Intelligent early warning method and system for wind turbine generator set

By analyzing the monitoring data of wind turbines and establishing early warning models, the problem of early warning of wind turbines in complex environments is solved, and the advance prediction and efficient operation and maintenance of faults are achieved, ensuring the stable operation of the wind farm.

CN114738204BActive Publication Date: 2024-11-15HUANENG JIUQUAN WIND POWER CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202210395026.5
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2022-04-15
Publication Date
2024-11-15
Estimated Expiration
2042-04-15

AI Technical Summary

Technical Problem

The long-term operation of wind turbines in complex environments leads to degradation of component performance and increased failure rate. It is difficult for existing technology to conduct in-depth mining through massive monitoring data, resulting in operation and maintenance personnel being unable to obtain early warning in a timely manner, affecting power generation efficiency.

Method used

By acquiring multiple continuous time period monitoring data of wind turbines, using variance-based data screening model and early warning model for data analysis, combining expert experience library and BIM model, establishing signal forward propagation and error backpropagation based on BP algorithm based signal algorithm early warning model predicts equipment health and outputs fault warning information.

Benefits of technology

It realizes the advance prediction of wind turbine failures, improves the perception of equipment health status by operation and maintenance personnel, reduces the occurrence of potential failures, ensures efficient output of wind farms, and improves its accuracy by correcting early warning information.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN114738204B_ABST
    Figure CN114738204B_ABST
Patent Text Reader

Abstract

The invention relates to the technical field of wind power generation monitoring, and provides an intelligent early warning method and system for a wind generator set. The method comprises the following steps: obtaining all kinds of monitoring data of a wind generator set in a plurality of continuous time periods, and sending the monitoring data of each time period into a variance-based data screening model to obtain data variances of a plurality of acquisition nodes of various types of monitoring data in the time period; classifying all devices in the wind generator set according to coupling degrees, and if the coupling degree between the two devices is higher than a preset first threshold, the two devices are classified as coupled devices, and the variances of the monitoring data of a plurality of continuous time periods related to the coupled devices are sent into an early warning model as input data; the early warning model predicts the equipment health of the wind generator set according to the trend change of the input data, and outputs fault early warning information if the health is lower than a second threshold; and a system application method forms a prediction of wind generator set faults by mining massive monitoring data.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the technical field of wind power generation monitoring, and in particular to an intelligent early warning method and system for a wind power generator set. Background Art

[0002] As a green and pollution-free new energy, wind energy is gaining more and more attention and is the main energy resource in the future. However, many wind farms are often located in remote Gobi deserts. Affected by complex and changeable environmental factors, the performance of wind turbine components will gradually decline under complex operating conditions for a long time, and the failure rate of the unit will continue to increase, eventually leading to component failure and unit shutdown. Therefore, real-time monitoring of wind turbine status and fault early warning diagnosis are very necessary.

[0003] At present, wind turbines have a large number of sensors that write data to servers, forming a large amount of complex, multi-source, heterogeneous wind turbine status monitoring data; the early fault information of many devices is hidden in these massive data. Due to the lack of in-depth mining, wind turbine operation and maintenance personnel cannot receive timely warnings. Often, after knowing the problem, it has already affected the work of the wind turbine, thereby reducing the efficiency of wind turbine power generation.

[0004] Therefore, an intelligent early warning method for wind turbines is needed. By mining and applying massive monitoring data, early prediction of wind turbine failures can be formed to facilitate maintenance of wind turbines by staff, so that wind turbines can operate stably for a long time. Summary of the invention

[0005] The purpose of the present invention is to provide an intelligent early warning method and system for wind turbines, which forms an early prediction of wind turbine failures by mining and applying massive monitoring data, so that operation and maintenance personnel can perceive the health status of the wind turbine in advance, and then prevent potential failures before they occur, thereby ensuring efficient output of the wind farm.

[0006] The embodiments of the present invention are implemented by the following technical solutions:

[0007] In a first aspect, a wind turbine generator intelligent early warning method is provided, comprising the following steps:

[0008] S1. Obtain all types of monitoring data of wind turbines in multiple continuous time periods, and send the monitoring data in each time period to a variance-based data screening model to obtain data variances of multiple acquisition nodes of various types of monitoring data in the time period;

[0009] S2. Classify all devices in the wind turbine generator set according to the coupling degree between each other. If the coupling degree between two devices is higher than a preset first threshold, the two devices are classified as coupled devices, and the variance of the monitoring data of multiple consecutive time periods related to the coupled devices is sent as input data to the early warning model;

[0010] S3. The warning model predicts the equipment health of the wind turbine generator set according to the trend change of the input data, and outputs fault warning information if the health is lower than a second threshold.

[0011] Furthermore, it also includes correcting the fault warning information and health status; specifically, setting up an expert experience database, and correcting the fault warning information according to the expert experience database; the expert experience database includes all fault type information, monitoring data of the wind turbine generator set within a preset time period before the occurrence of the corresponding fault, and measures for handling the corresponding fault.

[0012] Furthermore, it also includes establishing a BIM model of the wind turbine generator set, wherein the BIM model is established according to the equipment conditions in the wind turbine generator set, and the health status of each device is presented through the BIM model.

[0013] Furthermore, the establishment of the early warning model is specifically based on the forward propagation of signals and the back propagation of errors of the BP algorithm; during the forward propagation, the input signal acts on the output node through the hidden layer, and after nonlinear transformation, an output signal is generated; if the actual output does not match the expected output, the error is turned to the back propagation, and the output error is transmitted back to the input layer through the hidden layer layer by layer, and the error is apportioned to all units, and the error signal obtained from each layer is used as the basis for adjusting the weights of each unit; by adjusting the connection strength between the input node and the hidden layer node and the connection strength between the hidden layer node and the output node, the error is reduced along the gradient direction, and after multiple trainings, the optimal early warning model is obtained.

[0014] Furthermore, the method also includes obtaining the maximum difference in data of multiple acquisition nodes within the continuous time period, and adjusting the influencing factor in the early warning model according to the maximum difference, and the value of the maximum difference is positively correlated with the influencing factor.

[0015] Furthermore, the monitoring data are all collected and acquired through a SCADA system.

[0016] In a second aspect, an intelligent early warning system for a wind turbine is provided, comprising a data acquisition device, a communication device, a control center and a mobile terminal; the data acquisition device and the mobile terminal are both connected to the control center through the communication device; the data acquisition device is used to collect all types of monitoring data of the wind turbine; the control center is used to receive the monitoring data collected by the data acquisition device, and run the early warning model to provide fault early warning for the wind turbine and send the fault early warning information to the mobile terminal; the mobile terminal is used to receive and remind fault information of the wind turbine.

[0017] Furthermore, the data acquisition equipment at least includes acquisition sensors in the SCADA system and temperature sensors of the motor and the gear box.

[0018] In a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned intelligent early warning method for wind turbines when executing the computer program.

[0019] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned intelligent early warning method for wind turbines are implemented.

[0020] The technical solution of the embodiment of the present invention has at least the following advantages and beneficial effects:

[0021] (1) Through the mining and application of massive monitoring data, the wind turbine failure prediction is formed in advance, so that the operation and maintenance personnel can perceive the health status of the wind turbine in advance, and then prevent potential failures before they occur. Then, the wind turbine operation control strategy can be adjusted in time to ensure the efficient output of the wind farm;

[0022] (2) At the same time, the patent experience rule library is introduced to realize the correction of fault warning information, making the fault warning information more accurate;

[0023] (3) By correlating the monitoring data, accurate prediction of fault information of different equipment components can be achieved. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 A schematic diagram of the process provided by the present invention;

[0025] Figure 2 A schematic diagram of the structure of the system provided by the present invention. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0027] In a first aspect, a wind turbine generator intelligent early warning method is provided, comprising the following steps:

[0028] S1. Acquire all kinds of monitoring data of wind turbines in multiple continuous time periods, and send the monitoring data in each time period into a variance-based data screening model to obtain the data variance of multiple acquisition nodes of each type of monitoring data in the time period; only comparing data through variance may easily lead to inaccurate early warning, so the method also includes acquiring the maximum difference in data of multiple acquisition nodes in the continuous time period, and adjusting the influencing factor in the early warning model according to the maximum difference, and the value of the maximum difference is positively correlated with the influencing factor.

[0029] Among them, the impact factor is a preset constant value in the early warning model. The size of the constant value is positively correlated with the value of the maximum difference in data of multiple acquisition nodes in a continuous time period. That is, the larger the value of the maximum data difference, the larger the constant value, and the greater the possibility of a fault warning.

[0030] S2. Classify all devices in the wind turbine generator set according to the coupling degree. If the coupling degree between two devices is higher than a preset first threshold, the two devices are classified as coupled devices, and the variance of the monitoring data of multiple continuous time periods related to the coupled devices is sent as input data to the early warning model; the coupling degree is the degree of mutual influence between devices. For example, when a device A fails, it will obviously affect the operation of device B, then it is judged that the coupling degree of devices A and B is high.

[0031] S3. The warning model predicts the equipment health of the wind turbine generator set according to the trend change of the input data, and outputs fault warning information if the health is lower than a second threshold.

[0032] Although relying solely on data for early warning has a high accuracy, the probability of misjudgment still has room for further improvement. Therefore, the method also includes correcting the fault warning information and health status; specifically, setting up an expert experience library, and correcting the fault warning information based on the expert experience library; the expert experience library includes all fault type information, monitoring data of the wind turbine generator set within a preset time period before the occurrence of the corresponding fault, and measures for handling the corresponding fault.

[0033] The setting of the expert experience base can be a preset experience table that corresponds to the equipment status and response method through the data range, or it can be learned from the expert experience through artificial intelligence to improve the expert experience base and realize the correction of fault warning information through the expert experience base.

[0034] In order to intuitively observe the equipment health status of the wind turbine generator set, the method also includes establishing a BIM model of the wind turbine generator set, the BIM model is established according to the equipment status in the wind turbine generator set, and the health status of each device is presented through the BIM model.

[0035] It should be noted that one of the methods for establishing the early warning model described in the method is specifically based on the forward propagation of signals and the back propagation of errors of the BP algorithm; during forward propagation, the input signal acts on the output node through the hidden layer, and after nonlinear transformation, an output signal is generated; if the actual output does not match the expected output, the error is turned to the back propagation, and the output error is transmitted back to the input layer through the hidden layer layer by layer, and the error is apportioned to all units, and the error signal obtained from each layer is used as the basis for adjusting the weights of each unit; by adjusting the connection strength between the input node and the hidden layer node and the connection strength between the hidden layer node and the output node, the error is reduced along the gradient direction, and after multiple trainings, the optimal early warning model is obtained.

[0036] In order to reduce the cost of early warning, the monitoring data described in this method can be collected and acquired through the SCADA system. In order to make the fault warning more accurate, other parameter collection can also be introduced, such as the temperature information of the motor and gearbox.

[0037] In a second aspect, an intelligent early warning system for a wind turbine is provided, comprising a data acquisition device, a communication device, a control center and a mobile terminal; the data acquisition device and the mobile terminal are both connected to the control center through the communication device; the data acquisition device is used to collect all types of monitoring data of the wind turbine; the control center is used to receive the monitoring data collected by the data acquisition device, and run the early warning model to provide fault early warning for the wind turbine and send the fault early warning information to the mobile terminal; the mobile terminal is used to receive and remind fault information of the wind turbine.

[0038] The data acquisition equipment at least includes one or more of the acquisition sensors in the SCADA system and the temperature sensors of the motor and the gear box.

[0039] In a third aspect, an electronic device is provided, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the above-mentioned intelligent early warning method for wind turbines when executing the computer program.

[0040] In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the steps of the above-mentioned intelligent early warning method for wind turbines are implemented.

[0041] The above are only preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. An intelligent early warning method for a wind turbine generator set, characterized in that: The steps include: S1. Obtain all types of monitoring data of wind turbines in multiple continuous time periods, and send the monitoring data in each time period to a variance-based data screening model to obtain data variances of multiple acquisition nodes of various types of monitoring data in the time period; S2. Classify all devices in the wind turbine generator set according to the coupling degree, where the coupling degree refers to the degree of mutual influence between the devices; if the coupling degree between two devices is higher than the preset first threshold, the two devices are classified as coupled devices, and the variance of the monitoring data of multiple continuous time periods related to the coupled devices is sent to the early warning model as input data; the establishment of the early warning model is specifically based on the signal forward propagation and error back propagation of the BP algorithm; during forward propagation, the input signal acts on the output node through the hidden layer, and after nonlinear transformation, an output signal is generated; if the actual output does not match the expected output, the error is reversely propagated, and the output error is propagated back layer by layer through the hidden layer to the input layer, and the error is apportioned to all units, and the error signal obtained from each layer is used as the basis for adjusting the weight of each unit; by adjusting the connection strength between the input node and the hidden layer node and the connection strength between the hidden layer node and the output node, the error is reduced along the gradient direction, and after multiple training, the optimal early warning model is obtained; S3. The early warning model predicts the equipment health of the wind turbine generator set according to the trend change of the input data, and outputs fault early warning information if the health is lower than the second threshold value; and corrects the fault early warning information and the health; specifically, an expert experience library is set up, and the fault early warning information is corrected according to the expert experience library; the expert experience library includes all fault type information, monitoring data of the wind turbine generator set within a preset time period before the occurrence of the corresponding fault, and measures for handling the corresponding fault.

2. The intelligent early warning method for wind turbine generator set according to claim 1, characterized in that: It also includes establishing a BIM model of the wind turbine generator set, wherein the BIM model is established according to the equipment conditions in the wind turbine generator set, and the health status of each equipment is presented through the BIM model.

3. The intelligent early warning method for wind turbine generator set according to claim 2, characterized in that: The method also includes obtaining the maximum difference of data of multiple acquisition nodes in the continuous time period, and adjusting the influencing factor in the early warning model according to the maximum difference, and the value of the maximum difference is positively correlated with the influencing factor.

4. The intelligent early warning method for wind turbine generator set according to claim 3, characterized in that: The monitoring data are all collected and obtained through the SCADA system.

5. An intelligent early warning system for a wind turbine generator set, used to implement the intelligent early warning method for a wind turbine generator set as claimed in any one of claims 1 to 4, characterized in that: It includes data acquisition equipment, communication equipment, control center and mobile terminal; the data acquisition equipment and mobile terminal are both connected to the control center through the communication equipment; the data acquisition equipment is used to collect all kinds of monitoring data of wind turbine generator sets; The control center is used to receive the monitoring data collected by the data acquisition device, and run the early warning model to provide fault early warning for the wind turbine generator set and send the fault early warning information to the mobile terminal; the mobile terminal is used to receive and remind the fault information of the wind turbine generator set.

6. The intelligent early warning system for wind turbine generator set according to claim 5, characterized in that: The data acquisition equipment at least includes the acquisition sensor in the SCADA system and the temperature sensor of the motor and the gear box.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that: When the processor executes the computer program, the intelligent early warning method for wind turbine generator set according to any one of claims 1 to 4 is implemented.

8. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent early warning method for wind turbine generator set according to any one of claims 1 to 4 are implemented.