A system and method for detecting anomaly in wind speed measurement

Abstract

A SYSTEM AND METHOD FOR DETECTING ANOMALY IN WIND SPEED MEASUREMENT The present invention relates to a system and method for detecting anomaly in the wind speed measurement that comprises of a data collection module (8) to collect SCADA data from wind turbines, a data filtering module (9) to filter out data points where the turbine is underperforming, a wind speed correction module (11) to correct the wind speed for data points, and an alert generation module (10) to initiate alerts. Thereby resulting in improved accuracy of wind speed measurement data, reduced number of false alarms and alerts by means of more accurate wind speed records of the turbine, accurate power generation forecasts, and better wind turbine performance benchmarking. Figures. 1 and 2

Description

FIELD OF THE INVENTION:The present invention relates to wind turbine performance monitoring. More particularly, the present invention relates to a system and method for detecting anomalies in the wind speed measurement by nacelle mounted anemometers which detects inaccuracies and corrects the incorrect wind speed recorded by anemometer ensuring no change to any suspected derating of the turbine. BACKGROUND OF THE INVENTION:Wind turbines are a type of renewable energy system that can generate electricity from wind power, which is an abundant and clean source of energy. Wind turbines are typically equipped with anemometers or wind speed measurement sensors that measure the velocity of the wind and provide feedback to the control systems of the turbine. Wind speed measurements are critical to the operation of wind turbines, as they are used to optimize the turbine's performance and maximize power output.However, wind speed measurement sensors can suffer from measurement anomalies, which can lead to inaccurate or unreliable data. These anomalies can be caused by a variety of factors, such as mechanical wear and tear, environmental conditions, or electronic malfunctions. It is therefore important to monitor the quality of the anemometer data to ensure that the wind turbine is operating safely and efficiently.Therefore, there remains a need to solve this technical problem and provide a system and method for detecting measurement anomalies in wind speed measurement sensors / anemometers. Detecting and diagnosing anomalies in the anemometer data, alerting the operator to potential problems, and enabling proactive maintenance and repair of the anemometer is also an important factor.PRIOR ART AND ITS DISADVANTAGES:A Chinese patent application no. CN202010354633A titled as "Wind turbine generator system anomaly detection method using data adversarial learning" depicts about a method for detecting anomalies in wind turbine generator systems using data adversarial learning, which essentially detects abnormal operating states of wind turbines. Said method includes SCADA (Supervisory Control and Data Acquisition) data preprocessing steps, constructing and training a generative adversarial network, checking the abnormal state of the SCADA data-driven unit large volumes, pre-processing of SCADA data being used for routine data collection, monitoring parameters, data used to train the constructed synthetic adversarial network. The trained generative adversarial network is used to detect anomalies in wind turbines.However, though accurate wind speed measurements are assumed for all historical data sets or real-time SCADA data, the aforementioned prior art fails to detect incorrect wind speed measurements from the sensor / anemometer. Said may result in the incorrect detection of an operational anomaly in the wind turbines because the measured wind speed can be biased or erroneous. Furthermore, the aforementioned previous art makes no mention of correction in the wind speed measurement data.Another Chinese patent application no. CN202311118587A titled as "Wind turbine SCADA fault early warning and positioning method based on AVAE_SDL" depicts about a method for wind turbine fault detection in gearboxes based on AVAE_SDL which comprises of steps: collecting SCADA data of the wind turbine gearbox, preprocessing the collected data; preprocessing the data into a training set and test set; constructing the AVAE_SDL model; input the training set into the AVAE_SDL model for training; obtaining the abnormal score of the optimization model and wind turbine fault degree, and determining the fault threshold. Based on the test set and the optimization model, obtaining the prediction data of the optimization model, drawing a fault warning map; based on the residual error of each input parameter in the SCADA data; predicting the data of the optimization model; fault warning map; and determining the fault location of the wind turbine gearbox.However, here also, though accurate wind speed measurements are assumed for all historical data sets or real-time SCADA data, the aforementioned prior art fails to detect incorrect wind speed measurements from the sensor / anemometer. Said may result in the incorrect detection of an operational anomaly in the wind turbines because the measured wind speed can be biased or erroneous. Furthermore, the aforementioned previous art makes no mention of correction in the wind speed measurement data.Another PCT application no. PCT / US2022 / 072973 titled as "Method and system for building prescriptive analytics to prevent wind turbine failures" suggests about systems and methods for building predictive and prescriptive analytics of wind turbines generate a historical operational dataset by loading historical operational SCADA data of one or more wind turbines. Each sensor measurement is associated with an engineering tag and at least one component of a wind turbine. The system creates one or more performance indicators corresponding to one or more sensor measurements and applies at least one data clustering algorithm onto the dataset to identify and label normal operation data clusters. The system builds a normal operation model using normal operational data clusters with the Efficiency of Wind-To-Power (EWTP) and defines a statistical confidence range around the normal operation model as the criterion for monitoring wind turbine performance. As real-time SCADA data is received by the system, the system can detect an anomalous event, and issue an alert notification and prescriptive early-action recommendations to a user, such as a turbine operator, technician, or manager. However, the aforementioned prior art fails to detect incorrect wind speed measurements from the sensor / anemometer. Said may result in the incorrect detection of an operational anomaly in the wind turbines because the measured wind speed can be biased or erroneous. Furthermore, the aforementioned previous art makes no mention of correction in the wind speed measurement data.Further, two non-patent literature titled as "Data-driven wind turbine power anomaly detection using SCADA" and "Wind turbine anomaly detection based on SCADA data mining" discuss about multiple instances where the measurement error of the anemometer leads to incorrect performance evaluation for the period.Another non-patent literature titled as "SCADA data for wind turbine data-driven condition / performance monitoring: A review on state-of-art, challenges, and future trends" suggests about ongoing challenges in performance monitoring. Said art also mentions errors in anemometer measurements leading to incorrect performance evaluation, however, said art completely fails to provide any teaching or technicalities over solving the same problem.One more non-patent literature titled as "Probabilistic anomaly detection approach for data-driven wind turbine condition monitoring" where condition monitoring of the wind turbines is suggested, however, fails to disclose about anomalies in wind speed measurements.DISADVANTAGES OF THE PRIOR ART:Said prior art suffers from at least all or any of the following disadvantages:All the prior art fails to suggest about detecting measurement anomalies in the wind speed measurement sensor / anemometer.All the prior art fails to suggest about detecting measurement anomalies in the wind speed measurement sensor / anemometer using a data-driven approach. All the prior art fails to suggest about either excluding incorrect wind speed measurements by the anemometer or correcting such erroneous records and thus may lead to incorrect detection of operational anomalies in the Wind turbines as the measured wind speed can be biased or erroneous. All the prior art fails to suggest about the correction in the wind speed measurement records.Therefore, the aforementioned drawbacks and limitations of the prior arts have been solved by the present invention efficiently and correctly.OBJECTS OF THE INVENTION:The main object of the present invention is to provide a system and method for detecting anomaly in the wind speed measurement by nacelle mounted anemometers.A principle object of the present invention is to provide a system and method for detecting anomalies in the wind speed measurement anemometer that improves the accuracy of wind speed measurement by correcting the windspeed data where erroneous measurements are suspected.Another object of the present invention is to provide a system and method for detecting anomalies in the wind speed measurement anemometer that reduces the number of false alarms and alerts by focusing on the data points where there is no suspected derating of the turbine.Yet another object of the present invention is to provide a system and method for detecting anomalies in the wind speed measurement anemometer that results in more accurate power generation forecasts and better performance benchmarking of wind turbines by detecting anomalies in wind speed measurement.The further object of the present invention is to provide a system and method for detecting and correcting anomalies in the wind speed measurement anemometer that overcomes the drawbacks of conventional or existing technology.BRIEF DESCRIPTION OF DRAWING 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:Figure 1: Shows an isometric view of a wind turbine having the placement of an anemometer over the wind turbine according to the present invention.Figure 1.1: Shows a flow chart of a system for detecting anomaly in the wind speed measurement anemometer according to the present invention.Figure 1.2: Shows a flow chart of a method for detecting anomaly in the wind speed measurement anemometer according to the present invention.Figure 2: Shows a flow chart of a data collection module according to the present invention.Figure 3: Shows a flow chart of a data filtration module according to the present invention.Figure 4: Shows a flow chart of an alert generation module according to the present invention.Figure 5: Shows a flow chart of a wind speed correction module according to the present invention.Figure 6: Shows a flow chart of a method of a data collection module according to the present invention.Figure 7: Shows a flow chart of a method of a data filtration module according to the present invention.Figure 8: Shows a flow chart of a method of an alert generation module according to the present invention.Figure 9: Shows a flow chart of a method of a wind speed correction module according to the present invention.Figures 10 and 11: Demonstrate the detected abnormal wind speeds and corrected 16% of data points within the time period of 1st January, 2023 to 30th June, 2023 according to the present invention, which was then confirmed by faulty anemometer by the site team. Figures 12 and 13: Indicate the post-replacement of the faulty anemometer within the time period of 1st July 2023 to 31st Dec 2023, the algorithm did not correct so many data points indicating the correctness and accuracy of the detection of anomalies in the wind speed measurement according to the present invention. Figures 14 and 15: Demonstrate the detected abnormal wind speeds and corrected 6.24 % of data points within the time period of 1st January, 2024 to 3rd May 2024 according to the present invention, which was then confirmed by a faulty anemometer by the site team. Figures 16 and 17: Indicate the post-replacement of the faulty anemometer within the time period of 4th May 2024 to onwards, the algorithm did not correct so many data points indicating the correctness and accuracy of the detection of anomalies in the wind speed measurement according to the present invention.SUMMARY OF THE INVENTION:The present invention relates to a system and method for detecting anomaly in the wind speed measurement that comprises of a data collection module to collect SCADA data from wind turbines, a data filtering module to filter out data points where the turbine is underperforming, a wind speed correction module to correct the wind speed for data points, and an alert generation module to initiate alerts. Thereby resulting in improved accuracy of wind speed measurement data, reduced number of false alarms and alerts by means of more accurate wind speed records of the turbine, accurate power generation forecasts, and better wind turbine performance benchmarking.LIST OF REFERENCE NUMERALSFoundation (1)Tower (2)Hub (3)Blades (4)Rotor (5)Nacelle (6)Anemometer (7)Data collection module (8) Data filtering module (9)Alert generation module (10) Wind speed correction module (11)DETAILED DESCRIPTION OF THE INVENTION:The following description is presented to enable any person skilled in the art to make and use the invention. It 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.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.The present invention provides a system and method for detecting anomaly in the wind speed measurement anemometer that detects inaccuracies and corrects the wind speed only for the data points without any suspected derating of the turbine. In addition, the present invention improves the accuracy of wind speed measurement by filtering out data points affected by turbine underperformance, curtailment, or other factors. Further, said system reduces the number of false alarms and alerts by evaluating wind turbine performance on the corrected windspeed points. This system thus is essential in more accurate power generation forecasts and better wind forecasts by detecting anomalies in wind speed measurement and providing correction on historical data.Now referring to Figure 1, a wind turbine system consists of a foundation (1), a tower (2), a hub (3), three blades (4), a rotor (5), and a nacelle (6). The turbine system having said foundation (1) is positioned below ground level. The system includes a visible "turret" section, with the majority of the structure being submerged in soil. The foundation (1) consists of a substantial and dense composite of concrete and steel mesh, serving to link the tower to the ground and to secure the turbine, thereby withstanding the forces exerted upon it.In accordance with Figure 1, the tower (2) is constructed using round tubular steel with a diameter ranging from 3-4m. The base of the tower is wider than the top to enhance stability, and thicker steel is utilized at the bottom to bolster strength. In a standard turbine, the walls at the base may have double the width of the top section. The height of the tower is of paramount importance for wind turbines due to the increase in wind speed at higher elevations. Further, the rotor (5), comprising three blades (4) and the central hub (3), forms the rotating component of the turbine. Wherein, said hub (3) is configured to securely hold three turbine blades (4) together, allowing rotational movement with respect to the rest of the turbine body, and the blades (4) are adjustable in pitch to optimize speed, reaction, and wind capture. Said nacelle (6) which is a structure located on top of the turbine tower (2) and houses various components, comprising - a generator (not shown) responsible for generating energy by converting the rotational kinetic energy of the rotor into electrical energy, - a controller configured to initiate the operation of the turbine at wind speeds, to cease the operation of the turbine when wind speeds exceed the predefined range, and to deactivate the turbine at higher wind speeds in order to prevent damage to various components of the turbine, and - an anemometer (7) is utilized to measure wind speed and relay said measured data to the controller for appropriate action.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.Now referring to Figures 1.1 and 2 to 5, a system for detecting anomaly in the wind speed measurement anemometer according to an exemplary embodiment is shown. The system for detecting anomaly in the wind speed measurement comprises - a data collection module (8) configured to collect SCADA data from wind turbines, including wind speed, power output, and other specified parameters,- a data filtering module (9) configured to filter out data points where the turbine is underperforming or curtailed by comparing the power output of the turbine with the expected power output based on the wind speed and other parameters,- a wind speed correction module (11) configured to correct the wind speed for data points where on the algorithm detects anomalies in the windspeed measurements based on the power output and other relevant parameters, and - an alert generation module (10) configured to initiate alerts or reports upon anomaly detection, providing information on the suspected causes of the anomalies in the anemometer, such as sensor bias or erroneous measurement.According to embodiments shown in Figure 2, said data collection module (8) consists of:S8.1: collection of plurality of sensors data located on the wind turbine; wherein said data is collected through the plurality of 4-20mA wires respectively,S8.2: subsequently transmission of collected data of the plurality of sensors data to a programmable logic controller (PLC) device of the generator (not shown),S8.3: process and preparation of said sensors data for a Supervisory Control and Data Acquisition (SCADA) device through the PLC device,S8.4: communication of the PLC device with the SCADA server through an application protocol over a fiber optic (FO) cable; wherein said application protocol may be a Modbus TCP / IP (Transmission Control Protocol and Internet Protocol) to store all sensor-related data to SCADA server,S8.5: link said SCADA device to a cloud server means; wherein said link of SCADA device may be done through employing either OPC-DA (Open Platform Communication -Data Access) or OPC-UA (Open Platform Communication - Unified Architecture) or any other protocol for communication depending on the SCADA devices,S8.6: transmission of data from said SCADA system through OPC to a cloud-based server through a computer network,S8.7: identification and correction of wind speed within the cloud-based server; wherein enables the system to detect irregularities, rectify them, store and utilize the corrected data for further analysis.Now, in the embodiments shown in Figure 3, said data filtering module (9) consists of:S9.1: selection of various parameters from said sensor data stored on the cloud server; wherein said data may comprise of plurality of measurement parameters like temperature, wind speed, active power, wind direction, and set forth,S9.2: transmission of said selected data to a pre-analytic engine; wherein said pre-analytic engine is configured to receive selected data and perform optimal data selection through statistical analysis techniques selected from, but not limited to thresholding, clustering, or machine learning algorithms; thereby ensuring that the most relevant data is transmitted for further processing,S9.3: subsequently, said pre-processed data is sent to an AI Kernel, which includes a feedback loop for hyperparameter tuning; wherein said hyperparameters are used to control the learning process and may be adjusted to optimize the performance of the AI model,S9.4: transmission of said processed data to a post-analytics engine for anomaly detection and data correction using statistical techniques, S9.5: identification of anomalies in the data and correction of the said anomalies, thereby ensuring the accuracy of the final output, S9.6: storage of said processed data and results in a cloud database; wherein said database is configured with the said pre-analytic engine, the post-analytics engine, and a user interface for displaying the results.According to the embodiments shown in Figure 5, said wind speed correction module (11) consists of:S11.1: selection of parameters f-rom sensor data, including temperature, wind speed, active power, and wind direction, through the SCADA,S11.2: process said data using optimal data selection techniques and statistical analysis methods in the pre-analytic engine, S11.3: employ the AI kernel for anomaly detection, and correct wind speed based on various thresholds, S11.4: tune the AI model's hyper-parameters for improved performance, S11.5: store the corrected wind speed in the database, and S11.6: display said corrections on a user interface for user understanding.Referring to the embodiments shown in Figure 4, said alert generation module (10) comprises: S10.1: retrieval of stored processed data by the post-analytics engine of said data filtering module (9) from the cloud database, S10.2: analyze the data using analytical techniques and historical data analysis through an alert generation engine to automatically generate an alert for anemometer anomalies; and S10.3: display alerted information on a user equipment interface and subsequently notify relevant parties through email.According to the exemplary embodiment of the present invention, as shown in Figures 1.2 and 6 to 9, a method for detecting anomaly in the wind speed measurement anemometer is described herein. The method for detecting anomaly in the wind speed measurement comprises - collecting (S1.1), SCADA data from wind turbines, including wind speed, power output, and specified parameters through a data collection module (8),- filtering out (S1.2), data points where the turbine is underperforming or curtailed by comparing the power output of the turbine with the expected power output based on the wind speed and other parameters through a data filtering module (9),- correcting (S1.3), the wind speed for data points where there is no suspected derating of the turbine using an AI-driven algorithm based on the power output and other relevant parameters through a wind speed correction module (11), and- initiating (S1.4), alerts or reports upon anomaly detection, providing information on the suspected causes of the anomalies, such as sensor bias or erroneous measurement through an alert generation module (10). According to embodiments shown in Figure 6, said data collection module (8) consists of:Step 2.1: collecting (S8.1), plurality of sensors data located on the wind turbine; wherein said data is collected through the plurality of 4-20mA wires respectively,Step 2.2: transmitting (S8.2), collected data of the plurality of sensors data to a programmable logic controller (PLC) device of the generator (not shown),Step 2.3: processing and preparing (S8.3), said sensors data for a Supervisory Control and Data Acquisition (SCADA) device through the PLC device,Step 2.4: communicating (S8.4), said PLC device with the SCADA device through an application protocol over a fiber optic (FO) cable; wherein said application protocol may be a Modbus TCP / IP (Transmission Control Protocol and Internet Protocol),Step 2.5: linking (S8.5), said SCADA device to a cloud server means; wherein said link of SCADA device may be done through employing either OPC-DA (Open Platform Communication -Data Access) or OPC-UA (Open Platform Communication - Unified Architecture) protocols for communication,Step 2.6: transmitting (S8.6), said data from said SCADA server means to a cloud-based server through a computer network, andStep 2.7: identifying and correcting (S8.7), said wind speed within the cloud-based server; wherein enables the system to detect irregularities, rectify them, and utilize the corrected data for further analysis.Now, in the embodiments shown in Figure 7, said data filtering module (9) consists of:Step 3.1: selecting (S9.1), various parameters from said sensor data; wherein said data may comprise of plurality of measurement parameters like temperature, wind speed, active power, wind direction, and set forth,Step 3.2: transmitting (S9.2), said selected data to a pre-analytic engine; wherein said pre-analytic engine is configured to receive selected data and perform optimal data selection through statistical analysis techniques selected from, but not limited to thresholding, clustering, or machine learning algorithms; thereby ensuring that the most relevant data is transmitted for further processing,Step 3.3: sending (S9.3), said pre-processed data to an AI Kernel, which includes a feedback loop for hyperparameter tuning; wherein said hyperparameters are used to control the learning process and may be adjusted to optimize the performance of the AI model,Step 3.4: transmitting (S9.4), said processed data to a post-analytics engine for anomaly detection and data correction using statistical techniques, Step 3.5: identifying (S9.5), said anomalies in the data and correcting the said anomalies, thereby ensuring the accuracy of the final output, Step 3.6: storing (S9.6), said processed data and results in a database; wherein said database is configured with the said pre-analytic engine, the post-analytics engine, and a user interface for displaying the results.According to the embodiments shown in Figure 9, said wind speed correction module (11) consists of:Step 5.1: selecting (S11.1), parameters from sensor data, including temperature, wind speed, active power, and wind direction, through the SCADA,Step 5.2: processing (S11.2), said data using optimal data selection techniques and statistical analysis methods in the pre-analytic engine, Step 5.3: employing (S11.3), the AI kernel for anomaly detection, and correcting wind speed based on various thresholds, Step 5.4: tuning (S11.4), said AI model's hyper-parameters for improved performance, Step 5.5: storing (S11.5), said corrected wind speed in the database, and Step 5.6: displaying (S11.6), said corrections on a user interface for user understanding.Referring to the embodiments shown in Figure 8, said alert generation module (10) comprises: Step 4.1: retrieving (S10.1), said stored processed data by the post-analytics engine of said data filtering module (9) from the database, Step 4.2: analyzing (S10.2), said data using analytical techniques and historical data analysis through an alert generation engine to generate an alert for anemometer anomalies automatically, and Step 4.3: displaying (S10.3), alerted information on a user equipment interface and subsequently notifying relevant parties through email.The system and method for detecting anomalies in wind speed measurement using SCADA data from wind turbines monitors the performance of wind turbines and detects anomalies in wind speed measurement. Said improves the accuracy of wind speed measurement, reduces the number of false alarms and alerts, and provides more accurate power generation forecasts and better wind turbine maintenance.The foregoing is considered as illustrative only of the principles of the invention. Further, since numerous modifications and changes will readily occur to those skilled in the art, it is not desired to limit the invention to the exact construction and operation shown and described, and accordingly, all suitable modifications and equivalents may be resorted to, falling within the scope of the invention.Experimental results:According to the embodiments described herein above, the system has undergone numerous experiments. Said experiment was conducted to understand how much correction is provided by the system when the anemometer is faulty vs when the anemometer is not faulty. In the desired outcome, the amount of correction should be negligible for healthy anemometer and considerable correction should take place for faulty anemometer. The graphical representation of said test results is shown in Figures 10 to 17.- Referring to Figures 10 and 11, a. depicts a graphical representation of wind speed (without providing the present system) vs the active power in the duration of 1st Jan, 2023 to 30th June, 2023, with a known anemometer issue.b. depicts a graphical representation of wind speed (providing the present system) vs the active power in the duration of 1st Jan, 2023 to 30th June, 2023. Thus, shows a 16% correction in the data from the said period confirming the anomaly in windspeed measurements due to the faulty anemometer.- Referring to Figures 12 and 13, a. depicts a graphical representation of wind speed (without providing the present system) vs the active power in the duration of 1st July, 2023 to 31st Dec, 2023 post replacement of faulty anemometer referred earlier.b. depicts a graphical representation of wind speed (providing the present system) vs the active power in the duration of 1st July, 2023 to 31st Dec, 2023. Thus, shows a 3.6% correction in the data points of the said period, confirming that the newly installed anemometer is measuring windspeeds accurately.- Referring to Figures 14 and 15,a. depicts a graphical representation of wind speed (without providing the present system) vs the active power in the duration of 1st Jan, 2024 to 3rd May, 2024, with a known anemometer issue.b. depicts a graphical representation of wind speed (providing the present system) vs the active power in the duration of 1st Jan, 2024 to 3rd May, 2024. Thus, shows a 6.24% correction in data from the said period confirming the anomaly in windspeed measurements due to the faulty anemometer.- Referring to Figures 16 and 17,a. depicts a graphical representation of wind speed (without providing the present system) vs the active power in the duration of 4th May, 2024 to onwards, post replacement of faulty anemometer referred earlier.b. depicts a graphical representation of wind speed (providing the present system) vs the active power in the duration of 4th May, 2024 to onwards. Thus, shows a 0.5% correction in the data points of the said period, confirming that the newly installed anemometer is measuring windspeeds accurately.ADVANTAGES OF THE INVENTION:The system and method for detecting anomalies in wind speed measurement provide several advantages over existing systems and methods for wind turbine performance monitoring and wind speed measurement. Said advantages are as follows:- Improve the accuracy of wind speed measurement by filtering out data points affected by turbine underperformance, curtailment, or other factors.- Reduce the number of false alarms and alerts by focusing on the data points where there is no suspected derating of the turbine.- Correct historically recorded data points to have a more accurate record of historical performance and wind regime.- Provide more accurate power generation forecasts and better wind performance benchmarking by detecting anomalies in wind speed measurement.

Claims

1. A system for detecting anomaly in the wind speed measurement comprises - a data collection module (8) configured to collect SCADA data from wind turbines, including wind speed, power output, and other specified parameters, - a data filtering module (9) configured to filter out data points where the turbine is underperforming or curtailed by comparing the power output of the turbine with the expected power output based on the wind speed and other parameters, - a wind speed correction module (11) configured to correct the wind speed for data points where the algorithm detects anomalies in the windspeed measurements based on the power output and other relevant parameters, and - an alert generation module (10) configured to initiate alerts or reports upon anomaly detection, providing information on the suspected causes of the anomalies in the anemometer, such as sensor bias or erroneous measurement.

2. The system for detecting anomaly in the wind speed measurement as claimed in claim 1, wherein said data collection module (8) consists of: - collection (S8.1) of plurality of sensors data located on the wind turbine; wherein said data is collected through the plurality of 4-20mA wires respectively, - transmission (S8.2) of collected data of the plurality of sensors data to a programmable logic controller (PLC) device of the generator, - process and preparation (S8.3) of said sensors data for a Supervisory Control and Data Acquisition (SCADA) device through the PLC device, - communication (S8.4) of the PLC device with the SCADA device through an application protocol over a fiber optic (FO) cable; wherein said application protocol may be a Modbus TCP / IP (Transmission Control Protocol and Internet Protocol) to store all sensor-related data to SCADA server, - link (S8.5) said SCADA device to a cloud server means; wherein said link of SCADA device may be done through employing either OPC-DA (Open Platform Communication -Data Access) or OPC-UA (Open Platform Communication - Unified Architecture) or any other protocol for communication depending on the SCADA devices, - transmission (S8.6) of data from said SCADA system through OPC to a cloud-based server through a computer network, - identification and correction (S8.7) of wind speed within the cloud-based server.

3. The system for detecting anomaly in the wind speed measurement as claimed in claim 1, wherein said data filtering module (9) consists of: - selection (S9.1) of plurality of measurement parameters like temperature, wind speed, active power, wind direction, and set forth from said sensor data stored on the cloud server, - transmission (S9.2) of said selected data to a pre-analytic engine; wherein said pre-analytic engine is configured to receive selected data and perform optimal data selection through statistical analysis techniques selected from, but not limited to thresholding, clustering, or machine learning algorithms ensuring that the most relevant data is transmitted for further processing, - send (S9.3) said pre-processed data to an AI Kernel, which includes a feedback loop for hyperparameter tuning; wherein said hyperparameters are used to control the learning process and may be adjusted to optimize the performance of the AI model, - transmission (S9.4) of said processed data to a post-analytics engine for anomaly detection and data correction using statistical techniques, - identification (S9.5) of anomalies in the data and correction of the said anomalies, and - storage (S9.6) of said processed data and results in a database; wherein said database is configured with the said pre-analytic engine, the post-analytics engine, and a user interface for displaying the results.

4. The system for detecting anomaly in the wind speed measurement as claimed in claim 1, wherein said wind speed correction module (11) consists of: - selection (S11.1) of parameters from sensor data, including temperature, wind speed, active power, and wind direction, through the SCADA, - process (S11.2) said data using optimal data selection techniques and statistical analysis methods in the pre-analytic engine, - employ (S11.3) the AI kernel for anomaly detection, and correct wind speed based on various thresholds, - tune (S11.4) the AI model's hyper-parameters for improved performance, - store (S11.5) the corrected wind speed in the database, and - display (S11.6) said corrections on a user interface for user understanding.

5. The system for detecting anomaly in the wind speed measurement as claimed in claim 1, wherein said alert generation module (10) comprises: - retrieval (S10.1) of stored processed data by the post-analytics engine of said data filtering module (9) from the cloud database, - analyze (S10.2) the data using analytical techniques and historical data analysis through an alert generation engine to automatically generate an alert for anemometer anomalies; and - display (S10.3) alerted information on a user equipment interface and subsequently notify relevant parties through email.

6. A method for detecting anomaly in the wind speed measurement comprises of following steps: - collecting (S1.1), SCADA data from wind turbines, including wind speed, power output, and specified parameters through a data collection module (8), - filtering out (S1.2), data points where the turbine is underperforming or curtailed by comparing the power output of the turbine with the expected power output based on the wind speed and other parameters through a data filtering module (9), - correcting (S1.3), the wind speed for data points where there is no suspected derating of the turbine using an AI-driven algorithm based on the power output and other relevant parameters through a wind speed correction module (11), and - initiating (S1.4), alerts or reports upon anomaly detection, providing information on the suspected causes of the anomalies, such as sensor bias or erroneous measurement through an alert generation module (10).

7. The method for detecting anomaly in the wind speed measurement as claimed in claim 6, wherein said data collection module (8) consists of the following steps: - collecting (S8.1), plurality of sensors data through the plurality of 4-20mA wires located on the wind turbine, - transmitting (S8.2), collected data of the plurality of sensors data to a programmable logic controller (PLC) device of the generator, - processing and preparing (S8.3), said sensors data for a Supervisory Control and Data Acquisition (SCADA) device through the PLC device, - communicating (S8.4), said PLC device with the SCADA device through an application protocol over a fiber optic (FO) cable, - linking (S8.5), said SCADA device to a cloud server means, - transmitting (S8.6), said data from said SCADA server means to a cloud-based server through a computer network, and - identifying and correcting (S8.7), said wind speed within the cloud-based server.

8. The method for detecting anomaly in the wind speed measurement as claimed in claim 6, wherein said data filtering module (9) consists of the following steps: - selecting (S9.1), plurality of measurement parameters like temperature, wind speed, active power, and wind direction, and set forth from said sensor data stored on the cloud server, - transmitting (S9.2), said selected data to a pre-analytic engine, - sending (S9.3), said pre-processed data to an AI Kernel, which includes a feedback loop for hyperparameter tuning, - transmitting (S9.4), said processed data to a post-analytics engine for anomaly detection and data correction using statistical techniques, - identifying (S9.5), said anomalies in the data and correcting the said anomalies, - storing (S9.6), said processed data and results in a database.

9. The method for detecting anomaly in the wind speed measurement as claimed in claim 6, wherein said wind speed correction module (11) consists of the following steps: - selecting (S11.1), parameters from sensor data, including temperature, wind speed, active power, and wind direction, through the SCADA, - processing (S11.2), said data using optimal data selection techniques and statistical analysis methods in the pre-analytic engine, - employing (S11.3), the AI kernel for anomaly detection, and correcting wind speed based on various thresholds, - tuning (S11.4), said AI model's hyper-parameters for improved performance, - storing (S11.5), said corrected wind speed in the database, and - displaying (S11.6), said corrections on a user interface for user understanding.

10. The method for detecting anomaly in the wind speed measurement as claimed in claim 6, wherein said alert generation module (10) comprises of following steps: - retrieving (S10.1), said stored processed data by the post-analytics engine of said data filtering module (9) from the database, - analyzing (S10.2), said data using analytical techniques and historical data analysis through an alert generation engine to generate an alert for anemometer anomalies automatically, and - displaying (S10.3), alerted information on a user equipment interface, and notifying relevant parties through email.