Research On Signal Processing Method And System Development Of Bubble Detector In Sodium Based On Vortex Electromagnetic Induction Principle

The steam generator in the nuclear power plant of fast reactor may crack under the severe conditions of high temperature and high pressure for a long time.At this time,high-pressure steam will be sprayed into the high-temperature liquid sodium,causing a violent sodium-water reaction.The product of the sodium-water reaction will corrode cracks,and at the same time,the reaction process will release a large amount of heat,and will intensify the sodium-water reaction.Such a vicious cycle will cause serious nuclear safety accidents.Therefore,it is necessary to always check whether the steam generator has leaked to ensure the safety of the nuclear power plant of fast reactor.To this end,this article studied a bubble detector in sodium based on the principle of vortex electromagnetic induction.The specific research work and innovations are as follows:(1)The sensor of the bubble detector in sodium was researched and designed,and its working principle was analyzed.The hardware circuit system of the bubble detector in sodium was developed.The experiment of bubble detection in sodium was conducted,and the signals output by the sensor of the bubble detector in sodium were collected under different sodium flow rates at different gas injection flow rates.A simulation experiment scheme using water instead of liquid sodium was proposed,and the experiment of bubble detection in water was conducted,which provided an effective auxiliary method for studying the signal processing method of bubble detection in sodium.(2)The signal processing method of bubble detection in sodium was studied from the perspective of the amplitude.Starting from the sensor output signal of the bubble detector in water,it was proposed to use the local extreme value to reflect the change in signal amplitude,and to convert the study of the influence of bubble on the signal amplitude into the study of the influence of bubble on the local extreme value of the sensor output signal.The local extreme value of the signal were modeled using statistical methods.Based on the statistical model of the sensor output signal of the bubble detector in water,a signal processing method based on the standard deviation of peak-to-peak value for bubble detection in sodium was proposed.(3)The signal processing method of bubble detection in sodium was studied from the perspective of signal waveform.Starting from the sensor output signal of the bubble detector in sodium,the study of the influence of the bubble on the sensor output signal was converted into the study of the influence of the bubble on the change pattern of the sensor output signal waveform.The similarity between the two adjacent periodic signals of the sensor output signal was affected by the bubble.Therefore,the correlation coefficient was used to identify the similarity between the two adjacent periodic signals.Correlation coefficients of the sensor output signals were calculated under different gas injection flow rates,and the results were statistically modeled and analyzed.Based on the analysis results of the model,a signal processing method based on correlation coefficient for bubble detection in sodium was proposed.(4)The signal processing method of bubble detection in sodium was studied from the perspective of frequency domain.Starting from the sensor output signal of the bubble detector in sodium,the study of the influence of the bubble on the sensor output signal was converted into the study of the influence of bubble on the calculated frequency of the sensor output signal.For the case where sodium contains bubble,when the signals of different time lengths were used to calculate the frequency of the sensor output signal,the frequency variation range was quite different.The concept of frequency fluctuation coefficient was defined,and the curve relationship model of the number of sensor output signal periods and the frequency fluctuation coefficient was established,which could easily determine the number of signal period required to calculate the frequency fluctuation coefficient,so as to highlight the effect of bubbles on the frequency fluctuation coefficient of the sensor output signal,and a signal processing method based on frequency fluctuation coefficient for bubble detection in sodium was proposed.(5)The signal processing method of bubble detection in sodium was studied from the perspective of signal composition.Starting from the sensor output signal of the bubble detector in sodium,the study of the influence of the bubble on the sensor output signal is converted into the influence of the bubble on the composition of the sensor output signal.The baseline of the sensor output signal is defined,which is used to reflect the bubble noise signal.In this way,the model of the sensor output signal can be regarded as composed of the flow signal and the bubble noise signal.When sodium does not contain bubble,the sensor output signal is approximately the flow signal.When there is bubble in sodium,the bubble noise signal in the sensor output signal by will become more obvious as the gas injection flow rate increases.Therefore,the proportion of the flow signal in the sensor output signal will decrease as the bubble content in sodium increases.Then,the extraction of bubble noise signal was studied,the concept of the energy ratio of the sensor output signal was defined,and a signal processing method based on energy ratio for bubble detection in sodium was proposed.(6)The signal processing method based on the peak-peak standard deviation for bubble detection in sodium and the signal processing method based on the correlation coefficient for bubble detection in sodium were respectively realized in real time on a DSP based transmitter of the bubble detector in sodium to form a complete experimental prototype.The real-time verification experiments were performed.The verification results showed that the minimum gas injection flow rate in sodium that can be detected by the two algorithms was 0.66L/min,that is,the minimum leakage of water in the steam generator could be detected as 0.1g/s.Real-time implementation schemes of the signal processing method based on frequency fluctuation coefficients for bubble detection in sodium and the signal processing method based on energy ratio for bubble detection in sodium were studied separately,and the real-time implementation schemes were verified.The verification results showed that the minimum gas injection flow rate in sodium that could be detected by the two algorithms was also 0.66L/min,that is,the minimum leakage of water in the steam generator also could be detected as 0.1g/s.