<abstract xmlns="http://www.w3.org/1999/xhtml">

<p>In light of the increasing frequency of extreme weather events—including persistent low temperatures, rain, snow, freezing conditions, and haze—globally, the vulnerability of photovoltaic (PV) power projects to severe weather has escalated. This study presents a comprehensive analysis of the current construction practices and associated risks of PV power projects. Subsequently, it introduces two innovative risk prediction models: the gray risk prediction model and the CEMD-LSTM risk prediction model, which is enhanced by an improved k-shape clustering algorithm. These models are specifically designed for assessing the risks of new energy PV power projects under extreme weather conditions. To validate the effectiveness of these models, simulation tests are conducted, offering insights into the risk dynamics of new energy PV power projects in adverse meteorological scenarios. The assessment results show that the absolute error of the prediction results of the gray prediction model is large. The absolute value of the absolute error reaches a maximum of 8.0×104 W. The fluctuation of the prediction value is huge, so it is not possible to accurately carry out the risk prediction. The values of MAPE and MSE of the CEMD-LSTM model are 3.26% and 0.21, respectively, and the rate of identification of the risk level of the CEMD-LSTM model is 98.25%, which is higher than 93.48% of the LSTM model. The CEMDLSTM model is capable of predicting the risk of PV power projects with accuracy, and the recognition accuracy of the risk level is superior to that of the LSTM model. This study provides lessons and references for risk prediction of PV power projects in extreme weather conditions.</p>
</abstract>

In light of the increasing frequency of extreme weather events—including persistent low temperatures, rain, snow, freezing conditions, and haze—globally, the vulnerability of photovoltaic (PV) power projects to severe weather has escalated. This study presents a comprehensive analysis of the current construction practices and associated risks of PV power projects. Subsequently, it introduces two innovative risk prediction models: the gray risk prediction model and the CEMD-LSTM risk prediction model, which is enhanced by an improved k-shape clustering algorithm. These models are specifically designed for assessing the risks of new energy PV power projects under extreme weather conditions. To validate the effectiveness of these models, simulation tests are conducted, offering insights into the risk dynamics of new energy PV power projects in adverse meteorological scenarios. The assessment results show that the absolute error of the prediction results of the gray prediction model is large. The absolute value of the absolute error reaches a maximum of 8.0×104 W. The fluctuation of the prediction value is huge, so it is not possible to accurately carry out the risk prediction. The values of MAPE and MSE of the CEMD-LSTM model are 3.26% and 0.21, respectively, and the rate of identification of the risk level of the CEMD-LSTM model is 98.25%, which is higher than 93.48% of the LSTM model. The CEMDLSTM model is capable of predicting the risk of PV power projects with accuracy, and the recognition accuracy of the risk level is superior to that of the LSTM model. This study provides lessons and references for risk prediction of PV power projects in extreme weather conditions.

Research on Risk Prediction of New Energy Photovoltaic Power Projects Oriented to Extreme Weather Conditions

Beijing Energy International Holding Co., Ltd.

Applied Mathematics and Nonlinear Sciences

This work is licensed under the Creative Commons Attribution 4.0 International License.

In light of the increasing frequency of extreme weather events—including persistent low temperatures, rain, snow, freezing conditions, and haze—globally, the...

Research on Risk Prediction of New Energy Photovoltaic Power Projects Oriented to Extreme Weather Conditions

Published Online: Jul 02, 2024

Page range: -

Received: Mar 07, 2024

Accepted: Jun 01, 2024

DOI: https://doi.org/10.2478/amns-2024-1571

Keywords
Photovoltaic power, Risk prediction, Extreme weather, Gray model, CEMD-LSTM model

© 2024 Yanjun Qiao et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Research on Risk Prediction of New Energy Photovoltaic Power Projects Oriented to Extreme Weather Conditions

Published Online: Jul 02, 2024

Page range: -

Received: Mar 07, 2024

Accepted: Jun 01, 2024

DOI: https://doi.org/10.2478/amns-2024-1571

KeywordsPhotovoltaic power, Risk prediction, Extreme weather, Gray model, CEMD-LSTM model

© 2024 Yanjun Qiao et al., published by Sciendo

This work is licensed under the Creative Commons Attribution 4.0 International License.

Keywords
Photovoltaic power, Risk prediction, Extreme weather, Gray model, CEMD-LSTM model