Abnormality detection method and system for dispatching electric power through flywheel energy storage

Abstract

The invention relates to the technical field of power supply systems, in particular to an anomaly detection method and system for dispatching electric power through flywheel energy storage, and aims to solve the problem of how to improve the prediction accuracy of generated power according to different terrains in regions with large topographic changes. The embodiment of the invention provides a detection method, which comprises the following steps: acquiring historical power data of a wind turbine generator, and overall historical meteorological data and actual historical meteorological data corresponding to the historical power data, calculating a power influence factor according to the actual historical meteorological data and the historical power data, and establishing a power prediction model; and calculating an electric quantity supply difference value in the future time period according to the predicted power consumption data and the predicted deviation coefficient, and adjusting a power supply mode in the target area according to the electric quantity supply difference value.

Description

Technical Field

[0001] This invention relates to the field of power supply system technology, and more specifically, to a method and system for detecting anomalies in dispatched power through flywheel energy storage. Background Technology

[0002] Wind power generation is a common new energy power generation technology. It boasts high power output and is a clean energy source, making it an excellent power generation method. However, due to the significant fluctuations in wind power generation, when wind power accounts for a large proportion of the power grid, this volatility can lead to large grid fluctuations and unstable power supply. During the grid connection of wind power generation, to ensure stable power supply, it is necessary to obtain the fluctuation characteristics of wind power generation through power generation forecasting and optimize power dispatch. However, existing forecasting methods fail to adequately consider the issue of insufficient forecast accuracy caused by terrain variations. Therefore, improving the forecast accuracy of power generation based on different terrain conditions is a key approach to ensuring stable power supply to the grid. Summary of the Invention

[0003] The problem this invention addresses is: how to improve the accuracy of power generation prediction based on different terrains in areas with significant topographical variations.

[0004] To address the aforementioned issues, this invention provides a method for detecting power anomalies in a flywheel energy storage system. The method includes: acquiring historical power data of wind turbine generators, corresponding overall historical meteorological data, and actual historical meteorological data; calculating a power influence factor based on the actual historical meteorological data and historical power data, and establishing a power prediction model; monitoring wind turbine generators in real-time during a target time period to obtain real-time power data and corresponding real-time meteorological data; calculating theoretical power data of the wind turbine generators based on the real-time meteorological data and the power prediction model; correcting the power prediction model based on the theoretical power data and real-time power data to obtain a corrected prediction model; predicting meteorological data for a future time period, denoted as predicted meteorological data, and calculating a prediction deviation coefficient based on the predicted meteorological data and the corrected prediction model; selecting historical time periods corresponding to the future time period from a database and obtaining historical electricity consumption data; predicting electricity consumption for the future time period based on the historical electricity consumption data to obtain predicted electricity consumption data; calculating the electricity supply difference for the future time period based on the predicted electricity consumption data and the prediction deviation coefficient; and adjusting the power supply method within the target area based on the electricity supply difference.

[0005] Compared with existing technologies, the technical effects achieved by this solution are as follows: Power influence factors are calculated based on actual historical meteorological and power data, and a power prediction model is established. This fully considers the impact of different meteorological factors on the power generation of wind turbines, making the power prediction of wind turbines more consistent with reality. The power prediction model is then corrected based on theoretical and real-time power data to obtain a revised prediction model, which fully considers the influence of environmental and non-meteorological factors, making the power prediction of wind turbines more reasonable. Furthermore, the setting of the prediction deviation coefficient takes into account the inherent errors of meteorological factors and the revised prediction model itself, further improving the accuracy of wind turbine power prediction.

[0006] In one embodiment of the present invention, historical power data of wind turbines, as well as corresponding overall historical meteorological data and actual historical meteorological data, are obtained. A power influence factor is calculated based on the actual historical meteorological data and historical power data, and a power prediction model is established. Specifically, this includes: obtaining historical power data of each wind turbine at different geographical locations within a target area; determining the influence coefficient of geographical location on actual historical meteorological data based on the historical power data and the geographical location of the wind turbines; correcting the influence coefficient based on the actual historical meteorological data and overall historical meteorological data to obtain a correction coefficient; and calculating the power influence factor based on the actual historical meteorological data and historical power data of the wind turbines, and establishing a power prediction model.

[0007] Compared with existing technologies, the technical effects achieved by adopting this technical solution are as follows: the influence coefficient of geographical location on actual historical meteorological data is determined based on historical power data and the geographical location of wind turbines, the impact of different terrains on meteorological data is fully considered, and the acquisition of theoretical historical meteorological data helps to improve the prediction accuracy of subsequent meteorological data.

[0008] In one embodiment of the present invention, historical power data of each wind turbine at different geographical locations are obtained within a target area. The influence coefficient of geographical location on actual historical meteorological data is determined based on the historical power data and the geographical location of the wind turbine. Specifically, this includes: dividing the wind turbine into different associated groups according to its model and years of service; obtaining theoretical power data based on the technical parameters of the wind turbines in the same associated group and the overall historical meteorological data; and calculating the influence coefficient of geographical location on actual historical meteorological data based on the theoretical power data and the historical power data.

[0009] Compared with existing technologies, the technical effects achieved by adopting this technical solution are as follows: wind turbines are divided into different associated groups according to their model and service life. Theoretical power data are obtained based on the technical parameters of wind turbines in the same associated group and overall historical meteorological data. This fully considers the impact of different models or service lives on the power data of wind turbines and improves the accuracy of the influence coefficient.

[0010] In one embodiment of the present invention, the influence coefficient is corrected based on actual historical meteorological data and overall historical meteorological data to obtain a correction coefficient. Specifically, this includes: calculating theoretical historical meteorological data for different geographical locations based on overall historical meteorological data and influence coefficient; and correcting the influence coefficient based on actual historical meteorological data and theoretical historical meteorological data to obtain a correction coefficient.

[0011] Compared with existing technologies, the technical effects achieved by adopting this technical solution are as follows: the influence coefficient is corrected based on actual historical meteorological data and theoretical historical meteorological data to obtain a corrected coefficient. This fully considers the inaccuracies of the influence coefficient caused by measurement errors, local environmental influences, instrument errors, and other factors such as climate change, making the obtained corrected coefficient more consistent with the actual situation and improving the accuracy of the corrected coefficient.

[0012] In one embodiment of the present invention, the meteorological data for a future time period is predicted and denoted as the predicted meteorological data. The prediction deviation coefficient is calculated based on the predicted meteorological data and the modified prediction model. Specifically, this includes: calculating the theoretical meteorological data of each wind turbine based on the predicted meteorological data and the modification coefficient; obtaining the predicted power data for a future time period based on the theoretical meteorological data and the modified prediction model; monitoring the power data for a future time period; and obtaining the prediction deviation coefficient based on the power data and the predicted power data.

[0013] Compared with existing technologies, the technical effects achieved by adopting this technical solution are as follows: The theoretical meteorological data of each wind turbine is calculated based on the predicted meteorological data and correction coefficients, which fully considers the actual impact of different geographical locations on the meteorological data, making the obtained predicted power data more consistent with the actual situation. Power data in the future time period is monitored, and the prediction deviation coefficient is obtained based on the power data and the predicted power data. The performance of the wind turbine is fully considered on the predicted power data, thus improving the accuracy of the predicted power data.

[0014] In one embodiment of the present invention, the method of calculating the power supply difference in a future time period based on the predicted power consumption data and the prediction deviation coefficient, and adjusting the power supply mode in the target area based on the power supply difference, specifically includes: calculating the predicted power generation of the wind turbine in a future time period based on the prediction deviation coefficient; calculating the power supply difference in a future time period based on the predicted power generation and the predicted power consumption data; obtaining an alarm threshold based on the power supply difference; and adjusting the power supply mode in the target area based on the alarm threshold.

[0015] Compared with existing technologies, the technical effects achieved by adopting this technical solution are as follows: alarm thresholds are obtained based on the power supply difference, and the power supply method in the target area is adjusted according to the alarm thresholds, thereby improving response efficiency and stabilizing the power supply system.

[0016] In one embodiment of the present invention, an alarm threshold is obtained based on the power supply difference, and the power supply method in the target area is adjusted according to the alarm threshold. Specifically, this includes: real-time monitoring of the power consumption data of the working cable in the future time period to obtain real-time power consumption data, and real-time monitoring of the real-time power generation of wind turbines at various locations; when the power supply difference between the real-time power consumption data and the real-time power generation is less than the alarm threshold, the usage time of the energy storage battery is calculated based on the power supply difference between the real-time power consumption data and the real-time power generation; when the power supply difference between the real-time power consumption data and the predicted power consumption data is greater than or equal to the alarm threshold, the predicted power consumption data in the target area is reduced based on the power supply difference, and the usage time of the energy storage battery is planned.

[0017] Compared with existing technologies, the technical effects achieved by this solution are as follows: When the difference between real-time power consumption data and real-time power generation is less than the alarm threshold, the usage time of the energy storage battery is calculated based on the difference, fully considering the limited energy storage capacity of the energy storage battery, thereby achieving efficient resource utilization. When the difference between real-time power consumption data and real-time power generation is greater than or equal to the alarm threshold, the predicted power consumption data in the target area is reduced based on the power supply difference, and the usage time of the energy storage battery is planned, fully considering the possibility of insufficient power supply in special circumstances, thus improving power supply stability.

[0018] In one embodiment of the present invention, the present invention also provides a detection system for abnormal power dispatch monitoring data. The detection method described in the above embodiment is applied to the detection system. The detection system includes: a storage module: a database is stored in the storage module; an acquisition module: the acquisition module is used to acquire meteorological data; a calculation module: the calculation module is used to calculate theoretical power data and prediction deviation coefficients; and a judgment module: the judgment module is used to determine whether the power supply mode in the target area needs to be adjusted. The detection system has all the technical features of the above detection method, which will not be described in detail here. Attached Figure Description

[0019] Figure 1 This is one of the flowcharts for the anomaly detection method of power dispatching using flywheel energy storage according to the present invention; Figure 2 This is the second flowchart of the method for detecting abnormalities in power dispatching using flywheel energy storage according to the present invention; Figure 3 This is the third flowchart of the method for detecting anomalies in power dispatching using flywheel energy storage according to the present invention; Figure 4 This is the fourth flowchart of the method for detecting abnormalities in power dispatching using flywheel energy storage according to the present invention; Figure 5 This is the fifth flowchart of the method for detecting abnormalities in power dispatching using flywheel energy storage according to the present invention; Figure 6 This is a schematic diagram of the detection system of the present invention; Explanation of reference numerals in the attached figures: 100 - Detection system; 110 - Storage module; 120 - Acquisition module; 130 - Calculation module; 140 - Judgment module. Detailed Implementation

[0020] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0021] [First Embodiment] See Figure 1 This invention provides an anomaly detection method for dispatched power using flywheel energy storage, the detection method comprising: S100. Obtain historical power data of wind turbines, as well as overall historical meteorological data and actual historical meteorological data corresponding to the historical power data. Calculate the power impact factor based on the actual historical meteorological data and historical power data, and establish a power prediction model. S200. During the target time period, the wind turbine is monitored in real time to obtain real-time power data and real-time meteorological data at the corresponding time. The theoretical power data of the wind turbine is calculated based on the real-time meteorological data and the power prediction model. S300. The power prediction model is corrected based on theoretical power data and real-time power data to obtain the corrected prediction model; meteorological data for the future time period is predicted and recorded as predicted meteorological data; the prediction deviation coefficient is calculated based on the predicted meteorological data and the corrected prediction model. S400. Filter out the historical time periods corresponding to the future time period from the database and obtain historical electricity consumption data. Based on the historical electricity consumption data, predict the electricity consumption for the future time period and obtain predicted electricity consumption data. S500: Calculate the power supply difference in the future time period based on the predicted power consumption data and the prediction deviation coefficient, and adjust the power supply mode in the target area according to the power supply difference.

[0022] In step S100, the overall historical meteorological data refers to the historical meteorological data of the area where the wind turbine is located, while the actual historical meteorological data refers to the meteorological data recorded by a single wind turbine. The historical power data of the wind turbine is very important for evaluating the performance, operation and future planning of the wind turbine. Generally speaking, the historical power data of the wind turbine includes actual output power, cycle power generation, rated power, maximum power and system efficiency. The power influence factor can be obtained by using machine learning models and multiple regression analysis to measure the influence of various characteristics of meteorological data on power. The power prediction model can use a traditional statistical model.

[0023] In step S200, meteorological monitoring stations are typically established within the wind farm. These stations are equipped with various meteorological instruments, such as anemometers, wind vanes, temperature sensors, humidity sensors, and barometric pressure sensors, to collect and record meteorological data in real time. In addition, the wind turbine itself is also equipped with built-in sensors and monitoring systems to record meteorological data related to the operation of the wind turbine in real time. This data includes wind speed, wind direction, temperature, humidity, etc., which are recorded in real time in the turbine's control system and stored in a database. The real-time meteorological data is then substituted into the power prediction model to obtain theoretical power data.

[0024] In step S300, power prediction models may have problems such as overfitting, insufficient feedback mechanisms, and lack of dynamic correction in practical applications. For example, some models fail to be updated and adjusted according to real-time data after deployment. As weather patterns, equipment aging, or other external factors change, the models may gradually lose their prediction accuracy. Therefore, methods such as weighted correction, incremental correction, and model fusion can be used to correct the power prediction model. For example, the weighted average method can be used to combine the output values ​​of two models, and the weights are dynamically adjusted according to the error between real-time power data and theoretical power data.

[0025] It should be noted that there will be some error between the predicted meteorological data and the actual meteorological data, and there may also be some error in correcting the prediction model when calculating the power data of wind turbines. Therefore, it is necessary to calculate the prediction deviation coefficient in order to better predict the power data of wind turbines.

[0026] In step S400, electricity consumption data generally includes total electricity consumption, real-time electricity consumption, and voltage and current data. Typically, residents' electricity consumption is closely related to seasonal changes. Therefore, it is necessary to predict the electricity consumption data for the future time period by analyzing the changes in electricity consumption data for historical time periods corresponding to the future time period. The corresponding historical time period refers to a time period with the same month as the future time period but a different year. For example, to predict the electricity consumption data of a certain area in March 2025, it is necessary to obtain the electricity consumption data of the area in March for three consecutive years. Based on the trend of the electricity consumption data in March for three consecutive years, the electricity consumption data in March can be predicted.

[0027] In step S500, the power data of the wind turbines in the future time period is obtained based on the predicted meteorological data and the revised prediction model. The power data of the wind turbines in the future time period is corrected by the prediction deviation coefficient. For example, if the wind turbines generate 100 MWh on May 3, 2025, based on the predicted meteorological data and the revised prediction model, and the prediction deviation coefficient is 0.01, then the predicted power generation of the wind turbines on May 3, 2025, after correction, is 100 + 100 × 0.01 = 101 MWh. If the electricity consumption in a certain area is 120 MWh on May 3, 2025, then the power input other than wind power generation should be 19 MWh.

[0028] Power influence factors are calculated based on actual historical meteorological and power data, and a power prediction model is established. This model fully considers the impact of different meteorological factors on the power generation of wind turbines, making the power prediction of wind turbines more consistent with reality. The power prediction model is then corrected based on theoretical power data and real-time power data to obtain a modified prediction model. This modified model fully considers the influence of environmental and non-meteorological factors, making the power prediction of wind turbines more reasonable. In addition, the setting of the prediction deviation coefficient takes into account the prediction of meteorological factors and the inherent error of the modified prediction model itself, further improving the accuracy of the power prediction of wind turbines.

[0029] [Second Embodiment] See Figure 2 In one specific embodiment, historical power data of the wind turbine, along with corresponding overall historical meteorological data and actual historical meteorological data, are acquired. Based on the actual historical meteorological data and historical power data, a power impact factor is calculated, and a power prediction model is established. Specifically, this includes: S110. Obtain historical power data of each wind turbine in different geographical locations within the target area, and determine the influence coefficient of geographical location on actual historical meteorological data based on the historical power data and the geographical location of the wind turbine. S120. The influence coefficient is corrected based on actual historical meteorological data and overall historical meteorological data to obtain the correction coefficient; S130. Calculate the power influence factor based on the actual historical meteorological data and historical power data of the wind turbine, and establish a power prediction model.

[0030] In step S110, the target area can be a small wind farm or an industrial park with wind turbines installed. Under the same meteorological data, the meteorological data of different geographical locations will have certain differences due to the existence of terrain or obstructions. Small wind turbines are more susceptible to the influence of local terrain in complex terrain because of their relatively low height. For example, in valleys, due to the obstruction of mountains on both sides, they can only obtain good wind speeds at specific wind directions. At other times, the wind speed is low or even turbulence is formed. In areas with obstacles such as buildings or trees, small wind turbines will also experience a significant drop in wind speed due to the obstruction of obstacles, resulting in a reduction in power generation efficiency. In complex mountainous areas, medium-sized wind turbines may experience uneven stress on the blades due to the vertical shear of wind direction and speed at different heights. This not only affects the power generation efficiency but may also increase equipment fatigue wear and reduce the service life of the equipment.

[0031] Furthermore, for wind turbines built in the early stages, urban development may cause them to be blocked by buildings in a certain direction. These wind turbines are often built on the top floor of residential areas. Over time, various buildings have been added around the wind turbines, resulting in differences in power generation when the wind turbines face different wind directions.

[0032] In step S120, generally speaking, the accuracy of meteorological data prediction in the spatial range is several hundred meters to several kilometers. Therefore, even under the same meteorological data, the meteorological data received by different wind turbines are different. For example, if the influence coefficient of a certain geographical location on the meteorological data is 0.95 and the overall wind speed is 10, then the wind speed at this geographical location is 10 × 0.95 = 9.5.

[0033] Based on historical power data and the geographical location of wind turbines, the influence coefficient is made more consistent with the working environment of wind turbines. By obtaining the difference between actual historical meteorological data and overall historical meteorological data, the impact of different terrains on meteorological data is fully considered. The establishment of the power prediction model can correct the calculation method of power generation as the power generation environment changes, which helps to improve the prediction accuracy of power generation.

[0034] [Third Embodiment] In one specific embodiment, historical power data of each wind turbine at different geographical locations within the target area are acquired. The influence coefficient of geographical location on actual historical meteorological data is determined based on the historical power data and the geographical locations of the wind turbines. Specifically, this includes: S111. Based on the model and service life, wind turbines are divided into different associated groups. The theoretical power data is obtained based on the technical parameters of wind turbines in the same associated group and the overall historical meteorological data. S112. Calculate the influence coefficient of geographical location on actual historical meteorological data based on theoretical power data and historical power data.

[0035] In step S111, the power data of the wind turbine is related to its model and loss status. Generally, the loss status of the wind turbine is linearly related to its service life. Therefore, the wind turbines are divided into different association groups according to their model and service life. The theoretical power data of the wind turbines in the same association group is calculated. The technical parameters can be obtained from the official website of the wind turbine manufacturer, industry reports, third-party testing institutions, etc. The overall historical meteorological data can be obtained from meteorological departments or institutions or other meteorological data suppliers. The theoretical power data of the wind turbine at a specific time can be calculated by using the technical parameters of the wind turbine, the overall historical meteorological data, and the power curve provided by the wind turbine.

[0036] In step S112, generally speaking, under the same conditions, the influence coefficient of the same geographical location on different meteorological data is usually different. Different surface roughness will lead to significant differences in wind speed and turbulence in different wind directions. The overall historical meteorological data for three consecutive years is obtained to obtain the variation pattern of the overall historical meteorological data of this region. When the wind turbine is installed in the monsoon climate region, the wind direction in the same season usually does not change significantly. However, in the non-monsoon climate region or under specific conditions, the wind direction may change significantly due to the influence of various factors. Based on different overall historical meteorological data and theoretical power data under different overall historical meteorological data, the influence coefficient of geographical location on the actual historical meteorological data is calculated using models such as multiple linear regression model, random forest and neural network. For example, if the theoretical power data of a wind turbine is 10MWh and the historical power data is 9MWh, then the influence coefficient of geographical location on the actual historical meteorological data is 9÷10=0.9.

[0037] Wind turbines are divided into different association groups based on their model and service life. Theoretical power data are obtained from the technical parameters of wind turbines in the same association group and the overall historical meteorological data. This fully considers the impact of different models or service lives on the power data of wind turbines and improves the accuracy of the influence coefficient.

[0038] [Fourth Embodiment] In one specific embodiment, the influence coefficient is corrected based on actual historical meteorological data and overall historical meteorological data to obtain a correction coefficient, specifically including: S121. Calculate the theoretical historical meteorological data for different geographical locations based on the overall historical meteorological data and the influence coefficient; S122. The influence coefficient is corrected based on actual historical meteorological data and theoretical historical meteorological data to obtain the correction coefficient.

[0039] In steps S121 and S122, theoretical meteorological data for different geographical locations are obtained based on the overall historical meteorological data and the calculation model in step S112. Taking wind speed as an example, the overall wind speed in a certain area is 10 m / s, and the influence coefficient of the target geographical location on the wind speed in this area is 1.1. Therefore, the theoretical wind speed corresponding to the target geographical location is 10 × 1.1 = 11 m / s.

[0040] Due to factors such as measurement errors, local environmental influences, instrument errors, and climate change, there are certain differences between theoretical historical meteorological data and actual historical meteorological data. By comparing actual historical meteorological data with theoretical historical meteorological data, we can identify the differences between the two, analyze the causes of the differences, such as instrument errors, differences in observation conditions, and different data processing methods, and obtain correction coefficients through statistical models or machine learning models.

[0041] The influence coefficients are corrected based on actual and theoretical historical meteorological data to obtain corrected coefficients. This process fully considers the inaccuracies caused by measurement errors, local environmental influences, instrument errors, and other factors such as climate change, making the corrected coefficients more consistent with the actual situation and improving their accuracy.

[0042] [Fifth Embodiment] See Figure 3 In one specific embodiment, meteorological data for a future time period is predicted, denoted as predicted meteorological data. A prediction bias coefficient is calculated based on the predicted meteorological data and a revised prediction model, specifically including: S310. Calculate the theoretical meteorological data for each wind turbine based on the predicted meteorological data and correction coefficients. S320. Obtain the predicted power data for the future time period based on theoretical meteorological data and the modified prediction model; S330: Monitor power data for a future time period and obtain the prediction deviation coefficient based on the power data and the predicted power data.

[0043] In steps S310 to S330, forecast meteorological data for the future time period can be obtained through the official meteorological data platform. Taking wind direction as an example, if the overall wind direction of a specific area is 90° and the correction coefficient corresponding to the target location within the specific area is 0.95, then the wind direction corresponding to the target location should be 90 × 0.95 = 85.5°.

[0044] By substituting theoretical meteorological data into the modified prediction model, the predicted power data can be obtained. For example, for a small wind turbine, the theoretical meteorological data for a certain geographical location are: wind speed 15 m / s, wind direction 83°, air density 1.225 kg / m³, temperature 15℃, and air pressure 1013.25 hPa. Substituting these data into the modified prediction model, the predicted power data is 15.66 kW.

[0045] Typically, due to differences in the performance of wind turbines, there will be some discrepancy between the predicted power data and the real-time monitored power data of the wind turbines in the future. Therefore, it is necessary to obtain a prediction deviation coefficient based on the power data and the predicted power data. For example, if the power data is 15 kWh and the predicted power data is 16 kWh, then the prediction deviation coefficient is 15 ÷ 16 = 0.9375.

[0046] The theoretical meteorological data for each wind turbine is calculated based on forecast meteorological data and correction coefficients. This fully considers the actual impact of different geographical locations on meteorological data, making the obtained forecast power data more consistent with the actual situation. Power data in the future time period is monitored, and a prediction deviation coefficient is obtained based on the power data and the forecast power data. This fully considers the impact of wind turbine performance on forecast power data, thus improving the accuracy of forecast power data.

[0047] [Sixth Embodiment] See Figure 4 In one specific embodiment, the power supply difference for a future time period is calculated based on predicted power consumption data and a prediction deviation coefficient. The power supply method within the target area is then adjusted according to this difference. Specifically, this includes: S510. Calculate the predicted power generation of the wind turbine in the future time period based on the prediction deviation coefficient. S520. Calculate the difference in electricity supply over a future time period based on the predicted power generation and predicted electricity consumption data. S530: Obtain the alarm threshold based on the power supply difference, and adjust the power supply method in the target area according to the alarm threshold.

[0048] In step S510, the predicted power generation of the wind turbine in the future time period is calculated based on the predicted power data and the deviation coefficient. The deviation coefficient is denoted as K, the predicted power data as P1, and the predicted power generation as V1. The following relationship is satisfied between the deviation coefficient, the predicted power data, and the predicted power generation: V1 = P1 × K.

[0049] For example, if the deviation coefficient of a wind turbine is 0.9 and the predicted power data is 20 kWh, then the predicted power generation of this wind turbine is 20 × 0.9 = 18 kWh.

[0050] In steps S520 and S530, the difference in power supply within a future time period is calculated based on the predicted power generation and predicted power consumption data. For example, if the predicted power generation in a certain area is 30 MWh and the predicted power consumption is 35 MWh within a certain time period, then the difference between the predicted power generation and predicted power consumption is 5 MWh, and the corresponding alarm threshold is also 5 MWh.

[0051] By calculating the predicted power generation using the deviation coefficient, the predicted power generation is made more consistent with the actual power generation of wind turbines. The acquisition of predicted power consumption data makes the power supply difference more consistent with the actual power consumption in the target area at the current time period, thereby improving the accuracy of alarm thresholds and avoiding incorrect adjustments to the power supply method that could affect the stability of the power supply system.

[0052] [Seventh Embodiment] See Figure 5 In one specific embodiment, an alarm threshold is obtained based on the power supply difference, and the power supply method in the target area is adjusted according to the alarm threshold, specifically including: S531. Real-time monitoring of the power consumption data of the working cable in the future time period to obtain real-time power consumption data, and real-time monitoring of the real-time power generation of wind turbine units at various locations. S532. When the difference between the real-time power consumption data and the real-time power generation is less than the alarm threshold, the usage time of the energy storage battery is calculated based on the difference between the real-time power consumption data and the real-time power generation. S533. When the difference between the real-time power consumption data and the real-time power generation is greater than or equal to the alarm threshold, the predicted power consumption data in the target area is reduced according to the power supply difference, and the usage time of the energy storage battery is planned.

[0053] In step S532, it should be noted that energy storage batteries typically use flywheel energy storage. Due to the limited storage capacity of energy storage batteries, when the difference between the real-time power consumption data and the real-time power generation is less than the alarm threshold, it is necessary to calculate the usage time of the energy storage battery based on the difference between the real-time power consumption data and the real-time power generation. Then, based on the usage time of the energy storage battery, the input of other electrical types needs to be adjusted. For example, if the maximum storage capacity of the energy storage battery is 10 MKWh, and the usage time of the energy storage battery calculated based on the difference between the real-time power consumption data and the real-time power generation is 2 hours, then the power input method needs to be adjusted within 2 hours.

[0054] In step S533, when the power generation of the wind turbine is used as the main power supply, it is necessary to manage the power-consuming equipment according to the power supply difference. For example, if the wind turbine is built on an isolated island and supplies power to the monitoring work on the island, when facing extreme weather, the wind turbine may not be able to supply power for a long time or may be in a low power supply state for a long time. In this case, it is necessary to plan the use of the power-consuming equipment according to the power supply of the energy storage battery and give priority to ensuring the operation of necessary working equipment.

[0055] When the difference between real-time power consumption data and real-time power generation is less than the alarm threshold, the usage time of the energy storage battery is calculated based on the difference, fully considering the limited energy storage capacity of the energy storage battery, thereby achieving efficient resource utilization. When the difference between real-time power consumption data and real-time power generation is greater than or equal to the alarm threshold, the predicted power consumption data in the target area is reduced based on the power supply difference, and the usage time of the energy storage battery is planned, fully considering the possibility of insufficient power supply in special circumstances, thus improving power supply stability.

[0056] [Eighth Embodiment] See Figure 6 The present invention also provides a detection system 100, wherein the detection method described in the above embodiments is applied to the detection system 100. The detection system 100 includes: a storage module 110: a database is stored in the storage module 110; an acquisition module 120: the acquisition module 120 is used to acquire meteorological data; a calculation module 130: the calculation module 130 is used to calculate theoretical power data and prediction deviation coefficient; and a judgment module 140: the judgment module 140 is used to judge whether the power supply mode in the target area needs to be adjusted. The detection system 100 has all the technical features of the above detection method, which will not be described in detail here.

[0057] While the present invention has been disclosed above, it is not limited thereto. Any person skilled in the art can make various modifications and alterations without departing from the spirit and scope of the invention; therefore, the scope of protection of the present invention should be determined by the scope defined in the claims.

Claims

1. An abnormality detection method of dispatching power by flywheel energy storage, characterized by, The abnormality detection method comprises: Obtaining historical power data of a wind turbine, overall historical meteorological data corresponding to the historical power data, and actual historical meteorological data, calculating a power influence factor according to the actual historical meteorological data and the historical power data, and establishing a power prediction model; Real-time monitoring the wind turbine in a target time period to obtain real-time power data and real-time meteorological data at a corresponding time; Calculating theoretical power data of the wind turbine according to the real-time meteorological data and the power prediction model; Correcting the power prediction model according to the theoretical power data and the real-time power data to obtain a corrected prediction model; Predicting meteorological data in a future time period, denoted as predicted meteorological data, calculating a prediction deviation coefficient according to the predicted meteorological data and the corrected prediction model; Filtering historical time periods corresponding to the future time period from a database to obtain historical power consumption data, predicting power consumption in the future time period according to the historical power consumption data to obtain predicted power consumption data; Calculating an electricity supply difference value in the future time period according to the predicted power consumption data and the prediction deviation coefficient, and adjusting a power supply mode in a target region according to the electricity supply difference value.

2. The abnormality detection method according to claim 1, characterized by, The obtaining of the historical power data of the wind turbine, the overall historical meteorological data corresponding to the historical power data, and the actual historical meteorological data, the calculation of the power influence factor according to the actual historical meteorological data and the historical power data, and the establishment of the power prediction model specifically comprise: Obtaining the historical power data corresponding to different geographical positions of each wind turbine in the target region, and determining an influence coefficient of the geographical position on the actual historical meteorological data according to the historical power data and the geographical position of the wind turbine; Correcting the influence coefficient according to the actual historical meteorological data and the overall historical meteorological data to obtain a correction coefficient; Calculating the power influence factor according to the actual historical meteorological data and the historical power data of the wind turbine, and establishing the power prediction model.

3. The abnormality detection method according to claim 2, characterized by, The obtaining of the historical power data corresponding to different geographical positions of each wind turbine in the target region, and the determination of the influence coefficient of the geographical position on the actual historical meteorological data according to the historical power data and the geographical position of the wind turbine specifically comprise: Dividing the wind turbines into different associated groups according to the type and service life, and obtaining the theoretical power data according to the technical parameters of the wind turbines in the same associated group and the overall historical meteorological data; Calculating the influence coefficient of the geographical position on the actual historical meteorological data according to the theoretical power data and the historical power data.

4. The abnormality detection method according to claim 3, characterized by, The correction of the influence coefficient according to the actual historical meteorological data and the overall historical meteorological data to obtain a correction coefficient specifically comprises: Calculating theoretical historical meteorological data of different geographical positions according to the overall historical meteorological data and the influence coefficient; Correcting the influence coefficient according to the actual historical meteorological data and the theoretical historical meteorological data to obtain a correction coefficient.

5. The abnormality detection method according to claim 4, characterized by, The weather data in the future time period is predicted, denoted as predicted weather data, and a prediction deviation coefficient is calculated according to the predicted weather data and the corrected prediction model, specifically including: Theoretical weather data of each wind turbine is calculated according to the predicted weather data and the correction coefficient; The predicted power data in the future time period is obtained according to the theoretical weather data and the corrected prediction model; The prediction deviation coefficient is obtained according to the power data and the predicted power data.

6. The abnormality detection method according to claim 5, characterized by, The power supply difference in the future time period is calculated according to the predicted power consumption data and the prediction deviation coefficient, and the power supply mode in the target area is adjusted according to the power supply difference, specifically including: The predicted power generation of the wind turbine in the future time period is calculated according to the prediction deviation coefficient; The power supply difference in the future time period is calculated according to the predicted power generation and the predicted power consumption data; An alarm threshold is obtained according to the power supply difference, and the power supply mode in the target area is adjusted according to the alarm threshold.

7. The abnormality detection method according to claim 6, characterized by, The power supply difference in the future time period is calculated according to the predicted power consumption data and the prediction deviation coefficient, and the power supply mode in the target area is adjusted according to the power supply difference, specifically including: Real-time power consumption data of the working cable in the future time period is monitored to obtain real-time power consumption data, and real-time power generation of the wind turbine at each position is monitored; When the power supply difference between the real-time power consumption data and the real-time power generation is less than the alarm threshold, the use time of the energy storage battery is calculated according to the power supply difference between the real-time power consumption data and the real-time power generation; When the power supply difference between the real-time power consumption data and the predicted power consumption data is greater than or equal to the alarm threshold, the predicted power consumption data in the target area is reduced according to the power supply difference, and the use time of the energy storage battery is planned.

8. An abnormality detection system for dispatching power by a flywheel energy storage, characterized by, The detection method according to any one of claims 1 to 7 is applied to the detection system, and the detection system includes: A storage module: the database is arranged in the storage module; A collection module: the collection module is used to collect the weather data; A calculation module: the calculation module is used to calculate the theoretical power data and the prediction deviation coefficient; A judgment module: the judgment module is used to judge whether the power supply mode in the target area needs to be adjusted.

Images