Research And Application Of Noise Prediction Method In Oxygen Plant

Noise in oxygen plant is a kind of special noise produced by oxygen plant when it provides oxygen for iron and steel production.This kind of noise is characterized by large number of specific sound sources,wide distribution range,wide frequency bandwidth,high sound pressure level and obvious aliasing.These characteristics make the noise test of oxygen plant difficult,low precision of prediction and make it difficult to control.Based on geometrical acoustics theory,this thesis presents a noise prediction method combining image source method,acoustical radiosity and ray trace method,and establishes a noise prediction model for indoor and outdoor noise in oxygen plants.In addition,a fast multipole algorithm and a space equivalent partition optimization prediction model are developed to reduce the order of sound source structure and simplify the space model.Finally,the effective prediction of noise field in oxygen plant is realized.This topic has been funded by the Research on Technology of Noise Control in Ironmaking and the National Natural Science Foundation(61671262).The main research work includes:(1)This thesis analyzes the present indoor and outdoor noise prediction methods,aiming at the accuracy and influencing factors of industrial noise prediction,and carries out detailed test,analysis and research on the noise radiation characteristics of the workshop area and main equipment,and finally determines the main characteristics of indoor and outdoor noise of the workshop.(2)The application of virtual source method,sound radiometric method and sound ray method in noise prediction is studied in depth.Aiming at the indoor noise characteristics of oxygen plants,an indoor noise prediction model Image source methodAcoustical radiosity(ISM-AR)is established by combining virtual source method and sound ray method;and with the outdoor noise characteristics,an outdoor noise prediction model is established by combining the sound ray method and outdoor sound propagation criteria.(3)Aiming at the problems of high model complexity and low computational efficiency in the application of the noise prediction model,a model optimization algorithm for rapid multipole expansion and spatial equivalence partition is constructed.Simulation results show that the optimization algorithm can reduce the model complexity and computational complexity,and prove the feasibility and effectiveness of the optimization model.(4)The model is applied to noise field prediction of oxygen plant.Based on the measured data of oxygen plant and the optimized prediction model,the indoor and outdoor noise distribution of oxygen plant is reconstructed and calculated.According to the calculation results,the noise field distribution and main control objectives of noise in the plant area are further clarified,which provides the theoretical basis for the actual noise reduction.