阅读说明：本技术 一种基于异常检测的风电机组温度预警系统 (wind turbine generator system temperature early warning system based on anomaly detection ) 是由 宋光雄 张蒙 于 2019-08-01 设计创作，主要内容包括：本发明公开了一种基于异常检测的风电机组温度预警系统,其包括：工业大数据平台,包括有风电机组SCADA运行数据模块、实施数据ETL模块,通过对比分析,检测基于异常检测的温度预警,着重进行基于齿轮箱、轴承温度时变数据模型异常检测；监测平台,用于配合工业大数据平台进行风电机组各方位的检测；特征提取平台,用于配合工业大数据平台进行风电机组的各个特征数据提取；温度评估平台,调取工业大数据平台数据和实际风电机组温度数据,对风电机组的温度变化进行评估；温度预测平台,用于配合工业大数据平台的数据,对风电机组的温度进行预测；温度诊断平台,用于配合工业大数据平台的数据,对风电机组的温度进行诊断。(The invention discloses a wind turbine generator temperature early warning system based on anomaly detection, which comprises: the industrial big data platform comprises a wind turbine generator SCADA operation data module and an implementation data ETL module, detects temperature early warning based on abnormal detection through comparative analysis, and emphasizes on abnormal detection based on a gear box and bearing temperature time-varying data model; the monitoring platform is used for matching with an industrial big data platform to detect all directions of the wind turbine generator; the characteristic extraction platform is used for matching with an industrial big data platform to extract each characteristic data of the wind turbine generator; the temperature evaluation platform is used for calling industrial big data platform data and actual wind turbine temperature data and evaluating the temperature change of the wind turbine; the temperature prediction platform is used for predicting the temperature of the wind turbine generator set in cooperation with data of the industrial big data platform; and the temperature diagnosis platform is used for diagnosing the temperature of the wind turbine generator by matching with the data of the industrial big data platform.)

1. the utility model provides a wind turbine generator system temperature early warning system based on anomaly detection which characterized in that, it includes:

the industrial big data platform comprises a wind turbine generator SCADA operation data module and an implementation data ETL module, detects temperature early warning based on abnormal detection through comparative analysis, fully considers the dynamic change characteristics of temperature data of a gear box and a bearing, and emphasizes on abnormal detection based on a time-varying data model of the gear box and the bearing;

The monitoring platform is used for matching with an industrial big data platform to detect all directions of the wind turbine generator;

The characteristic extraction platform is used for matching with an industrial big data platform to extract each characteristic data of the wind turbine generator;

The temperature evaluation platform is used for calling industrial big data platform data and actual wind turbine temperature data and evaluating the temperature change of the wind turbine;

The temperature prediction platform is used for predicting the temperature of the wind turbine generator set in cooperation with data of the industrial big data platform;

And the temperature diagnosis platform is used for diagnosing the temperature of the wind turbine generator by matching with the data of the industrial big data platform.

2. The wind turbine generator temperature early warning system based on anomaly detection according to claim 1, characterized in that: the monitoring platform comprises a vibration detection module, a noise detection module, a temperature detection module, a pressure detection module, a voltage detection module and an acoustic emission detection module.

3. The wind turbine generator temperature early warning system based on anomaly detection according to claim 1, characterized in that: the device comprises a vibration data extraction module, a noise data extraction module, a temperature data extraction module, a pressure data extraction module, a voltage data extraction module and an acoustic emission data extraction module.

4. The wind turbine generator temperature early warning system based on anomaly detection according to claim 1, characterized in that: the temperature evaluation platform comprises a logistic regression module, a feature mapping module, a statistical pattern recognition module, a physical mechanism model module and a temperature evaluation module.

5. the wind turbine generator temperature early warning system based on anomaly detection according to claim 1, characterized in that: the temperature prediction platform comprises a time sequence module, a composite matching matrix module, a temperature network module, a time-frequency domain analysis module and a temperature prediction module.

6. the wind turbine generator temperature early warning system based on anomaly detection according to claim 1, characterized in that: the temperature diagnosis platform comprises a statistical diagnosis module, a time-frequency domain diagnosis module, an information theory analysis module, an integrated diagnosis module and a temperature diagnosis module.

Technical Field

The invention relates to the technical field of temperature early warning, in particular to a wind turbine generator temperature early warning system based on anomaly detection.

background

At present, most wind turbine generator gearbox fault early warning mechanisms are single, if the temperature of an oil pool is set to be higher than 75 ℃, the wind turbine generator gearbox is alarmed, and the wind turbine generator gearbox is shut down when the temperature is higher than 80 ℃; the traditional regulations stipulate that the maximum temperature of the rolling bearing does not exceed 95 ℃ (threshold); the maximum temperature of the sliding bearing does not exceed 80 ℃ (threshold); the temperature rise did not exceed 55 deg.c (threshold). The technology for quantitatively analyzing, monitoring and identifying the temperature abnormality of industrial equipment at the industrial pain point is seriously lagged, and is limited to threshold-based discrimination for a long time. In fact, temperature anomalies in the field of industrial production, including wind turbine plants, are widespread. The temperature management of the device is often limited to the setting and application of temperature thresholds (alarm values, shutdown values).

Typical research work in recent years includes establishing a calculation model of the heat generation amount of the counter-rotating bearing, and performing interaction analysis between bearing elements and bearing heat generation analysis by using a pseudo-kinetic analysis method. A high-speed ball bearing transient thermal calculation model is established by using a heat grid method, and the influence rule of working condition parameters on bearing heat generation quantity, temperature and heat induction load is researched. The effects of reliability, temperature, lubricants and additives, surface roughness, materials, load distribution, hoop stress and interfacial sliding on the fatigue life of the rolling bearing were investigated. Rolling and sliding friction between the rolling element and the inner and outer raceways, sliding friction between the cage and the guide surface of the ferrule, sliding friction between the rolling element and the cage pocket, sliding friction between the roller end surface and the flange, and lubricant viscous friction. Besides self-heating, the bearing system has a mutual heat transfer process with the outside of the system. If the heat generated in the bearing can not be effectively dissipated in time, the temperature of the bearing is abnormally increased along with the continuous accumulation of the heat in the bearing, and the viscosity of the lubricating oil is reduced. The method comprises the steps of analyzing the association degree of lubricating grease temperature, friction torque and vibration acceleration signals and faults by carrying out an accelerated fatigue life test on a certain wind power turntable bearing. SCADA data analysis of the wind turbine indicates that the efficiency of the gear box, the temperature of lubricating oil and the degradation of oil products have corresponding relations.

the temperature performance of the gear box and the bearing is related to the operation process, and is also closely related to damage development and fault evolution. The early failure of the rolling bearing has the characteristics of unobvious characteristics, weak signals, low signal-to-noise ratio, difficult failure identification and the like. For a main shaft bearing, a low-speed shaft bearing of a gear box, a yaw bearing and a pitch bearing, the vibration monitoring effect is limited. The low-speed shaft bearing of the gear box can adopt an online monitoring method of lubricating oil, but for a main shaft bearing, a yaw bearing and a pitch bearing, lubricating grease lubrication or lubricating grease and lubricating oil mixed lubrication is mostly adopted, the online monitoring method is difficult to adopt, offline sample collection is also difficult to ensure that a sample participates in the lubrication work, and the monitoring effect is difficult to ensure. Therefore, the research work of early warning and prediction of the temperature of the gearbox and the bearing of the wind turbine generator is carried out, and the method has great application development space and necessity.

disclosure of Invention

In order to overcome the problems, the invention provides a wind turbine generator temperature early warning system based on abnormal detection.

The technical scheme of the invention is to provide a wind turbine generator temperature early warning system based on abnormal detection, which is characterized by comprising the following components:

The industrial big data platform comprises a wind turbine generator SCADA operation data module and an implementation data ETL module, detects temperature early warning based on abnormal detection through comparative analysis, fully considers the dynamic change characteristics of temperature data of a gear box and a bearing, and emphasizes on abnormal detection based on a time-varying data model of the gear box and the bearing;

The monitoring platform is used for matching with an industrial big data platform to detect all directions of the wind turbine generator;

The characteristic extraction platform is used for matching with an industrial big data platform to extract each characteristic data of the wind turbine generator;

The temperature evaluation platform is used for calling industrial big data platform data and actual wind turbine temperature data and evaluating the temperature change of the wind turbine;

The temperature prediction platform is used for predicting the temperature of the wind turbine generator set in cooperation with data of the industrial big data platform;

and the temperature diagnosis platform is used for diagnosing the temperature of the wind turbine generator by matching with the data of the industrial big data platform.

Furthermore, the monitoring platform comprises a vibration detection module, a noise detection module, a temperature detection module, a pressure detection module, a voltage detection module and an acoustic emission detection module.

Furthermore, the device comprises a vibration data extraction module, a noise data extraction module, a temperature data extraction module, a pressure data extraction module, a voltage data extraction module and an acoustic emission data extraction module.

Furthermore, the temperature evaluation platform comprises a logistic regression module, a feature mapping module, a statistical pattern recognition module, a physical mechanism model module and a temperature evaluation module.

Furthermore, the temperature prediction platform comprises a time sequence module, a composite matching matrix module, a temperature network module, a time-frequency domain analysis module and a temperature prediction module.

furthermore, the temperature diagnosis platform comprises a statistical diagnosis module, a time-frequency domain diagnosis module, an information theory analysis module, an integrated diagnosis module and a temperature diagnosis module.

the invention has the beneficial effects that: the wind turbine temperature early warning system based on the anomaly detection is based on wind turbine SCADA operation data, and a wind turbine key component temperature early warning system oriented to an industrial large data platform is researched and developed to provide accurate, reliable and effective basis for operation and maintenance decision of wind turbine equipment. According to the invention, large-scale wind turbine SCADA operation data are utilized, a temperature anomaly detection model is optimized, a temperature anomaly standard threshold value acquisition method is combined, the dynamic change characteristics of temperature data of a gear box and a bearing are fully considered, a temperature early warning method is researched, a temperature early warning data model is designed and established, and a temperature early warning algorithm of key components such as the gear box and the bearing is repeatedly verified and optimized; the method fully combines the temperature anomaly occurrence and development rules of key components such as a gear box and a bearing, researches and defines evaluation indexes and makes evaluation strategies, ensures the consistency of temperature anomaly detection quantitative parameters and state degradation quantitative evaluation, develops a quantitative evaluation algorithm of the degradation process of the key components, and provides accurate, reliable and effective basis for operation and maintenance decision of wind power equipment.

Detailed Description

in order to make the technical means, the creation characteristics, the achievement purposes and the effects of the invention easy to understand, the invention is further described with the specific embodiments.

The invention discloses a wind turbine generator temperature early warning system based on anomaly detection, which is characterized by comprising the following components:

The industrial big data platform comprises a wind turbine generator SCADA operation data module and an implementation data ETL module, detects temperature early warning based on abnormal detection through comparative analysis, fully considers the dynamic change characteristics of temperature data of a gear box and a bearing, and emphasizes on abnormal detection based on a time-varying data model of the gear box and the bearing;

The monitoring platform is used for matching with an industrial big data platform to detect all directions of the wind turbine generator;

The characteristic extraction platform is used for matching with an industrial big data platform to extract each characteristic data of the wind turbine generator;

The temperature evaluation platform is used for calling industrial big data platform data and actual wind turbine temperature data and evaluating the temperature change of the wind turbine;

The temperature prediction platform is used for predicting the temperature of the wind turbine generator set in cooperation with data of the industrial big data platform;

and the temperature diagnosis platform is used for diagnosing the temperature of the wind turbine generator by matching with the data of the industrial big data platform.

in a preferred embodiment of the present invention, the monitoring platform comprises a vibration detection module, a noise detection module, a temperature detection module, a pressure detection module, a voltage detection module, and an acoustic emission detection module.

in a preferred embodiment of the present invention, the vibration data extraction module, the noise data extraction module, the temperature data extraction module, the pressure data extraction module, the voltage data extraction module, and the acoustic emission data extraction module are provided.

in a preferred embodiment of the present invention, the temperature evaluation platform comprises a logistic regression module, a feature mapping module, a statistical pattern recognition module, a physical mechanism model module, and a temperature evaluation module.

in a preferred embodiment of the present invention, the temperature prediction platform comprises a time sequence module, a composite matching matrix module, a temperature network module, a time-frequency domain analysis module, and a temperature prediction module.

In a preferred embodiment of the present invention, the temperature diagnosis platform comprises a statistical diagnosis module, a time-frequency domain diagnosis module, an information theory analysis module, an integrated diagnosis module, and a temperature diagnosis module.

The above embodiment is only one embodiment of the present invention, and the description thereof is specific and detailed, but not construed as limiting the scope of the present invention. It should be noted that, for a person skilled in the art, several variations and modifications can be made without departing from the inventive concept, which falls within the scope of the present invention. Therefore, the protection scope of the present patent shall be subject to the appended claims.