CN-121976950-A - Medical gas system vortex air compressor fault prediction method, system and medium

CN121976950ACN 121976950 ACN121976950 ACN 121976950ACN-121976950-A

Abstract

A medical gas system vortex air compressor fault prediction method, system and medium relate to the field of air compressor fault prediction, in the method, the fault diagnosis is realized from multiple dimensions of machinery, electricity and performance by acquiring multidimensional data such as pressure pulsation spectrum, vibration envelope spectrum, current waveform and temperature change rate; the method comprises the steps of detecting the fault of a vortex disc and a bearing through harmonic analysis of a pressure pulsation spectrum and characteristic frequency of a vibration envelope spectrum, judging the electrical fault through current waveform analysis, improving the accuracy of fault diagnosis, establishing a performance abnormality evaluation system through introducing a compressed air flow ratio and a pressure-current correlation coefficient, simultaneously monitoring the pollution risk of a system through combining a temperature change rate, a dew point and an oil vapor concentration, realizing comprehensive monitoring of equipment state and gas quality, comprehensively analyzing and predicting various fault characteristics through a deep learning model, finding potential faults in time, generating early warning signals, and improving the operation reliability of a medical gas system.

Inventors

DU QIN
YE XIAOJIE
XU LINFEI
CHENG QUAN

Assignees

广州市桂勤器械设备工程有限公司

Dates

Publication Date: 20260505
Application Date: 20260126

Claims (10)

1. The fault prediction method for the vortex air compressor of the medical gas system is characterized by comprising the following steps of: Acquiring operation data of a vortex air compressor, and acquiring a pressure pulsation spectrum, a vibration envelope spectrum, a current waveform and a temperature change rate; Analyzing the first-order harmonic amplitude ratio and the second-order harmonic amplitude ratio according to the pressure pulsation spectrum, judging the meshing state of the vortex plate, and determining the type of mechanical faults by combining the bearing fault characteristic frequency in the vibration envelope spectrum; analyzing harmonic distortion rate and fluctuation trend of the current waveform, and judging the type of the electrical fault; calculating the ratio of the actual flow to the rated flow of the compressed air, and combining the pearson correlation coefficient of the pressure and the current to obtain a performance abnormality index; Trend analysis is carried out on the temperature change rate, and the risk of system pollution is estimated by combining dew point monitoring data and the concentration of oil vapor; and predicting the fault development trend through a preset deep learning model according to the mechanical fault type, the electrical fault type, the performance abnormality index and the system pollution risk, and generating an early warning signal.
2. The method according to claim 1, wherein the mechanical fault types include scroll wear and bearing faults, the analyzing the first order and second order harmonic amplitude ratios according to the pressure pulsation spectrum, determining the scroll engagement state, and determining the mechanical fault types in combination with the bearing fault characteristic frequency in the vibration envelope spectrum, specifically includes: When the first-order harmonic amplitude ratio and the second-order harmonic amplitude ratio in the pressure pulsation spectrum exceed a preset harmonic ratio and the exhaust pressure fluctuation exceeds a preset fluctuation range, judging that the vortex disc is worn; and judging that the bearing is in fault when the energy of the bearing fault characteristic frequency in the vibration envelope spectrum exceeds a preset energy threshold value, and the vibration effective value exceeds a preset vibration limit value or the temperature change rate exceeds a preset temperature change threshold value.
3. The method of claim 2, wherein the type of mechanical fault further comprises a host lock-up, and wherein after the step of analyzing the harmonic distortion rate and the trend of the current waveform to determine the type of electrical fault, the method further comprises: Judging whether overload characteristics exist in the current waveform or not; Detecting the rotation resistance of the vortex plate and acquiring rotation resistance characteristic data; And when the overload characteristic of the current waveform is detected, and the rotation resistance characteristic data exceeds a preset resistance threshold value, judging that the host is locked.
4. The method according to claim 1, wherein before the predicting of the failure development tendency by the preset deep learning model, the method further comprises: acquiring fault type, fault characteristics, maintenance measures and equipment operation time length data in a historical fault record; Layering fault samples according to the equipment operation time length, and establishing fault feature libraries of different operation stages; Calculating the influence weight of various fault characteristics in each fault characteristic library based on the fault type, the fault characteristics and the maintenance measures; and optimizing the prediction precision of the deep learning model by using the influence weight.
5. The method of claim 1, wherein after the step of collecting scroll air compressor operation data, the method further comprises: acquiring data of the environmental temperature, the environmental humidity, the atmospheric pressure and the ventilation condition of a machine room where the vortex air compressor is positioned; generating an environmental suitability index based on the ambient temperature, ambient humidity, barometric pressure, and ventilation status data; and dynamically adjusting a fault characteristic threshold according to the environment adaptability index, wherein when the environment adaptability index is lower than a preset standard, a machine room environment improvement suggestion is generated.
6. The method of claim 1, wherein after the step of trending the rate of temperature change, in combination with dew point monitoring data and oil vapor concentration, the method further comprises: Acquiring data of moisture content, oil content and particulate matter concentration in the gas; Determining medical gas purity grading standards and gas quality grading rules; Calculating a gas quality comprehensive score according to the moisture content, the oil content and the particulate matter concentration data and combining the dew point monitoring data and the oil vapor concentration; and when the gas quality comprehensive score is lower than a preset standard corresponding to the purity grade, generating a grading treatment suggestion according to the exceeding condition of each pollutant index.
7. The method of claim 6, wherein after the step of generating a classification proposal based on the overstandard condition of each contaminant indicator when the gas quality composite score is below a preset level of the corresponding purity level, the method further comprises: Acquiring filter state data and pipeline system pressure distribution data of an air compressor system; Determining the position and monitoring parameters of key control points of the system; Analyzing the propagation path of the pollutant in the system according to the filter state data and the pipeline system pressure distribution data; Based on the propagation path and the location and monitoring data of the critical control points, a system decontamination scheme and maintenance strategy are generated.
8. Medical gas system vortex air compressor machine trouble prediction system, its characterized in that, the system includes: One or more processors and memory coupled with the one or more processors, the memory to store computer program code, the computer program code comprising computer instructions that the one or more processors invoke to cause the system to perform the method of any of claims 1-7.
9. A computer readable storage medium comprising instructions which, when run on a system, cause the system to perform the method of any of claims 1-7.
10. A computer program product, characterized in that the computer program product, when run on a system, causes the system to perform the method according to any of claims 1-7.

Description

Medical gas system vortex air compressor fault prediction method, system and medium Technical Field The application belongs to the field of air compressor fault prediction, and particularly relates to a method, a system and a medium for predicting faults of a vortex air compressor of a medical gas system. Background The vortex air compressor in the medical gas system is used as core equipment, and the reliable operation of the vortex air compressor is critical to medical safety. Traditional vortex air compressor fault diagnosis mainly relies on single vibration or temperature monitoring, can't realize accurate discernment and the early warning of trouble. However, various fault modes in the running process of the equipment are complex and changeable, potential faults cannot be found in time only by a simple threshold judgment mode, and the fault development trend is difficult to accurately predict. In the related art, the fault diagnosis method of the vortex air compressor comprises operation state evaluation based on pressure monitoring, bearing temperature detection, equipment start-stop state recording and the like. In addition, some medical gas systems employ periodic sampling detection to evaluate gas quality. However, the medical gas system has strict requirements on equipment reliability and gas quality, and the prior art is difficult to meet the requirements of timely early warning and preventive maintenance. In particular, in the aspects of fault diagnosis and gas quality monitoring under complex working conditions, a systematic solution is lacking, and the situation needs to be further improved. Disclosure of Invention The application provides a fault prediction method for a vortex air compressor of a medical gas system, which is used for solving the technical problems of low fault diagnosis precision and untimely early warning of the vortex air compressor. According to the method, the accurate prediction and early warning of the vortex air compressor faults are realized by collecting multidimensional data such as a pressure pulsation spectrum, a vibration envelope spectrum, a current waveform and a temperature change rate and analyzing the mechanical fault type, the electrical fault type, the performance abnormality index and the system pollution risk by combining a deep learning model. In a first aspect, the application provides a failure prediction method for a vortex air compressor of a medical gas system, Acquiring operation data of a vortex air compressor, and acquiring a pressure pulsation spectrum, a vibration envelope spectrum, a current waveform and a temperature change rate; Analyzing the first-order harmonic amplitude ratio and the second-order harmonic amplitude ratio according to the pressure pulsation spectrum, judging the meshing state of the vortex plate, and determining the type of mechanical faults by combining the bearing fault characteristic frequency in the vibration envelope spectrum; analyzing harmonic distortion rate and fluctuation trend of the current waveform, and judging the type of the electrical fault; calculating the ratio of the actual flow to the rated flow of the compressed air, and combining the pearson correlation coefficient of the pressure and the current to obtain a performance abnormality index; Trend analysis is carried out on the temperature change rate, and the risk of system pollution is estimated by combining dew point monitoring data and the concentration of oil vapor; and predicting the fault development trend through a preset deep learning model according to the mechanical fault type, the electrical fault type, the performance abnormality index and the system pollution risk, and generating an early warning signal. In the embodiment, fault diagnosis is realized from multiple dimensions of machinery, electricity and performance by acquiring multidimensional data such as a pressure pulsation spectrum, a vibration envelope spectrum, a current waveform and a temperature change rate, faults of a vortex disk and a bearing are identified through harmonic analysis of the pressure pulsation spectrum and characteristic frequency of the vibration envelope spectrum, the electrical faults are judged by combining current waveform analysis, the accuracy of fault diagnosis is improved, a performance abnormality evaluation system is established by introducing a compressed air flow ratio and a pressure-current correlation coefficient, meanwhile, the pollution risk of a temperature change rate, a dew point and an oil vapor concentration monitoring system is combined, the comprehensive monitoring of equipment state and gas quality is realized, in addition, comprehensive analysis and prediction are carried out on various fault characteristics by adopting a deep learning model, potential faults are found timely, early warning signals are generated, and the operation reliability of a medical gas system is improved. With reference to some embodiments of the first aspect, in