Measurement Of Urban Sewage Flow Parameters Based On Sound And Light Information

In recent years,along with the rapid development of the economy and the advancement of the urbanization process,urban sewage discharge and urban waterlogging,and other urban water problems have seriously affected daily life and caused great economic losses.Urban sewage pipes are often located in harsh conditions,where conventional measuring instruments cannot be applied,and automatic online monitoring cannot be realized.The measurement of the urban sewage flow rate and the monitoring of the hazardous composition are chronically absent.Therefore,it is necessary to investigate the flow measurement techniques for sewage flow and the phase distribution mechanism,and the online monitoring and processing techniques.The research provides a technical basis for the hierarchical treatment of sewage,and the reuse and recycling of the sewage,which could promote the development of the green economy.Research on sewage flow rate measurement methods based on the near-infrared absorption spectra and the ultrasonic phased array is carried out.Prediction models for the phase volume fraction of gas-liquid flow and the flow rate of the liquid-solid two-phase flow are proposed.The content and innovative results are summarized as follows:1.The experimental system for measuring sewage flow parameters was designed.Using transmission spectroscopy technology combined with the static experimental results analysis of near-infrared probes,a vertical through-beam dual-optical path near-infrared measurement system was designed,and a gas-liquid two-phase flow interface wave velocity measurement sensor was constructed,which was used to realize the fine measurement of gas-liquid twophase flow point liquid film.According to the assembly requirements of the array ultrasonic sensor probe and the gas-liquid two-phase flow measurement pipeline,a wedge suitable for the connection between the array ultrasonic sensor and the circular pipeline was designed,and the array ultrasonic sensing experimental measurement device was constructed.2.The void fraction of stratified flow,plug flow,and bubble flow is measured using the constructed measurement device based on specially designed focusing rules.The image reconstruction of the cross-section under the three flow patterns is realized by matrix coordinate transformation and image optimization,and the measurement of the void fraction is realized by relevant algorithms such as lattice statistics.The prediction models of the void fraction under the three flow patterns are evaluated based on the functional relationship between the void fraction and the gas volume fraction and the maximum errors are 10.1%,5.08%,and 1.6%,respectively.3.Interfacial disturbance wave is the key medium for the momentum transfer between the gas core and liquid film at the two-phase interface,the velocity of which is the key parameter when studying the momentum transfer.Since an exact analytical solution of the interfacial disturbance velocity cannot be derived theoretically,an ensemble learning framework for the disturbance wave velocity is constructed and a new model for predicting different flow conditions in vertical two-phase flow is proposed.The dimensionless velocityrelated parameters of interfacial disturbance waves are obtained by feature selection based on the interfacial shear force model,and the grid search method is equipped to tune the important parameters.The ExtraTree model is constructed and optimized,with the Mean Absolute Percentage Error of the model being less than 20%,and the relative measurement uncertainty being within ±25% for 95.67% of the results.The predicted MAPE value of the optimized ExtraTree model is only 0.31%.By comparing the prediction results of existing and literature data(MAPE values range from 2.93% to 15.62%),the extrapolation and applicability of the ensemble learning model are further verified.4.Based on grayscale co-existence matrix(GLCM),the images obtained by the array ultrasonic sensor were extracted from energy,and a particle concentration prediction model based on the integrated algorithm XGBoos was established.Experimental results show that the relative error of about 95.5% of the working point is within ±30%,and the relative error of about 82.9% of the working point is within ±20%,MAPE value is 8.7%,indicating that the particle concentration prediction model fitting has a good effect.