TWI637332B - Power generation prediction system and method for solar power plant - Google Patents

Abstract

A power generation prediction system and method for a solar power plant, comprising the steps of: looking at a power generation of a second power plant around a first power plant; when the power generation of the second power plant decreases in the first time interval, looking for a third The power generation of the power plant decreases; when the power generation of the third power plant decreases in the second time interval, the power generation loss ratio is calculated according to the power generation of the second power plant and the third power plant; and the second power plant and the third power plant are affected by the cloud layer. Calculating the moving speed of the cloud layer; estimating the time when the cloud layer moves to the first power plant according to the moving speed of the cloud layer; and calculating the possible power generation amount of the first power plant according to the power loss amount of the power generation amount and the power generation amount of the first power plant historical time.

Description

Power generation prediction system and method for solar power plant

This creation is about a power generation prediction system and method, and in particular, a power generation prediction system and method for a solar power plant.

With the advancement of science and technology, energy consumption has increased dramatically, and fossil energy has become increasingly exhausted. Therefore, finding alternative energy sources is an urgent task. Solar energy has inexhaustible and clean and pollution-free characteristics. Solar power has the advantages of simple structure, small size, light weight, easy installation, convenient transportation, easy construction, convenient maintenance, cleanliness and safety. Therefore, solar energy has become a renewable energy development and utilization. One of the main options.

Solar power generation has the disadvantage of unstable power generation. Therefore, if the output power of solar power plants can be predicted, it will help the overall arrangement of power system dispatch, which is conducive to the coordination of conventional energy and solar power generation. The power company can adjust and adjust instantly. Scheduling and rationally arranging the grid operation mode can effectively reduce the impact of solar energy on the original grid when it is integrated into the grid, improve the safety and stability of the grid operation, and reduce the standby and operating costs of the power system to fully utilize solar energy resources. Get more benefits.

The output power of the solar power system will be affected by solar radiation, weather factors and the material of the solar panel itself. The power generation and output power are random, fluctuating and uncontrollable. When the weather is abrupt, the cloud clusters are rapidly changing and moving, the situation in which power generation is affected is particularly obvious. When solar power is integrated into the grid, it will affect the original power quality and safety of the grid. Therefore, it is necessary to monitor the dynamic changes of the cloud clusters in the sky. Currently used for sky cloud monitoring, it can use equipment such as full sky imager, and the full sky imager can 360 degrees all day in the local visible range. It is used for real-time data collection and analysis, and it is often used for astronomical observation, weather forecasting or videography. In recent years, research units have used this type of equipment to analyze the dynamic characteristics of sky clouds to predict the power of solar power generation. The all-sky imager is mainly composed of a mirror, which images the sky within a 360-degree range on the surface of the mirror, and then uses a lens with a photosensitive coupling element (CCD) to capture the image data, and then acquires the image using the image software. Hemispherical photos or videos are expanded and corresponding target recognition and positioning processes are performed.

Although the full sky imager can collect and process the sky cloud image in a wide range of angles of view, the shape and position distortion are present in the acquired image, even if the image software can be used to image the distortion image to a certain extent. The reply, but there is still a big difference between the characteristics of the actual object and the object image accuracy and accuracy, the farther away from the imaging center, the more serious the problem of image deformation.

Therefore, it is necessary to propose a solar power plant power generation prediction method that can predict the moving path of the cloud layer and effectively evaluate the degree of influence of the solar power plant.

The purpose of this creation is to create a power generation prediction method for a solar power plant. It is predicted by the power generation prediction method whether the solar power plant will be affected by the cloud layer, and the magnitude of the power generation affected by it is estimated, and the power generation capacity of the solar power plant is further predicted.

According to the above object, a power generation prediction method for a solar power plant is proposed, which comprises the steps of: looking at a first power plant to see if a second power plant in the plurality of power plants around the first power plant generates a power generation in a first time interval. The quantity decreases; when the power generation of the second power plant decreases in the first time interval, the second power plant and the first power plant are connected to form an imaginary straight line to find whether there is a region within a region of the imaginary straight line. a third power plant, and the third power plant has a decrease in the power generation amount in a second time interval, wherein the second time interval is earlier than the first time interval; calculating a power generation of the second power plant in the first time interval a ratio of the amount of loss reduction and a ratio of the amount of power generation loss of the third power plant in the second time interval; according to a relative distance between the second power plant and the third power plant, the first time interval and the second time interval Calculating a moving speed of the cloud layer moving from the third power plant to the second power plant according to the time difference; and calculating, according to the moving speed of the cloud layer, the moving of the cloud layer to a third time interval of the first power plant, wherein the third time The interval is a future time interval, and the third time interval is later than the first time interval and the second time interval; and the first power plant is estimated according to the power generation amount falling loss ratio and a reference power generation amount of the first power plant The amount of power generated in the third time interval.

Another purpose of this creation is to create a power generation prediction system for a solar power plant, through which the power generation prediction system can be used to calculate whether the power plant will be affected by cloud cover and cause a decrease in power generation, and further predict the power generation of the power plant.

According to the above objective, a power generation prediction system for a solar power plant is provided, comprising: a power generation falling search device, connected to a solar energy management system, and searching for a surrounding area in a database of the solar power management system Among the plurality of power plants, whether there is a second power plant that has a decrease in power generation in a first time interval, and whether there is a third power plant that has a power generation decrease in a second time interval, wherein the second time interval is earlier than the In a first time interval, the first power plant and the second power plant are connected to an imaginary line, and the third power plant is located in a region extending outward from the imaginary line; a power generation falling calculation device is connected to the power generation amount Searching for a device to calculate a ratio of power generation loss of the second power plant to the third power plant; a moving speed calculating device connecting the power generation amount lowering searching device and the solar energy management system, calculating a relative distance between the second power plant and the third power plant, and calculating a time difference between the first time interval and the second time interval a moving speed of a cloud layer; a moving time computing device connecting the moving speed computing device and the solar energy management system, and estimating a third time interval of the cloud layer reaching the first power plant according to the moving speed of the cloud layer, wherein the The third time interval is later than the second time interval and the first time interval; a power generation computing device is connected to the mobile time calculation device and the power generation amount reduction calculation device, according to the power generation amount of the second power plant and the third power plant The ratio of the falling loss and a reference power generation amount of the first power plant calculate the amount of power generated by the first power plant in the third time interval.

The advantage of this creation is that it can effectively predict whether the first power plant may be affected by cloud interference and cause a decrease in power generation. Compared with the full sky imager, the image is used to judge the influence of cloud movement and cloud layer on the power plant. The proposed solar power plant prediction system and method have no shortcomings of image processing, and can calculate the amount of power generation that may be reduced.

10‧‧‧Power Forecasting System

11‧‧‧Power generation drop search device

12‧‧‧Power generation reduction calculation device

13‧‧‧Mobile speed computing device

14‧‧‧Mobile time computing device

15‧‧‧Power generation calculation device

20‧‧‧Solar management system

21‧‧‧Database

R1, r2‧‧‧ power generation

T1, t2‧‧‧ time

Figure 1 is a block diagram of the power generation prediction system of the solar power plant of the present invention.

2A-2B are schematic views of the distribution of the solar power plant of the present invention.

Figure 3 is a waveform diagram of the power generation of a solar power plant on a sunny day.

Figure 4 is a waveform diagram of the power generation of solar power plants at different times and in different seasons.

Figure 5 is a schematic diagram of the running orbit of the sun at different times and in different seasons.

FIG. 6 is a flow chart showing the steps of the power generation prediction method of the solar power plant according to the present invention.

7A-7C are waveform diagrams of power generation of the second power plant, the third power plant, and the first power plant of the present invention.

Referring to Fig. 1, the power generation prediction system 10 of the solar power plant of the present invention includes a power generation amount lowering search device 11, a power generation amount decrease calculating device 12, a moving speed calculating device 13, a moving time calculating device 14, and a power generating computing device 15. In addition, it should be noted that the solar power plant of the power generation prediction system of the present invention may include one Maximum Power Point Tracking (MPPT), a plurality of MPPTs, one inverter or a plurality of inverters. The solar power plant predicted can be the entire solar power plant or part of the solar power plant, which is not limited here.

The power generation prediction system 10 can be installed in a solar energy management system 20, or the power generation prediction system 10 can be a computer system and connected to the solar energy management system 20, which is not limited herein.

The solar energy management system 20 is a system device for managing a solar power plant, and is used for collecting power plant data, power generation data or component signal output and input data of a solar power plant, and mainly includes a database 21, which can store a plurality of solar power plants. The solar energy management system 20 is well known to those of ordinary skill in the art for power plant data, power generation data or component signal output and input data, and therefore the structure and function of the solar energy management system 20 will not be described in detail herein.

The power generation reduction search device 11 is connected to the solar energy management system 20, and in the database 21 of the solar energy management system 20, by comparing whether the solar power plant's power generation amount and actual power generation amount decrease during the historical time, for example, whether there are a plurality of power plants The phenomenon of a decrease in power generation occurs, and it is determined whether the power generation of the power plant is degraded by the influence of the cloud. The power generation amount lowering search means 11 may be a software that can read the power generation status of all the solar power plants in the data base 21, and search for whether or not the power generation amount drops due to the power generation condition.

For example, centering on a first power plant, which is a target power plant that needs to predict whether it is subject to cloud interference, the power generation reduction search device 11 searches for a second power plant among a plurality of power plants around the first power plant. The power generation amount decreases in a first time interval, and the first power plant and the second power plant are assumed to be imaginary straight lines, and the power generation amount falling search device 11 finds the imaginary straight line external expansion station. Whether there is a third power plant in a region where a similar power generation decline occurs in a second time interval.

As shown in FIG. 2A, assuming that the location of the first power plant is indicated as C, the range of the area expanded from the imaginary straight line may be a sector-shaped area ranging from the position of the first power plant to the apex of the sector, which The fan-shaped area has a central angle, and the power generation reduction search device 11 determines whether there is a second power plant (marker A) and a third power plant (marker H) within the sector area, and the second power plant and the third power plant The power generation amount decreases in a time interval; or as shown in FIG. 2B, the range of the region extending outward from the imaginary straight line may be surrounded by two imaginary parallel lines L1, L2 separated by a distance S, between two parallel The center line L3 between the lines L1, L2 passes through the first power plant C.

The power generation amount lowering search device 11 determines whether the percentage of error between the normal power generation amount and the power generation amount of the first power plant exceeds a threshold value, or according to the power generation amount of the power plant in the historical time interval and the power generation amount Whether the percentage of error between the values exceeds the threshold value and confirms that the first power plant has a decrease in power generation amount as a search method for the power generation of the first power plant. The normal power generation amount and the time for the power generation reduction may be the difference between the instantaneous power generation amount of the front and rear power generation reduction amount, and different threshold values are set in the morning and afternoon. It is also possible to use the data of the power generation in the same period of the same period as the reference data, and compare the actual power generation at the same time point with the power generation on the sunny day. It can be found whether the power generation in a certain moment or a certain period of time has decreased. . The solar power plant's power generation on a sunny day can be used as a reference for finding the amount of solar power generated by the solar power plant during the historical time. If the amount of solar power generated in historical time is not found, you can find the solar power plant in the days before and after the historical time (different years, same Look for the amount of power generated on a sunny day in the amount of electricity generated in the month and the days before and after the same date.

For example, the power generation forecasting system 10 looks for the amount of sunny day power generation during the historical time of September 8, 105 in the Republic of China. It can be found in the database 21 of the solar energy management system 20 in the Republic of China 104 or before the Republic of China 104 years ago on September 8 or Whether there is a sunny day's power generation in the power generation around September 8th. Solar power plant There is no sunny day power generation in the historical time or in the days before and after the historical time, or you can use the solar power plant in historical time (different years, days before and after the same date, data acquisition points at the same time every day, such as 6 : 00, 6:05, ... or 18:00, etc.) The power generation is sampled and screened and combined into a sunny day power generation, and then the interpolation method is used to smooth the sunny power generation curve, as shown in Fig. 3. How to apply the data storage 21 of the solar power plant management system 20 for the comparison and processing of the power generation data is well known to those skilled in the art of software design and data processing, and will not be described herein. In addition, as shown in Fig. 4 and Fig. 5, in the 365 days of the year, the angle of the sun entering the earth is constantly changing, and the amount of power generated by the sunny day is also different. Therefore, it is necessary to separately calculate the sunny days of the third, second, and first power plants on different days. The amount can be used to calculate the power generation loss of the third power plant at the next time point by calculating the power loss of the third and second power plants.

The power generation amount reduction calculation means 12 is connected to the power generation amount reduction search means 11 for calculating the power generation amount reduction loss ratio of the solar power generation power generation amount affected by the cloud layer. For example, the power generation loss reduction ratio can be calculated by calculating the error percentage of the error percentage between the normal power generation amount and the falling power generation amount, or using the power generation data of the same historical period of the power plant as a reference material, and the same The decrease in power generation at the time point is compared with the amount of power generated on a sunny day, and the proportion of power generation loss at the moment or the time interval is calculated, and the calculated drop loss ratio represents the actual power generation loss of the solar power plant.

In addition, in order to improve the prediction accuracy of the power generation prediction system of the solar power plant of the present creation, an additional solar power plant (for example, a fourth power plant or a fifth power plant or even a sixth power plant) may be additionally selected to view the second power plant and the third power plant. One of the power plants is connected in a straight line, so that there is a situation in which the power generation is declining. If the fourth power plant, the fifth power plant, or the sixth power plant is also found to have a decrease in power generation, it is more certain that the cloud layer is along The linear movement of the second power plant and the third power plant will affect the first power plant in the third time interval.

In order to further confirm that the first power plant will have a decrease in power generation in the third time interval, the power generation reduction search device 11 finds that the second power plant around the first power plant has a decrease in power generation amount in the first time interval, and informs the power generation amount to decrease. The device 12 is calculated by the power generation amount reduction calculation device 12 Calculate the proportion of the loss of power generation in the second power plant, such as the decreasing slope of the power generation or the ratio of the decrease and recovery of the power generation. In addition, when the third power plant similarly drops the power generation amount in the second time interval, the second time interval is earlier than the first time interval, and the power generation amount reduction calculation device 12 calculates the power generation loss of the third power plant. proportion.

The moving speed calculating device 13 connects the power generation amount lowering searching device 11 and the solar energy management system 20, and may be based on the distance between the second power plant and the third power plant and the first time interval in which the second power plant and the third power plant respectively reduce the power generation amount. In the second time interval, the moving speed of the cloud layer is calculated. After the power generation reduction search device 11 finds the second power plant and the third power plant that are disturbed by the cloud, the information of the second power plant and the third power plant is transmitted to the moving speed calculation device 13, and the data of the mobile speed calculation device 13 from the solar energy management system 20 The coordinates of the second power plant and the third power plant are found in the library 21, that is, according to the relative distance between the second power plant and the third power plant (as shown in FIG. 2A, the distance d between the mark A and the mark H), first The time difference between the time interval and the second time interval calculates the moving speed of the cloud layer.

The moving time calculating device 14 is connected to the moving speed calculating device 13 and the solar energy management system 20, and can calculate the third time interval of the cloud layer reaching the first power plant according to the moving speed of the cloud layer and the relative distance between the second power plant and the first power plant, according to The moving speed of the cloud layer, and the coordinate position of the first power plant and the second power plant can be found from the solar energy management system 20, and the speed and distance are known to calculate the third time interval in which the cloud layer reaches the first power plant. The third time interval is a future time interval, and the third time interval is later than the second time interval and the first time interval.

The power generation calculation device 15 is connected to the movement time calculation device 14, the power generation amount reduction calculation device 12, and the solar energy management system 20, and multiplies the power generation amount of the second power plant or the third power plant by the power generation amount of the first power plant at the historical time, Or the average power generation of the first power plant in historical time multiplied by the average of the ratio of the power generation loss of the second power plant to the third power plant, and the power generation amount of the first power plant after being affected by the cloud layer in the third time interval can be predicted, according to This amount of power generation allows the power dispatcher to schedule other power supplements early, or to transfer excess power to other power units.

Referring to FIG. 6 and FIG. 2A - FIG. 2B, in step S301, a plurality of power plants around the first power plant are viewed centered on a first power plant (as shown in FIG. 2A - FIG. 2B) (see FIG. 2A - At the power generation status of the A, B, D or G point power plant in Fig. 2B, is there a second power plant in which the power generation is reduced? The first power plant is a target power plant that needs to predict the amount of power generated, and the power generation status of a plurality of power plants around the first power plant is viewed. When a cloud layer moves above a second power plant, the cloud layer shields the sunlight that is irradiated to the second power plant, resulting in sunlight. The power generation of the second power plant has declined. Therefore, by searching for the power generation status of the second power plant around the first power plant, it is known whether there are clouds in the nearby second power plants, causing the power generation to drop. May be affected by the same cloud. The plurality of power plants (including the first power plant, the second power plant, and the third power plant) are power plants managed by a solar energy management system, and the power generation status of the plurality of power plants is recorded in a database of the solar energy management system, and the data can be obtained from the data. The warehouse looks for the power generation status of a number of power plants, and then finds a second power plant that has a reduced power generation around the first power plant.

In step S302, when a second power plant has a power generation decrease phenomenon in a first time interval, the second power plant and the first power plant are imaginary straight lines, and find whether a region exists in a region where the imaginary straight line is located. The third power plant (such as the H-point power plant in Fig. 2A - Fig. 2B) has a similar phenomenon of a decrease in power generation. Because only one second power plant has a decline in power generation, it is impossible to determine whether the direction of movement of the cloud will affect the first power plant. Therefore, it is necessary to find whether there is a similar power generation drop in a third power plant farther away from the first power plant. situation. In addition, it should be noted here that those skilled in the art can understand from the degree of decrease of the power generation amount in the first time interval of the second power plant, and the phenomenon that the power generation amount is decreased is caused by the internal component failure of the solar power plant or because Caused by environmental factors. In addition, the third power plant is also the power plant managed by the solar energy management system, and the distance between the third power plant and the first power plant is farther than the distance between the second power plant and the first power plant, and the second time interval is earlier than the first time. Interval.

In step S303, when the third power plant has a power generation amount decrease in the second time interval, the power generation amount decrease loss ratio is calculated according to the reduced power generation amount of the second power plant and the third power plant, respectively. The proportion of the loss of power generation loss can be calculated by calculating the error percentage or the decreasing slope of the power generation (normal power generation and the reduction power generation) before the power generation decline, or the actual decline of the power generation decline section according to the historical sunny power generation of the power plant. Power generation.

The formula for the falling slope is: R = (r1 - r2) / (t1 - t2).

Here, as shown in FIG. 7A, r2 is the maximum power generation amount in one time interval (first time interval or second time interval), t2 is the time point at which the maximum power generation amount r2 occurs, and r1 is the minimum power generation amount in the time interval, T1 is the time point at which the minimum power generation amount r1 occurs. Alternatively, in different embodiments, the amount of power generated by the first power plant after being affected by the cloud layer in the third time interval may also be calculated by the power generation loss ratio.

The formula for reducing the loss of power generation is: (r1-r2)/r2.

And in step S304, when both the second power plant and the third power plant have a decrease in power generation, it is determined that the second power plant and the third power plant are affected by the same cloud layer. When the power generation of the second power plant and the third power plant decreases in different time intervals, it is determined that the second power plant and the third power plant are affected by the same cloud layer. Figure 7A and Figure 7B show waveform diagrams of the power generation of the power plant caused by the influence of the same cloud layer on two different power plants. Waveform B and waveform A represent the power generation of the second power plant and the third power plant, respectively, and Figure 7B and Figure 7B The waveform diagram of Fig. 7A can be used to understand that the second power plant and the third power plant have a decrease in power generation in the first time interval and the second time interval, respectively, and it can be clearly seen from the waveform diagram that the power generation amount does not directly drop to zero, but The instantaneous power generation is reduced, and it can be seen that both the second power plant and the third power plant are affected by the cloud layer, resulting in a decrease in power generation.

In step S305, the moving speed of the cloud layer is calculated by the moving speed calculating means 13 based on the relative distance between the second power plant and the third power plant and the time difference between the first time interval and the second time interval. The setting of the solar power plant is not based on a fixed distance to set up a solar power plant, solar energy The distance between the power plants is different. It is necessary to calculate the speed at which the cloud layer moves from the third power plant to the second power plant to predict the third time interval in which the cloud layer reaches the first power plant.

In step S306, according to the moving speed of the cloud layer, the moving time calculation means 14 estimates that the cloud layer moves to a third time interval of the first power plant. After obtaining the moving speed of the cloud layer, the distance between the first power plant and the second power plant is also known, so it is possible to calculate the third time interval of the cloud layer reaching the first power plant, and then predict the first power plant in the third time. The interval will be affected by the cloud layer, and the probability of a decrease in power generation is high.

In step S307, based on the power generation amount drop loss ratio and a reference power generation amount of the first power plant, the power generation amount of the first power plant in the third time interval is estimated, as shown in FIG. 7C. The first power plant was not disturbed by any cloud layer during this historical period. In other words, the first power plant generated electricity in the historical time was the amount of power generated on a sunny day, through the first power plant on the sunny day of historical time or the first power plant. The known power generation amount and the power generation loss reduction ratio before the power generation loss due to the cloud layer can be used to estimate the proportion of the power generation loss reduction (or power generation loss) caused by the interference of the cloud layer in the third time interval. Therefore, according to the estimated possible power generation of the first power plant in the third time interval, the power company can effectively allocate power in response to the decline of the power generation of the power plant, and avoid the shortage of power.

In addition, it should be noted that the cloud layer may move arbitrarily in any direction, but because of the unstable solar power generation, the power generation prediction method and system of the present invention still have reference value to prevent the power generation from generating electricity. A steady situation arises.

Claims (8)

A method for predicting power generation of a solar power plant includes the steps of: looking at a first power plant to see if a second power plant in the plurality of power plants around the first power plant has a power generation drop in a first time interval; The power generation of the power plant decreases in the first time interval, and the second power plant and the first power plant are connected to form an imaginary line to find whether a third power plant exists in a region of the imaginary straight line, and the The third power plant has a decrease in the power generation amount in a second time interval, wherein the second time interval is earlier than the first time interval; calculating a power generation loss loss ratio of the second power plant in the first time interval and the first The ratio of the power generation loss of the third power plant in the second time interval; calculating a first time according to the relative distance between the second power plant and the third power plant, and the time difference between the first time interval and the second time interval The moving speed of the cloud layer moving from the third power plant to the second power plant; according to the moving speed of the cloud layer, estimating the third time when the cloud layer moves to the first power plant And wherein the third time interval is a future time interval, and the third time interval is later than the first time interval and the second time interval; and a reference loss generation ratio of the first power plant according to the power generation amount Calculating the amount of power generated by the first power plant in the third time interval; wherein the ratio of the power generation loss of the second power plant to the third power plant is an error between calculating a normal power generation amount and a falling power generation amount Percentage, or referring to the amount of power generated by the second power plant and the third power plant at the same historical time, the power generation amount at the same time point at which the power generation amount decreases is compared with the power generation amount of the sunny day to obtain the power generation The ratio of the amount of loss reduction; further comprising whether the ratio of the power generation loss between the normal power generation amount and the power generation amount exceeds a threshold value according to the first power plant, or according to the sunny day of the power plant in the historical time interval Whether the ratio of the amount of power generation drop loss between the amount of electricity and the amount of power generation exceeds the threshold value confirms that the power generation amount of the first power plant decreases. The power generation prediction method of the solar power plant according to claim 1, wherein the power generation amount of the sunny day is a combination of sampling and filtering the power generation amount of the historical time, or the power generation amount of the sunny day is power generation by using an interpolation method and the historical time. Obtained by quantity. The method for predicting power generation of a solar power plant according to claim 1, wherein in the step of calculating a moving speed of the cloud layer, the second power plant and the third are obtained according to coordinates of the second power plant and the third power plant The distance between the power plants. A power generation prediction system for a solar power plant, comprising: a power generation falling search device, connected to a solar energy management system, in a database of the solar energy management system, searching for a plurality of power plants around the first power plant Whether there is a second power plant that has a decrease in power generation in a first time interval, and whether there is a third power plant that has a power generation decrease in a second time interval, wherein the second time interval is earlier than the first time interval, The first power plant and the second power plant are connected to an imaginary straight line, and the third power plant is located in a region extending outward from the imaginary straight line; a power generation falling calculation device is connected to the power generation falling search device to calculate the a ratio of power generation loss to loss of the second power plant and the third power plant; a moving speed calculation device connecting the power generation falling search device and the solar energy management system according to a relative distance between the second power plant and the third power plant And a time difference between the first time interval and the second time interval, and calculating a moving speed of the cloud layer; a dynamic time calculating device, connected to the moving speed calculating device and the solar energy management system, estimating a third time interval of the cloud layer reaching the first power plant according to the moving speed of the cloud layer, wherein the third time interval is later than the first time interval a second time interval and the first time interval; a power generation calculation device that connects the movement time calculation device and the power generation amount reduction calculation device, and estimates the first according to the power generation amount drop loss ratio of the second power plant and the third power plant and a reference power generation amount of the first power plant a power generation amount of the power plant in the third time interval; wherein the power generation amount reduction calculating device calculates a normal power generation amount and a decrease power generation amount according to a ratio of the power generation amount drop loss of the second power plant to the third power plant a percentage error, or referring to a sunny day power generation amount of the second power plant and the third power plant at the same historical time, comparing the decreased power generation amount at the same time point of the power generation decrease to the sunny power generation amount Obtaining a ratio of the power generation amount drop loss; determining, by the power generation amount reduction device, whether the ratio of the power generation amount drop loss between the normal power generation amount and the power generation amount of the first power plant exceeds a threshold value, or according to the power plant Whether the ratio of the power generation amount drop between the sunny day power generation amount and the down power generation amount in the historical time interval exceeds the threshold value Recognize the occurrence of the first power plant generating capacity decline. The power generation prediction system of the solar power plant according to claim 4, wherein the power generation quantities of the power plants are stored in a database of the solar energy management system. The power generation prediction system of the solar power plant according to claim 4, wherein the power generation prediction system of the solar power plant is a computer system and is connected to the solar energy management system. The power generation prediction system of the solar power plant according to claim 4, wherein the power generation prediction system of the solar power plant is installed in the solar energy management system. The power generation prediction system of the solar power plant of claim 6, wherein the movement speed calculation device is based on coordinates of the second power plant and the third power plant to obtain the relative distance between the second power plant and the third power plant .