| Power transformer is an important power system element,which can transmit electricity efficiently and reduce the series voltage.Its stability performance plays an important role in the operation of the power system,because the loss of a key unit will have a great influence on the safety,reliability and cost of the power supply.It is the load and the internal temperature,especially the hot temperature(HST)and the top oil temperature(TOT),which all affect the insulation aging of the transformer,thus affecting the service life of the equipment.In this paper,the current status of the current state monitoring and control system of power transformer is studied.The status monitoring of the transformer is mainly focused on the monitoring of the hot spot temperature and load of the transformer.Because the hot temperature of the transformer winding is difficult to measure in the actual project,it is estimated in the same way according to the thermodynamic model of the transformer.The control of transformer is mainly focused on the control of transformer cooling system and load.Based on the analysis and research of existing theory and technology,this paper predicts the hot temperature of transformer and uses the least square support vector machine(LS-SVM)estimation method based on nonlinear kernel function.Control,the fuzzy proportional integral differential(PID)modulation pulse width modulation(PWM)control method is used to adjust the working state of the cooling fan group according to the monitoring temperature of the transformer.For the on-line monitoring of the state of the power transformer,a wireless communication scheme based on GPRS is adopted to upload the collected transformer status data to the server.This paper will focus on the design principle of power transformer state monitoring and control system,and expounds the working process of power transformer state monitoring and control system.The system designed can realize real-time,intelligent monitoring and control of power transformer.The hardware part of the system mainly includes data acquisition module and STM32 single chip microcomputer control.System module,GPRS communication module,cooling system drive module,system power supply module,display module,input module and so on.System software design mainly includes: cooling fan frequency conversion speed control program design,data processing program design,data upload interface program design,module driver design and so on.The hot spot estimation and condition monitoring control system of power transformer designed in this paper is tested on the experimental platform.The test results show that the data collected by the terminal can be displayed in real time in the application program of the cloud server.The hot spot estimation model of transformer based on ls-svm realizes the prediction of hot spot temperature and the error is less than 6%,which satisfies the real-time and accuracy requirements of hot spot prediction.The control system can adjust the working state of the cooling system at different hot-spot temperatures,improve the control mode of the cooling system and reduce the power consumption. |
