Research And Application Of Power Plant Electrical Monitoring System Based On Fieldbus Technology

Power plant electrical monitoring system(ECS system)is an important core part of power plant electrical specialty.Its current main research direction is to carry out distributed control transformation based on fieldbus technology.However,due to the different operating conditions of the power plants,it is necessary to select the transformation plan in a targeted manner.Therefore,through the research and comparison of the three current power plant monitoring methods,this paper proposes a system networking method that combines multiple methods,and evaluates and analyzes its networking methods from both technical and economic aspects.In the practical application of the ECS system based on fieldbus,this paper uses the two problems of measuring point noise,jump and difficult to monitor nonlinear,multi-parameter systems(unable to find a clear correspondence between parameters).The fieldbus-based ECS system can record and store the real-time and historical data of the equipment,filter out reasonable data samples from its database,and combine the Kalman filter algorithm and the BP neural network algorithm to solve the above problems.In terms of measuring point noise and jump,it is proposed to use Kalman filter algorithm to filter the noise carried by the signal.More importantly,when the measuring point fails or jumps,the principle of self-linear regression of the Kalman filter algorithm can converge to the value of the jump(or fault)measuring point,although the algorithm is correct for the jump(or fault)measuring point.The convergence time is longer than the convergence time of the normal measuring point,but the convergence result is still close to the true value.Aiming at the complex relationship in the nonlinear and multi-parameter system,firstly,the grey relational analysis method is used to process the multi-variable data to filter out the secondary data.Secondly,the BP neural network is used to find the more complicated mapping relationships in the nonlinear system.According to the output value of the neural network and the measured value of the system,the error between the two can be compared to find possible hidden dangers,which is beneficial to prevent accidents.The above researches have been simulated on MATLAB software for many times,and the simulation results have confirmed the reliability and practicability of the research.