CN-121828117-B - Online intelligent monitoring method for verticality of tower barrel of fan

Abstract

The invention relates to the technical field of data processing, in particular to an online intelligent monitoring method for the verticality of a fan tower, which comprises the steps of obtaining comprehensive vibration data and multidimensional environment data at each monitoring time in a current monitoring time and a preset historical time, obtaining the influence degree of each environment data according to the multidimensional environment data and the data fluctuation characteristics of the comprehensive vibration data in the preset historical time, clustering the multidimensional environment data in the preset historical time to obtain normal comprehensive vibration data corresponding to each cluster, obtaining correction values corresponding to the current monitoring time according to the difference between the multidimensional environment data in each cluster and the multidimensional environment data at the current monitoring time, the normal comprehensive vibration data corresponding to each cluster and the influence degree of each environment data, and obtaining the verticality of the fan tower according to the correction values corresponding to each direction of each monitoring position in the fan tower at the current monitoring time, thereby improving the accuracy of the verticality monitoring of the fan tower.

Inventors

• MA SHAOLI
• LI GUIMIN
• JIANG XUEJIN

Assignees

• 深圳前海慧联科技发展有限公司

Dates

Publication Date

20260512

Application Date

20260313

Claims (7)

1. 1. An online intelligent monitoring method for the verticality of a fan tower is characterized by comprising the following steps: Aiming at any direction of any monitoring position in a fan tower, acquiring comprehensive vibration data of the current monitoring moment and the monitoring moment in a preset history period according to vibration data of each sampling moment in any direction, and acquiring multidimensional environment data of the current monitoring moment and the monitoring moment in the preset history period; Acquiring the influence degree of each environmental data according to the data fluctuation characteristics of the multidimensional environmental data and the comprehensive vibration data in a preset historical period, clustering the multidimensional environmental data in the preset historical period by utilizing the influence degree of each environmental data to acquire at least one cluster, and acquiring the normal comprehensive vibration data corresponding to each cluster according to the fluctuation characteristics of the vibration data corresponding to the monitoring moment of the multidimensional environmental data in each cluster; According to the difference between the multidimensional environmental data in each type of cluster and the multidimensional environmental data at the current monitoring moment, the normal comprehensive vibration data corresponding to each type of cluster and the influence degree of each environmental data, the influence degree of environmental factors at the current monitoring moment is obtained, and each vibration data corresponding to the current monitoring moment is corrected by utilizing the influence degree of the environmental factors to obtain a correction value of each vibration data; acquiring the verticality of the fan tower by using the correction value of each vibration data corresponding to each monitoring position in each direction of the fan tower at the current monitoring moment; the step of obtaining the comprehensive vibration data of the current monitoring time and each monitoring time in a preset history period according to the vibration data of each sampling time in any direction comprises the following steps: Constructing a time window with a preset length taking any monitoring time as a cut-off time according to any monitoring time in the current monitoring time and a preset history period of the current monitoring time, wherein the preset length is the interval time between the any monitoring time and the last monitoring time; Acquiring inclination angle monitoring data at any monitoring time, acquiring a vibration data gravity component at any sampling time by utilizing the inclination angle monitoring data aiming at any sampling time in the time window, and calculating a difference value between the vibration data at any sampling time and the vibration data gravity component to obtain target vibration data at any sampling time; Acquiring an average value of target vibration data at each sampling moment in the time window, and acquiring comprehensive vibration data at any monitoring moment; The method for obtaining the influence degree of each environmental data according to the data fluctuation characteristics of the multidimensional environmental data and the comprehensive vibration data in the preset history period comprises the following steps: The method comprises the steps of forming a multi-dimensional environment data sequence from multi-dimensional environment data under each monitoring time in a preset history period, dividing the multi-dimensional environment data sequence into at least two environment data sequences according to dimensions, and clustering the environment data sequences according to the environment data sequence corresponding to any environment data to obtain at least one class cluster; For any two environmental data in any type of cluster, acquiring the absolute value of the difference value of the comprehensive vibration data of the monitoring moment of the any two environmental data, and calculating the reciprocal of the absolute value of the difference value and a preset constant to obtain the comprehensive vibration data proximity degree of the any two environmental data; And acquiring the comprehensive vibration data proximity degree of each two environmental data in any type of cluster, correspondingly acquiring the average comprehensive vibration data proximity degree, acquiring the average value of the average comprehensive vibration data proximity degree corresponding to each type of cluster, and carrying out normalization processing to acquire the influence degree of any type of environmental data.
2. 2. The online intelligent monitoring method of the verticality of a tower of a fan according to claim 1, wherein the obtaining normal integrated vibration data corresponding to each type of cluster according to fluctuation characteristics of vibration data corresponding to monitoring moments where multidimensional environment data in each type of cluster are located comprises: Aiming at any cluster, acquiring the comprehensive vibration data credibility of the monitoring moment of each multi-dimensional environment data in any cluster according to the fluctuation characteristics of the vibration data corresponding to the monitoring moment of each multi-dimensional environment data in the any cluster; Acquiring the integrated vibration data reliability accumulated value of the monitoring moment of each multi-dimensional environmental data in any cluster, respectively calculating the ratio of the integrated vibration data reliability of the monitoring moment of each multi-dimensional environmental data to the integrated vibration data reliability accumulated value to obtain the weight of the integrated vibration data of the monitoring moment of each multi-dimensional environmental data, and carrying out weighted summation on the integrated vibration data of the monitoring moment of each multi-dimensional environmental data in any cluster to obtain the normal integrated vibration data corresponding to any cluster.
3. 3. The online intelligent monitoring method of the verticality of a tower of a fan according to claim 2, wherein the obtaining the integrated vibration data reliability of the monitoring time of each multi-dimensional environmental data in any cluster according to the fluctuation characteristics of the vibration data corresponding to the monitoring time of each multi-dimensional environmental data in any cluster comprises: And aiming at the monitoring moment of any multi-dimensional environmental data in any cluster, acquiring the standard deviation of the target vibration data at each sampling moment in a time window corresponding to the monitoring moment of any multi-dimensional environmental data, and carrying out normalization processing on the inverse of the addition result of the standard deviation and a preset constant to obtain the reliability degree of the comprehensive vibration data of the monitoring moment of any multi-dimensional environmental data.
4. 4. The online intelligent monitoring method of the verticality of a tower of a fan according to claim 1, wherein the obtaining the influence degree of the environmental factors at the current monitoring time according to the difference between the multidimensional environmental data in each cluster and the multidimensional environmental data at the current monitoring time, the normal integrated vibration data corresponding to each cluster, and the influence degree of each environmental data comprises: aiming at any environmental data, acquiring the environmental proximity degree of each type of cluster except any environmental data and the current monitoring moment according to the difference between the multi-dimensional environmental data in each type of cluster and the multi-dimensional environmental data at the current monitoring moment; Acquiring an environmental data average value of any environmental data in each type of cluster, taking the environmental data average value of any environmental data in each type of cluster as an abscissa, taking normal comprehensive vibration data corresponding to each type of cluster as an ordinate, constructing a graph, and fitting data points in the graph by utilizing the proximity degree of each type of cluster except any environmental data to other environments at the current monitoring moment to obtain a fitting curve; Acquiring reference environmental data of any environmental data, acquiring a fitting value of the reference environmental data by using the fitting curve to obtain reference normal comprehensive vibration data, recording the environmental data corresponding to any environmental data at the current monitoring moment as current environmental data, acquiring the fitting value of the current environmental data by using the fitting curve to obtain current normal comprehensive vibration data, and calculating a difference value between the current normal comprehensive vibration data and the reference normal comprehensive vibration data to obtain a normal comprehensive vibration data difference value of any environmental data at the current monitoring moment; And acquiring a normal comprehensive vibration data difference value of each environmental data at the current monitoring time, and acquiring the influence degree of environmental factors at the current monitoring time according to the normal comprehensive vibration data difference value of each environmental data at the current monitoring time and the influence degree of each environmental data.
5. 5. The online intelligent monitoring method of the verticality of a fan tower according to claim 4, wherein the obtaining the proximity of each cluster except any one of the environmental data to the other environment at the current monitoring time according to the difference between the multi-dimensional environmental data in each cluster and the multi-dimensional environmental data at the current monitoring time comprises: aiming at any cluster, the multidimensional environment data in the cluster form a multidimensional environment data subsequence, and the multidimensional environment data subsequence is divided into at least two environment data subsequences according to dimensions; For any environmental data subsequence corresponding to the environmental data, marking the environmental data subsequences except the environmental data subsequence as other environmental data subsequences, and for any other environmental data subsequence, marking the environmental data with the same kind as the any other environmental data subsequence in the multi-dimensional environmental data at the current monitoring time as corresponding environmental data, and obtaining an average value of the any other environmental data subsequence and an absolute value of a difference value of the corresponding environmental data to obtain an environmental difference value of the any other environmental data subsequence; And obtaining an environmental difference value of each other environmental data subsequence, correspondingly obtaining an environmental difference value accumulated value, and carrying out normalization processing on the reciprocal of the added result of the environmental difference value accumulated value and a preset constant to obtain the other environmental proximity degree of any cluster except any environmental data and the current monitoring moment.
6. 6. The online intelligent monitoring method of the verticality of a tower of a fan according to claim 4, wherein the obtaining the influence degree of the environmental factors at the current monitoring time according to the difference value of the normal integrated vibration data of each environmental data at the current monitoring time and the influence degree of each environmental data comprises: For any environmental data, obtaining a product of a difference value of normal comprehensive vibration data of the any environmental data at the current monitoring moment and the influence degree of the any environmental data to obtain a weighted influence degree of the any environmental data; and obtaining the average value of the weighted influence degree of each environmental data, and obtaining the influence degree of the environmental factors at the current monitoring moment.
7. 7. The online intelligent monitoring method of the verticality of a tower of a fan according to claim 1, wherein the correcting the vibration data corresponding to the current monitoring time by using the influence degree of the environmental factors to obtain the correction value of each vibration data comprises the following steps: And aiming at any vibration data corresponding to the current monitoring moment, acquiring a difference value between target vibration data of the sampling moment of any vibration data and the influence degree of the environmental factors, and obtaining a correction value of any vibration data.

Description

Online intelligent monitoring method for verticality of tower barrel of fan Technical Field The invention relates to the technical field of data processing, in particular to an online intelligent monitoring method for the verticality of a fan tower. Background The wind turbine tower is a bearing structure for supporting the cabin and the impeller of the wind turbine, if the perpendicularity of the tower is greatly deviated, the running unbalance of the unit can be caused, the power generation efficiency of the wind turbine is influenced, the tower can be caused to generate additional structural stress, the structural stability is influenced, and even the risk of accidents such as tower inversion exists. Therefore, the method and the device monitor the verticality of the tower in real time, and have important significance for improving the safety and reliability of the wind turbine generator. The existing method for monitoring the verticality of the tower barrel of the fan mainly utilizes monitoring equipment such as an inclinometer, an acceleration sensor and the like to acquire vibration data of the tower barrel in real time, and utilizes the vibration data of the tower barrel to calculate the verticality of the tower barrel. However, the vibration data of the tower is easily affected by environmental factors such as wind speed, wind direction, temperature, humidity and the like, so that the vibration data cannot accurately reflect the true verticality of the tower. In order to reduce the influence of environmental factors, the prior art generally adopts a fixed threshold method to reject part of tower vibration data, or utilizes a simple linear model to perform environmental compensation on the tower vibration data. However, the combination of multiple environmental factors and the complexity of environmental changes make it difficult to accurately represent the influence of the environment on the tower vibration data by using a fixed threshold or linear model, so that the final tower verticality still has deviation. Therefore, how to reduce the influence of environmental factors on the vibration data of the tower, and to improve the accuracy of monitoring the verticality of the tower become a urgent problem to be solved. Disclosure of Invention In view of the above, the embodiment of the invention provides an online intelligent monitoring method for the verticality of a tower of a fan, which aims to solve the problems of reducing the influence of environmental factors on vibration data of the tower and improving the accuracy of monitoring the verticality of the tower. The embodiment of the invention provides an online intelligent monitoring method for the verticality of a fan tower, which comprises the following steps: Aiming at any direction of any monitoring position in a fan tower, acquiring comprehensive vibration data of the current monitoring moment and the monitoring moment in a preset history period according to vibration data of each sampling moment in any direction, and acquiring multidimensional environment data of the current monitoring moment and the monitoring moment in the preset history period; Acquiring the influence degree of each environmental data according to the data fluctuation characteristics of the multidimensional environmental data and the comprehensive vibration data in a preset historical period, clustering the multidimensional environmental data in the preset historical period by utilizing the influence degree of each environmental data to acquire at least one cluster, and acquiring the normal comprehensive vibration data corresponding to each cluster according to the fluctuation characteristics of the vibration data corresponding to the monitoring moment of the multidimensional environmental data in each cluster; According to the difference between the multidimensional environmental data in each type of cluster and the multidimensional environmental data at the current monitoring moment, the normal comprehensive vibration data corresponding to each type of cluster and the influence degree of each environmental data, the influence degree of environmental factors at the current monitoring moment is obtained, and each vibration data corresponding to the current monitoring moment is corrected by utilizing the influence degree of the environmental factors to obtain a correction value of each vibration data; And acquiring the verticality of the fan tower by using the corrected value of each vibration data corresponding to each monitoring position in each direction of the fan tower at the current monitoring moment. Preferably, the obtaining, according to the vibration data at each sampling time in any direction, the comprehensive vibration data at each monitoring time in the current monitoring time and the preset history period thereof includes: Constructing a time window with a preset length taking any monitoring time as a cut-off time according to any monitoring time in the curr