Unit online monitoring method based on big data

Abstract

The invention discloses a unit online monitoring method based on big data, and the method comprises the steps: firstly enabling real-time data to be directly connected to a single-point evaluation model during monitoring, and quickly detecting the operation state of equipment or whether the data is noise through the single-point evaluation model; or when the single-point evaluation model judges that the equipment operates and the real-time data is non-noise data, the real-time data can be accessed to the abnormal data model, and the abnormal data model is established based on the abnormal historical data, so that when the real-time data and the abnormal data model are successfully matched, the system can quickly and accurately send out an equipment exception signal, so that a worker can check and overhaul an exceptional state, and the fault monitoring efficiency and accuracy of the monitoring system are improved.

Description

technical field [0001] The invention relates to the industrial application field of generator sets, in particular to an on-line monitoring method for generator sets based on big data. Background technique [0002] my country's wind power generation is developing rapidly, but it faces the challenge of frequent failures in the initial stage. The wind power industry expects to transform from extensive operation to intensive operation, which not only needs to increase the installed capacity, but also needs to achieve high output and high efficiency. The reliability of the fan is low, and due to the lack of fault early warning function, minor faults cannot be found and repaired and developed into major safety and equipment accidents, which not only cause downtime losses, but also expensive maintenance costs. There have been reports in foreign countries that a wind turbine manufacturer went bankrupt because of the warranty of its fault-prone products, which caused the enterprise ...

AI Technical Summary

Problems solved by technology

[0004] The main purpose of the present invention is to solve the problem that when the real-time data does not match the model because only the health status monitoring model is integrated in the monitoring system, the system will determine the equipment abnormality through the elimination method, which will affect the sensitivity and accuracy of the health status monitoring system of the wind power generating set Low problem, and provide an online unit monitoring method based on big data

Images