CN-121981702-A - Marine wind power plant collaborative operation and maintenance method based on atmospheric marine environment dynamic simulation

CN121981702ACN 121981702 ACN121981702 ACN 121981702ACN-121981702-A

Abstract

The invention relates to a marine wind farm collaborative operation and maintenance method based on dynamic simulation of an atmospheric marine environment, which is applicable to the field of new energy. The method comprises the steps of S1, carrying out clock synchronization, coordinate conversion, anomaly rejection, missing value restoration and dimension standardization on collected multi-source heterogeneous data to generate a standardized data set in a unified format, S2, constructing an atmospheric ocean environment field dynamic model and an equipment state evolution model, carrying out coupling training on the two models through a multi-scale space-time fusion algorithm to generate an environment change trend prediction result and an equipment running state prediction result, dynamically correcting model parameters according to prediction errors, S3, carrying out quantization grading on equipment fault early warning and environment risks, carrying out matching scheduling on the combination of real-time operation and maintenance resource states, generating a collaborative operation and maintenance plan through a space-time optimization algorithm, executing the collaborative operation and maintenance plan and feeding back an execution state in real time, and realizing iterative optimization of the operation and maintenance plan.

Inventors

WANG BIN
CAI LI
LAI YONGQING
HUAN CAIYUN
WANG WENQIAN

Assignees

中国电建集团华东勘测设计研究院有限公司

Dates

Publication Date: 20260505
Application Date: 20251226

Claims (13)

1. The method for collaborative operation and maintenance of the offshore wind farm based on the dynamic simulation of the atmospheric and marine environments is characterized by comprising the following steps of: S1, acquiring atmospheric marine environment data, wind power plant equipment running state data and operation and maintenance resource state data of a marine wind power plant area through multiple types of monitoring points, and performing clock synchronization, coordinate conversion, abnormal rejection, missing value restoration and dimension standardization processing on the acquired multiple-source heterogeneous data to generate a standardized data set in a uniform format; S2, constructing an atmospheric ocean environmental field dynamic model and an equipment state evolution model based on the standardized data set, performing coupling training on the two models through a multi-scale space-time fusion algorithm to generate an environmental change trend prediction result and an equipment running state prediction result, and dynamically correcting model parameters according to prediction errors; And S3, quantitatively grading equipment fault early warning and environmental risks based on the environmental change trend prediction result, the equipment running state prediction result and the pre-stored operation and maintenance rule, carrying out matched scheduling by combining the real-time operation and maintenance resource states, generating a collaborative operation and maintenance plan through a space-time optimization algorithm, executing the collaborative operation and maintenance plan, and feeding back the execution state in real time to realize iterative optimization of the operation and maintenance plan.
2. The offshore wind farm collaborative operation and maintenance method based on the dynamic simulation of the atmospheric and marine environment according to claim 1, wherein: the atmospheric marine environment data comprise wind field parameters, wave field parameters and flow field parameters, and are cooperatively collected through meteorological buoys deployed on the sea surface, an underwater observation platform on the sea bottom and monitoring equipment of a land anemometer tower; The equipment operation state data comprise the rotating speed, the vibration amplitude, the temperature of the wind turbine generator and the insulation resistance of the submarine cable, and are collected through sensors arranged at key parts of the equipment; The operation and maintenance resource state data comprise the positions of operation and maintenance personnel, the availability of equipment and the stock of materials, and are collected through GPS positioning equipment and a stock management system.
3. The method for collaborative operation and maintenance of the offshore wind farm based on the dynamic simulation of the atmospheric and ocean environment according to claim 1 is characterized in that in the step S1, the clock synchronization is realized through a satellite time service system, the time stamp error is controlled, and the coordinate conversion converts longitude and latitude of a monitoring point into plane coordinates through a UTM projection coordinate system.
4. The method for collaborative operation and maintenance of the offshore wind farm based on the dynamic simulation of the atmospheric and ocean environment according to claim 1 is characterized in that in the step S1, abnormal data points are identified by adopting a 3 sigma principle, impulse noise is removed through sliding median filtering, a linear interpolation method is adopted for repairing missing values, a data segment which is continuously in missing state for 2 hours is marked as invalid, and Min-Max standardization is adopted for dimension standardization to convert data into a [0,1] interval.
5. The marine wind farm collaborative operation and maintenance method based on the atmospheric marine environment dynamic simulation is characterized in that in the step S2, the atmospheric marine environment dynamic model comprises a wind speed field model, a sea wave field model and a sea current field model, wherein the wind speed field model is built by fusing a WRF meteorological model with a CFD method, the sea wave field model is built based on a SWAN model, the sea current field model is built based on a FVCOM model, and the equipment state evolution model quantifies equipment health state evolution rules by extracting equipment fault characteristic frequencies and combining an index degradation model.
6. The method for collaborative operation and maintenance of an offshore wind farm based on dynamic simulation of an atmospheric and marine environment according to claim 1, wherein in step S2 the multi-scale spatiotemporal fusion algorithm comprises the sub-steps of: S21, synchronizing the standardized data set with the time stamp, and resampling by adopting a sliding time window to generate a unified time sequence; S22, performing 3-layer Daubechies-4 wavelet decomposition on the unified time sequence to obtain a low-frequency approximate component With high-frequency detail components Calculating the energy ratio of each high-frequency component ; S23, generating grids covering the offshore wind farm area, constructing a Gaussian semi-variation function based on a Kriging interpolation method, and generating an environmental field matrix of each grid in the grids by combining the positions of each monitoring point and the acquired atmospheric marine environmental data; s24, splicing the time decomposition components and the environmental field matrix to form a space-time characteristic tensor, and combining the energy duty ratio Input space-time convolutional neural network, adopt The weighted mixed loss function training model of (2) and outputting the prediction result.
7. The method for collaborative operation and maintenance of a marine wind farm based on dynamic simulation of atmospheric and marine environments according to claim 6, wherein in step S2, the prediction error is determined by continuous Average relative error of individual time windows Quantification when When the model parameter is larger than the preset value, historical data in the latest preset time period is extracted to construct an updated training set, 3 layers of parameters at the front end of the space-time convolutional neural network are frozen, the weight of the last 1 layers of full-connection layers is finely adjusted, and the model parameters are dynamically corrected.
8. The method for collaborative operation and maintenance of an offshore wind farm based on dynamic simulation of an atmospheric and ocean environment according to claim 1, wherein in step S3, the quantitatively classifying the equipment fault pre-warning and the environmental risk based on the environmental change trend prediction result, the equipment operation state prediction result and the pre-stored operation and maintenance rule comprises: Constructing a device fault feature vector, wherein the device fault feature vector comprises a fault position weight, a parameter deviation degree and a fault development rate, the parameter deviation degree is determined based on a difference value between a device running state prediction result and a device running state threshold corresponding to an environment change trend prediction result, and the fault development rate is a development rate after the device running state prediction result deviates from the threshold; calculating a fault grade value based on the equipment fault feature vector, and determining an equipment fault early warning grade based on the fault grade value; and determining the environmental risk level based on the duration of the deviation from the safety threshold in the environmental change trend prediction result.
9. The method for collaborative operation and maintenance of an offshore wind farm based on dynamic simulation of an atmospheric and ocean environment according to claim 1, wherein in step S3, the resource matching comprises screening adaptive operation and maintenance personnel according to task skill requirements, planning operation and maintenance paths of an operation and maintenance ship and an unmanned aerial vehicle in combination with environmental prediction results, checking stock of materials and triggering a shortage material replenishment process.
10. The method for collaborative operation and maintenance of a marine wind farm based on dynamic simulation of an atmospheric and marine environment according to claim 1, wherein in step S3, the space-time optimization algorithm is a space-time collaborative genetic algorithm, and the task priority, the environmental risk period and the spatial clustering efficiency are weighted and considered by an fitness function by using "task-resource-time" as chromosome coding.
11. Offshore wind farm collaborative operation and maintenance device based on atmospheric marine environment dynamic simulation, which is characterized by comprising: the data acquisition module is used for acquiring the atmospheric marine environment data, the wind power plant equipment running state data and the operation and maintenance resource state data of the offshore wind power plant area through multiple types of monitoring points, performing clock synchronization, coordinate conversion, abnormal rejection, missing value restoration and dimension standardization on the acquired multi-source heterogeneous data, and generating a standardized data set in a uniform format; The dynamic simulation module is used for constructing an atmospheric ocean environmental field dynamic model and an equipment state evolution model based on the standardized data set, performing coupling training on the two models through a multi-scale space-time fusion algorithm, generating an environmental change trend prediction result and an equipment running state prediction result, and dynamically correcting model parameters according to prediction errors; And the operation and maintenance decision module is used for quantitatively grading equipment fault early warning and environmental risks based on the environmental change trend prediction result, the equipment operation state prediction result and the pre-stored operation and maintenance rule, carrying out matched scheduling by combining the real-time operation and maintenance resource state, generating a collaborative operation and maintenance plan through a space-time optimization algorithm, executing the collaborative operation and maintenance plan and feeding back the execution state in real time, and realizing iterative optimization of the operation and maintenance plan.
12. A storage medium having stored thereon a computer program executable by a processor, characterized in that the computer program when executed realizes the steps of the offshore wind farm co-operation method according to any of claims 1-10.
13. An offshore wind farm co-operation and maintenance device having a memory and a processor, the memory having stored thereon a computer program executable by the processor, characterized in that the computer program when executed implements the steps of the offshore wind farm co-operation and maintenance method according to any of claims 1-10.

Description

Marine wind power plant collaborative operation and maintenance method based on atmospheric marine environment dynamic simulation Technical Field The invention relates to a marine wind farm collaborative operation and maintenance method based on dynamic simulation of an atmospheric marine environment. Is suitable for the field of new energy. Background With the large-scale development of offshore wind power to deep open sea, operation and maintenance operations face three challenges of strong dynamic coupling of the atmosphere and ocean environment, delayed equipment fault response and low multi-task cooperative efficiency. The space-time abrupt change characteristics of the environmental fields (wind, wave and current) are in strong coupling correlation with the running state degradation of equipment such as fans, cables and the like, but the traditional operation and maintenance mode only depends on a static threshold early warning and discretization decision mechanism, and is difficult to adapt to the real-time requirement of resource scheduling in a complex marine environment. Disclosure of Invention Aiming at the problems, the invention provides a marine wind power plant collaborative operation and maintenance method based on the dynamic simulation of the atmospheric marine environment. The technical scheme adopted by the invention is that the method for collaborative operation and maintenance of the offshore wind farm based on the dynamic simulation of the atmospheric ocean environment comprises the following steps: S1, acquiring atmospheric marine environment data, wind power plant equipment running state data and operation and maintenance resource state data of a marine wind power plant area through multiple types of monitoring points, and performing clock synchronization, coordinate conversion, abnormal rejection, missing value restoration and dimension standardization processing on the acquired multiple-source heterogeneous data to generate a standardized data set in a uniform format; S2, constructing an atmospheric ocean environmental field dynamic model and an equipment state evolution model based on the standardized data set, performing coupling training on the two models through a multi-scale space-time fusion algorithm to generate an environmental change trend prediction result and an equipment running state prediction result, and dynamically correcting model parameters according to prediction errors; And S3, quantitatively grading equipment fault early warning and environmental risks based on the environmental change trend prediction result, the equipment running state prediction result and the pre-stored operation and maintenance rule, carrying out matched scheduling by combining the real-time operation and maintenance resource states, generating a collaborative operation and maintenance plan through a space-time optimization algorithm, executing the collaborative operation and maintenance plan, and feeding back the execution state in real time to realize iterative optimization of the operation and maintenance plan. The atmospheric marine environment data comprise wind field parameters, wave field parameters and flow field parameters, and are cooperatively collected through meteorological buoys deployed on the sea surface, an underwater observation platform on the sea bottom and monitoring equipment of a land anemometer tower; The equipment operation state data comprise the rotating speed, the vibration amplitude, the temperature of the wind turbine generator and the insulation resistance of the submarine cable, and are collected through sensors arranged at key parts of the equipment; The operation and maintenance resource state data comprise the positions of operation and maintenance personnel, the availability of equipment and the stock of materials, and are collected through GPS positioning equipment and a stock management system. In the step S1, the clock synchronization is realized through a satellite time service system, the time stamp error is controlled, and the coordinate conversion converts the longitude and latitude of the monitoring point into a plane coordinate through a UTM projection coordinate system. In the step S1, abnormal data points are identified by adopting a3 sigma principle, impulse noise is removed through sliding median filtering, the missing value restoration adopts a linear interpolation method, a data segment which is continuously in missing and quenching for 2 hours is marked as invalid, and the dimension standardization adopts Min-Max standardization to convert data into a [0,1] section. In the step S2, the atmospheric marine environment field dynamic model comprises a wind speed field model, a sea wave field model and a ocean current field model, wherein the wind speed field model is built by fusing a WRF meteorological model and a CFD method, the sea wave field model is built based on a SWAN model, the ocean current field model is built based on a FVCOM mod