System and method for automatically estimating azimuth of solar panel based on power generation and astronomical information

Abstract

The solar panel azimuth angle automatic estimation system of the present invention comprises: a power generation data collection unit that collects time-series power generation data in predetermined time units from a first inverter connected to a first solar panel for which the azimuth angle is to be estimated among a plurality of inverters provided in the solar power generation system; a solar vector calculation unit that calculates a solar vector for each time-series power generation data from the installation location information of the first inverter and the time information in which the time-series power generation data is measured; a candidate azimuth angle calculation unit that calculates a candidate azimuth angle of the first solar panel by searching for a panel vector in which the inner product is maximized for each solar vector at each calculation time point for calculating the solar vector; and an aggregation processing unit that determines a single final azimuth angle for the first solar panel by statistically integrating a plurality of candidate azimuth angles calculated at each calculation time point. By doing so, the azimuth angle of the solar panel can be automatically estimated using actual power generation data and astronomical information of the sun, and thereby, the problem of difficulty in accurately predicting power generation due to the omission of azimuth angle information of the solar panel is resolved, and the accuracy of power generation prediction is improved.

Description

Technology Field

[0001] The present invention relates to a technology for estimating the azimuth angle of a solar panel, and more specifically, to a system and method for automatically estimating the azimuth angle of a solar panel based on power generation and astronomical information. Background Technology

[0002] As the importance of responding to climate change and the energy transition grows, the role of new and renewable energy sources such as solar and wind power is expanding. However, these energy sources are characterized by intermittency, where power generation fluctuates rapidly depending on weather conditions. Therefore, accurate forecasting of power generation is essential for the stable operation of the power grid.

[0003] In the case of photovoltaic power generation systems, the azimuth angle of the solar panels is essential for accurately predicting power generation. This is because power generation prediction models utilize the azimuth and tilt angles of the solar panels, along with astronomical information about the sun, to calculate the transmission effect (i.e., the angle of incidence effect) of the solar radiation reaching the panel surface, thereby determining the effective light energy and enabling accurate prediction of power generation.

[0004] In this regard, Korean Registered Patent No. 10-2404397 discloses a solar power generation prediction device and method, wherein the device and method predict solar irradiance based on location information of a planned area for installation of a solar power generation system, weather information of the planned area, geographical characteristics of the planned area, and the angle of the solar panels of the solar power generation system, and predict solar power generation by considering the power generation efficiency and the conversion efficiency of the solar power generation system according to the solar irradiance and the temperature change of the solar panels.

[0005] According to the above patent, solar irradiance is predicted based on the angle of the solar panels of a photovoltaic power generation system, and solar power generation is predicted by considering the power generation efficiency and the conversion efficiency of the photovoltaic power generation system according to the solar irradiance and the temperature change of the solar panels.

[0006] However, conventionally, since the azimuth information of these solar panels was obtained by manually measuring it or relying on design drawings and construction data, there were frequent instances where the information was inaccurate or missing, making it difficult to predict power generation based on accurate azimuth information.

[0007] Furthermore, conventionally, if azimuth information was inaccurate or missing, structural deformations such as changes in the azimuth of solar panels could not be detected early. Consequently, the resulting short-term drop in power generation was mistaken for weather factors, leading to a decline in the accuracy of power generation performance evaluation and prediction. Moreover, by mistaking fluctuations in power generation caused by installation and equipment factors for weather conditions, there was a problem in being unable to adequately respond to malfunctions or failures of the installations and equipment. Prior art literature

[0008] Republic of Korea Registered Patent No. 10-2404397 The problem to be solved

[0009] Accordingly, the present invention aims to provide a solar panel azimuth automatic estimation system and method that improves the accuracy of power generation prediction by solving the problem of difficulty in accurately predicting power generation due to the omission of solar panel azimuth information, by automatically estimating the azimuth of a solar panel using actual power generation data and astronomical information of the sun.

[0010] In addition, the present invention aims to provide a solar panel azimuth angle automatic estimation system and a method thereof, which enables monitoring of changes in the azimuth angle of a solar panel even from a remote location by automatically estimating the azimuth angle of the solar panel at predetermined intervals, and thereby enables the stable operation of a solar power system by detecting and responding to structural deformation of the solar panel at an early stage.

[0011] Furthermore, the present invention aims to provide an automatic solar panel azimuth angle estimation system and method that accurately estimates the azimuth angle of a solar panel to enable accurate detection of sensor abnormalities based on this, thereby resolving the problem of neglecting maintenance of installations and facilities by mistaking power generation fluctuations caused by installation and facility factors, such as sensor abnormalities, for fluctuations caused by weather factors, and consequently enabling efficient maintenance of installations and facilities. means of solving the problem

[0012] To achieve the above objective, the solar panel azimuth angle automatic estimation system provided in the present invention is a solar panel azimuth angle automatic estimation system for automatically estimating the azimuth angle of a solar panel of a solar power generation system, and comprises: a power generation data collection unit that collects time-series power generation data in predetermined time units from a first inverter connected to a first solar panel for which the azimuth angle is to be estimated among a plurality of inverters provided in the solar power generation system; a solar vector calculation unit that calculates a solar vector for each time-series power generation data from installation location information of the first inverter and time information in which the time-series power generation data is measured; a candidate azimuth angle calculation unit that calculates a candidate azimuth angle of the first solar panel by searching for a panel vector in which the inner product is maximized for each solar vector at each calculation time point for calculating the solar vector; and an aggregation processing unit that determines a single final azimuth angle for the first solar panel by statistically integrating a plurality of candidate azimuth angles calculated at each calculation time point.

[0013] Preferably, the solar panel azimuth automatic estimation system further includes an estimation interval extraction unit that extracts a time interval in which the power generation amount is above a predetermined level as an azimuth estimation interval using the time series power generation data and the corresponding solar vector, and the candidate azimuth calculation unit can search for a panel vector in which the inner product of the first solar vectors is maximized after selecting one or more first solar vectors in which the calculation time of the solar vector is included in the azimuth estimation interval.

[0014] Preferably, the candidate azimuth calculation unit can fix the slope, which is one of the parameters of the panel vector, to an arbitrary setting value and vary the azimuth, which is the other parameter of the panel vector, and calculate the azimuth at which the inner product with the sun vector is maximized as the candidate azimuth of the first solar panel.

[0015] Preferably, the solar panel azimuth angle automatic estimation system further includes a panel tilt estimation unit that learns the time series power generation data and the corresponding solar vector to estimate the tilt of the first solar panel, and the candidate azimuth angle calculation unit can fix the tilt of the first solar panel estimated by the panel tilt estimation unit as the tilt of the panel vector.

[0016] Preferably, the aggregation processing unit can determine the final azimuth angle by calculating a temporary circular average of the candidate azimuth angles by reflecting the circular characteristics of the angle data, removing outliers by reflecting the temporary circular average and the circular distance of each of the candidate azimuth angles, and then calculating a circular average for the remaining candidate azimuth angles.

[0017] Preferably, the solar panel azimuth angle automatic estimation system may further include a monitoring unit that periodically monitors the final azimuth angle determined by the aggregation processing unit and generates an error notification if the value deviates from a preset normal range.

[0018] Meanwhile, to achieve the above objective, the method for automatically estimating the azimuth angle of a solar panel provided in the present invention is a method for automatically estimating the azimuth angle of an arbitrary first solar panel among a plurality of solar panels constituting a solar power generation system using a solar panel azimuth angle automatic estimation system (hereinafter abbreviated as 'estimation system'), wherein the estimation system comprises: a step of collecting time-series power generation data in a predetermined time unit from a first inverter connected to the first solar panel; a step of the estimation system calculating a solar vector for each time-series power generation data from the installation location information of the first inverter and the time information in which the time-series power generation data was measured; a step of the estimation system searching for a panel vector in which the inner product is maximized for each solar vector at each calculation time point for calculating the solar vector to calculate a candidate azimuth angle of the first solar panel; and a step of the estimation system statistically integrating a plurality of candidate azimuth angles calculated at each calculation time point to determine a single final azimuth angle for the first solar panel.

[0019] Preferably, the method further includes the step of the estimation system extracting a time interval in which the power generation amount is above a predetermined level as an azimuth estimation interval using the time series power generation data and the corresponding solar vector, and the step of calculating the candidate azimuth may include selecting one or more first solar vectors in which the time point of calculation of the solar vector is included in the azimuth estimation interval, and then searching for a panel vector in which the inner product of the first solar vectors is maximized.

[0020] Preferably, the step of calculating the candidate azimuth angle may be performed by fixing the slope, which is one of the parameters of the panel vector, to an arbitrary setting value and varying the azimuth angle, which is the other parameter of the panel vector, so that the azimuth angle at which the inner product with the sun vector is maximized is calculated as the candidate azimuth angle of the first solar panel.

[0021] Preferably, the method further includes the step of the estimation system learning the time series power generation data and the corresponding solar vector to estimate the inclination of the first solar panel, and the step of calculating the candidate azimuth angle may fix the inclination of the first solar panel estimated in the step of estimating the inclination as the inclination of the panel vector.

[0022] Preferably, the step of determining the final azimuth angle may include: a step of calculating a provisional circular average of the candidate azimuth angles by reflecting the circular characteristics of the angle data; a step of removing outliers by reflecting the provisional circular average and the circular distance of each of the candidate azimuth angles; and a step of determining the final azimuth angle by calculating a circular average for the remaining candidate azimuth angles.

[0023] Preferably, the step of determining the final azimuth angle may further include the step of the estimation system periodically monitoring the final azimuth angle and generating an error notification if the value deviates from a preset normal range. Effects of the invention

[0024] The solar panel azimuth automatic estimation system and method of the present invention as described above has the effect of improving the accuracy of power generation prediction by solving the problem of difficulty in accurately predicting power generation due to the omission of solar panel azimuth information, by automatically estimating the azimuth of a solar panel using actual power generation data and astronomical information of the sun.

[0025] In addition, the present invention has the effect of enabling the stable operation of a solar power system by automatically estimating the azimuth angle of a solar panel at predetermined intervals, thereby allowing for monitoring changes in the azimuth angle of the solar panel even from a remote location, and enabling early detection and response to structural deformation of the solar panel.

[0026] Furthermore, the present invention enables accurate detection of sensor abnormalities based on the azimuth angle of a solar panel by accurately estimating it. Consequently, it resolves the problem of neglecting maintenance of installations and facilities by mistaking fluctuations in power generation caused by installation and equipment factors, such as sensor abnormalities, for fluctuations caused by weather factors, thereby enabling efficient maintenance of installations and facilities. Brief explanation of the drawing

[0027] FIG. 1 is a schematic block diagram of an automatic solar panel azimuth estimation system according to one embodiment of the present invention. FIGS. 2 and FIGS. 3 are processing flowcharts for a method for automatically estimating the azimuth angle of a solar panel according to an embodiment of the present invention. Specific details for implementing the invention

[0028] Embodiments of the present invention are described below with reference to the attached drawings. The description is provided in detail to enable those skilled in the art to easily practice the present invention. However, the present invention may be embodied in various different forms and is not limited to the embodiments described herein. Meanwhile, in order to clearly explain the present invention in the drawings, parts unrelated to the explanation have been omitted, and similar parts throughout the specification have been given similar reference numerals. Furthermore, explanations of parts that can be easily understood by those skilled in the art even without detailed description have been omitted.

[0029] Throughout the specification and claims, when a part is described as including a certain component, this means that, unless specifically stated otherwise, it does not exclude other components but may include additional components.

[0030] FIG. 1 is a schematic block diagram of a solar panel azimuth angle automatic estimation system according to an embodiment of the present invention. Referring to FIG. 1, the solar panel azimuth angle automatic estimation system (100) according to an embodiment of the present invention includes a power generation data collection unit (110), a solar vector calculation unit (120), an estimation interval extraction unit (130), a panel tilt estimation unit (140), a candidate azimuth angle calculation unit (150), an aggregation processing unit (160), a monitoring unit (170), and a control unit (180) to automatically estimate the azimuth angle of a solar panel of a solar power generation system. For convenience of explanation, the azimuth angle of a first solar panel, which is an arbitrary solar panel, will be described as an example. However, it is obvious that the solar panel azimuth angle automatic estimation system (100) of the present invention can simultaneously estimate the azimuth angles of a plurality of solar panels installed in different directions on an inverter unit basis.

[0031] The power generation data collection unit (110) collects power generation data output from the first solar panel (which is to estimate the azimuth angle) ) is collected as a time series. To this end, the power generation data collection unit (110) collects time series power generation data in predetermined time units (e.g., 15 minutes, 30 minutes, 60 minutes, etc.) from the first inverter connected to the first solar panel for which the azimuth angle is to be estimated among the plurality of inverters provided in the solar power generation system. This is because, typically, the inverter periodically updates the power output value or power generation amount measured by itself, and although there may be solar panels with different azimuth angles within a single power plant, only solar panels facing the same direction are connected to a single inverter. To this end, the power generation data collection unit (110) is connected to the first inverter via a communication network and can collect time series power generation data per unit time provided periodically by the first inverter through the communication network. At this time, the reason the power generation data collection unit (110) collects time-series power generation data is that the time-series power generation data reflects the relative pattern of power generation according to the change in the position of the sun (i.e., change in time), and thus the azimuth angle of the solar panel can be automatically estimated by comparing and analyzing the relative pattern with solar astronomical information.

[0032] Meanwhile, the power generation data collection unit (110) can perform preprocessing such as timestamp matching, unit conversion, and outlier removal of the collected time-series power generation data to ensure accuracy in azimuth angle estimation.

[0033] The solar vector calculation unit (120) collects time series power generation data (from the power generation data collection unit (110) The solar vector corresponding to ) ) calculates. That is, the solar vector calculation unit (120) calculates time series power generation data ( For every collection period of ) solar vector( It calculates the solar vector at each point in time ( ) and the amount of power generated at that point in time ( This is because the azimuth angle is estimated by comparing ). In addition, the solar vector calculation unit (120) includes the installation location information (i.e., latitude, longitude) of the first inverter, and time-series power generation data ( Solar vectors for each time-series power generation data from the time information (i.e., timestamp) where ) was measured ( ) is calculated. To this end, the solar vector calculation unit (120) may receive installation location information of the first inverter from an external management server or database that stores installation location information of the first inverter or manages installation location information of the entire photovoltaic power generation system. In addition, the time information at which the time series power generation data is measured may be information included in the time series power generation data, as time information measured by the first inverter.

[0034] Meanwhile, the solar vector calculation unit (120) can be implemented by applying a predetermined algorithm (e.g., PVLib (Photovoltaic Library)) that can calculate the azimuth and altitude of the sun using the latitude and longitude of the inverter.

[0035] The solar vector calculated in this way ( ) is a value representing the position of the sun as a three-dimensional unit vector, and can be expressed as shown in Equation 1 below.

[0036]

[0037] At this time, represents the sun elevation angle, and represents the sun azimuth. That is, the sun vector ( ) can be calculated using the azimuth and elevation angles at the time of power generation.

[0038] The solar vector calculated in this way ( ) is the panel vector to be described later ( It is used in the calculation of the inner product with ), and is used as an input for numerical optimization under the assumption that the panel azimuth angle that maximizes the inner product (or minimizes the negative inner product) is the condition where the power generation is maximum.

[0039] The estimation interval extraction unit (130) is time series power generation data ( ) and the corresponding solar vector( The time point (or condition) (also known as the azimuth estimation interval) most suitable for estimating the solar panel azimuth angle is extracted from the solar panel. In particular, the estimation interval extraction unit (130) extracts the time series power generation data ( ) and the corresponding solar vector( By using ), a time interval in which the power generation amount is above a predetermined level can be extracted as an azimuth estimation interval. This is intended to improve accuracy by utilizing the characteristic that the power generation amount is maximized as the solar panel azimuth angle receives the sun directly, extracting the time when there is high output (peak) and good weather conditions (e.g., clear) as the azimuth estimation interval, and estimating the panel direction based on the sun position at the time corresponding to the azimuth estimation interval. To this end, the estimation interval extraction unit (130) [uses] time-series power generation data ( collected from the power generation data collection unit (110) Among these, the top 3 days with the highest total daily power generation can be selected based on data between 11:00 and 14:00, which is the time period when high solar radiation is expected, by classifying them by season (e.g., spring (March–May), summer (June–August), autumn (September–November), winter (December–February)), and the 3 times with the highest power generation for each day (e.g., among power generation times in 30-minute intervals) can be extracted as the above-mentioned azimuth angle estimation intervals. This is intended to mitigate bias caused by environmental factors such as clouds or shadows by estimating the azimuth angle based on high-output times distributed by season and time period.

[0040] Meanwhile, the estimation section extraction unit (130) may further reflect the facility capacity of the first solar panel, which is the target for estimating the azimuth angle, and the power generation quality indicator, etc., in order to extract the azimuth angle estimation section.

[0041] The panel tilt estimation unit (140) estimates the tilt of the first solar panel. To do this, the panel tilt estimation unit (140) uses past time-series power generation data output from the inverter connected to the first solar panel ( ) and the corresponding solar vector( ) can learn and, based on the result, estimate the slope of the first solar panel. That is, the panel slope estimation unit (140) uses past time series power generation data ( ) and the corresponding solar vector( The power generation response pattern according to the change in the sun's elevation angle is learned from ), and based on the result, the slope value most similar to the change pattern of the sun's elevation angle and the actual power generation can be estimated as the slope of the first solar panel.

[0042] The candidate azimuth calculation unit (150) is a solar vector ( At each calculation point where ) is calculated, that solar vector ( Internal (by ) The panel vector that maximizes ) ) is searched to calculate the candidate azimuth angle of the first solar panel. In particular, the candidate azimuth angle calculation unit (150) calculates the solar vector ( After selecting one or more first solar vectors that are included in the azimuth estimation interval extracted by the estimation interval extraction unit (130) at the time of calculation of ), a panel vector ( By searching for ), the accuracy can be improved.

[0043] At this time, panel vector( ) can be expressed as a three-dimensional unit vector with the tilt and azimuth angles of the solar panel as parameters. That is, the panel vector of the first solar panel ( ) takes the tilt and azimuth angle of the first solar panel as parameters and can be expressed as Equation 2.

[0044]

[0045] At this time, is the tilt of the panel, and is the azimuth of the panel, which is the subject of estimation.

[0046] That is, the candidate bearing angle calculation unit (150) is a panel vector (where mathematical formula 3 is maximized) Searches for ).

[0047]

[0048] To this end, the objective function of the candidate azimuth calculation unit (150) can search for an azimuth that minimizes the negative inner product through numerical optimization, as exemplified in Equation 4.

[0049]

[0051] At this time, the actually implementable azimuth ( Since the range of ) is 60° or more and less than 300°, the panel vector of Equation 4( Azimuth angle, which is one of the parameters of ) The range of ) may be 60° or more and less than 300°.

[0052] Meanwhile, the candidate bearing angle calculation unit (150) is a panel vector ( The slope, which is one of the parameters of ) Fix ) to an arbitrary setting value, and panel vector( The azimuth angle, which is the remaining one of the parameters of ) While varying only ) the solar vector( The azimuth angle where the inner product with ) is maximized ( ) can be calculated as the candidate azimuth angle of the first solar panel. At this time, the inclination ( The reason for fixing ) to an arbitrary setting value is that it does not change over time, so the accuracy of azimuth estimation is maintained even when processed as a fixed value.

[0053] To this end, the candidate azimuth calculation unit (150) calculates the inclination of the first solar panel estimated by the panel inclination estimation unit (140) as a panel vector ( The slope of ) Fixing it as ), or using the tilt set using the information specified in the design drawings of the photovoltaic power generation system, or any one of the values ​​set as the standard tilt of the solar panels in the region where the photovoltaic power generation system is installed as the panel vector ( The slope of ) It can be fixed as ). In this case, the tilt set using the information described in the above design drawing refers to the design tilt angle where each solar panel or array is to be installed, and this can be used as a reference value for setting the support structure (e.g., rack, frame, etc.) of the panel during construction. Meanwhile, the value set as the standard tilt of the solar panel in the region where the solar power generation system is installed may refer to the recommended panel tilt angle set to maximize the average annual power generation by considering the latitude of the region and the characteristics of the annual elevation angle change of the sun.

[0054] The aggregation processing unit (160) is the above solar vector ( Multiple candidate bearings calculated at each calculation point of ) ) statistically integrates to determine a single final azimuth angle for the first solar panel. That is, the aggregation processing unit (160) collects multiple candidate azimuth angles calculated under various conditions such as season, date, and peak time from the candidate azimuth angle calculation unit (150) to form a set of candidate values ​​( After preparing ), the set of candidate values ​​( One of the candidate bearings included in ) is the final bearing ( ) is determined. To this end, the aggregation processing unit (160) reflects the original characteristics of the angle data and the above candidate azimuths ( The temporary circular average of ) Calculate ) and the above temporary circular average (as in mathematical formula 5) ) and the above candidate bearings ( ) Each circular distance( After saving ), they ( Outliers are removed by reflecting the median (Circular Median Absolute Deviation, CMAD).

[0055]

[0056]

[0057]

[0059] For example, the aggregation processing unit (160) has a circular distance (when z is 2.5 in the above mathematical formula 5) that exceeds the median (CMAD). Candidate defense angles corresponding to ) ) can be considered an outlier and removed.

[0060] In addition, the aggregation processing unit (160) has candidate bearing angles remaining after outlier removal ( For ), as exemplified in Equation 6, the final azimuth angle ( Determines ).

[0061]

[0063] At this time, is the candidate bearing remaining after removing outliers, and is the number.

[0064] The monitoring unit (170) periodically monitors the final azimuth angle determined by the aggregation processing unit (160) and generates an error notification if the value deviates from a preset normal range. For example, the monitoring unit (170) may transmit an error notification message to a pre-registered administrator terminal device or display an error notification warning sound or warning light based on the monitoring result. Accordingly, the present invention has the feature of enabling rapid response to structural deformation of the solar panel by detecting structural deformation of the solar panel at an early stage and notifying thereof.

[0065] The control unit (180) stores a preset azimuth angle automatic estimation algorithm and controls the overall operation of the solar panel azimuth angle automatic estimation system of the present invention by the azimuth angle automatic estimation algorithm. For example, the control unit (180) can control the operation of each of the power generation data collection unit (110), solar vector calculation unit (120), estimation section extraction unit (130), panel tilt estimation unit (140), candidate azimuth angle calculation unit (150), aggregation processing unit (160), and monitoring unit (170) by the azimuth angle automatic estimation algorithm.

[0066] FIGS. 2 and 3 are process flowcharts for a method for automatically estimating the azimuth angle of a solar panel according to an embodiment of the present invention. With reference to FIGS. 1 to 3, a method for automatically estimating the azimuth angle of a solar panel according to an embodiment of the present invention, which automatically estimates the azimuth angle of any first solar panel among a plurality of solar panels constituting a photovoltaic power generation system, is described as follows.

[0067] First, in step S110, the power generation data collection unit (110) obtains power generation data output from the first solar panel to estimate the azimuth angle ( ) is collected as a time series. To this end, the power generation data collection unit (110) collects time series power generation data in predetermined time units (e.g., 15 minutes, 30 minutes, 60 minutes, etc.) from the first inverter connected to the first solar panel to estimate the azimuth angle among the plurality of inverters provided in the photovoltaic power generation system.

[0068] In step S120, the solar vector calculation unit (120) obtains the time series power generation data (110) collected from the power generation data collection unit (110). The solar vector corresponding to ) ) calculates. That is, the solar vector calculation unit (120) calculates time series power generation data ( For each collection cycle of ), the installation location information of the first inverter and the time series power generation data ( Solar vectors for each time series power generation data from the time information where ) was measured ( Produces ).

[0069] In step S130, the estimation interval extraction unit (130) obtains time series power generation data ( ) and the corresponding solar vector( The time point (or condition) (also known as the azimuth estimation interval) most suitable for estimating the solar panel azimuth angle is extracted from the solar panel. In particular, the estimation interval extraction unit (130) extracts the time series power generation data ( ) and the corresponding solar vector( By using ), time intervals where the amount of power generated is above a predetermined level can be extracted as azimuth estimation intervals.

[0070] In step S140, the panel tilt estimation unit (140) estimates the tilt of the first solar panel. To do this, the panel tilt estimation unit (140) uses past time-series power generation data output from the inverter to which the first solar panel is connected ( ) and the corresponding solar vector( ) can be learned, and the inclination of the first solar panel can be estimated based on the result.

[0071] In step S150, the candidate azimuth calculation unit (150) calculates the solar vector ( At each calculation point where ) is calculated, that solar vector ( Internal (by ) The panel vector that maximizes ) ) is searched to calculate the candidate azimuth angle of the first solar panel. In particular, in step S150, the candidate azimuth angle calculation unit (150) calculates the solar vector ( The calculation point of ) is, after selecting one or more first solar vectors included in the azimuth estimation interval extracted in step S130, the panel vector ( ) can be searched. Also, in step S150, the candidate bearing angle calculation unit (150) is a panel vector ( The slope, which is one of the parameters of ) Fix ) to an arbitrary setting value, and panel vector( The azimuth angle, which is the remaining one of the parameters of ) While varying ), the solar vector ( The azimuth angle where the inner product with ) is maximized ( ) can be calculated as the candidate azimuth angle of the first solar panel. In particular, in step S150, the candidate azimuth angle calculation unit (150) calculates the inclination of the first solar panel estimated in step S140 ( ) panel vector( The slope of ) It can be fixed with ).

[0072] In step S160, the aggregation processing unit (160) [complies] with the solar vector ( Multiple candidate bearings calculated at each calculation point of ) ) statistically integrates to determine a single final azimuth angle for the first solar panel. To this end, the aggregation processing unit (160), in step S161, reflects the original characteristics of the angle data and the candidate azimuth angles ( The temporary circular average of ) Calculate ) and, in step S162, the above temporary circular average ( ) and the above candidate bearings ( ) Each circular distance( After saving ), they ( Remove outliers by reflecting the median (Circular Median Absolute Deviation, CMAD) of ), and in step S163, the remaining candidate azimuths ( The final azimuth is determined by calculating the circular average for ).

[0073] In addition, the method for automatically estimating the azimuth angle of a solar panel according to the present invention may further include a step (not shown) in which a monitoring unit (170) periodically monitors the final azimuth angle and generates an error notification when the value deviates from a preset normal range.

[0074] In the description of the method for automatically estimating the azimuth angle of a solar panel according to the present invention with reference to FIGS. 1 to 3, redundant descriptions regarding the content mentioned in the description of the system for automatically estimating the azimuth angle of a solar panel according to the present invention with reference to FIG. 1 have been omitted.

[0075] As such, the solar panel azimuth angle automatic estimation system and method of the present invention is characterized by improving the accuracy of power generation prediction by solving the problem of difficulty in accurately predicting power generation due to the omission of solar panel azimuth angle information, by automatically estimating the azimuth angle of a solar panel using actual power generation data and astronomical information of the sun.

[0076] In addition, the present invention is characterized by the ability to monitor changes in the azimuth angle of a solar panel even from a remote location by automatically estimating the azimuth angle of the solar panel at predetermined intervals, thereby enabling the stable operation of the solar power system by detecting and responding to structural deformation of the solar panel at an early stage.

[0077] In addition, the present invention is characterized by accurately estimating the azimuth angle of a solar panel to accurately detect sensor abnormalities based on this, thereby resolving the problem of mistaking fluctuations in power generation caused by installation and equipment factors, such as sensor abnormalities, for fluctuations caused by weather factors, which leads to the omission of maintenance of the installation and equipment, and consequently enabling efficient maintenance of the installation and equipment.

[0078] Although embodiments of the present invention have been described above, the scope of the present invention is not limited thereto and includes all changes and modifications within the scope recognized as equivalents that can be easily changed by a person skilled in the art from the embodiments to which the present invention belongs. Explanation of the symbols

[0079] 100: Automatic Solar Panel Azimuth Estimation System 110: Power Generation Data Collection Unit 120: Solar Vector Calculation Unit 130: Estimation interval extraction unit 140: Panel slope estimation unit 150: Candidate Bearing Angle Calculation Unit 160: Aggregation Processing Unit 170: Monitoring unit 180: Control unit

Claims

Claim 1 A solar panel azimuth estimation automatic estimation system for automatically estimating the azimuth of a solar panel of a solar power generation system, comprising: a power generation data collection unit that collects time-series power generation data in predetermined time units reflecting the relative pattern of power generation according to the change in position of the sun from a first inverter connected to a first solar panel for which the azimuth is to be estimated among a plurality of inverters provided in the solar power generation system; a solar vector calculation unit that calculates a solar vector for each time-series power generation data from the installation location information of the first inverter and the time information in which the time-series power generation data is measured; an estimation section extraction unit that extracts a time section in which the power generation is above a predetermined level as an azimuth estimation section using the time-series power generation data and the corresponding solar vector; and a candidate azimuth calculation unit that calculates a candidate azimuth of the first solar panel by searching for a panel vector in which the inner product is maximized among the solar vectors calculated by the solar vector calculation unit, with respect to first solar vectors whose calculation time is included in the azimuth estimation section. A solar panel azimuth angle automatic estimation system characterized by including an aggregation processing unit that statistically integrates a plurality of candidate azimuth angles calculated by the candidate azimuth angle calculation unit to determine one final azimuth angle for the first solar panel. Claim 2 delete Claim 3 A solar panel azimuth angle automatic estimation system according to claim 1, wherein the candidate azimuth angle calculation unit fixes the slope, which is one of the parameters of the panel vector, to an arbitrary set value, and varies the azimuth angle, which is the other parameter of the panel vector, and calculates the azimuth angle at which the inner product with the solar vector is maximized as the candidate azimuth angle of the first solar panel. Claim 4 A solar panel azimuth angle automatic estimation system according to claim 3, further comprising a panel tilt estimation unit that learns the time series power generation data and the corresponding solar vector to estimate the tilt of the first solar panel, and wherein the candidate azimuth angle calculation unit fixes the tilt of the first solar panel estimated by the panel tilt estimation unit as the tilt of the panel vector. Claim 5 A solar panel azimuth angle automatic estimation system according to claim 1, wherein the aggregation processing unit calculates a temporary circular average of the candidate azimuth angles by reflecting the circular characteristics of the angle data, removes outliers by reflecting the temporary circular average and the circular distance of each of the candidate azimuth angles, and then calculates a circular average for the remaining candidate azimuth angles to determine the final azimuth angle. Claim 6 A solar panel azimuth angle automatic estimation system according to claim 1, further comprising a monitoring unit that periodically monitors the final azimuth angle determined by the aggregation processing unit and generates an error notification when the value deviates from a preset normal range. Claim 7 A method for automatically estimating the azimuth of a solar panel among a plurality of solar panels constituting a solar power generation system, wherein the estimation system automatically estimates the azimuth of an arbitrary first solar panel using a solar panel azimuth automatic estimation system (hereinafter abbreviated as "estimation system"), the method comprises: a step in which the estimation system collects time-series power generation data in a predetermined time unit that reflects the relative pattern of power generation according to the change in position of the sun from a first inverter connected to the first solar panel; a step in which the estimation system calculates a solar vector for each time-series power generation data from the installation location information of the first inverter and the time information in which the time-series power generation data is measured; a step in which the estimation system extracts a time interval in which the power generation is above a predetermined level as an azimuth estimation interval using the time-series power generation data and the corresponding solar vector; and a step in which the estimation system searches for a panel vector in which the inner product is maximized among the solar vectors calculated in the step of calculating the solar vectors, with respect to first solar vectors whose calculation time is included in the azimuth estimation interval, and calculates a candidate azimuth of the first solar panel. A method for automatically estimating a solar panel azimuth angle, characterized in that the estimation system includes a step of statistically integrating a plurality of candidate azimuth angles calculated in the step of calculating the candidate azimuth angle to determine one final azimuth angle for the first solar panel. Claim 8 delete Claim 9 A method for automatically estimating the azimuth angle of a solar panel according to claim 7, wherein the step of calculating the candidate azimuth angle is characterized by fixing the slope, which is one of the parameters of the panel vector, to an arbitrary set value, and varying the azimuth angle, which is the other parameter of the panel vector, and calculating the azimuth angle at which the inner product with the solar vector is maximized as the candidate azimuth angle of the first solar panel. Claim 10 A method for automatically estimating the azimuth angle of a solar panel according to claim 9, wherein the estimation system further includes the step of estimating the inclination of the first solar panel by learning the time series power generation data and the corresponding solar vector, and the step of calculating the candidate azimuth angle is characterized by fixing the inclination of the first solar panel estimated in the step of estimating the inclination as the inclination of the panel vector. Claim 11 A method for automatically estimating the azimuth angle of a solar panel according to claim 7, wherein the step of determining the final azimuth angle comprises: a step of calculating a provisional circular average of the candidate azimuth angles by reflecting the circular characteristics of the angle data; a step of removing outliers by reflecting the provisional circular average and the circular distance of each of the candidate azimuth angles; and a step of determining the final azimuth angle by calculating a circular average for the remaining candidate azimuth angles. Claim 12 A method for automatically estimating the azimuth angle of a solar panel according to claim 7, wherein the step of determining the final azimuth angle further includes the step of the estimation system periodically monitoring the final azimuth angle and generating an error notification when the value deviates from a preset normal range.

Images