Font Size: a A A

Design Of Liquid Level Measurement System Based On Information Fusion

Posted on:2021-05-25Degree:MasterType:Thesis
Country:ChinaCandidate:D F FangFull Text:PDF
GTID:2392330605451191Subject:Control Engineering
Abstract/Summary:PDF Full Text Request
The liquid level measurement method is widely used in the fields of petroleum,chemical,sewage treatment,geological disasters,etc.,and plays an important role in state monitoring and alarming in production,processing,transportation,storage,and environmental management.The liquid level measurement method based on low-frequency acoustic resonance can effectively overcome the parasitic reflection problem faced by ultrasonic measurement,and can obtain more accurate measurement results according to the diffraction principle even though foreign matter exists in the liquid surface.However,under big environment noises,when the liquid level conversion is performed by extracting the acoustic resonance frequency point sequence,a false resonance point or a missing resonance point may occur,which may affect the measurement accuracy.Aiming at this problem,this paper first analyzes the causes and their specific manifestations of the above non-ideal conditions,and implements the correction and error compensation for resonance points based on rule-based reasoning and information fusion methods.Then,the corresponding liquid level measurement system is designed,and the effectiveness of the proposed method is verified by a large number of liquid level measurement experiments.The main research work is as follows:(1)Analysis of the principle of low frequency acoustic resonance liquid level measurement.The acoustic resonance measurement principle combines a transmitted wave and a reflected wave to form a composite wave by transmitting a sound wave scanning signal to a liquid surface and reflecting it.A multi-stage filtering algorithm is used to extract a sequence of resonance frequency points in the synthesized wave.Since the length of the resonance sequence and the value of the resonance frequency vary with the height of the liquid level,the liquid level height can be converted according to the resonance sequence.The liquid level measurement system designed based on this principle has the advantages of high measurement accuracy,simple structure,low cost and wide application range.(2)Acoustic resonance liquid level measurement error compensation method based on information fusion.Firstly,the acoustic resonance sequence generated under non-ideal conditions is analyzed,and the relationship between the local features of adjacent resonance points and the false resonance points and the missing resonance points is summarized.The belief rule base is built to model the nonlinear mapping relationship,and the non-ideal resonance points are corrected and compensated by the ER(Evidence reasoning)fusion reasoning method to generate a complete resonance point sequence.In order to further reduce the error,the complete resonance sequences obtained in multiple sampling periods are iteratively fused,and the accurate liquid level measurement values are obtain by the fused resonance sequence.(3)Acoustic resonance level measurement system design and performance test.Use the microphone,speaker,temperature sensor,development board and other components to build an acoustic resonance liquid level measurement hardware platform,and load the measurement software developed in(2)to form the final measurement system.In the noisy environment,a large-scale(0.6m?10.4m)liquid level measurement experiment is carried out,and the obtained resonance sequences are corrected and compensated in real time by using the proposed method.Through a large number of experimental data analysis,it can be found that after two steps of reasoning fusion and iterative fusion,the measurement error can be effectively reduced,and the liquid level measurement accuracy of the acoustic resonance method can be significantly improved.
Keywords/Search Tags:Liquid level measurement, Acoustic resonance, Confidence rule base, Error compensation
PDF Full Text Request
Related items