summary
Researchers Wang Shijun and Ping Chang, from the Economic and Technological Research Institute of State Grid Gansu Electric Power Company and the Transmission and Transformation Engineering Co., Ltd. of State Grid Gansu Electric Power Company, published an article in the 8th issue of "Electrical Technology" in 2018, analyzing the factors affecting the output power of photovoltaic power plants. They focused on discussing the impact of the photovoltaic system's own electrical parameters and meteorological factors such as irradiance, weather conditions, temperature, and season on photovoltaic power generation.
By analyzing and compiling historical power generation data from a 30MW photovoltaic power plant, corresponding conclusions were drawn. These conclusions also provide a theoretical basis for selecting input samples for predicting power generation.
In recent years, with increased national subsidies and technological advancements in the photovoltaic industry, the secondary energy photovoltaic industry has developed significantly, and the cost of its components has continued to decline. However, solar energy is a secondary energy source with low stability and intermittent operation, resulting in poor stability and intermittent output power of photovoltaic power plants. Among these factors, the electrical properties of photovoltaic power plants and weather and climate conditions are crucial influencing photovoltaic power generation. Therefore, accurately understanding the impact of these two factors on predicting photovoltaic power generation, and thus enabling appropriate planning and scheduling, is of significant research importance.
There are many factors that affect photovoltaic power, generally falling into two categories: the electrical properties of the photovoltaic power station itself and the external meteorological environment, or external meteorological attributes. Irradiance, weather type, season, and temperature are generally considered external meteorological attributes. This article will analyze these two main factors based on historical data collected from a 30MW photovoltaic power station, illustrating their specific impact on photovoltaic output through a case study.
1. The impact of inherent electrical properties on photovoltaic output
The inherent properties of a photovoltaic power station involve numerous and complex factors, and some performance parameters are not used for reference in engineering. However, once a photovoltaic power station is connected to the grid and put into operation, the output power of the photovoltaic system exhibits a high correlation. This is due to the inherent configuration information of the photovoltaic panels, including their geographical location, installation height and angle, the efficiency of the cells in converting energy into electricity, the conversion efficiency achieved by the inverter, and the cleanliness of the cell surfaces.
These relevant information factors have been taken into account in historical output power. They change very little during the lifespan of the relevant photovoltaic cell components and are treated as constant quantities in short-term power forecasting. Therefore, historical output data samples of photovoltaics are usually used as sample inputs for the prediction model.
2. The impact of meteorological factors on photovoltaic output
2.1 The Influence of Solar Irradiance
This section uses historical data from a 30MW photovoltaic power station, selecting the 15th of each of the 12 months in 2017, to illustrate the relationship between its power generation, radiation, and peak sunshine hours. Specific data are shown in Table 1.
Table 1 Historical Data of Photovoltaic Power Plants
1) Relationship between power generation and radiation
The concept of solar irradiance is the amount of radiant energy received by a unit area of the Earth's surface from the vertical projection of sunlight per unit time. The output power of a photovoltaic panel increases as the absorbed solar irradiance increases. The relationship is shown in Figure 1. The trend graph clearly shows that the photovoltaic output power changes with the trend of solar irradiance, and the fluctuation trends of the two are similar.
Figure 1. Trends in power generation and radiation.
2) Relationship between power generation and peak sunshine hours
Let's first explain two definitions: Sunshine duration refers to the sum of all time periods during which the local direct solar irradiance exceeds 0.12 kW/m², and the unit is h. Peak sunshine duration refers to the sum of the cumulative irradiance received on the ground in a certain area throughout the day, divided by 1 kW/m², and the unit is hours (h).
The radiation collected by this photovoltaic power station is measured in MJ/m². According to the definition of peak sunshine duration, when deriving peak sunshine duration from irradiance, the unit needs to be converted to kW•h/m² to match the definition. The conversion formula is: 1J = 2.778 × 10⁻⁷ kW•h. For example, Table 1 shows that on January 15th, there were 10 hours of solar irradiance greater than 0.12 kW/m², so the sunshine duration was 10 hours. The total solar irradiance for the entire day was 18.93 MJ/m², which is first converted to 5.3 kW•h/m², resulting in a peak sunshine duration of 5.3 (kW•h/m²) / 1 (kW/m²) = 5.3 (h).
Based on the relationship between peak sunshine hours, sunshine duration, and power generation, its trend throughout the day can be plotted, as shown in Figure 2.
Figure 2. Trends in power generation, sunshine hours, and peak sunshine hours.
As can be seen from the trend chart, the trend of power generation is very close to that of peak sunshine hours, showing a strong correlation, while the relationship with sunshine hours is not very close.
This section directly analyzes the impact of radiation, and then indirectly analyzes its impact through "peak sunshine duration," concluding that irradiance intensity is reflected in power generation, and the two have a close linear relationship. It also demonstrates that peak sunshine duration can be used as one of the reference standards for predicting photovoltaic output power.
2.2 Impact of Weather Types
Due to varying weather conditions, the intensity of irradiance, humidity, and temperature received by photovoltaic panels can differ significantly, leading to substantial variations in photovoltaic output. This section presents power generation data from July 18th to 25th, as shown in Figure 3.
The results show that power generation is highest on sunny days and lower on cloudy/rainy days. This indicates a significant difference in photovoltaic output under different weather conditions, while under the same weather conditions, such as sunny days, the total power generation within a day is essentially the same.
Figure 3. Power generation data under different weather conditions
As the analysis of the impact of weather types in this section shows, in order to maximize the short-term forecasting level of photovoltaic power generation, it is necessary to classify and organize the historical power generation of photovoltaic power plants according to the characteristics of different weather conditions, and use them as input samples for power generation forecasting in order to achieve good forecasting results.
2.3 The Influence of Seasons
The effective sunshine duration and solar incidence angle of grid-connected photovoltaic power stations vary continuously with different weather conditions throughout the seasons, causing fluctuations in output power. This section illustrates this with historical data measured from photovoltaic power stations, as shown in Table 2. This article selects the hourly instantaneous power output on four effective periods: February 20th (spring), May 9th (summer), September 13th (autumn), and December 18th (winter). All four days were sunny.
Table 2 Statistical data of a photovoltaic power station in different seasons
To visually assess the impact of seasons on power output, Figure 4 was created based on the data in Table 2. The figure illustrates the influence of varying sunshine duration and irradiance across the four seasons on photovoltaic power output. The trends of the four curves also indicate that summer and autumn have the longest effective output time, while winter has the shortest; furthermore, spring and autumn have significantly higher maximum output power compared to winter.
Figure 4 Seasonal power data graph
Therefore, as this section shows, under the same weather conditions in different seasons, the variation patterns of photovoltaic power generation curves are basically similar, but the magnitude of the power output and the length of the effective power generation period differ. Modularizing the power generation data according to the four seasons and dividing and conquering photovoltaic forecasting can help reduce forecast complexity and improve accuracy.
2.4 Effect of Temperature
As ambient temperature rises, the output current I of photovoltaic modules will increase to some extent, but the conversion efficiency of the photovoltaic power generation system will decrease significantly, resulting in a reduction in the operating current V. However, the decrease in V is relatively more pronounced, so the total output power P will gradually decrease as the temperature rises. This article will use the performance parameters of the monocrystalline module DM60-285 as an example to illustrate this. The electrical performance of the photovoltaic module is shown in Table 3.
Table 3 Electrical properties of DM60-285 photovoltaic modules
The "STC" in the table refers to the condition where the photovoltaic module's peak output power reaches 285W when the ambient conditions are irradiance of 1kW/m2, cell temperature of 25℃, and atmospheric mass of 1.5. The temperature coefficient of the photovoltaic module is shown in Table 4.
Table 4 Temperature coefficient of photovoltaic modules
As shown in Table 4, the "maximum power temperature coefficient" is -0.42%/℃, meaning that for every 1℃ increase in the ambient temperature of the photovoltaic panel's operating environment, the power generation decreases by 0.42%. Assuming the ambient temperature of the photovoltaic cell is 55℃, the percentage decrease in the cell's output power relative to the peak power is 0.42% × (55℃ - 25℃) = 12.6%.
Therefore, a certain degree of temperature increase will reduce the power output of the solar panels. In actual photovoltaic systems, in addition to photovoltaic cells, other components, including inverters, are also affected by temperature, thus impacting photovoltaic output. Therefore, temperature is one of the factors that must be considered in photovoltaic forecasting.
in conclusion
This paper examines photovoltaic (PV) power generation from two perspectives: ① it analyzes the power characteristics of PV systems and organizes and plots the data; ② it analyzes the impact of meteorological factors, namely solar irradiance, weather conditions, season, and temperature, and draws the following conclusions and recommendations.
1) The meteorological factors affecting photovoltaic (PV) power generation are diverse. For example, solar irradiance, weather type, season, and ambient temperature all have positive or negative effects on PV power output. Therefore, it is recommended to combine historical power generation data with meteorological factors to predict PV output power.
This involves selecting dates with similar weather conditions to the "date to be predicted," using their historical power generation load as input data, and then making predictions for the "date to be predicted." The method for selecting similar dates is as follows: based on the environmental type of the date to be predicted, select dates with the same weather type within the same period of the previous three years as similar dates.
2) In the climate environment where this photovoltaic power plant is located, summers are very hot and winters are the coldest. If temperature is considered alone, winter should have the highest power generation. However, spring and autumn have dry climates, long effective peak sunshine hours, and abundant sunshine, so the irradiance is significantly better in both time and intensity than in winter. Therefore, spring and autumn are the best seasons for photovoltaic power plants to generate electricity. In summary, the best time for power generation is at noon on sunny days in spring and autumn. It is recommended that photovoltaic power plants operate during spring and autumn and adjust maintenance periods to winter as much as possible.
3) Even within the same season, weather conditions vary from moment to moment and are difficult to quantify, causing fluctuations in the power output of photovoltaic modules due to changes in actual weather conditions, thus increasing the difficulty of forecasting. Therefore, to accurately predict weather conditions, it is recommended to shorten the forecast time scale, setting the forecast time to 0.25 to 4 hours, depending on the actual situation.
