On Line Prediction Technology For Inner Winding Temperature Rise Of EHV Power Transformer

Transformer is one of the most important and expensive equipment in power grid.Its safe and reliable operation is directly related to whether the power transmission and transformation network is running stably.The end of the life of a power transformer is due to the loss of its due insulation,and the main factor affecting the insulation is the temperature of the winding during the operation of the transformer,that is,the hot temperature of the winding.The winding temperature is an important factor that affects the load capacity and the insulation life of the winding,and it is also an important index of the design of the transformer winding.So monitoring.It is important to predict the hot spot temperature of transformer windings for estimating the life of transformers,ensuring the safe and reliable operation of the system,and improving the economic benefits of stable operation.The main work of this article is as follows:(1)In view of the structure of oil immersed transformer,the mechanism of heat generation inside the oil immersed transformer and the transfer process of heat inside the transformer are studied.At the same time,on the basis of the analysis of the heat transfer path,the flow of the oil flow in the vertical flow channel of the vertical flow channel inside the coil is further analyzed,and the formulas for calculating the heat transfer coefficient of the convection are obtained.It lays the foundation for the theory of numerical computation model put forward below.(2)Through the analysis of the internal oil flow in the transformer,the temperature calculation model of the inner winding of the transformer based on the thermoelectric analogy and the IEEE guide model is put forward,and the model is simulated.The comparison between the simulation prediction value and the IEEE guide model value shows that the simulation value of the hot spot temperature of the transformer winding and the IEEE guide model value have good unity under the condition of the under load,the rated load and the overload.(3)To study the current research status of optical fiber temperature sensor,design a specific scheme of fiber Bragg grating temperature measurement according to the specific requirements of internal temperature measurement in transformer and the simulation model of temperature calculation in transformer inner winding,and choose the appropriate hardware,and build the fiber based on the test transformer of oil immersion temperature rise test.Grating temperature measurement platform.Fiber Bragg grating temperature sensor is applied to measure the heating condition of transformer under the condition of under load,rated load and changing load.And the related curves are drawn.The experimental results show that the fiber grating temperature sensor is accurate and stable,and can reflect the change of temperature rapidly,and can effectively measure the internal temperature of the transformer.On the basis of measurement and monitoring,the temperature of the inner winding of transformer is predicted by artificial neural network based on L-M algorithm,and the temperature test data are obtained.The data of the test are divided into two groups.One group is used to train the improved neural network and the other is used to verify the result of the experiment.The prediction results show that the BP neural network based on L-M algorithm can better predict the hot spot temperature of transformer winding.