability Contribution of Working Group II to the Sixth Assessment Report of the Integovernmental Panel on Climate Change and Update Report on Assessment and Analysis of Climate Change Impacts in Taiwan. (In Chinese)
Taiwan Climate Change Projection Information and Adaptation Knowledge Platform (TCCIP). https://tccip.ncdr.nat.gov.tw/
Tung CP, Lee PF, Lin HZ, Lee MH, Lu HS, Su HZ, Chang JZ, Liang SL, Hsu TW, Lee HQ (2017). Climate Change in Taiwan 2017: Scientific Report – Impacts and Adaptation. National Science and Technology Council. Pp17 (In Chinese)

Typhoon Disaster Research on Solar Photovoltaic Equipment in Taiwan

Status:

Version 1

Abstract

Figures

1 Introduction

2 Data

3 Results

4 Discussion

5 Conclusion

Declarations

References

Status:

Version 1