Failure prediction method and system for vortex air compressor of medical gas system and medium

Abstract

The invention discloses a medical gas system vortex air compressor fault prediction method and system and a medium, and relates to the field of air compressor fault prediction, in the method, multi-dimensional data such as a pressure pulsation frequency spectrum, a vibration envelope spectrum, a current waveform and a temperature change rate are obtained, and fault diagnosis is achieved from multiple dimensions of machinery, electricity and performance; vortex plate and bearing faults are identified through harmonic analysis of a pressure pulsation spectrum and vibration envelope spectrum characteristic frequency, electrical faults are analyzed and judged in combination with current waveforms, and the accuracy of fault diagnosis is improved. A performance abnormity evaluation system is established by introducing a compressed air flow ratio and a pressure-current correlation coefficient, and meanwhile, the system pollution risk is monitored in combination with a temperature change rate, a dew point and an oil vapor concentration, so that comprehensive monitoring of an equipment state and gas quality is realized; in addition, the deep learning model is adopted to comprehensively analyze and predict various fault features, potential faults are found in time, early warning signals are generated, and the operation reliability of the medical gas system is improved.

Description

Technical Field

[0001] This application belongs to the field of air compressor failure prediction, and particularly relates to methods, systems and media for predicting failures of scroll air compressors in medical gas systems. Background Technology

[0002] As a core component of medical gas systems, the reliable operation of scroll air compressors is crucial for medical safety. Traditional fault diagnosis for scroll air compressors mainly relies on single vibration or temperature monitoring, which cannot achieve accurate fault identification and early warning. However, various fault modes occur during equipment operation, which are complex and changeable. Simple threshold judgment methods alone cannot detect potential faults in a timely manner, nor can they accurately predict the development trend of faults.

[0003] In related technologies, fault diagnosis methods for scroll air compressors include operational status assessment based on pressure monitoring, bearing temperature detection, and equipment start-up and shutdown status recording. In addition, some medical gas systems employ periodic sampling tests to assess gas quality.

[0004] However, medical gas systems have stringent requirements for equipment reliability and gas quality, and existing technologies are insufficient to meet the needs for timely early warning and preventative maintenance. In particular, there is a lack of systematic solutions for fault diagnosis and gas quality monitoring under complex operating conditions, and this situation requires further improvement. Summary of the Invention

[0005] This application provides a method for predicting the faults of scroll air compressors in medical gas systems, addressing the technical problems of low fault diagnosis accuracy and untimely early warning in scroll air compressors. This method collects multi-dimensional data such as pressure pulsation spectrum, vibration envelope spectrum, current waveform, and temperature change rate, and combines this data with a deep learning model to analyze mechanical fault types, electrical fault types, abnormal performance indicators, and system contamination risks, thereby achieving accurate prediction and early warning of scroll air compressor faults.

[0006] Firstly, this application provides a method for predicting the failure of a scroll air compressor in a medical gas system. Collect operating data of the scroll air compressor to obtain pressure pulsation spectrum, vibration envelope spectrum, current waveform and temperature change rate; Based on the analysis of the first and second harmonic amplitude ratios of the pressure pulsation spectrum, the meshing state of the scroll plate is determined, and the type of mechanical fault is determined by combining the bearing fault characteristic frequency in the vibration envelope spectrum. Analyze the harmonic distortion rate and fluctuation trend of the current waveform to determine the type of electrical fault; Calculate the ratio of actual compressed air flow rate to rated flow rate, and combine it with the Pearson correlation coefficient between pressure and current to obtain performance anomaly indicators; Trend analysis was performed on the temperature change rate, and the system contamination risk was assessed by combining dew point monitoring data and oil vapor concentration. Based on the mechanical fault type, electrical fault type, abnormal performance indicators, and system pollution risk, a pre-set deep learning model is used to predict the fault development trend and generate an early warning signal.

[0007] In the above embodiments, fault diagnosis is achieved from multiple dimensions, including mechanical, electrical, and performance, by acquiring multi-dimensional data such as pressure pulsation spectrum, vibration envelope spectrum, current waveform, and temperature change rate. Faults in the scroll plate and bearings are identified through harmonic analysis of the pressure pulsation spectrum and characteristic frequencies of the vibration envelope spectrum, while electrical faults are judged by combining current waveform analysis, thus improving the accuracy of fault diagnosis. A performance anomaly evaluation system is established by introducing the compressed air flow ratio and pressure-current correlation coefficient. Simultaneously, combined with the pollution risk monitoring system for temperature change rate, dew point, and oil vapor concentration, comprehensive monitoring of equipment status and gas quality is achieved. Furthermore, a deep learning model is used to comprehensively analyze and predict various fault characteristics, promptly identifying potential faults and generating early warning signals, thereby improving the operational reliability of the medical gas system. In conjunction with some implementations of the first aspect, in some implementations, the mechanical fault types include scroll plate wear and bearing failure. The step of determining the scroll plate meshing state based on the ratio of first and second harmonic amplitudes analyzed by the pressure pulsation spectrum, and determining the mechanical fault type by combining the bearing failure characteristic frequency in the vibration envelope spectrum, specifically includes: When the ratio of the amplitude of the first and second harmonics in the pressure pulsation spectrum exceeds the preset harmonic ratio value, and the exhaust pressure fluctuation exceeds the preset fluctuation range, it is determined that the scroll plate is worn. When the energy of the bearing fault characteristic frequency in the vibration envelope spectrum exceeds a preset energy threshold, and the effective vibration value exceeds a preset vibration limit or the temperature change rate exceeds a preset temperature change threshold, it is determined to be a bearing fault.

[0008] In the above embodiments, this application achieves mechanical fault identification by setting judgment rules for scroll disk wear and bearing failure. For scroll disk wear, the judgment is made by combining two parameters: the harmonic amplitude ratio in the pressure pulsation spectrum and the exhaust pressure fluctuation, thus avoiding the problem of misjudgment by a single indicator. For bearing failure, the thresholds of three dimensions—characteristic frequency energy, vibration effective value, and temperature change rate—are comprehensively considered, which improves the reliability of bearing fault diagnosis. This multi-parameter collaborative judgment mechanism can not only accurately distinguish different types of mechanical faults, but also effectively reduce the false alarm rate. In conjunction with some embodiments of the first aspect, in some embodiments, the mechanical fault type further includes main unit seizure. After the step of analyzing the harmonic distortion rate and fluctuation trend of the current waveform to determine the electrical fault type, the method further includes: Determine whether the current waveform exhibits overload characteristics; Detect the rotational resistance of the scroll disk and obtain rotational resistance characteristic data; When an overload characteristic is detected in the current waveform and the rotational resistance characteristic data exceeds a preset resistance threshold, it is determined that the main unit is locked.

[0009] In the above embodiments, this application establishes a dual judgment mechanism for main unit seizure by analyzing the overload characteristics of the current waveform and the rotational resistance characteristics of the scroll plate; when the current overload and rotational resistance both exceed the preset threshold, it is judged as main unit seizure, which can promptly detect serious faults in the operation of the scroll air compressor; it not only avoids misjudgment that may be caused by single parameter monitoring, but also provides timely warnings before the fault occurs, effectively preventing equipment damage and medical gas supply interruption caused by main unit seizure. In conjunction with some implementations of the first aspect, in some implementations, before predicting the fault development trend through a preset deep learning model, the method further includes: Obtain fault type, fault characteristics, maintenance measures, and equipment runtime data from historical fault records; The fault samples are stratified according to the operating time of the equipment to establish a fault feature library for different operating stages; Based on the fault type, fault characteristics and maintenance measures, calculate the influence weight of each type of fault characteristic in each fault characteristic library; The prediction accuracy of the deep learning model is optimized using the influence weights.

[0010] In the above embodiments, this application stratifies fault samples according to equipment operating time, establishes fault feature libraries for different operating stages, and considers the impact of equipment aging on fault features; by analyzing the correspondence between fault types, fault features and maintenance measures in historical fault records, the influence weight of various fault features is calculated, and dynamic evaluation of fault features is realized; this hierarchical feature weight optimization mechanism based on operating stages improves the prediction accuracy of deep learning models for faults in different operating stages, making fault warnings more targeted. In some embodiments, in conjunction with the first aspect, after the step of collecting the operating data of the scroll air compressor, the method further includes: Obtain data on ambient temperature, humidity, atmospheric pressure, and ventilation conditions in the machine room where the scroll air compressor is located; Based on the environmental temperature, humidity, atmospheric pressure, and ventilation data, an environmental adaptability index is generated. The fault characteristic threshold is dynamically adjusted based on the environmental adaptability index. When the environmental adaptability index is lower than the preset standard, suggestions for improving the data center environment are generated.

[0011] In the above embodiments, this application generates environmental adaptability indicators by acquiring data on ambient temperature, humidity, atmospheric pressure, and ventilation conditions of the computer room, thereby achieving dynamic adjustment of fault characteristic thresholds. When the environmental adaptability indicators fail to meet the standards, timely environmental improvement suggestions are generated, which helps to optimize the operating environment of the scroll air compressor and ensure the stable operation of the equipment. In some embodiments, in conjunction with the first aspect, after the step of performing trend analysis on the temperature change rate, combining dew point monitoring data and oil vapor concentration to assess the system contamination risk, the method further includes: To obtain data on the moisture content, oil content, and particulate matter concentration in the gas; Establish standards for grading the purity of medical gases and rules for scoring gas quality; Based on the moisture content, oil content, and particulate matter concentration data, combined with the dew point monitoring data and oil vapor concentration, a comprehensive gas quality score is calculated. When the overall gas quality score is lower than the preset standard for the corresponding purity level, a graded treatment suggestion is generated based on the exceedance of various pollutant indicators.

[0012] In the above embodiments, this application establishes a comprehensive evaluation system for the quality of medical gases by monitoring the moisture, oil, and particulate matter content in the gas, combined with dew point and oil vapor concentration data; it calculates a comprehensive gas quality score based on the medical gas purity grading standard, and generates targeted graded treatment suggestions when the quality does not meet the standard; this not only achieves accurate monitoring of medical gas quality, but also enables timely detection and handling of different types of pollution problems, ensuring the gas supply safety of the medical gas system. In conjunction with some embodiments of the first aspect, in some embodiments, after the step of generating a graded treatment suggestion based on the exceedance of various pollutant indicators when the overall gas quality score is lower than the preset standard for the corresponding purity level, the method further includes: Acquire filter status data and pipeline system pressure distribution data of the air compressor system; Determine the location and monitoring parameters of the system's key control points; Based on the filter status data and pipeline system pressure distribution data, the propagation path of pollutants in the system is analyzed; Based on the propagation path and the location and monitoring data of the key control points, a system purification plan and maintenance strategy are generated.

[0013] In the above embodiments, this application obtains filter status and pipeline pressure distribution data to analyze the propagation path of pollutants in the system, thereby achieving precise location of the pollution source; by setting monitoring parameters at key control points and combining the results of pollutant propagation path analysis, targeted system purification plans and maintenance strategies are formulated; the efficiency of pollution treatment is improved, and the recurrence of pollution problems can be prevented from the source, ensuring the long-term stable operation of the medical gas system. In a second aspect, embodiments of this application provide a fault prediction system for a scroll air compressor in a medical gas system, comprising: one or more processors and a memory; the memory is coupled to one or more processors, the memory is used to store computer program code, the computer program code including computer instructions, and one or more processors call the computer instructions to cause the system to perform the method described in the first aspect and any possible implementation thereof. Thirdly, embodiments of this application provide a computer-readable storage medium including instructions that, when executed on a system, cause the system to perform the method described in the first aspect and any possible implementation thereof. Fourthly, embodiments of this application provide a computer program product that, when run on a system, causes the system to execute the method described in any possible implementation of the first aspect. One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages: 1. This application provides a method for predicting the faults of a scroll air compressor in a medical gas system. By employing a multi-dimensional data analysis method based on pressure pulsation spectrum, vibration envelope spectrum, current waveform, and temperature change rate, as well as a collaborative diagnostic mechanism for mechanical faults, electrical faults, and system contamination risks, the system can automatically complete the entire process from data acquisition to fault warning. This effectively solves the problems in the prior art where single parameter monitoring is difficult to fully reflect the equipment status and cannot predict the fault development trend in a timely manner, thereby achieving the accuracy of fault diagnosis.

[0014] 2. This application provides a fault prediction method for a scroll air compressor in a medical gas system. By adopting a fault sample hierarchical and feature weight optimization method based on runtime, combined with a dynamic adjustment mechanism for the threshold of environmental adaptability indicators, the system can accurately assess the fault characteristics at different operating stages. This effectively solves the problems of ignoring the impact of equipment aging and misjudgment caused by environmental changes in the prior art, thereby achieving reliable fault early warning.

[0015] 3. This application provides a method for predicting the failure of a vortex air compressor in a medical gas system. Because it adopts a gas quality assessment method based on multi-index synergy, as well as pollutant propagation path analysis and key control point monitoring strategies, the system can achieve accurate monitoring of medical gas quality and location of pollution sources, thus enabling the safe and stable operation of the medical gas system. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating a method for predicting the failure of a scroll air compressor in a medical gas system, as described in this application.

[0017] Figure 2 This is another flowchart illustrating a method for predicting the failure of a scroll air compressor in a medical gas system, as described in this application.

[0018] Figure 3 This is a schematic diagram of the physical device structure of a medical gas system vortex air compressor fault prediction system provided in an embodiment of this application. Detailed Implementation

[0019] The terminology used in the following embodiments of this application is for the purpose of describing particular embodiments only and is not intended to be limiting of this application. As used in the specification and appended claims of this application, the singular expressions “a,” “an,” “the,” “the,” “the,” and “this” are intended to include the plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term “and / or” as used in this application refers to any or all possible combinations including one or more of the listed items.

[0020] Hereinafter, the terms "first" and "second" are used for descriptive purposes only and should not be construed as implying or suggesting relative importance or implicitly indicating the number of indicated technical features. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature, and in the description of the embodiments of this application, unless otherwise stated, "multiple" means two or more. In the field of medical devices, medical gas systems are one of the important infrastructures of hospitals. Among them, the scroll air compressor is the core component of the gas source equipment, and its operating status directly affects the supply quality and safety of medical gases.

[0021] In related technologies, fault prediction for scroll air compressors mainly relies on single parameter monitoring or periodic inspections, making it difficult to detect potential faults in a timely manner. At the same time, due to the lack of comprehensive assessment of the equipment's operating environment and gas quality, problems such as misjudgment of faults or exceeding quality standards are prone to occur, affecting the reliability of medical gas systems.

[0022] This application is primarily applied to medical gas supply scenarios such as hospital oxygen plants and compressed air systems, where gas quality and supply reliability are extremely critical. In these applications, comprehensive monitoring of the mechanical condition, electrical performance, and gas quality of the scroll air compressor is required, along with the ability to predict potential faults in a timely manner. To address these technical problems, this application provides a fault prediction method for scroll air compressors in medical gas systems. An embodiment is described below, combined with… Figure 1 The present application describes a method for predicting the failure of a scroll air compressor in a medical gas system. Please see Figure 1 This is a flowchart illustrating a method for predicting faults in a medical gas system vortex air compressor, as described in this application.

[0023] S101. Collect operating data of the scroll air compressor and obtain pressure pulsation spectrum, vibration envelope spectrum, current waveform and temperature change rate.

[0024] Among them, the pressure pulsation spectrum refers to the frequency distribution characteristics of the exhaust pressure of the scroll air compressor changing with time, the vibration envelope spectrum is the modulation characteristics in the equipment vibration signal obtained through demodulation analysis, the current waveform reflects the current change curve during motor operation, and the temperature change rate represents the range of equipment temperature change per unit time.

[0025] Specifically, firstly, exhaust pressure data is collected using a pressure sensor, and the pressure pulsation spectrum is obtained through a fast Fourier transform; then, vibration signals are collected using an accelerometer, and the vibration envelope spectrum is obtained through a Hilbert transform; simultaneously, a current transformer is used to collect the motor operating current and record the current waveform data; finally, temperature sensors are deployed to monitor the temperature of key components and calculate the rate of temperature change.

[0026] Understandably, in practical applications, sensors can be deployed at multiple locations simultaneously to obtain more comprehensive monitoring data; the data acquisition frequency can be adjusted according to the equipment's operating status to increase the sampling density during critical periods; and supplementary temperature monitoring points such as oil temperature and shaft temperature can be added to expand the temperature monitoring range.

[0027] In some embodiments, sensor signal interference or data loss may occur. This can be addressed by setting a reasonable effective data range to filter out outliers; then using data from adjacent time points for interpolation compensation; and simultaneously establishing a multi-sensor data cross-validation mechanism to ensure the accuracy and reliability of the collected data.

[0028] S102. Based on the pressure pulsation spectrum analysis, the amplitude ratio of the first and second harmonics is analyzed to determine the meshing state of the scroll plate, and combined with the bearing fault characteristic frequency in the vibration envelope spectrum, the mechanical fault type is determined.

[0029] Mechanical failure types include scroll plate wear, bearing failure, and main engine seizure. The ratio of the amplitudes of the first and second harmonics reflects the meshing state of the scroll plate and is an important indicator for assessing the degree of scroll plate wear. Bearing failure characteristic frequencies include inner ring failure frequencies, outer ring failure frequencies, rolling element failure frequencies, and cage failure frequencies. The energy changes of these frequencies in the vibration envelope spectrum can indicate different types of bearing failures.

[0030] Specifically, when the ratio of the first and second harmonic amplitudes in the pressure pulsation spectrum exceeds a preset harmonic ratio, and the exhaust pressure fluctuation exceeds a preset fluctuation range, it is determined that the scroll plate is worn. When the energy of the bearing fault characteristic frequency in the vibration envelope spectrum exceeds a preset energy threshold, and the effective vibration value exceeds a preset vibration limit or the temperature change rate exceeds a preset temperature change threshold, it is determined that the bearing is faulty. Furthermore, for determining main engine seizure, it is necessary to determine whether the current waveform exhibits overload characteristics and to detect the rotational resistance of the scroll plate to obtain rotational resistance characteristic data. When an overload characteristic is detected in the current waveform, and the rotational resistance characteristic data exceeds a preset resistance threshold, it is determined that the main engine is seized.

[0031] In practical applications, optionally, for scroll disc wear, a graded early warning threshold can be set to determine the degree of wear based on different combinations of harmonic ratios and pressure fluctuations; optionally, for bearing failure, a multi-parameter weighted scoring mechanism can be established to comprehensively assess the severity of the failure; for main engine seizure, auxiliary judgment indicators such as speed change rate and starting current characteristics can be added.

[0032] Furthermore, during implementation, there may be situations where fault characteristics are not obvious. To address this, the system establishes a fault characteristic database and utilizes pattern recognition algorithms to match and analyze currently detected fault characteristics with typical fault characteristics in the database. Even when fault characteristics are not obvious, the system can accurately identify the fault type, thereby improving the reliability of fault diagnosis.

[0033] S103. Analyze the harmonic distortion rate and fluctuation trend of the current waveform to determine the type of electrical fault.

[0034] Among them, the harmonic distortion rate (THD) of the current waveform refers to the ratio of each harmonic component to the fundamental component in the motor operating current, and the fluctuation trend reflects the characteristics of the current value changing over time. Electrical fault types mainly include two categories: motor faults and sensor malfunctions.

[0035] Specifically, the system first performs Fourier analysis on the acquired current waveform to calculate the total harmonic distortion (THD). A THD greater than 7% indicates a defect in the motor windings. Simultaneously, the system analyzes current fluctuation trends: periodic fluctuations exceeding ±10% of the rated value indicate abnormal motor load; a continuous upward trend exceeding a preset threshold indicates motor overload. For sensor fault diagnosis, cross-validation is used, comparing data from three redundant sensors to determine if a sensor is malfunctioning. Specifically, if the data from one sensor deviates from the data from the other two sensors by more than a preset threshold, the system determines that sensor is malfunctioning and automatically switches to the backup sensor. The system uses a soft starter to reduce the inrush current during motor startup and follows the "six firsts, six lasts" rule for air compressor maintenance to protect the motor. IP67-rated sensors are used to ensure reliable operation in dusty and humid environments. The system also performs regular sensor self-checks and calibrations to promptly detect and address sensor anomalies.

[0036] In some embodiments, the data processing system can establish a multidimensional electrical fault diagnosis model: first, identify motor fault characteristics through harmonic analysis; then, analyze load anomalies by combining current fluctuation trends; next, evaluate the consistency of sensor data; and finally, generate a comprehensive fault diagnosis result. Optionally, the data processing system can also adopt an adaptive threshold adjustment method to dynamically adjust the THD judgment threshold and the early warning standard of fluctuation trends according to the equipment operating time and operating conditions, thereby improving the accuracy of fault diagnosis.

[0037] Furthermore, during implementation, there may be situations where multiple sensors fail simultaneously. To address this, the system is equipped with a sensor backup mechanism and an emergency response plan. When two or more sensors fail simultaneously, the system will activate emergency mode, utilizing remaining available sensors and historical data to establish a temporary monitoring plan, ensuring safe equipment operation, and simultaneously issuing a sensor replacement warning.

[0038] S104. Calculate the ratio of the actual flow rate of compressed air to the rated flow rate, and combine it with the Pearson correlation coefficient of pressure and current to obtain the performance anomaly index.

[0039] The performance anomaly indicators are primarily based on the ratio of actual compressed air flow to rated flow, and the correlation analysis between pressure and current. The Pearson correlation coefficient is used to characterize the consistency of pressure and current changes, reflecting the stability of the equipment's operating status.

[0040] Specifically, the system first monitors the exhaust pressure, triggering an emergency alarm when it exceeds the nominal value by ±20%. Simultaneously, it monitors the exhaust temperature; if it consistently exceeds 100°C, immediate shutdown is required. For determining abnormal flow, the system classifies insufficient flow as occurring when the actual flow rate is 15% lower than the rated value for more than 10 minutes. Furthermore, the system assesses the coordination of equipment operation by calculating the Pearson correlation coefficient between pressure and current; a significant decrease in this coefficient typically indicates abnormal equipment performance.

[0041] The system also establishes a graded response mechanism for performance anomalies. When a blockage in the intake filter or a pressure sensor drift is detected, the system issues a Level 1 warning; when a cooling system malfunction (fan or heat exchanger failure) or an ambient temperature exceeding the limit is detected, the system issues a Level 2 warning; and when a pipeline leak is confirmed (detection rate must reach 90%) or the scroll plate is worn, the system issues a Level 3 warning.

[0042] Furthermore, performance parameters may fluctuate during implementation. To address this, the system employs a dynamic benchmark correction method, automatically adjusting the performance evaluation benchmark based on factors such as ambient temperature and load changes to avoid false alarms. For example, when the ambient temperature is high, the system will correspondingly increase the alarm threshold for exhaust temperature to ensure accurate judgment.

[0043] S105. Conduct trend analysis on the temperature change rate, and assess the system contamination risk by combining dew point monitoring data and oil vapor concentration.

[0044] The system contamination risk mainly involves two aspects: moisture contamination and oil contamination. Trend analysis of temperature change rate can reflect changes in the system's thermodynamic characteristics, dew point monitoring data is used to assess the moisture content in compressed air, and oil vapor concentration is directly related to the purity of medical gases.

[0045] Specifically, the system monitors the dew point temperature in real time, ensuring it remains below -40°C. When the relative humidity exceeds 60%, the system issues a moisture contamination warning, as the microbial proliferation rate increases to three times the normal level at this point. Simultaneously, it continuously monitors oil vapor concentration, triggering an alarm when the detected value approaches 0.1 mg / m³ to prevent respiratory damage caused by oil contamination. Furthermore, an ultrasonic leak detector (with a sensitivity of up to 0.5 mL / min) is used to detect gas leaks.

[0046] In some embodiments, the data processing system can establish a multi-level pollution risk assessment model: first, it identifies abnormal patterns in temperature, dew point, and oil and gas concentration through real-time data analysis; then, it analyzes the causes of the abnormalities by combining factors such as ambient temperature and humidity and equipment operating time; next, it assesses the pollution risk level under different operating conditions; and finally, it generates graded early warning information and disposal suggestions. Optionally, the data processing system can also adopt a dynamic threshold adjustment method, first establishing a baseline data model under different seasons and operating conditions, then adjusting the early warning thresholds of various monitoring indicators in real time, then optimizing the system's response strategy, and finally forming an adaptive pollution prevention and control mechanism.

[0047] Understandably, AI prediction technology can also be used to analyze time-series data of various parameters through LSTM models, enabling early warnings of more than 72 hours in advance. For example, when an abnormal rate of temperature change is detected and the dew point begins to rise, the system can predict potential pollution risks in advance, providing sufficient preparation time for equipment maintenance and adjustments.

[0048] S106. Based on the mechanical fault type, electrical fault type, abnormal performance indicators, and system pollution risk, predict the fault development trend through a preset deep learning model and generate an early warning signal.

[0049] The early warning signals include information such as fault type, severity, development trend, and handling suggestions, providing a basis for equipment maintenance decisions.

[0050] Specifically, the system first inputs data such as mechanical fault characteristics (including characteristic parameters of scroll plate wear, bearing failure, and main unit seizure), electrical fault characteristics (including harmonic distortion rate and current fluctuation trends), abnormal performance indicators (including pressure, flow, and temperature anomalies), and system pollution risks (including dew point and oil and gas concentrations) into a pre-set deep learning model. The model uses a Long Short-Term Memory (LSTM) network structure and predicts the development trend of faults by analyzing the temporal characteristics of historical fault cases and real-time monitoring data. When the prediction results show that the fault risk exceeds the warning threshold, the system generates a warning signal that includes the fault type, the expected development time, and recommended handling measures.

[0051] In some embodiments, the deep learning model can employ a multi-level early warning mechanism. Optionally, it can first establish a time-series evolution model of fault characteristics to analyze the development patterns of different types of faults, then combine equipment operating conditions and environmental factors to predict key nodes in fault development, then assess the risk level of various fault combinations, and finally generate graded early warning signals. Optionally, the model can also establish a fault feature fusion analysis framework to improve the accuracy and predictability of fault prediction through collaborative analysis of multi-source data, thereby achieving early warning of complex faults.

[0052] Furthermore, frequent triggering of warning signals may occur during implementation. To address this, the system employs a warning optimization strategy, setting fault priorities and warning time windows to avoid redundant warning information. For example, when multiple related faults trigger warnings simultaneously, the system will prioritize displaying fault information with higher risk levels and integrate related faults into a single warning sequence for centralized handling by maintenance personnel.

[0053] In the above embodiments, the system collects operating parameters of the scroll air compressor by deploying multiple sensors, establishing a comprehensive monitoring system including pressure pulsation spectrum and vibration envelope spectrum. For fault diagnosis, it not only uses harmonic amplitude ratio and bearing characteristic frequency to identify mechanical faults, but also combines current waveform analysis to judge electrical system anomalies. Pearson correlation coefficient is also introduced for performance evaluation, realizing the correlation analysis between pressure and current changes. Addressing the special requirements of medical gases, this application focuses on system contamination risk, ensuring gas supply safety through dew point monitoring and oil / gas concentration analysis. Finally, deep learning technology is used to model various fault characteristics, achieving accurate prediction of equipment status and providing a scientific basis for preventative maintenance.

[0054] In the above embodiments, the system achieves early warning of scroll air compressor failures through multi-source data analysis and deep learning prediction. However, medical gas systems not only need to monitor equipment operating status but also ensure that the gas supply quality meets medical standards. To further ensure the purity and safety of medical gases, this application also provides another method for predicting scroll air compressor failures in medical gas systems. The following is a combination of... Figure 2 Another method for predicting the failure of a scroll air compressor in a medical gas system, as described in the embodiments of this application, is as follows: Please see Figure 2 This is another flowchart illustrating a method for predicting the failure of a vortex air compressor in a medical gas system, as described in this application.

[0055] Following step S105, the following steps are also included: S201. Obtain data on moisture content, oil content, and particulate matter concentration in the gas.

[0056] S202. Determine the purity grading standards and gas quality scoring rules for medical gases.

[0057] Medical gases are classified into four levels based on their moisture, oil, and particulate matter content. The gas quality scoring rule adopts a weighted scoring method, setting weight coefficients according to the degree of impact of each pollutant indicator on medical safety, and taking into account the degree of deviation of pollutant concentration from the standard limit.

[0058] S203. Calculate the comprehensive gas quality score based on the moisture content, oil content, and particulate matter concentration data, combined with dew point monitoring data and oil vapor concentration.

[0059] Specifically, the monitoring data is first standardized to eliminate dimensional differences; then, a weighted score is calculated based on preset weighting coefficients; finally, the scoring results are dynamically adjusted based on the changing trends of dew point and oil vapor concentration. For example, if a continuous rise in dew point temperature or increased fluctuations in oil vapor concentration are detected, the overall score is lowered accordingly.

[0060] In some embodiments, the system first determines the baseline range for each pollutant indicator based on historical data, then analyzes the fluctuation patterns of the indicators under different operating conditions, next establishes an indicator correlation analysis model, and finally forms a dynamic scoring mechanism. For example, in seasons with high humidity, the system will appropriately increase the allowable fluctuation range of the moisture index to avoid misjudgment.

[0061] Furthermore, during implementation, multiple pollutant indicators may fluctuate simultaneously. To address this, the system first establishes a pollutant interaction model to analyze the correlation effects between various indicators; then, it adjusts the weighting coefficients based on the degree of interaction; finally, it uses a fuzzy comprehensive evaluation method to obtain a more accurate quality score. For example, when both moisture and oil content are detected to be rising simultaneously, the system analyzes whether there are common causes such as condenser malfunction and adjusts the scoring strategy accordingly.

[0062] S204. When the overall gas quality score is lower than the preset standard for the corresponding purity level, a graded treatment suggestion is generated based on the exceedance of various pollutant indicators.

[0063] Specifically, a Level 1 warning is triggered when only a single indicator slightly exceeds the limit, suggesting an increase in monitoring frequency; a Level 2 warning is triggered when multiple indicators exceed the limit or a single indicator severely exceeds the limit, suggesting a check of the filtering system; and a Level 3 warning is triggered when the overall score significantly decreases and shows a continuous deteriorating trend, suggesting system shutdown for maintenance. Furthermore, the system will provide specific handling reference solutions for each type of exceedance based on historical maintenance records.

[0064] In some embodiments, the system first establishes a mapping relationship between pollutant characteristics and treatment solutions. Then, it analyzes the effectiveness of different treatment solutions by combining equipment operating status and maintenance records. Finally, it intelligently matches the optimal treatment solution based on the current exceedance situation. For example, when it is found that particulate matter concentration exceedance is highly correlated with filter efficiency decline, the system will prioritize recommending filter maintenance solutions.

[0065] S205. Obtain filter status data and pipeline system pressure distribution data of the air compressor system.

[0066] The filter status data includes parameters such as differential pressure, usage time, and filtration efficiency. Pipeline system pressure distribution data is collected by pressure sensors deployed at key nodes of the gas supply network to analyze system pressure loss and gas flow characteristics. Specifically, the filter differential pressure is monitored in real time by a differential pressure sensor; when the differential pressure exceeds 50 kPa, it indicates that the filter element needs to be replaced. Usage time is automatically accumulated based on actual operating time. Filtration efficiency is calculated by comparing the inlet and outlet particulate matter concentrations.

[0067] S206. Determine the location and monitoring parameters of the system's key control points.

[0068] The key control points include locations such as the air compressor outlet, air tank, before and after the dryer, and end-user gas consumption points. Each control point is equipped with corresponding monitoring parameters, such as pressure, temperature, dew point, and contaminant concentration. Specifically, pressure and temperature are monitored at the air compressor outlet; pressure and moisture content are monitored at the air tank; dew point changes are measured before and after the dryer; and gas purity is detected at the end-user gas consumption points. The system monitors these key control points online 24 hours a day, recording the changing trends of each parameter in real time.

[0069] Optionally, the system can employ an intelligent point selection strategy: first, it analyzes the gas flow characteristics using a computational fluid dynamics model; then, it identifies key locations where pollutants are prone to accumulate; and finally, it optimizes the monitoring point layout. For example, monitoring points can be added at locations where airflow is prone to eddies, such as pipe bends, to improve the targeting of monitoring.

[0070] Furthermore, during implementation, there may be instances of asynchronous data from monitoring points. To address this, the system employs a data synchronization acquisition strategy: first, a time-series correlation model is established between monitoring points; then, the data acquisition time is automatically calibrated based on the gas flow time; and finally, a unified monitoring result is generated through a data fusion algorithm to ensure the temporal consistency of data from all monitoring points.

[0071] S207. Based on the filter status data and pipeline system pressure distribution data, analyze the propagation path of pollutants in the system.

[0072] The pollutant propagation path analysis is based on the filter's capture efficiency and the pressure gradient distribution in the pipeline network. Specifically, by monitoring changes in pollutant concentration before and after each stage of the filter, combined with pressure distribution data, the migration patterns of pollutants in the system are tracked. For example, when abnormal pressure fluctuations occur in a section of the pipeline, and the pollutant retention capacity of the downstream filter suddenly increases, it indicates that pollutant accumulation may exist in that section of the pipeline.

[0073] Optionally, the system first constructs a numerical simulation model of the pipeline flow field to analyze the airflow distribution characteristics under different operating conditions; then, combining the physical and chemical characteristics of pollutants, it predicts their diffusion behavior in the system; finally, it verifies and optimizes the model parameters using measured data. For example, CFD simulation analysis can be used to identify airflow dead zones and vortex regions, predicting locations where pollutants are likely to accumulate.

[0074] Furthermore, the system first uses characteristic pollutant tracing technology to label pollutants from different sources, then analyzes the spatiotemporal distribution patterns of various pollutants, and finally establishes a quantitative assessment model for pollutant transmission. When an abnormal pollutant concentration is detected, the system can quickly trace the pollution source and determine the transmission path, providing precise guidance for systemic purification.

[0075] S208. Based on the propagation path and the location and monitoring data of key control points, generate system purification plans and maintenance strategies.

[0076] Finally, the system develops specific treatment measures for different types of pollutants, and the maintenance strategy determines the maintenance cycle and method based on the equipment status and the degree of pollution.

[0077] In the above embodiments, by establishing a medical gas purity grading standard and a quality scoring system, accurate assessment of gas quality is achieved. The system not only considers key pollutant indicators such as moisture, oil, and particulate matter, but also adapts to fluctuations under different operating conditions through a dynamic scoring mechanism. Based on this, by analyzing filter status and pipeline pressure distribution, combined with real-time monitoring data from key control points, a pollutant propagation path model is constructed, providing a scientific basis for system purification and maintenance.

[0078] In some embodiments, before predicting the fault development trend using a preset deep learning model, the system also optimizes the prediction performance of the deep learning model by analyzing historical fault data. First, it acquires fault type, fault characteristics, maintenance measures, and equipment runtime data from historical fault records. Fault samples are then stratified according to equipment runtime to establish fault feature libraries for different operating stages. Based on fault type, fault characteristics, and maintenance measures, the influence weights of various fault features in each fault feature library are calculated. Finally, the calculated influence weights are used to optimize the prediction accuracy of the deep learning model.

[0079] Specifically, fault types include the aforementioned mechanical faults, electrical faults, performance anomalies, and pollution risks; fault characteristics include the characteristic parameters and development process of each type of fault; maintenance measures record the specific handling methods and their effects for different faults; and equipment runtime reflects the equipment's usage status and life cycle characteristics. The system divides runtime into three stages: initial, middle, and late stages, and establishes corresponding feature libraries for each. A correlation analysis method is used to determine the importance of different features in fault prediction by evaluating the correlation between fault features and maintenance effects. For example, for bearing faults, the weight of vibration features may be significantly higher in the later stages than in the initial stages. By integrating the weight information into the model's loss function, the model's sensitivity to important features is improved. Simultaneously, the system periodically updates the feature weights, enabling the model to adapt to dynamic changes in equipment status.

[0080] In some embodiments, after collecting equipment operation data, the fault diagnosis strategy is further optimized by analyzing the computer room environmental parameters. First, the ambient temperature, humidity, atmospheric pressure, and ventilation data of the computer room where the scroll air compressor is located are acquired. Based on these environmental parameters, an environmental adaptability index is generated. The fault characteristic threshold is dynamically adjusted according to the environmental adaptability index. When the environmental adaptability index is lower than the preset standard, suggestions for improving the computer room environment are generated.

[0081] Specifically, ambient temperature and humidity are monitored in real time using temperature and humidity sensors, requiring temperature control within the range of 18-28℃ and relative humidity maintained between 45%-65%. Atmospheric pressure is measured using a digital barometer to assess air intake conditions. Ventilation status is evaluated using wind speed sensors and airflow monitoring devices to assess the ventilation effect of the computer room. The system uses a multi-parameter weighted scoring method to calculate the environmental adaptability index: first, each environmental parameter is standardized; then, weighting coefficients are set according to the degree of influence of each parameter on equipment operation; finally, a comprehensive score is calculated. When the ambient temperature is too high, it will increase the heat dissipation burden on the equipment, correspondingly reducing the environmental adaptability index; when ventilation is insufficient, it may lead to heat accumulation and pollutant buildup in the computer room, also reducing the score. In high-temperature environments, the system will appropriately lower the alarm threshold for temperature-related faults; under high humidity conditions, the sensitivity of moisture content monitoring will be increased. When the environmental adaptability index is lower than the preset standard, the system will generate specific improvement suggestions, such as adding ventilation equipment or installing dehumidifiers.

[0082] In some embodiments, optionally, the system can establish a correlation model between environmental parameters and equipment performance: first, analyze the impact of environmental changes on equipment operation in historical data; then, establish an environmental adaptability assessment model; and finally, predict potential operational risks based on real-time environmental data. For example, if the system finds that the equipment failure rate increases significantly under certain combinations of environmental conditions, it will issue an early warning and suggest preventative measures. Furthermore, sudden changes in environmental parameters may occur during implementation. Optionally, the system first confirms the environmental anomaly through cross-validation using multiple sensors; then, it assesses the potential impact of the anomaly on equipment operation; and finally, it automatically adjusts equipment operating parameters or triggers protective measures. For example, when a sharp increase in the computer room temperature is detected, the system will automatically increase the equipment cooling water flow and issue an environmental anomaly alarm.

[0083] The system in the embodiments of this invention is described below from the perspective of hardware processing. Please refer to [link / reference needed]. Figure 3 This is a schematic diagram of the physical device structure of a medical gas system vortex air compressor fault prediction system provided in an embodiment of this application.

[0084] It should be noted that, Figure 3 The structure of the system shown is merely an example and should not impose any limitations on the functionality and scope of use of the embodiments of the present invention.

[0085] like Figure 3 As shown, the system includes a CPU 301, which can perform various appropriate actions and processes according to a program stored in ROM 302 or a program loaded from storage section 308 into RAM 303, such as executing the methods described in the above embodiments. RAM 303 also stores various programs and data required for system operation. The CPU 301, ROM 302, and RAM 303 are interconnected via bus 304. I / O interface 305 is also connected to bus 304.

[0086] The following components are connected to I / O interface 305: input section 306 including a camera, infrared sensor, etc.; output section 307 including a liquid crystal display (LCD) and speakers, etc.; storage section 308 including a hard disk, etc.; and communication section 309 including a network interface card such as a LAN (Local Area Network) card and a modem, etc. Communication section 309 performs communication processing via a network such as the Internet. Drive 310 is also connected to I / O interface 305 as needed. Removable media 311, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 310 as needed so that computer programs read from them can be installed into storage section 308 as needed.

[0087] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing computer programs for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 309, and / or installed from removable medium 311. When the computer program is executed by CPU 301, it performs the various functions defined in the present invention.

[0088] It should be noted that the computer-readable medium shown in the embodiments of the present invention can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, wherein a computer-readable computer program is carried. The transmitted data signal can take many forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.

[0089] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of the present invention. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, may be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0090] In another aspect, the present invention also provides a computer-readable storage medium, which may be included in the system described in the above embodiments; or it may exist independently and not assembled into the system. The storage medium carries one or more computer programs that, when executed by a processor of a system, cause the system to implement the methods provided in the above embodiments.

[0091] The above-described embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit it. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of this application.

[0092] As used in the above embodiments, depending on the context, the term "when..." can be interpreted as "if...", "after...", "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if (the stated condition or event) is interpreted as "if determining...", "in response to determining...", "when (the stated condition or event) is detected", or "in response to detecting (the stated condition or event)".

[0093] In the above embodiments, implementation can be achieved entirely or partially through software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented entirely or partially in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., solid-state drive), etc.

[0094] Those skilled in the art will understand that all or part of the processes in the methods of the above embodiments can be implemented by a computer program instructing related hardware. This program can be stored in a computer-readable storage medium, and when executed, it can include the processes described in the above method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A method for predicting faults in a scroll air compressor for a medical gas system, characterized in that, include: Collect operating data of the scroll air compressor to obtain pressure pulsation spectrum, vibration envelope spectrum, current waveform and temperature change rate; Based on the analysis of the first and second harmonic amplitude ratios of the pressure pulsation spectrum, the meshing state of the scroll plate is determined, and the type of mechanical fault is determined by combining the bearing fault characteristic frequency in the vibration envelope spectrum. Analyze the harmonic distortion rate and fluctuation trend of the current waveform to determine the type of electrical fault; Calculate the ratio of actual compressed air flow rate to rated flow rate, and combine it with the Pearson correlation coefficient between pressure and current to obtain performance anomaly indicators; Trend analysis was performed on the temperature change rate, and the system contamination risk was assessed by combining dew point monitoring data and oil vapor concentration. Based on the mechanical fault type, electrical fault type, abnormal performance indicators, and system pollution risk, a pre-set deep learning model is used to predict the fault development trend and generate an early warning signal.

2. The method according to claim 1, characterized in that, The mechanical fault types include scroll plate wear and bearing failure. The method involves analyzing the amplitude ratio of the first and second harmonics in the pressure pulsation spectrum to determine the scroll plate meshing state, and combining this with the bearing failure characteristic frequencies in the vibration envelope spectrum to determine the mechanical fault type, specifically including: When the ratio of the amplitude of the first and second harmonics in the pressure pulsation spectrum exceeds the preset harmonic ratio value, and the exhaust pressure fluctuation exceeds the preset fluctuation range, it is determined that the scroll plate is worn. When the energy of the bearing fault characteristic frequency in the vibration envelope spectrum exceeds a preset energy threshold, and the effective vibration value exceeds a preset vibration limit or the temperature change rate exceeds a preset temperature change threshold, it is determined to be a bearing fault.

3. The method according to claim 2, characterized in that, The mechanical fault types also include main unit seizure. After analyzing the harmonic distortion rate and fluctuation trend of the current waveform to determine the electrical fault type, the method further includes: Determine whether the current waveform exhibits overload characteristics; Detect the rotational resistance of the scroll disk and obtain rotational resistance characteristic data; When an overload characteristic is detected in the current waveform and the rotational resistance characteristic data exceeds a preset resistance threshold, it is determined that the main unit is locked.

4. The method according to claim 1, characterized in that, Before predicting the fault development trend using a preset deep learning model, the method further includes: Obtain fault type, fault characteristics, maintenance measures, and equipment runtime data from historical fault records; The fault samples are stratified according to the operating time of the equipment to establish a fault feature library for different operating stages; Based on the fault type, fault characteristics and maintenance measures, calculate the influence weight of each type of fault characteristic in each fault characteristic library; The prediction accuracy of the deep learning model is optimized using the influence weights.

5. The method according to claim 1, characterized in that, After the step of collecting the operating data of the scroll air compressor, the method further includes: Obtain data on ambient temperature, humidity, atmospheric pressure, and ventilation conditions in the machine room where the scroll air compressor is located; Based on the environmental temperature, humidity, atmospheric pressure, and ventilation data, an environmental adaptability index is generated. The fault characteristic threshold is dynamically adjusted based on the environmental adaptability index. When the environmental adaptability index is lower than the preset standard, suggestions for improving the data center environment are generated.

6. The method according to claim 1, characterized in that, After the step of performing trend analysis on the temperature change rate, combining dew point monitoring data and oil vapor concentration to assess the system contamination risk, the method further includes: Acquire data on moisture content, oil content, and particulate matter concentration in the gas; Establish standards for grading the purity of medical gases and rules for scoring gas quality; Based on the moisture content, oil content, and particulate matter concentration data, combined with the dew point monitoring data and oil vapor concentration, a comprehensive gas quality score is calculated. When the overall gas quality score is lower than the preset standard for the corresponding purity level, a graded treatment suggestion is generated based on the exceedance of various pollutant indicators.

7. The method according to claim 6, characterized in that, After the step of generating a graded treatment suggestion based on the exceedance of various pollutant indicators when the overall gas quality score is lower than the preset standard for the corresponding purity level, the method further includes: Acquire filter status data and pipeline system pressure distribution data of the air compressor system; Determine the location and monitoring parameters of the system's key control points; Based on the filter status data and pipeline system pressure distribution data, the propagation path of pollutants in the system is analyzed; Based on the propagation path and the location and monitoring data of the key control points, a system purification plan and maintenance strategy are generated.

8. A fault prediction system for a scroll air compressor in a medical gas system, characterized in that, The system includes: One or more processors and a memory; the memory is coupled to the one or more processors, the memory being used to store computer program code, the computer program code including computer instructions, the one or more processors invoking the computer instructions to cause the system to perform the method as described in any one of claims 1-7.

9. A computer-readable storage medium comprising instructions, characterized in that, When the instructions are executed on the system, the system performs the method as described in any one of claims 1-7.

10. A computer program product, characterized in that, When the computer program product is run on the system, the system performs the method as described in any one of claims 1-7.

Images