Tower vibration anomaly detection and early warning method and system

Abstract

The application discloses a tower vibration anomaly detection and early warning method and system, and the method comprises the following steps: S1, collecting tower vibration data in a preset time period and preprocessing the data to obtain processed vibration data; S2, obtaining spatial anomaly degree and time anomaly degree according to the processed vibration data; S3, calculating vibration anomaly degree by weighting according to the spatial anomaly degree and the time anomaly degree; and S4, comparing the vibration anomaly degree with a preset threshold value, and judging the state of the tower according to the comparison result. The application fully considers the common characteristics of the whole field unit and the individual differences between units, constructs two standards of spatial dimension and time dimension, respectively calculates the anomaly degrees of the two dimensions, then calculates the unit vibration anomaly degree by weighting, and the obtained result is more referable; the standard curves of the spatial dimension and the time dimension are calculated by using the warehouse calculation method, the interpretability is relatively strong, and the method has great advantages for industrial scene model landing.

Description

Technical Field

[0001] This invention mainly relates to the field of wind power technology, specifically to a method and system for detecting and warning of abnormal tower vibration. Background Technology

[0002] Wind turbine generators are large mechanical devices, mainly composed of blades, a transmission system, and a tower. Due to their long-term location in areas with abundant wind energy resources but harsh environments, their components age rapidly. In recent years, tower collapse accidents have become increasingly common, and post-accident analyses in most cases have concluded that abnormal tower vibrations existed days or even weeks before the collapse. However, existing vibration anomaly detection methods only consider the turbine itself, ignoring the time dimension, and suffer from poor interpretability, making it difficult to pinpoint the root cause of problems. Summary of the Invention

[0003] The technical problem to be solved by the present invention is to provide a tower vibration anomaly detection and early warning method and system with strong interpretability and more referential results, in view of the technical problems existing in the prior art.

[0004] To solve the above-mentioned technical problems, the technical solution proposed by this invention is as follows:

[0005] A method for detecting and warning of abnormal tower vibration includes the following steps:

[0006] S1. Collect tower vibration data for a preset time period and preprocess it to obtain processed vibration data;

[0007] S2. Obtain spatial anomaly and temporal anomaly based on the processed vibration data;

[0008] S3. Calculate the vibration anomaly degree by weighting the spatial anomaly degree and the temporal anomaly degree;

[0009] S4. Compare the vibration anomaly with the preset threshold, and determine the state of the tower based on the comparison result.

[0010] Preferably, in step S1, the preprocessing includes:

[0011] Calculate the speed-vibration curve for each unit at each time period, divide the data into compartments, and calculate the vibration RMS data for each compartment.

[0012] Preferably, in step S1, the preprocessing further includes outlier processing, data resampling, and removal of abnormal operating condition data for the time series data of tower vibration.

[0013] Preferably, in step S2, the specific process of obtaining the spatial anomaly degree is as follows:

[0014] S211. Collect the vibration RMS data of each compartment of the entire unit together, filter out outliers, calculate the reference point of each compartment, and stitch the reference points together in order of speed from small to large to form a spatial standard baseline.

[0015] S212. Based on the spatial standard baseline and the speed-vibration curve of each unit, calculate the vibration difference of each unit by means of differential integral calculation.

[0016] S213. Convert vibration difference into spatial anomaly.

[0017] Preferably, in step S2, the specific process for obtaining the time anomaly degree is as follows:

[0018] S221. Select the speed-vibration curve of a unit in the most recent period, filter out the outliers, calculate the reference point of each unit, and stitch the reference points together to form a time standard curve.

[0019] S222. Based on the time standard baseline and the speed-vibration curve of each unit, calculate the vibration difference of each unit by means of differential integral calculation.

[0020] S223. Convert vibration difference into time anomaly.

[0021] Preferably, in step S4, the preset threshold corresponds to the health level of the tower.

[0022] Preferably, when the vibration anomaly degree is within the preset threshold of 0.8 to 1.0, it corresponds to the tower's health level being severely abnormal;

[0023] When the vibration anomaly is within the preset threshold of 0.6 to 0.8, it indicates that the health level of the tower is abnormal.

[0024] When the vibration anomaly is within the preset threshold of 0.3 to 0.6, it indicates that the health level of the tower is sub-healthy.

[0025] When the vibration anomaly is within the preset threshold of 0 to 0.3, it indicates that the health level of the tower is healthy.

[0026] Preferably, in step S1, the frequency of the tower vibration data is greater than 1 Hz, and the preset power range is greater than 0 kW.

[0027] The present invention also discloses a computer-readable storage medium having a computer program stored thereon, the computer program performing the steps of the method described above when run by a processor.

[0028] The present invention further discloses a computer device including a memory and a processor interconnected thereon, wherein the memory stores a computer program that, when run by the processor, performs the steps of the method described above.

[0029] Compared with the prior art, the advantages of the present invention are as follows:

[0030] This invention, by deeply mining the vibration information in SCADA data and combining the characteristics of industrial big data with data anomaly detection, can detect abnormal vibration information of the unit several days in advance, enabling on-site maintenance personnel to carry out targeted maintenance and avoid accidents.

[0031] This invention fully considers the common characteristics of all units in the field and the individual differences between units, and constructs two standards, spatial dimension and time dimension, to calculate the anomaly degree of the two dimensions respectively. Then, the weighted calculation is used to obtain the unit vibration anomaly degree, and the result is more reliable.

[0032] This invention calculates the vibration anomaly degree of each generator unit in both time and space dimensions through anomaly data filtering and compartmentalized calculation. The final anomaly degree for each unit is obtained through a weighted average. Specifically, the rotational speed-vibration curve for each unit in each time period is calculated using compartmentalized calculation. Then, the standard curve for the entire field and the standard curve for each unit are calculated. The difference is calculated using integration, and finally, the difference is converted into anomaly indicators. This method fully considers the common characteristics of all units in the field as well as the individual differences between units. The compartmentalized calculation method for standard curves in both spatial and time dimensions provides strong interpretability and has significant advantages for the practical application of industrial scenario models. Attached Figure Description

[0033] Figure 1 This is a flowchart of an embodiment of the anomaly detection and early warning method of the present invention.

[0034] Figure 2 This is a graph showing the speed-vibration curves of all 33 units in the field for a certain month. Detailed Implementation

[0035] The present invention will be further described below with reference to the accompanying drawings and specific embodiments.

[0036] like Figure 1 As shown, the tower vibration anomaly detection and early warning method of this invention includes the following steps:

[0037] S1, Data Preprocessing

[0038] Collect tower vibration data for a preset time period (at least six months); the data acquisition frequency is greater than 1Hz, and a preset power range (greater than 0kW) is selected;

[0039] The time-series data of tower vibration were processed by outlier removal, data resampling, and removal of abnormal operating conditions to obtain the processing results.

[0040] The formula for outlier filtering is shown below:

[0041] P(μ-3σ <x≤μ+3σ)=λ%

[0042] The RMS (Root Mean Square) value of vibration data for each unit is calculated for each time period (one month is optional), as shown in the following formula:

[0043]

[0044] In the formula, a1, a2...a N This represents the vibration acceleration of a certain unit within 1 minute, and N represents the total number of units in the field.

[0045] like Figure 2 As shown, the speed-vibration curves for each unit are calculated for each time period (one month is optional). Each data point is divided into sections, and the average vibration RMS data (BIN) for each section is calculated. (i) The calculation process is as follows:

[0046] BIN (i) =Mean{a RMS v(i) |v (i) <v<v (i+1)}

[0047] In the formula, v (i) a represents the wind speed in the i-th wind chamber; RMS v(i) Let RMS represent the vibration data of the i-th wind speed chamber.

[0048] S2. Obtain spatial anomaly and temporal anomaly based on the processed vibration data;

[0049] S21. Spatial Anomaly Calculation

[0050] S211, Calculate the standard curve in space

[0051] The spatial standard curve represents the speed-vibration curve of all units in the current time period (generally the current month). The vibration RMS values ​​of each compartment of all units in the field are collected together, outliers are filtered out using the IQR method, and then the baseline IQR of each compartment is calculated, as shown in the following formula:

[0052] IQR = Q3 - Q1

[0053] (a RMS >Q1-IQR×1.5)&(aRMS <Q3+IQR×1.5)

[0054] In the formula, Q1 is the quarter quantile and Q3 is the third quantile;

[0055]

[0056] In the formula, N represents the total number of units in the field, and i represents each speed compartment; in the formula, This represents the benchmark value for the corresponding sub-account. This indicates the vibration value of the corresponding compartment for each unit;

[0057] Then, the reference points are spliced ​​together according to the rotational speed from small to large to form a spatial standard baseline;

[0058] S212, Calculate spatial anomaly

[0059] The spatial standard baseline and the speed-vibration curves for each unit have been calculated in the previous step. Now, the vibration difference of each unit is calculated by differential integration, and the specific formula is shown below:

[0060]

[0061] Where k represents the vibration difference of each unit.

[0062] S213. Convert vibration difference degree into spatial anomaly degree d space :

[0063]

[0064] In the formula, k i This indicates the vibration difference of the i-th unit; σ represents the average vibration difference of all units in the field; σ represents the variance of the vibration difference of all units in the field.

[0065] S22, Calculation of Time Anomalies

[0066] S221, Calculate the time standard curve

[0067] Select the speed-vibration compartment curves of a unit over a recent period (excluding the current period), filter out outliers, and calculate the baseline point for each unit.

[0068]

[0069] The reference points are stitched together to form a time standard curve, where M is the number of time intervals;

[0070] S222. Based on the time standard baseline and the speed-vibration curve of each unit, calculate the vibration difference of each unit by means of differential integral calculation.

[0071] S223. Convert vibration difference into time anomaly d. time .

[0072] S3, Weighted calculation of vibration anomaly degree

[0073] Vibration anomaly is calculated using a weighted average of spatial and temporal anomalies:

[0074] d = w space d space +w time d time

[0075] In the formula, w space w represents the spatial anomaly weight. time This indicates the weight of time anomalies.

[0076] S4. Threshold Setting and Alarm

[0077] By comparing the tower vibration anomaly results with the predicted anomaly thresholds shown in Table 1, the health level of the tower vibration can be determined based on the predicted anomaly threshold at which the tower vibration anomaly prediction results are located, and corresponding adjustments can be made to the tower vibration to ensure that the tower vibration is in a healthy state.

[0078] Table 1:

[0079] Anomaly Health level 0.8～1.0 Serious abnormality 0.6～0.8 abnormal 0.3～0.6 Sub-health 0～0.3 healthy

[0080] This invention, by deeply mining the vibration information in SCADA data and combining the characteristics of industrial big data with data anomaly detection, can detect abnormal vibration information of the unit several days in advance, enabling on-site maintenance personnel to carry out targeted maintenance and avoid accidents.

[0081] This invention fully considers the common characteristics of all units in the field and the individual differences between units, and constructs two standards, spatial dimension and time dimension, to calculate the anomaly degree of the two dimensions respectively. Then, the weighted calculation is used to obtain the unit vibration anomaly degree, and the result is more reliable.

[0082] This invention calculates the vibration anomaly degree of each generator unit in both time and space dimensions through anomaly data filtering and compartmentalized calculation. The final anomaly degree for each unit is obtained through a weighted average. Specifically, the rotational speed-vibration curve for each unit in each time period is calculated using compartmentalized calculation. Then, the standard curve for the entire field and the standard curve for each unit are calculated. The difference is calculated using integration, and finally, the difference is converted into anomaly indicators. This method fully considers the common characteristics of all units in the field as well as the individual differences between units. The compartmentalized calculation method for standard curves in both spatial and time dimensions provides strong interpretability and has significant advantages for the practical application of industrial scenario models.

[0083] This invention also provides a computer-readable storage medium storing a computer program thereon, which, when run by a processor, performs the steps of the method described above. This invention further provides a computer device including a memory and a processor interconnected, wherein the memory stores a computer program that, when run by a processor, performs the steps of the method described above. The medium and system of this invention, corresponding to the methods described above, also possess the advantages described above.

[0084] The present invention can implement all or part of the processes in the methods of the above embodiments, or it can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium. When the computer program is executed by a processor, it can implement the steps of the above method embodiments. The computer program includes computer program code, which can be in the form of source code, object code, executable file, or some intermediate form. Computer-readable media include: any entity or device capable of carrying computer program code, recording media, USB flash drives, portable hard drives, magnetic disks, optical disks, computer memory, read-only memory (ROM), random access memory (RAM), electrical carrier signals, telecommunication signals, and software distribution media, etc. The memory is used to store computer programs and / or modules. The processor implements various functions by running or executing the computer programs and / or modules stored in the memory, and by calling data stored in the memory. The memory may include high-speed random access memory, as well as non-volatile memory, such as hard disks, RAM, plug-in hard disks, smart media cards (SMC), secure digital (SD) cards, flash cards, at least one disk storage device, flash memory device, or other volatile solid-state storage devices.

[0085] The above are merely preferred embodiments of the present invention. The scope of protection of the present invention is not limited to the above embodiments. All technical solutions falling within the scope of the present invention's concept are within the scope of protection of the present invention. It should be noted that for those skilled in the art, any improvements and modifications made without departing from the principles of the present invention should be considered within the scope of protection of the present invention.

Claims

1. A method for detecting and warning of abnormal tower vibration, characterized in that, Including the following steps: S1. Collect tower vibration data for a preset time period and preprocess it to obtain processed vibration data; S2. Obtain spatial anomaly and temporal anomaly based on the processed vibration data; S3. Calculate the vibration anomaly degree by weighting the spatial anomaly degree and the temporal anomaly degree; S4. Compare the vibration anomaly with the preset threshold, and determine the state of the tower based on the comparison result; In step S2, the specific process of obtaining the spatial anomaly degree is as follows: S211. Collect the vibration RMS data of each compartment of the entire unit together, filter out outliers, calculate the reference point of each compartment, and stitch the reference points together in order of speed from small to large to form a spatial standard baseline. S212. Based on the spatial standard baseline and the speed-vibration curve of each unit, calculate the vibration difference of each unit by means of differential integral calculation. S213. Convert vibration difference into spatial anomaly. In step S2, the specific process of obtaining the time anomaly degree is as follows: S221. Select the speed-vibration curve of a unit in the most recent period, filter out the outliers, calculate the reference point of each unit, and stitch the reference points together to form a time standard curve. S222. Based on the time standard baseline and the speed-vibration curve of each unit, calculate the vibration difference of each unit by means of differential integral calculation. S223. Convert vibration difference into time anomaly. Vibration anomaly degree in step S3 The calculation formula is: In the formula, Indicates the spatial anomaly weight. Indicates the weight of time anomalies; Indicates spatial anomaly; Indicates the degree of time anomaly.

2. The tower vibration anomaly detection and early warning method according to claim 1, characterized in that, In step S1, the preprocessing includes: Calculate the speed-vibration curve for each unit at each time period, divide the data into compartments, and calculate the vibration RMS data for each compartment.

3. The tower vibration anomaly detection and early warning method according to claim 1, characterized in that, In step S1, the preprocessing also includes outlier processing, data resampling, and removal of abnormal operating condition data for the time series data of tower vibration.

4. The tower vibration anomaly detection and early warning method according to claim 1, 2, or 3, characterized in that, In step S4, the preset threshold corresponds to the health level of the tower.

5. The tower vibration anomaly detection and early warning method according to claim 4, characterized in that, When the vibration anomaly is within the preset threshold of 0.8 to 1.0, it indicates that the health level of the tower is severely abnormal. When the vibration anomaly is within the preset threshold of 0.6 to 0.8, it indicates that the health level of the tower is abnormal. When the vibration anomaly is within the preset threshold of 0.3 to 0.6, it indicates that the health level of the tower is sub-healthy. When the vibration anomaly is within the preset threshold of 0 to 0.3, it indicates that the health level of the tower is healthy.

6. The tower vibration anomaly detection and early warning method according to claim 1, 2, or 3, characterized in that, In step S1, the frequency of the tower vibration data is greater than 1 Hz, and the preset power range is greater than 0 kW.

7. A computer-readable storage medium having a computer program stored thereon, characterized in that, The computer program, when run by a processor, performs the steps of the method as described in any one of claims 1 to 6.

8. A computer device comprising a memory and a processor interconnected thereon, the memory storing a computer program, characterized in that, The computer program, when run by a processor, performs the steps of the method as described in any one of claims 1 to 6.

Images