A system and method for solar irradiance monitoring in weather monitoring systems

Abstract

The present invention relates to a system and method for solar irradiance monitoring in weather monitoring systems (WMS) comprises a data collection module (21), a data filtration module (22), and an alert generation module (23); thereby ensures optimal solar energy generation by improving the precision of solar irradiance measurements, incorporate advanced machine learning techniques and a sophisticated clear sunny period to address the challenges associated with weather data accuracy and anomaly detection in solar energy, enhances the accuracy and reliability of solar irradiance monitoring, results in a significant impact on the solar energy industry by minimizing energy inefficiencies, reducing financial losses, and contributing to a more sustainable and efficient energy generation process.

Description

[0001] “A SYSTEM AND METHOD FOR SOLAR IRRADIANCE MONITORING IN WEATHER MONITORING SYSTEMS”

[0002] FIELD OF THE INVENTION:

[0003] The present invention relates to the field of weather monitoring and data collection in the context of solar energy generation. More particularly, the present invention relates a system and method for solar irradiance monitoring in weather monitoring systems that leverages advanced machine learning techniques and a sophisticated clear sunny period to enhance the accuracy and reliability of solar irradiance monitoring and anomaly detection in Weather Monitoring Systems (WMS).

[0004] BACKGROUND OF THE INVENTION:

[0005] The field of solar energy generation is gaining momentum globally, and efficient and optimal operations of solar power systems are essential to ensure maximum energy generation. Accurate and dependable weather data, including solar irradiance intensity, is indispensable for maximizing solar energy generation and overall system performance. However, traditional weather monitoring systems have faced challenges related to the accuracy and reliability of data, especially concerning solar irradiance measurements.

[0006] Sky conditions, which encompass cloud cover and atmospheric phenomena, are a key determinant of solar irradiance. Accurate solar irradiance measurements are essential for optimizing solar power generation. However, accurately classifying sky conditions and detecting anomalies in solar irradiance data have been complex challenges. Anomalies in weather data can have misguiding information regarding the solar plant performance leading to inaccuracies in performance benchmarking of various components of the solar plant.

[0007] Therefore, there remains a need to solve this technical problem and challenges associated with solar irradiance data accuracy and anomaly detection in such measurements for the context of solar energy generation, by leveraging advanced machine learning techniques and a sophisticated algorithm to detect clear sunny period.

[0008] PRIOR ART AND ITS DISADVANTAGES:

[0009] A Chinese patent application no. CN202310039508A titled as “Photovoltaic fault sensing method and device based on sunny day state, equipment, and medium” suggests a method for identifying standard sunny data in a historical period by utilizing reference sunny data and error correction fitting. This process aims to improve the accuracy of solar energy generation predictions by accounting for variations in solar irradiance. Additionally, said art suggests identifying faults in photovoltaic distributed plants by analyzing historical solar irradiance measurements and energy / power generation data.

[0010] In the realm of photovoltaic plants, the critical factors of energy / power generation and solar irradiance play a pivotal role in the concept of solar energy generation. Said art lacks a clear description of the specific equipment and faults associated with photovoltaic plants. Within the art, data processing, error fitting correction, and identification of photovoltaic plant faults are conducted using historical data. The Root Mean Square Error (RMSE) values of the parameters are employed to pinpoint and address faults effectively. A Chinese patent application no. titled “Photovoltaic system equipment operation data-oriented pre-training method based on self- learning” depicts about the development of a pretraining model based on time series data from solar photovoltaic equipment. Said art also focuses on data processing and self-learning behavior of the pre-trained data.

[0011] A non-patent literature titled as “Anomaly detection in monitoring sensor data for preventive maintenance” is focused on automatic detection of anomalies in railway sensor data with the aim of predicting potential failures in advance. By analyzing normal behavior and identifying contextual criteria related to railway data, anomalies are pinpointed and evaluated. The process begins with data preprocessing, where raw sensor data collected from trains is transformed into a format suitable for data mining tasks, with a specific focus on sensors located in critical components such as bogies. Following this, normal behavior patterns are extracted using data mining techniques. Anomaly detection is then carried out by comparing new data with the extracted knowledge, utilizing sequential patterns to describe component interactions during train journeys. Experimental results from real railway data collected from 12 trains over the course of a year are presented to assess the effectiveness of the approach, offering valuable insights into its performance.

[0012] Another non-patent literature titled as “Machine Learning Schemes for Anomaly Detection in Solar Power Plants” described about a thorough evaluation of machine learning schemes for detecting anomalies in solar PV systems with the goal of enhancing system reliability. Data collected from two solar power plants in India over a period of 34 days was analyzed. The Autoencoder Long Short-Term Memory (AE-LSTM) model emerged as the most effective in accurately identifying anomalies. Data was meticulously gathered from the two solar power plants, encompassing measurements from 22 inverter sensors that captured a range of parameters. The study compared the anomaly detection performance of Isolation Forest, AE-LSTM, and Prophet algorithms, with AE-LSTM showcasing superior accuracy in anomaly detection without misclassifying normal data points.

[0013] In conclusion, the utilization of AE-LSTM for anomaly detection is deemed essential in minimizing downtime and optimizing efficiency in solar power plants. Future research endeavors could delve into innovative anomaly mitigation strategies and the application of distributed machine learning in expansive solar power grids.

[0014] One more non-patent literature titled as “An Intelligent Anomaly Detection Approach for Accurate and Reliable Weather Forecasting at loT Edges: A Case Study” suggests that the study on urbanization and digitalization have posed challenges to weather forecasting accuracy, primarily due to flawed loT sensor data. In response to this issue, a recent study has put forth an anomaly detection approach utilizing five machine learning algorithms. The introduction highlights the impact of urbanization on weather forecasting and emphasizes the role of loT and machine learning in enhancing predictive analytics for real-time monitoring. The discussion delves into the proposed model, which aims to improve loT-based forecasting by identifying anomalies between sensing and network layers. Furthermore, the study evaluates five algorithms for real-time anomaly detection. In conclusion, simulated models provide valuable insights into algorithm performance, with the RF algorithm demonstrating strong detection capabilities but facing challenges in execution time. On the other hand, the SVC algorithm shows promise but struggles with speed.

[0015] However, said cited arts are completely silent over:

[0016] 1. identifying anomalies in solar radiation measurement sensors (pyranometers and the like) using the energy / electricity production and solar irradiation of photovoltaic systems

[0017] 2. detecting anomalies automatically based on real-time solar energy and radiation measurements

[0018] 3. detecting different types of anomalies of in solar irradiance measurement devices through training data set derived from actual field testing under different weather conditions over a period of about 45 days to 2 months.

[0019] DISADVANTAGES OF THE PRIOR ART:

[0020] Said prior art suffers from at least all or any of the following disadvantages :

[0021] All of the prior arts fail to suggest a system and method for detection of measurement anomalies in solar irradiance sensors.

[0022] All of the prior arts fail to suggest identifying abnormalities in solar radiation measuring sensors (pyranometers and the like) using the energy / electricity production and solar irradiation of photovoltaic systems.

[0023] All of the prior arts fail to suggest detecting anomalies based on realtime solar energy and radiation measurements. All of the prior arts fail to suggest evaluating possible anomalies in different types of solar measuring devices through actual field testing under various weather conditions over an extended period (approximately 45 days to 2 months). Additionally, they do not address the automatic detection of anomalies in solar irradiance sensors based on training data obtained from these field tests.Therefore, the aforementioned drawbacks and limitations of the prior arts have been solved by the present invention efficiently and correctly.

[0024] OBJECTS OF THE INVENTION:

[0025] The main object of the present invention is to provide a system and method for solar irradiance monitoring in weather monitoring systems.

[0026] The principle object of the present invention is to ensure accurate performance evaluation of a solar plant by improving the precision of solar irradiance measurements.

[0027] A further object of the present invention is to incorporate advanced machine learning techniques and a sophisticated clear sunny period algorithm to address the challenges associated with weather data accuracy and anomaly detection in solar energy applications.

[0028] Another object of the present invention is to provide a system and method for solar irradiance monitoring that first checks for cloudy conditions using a proprietary algorithm. Once the cloud condition is assessed, the system flags potential faults and automatically triggers alerts via email, thereby enhancing the accuracy and reliability of solar irradiance monitoring within Weather Monitoring Systems (WMS) and solar energy generation through actual field testing under various weather conditions over an extended period.

[0029] Yet another object of the present invention is to provide a system and method for solar irradiance monitoring in weather monitoring systems that result in a significant impact on the solar energy industry by minimizing energy inefficiencies, reducing financial losses, and contributing to a more sustainable and efficient energy generation process.

[0030] One more object of the present invention is to provide a system and method for solar irradiance monitoring in weather monitoring systems that employ advanced machine learning models for anomaly detection and data enhancement.

[0031] Another object of the present invention is to provide a system and method for solar irradiance monitoring in weather monitoring systems that overcomes the drawbacks associated with the existing technology.

[0032] BRIEF DESCRIPTION OF DRAWING

[0033] Various other objects, features, and attendant advantages of the present invention will become fully appreciated as the same becomes better understood when considered in conjunction with the accompanying drawings, in which like reference characters designate the same or similar parts throughout the several views, and wherein:

[0034] Figure 1 : Shows perspective view of a solar power plant having a Weather Monitoring Station (WMS) according to the present invention. Figure 2: Shows a flow chart of a system for solar irradiance monitoring in weather monitoring systems according to an exemplary embodiment of the present invention.

[0035] Figure 3: Shows a flow chart of a data collection module according to the present invention.

[0036] Figure 4: Shows a flow chart of a data filtration module according to the present invention.

[0037] Figure 5: Shows a flow chart of an alert generation module according to the present invention.

[0038] Figure 6: Shows a flow chart of a method for solar irradiance monitoring in weather monitoring systems according to an exemplary embodiment of the present invention.

[0039] Figure 7(a): Shows the graphical representation of the detection of dust on the pyranometer according to the present invention.

[0040] Figure 7(b): Shows the graphical representation of the post-correction data according to the present invention.

[0041] SUMMARY OF THE INVENTION:

[0042] The present invention relates to a system and method for solar irradiance monitoring in weather monitoring systems (WMS) comprises a data collection module, a data filtration module, and an alert generation module; thereby ensures optimal solar energy generation by improving the precision of solar irradiance measurements, incorporate advanced machine learning techniques and a sophisticated clear sunny period to address the challenges associated with weather data accuracy and anomaly detection in solar energy, enhances the accuracy and reliability of solar irradiance monitoring, results in a significant impact on the solar energy industry by minimizing energy inefficiencies, reducing financial losses, and contributing to a more sustainable and efficient energy generation process.

[0043] LIST OF REFERENCE NUMERALS

[0044] Sun (1)

[0045] Solar cell (2)

[0046] Inverter (3)

[0047] Power generation measuring meter (4)

[0048] Weather monitoring stations (5)

[0049] Temperature sensor (6)

[0050] Pyranometer (7)

[0051] Data collection module (21)

[0052] Data filtration module (22)

[0053] Alert generation module (23)

[0054] DETAILED DESCRIPTION OF THE INVENTION:

[0055] The following description is presented to enable any person skilled in the art to make and use the invention and is provided in the context of a particular application and its requirements. Various modifications to the disclosed embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be applied to other embodiments and applications without departing from the spirit and scope of the present invention. Thus, the present invention is not intended to be limited to the embodiments shown but is to be accorded the widest scope consistent with the principles and features disclosed herein.

[0056] It is to be understood that the term “comprising” or “comprises” used in the specification and claims refers to the element of the invention which comprises X, Y, and Z, which means that the invention might have other elements in addition to X, Y, and Z. For example, their invention could include A, B, and / or C as long as it also has X, Y, and Z.

[0057] The present invention relates to a system and method for solar irradiance monitoring in weather monitoring systems that provides unprecedented accuracy and reliability in solar irradiance data, utilizing advanced machine learning techniques and clear sunny periods to enhance the quality of the data.

[0058] Now, according to the embodiments shown in Figure 1 , a solar power plant is having various components / equipment. For this invention, major components / equipment are solar cells / panels (2), inverters (3), power generation measuring meters (4), and weather monitoring stations (5) are described herein. Wherein said solar power plant is based on the conversion of sunlight into electricity using solar cells (2).

[0059] Said solar cell (2) is an apparatus designed to harness solar energy emitted by the sun (1) and convert it into electrical energy. The solar energy, carried by photons, is absorbed by a semiconductor material within the solar cell (2). Upon absorption, the photons dislodge electrons from the material's atoms, thereby generating an electric current. The surface of the semiconductor material is meticulously engineered during manufacturing to facilitate the migration of free electrons to the front surface of the cell, thus enabling the efficient generation of electricity.

[0060] - The inverter (3) is a device configured for converting the electricity produced by a solar panel / cell (2) from DC (Direct Current) to AC (Alternating current), thereby facilitating its use by the electrical grid. Said conversion is enabled by the inverter (3), which falls under the category of power electronics and plays a crucial role in regulating the flow of electrical power.

[0061] - The power generation measuring meter (4) is a device configured for measuring solar power.

[0062] - The weather monitoring stations (5) are configured to collect weather data, including solar radiation, module surface temperature, ambient temperature, wind speed, etc., at any solar PV (Photovoltaic) site to monitor the efficiency and performance of the power plant. Further analysis of this data, in relation to the plant output, assists in analyzing and improving the plant's performance. Said WMS comprises of pyranometers (7) configured to measure solar irradiance from a hemispherical field of view incident on a flat surface. Said is traditionally used for climate research, weather monitoring, and solar energy applications. The pyranometers measure Global Horizontal Irradiance (GHI), Global Inclined Irradiance (GII) and are crucial in assessing the performance and efficiency of photovoltaic (PV) power plants. They are also used to determine suitable sites for PV power plants. Pyranometers operate based on the Seebeck- or thermoelectric effect. a temperature sensor (6) which is combined with a humidity sensor to provide comprehensive meteorological data. The temperature sensor utilizes a resistive thermal device for accurate temperature measurement, while the humidity sensor utilizes a humidity-sensitive condenser. These sensors are installed within radiation screens to prevent the thermal effects of direct and indirect radiation.

[0063] It is to be understood to the person skilled in the art that the aforementioned details are well known to an ordinary skilled person and said details have been incorporated for better clarity of the invention. Hence, the same shall not be considered as a scope of protection of the present invention.

[0064] Now referring to Figure 2, a system for solar irradiance monitoring in weather monitoring systems according to an exemplary embodiment is shown. The system for solar irradiance monitoring in weather monitoring systems comprises of a data collection module (21), a data filtration module (22), and an alert generation module (23).

[0065] According to embodiments shown in Figure 3, said data collection module (21) consists of: acquisition (S21.1) of data from Weather Monitoring Stations (WMS) and plant meters located at the solar plants; wherein, said stations are equipped with pyranometers, temperature sensors, and MFM (Multi-function meters) to measure solar irradiance, ambient temperature, and power generation, respectively, transmission (S21.2) of said data via hardwired connections to a PLC (Programmable Logic Controller) system for reliability and accuracy; wherein, said PLC interprets and transforms the raw data into a format suitable for a SCADA (Supervisory Control and Data Acquisition) system, communication (S21.3) of the PLC system with the Supervisory Control and Data Acquisition (SCADA) system utilizing the Modbus TCP / IP (Transmission Control Protocol / Internet Protocol) protocol; wherein said communication is facilitated over a fiber optic (FO) cable, ensuring enhanced speed and reliability, integration (S21.4) of the SCADA system with an OPC (OLE for Process Control) Server; wherein said server leverages both OPC- DA (Data Access) and OPC-UA (Unified Architecture) protocols for efficient data exchange, and subsequently, transmission (S21.5) of the processed data from the OPC server over the internet to a cloud platform; wherein, the system actively identifies and rectifies any anomalies or irregularities in the pyranometer data, which represents solar irradiance, thereby ensuring the generation of accurate and reliable data for subsequent analysis and decision-making, ultimately contributing to optimal solar plant performance and enabling predictive maintenance. In accordance with the embodiments shown in Figure 4, said data filtration module (22) consists of: perform (S22.1) field testing rigorously for fault anomaly model training over a period of 45 days encompassing various scenarios; wherein, said tests are conducted for healthy conditions, dust detection, shadow fault, and identification of clear sky days and cloudy days, employ (S22.2) a computation system to differentiate between clear sky and cloudy days using clear sky data, thereby establishing a baseline for solar irradiance, calculation of slopes to represent expected irradiance changes, and analyze deviations from said slopes to accurately classify sky conditions, acquisition (S22.3) of relevant sensor data from the SCADA system, which includes temperature (°C) that affects plant generation, global inclined irradiance (GII, W / m2) that is the primary measurement of solar radiation, and active power (MW) that provides context for solar energy production, to facilitate parameter selection, subsequently, process (S22.4) said acquired data through a pre- analytic engine, which employs statistical techniques to filter the data, extract meaningful features for anomaly detection, and standardize the data for further processing, process (S22.5) the pre-processed data through an Al (Artificial Intelligence) Kernel; wherein, said Al kernel utilizes machine learning methods such as deep learning networks, support vector machines, and ensemble methods to detect anomalies and integrates a feedback loop for hyperparameter tuning to optimize performance, transmission (S22.6) of said output to a post- analytics engine for identification of anomalies or data points deviating significantly from expected behavior, store (S22.7) the raw data, processed data, and data of anomalies in a central database; wherein said database is configured with the pre-analytic and the post-analytic engines respectively, and access (S22.8) said data through a user equipment interface for result visualization and insights into pyranometer performance.

[0066] It is to be understood that said clear sunny period is an integral part of the system and its function is crucial to the accuracy of solar irradiance measurements. The clear sunny period determines the prevailing sky conditions and is configured to work with data collected at precise 5- minute intervals. Said specifically focuses on data obtained during clear sunny days, which are considered ideal for precise solar irradiance measurements. Said 45 days are essential for developing the system’s model, as the data collected during said periods provides the baseline for accurate measurements and anomaly detection.

[0067] Said machine learning models are employed to identify anomalies in the data collected, contributing to the accuracy and reliability of the system. The models analyze the data with machine learning models and statistical techniques to enhance the quality of the data and minimize energy inefficiencies. This ensures that accurate and reliable data is generated for solar irradiance monitoring in weather monitoring systems. Now, according to the embodiments shown in Figure 5, said alert generation module (23) consists of: retrieval (S23.1) of data from the database and apply various analytical techniques through an alert generation engine; wherein, said engine is configured to combine with thresholds and limits established through historical data analysis and trigger alerts when pyranometer behaviors deviate significantly from the expected norms through an alert generation engine, thereby indicates a potential fault, deliver (S23.2) said alert on the user-equipment interface, thereby allowing operators to monitor the system and investigate the reported issue, and simultaneously, send (S23.3) the alert to relevant personnel via email, thereby ensuring timely awareness and enabling immediate action.

[0068] Said multi-channel (i.e. delivery of alert on user interface and as an email notification) aims to promptly detect and address potential pyranometer faults, minimizing disruptions and maximizing the reliability of solar energy production.

[0069] The system of solar irradiance monitoring in weather monitoring systems impact on the solar energy industry is expected to be significant, as it minimizes energy inefficiencies, reduces financial losses, and contributes to a more sustainable and efficient energy generation process. By providing accurate and reliable solar irradiance data, the system enhances the efficiency of solar energy generation, making it a more viable and sustainable solution. Now referring to the Figure 6 of the present invention, a method for solar irradiance monitoring in weather monitoring systems according to an exemplary embodiment is shown. The method for solar irradiance monitoring in weather monitoring systems comprises of the following steps:

[0070] SI : utilizing a central database to store and retrieve processed pyranometer data through a data collection module (21),

[0071] S2: selecting relevant parameters such as temperature (°C), global irradiance (GII, W / m2), and active power (MW) from a SCADA system,

[0072] S3: subjecting said acquired data to a pre-processing stage involving filtering, feature extraction, and data normalization using a pre- analytic engine of a data filtration module (22),

[0073] S4: analyzing the processed data for detecting anomalies and suggesting correct pyranometer readings through an Al (Artificial Intelligence) kernel,

[0074] S5: optimizing said Al Kernel performance by adjusting the parameters through a feedback loop,

[0075] S6: storing the results, including any detected faults or corrections, back in the database for future reference and analysis, and

[0076] S7: displaying a clear visualization of the detected anomalies and suggested corrections on a user-equipment interface and an email notification to enable users to comprehend the system's performance and take appropriate actions through an alert generation module (23). The present invention provides an effective solution for enhancing the accuracy and reliability of solar irradiance monitoring in Weather Monitoring Systems (WMS), with a specific emphasis on solar energy generation. The integration of advanced machine learning techniques and a sophisticated clear sunny period offers precise anomaly detection and sky condition determination, improving the quality of solar irradiance data and contributing to optimal solar energy generation.

[0077] In the foregoing, it is understood that the terms "solar plant", "PLC", "SCADA system", "OPC Server", "anomaly detection", "field testing", and "Al kernel" are intended to have their broadest meaning as understood by those skilled in the art.

[0078] The present invention has been described in terms of specific embodiments, and it is understood that various other modifications may be made without departing from the scope of the invention. The abovedescribed embodiments are illustrative and not restrictive. The application is thus intended to cover various modifications and equivalent arrangements included within the spirit and scope of the appended claims.

[0079] Experiment and test data:

[0080] 1. Dust induced fault: The dust-induced fault is performed on a pyranometer to understand how Dust affects light reaching the thermopile (measuring surface). The greater the density of dust accumulation lesser the responsivity of the pyranometer.

[0081] Test Methodology: >- Select one pyranometer and apply some quantity of sand deliberately and evenly on the dome.

[0082] >- Carry out the test between 06:00 AM to 06:00 PM on a clear sky day, ensuring that other pyranometer sensors are clean.

[0083] 2. Tilt angle fault: A tilt angle test aims to evaluate how the tilt angle affects the responsivity of the pyranometer sensor

[0084] Test Methodology:

[0085] >- Clean all the pyranometer sensors in a plant using lint-free clothes and alcohol / distilled water.

[0086] >- Carry out the test between 06:00 AM to 06:00 PM on a clear sky day, by changing the orientation of one sensor from its fixed position and recording the results. E.g.: if the tilt angle is on the baseline, change it to +-5°, +-10 °, and record it on a clear sky day, ensuring other pyranometers are aligned to the baseline position.

[0087] 3. Shadow fault: A shadow fault test aims to analyze the impact of shadow on the responsivity of the pyranometer.

[0088] Test Methodology:

[0089] >- Clean all the pyranometer sensors in a plant using lint-free clothes and alcohol / distilled water.

[0090] >- Create a shadow on the pyranometer, and conduct the test between 06:00 AM to 06:00 PM on a clear sky day, ensuring other pyranometers are clean and functioning without any shadow. Healthy case: A healthy case refers to the ideal working condition of a pyranometer, where it is regularly cleaned, kept at the desired angle, and unaffected by dust and shadow.

[0091] Test Methodology: >- Clean all the pyranometer sensors in a plant using lint-free clothes and alcohol / distilled water.

[0092] >- Carry out the test between 06:00 AM to 06:00 PM on clear sky and cloudy days, ensuring other pyranometers are clean and functioning without having any influence of dust and shadow. Test Data Collection: Data collection should be carried out from the site for the aforementioned test cases for one-minute duration, ensuring 100% data availability. Table 1 shown below summarize the above experiment performed on the field in numerical values.

[0093] Table 1

[0094] Results:

[0095] In accordance with the disclosed embodiments, the system has been subjected to numerous experiments. The visual representation of the test results is depicted in the accompanying figures 7a and 7b. The present system has successfully identified an anomaly, wherein one of the pyranometers is affected by dust. Subsequent physical verification of the pyranometer confirmed the existence of the issue. The graphical data illustrating the detection of dust on the pyranometer is portrayed in the provided figure 7a. Upon the removal of dust from the pyranometer, the Global In-plane Irradiation was appropriately rectified. The graphical representation of the post-correction data is presented in the provided figure 7b.

[0096] ADVANTAGES OF THE INVENTION:

[0097] • A system and method for solar irradiance monitoring in weather monitoring systems ensures optimal solar energy generation by improving the precision of solar irradiance measurements.

[0098] • A system and method for solar irradiance monitoring in weather monitoring systems incorporate advanced machine learning techniques and a sophisticated clear sunny period to address the challenges associated with weather data accuracy and anomaly detection in solar energy applications.

[0099] • A system and method for solar irradiance monitoring in weather monitoring systems enhances the accuracy and reliability of solar irradiance monitoring in the context of Weather Monitoring Systems (WMS), with a specific emphasis on solar energy generation.

[0100] • A system and method for solar irradiance monitoring in weather monitoring systems result in a significant impact on the solar energy industry by minimizing energy inefficiencies, reducing financial losses, and contributing to a more sustainable and efficient energy generation process.

[0101] • A system and method for solar irradiance monitoring in weather monitoring systems employs advanced machine learning models for anomaly detection and data enhancement.

[0102] • A system and method for solar irradiance monitoring in weather monitoring systems overcome the drawbacks associated with the existing technology.

Claims

We Claim,1. A system for solar irradiance monitoring in weather monitoring systems (WMS) comprises of a data collection module (21), a data filtration module (22), and an alert generation module (23); wherein, said data collection module (21) consists of: acquisition (S21.1) of data from Weather Monitoring Stations (WMS) and plant meters located at the solar plants, transmission (S21.2) of said data via hardwired connections to a PLC (Programmable Logic Controller) system for reliability and accuracy, communication (S21.3) of the PLC system with the Supervisory Control and Data Acquisition (SCADA) system utilizing the Modbus TCP / IP (Transmission Control Protocol / Internet Protocol) protocol, integration (S21.4) of the SCADA system with an OPC (OLE for Process Control) Server, and transmission (S21.5) of the processed data from the OPC server over the internet to a cloud platform; said data filtration module (22) consists of: perform (S22. 1) field testing rigorously for fault anomaly model training over a period of 45 days encompassing various scenarios,employ (S22.2) a computation system to differentiate between clear sky and cloudy days using clear sky data, acquisition (S22.3) of relevant sensor data from the SCADA system, which includes temperature (°C), global inclined irradiance (GII, W / m2), and active power (MW), process (S22.4) said acquired data through a pre-analytic engine, to filter the data, extract meaningful features for anomaly detection, and standardize the data, process (S22.5) the pre-processed data through an Al (Artificial Intelligence) Kernel, transmission (S22.6) of said output to a post-analytics engine for identification of anomalies or data points deviating significantly from expected behavior, store (S22.7) the raw data, processed data, and data of anomalies in a central database, and access (S22.8) said data through a user equipment interface for result visualization and insights into pyranometer performance; said alert generation module (23) consists of: retrieval (S23.1) of data from the database and apply various analytical techniques through an alert generation engine, deliver (S23.2) said alert on the user-equipment interface, and send (S23.3) the alert to relevant personnel via email.

2. The system as claimed in claim 1, wherein said stations are equipped with pyranometers, temperature sensors, and MFM (Multi-function meters) to measure solar irradiance, ambient temperature, and power generation, respectively.

3. The system as claimed in claim 1, wherein said PLC interprets and transforms the raw data into a format suitable for a SCADA (Supervisory Control and Data Acquisition) system.

4. The system as claimed in claim 1 , wherein said communication is facilitated over a fiber optic (FO) cable, ensuring enhanced speed and reliability.

5. The system as claimed in claim 1, wherein said server leverages both OPC-DA (Data Access) and OPC-UA (Unified Architecture) protocols for efficient data exchange.

6. The system as claimed in claim 1, wherein said tests are conducted for healthy conditions, dust detection, shadow fault, and identification of clear sky days and cloudy days.

7. The system as claimed in claim 1, wherein said Al kernel utilizes machine learning methods such as deep learning networks, support vector machines, and ensemble methods to detect anomalies and integrates a feedback loop for hyperparameter tuning to optimize performance.

8. The system as claimed in claim 1, wherein said database is configured with the pre-analytic and the post-analytic engines respectively.

9. The system as claimed in claim 1, wherein said engine is configured to combine with thresholds and limits established through historical data analysis and trigger alerts when pyranometer behaviors deviate significantly from the expected norms through an alert generation engine, and indicates a potential fault.

10. A method for solar irradiance monitoring in weather monitoring systems comprises of the following steps: utilizing a central database to store and retrieve processed pyranometer data through a data collection module (21), selecting relevant parameters such as temperature (°C), global irradiance (GII, W / m2), and active power (MW) from a SCADA system, subjecting said acquired data to a pre-processing stage involving filtering, feature extraction, and data normalization using a pre- analytic engine of a data filtration module (22), analyzing the processed data for detecting anomalies and suggesting correct pyranometer readings through an Al (Artificial Intelligence) kernel, optimizing said Al Kernel performance by adjusting the parameters through a feedback loop, storing the results, including any detected faults or corrections, back in the database for future reference and analysis, and displaying a clear visualization of the detected anomalies and suggested corrections on a user-equipment interface and an email notification to enable users to comprehend the system'sperformance and take appropriate actions through an alert generation module (23).

Images