Study Of Multi-Path Ultrasonic Gas Flowmeter Based On Time Difference Method

Ultrasonic gas flowmeter has the characteristics of wide turn-down ratio,low pressure loss and no movable parts.It is widely used in the fields of natural gas measurement,trade settlement and chemical industry.However,the working environment of gas ultrasonic flowmeter is complex in practical application.So the ultrasonic echo signal is easy to be disturbed by noises and the echo waveform changes greatly under different working conditions.At the same time,the integration method of multi-path has low accuracy in complex flow field.These factors reduce the accuracy and stability of multi-path ultrasonic gas flowmeter.In order to solve the above problems,a multi-path ultrasonic gas flowmeter based on time difference method is designed.Studies on the excitation method of ultrasonic transducer,signal processing method and multi-channel velocity data fusion method are done.The system design of multi-path ultrasonic gas flowmeter includes hardware design and software design.The hardware design mainly includes variable amplitude and coding excitation circuit,channel switching circuit,echo signal receiving and processing circuit.The system software includes MSP430 program and FPGA program.The variable amplitude and coding excitation is adopted in the excitation of ultrasonic transducer.FPGA is used to generate the excitation control signal.The start position of high-voltage excitation and the amplitude of low-voltage excitation are studied in the experiments.The better variable amplitude and coding excitation signal is obtained.So the rising characteristics of ultrasonic echo signal are distinguished obviously.An adaptive threshold adjustment method based on similarity principle is designed in the signal processing.Firstly,the echo peak holding circuit is collected at high speed to obtain the envelope steps and the step group.Then,the similarity between each step group and the standard step group is calculated according to the similarity principle.So the actual feature step position under the current working condition is identified and the output threshold is calculated in order to ensure the consistency of feature points.The multi-path data fusion method based on BP neural network is proposed in data fusion.An improved BP neural network is designed to identify the flow field and calculate the corresponding optimal weight under the different pipe installation.The stability of the flow measurement is also improved.Finally,the prototype of ultrasonic gas flowmeter is studied in the experiments.The variable amplitude and coding excitation experiments are studied.The flow experiments of different pipe installation are also done.The experimental results show that the multi-path ultrasonic gas flowmeter meets the accuracy requirements of level 1.