Wind power plant network-related assessment early warning system and method

Abstract

The invention provides a wind power plant network-related assessment early warning system and method, and the system comprises seven modules: a station assessment information statistics module, an assessment data exception analysis module, a one-key collection assessment-free module, a total station data analysis and storage module, an exception alarm module, an information instant alarm module, and a master station centralized management module. The system gets rid of a management mode of top-to-bottom supervision from a scheduling end to a plant station end, monitors uploaded data in real time, analyzes the uploaded data, prompts in real time, actively explores an early warning analysis system without a kernel based on big data analysis, and can quickly, flexibly and efficiently perform early warning in a short time. And the network-related equipment assessment problem is processed in an omnibearing and multi-dimensional manner with less manpower and material resources.

Description

Technical Field

[0001] The present invention relates to the technical field of wind farms, and in particular to a wind farm network-related assessment and early warning system and method. Background Art

[0002] The existing wind farm network-related equipment, such as AGC / AVC system, wind power prediction system, primary frequency regulation system, etc., operate independently and form their own systems. They communicate and transmit data with the centralized control and dispatching end separately, and the data and information are not analyzed or identified. This networking method cannot use network-related equipment to provide auxiliary data real-time monitoring, assessment warning, and abnormal analysis for wind farm operators. It is impossible to understand the information and assessment data of network-related equipment and dispatch, communication of important equipment in the wind farm in real time. Centralized management is required to reduce the double losses of economy and management caused by dispatching assessment. Summary of the invention

[0003] The present invention provides a wind farm network-related assessment early warning system and method, which can realize real-time early warning of plant and station communication data with dispatching, and can also efficiently collect key data and apply to dispatching for assessment exemption in a timely manner.

[0004] In order to solve the above technical problems, the technical solution adopted by the present invention is: A wind farm network-related assessment early warning system, the early warning system comprises seven modules, namely, a station assessment information statistics module, an assessment data abnormality analysis module, a one-key collection and exemption module, a full-station data analysis and storage module, an abnormality alarm module, an information instant alarm module and a master station centralized management module. The station assessment information statistics module is connected to a station-side server via a centralized control-side communication bus, reads data information of each wind turbine set from the station server, and sends the data information to the assessment data abnormality analysis module after statistical classification. The assessment data abnormality analysis module judges the data information of the statistically classified wind turbine set and outputs the judgment result. The one-key collection and exemption module collects abnormal information other than the data collected by the station assessment information statistics module. The full-station data analysis and storage module stores data of the assessment data abnormality analysis module and the one-key collection and exemption module. The abnormality alarm module warns of the abnormality of the station. The information instant alarm module monitors and informs the monitoring information of the station.

[0005] The wind turbine data information read by the above-mentioned station assessment information statistics module includes AGC assessment information, AVC assessment information, communication status with the dispatcher, power prediction assessment information and primary frequency regulation assessment information.

[0006] The judgment results output by the above-mentioned assessment data anomaly analysis module include judgment abnormal results and statistical results. The judgment abnormal results include unqualified AVC adjustment performance, unqualified AGC adjustment performance, fast frequency modulation performance, remote adjustment / remote control failure, poor communication quality, data jump and data non-refresh. The statistical results include year-on-year and month-on-month statistical results.

[0007] The above-mentioned one-click collection and assessment-free module collects information on the plant station being subject to dispatch assessment due to failure to deliver instructions, different data sources, poor channel quality and data network equipment failure between the dispatching master station and the plant station grid-connected equipment. According to the dispatching information and format requirements, the frequency, AGC instructions, active power measurement summary and other data at the specified time are collected for opinions.

[0008] The above-mentioned full-station data analysis and storage module performs communication message capture and data analysis, AGC / AVC / fan monitoring / SVG / NCS and other multi-channel data analysis, scheduling, and AGC / AVC instruction monitoring within the plant station, as well as storage and export of historical data and messages.

[0009] The site abnormalities warned by the above-mentioned abnormal alarm module include communication interruption between telecontrol and dispatch, communication interruption between PMU and dispatch, communication interruption between telecontrol and other important systems or devices in the plant, AVC exit, AGC exit, data jump, data non-refresh and remote control failure.

[0010] The above-mentioned information instant alarm module monitors the information of the entire station and optionally provides instant and summary alarm notifications via SMS.

[0011] The above-mentioned master station centralized management module includes the plant EVD overview center, real-time monitoring center, intelligent alarm center, benchmarking decision center, assessment statistics center, data analysis center and data display center.

[0012] Using the above-mentioned early warning method of a wind farm grid-related assessment early warning system, the AGC pass rate assessment includes control performance assessment indicators 1 and 2. Each type of assessment indicator has a 1-minute assessment function within the tth minute. , and the n-minute level assessment function within the first n minutes , ,Right now: ; in, is the actual system frequency in the tth minute With the system rated frequency The average deviation of , is the actual frequency of the system at the initial moment, is the frequency increment of the system in the kth minute compared with the previous minute; B is the frequency deviation coefficient of the control area, which is positive; As the frequency control target, the statistical value of the mean square root of the 1-minute frequency deviation of the interconnected power grid in the previous year is taken; is the average regional control deviation value in the tth minute.

[0013] ; In the formula, is the actual power of the interconnection system tie line in the tth minute Planned power of tie lines The average deviation of .

[0014] Assuming the outbound regions are balanced, for: ; In the formula, is the predicted load in the control area within the minute; is the actual load in the control area in the k-1th minute; is the total number of AGC units; is the first i Secondary frequency regulation power generation increment; The power is adjusted for the system units of the control area and the external area respectively.

[0015] and The calculation formula is: ; In the formula, The frequency control target is the statistical value of the average root mean square of the frequency deviation of the interconnected power grid in the previous year within n minutes; is the frequency deviation coefficient of the entire interconnected power grid.

[0016] When conducting AGC assessment on the power grid, the cycle is 30 minutes, and the following conditions must be met to pass the assessment: 1. 30-minute AGC pass, that is: ; 2. The minute-level AGC pass rate should reach more than 70% within 30 minutes, that is: ; Where n is time, which takes the value of 30; I(*) is a function to determine whether (*) is true or not, and its value is 1 if it is true, otherwise it is 0.

[0017] The present invention provides a short-term wind power prediction method based on the Frandsen wake model, which has the following beneficial effects: (1) Downscaling of predicted wind speeds is achieved through the Frandsen wake model, which can more finely characterize the wake effect inside the wind farm. Based on physical prior knowledge, the robustness of the prediction model is enhanced and the prediction accuracy is improved.

[0018] (2) Recursive feature elimination is used to automatically screen out key wind power features, reducing the reliance on manual feature engineering.

[0019] (3) Utilize deep learning algorithms and model integration technology to fully explore the nonlinear relationship of wind power data and enhance the generalization ability of the model.

[0020] (4) Linear combination of multi-class models can adaptively adjust weights according to the predictive capabilities of different models to further improve the predictive performance.

[0021] In summary, the present invention provides a wind farm network-related assessment early warning system and method, which combines the actual situation on site and gets rid of the top-down supervision mode from the dispatching end to the plant end. It uses a mirror monitoring system to monitor uploaded data in real time, analyze uploaded data, and give real-time reminders. It actively explores an early warning analysis system based on big data analysis without a kernel, which can handle network-related equipment assessment issues in a short time, quickly and flexibly with less manpower and material resources in an all-round and multi-dimensional manner, and can ensure that the communication quality between the substation and the centralized control and dispatching end meets the relevant requirements, greatly reducing the loss of the dispatched assessment. It is also the first time that the early warning analysis system has been used as a big data analysis method in an offshore wind farm, which provides an extremely important reference for the selection of large-scale data analysis directions and the establishment of data models in future wind farms. BRIEF DESCRIPTION OF THE DRAWINGS

[0022] The present invention will be further described below in conjunction with the accompanying drawings and embodiments: Figure 1 It is the structure diagram of the information management system of the wind and solar power plant; Figure 2 It is a structural diagram of the early warning system. DETAILED DESCRIPTION

[0023] In order to make the purpose, technical solutions and advantages of the present invention clearer, the following content will systematically and completely describe the specific technical solutions of the present invention in combination with the drawings provided according to the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work are within the scope of protection of the present invention.

[0024] like Figure 1 and 2As shown in the figure, a wind farm grid-related assessment early warning system, the early warning system includes seven modules, namely, a station assessment information statistics module, an assessment data abnormality analysis module, a one-key collection and exemption module, a whole station data analysis and storage module, an abnormality alarm module, an information instant alarm module and a master station centralized management module. The station assessment information statistics module is connected to the station side server through the centralized control side communication bus, reads the data information of each wind turbine set from the station server, and sends it to the assessment data abnormality analysis module after statistical classification. The assessment data abnormality analysis module judges the data information of the wind turbine set after statistical classification and outputs the judgment result. The one-key collection and exemption module collects abnormal information other than the data collected by the station assessment information statistics module. The whole station data analysis and storage module stores the data of the assessment data abnormality analysis module and the one-key collection and exemption module. The abnormality alarm module warns of the abnormality of the station, and the information instant alarm module monitors and informs the station monitoring information.

[0025] The wind turbine data information read by the above-mentioned station assessment information statistics module includes AGC assessment information, AVC assessment information, communication status with the dispatcher, power prediction assessment information and primary frequency regulation assessment information.

[0026] The assessment information of AGC includes the qualification rate and operation rate, the assessment information of AVC includes the qualification rate and operation rate, the dispatching communication status includes the communication quality and channel availability, the power prediction assessment information includes the transmission rate and accuracy, and the primary frequency modulation assessment information includes the fast frequency response.

[0027] The above AGC pass rate assessment includes control performance assessment indicators 1 and 2. Each type of assessment indicator has a 1-minute assessment function within the tth minute. , and the n-minute level assessment function within the first n minutes , ,Right now: ; in, is the actual system frequency in the tth minute With the system rated frequency The average deviation of , is the actual frequency of the system at the initial moment, is the frequency increment of the system in the kth minute compared with the previous minute; B is the frequency deviation coefficient of the control area, which is positive; As the frequency control target, the statistical value of the mean square root of the 1-minute frequency deviation of the interconnected power grid in the previous year is taken; is the average regional control deviation value in the tth minute.

[0028] ; In the formula, is the actual power of the interconnection system tie line in the tth minute Planned power of tie lines The average deviation of .

[0029] Assuming the outbound regions are balanced, for: ; In the formula, is the predicted load in the control area within the minute; is the actual load in the control area in the k-1th minute; is the total number of AGC units; is the first i Secondary frequency regulation power generation increment; The power is adjusted for the system units of the control area and the external area respectively.

[0030] and The calculation formula is: ; In the formula, The frequency control target is the statistical value of the average root mean square of the frequency deviation of the interconnected power grid in the previous year within n minutes; is the frequency deviation coefficient of the entire interconnected power grid.

[0031] When conducting AGC assessment on the power grid, the cycle is 30 minutes, and the following conditions must be met to pass the assessment: 1. 30-minute AGC pass, that is: ; 2. The minute-level AGC pass rate should reach more than 70% within 30 minutes, that is: ; Where n is time, which takes the value of 30; I(*) is a function to determine whether (*) is true or not, and its value is 1 if it is true, otherwise it is 0.

[0032] The judgment results output by the above-mentioned assessment data anomaly analysis module include judgment abnormal results and statistical results. The judgment abnormal results include unqualified AVC adjustment performance, unqualified AGC adjustment performance, fast frequency modulation performance, remote adjustment / remote control failure, poor communication quality, data jump and data non-refresh. The statistical results include year-on-year and month-on-month statistical results.

[0033] The above-mentioned one-click collection and assessment-free module collects information on the plant station being subject to dispatch assessment due to failure to deliver instructions, different data sources, poor channel quality and data network equipment failure between the dispatching master station and the plant station grid-connected equipment. According to the dispatching information and format requirements, the frequency, AGC instructions, active power measurement summary and other data at the specified time are collected for opinions.

[0034] The above-mentioned full-station data analysis and storage module performs communication message capture and data analysis, AGC / AVC / fan monitoring / SVG / NCS and other multi-channel data analysis, scheduling, and AGC / AVC instruction monitoring within the plant station, as well as storage and export of historical data and messages.

[0035] The site abnormalities warned by the above-mentioned abnormal alarm module include communication interruption between telecontrol and dispatch, communication interruption between PMU and dispatch, communication interruption between telecontrol and other important systems or devices in the plant, AVC exit, AGC exit, data jump, data non-refresh and remote control failure.

[0036] The above-mentioned information real-time alarm module monitors the entire station information, and can optionally provide real-time and summary alarm notifications via SMS to solve the problem of scattered plant sites, small number of maintenance personnel, and inability to timely understand the operation status of the plant. It has customizable alarm level real-time notifications, custom information summary scheduled notifications, and custom management position classification for SMS notifications.

[0037] The above-mentioned master station centralized management module includes the plant EVD overview center, real-time monitoring center, intelligent alarm center, benchmarking decision center, assessment statistics center, data analysis center and data display center.

Claims

1. A wind farm grid-related assessment and early warning system, characterized in that: The early warning system includes seven modules, namely, station assessment information statistics module, assessment data anomaly analysis module, one-click assessment-free collection module, whole-station data analysis and storage module, anomaly alarm module, information instant alarm module and master station centralized management module. The station assessment information statistics module is connected to the station side server through the centralized control side communication bus, reads the data information of each wind turbine from the station server, and sends it to the assessment data anomaly analysis module after statistical classification. The assessment data anomaly analysis module judges the data information of the wind turbine after statistical classification and outputs the judgment result. The one-click assessment-free collection module collects abnormal information other than the data collected by the station assessment information statistics module. The whole-station data analysis and storage module stores the data of the assessment data anomaly analysis module and the one-click assessment-free collection module. The anomaly alarm module warns of the station anomaly, and the information instant alarm module monitors and informs the station monitoring information.

2. According to the wind farm grid-related assessment and early warning system described in claim 1, it is characterized in that: The wind turbine data information read by the station assessment information statistics module includes AGC assessment information, AVC assessment information, dispatch communication status, power prediction assessment information and primary frequency regulation assessment information.

3. According to the wind farm grid-related assessment and early warning system described in claim 2, it is characterized in that: The judgment results output by the assessment data abnormality analysis module include abnormal judgment results and statistical results. The abnormal judgment results include unqualified AVC adjustment performance, unqualified AGC adjustment performance, fast frequency modulation performance, remote adjustment / remote control failure, poor communication quality, data jump and data non-refresh. The statistical results include year-on-year and month-on-month statistical results.

4. According to the wind farm grid-related assessment and early warning system described in claim 3, it is characterized in that: The one-key collection and assessment-free module collects information on the plant station being subject to dispatch assessment due to failure to deliver instructions, different data sources, poor channel quality, and data network equipment failure between the dispatching master station and the plant station grid-connected equipment. According to the dispatching information and format requirements, the frequency, AGC instructions, active power measurement summary and other data at the specified time are collected for opinions.

5. According to the wind farm grid-related assessment and early warning system described in claim 4, it is characterized in that: The whole station data analysis and storage module performs communication message capture and data analysis, AGC / AVC / fan monitoring / SVG / NCS and other multi-channel data analysis, scheduling, and AGC / AVC instruction monitoring within the plant station, as well as storage and export of historical data and messages.

6. According to the wind farm grid-related assessment and early warning system described in claim 5, it is characterized in that: The station abnormalities warned by the abnormal alarm module include communication interruption between telecontrol and dispatch, communication interruption between PMU and dispatch, communication interruption between telecontrol and other important systems or devices in the plant, AVC exit, AGC exit, data jump, data non-refresh and remote control failure.

7. A wind farm grid-related assessment and early warning system according to claim 6, characterized in that: The information instant alarm module monitors the entire station information and optionally provides instant and summary alarm notifications via text messages.

8. A wind farm grid-related assessment and early warning system according to claim 7, characterized in that: The master station centralized management module includes a plant EVD overview center, a real-time monitoring center, an intelligent alarm center, a benchmarking decision center, an assessment statistics center, a data analysis center, and a data display center.

9. The early warning method using the wind farm grid-related assessment early warning system according to claim 8 is characterized in that: The AGC pass rate assessment includes control performance assessment indicators 1 and 2. Each type of assessment indicator has a 1-minute assessment function within the tth minute. , and the n-minute level assessment function within the first n minutes , ,Right now: ； in, is the actual system frequency in the tth minute With the system rated frequency The average deviation of , is the actual frequency of the system at the initial moment, is the frequency increment of the system in the kth minute compared with the previous minute; B is the frequency deviation coefficient of the control area, which is positive; As the frequency control target, the statistical value of the mean square root of the 1-minute frequency deviation of the interconnected power grid in the previous year is taken; is the average regional control deviation value in the tth minute; ； In the formula, is the actual power of the interconnection system tie line in the tth minute Planned power of tie lines The average deviation of Assuming the outbound regions are balanced, for: ； In the formula, is the predicted load in the control area within the minute; is the actual load in the control area in the k-1th minute; is the total number of AGC units; is the first i Secondary frequency regulation power generation increment; Adjust power for system units in control area and external connection area respectively; and The calculation formula is: ； In the formula, The frequency control target is the statistical value of the average root mean square of the frequency deviation of the interconnected power grid in the previous year within n minutes; is the frequency deviation coefficient of the entire interconnected power grid; When conducting AGC assessment on the power grid, the cycle is 30 minutes, and the following conditions must be met to pass the assessment: 1) 30-minute AGC pass, that is: ； 2) The minute-level AGC pass rate reaches more than 70% within 30 minutes, that is: ； Where n is time, which takes the value of 30; I(*) is a function to determine whether (*) is true or not, and its value is 1 if it is true, otherwise it is 0.

Images