A solar thermal power generation system and its operation method

Abstract

The present invention discloses a solar thermal power generation and supply system and its operation method, which relates to the technical field of power generation and supply system. In order to solve the problem of abandoned light, abandoned wind and abandoned electricity during the off-peak period of power grid load, and the problem of insufficient supply of online electricity during the peak period of power grid load, the meteorological forecast unit and the power coordination control unit are closely coordinated. When there is excess photovoltaic and / or wind power, the solar thermal power generation system stores the excess electricity as thermal energy; when there is insufficient electricity, the thermal energy is released for supplement. Through this peak-cutting and valley-filling method, the energy configuration is optimized, the phenomenon of abandoned wind and abandoned light is effectively reduced, and the efficient use of clean energy and the stable operation of the power grid are guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of power generation and power supply systems, and in particular to a solar thermal power generation and power supply system and an operation method thereof. Background Art

[0002] With the rapid development of clean energy such as photovoltaic power generation and wind power generation, the proportion of clean energy in the power grid has increased year by year. However, since the power generation of photovoltaic power generation and wind power generation is greatly affected by environmental factors, they cannot generate electricity steadily within a day and can only generate electricity during periods of sufficient sunlight and wind power. Usually, the periods when photovoltaic power generation and wind power generation can generate a large amount of electricity are not the periods when the power grid has the largest electricity load, and are even the periods when the power grid has the lowest electricity load. This will inevitably result in the power generated by photovoltaic power generation and wind power generation having nowhere to be consumed, resulting in the abandonment of light, wind and electricity. During the peak period of the power grid load, photovoltaic and wind power are often in the trough period and cannot provide large-load output to the power grid, resulting in a shortage of online electricity.

[0003] Therefore, a solar thermal power generation system and an operating method thereof are needed. Summary of the invention

[0004] In order to solve all or part of the above problems, the purpose of the present invention is to provide a solar thermal power generation and supply system and its operation method, so as to solve the problem of abandoned light, abandoned wind and abandoned electricity during the low period of power load of the power grid, and the problem of insufficient supply of online electricity during the peak period of power load of the power grid.

[0005] To achieve the above-mentioned object, the present invention provides the following technical solutions: A solar thermal power generation and power supply system, comprising: a power generation database: storing the daily power generation of past power generation and the corresponding daily natural weather data;

[0006] Regional power grid database: stores the regional daily power consumption corresponding to the power generation database in the past, and updates the regional power grid daily power consumption before the forecast time in real time. At the same time, it calculates and stores the consumption fluctuation ratio between the past forecast time and before the forecast time, as well as the daily power consumption fluctuation ratio before the forecast time in real time;

[0007] Meteorological forecast unit: cooperate with regional weather forecast, retrieve future regional weather data, match future regional weather data with power generation database, predict future regional daily power generation, and calculate future regional power grid daily power consumption using weight algorithm based on regional power grid database, predict future regional daily power consumption;

[0008] Power coordination control unit: receives the results of the weather forecast module, calculates power compensation or energy storage, and issues instructions to the thermal energy storage unit;

[0009] Thermal energy storage unit: receives instructions from the power coordination control unit to compensate for regional electricity consumption or store excess electricity;

[0010] Regional power grid: Centrally allocates the electrical energy and power generation of thermal energy storage units to supply power to the region.

[0011] Furthermore, the weather forecasting unit includes:

[0012] Regional weather data acquisition module: uses regional weather forecast as real-time database, retrieves future regional daily weather data from the real-time database, cleans and organizes the acquired data, removes invalid and erroneous data, and integrates data from different sources in a unified format;

[0013] Daily power generation data matching module: receives data from the regional weather data acquisition module, inputs future regional weather data into the power generation database for data matching analysis, and uses a matching algorithm based on feature vectors, focusing on weather data and solar term information. By converting weather data and solar term information into feature vectors, the similarity between the feature vectors and the data records in the power generation database is calculated, and the most matching forecast data is locked;

[0014] Regional power grid data matching module: Based on the forecast data matched by the daily power generation data matching module, it is input into the regional power grid database to lock the regional daily power consumption at that time under this forecast data, and the regional power grid daily power consumption before the forecast time point is synchronously retrieved to provide data support for the power grid power consumption forecast module;

[0015] Electricity prediction module: conducts in-depth analysis on the prediction data obtained by the daily power generation data matching module, combines the current weather data and solar terms factors, and uses statistical analysis algorithms to predict future power generation;

[0016] Grid electricity consumption prediction module: obtains electricity consumption data before the prediction time, the consumption fluctuation ratio between the past prediction time and before the prediction time, and the daily electricity consumption fluctuation ratio before the prediction time from the regional grid data matching module, and uses the weight algorithm to calculate the future daily electricity consumption of the regional grid.

[0017] Furthermore, the matching algorithm based on the feature vector in the daily power generation data matching module sets the feature vector transformed from the future regional weather data and solar term information as ,in: Can represent solar terms, - Represent different climate conditions in weather data;

[0018] Suppose the characteristic vector of a data record in the power generation database is , the vector dimension is Consistent, the meanings of each dimension correspond to the same;

[0019] The cosine similarity algorithm is used to calculate the similarity between two vectors. The formula is: ,

[0020] The larger the calculated similarity value is, the more the future regional weather data and solar term information match the record in the power generation database. By traversing all records in the power generation database, the record with the greatest similarity is found, which is the most matching prediction data.

[0021] Furthermore, the statistical analysis algorithm formula is as follows:

[0022] ,

[0023] Among them: The predicted daily power generation in the future is , power generation data at past forecast points , the characteristic vector value converted from the current weather data is , the characteristic vector value of weather data conversion in historical data is , the characteristic vector value of the current solar term factor conversion is , the eigenvector value corresponding to the solar term factor transformation in the historical data is , the weight coefficient of weather data factors on power generation is , the weight coefficient of the influence of solar term factors on power generation is .

[0024] Furthermore, the weight coefficient of the weather data on the power generation Weight coefficient of the influence of solar terms on power generation The specific calculation formula is as follows:

[0025]

[0026] Among them: The power generation is , the vector of weather data factors is , solar term factors are , for Group observation data.

[0027] Furthermore, the specific formula of the weight algorithm is as follows:

[0028] ,

[0029] Among them: The predicted daily power consumption of the regional power grid at the prediction time is ;

[0030] The daily power consumption of the regional power grid one day before the forecast time is ;

[0031] The fluctuation ratio of daily electricity consumption between the previous forecast time point and the previous day is .

[0032] The daily electricity consumption fluctuation ratio between the day before and the two days before the forecast time is .

[0033] Furthermore, the and The calculation formula is as follows:

[0034] ,

[0035] ,

[0036] Including: Daily power consumption of the region at the previous forecast time point , the daily electricity consumption of the region one day before the previous prediction time point , the daily electricity consumption of the region one day before the prediction time , the daily electricity consumption of the region two days before the forecast time .

[0037] Furthermore, the For the compensation coefficient, an initial value is determined by analyzing the difference between historical forecast data and actual electricity consumption, using statistical methods. If it is found that the long-term forecast value is less than the actual value, the compensation coefficient is appropriately increased; if the long-term forecast value is more than the actual value, the compensation coefficient is appropriately reduced.

[0038] Furthermore, the power coordination control unit analyzes the relationship between the predicted daily power generation and the daily power consumption of the regional power grid. If the predicted daily power generation is greater than the daily power consumption of the regional power grid, the excess power is calculated for storage in the thermal energy storage unit; if the predicted daily power generation is less than the daily power consumption of the regional power grid, the power released by the thermal energy storage unit is calculated for replenishment, and based on the calculation result, the instruction to store or release power is sent to the thermal energy storage unit.

[0039] Another technical solution proposed by the present invention is to provide an operation method of a solar thermal power generation system, comprising the following steps:

[0040] S1: Use the regional weather forecast as a real-time database and retrieve the future regional daily weather data from the real-time database;

[0041] S2: Input the future regional weather data into the power generation database for data matching analysis. Based on the feature vector matching algorithm, calculate the similarity between the feature vectors of the data records in the power generation database and lock the most matching forecast data;

[0042] S3: Based on the prediction data matched by the daily power generation data matching module, the prediction data is input into the regional power grid database to lock the regional daily power consumption at that time under the prediction data, and the regional power grid daily power consumption before the prediction time is synchronously retrieved;

[0043] S4: Conduct in-depth analysis on the forecast data obtained by the daily power generation data matching module, combine the current weather data and solar terms factors, and use statistical analysis algorithms to predict future power generation;

[0044] S5: Obtain the power consumption data before the prediction time point, the consumption fluctuation ratio between the past prediction time point and before the prediction time point, and the daily power consumption fluctuation ratio before the prediction time point from the regional power grid data matching module, and calculate the future daily power consumption of the regional power grid using a weight algorithm;

[0045] S6: Calculate power compensation or energy storage, issue instructions, compensate for regional electricity consumption or store excess electricity.

[0046] Compared with the prior art, the present invention has the following beneficial effects:

[0047] The present invention proposes a solar thermal power generation and supply system and an operation method thereof, and the cooperation between a meteorological forecast unit and a power coordination control unit. Therefore, according to the dispatching requirements of the regional power grid, when the photovoltaic power generation system and / or the wind power generation system outputs too much power, the excess power can be stored in the solar thermal power generation system in the form of thermal energy; when the photovoltaic power generation system and / or the wind power generation system outputs insufficient power, the solar thermal power generation system can release the stored thermal energy, thereby achieving peak shaving and valley filling for the photovoltaic power generation system and / or the wind power generation system, so that the combined output curve of the solar thermal power generation system, the photovoltaic power generation system and / or the wind power generation system meets the requirements of the regional power grid, effectively reducing the occurrence of wind and solar power abandonment. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 This is a block diagram of the solar thermal power generation system of the present invention. DETAILED DESCRIPTION

[0049] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0050] like Figure 1 As shown, a solar thermal power generation system comprises:

[0051] Power generation database: stores the daily power generation of past power generation and the corresponding daily natural weather data, which is mainly divided by 24 solar terms. The existing division is mostly divided by four seasons. This division data is not accurate enough, because the division of 24 solar terms can make a more detailed division of the four seasons of the year, and it is also an existing division, which is convenient for data aggregation. The changes in the 24 solar terms are also directly a sign of changes in weather factors such as daily sunlight radiation intensity, overlooking time, and wind speed and wind volume. 365 days of data are divided within the 24 solar terms, and then the weather conditions of the day are recorded every day under fixed photovoltaic and / or wind equipment (including: sunny, cloudy, overcast, rainy and snowy, or strong wind, light wind and moderate wind, etc.), and the power generation of the day is recorded according to the above.

[0052] Regional power grid database: stores the regional daily electricity consumption corresponding to the power generation database in the past, which corresponds to the data recorded in the power generation database one by one, and the recording time is also the same. The purpose is to record the total electricity consumption in the region under different weather conditions in different solar terms, and record and update the regional power grid daily electricity consumption before the prediction time point in real time. The current data recorded here mainly records the daily electricity consumption at least one week before the prediction time point. At the same time, the consumption fluctuation ratio between the past prediction time point and the time before the prediction time point and the daily electricity consumption fluctuation ratio before the prediction time point are calculated and stored in real time. Here is an example: for example, the prediction time point is November 7, then the consumption fluctuation ratio between the past prediction time point and the time before the prediction time point is the regional electricity consumption ratio between November 7 and November 6 in the same solar term in the previous 1-2 years, and the daily electricity consumption fluctuation ratio before the prediction time point is the regional electricity consumption ratio between November 6 and November 5 at present.

[0053] Meteorological forecast unit: cooperate with regional weather forecast, retrieve future regional weather data, match future regional weather data with power generation database, predict future regional daily power generation, and calculate future regional power grid daily power consumption using weight algorithm based on regional power grid database, and predict future regional daily power consumption.

[0054] Regional weather data acquisition module: uses regional weather forecast as real-time database, and retrieves future regional daily weather data from the real-time database, cleans and organizes the acquired data, removes invalid and erroneous data, and integrates data from different sources in a unified format.

[0055] Daily power generation data matching module: used to input future regional weather data into the power generation database for data matching analysis. The matching algorithm based on feature vectors focuses on weather data and solar term information. By converting weather data and solar term information into feature vectors, the similarity between the feature vectors of data records in the power generation database is calculated to lock the most matching forecast data.

[0056] The matching algorithm based on feature vector in the daily power generation data matching module sets the feature vector transformed from future regional weather data and solar term information as ,in: Can represent solar terms, - Represents different climate conditions in weather data (e.g., sunny, cloudy, overcast, rainy, snowy, etc.);

[0057] Suppose the characteristic vector of a data record in the power generation database is , the vector dimension is Consistent, the meanings of each dimension correspond to the same;

[0058] The cosine similarity algorithm is used to calculate the similarity between two vectors. The formula is: ,

[0059] The larger the calculated similarity value is, the more the future regional weather data and solar term information match the record in the power generation database. By traversing all records in the power generation database, the record with the greatest similarity is found, which is the most matching prediction data.

[0060] Regional power grid data matching module: Based on the forecast data matched by the daily power generation data matching module, it is input into the regional power grid database to lock the regional daily power consumption at that time under this forecast data, and synchronously retrieve the regional power grid daily power consumption before the forecast time point to provide data support for the power grid power consumption forecast module.

[0061] Electricity prediction module: conducts in-depth analysis on the prediction data obtained by the daily power generation data matching module, combines the current weather data and solar terms factors, and uses statistical analysis algorithms to predict future power generation;

[0062] The statistical analysis algorithm formula is as follows:

[0063] ,

[0064] Among them: The predicted daily power generation in the future is , power generation data at past forecast points , the characteristic vector value converted from the current weather data is , the characteristic vector value of weather data conversion in historical data is , the characteristic vector value of the current solar term factor conversion is , the eigenvector value corresponding to the solar term factor transformation in the historical data is , the weight coefficient of weather data factors on power generation is , the weight coefficient of the influence of solar term factors on power generation is ;

[0065] It is further necessary to add that the characteristic vector value converted from the current weather data refers to the value obtained by converting the current weather conditions (such as sunny, cloudy, rainy, snowy, etc.) and other related weather data (such as temperature, humidity, etc.) into a characteristic vector according to certain rules and coding methods. This characteristic vector can comprehensively represent various characteristics and attributes of the current weather, and the current is the specific value corresponding to this characteristic vector, which reflects the overall state of the current weather;

[0066] For example, suppose sunny is coded as 1, cloudy is coded as 2, overcast is coded as 3, and the temperature is coded as 1 between 0-10℃ and 2 between 10-20℃. If the current weather is sunny and the temperature is 15℃, the feature vector obtained after encoding and conversion may be (1,2). Then the current value is a specific value corresponding to this feature vector. This value is calculated according to specific algorithms and rules, and it can uniquely represent the current weather conditions.

[0067] The eigenvector values ​​converted from the weather data in the historical data, the eigenvector values ​​converted from the current solar term factors, and the eigenvector values ​​converted from the corresponding solar term factors in the historical data are realized in the same way as the eigenvector values ​​converted from the current weather data.

[0068] Furthermore, the weight coefficient of the impact of weather data on power generation Weight coefficient of the influence of solar terms on power generation The calculation process is as follows:

[0069] Assume that the power generation is , the weather data is , solar term factors are , establish a multiple linear regression model: ,in is the error term;

[0070] The least squares method is used to solve the regression model and we get and The specific calculation process is as follows:

[0071] Assumptions There is a set of observation data, , , define the objective function: , in order to find the smallest and , respectively and Find the partial derivative and set it to 0: ,

[0072] ,

[0073] Arranging the above two equations gives:

[0074] ,

[0075] , write the above equations into matrix form:

[0076] ,

[0077] By solving this matrix equation, we can get and The value of .

[0078] Grid electricity consumption prediction module: obtains electricity consumption data before the prediction time, the consumption fluctuation ratio between the past prediction time and before the prediction time, and the daily electricity consumption fluctuation ratio before the prediction time from the regional grid data matching module, and uses the weight algorithm to calculate the future daily electricity consumption of the regional grid.

[0079] The specific formula of the weight algorithm is as follows:

[0080] The consumption fluctuation ratio between this point in the past and the previous forecast time point: In the regional power grid database, based on the date and time corresponding to the current forecast time point, find the relevant data at the same time point last year. Calculate the daily consumption fluctuation ratio between this point in the past and the previous day, that is,

[0081] ,in It indicates the daily consumption of the region at this time last year. It indicates the daily consumption of the region on the day before this time point last year;

[0082] The fluctuation ratio of the consumption of the day before the forecast time point: Calculate the fluctuation ratio of the daily consumption of yesterday and the day before yesterday. The formula is: ,in Indicates yesterday's regional daily consumption, Indicates the daily consumption of the area on the day before yesterday;

[0083] Compensation factor: Compensation factor Adjustments are made based on the actual operation of the power system and historical prediction errors. A suitable initial value can be determined by analyzing the difference between historical prediction data and actual power consumption, and using statistical methods. In the subsequent operation process, adjustments are made continuously based on the prediction errors fed back in real time. For example, if the long-term prediction value is found to be less than the actual value, the compensation coefficient is appropriately increased; conversely, if the long-term prediction value is more than the actual value, the compensation coefficient is appropriately reduced.

[0084] Assume that the daily power consumption of the regional power grid at the prediction time is , the daily power consumption of the regional power grid one day before the forecast time (yesterday) was , the prediction formula is: , where: and are weight coefficients, respectively representing the recent fluctuation ratio and the fluctuation ratio in the same period last year in the importance degree of prediction, and .

[0085] Power coordination control unit: By analyzing the relationship between the predicted daily power generation and the daily power consumption of the regional power grid, if the predicted daily power generation is greater than the daily power consumption of the regional power grid, calculate the excess power for energy storage in the thermal energy storage unit; if the predicted daily power generation is less than the daily power consumption of the regional power grid, calculate the power released by the thermal energy storage unit for supplementation, and according to the calculation results, send the command of energy storage or power release to the thermal energy storage unit. For example: Let the predicted daily power generation be P1 and the daily power consumption of the regional power grid be P2. When P1 > P2, the energy storage power = (P1 - P2) × energy storage ratio (the energy storage ratio is set according to the performance and strategy of the energy storage device); when P1 < P2, the energy storage release power = (P2 - P1) × release ratio (the release ratio is also set according to the relevant strategy).

[0086] Thermal energy storage unit: Receive the command issued by the power coordination control unit to compensate for the regional power consumption or store the excess power consumption.

[0087] Regional power grid: Centralize and allocate the electric energy and power generation of the thermal energy storage unit for power supply to the region.

[0088] Another technical solution proposed by the present invention: Provide an operation method for a solar thermal power generation and power supply system, including the following steps:

[0089] Step 1: Use the regional weather forecast as a real-time database and retrieve the future regional daily weather data in the real-time database;

[0090] Step 2: Input the future regional weather data into the power generation database for data matching analysis. Based on the matching algorithm of feature vectors, calculate the similarity with the feature vectors of the data records in the power generation database, and lock the most matching prediction data;

[0091] Step 3: Based on the prediction data matched by the daily power generation data matching module, input it into the regional power grid database to lock the current regional daily power consumption under this prediction data, and synchronously retrieve the regional daily power consumption before the prediction time point;

[0092] Step 4: Conduct in-depth analysis on the prediction data obtained by the daily power generation data matching module, combine the current weather data and solar term factors, and use statistical analysis algorithms to predict the future power generation;

[0093] Step 5: Obtain the power consumption data before the prediction time point, the consumption fluctuation ratio between the past prediction time point and before the prediction time point, and the daily power consumption fluctuation ratio before the prediction time point from the regional power grid data matching module, and calculate the future daily power consumption of the regional power grid using the weight algorithm;

[0094] Step 6: Calculate power compensation or energy storage, issue instructions, and compensate for regional electricity consumption or store excess electricity.

[0095] It should be noted that, in the description of the present application, it should be understood that the terms "length", "thickness", "inside", "outside", "axial", "radial", etc., indicating orientations or positional relationships are based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific orientation, be constructed and operated in a specific orientation, and therefore cannot be understood as a limitation on the present invention.

[0096] In addition, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprises", or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article, or apparatus that includes a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article, or apparatus.

[0097] The above description is only a preferred specific implementation manner of the present invention, but the protection scope of the present invention is not limited thereto. Any technician familiar with the technical field can make equivalent replacements or changes according to the technical scheme and inventive concept of the present invention within the technical scope disclosed by the present invention, which should be covered by the protection scope of the present invention.

Claims

1. A solar thermal power generation system, characterized in that: Includes: Power generation database: stores the daily power generation and the corresponding daily natural weather data of past power generation; Regional power grid database: stores the regional daily power consumption corresponding to the power generation database in the past, and updates the regional power grid daily power consumption before the forecast time in real time. At the same time, it calculates and stores the consumption fluctuation ratio between the past forecast time and before the forecast time, as well as the daily power consumption fluctuation ratio before the forecast time in real time; Meteorological forecast unit: cooperate with regional weather forecast, retrieve future regional weather data, match future regional weather data with power generation database, predict future regional daily power generation, and calculate future regional power grid daily power consumption using weight algorithm based on regional power grid database, predict future regional daily power consumption; The weather forecast unit includes: Regional weather data acquisition module: uses regional weather forecast as real-time database, retrieves future regional daily weather data from the real-time database, cleans and organizes the acquired data, removes invalid and erroneous data, and integrates data from different sources in a unified format; Daily power generation data matching module: receives data from the regional weather data acquisition module, inputs future regional weather data into the power generation database for data matching analysis, and uses a matching algorithm based on feature vectors, focusing on weather data and solar term information. By converting weather data and solar term information into feature vectors, the similarity between the feature vectors and the data records in the power generation database is calculated, and the most matching forecast data is locked; Electricity prediction module: conducts in-depth analysis on the prediction data obtained by the daily power generation data matching module, combines the current weather data and solar terms factors, and uses statistical analysis algorithms to predict future power generation; The statistical analysis algorithm formula is as follows: ， Among them: The predicted daily power generation in the future is , power generation data at past forecast points , the characteristic vector value converted from the current weather data is , the characteristic vector value of weather data conversion in historical data is , the characteristic vector value of the current solar term factor conversion is , the eigenvector value corresponding to the solar term factor transformation in the historical data is , the weight coefficient of weather data factors on power generation is , the weight coefficient of the influence of solar term factors on power generation is ; The weight coefficient of the impact of the weather data on power generation Weight coefficient of the influence of solar terms on power generation The specific calculation formula is as follows: Among them: The power generation is , the vector of weather data factors is , solar term factors are , for Group observation data; Regional power grid data matching module: Based on the forecast data matched by the daily power generation data matching module, it is input into the regional power grid database to lock the regional daily power consumption at that time under the forecast data, and the regional power grid daily power consumption before the forecast time point is synchronously retrieved to provide data support for the power grid power consumption forecast module; Power grid electricity consumption prediction module: obtains electricity consumption data before the prediction time point, the consumption fluctuation ratio between the past prediction time point and before the prediction time point, and the daily electricity consumption fluctuation ratio before the prediction time point from the regional power grid data matching module, and calculates the future regional power grid daily electricity consumption using a weight algorithm; Power coordination control unit: receives the results of the weather forecast module, calculates power compensation or energy storage, and issues instructions to the thermal energy storage unit; Thermal energy storage unit: receives instructions from the power coordination control unit to compensate for regional electricity consumption or store excess electricity; Regional power grid: Centrally allocates the electrical energy and power generation of thermal energy storage units to supply power to the region.

2. A solar thermal power generation system according to claim 1, characterized in that: The matching algorithm based on the feature vector in the daily power generation data matching module sets the feature vector transformed from the future regional weather data and solar term information as ,in: Represents solar terms, - Represent different climate conditions in weather data; Suppose the characteristic vector of a data record in the power generation database is , the vector dimension is Consistent, the meanings of each dimension correspond to the same; The cosine similarity algorithm is used to calculate the similarity between two vectors. The formula is: , The larger the calculated similarity value is, the more the future regional weather data and solar term information match the data record in the power generation database. By traversing all records in the power generation database, the record with the greatest similarity is found, which is the most matching prediction data.

3. A solar thermal power generation system according to claim 1, characterized in that: The specific formula of the weight algorithm is as follows: ， Among them: The predicted daily power consumption of the regional power grid at the prediction time is , the daily power consumption of the regional power grid one day before the forecast time is , the fluctuation ratio of daily electricity consumption between the previous forecast time point and the previous day is The daily electricity consumption fluctuation ratio between the day before the forecast time and the two days before is , and are weight coefficients, representing the recent volatility ratio Compared with the same period last year importance in the prediction, and , is the compensation coefficient.

4. A solar thermal power generation system as claimed in claim 3, characterized in that: Said and The calculation formula is as follows: ， ， Including: Daily electricity consumption of the region at the previous forecast time point , the daily electricity consumption of the region one day before the previous prediction time point , the daily electricity consumption of the region one day before the prediction time , the daily electricity consumption of the region two days before the forecast time .

5. A solar thermal power generation system as claimed in claim 4, characterized in that: Said For the compensation coefficient, an initial value is determined by analyzing the difference between historical forecast data and actual electricity consumption, using statistical methods. If it is found that the long-term forecast value is less than the actual value, the compensation coefficient is increased; if the long-term forecast value is more than the actual value, the compensation coefficient is reduced.

6. A solar thermal power generation system according to claim 1, characterized in that: The power coordination control unit analyzes the relationship between the predicted daily power generation and the daily power consumption of the regional power grid. If the predicted daily power generation is greater than the daily power consumption of the regional power grid, the excess power is calculated for storage in the thermal energy storage unit; if the predicted daily power generation is less than the daily power consumption of the regional power grid, the power that needs to be released by the thermal energy storage unit is calculated for replenishment, and based on the calculation results, instructions for storing or releasing power are sent to the thermal energy storage unit.

7. The method for operating a solar thermal power generation system according to any one of claims 1 to 6, characterized in that: The steps include: S1: Use the regional weather forecast as a real-time database and retrieve the future regional daily weather data from the real-time database; S2: Input the future regional weather data into the power generation database for data matching analysis. Based on the feature vector matching algorithm, calculate the similarity between the feature vectors of the data records in the power generation database and lock the most matching forecast data; S3: Based on the prediction data matched by the daily power generation data matching module, the prediction data is input into the regional power grid database to lock the regional daily power consumption at that time under the prediction data, and the regional power grid daily power consumption before the prediction time is synchronously retrieved; S4: Conduct in-depth analysis on the forecast data obtained by the daily power generation data matching module, combine the current weather data and solar terms factors, and use statistical analysis algorithms to predict future power generation; S5: Obtain the power consumption data before the prediction time point, the consumption fluctuation ratio between the past prediction time point and before the prediction time point, and the daily power consumption fluctuation ratio before the prediction time point from the regional power grid data matching module, and calculate the future daily power consumption of the regional power grid using a weight algorithm; S6: Calculate power compensation or energy storage, issue instructions, compensate for regional electricity consumption or store excess electricity.

Images