A method and system for testing the sealing performance of pneumatic connectors

Abstract

The application discloses a kind of based on dynamic pressure spectrum and multi-parameter fusion's pneumatic connector sealing performance intelligent test method and system, belong to pneumatic component sealing detection technical field.The method is by to the measured joint simulated actual working condition dynamic pressure spectrum, synchronous acquisition pressure, temperature and ultrasonic signal, using temperature real-time compensation algorithm eliminates environmental thermodynamic interference, and by deducting system baseline data separates out the pressure change caused purely by leakage;Finally, the fusion pressure leakage rate and ultrasonic energy double characteristic parameters, through intelligent decision model, the sealing performance is comprehensively evaluated.The system includes gas path tool module, multi-sensing module and control processing module, can effectively solve the problem that traditional static pressure maintenance method is greatly influenced by temperature, cannot detect transient leakage and the like, with the advantages of high detection precision, strong anti-interference ability, can simulate real working condition, significantly improve test reliability and efficiency.

Description

Technical Field

[0001] This invention relates to the field of fluid transmission and sealing performance testing technology, and in particular to a high-precision, intelligent testing method for the sealing performance of pneumatic components, especially pneumatic joints, and a system for implementing the method. Background Technology

[0002] As an indispensable basic component in pneumatic systems, the quality of pneumatic joints directly affects the overall system's working efficiency, energy consumption, reliability, and safety. In high-end fields such as aerospace, automotive manufacturing, precision instruments, and industrial automation, the requirements for "zero leakage" or "micro-leakage" pneumatic joints are becoming increasingly stringent.

[0003] Currently, the mainstream method for testing the sealing performance of pneumatic joints in the industry is the static pressure holding method (pressure drop method). Its basic principle is: a certain pressure of gas (usually clean compressed air or nitrogen) is introduced into the sealed test chamber containing the joint under test. After the gas source is turned off, the pressure is held for a period of time. The pressure drop in the chamber is monitored by a high-precision pressure sensor. The leakage rate is calculated according to the ideal gas law and compared with a preset qualified threshold.

[0004] However, the traditional static pressure holding method has several inherent drawbacks and is difficult to meet the increasingly demanding testing requirements:

[0005] High sensitivity to ambient temperature: According to the ideal gas law (PV = nRT), the pressure of a gas within a sealed cavity is positively correlated with its temperature. Even small fluctuations in ambient temperature or gas temperature during testing (e.g., ±1°C) can cause significant pressure changes, potentially much larger than the pressure drop caused by a minor leak, leading to misjudgments. Although mitigation methods such as using a constant temperature chamber or testing at night can be employed, these significantly reduce testing efficiency and cannot be applied to online detection.

[0006] Poor simulation of operating conditions: In actual pneumatic systems, the pressure on the joints is often dynamically changing (such as the reciprocating motion of the cylinder and the frequent start and stop of the solenoid valve). Static pressure holding tests cannot reproduce this alternating stress condition, and therefore may not be able to detect "transient leakage" that only occurs when the pressure fluctuates drastically or leakage caused by material fatigue or stress relaxation.

[0007] The trade-off between testing efficiency and sensitivity: To detect even the smallest leaks, it is usually necessary to extend the holding time or increase the test chamber volume to amplify the pressure drop signal. However, this inevitably leads to a significant increase in the testing time per unit, making it impossible to meet production cycle requirements.

[0008] The information dimension is limited, and the misjudgment rate is high: relying on only one parameter, pressure, for judgment cannot distinguish whether the pressure drop is due to a real leak or the system's own adsorption, deformation, or temperature change, which easily generates "false leak" alarms and increases the cost of re-inspection.

[0009] Unable to pinpoint the leak: The pressure method is a holistic test. Once a leak is detected, it cannot directly indicate the specific location of the leak, which makes subsequent process improvements or product rework difficult.

[0010] To address these issues, several improved techniques have been proposed, such as the differential pressure method, which uses a differential pressure sensor instead of an absolute pressure sensor to improve the resolution of small pressure drops; or using a flow meter to directly measure leakage flow. However, these methods still fail to fundamentally solve the problems of temperature interference and dynamic operating condition simulation, and high-precision flow meters are expensive.

[0011] Therefore, developing a highly efficient and reliable method and system for testing the sealing performance of pneumatic joints that is immune to temperature interference, simulates dynamic working conditions, and achieves multi-parameter fusion interpretation has become a technical challenge that urgently needs to be solved in this field. Summary of the Invention

[0012] The primary objective of this invention is to overcome the shortcomings of the prior art and provide a method for testing the sealing performance of pneumatic joints that has strong anti-interference capabilities, can simulate dynamic working conditions, and has high testing accuracy and efficiency.

[0013] To achieve the above objectives, the present invention provides the following technical solution:

[0014] A method for testing the sealing performance of a pneumatic joint, characterized by comprising the following steps:

[0015] System baseline self-learning steps: Perform dynamic pressure testing using a standard leak-free workpiece to acquire and store the pressure baseline data of the test system itself;

[0016] Dynamic pressure test procedure: Apply a preset dynamic pressure change curve to the test chamber containing the pneumatic connector under test.

[0017] Multi-parameter synchronous acquisition step: During the dynamic pressure test, the pressure signal, temperature signal and ultrasonic signal in the test chamber are acquired synchronously and at high speed.

[0018] Data fusion analysis steps: Based on the temperature signal, perform real-time temperature compensation on the pressure signal to obtain temperature-compensated pressure data; compare and subtract the temperature-compensated pressure data from the pressure baseline data to obtain net pressure change data; simultaneously, analyze the energy characteristics of the ultrasonic signal.

[0019] Intelligent comprehensive judgment steps: Based on the leakage rate characteristic value calculated from the net pressure change data and the energy characteristic value of the ultrasonic signal, a comprehensive judgment model is used to output the evaluation result of the sealing performance of the pneumatic joint.

[0020] The present invention further provides that the dynamic pressure change curve is a periodically changing waveform, including square wave, sine wave, triangle wave or a custom pressure-time curve simulating actual working conditions.

[0021] The present invention further specifies that the real-time temperature compensation in the data fusion analysis step is specifically based on the ideal gas law, which calculates the original pressure value P collected at each moment. raw Compensation to a fixed reference temperature T ref The calculation formula is: P compensated (t)=P raw (t)*(T ref / T(t)), where T(t) is the real-time temperature value collected synchronously.

[0022] The present invention further provides that the energy characteristic analysis of the ultrasonic signal in the data fusion analysis step refers to calculating the root mean square (RMS) or peak value of the acquired raw ultrasonic signal within a specific time window as a characteristic value after bandpass filtering.

[0023] The present invention further specifies that the comprehensive judgment model in the intelligent comprehensive judgment step is a judgment logic based on a rule base:

[0024] If the leakage rate characteristic value is less than the first threshold and the ultrasonic energy characteristic value is less than the second threshold, it is considered qualified.

[0025] If the leakage rate characteristic value is greater than the first threshold but the ultrasonic energy characteristic value is less than the second threshold, it is determined to be a micro-leakage;

[0026] If the leakage rate characteristic value is less than the first threshold but the ultrasonic energy characteristic value is greater than the second threshold, it is determined that there is interference.

[0027] If the leakage rate characteristic value is greater than the first threshold and the ultrasonic energy characteristic value is greater than the second threshold, it is determined to be a serious leakage.

[0028] The present invention further provides that the comprehensive judgment model in the intelligent comprehensive judgment step is an algorithm model based on machine learning or fuzzy logic, which takes the leakage rate feature value and the ultrasonic energy feature value as input and outputs a continuous sealing quality score.

[0029] The present invention also provides a pneumatic joint sealing performance testing system for the method, characterized in that it comprises:

[0030] The air circuit and tooling module includes a test chamber and a high-speed pneumatic control valve, which are used to connect the connector under test and generate the dynamic pressure change curve.

[0031] The sensing module includes a pressure sensor for detecting the pressure inside the test chamber, a temperature sensor for detecting the temperature of the gas inside the test chamber, and an ultrasonic sensor for capturing high-frequency sound waves generated by the leak.

[0032] The control and data acquisition module includes a high-speed synchronous data acquisition card and a main controller. The data acquisition card is connected to the sensing module for synchronously acquiring sensor signals. The main controller is connected to the data acquisition card and the pneumatic control valve for controlling the generation of pressure curves and executing the data fusion analysis and intelligent comprehensive judgment steps.

[0033] In a further embodiment of the present invention, the high-speed pneumatic control valve is a high-speed proportional valve or a combination valve consisting of two high-speed switching solenoid valves.

[0034] The present invention further provides that the volume of the test cavity is accurately measurable and replaceable to accommodate pneumatic connectors of different diameters; the temperature sensor is a PT100 platinum resistance thermometer or an NTC thermistor, and its probe extends directly into the test cavity to contact the gas.

[0035] The present invention also provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the method described.

[0036] This invention includes at least one of the following beneficial effects:

[0037] Eliminating temperature interference: By synchronously acquiring temperature and performing real-time compensation based on a physical model, the measurement errors caused by ambient temperature fluctuations and gas thermodynamic effects are fundamentally solved, freeing the test from the constant temperature laboratory and making it suitable for workshop sites, while significantly improving measurement accuracy and reliability.

[0038] Achieving realistic working condition simulation: Dynamic pressure spectrum testing can effectively reproduce the pressure and stress changes that the joint experiences in actual operation, thereby detecting transient leaks and stress failures hidden under static pressure, and the test results are more instructive for engineering.

[0039] Multi-dimensional information fusion leads to more scientific judgment: Combining pressure drop and ultrasonic waves as two physical signals for comprehensive judgment, they mutually verify and complement each other. This allows for both quantitative assessment of the overall leakage rate and qualitative location of the leak point, significantly reducing false alarm and false alarm rates.

[0040] High efficiency and high sensitivity are combined: dynamic testing is typically faster than long-duration static pressure holding. Simultaneously, high-precision sensors and advanced signal processing algorithms ensure extremely high detection sensitivity while shortening testing time, meeting the dual requirements of efficiency and quality in modern production.

[0041] Intelligent and digital systems: The system automatically completes all testing, analysis, and judgment processes, and is easy to operate. Full data recording facilitates quality traceability, statistical analysis, and process improvement, providing fundamental data support for achieving intelligent manufacturing and digital factories. Attached image description:

[0042] Figure 1 This is a flowchart illustrating an embodiment of the present invention. Detailed Implementation

[0043] The following will describe in detail the implementation of this application with reference to the accompanying drawings and embodiments, so that the implementation process of how this application uses technical means to solve technical problems and achieve technical effects can be fully understood and implemented accordingly.

[0044] In existing technologies, pneumatic joint sealing tests primarily employ the static pressure holding method, which determines leakage by monitoring the pressure drop within the sealed test chamber. This method suffers from inherent drawbacks, including high temperature sensitivity, inability to simulate dynamic operating conditions, a trade-off between testing efficiency and sensitivity, and a high false positive rate for single parameters. In practical applications, ambient temperature fluctuations can distort pressure data, static testing cannot capture transient leaks, extending the pressure holding time impacts production efficiency, and relying solely on pressure parameters makes it difficult to distinguish between actual leaks and system interference.

[0045] To address the aforementioned issues, the research and development process revealed that temperature compensation and dynamic pressure excitation are key breakthroughs compared to traditional methods. By analyzing the correlation between pressure signals and temperature changes, a system baseline data model was established to eliminate inherent interference. Considering the prevalence of pressure fluctuations in actual working conditions, a dynamic pressure testing environment was designed to induce potential leaks. To overcome the limitations of single parameters, ultrasonic signals were introduced as an auxiliary criterion. Finally, a multi-dimensional data fusion mechanism was formed, and a cross-validation model was constructed to improve the accuracy of the judgment.

[0046] like Figure 1 As shown in the figure, this embodiment proposes a method for testing the sealing performance of a pneumatic joint, which includes the following steps:

[0047] S1: System baseline self-learning and calibration

[0048] During initial use or after periodic maintenance, replace the workpiece under test with a calibrated, leak-free "standard workpiece" to perform one or more complete dynamic pressure tests. Simultaneously record the pressure change data P within the test chamber throughout the entire process. baseline(t) and temperature change data T baseline (t). After averaging and processing these data, they are stored as a system baseline file. This file characterizes the inherent properties of the test system itself (piping, cavities, valve internal leakage) under dynamic pressure, providing a benchmark for subsequent subtraction of system errors from the test data.

[0049] S2: Installing the workpiece and preparing for testing

[0050] The pneumatic connector under test is installed and sealed in the test fixture. The operator selects or inputs the model of the connector under test through the human-machine interface, and the system automatically calls up the preset test parameters, including: dynamic pressure curve waveform, amplitude, period, number of cycles, test cavity volume, and judgment thresholds for pressure and ultrasonic signals.

[0051] S3: Apply dynamic pressure spectrum

[0052] The system controller drives a high-speed pneumatic control valve (such as a proportional valve or a high-speed switching valve) to apply a pre-set dynamic pressure change curve to the test chamber. This curve is a function of time, P. set (t), designed to simulate the pressure conditions that the joint may experience in practical applications. Its waveform can be, but is not limited to:

[0053] Square wave: A step change between two or more pressure values, used to test the instantaneous sealing performance of a joint under sudden pressure changes.

[0054] Sine wave: The pressure changes in a sinusoidal pattern, simulating a periodic load.

[0055] Sawtooth / triangular wave: Pressure rises and falls at a constant rate.

[0056] Custom waveform: Any pressure curve recorded and reproduced based on user-specific actual working condition data.

[0057] S4: High-speed synchronous acquisition of multiple parameters

[0058] Throughout the dynamic stress test, the system uses a high-speed synchronous data acquisition card to synchronously trigger and acquire signals of the following three key physical quantities at a sampling rate of no less than 1kHz:

[0059] Pressure signal P raw (t): Derived from a high-precision pressure sensor, directly reflecting the instantaneous air pressure value inside the test chamber.

[0060] Temperature signal T(t): originates from a high-response-speed temperature sensor, which monitors the temperature change of the gas inside the test chamber in real time.

[0061] Ultrasonic signal U raw(t): Originates from an ultrasonic sensor, capturing high-frequency acoustic signals (typically above 40kHz) generated by the turbulence of leaking gas.

[0062] S5: Data Fusion and Compensation Analysis

[0063] Includes the following sub-steps:

[0064] S51: Signal Preprocessing: Filtering and denoising the acquired raw signals. Low-pass filtering is primarily used for pressure and time signals to remove high-frequency electrical noise. Band-pass filtering is used for ultrasonic signals to retain specific high-frequency bands (e.g., 40kHz-100kHz) relevant to the leak.

[0065] S52: Temperature Compensation Calculation: Using the synchronously acquired temperature data T(t), and based on the ideal gas law, the original pressure value P acquired at each moment is calculated. raw Compensation to a fixed reference temperature T ref The calculation formula is: P compensated (t)=P raw (t)*(T ref / T(t)), where T(t) is the real-time temperature value collected synchronously.

[0066] After this step, P compensated The change in (t) will be caused by only two factors: a) the actual leakage of the workpiece under test; b) the inherent volume change or adsorption effect of the test system (characterized by baseline).

[0067] S53: System error subtraction (baseline subtraction): Subtracts the temperature-compensated pressure data P compensated (t) and the system baseline pressure data P obtained in step S1 baseline Subtracting the corresponding values ​​from (t) yields the final net pressure change curve P, which is purely caused by leakage from the tested workpiece. leak (t):

[0068] P leak (t)=P compensated (t)-P baseline (t);

[0069] S54: Ultrasonic Signal Analysis: Calculates the energy characteristics of the bandpass-filtered ultrasonic signal within a specific time window, such as the root mean square (RMS) value or peak value. This energy value, Vu, is directly related to the intensity of the leakage flow.

[0070] S6: Intelligent Comprehensive Judgment

[0071] Extracting feature values ​​from the processed data:

[0072] Pressure characteristic value Vp: For each "pressure holding plateau segment" of the dynamic pressure curve, the value Vp represents the pressure characteristic value for P. leak (t) data are linearly fitted, and the slope (dp / dt), i.e. leakage rate, is calculated and can be converted to standard milliliters per minute (mL / min).

[0073] Ultrasonic characteristic value Vu: The average energy of the ultrasonic signal within the corresponding pressure holding platform segment.

[0074] Vp and Vu are input into a preset decision model for comprehensive judgment. This model can be:

[0075] Rule base model: Set pressure leakage rate threshold Th p and ultrasonic energy threshold Th u .

[0076] If Vp <Th p And Vu <Th u It was judged as "qualified".

[0077] If Vp>Th p But Vu <Th u It was determined to be a "micro-leak" (possibly molecular flow, which is not sensitive to ultrasound).

[0078] If Vp <Th p But Vu>Th u The error was identified as "interference" (possibly due to environmental noise; we recommend trying again).

[0079] If Vp>Th p And Vu>Th u It was determined to be a "serious leak".

[0080] Fuzzy logic or machine learning models: Weighted or nonlinearly combined Vp and Vu to output a sealing score of 0-100, providing a more refined quality grading.

[0081] The specific details are explained as follows:

[0082] System baseline self-learning refers to establishing reference data for inherent interferences in the test system using standard leak-free workpieces. This can be achieved by averaging multiple repeated tests to eliminate the effects of equipment deformation and sensor drift. Dynamic pressure testing involves generating a pressure fluctuation environment through programmed control, which can be achieved by adjusting gas flow using a high-speed proportional valve to replicate the alternating stress state in the actual test conditions. Multi-parameter synchronous acquisition refers to acquiring multi-sensor data with millisecond-level time accuracy, which can be achieved using a multi-channel synchronous sampling card to ensure the temporal consistency of pressure, temperature, and ultrasonic signals. Real-time temperature compensation corrects pressure data based on thermodynamic equations, which can be achieved by numerical conversion using the ideal gas law to eliminate pressure changes caused by gas temperature fluctuations. Net pressure change data refers to the effective pressure difference after deducting the system baseline, which can be achieved using a differential algorithm to isolate the actual leak signal from system noise. Ultrasonic energy characteristic analysis extracts the acoustic characteristics of leak turbulence, which can be achieved by combining bandpass filtering with root mean square calculation to capture high-frequency sound waves generated by micro-leakage.

[0083] Specifically, before testing, baseline data for the system is established using standard workpieces to eliminate equipment deformation and sensor errors. During testing, a programmed pressure curve simulates actual operating conditions, stimulating potential leakage sources. Pressure, temperature, and ultrasonic data are simultaneously acquired. First, temperature compensation is applied to the pressure data to eliminate thermodynamic interference. Then, the data is compared with the baseline data to remove inherent system noise, yielding the net pressure change reflecting the actual leakage. Simultaneously, the energy characteristics of the ultrasonic signal are analyzed to capture turbulent sound waves generated by micro-leakage. Finally, the leakage rate characteristics and acoustic characteristics are input into the judgment model. A two-parameter cross-validation mechanism distinguishes between actual leakage and external interference, improving the accuracy of the judgment.

[0084] Compared with existing technologies, this method overcomes the limitations of static testing by inducing transient leakage under dynamic pressure environments, thus solving the problem that traditional methods cannot detect leaks under pressure fluctuation conditions. A dual correction mechanism of temperature compensation and baseline subtraction effectively isolates the influence of temperature changes and system noise on the pressure signal. The introduction of ultrasonic signals as an independent criterion and the construction of a multi-dimensional data fusion model overcomes the drawback of single parameters being susceptible to interference, significantly reducing the false positive rate. Synchronous acquisition technology ensures the temporal consistency of multi-source data, providing a reliable foundation for data fusion analysis.

[0085] Through the above technical solutions, this application achieves accurate leak detection of pneumatic joints under dynamic operating conditions, effectively eliminating the impact of temperature fluctuations on test results. Valid data can be obtained without prolonged pressure holding during testing, resolving the contradiction between testing efficiency and detection sensitivity. The multi-parameter fusion judgment mechanism can distinguish between actual leaks and system interference, avoiding the re-inspection costs caused by misjudgments. Ultrasonic signal analysis can assist in locating the leak source, providing data support for subsequent process improvements.

[0086] A periodically changing waveform refers to a signal pattern in which pressure fluctuates regularly over time. This can be achieved by using a waveform generator to control the opening and closing timing of a high-speed pneumatic valve, generating corresponding waveforms by setting different frequency and amplitude parameters. A custom pressure-time curve refers to an arbitrary pressure control curve constructed based on the actual pressure variation patterns of the target application scenario. This can be achieved by importing preset timing data packages into the host computer software, ensuring that the test pressure changes are synchronized with actual operating conditions.

[0087] Specifically, the square wave pattern, by rapidly switching between pressure boosting and depressurization stages, can reproduce the pressure surge caused by frequent opening and closing of solenoid valves in pneumatic systems, thereby detecting sealing failure of joints under instantaneous pressure shocks. The sine wave pattern, through periodic pressure fluctuations, simulates the alternating stress environment generated by the reciprocating motion of a cylinder, effectively inducing micro-leakage caused by elastic deformation of the material. The triangular wave pattern, through linearly increasing and decreasing pressure changes, can evaluate the sealing stability of joints under gradual stress. The custom pressure-time curve, by importing time-series data of pressure changes during actual equipment operation, perfectly matches the test conditions to real-world conditions, thereby detecting leakage phenomena under specific dynamic load modes.

[0088] Compared to existing technologies, traditional static pressure holding methods can only detect the overall leakage under constant pressure and cannot reproduce transient leakage and material fatigue leakage caused by dynamic pressure changes. This solution constructs multiple dynamic pressure curves to cover pressure fluctuation scenarios with different frequencies, amplitudes, and variation patterns in actual working conditions, enabling the testing process to effectively stimulate potential leakage defects under dynamic stress.

[0089] Through the above technical solution, this application can detect transient leaks and material fatigue leaks that traditional static testing methods cannot detect, such as sealing failures during rapid switching of solenoid valves and micro-leakage caused by high-frequency reciprocating motion of cylinders. The application of a customized pressure curve ensures that the test conditions are completely consistent with the actual operating conditions of the target equipment, ensuring that the test results directly reflect the sealing performance of the joint under real-world usage conditions.

[0090] Real-time temperature compensation refers to eliminating the influence of temperature changes on pressure measurements through mathematical calculations. Specifically, it can be achieved using the temperature-pressure relationship from the ideal gas law, which describes the direct proportionality between gas pressure and temperature in a closed system. The fixed reference temperature refers to a pre-set constant temperature value, which can be implemented using a standard laboratory temperature or the design temperature of the testing environment, serving as a benchmark for comparing pressure data. The synchronously acquired real-time temperature value refers to acquiring the temperature data within the testing chamber at the same time as the pressure signal is acquired. This can be achieved using a high-response-speed temperature sensor, ensuring time alignment between temperature and pressure data.

[0091] Specifically, during dynamic pressure testing, the gas temperature within the test chamber fluctuates due to environmental changes or gas flow. According to the ideal gas law, pressure and temperature are positively correlated; an increase in temperature leads to an increase in pressure, and a decrease in temperature leads to a decrease in pressure. By acquiring temperature signals in real time, the raw pressure value at each moment is converted to an equivalent pressure value at a fixed reference temperature using a formula, thus eliminating the interference of temperature changes in the pressure data. For example, when the real-time temperature is higher than the reference temperature, the raw pressure value is corrected downwards; when the real-time temperature is lower than the reference temperature, the raw pressure value is corrected upwards. This compensation method, based on thermodynamic laws, can accurately separate temperature fluctuations from pressure changes caused by actual leakage, thereby ensuring the accuracy of subsequent leakage rate calculations.

[0092] Compared to existing technologies, traditional static pressure holding methods only reduce temperature interference through a constant temperature environment or post-test temperature correction, failing to eliminate the impact of temperature fluctuations in real time during dynamic testing. This solution, however, achieves real-time temperature-independent reconstruction of pressure data by synchronously acquiring temperature signals and performing dynamic compensation based on a physical model, significantly improving the accuracy and anti-interference capability of leak detection.

[0093] Through the above technical solution, this application solves the problem of pressure measurement distortion caused by temperature fluctuations in traditional methods, and can accurately distinguish the pressure difference caused by temperature changes and actual leakage, thereby avoiding misjudgment and improving the reliability of sealing performance testing.

[0094] Bandpass filtering refers to removing low-frequency mechanical vibration noise and high-frequency electromagnetic interference by using a filter with set upper and lower frequency limits. This can be achieved using Butterworth or Chebyshev filters, while retaining the ultrasonic frequency bands relevant to gas leaks. A specific time window refers to a signal analysis interval divided according to the periodic characteristics of the dynamic pressure change curve. This can be achieved by using a time period during the pressure rise phase or a fixed duration during the pressure stabilization phase, ensuring that the energy characteristics are synchronized with the pressure change conditions. The root mean square (RMS) value is calculated by taking the square root of the average value of the filtered ultrasonic signal. This can be implemented using a sliding window algorithm, reflecting the average energy level of the leaking sound wave within the time window. The peak value is the maximum amplitude of the filtered ultrasonic signal, which can be extracted using an extremum detection algorithm to capture sudden energy pulses generated by intermittent leaks.

[0095] Specifically, the raw signal captured by the ultrasonic sensor includes environmental noise and inherent equipment vibration interference. After filtering out non-leakage-related frequency components through a bandpass filter, the high-frequency acoustic signal unique to gas leaks is retained. During dynamic pressure testing, time windows are divided according to the phase of the pressure change curve; for example, the first 200 milliseconds window is selected during the pressure rise phase, and the middle 500 milliseconds window is selected during the pressure holding phase. Within each time window, the root mean square value of the filtered ultrasonic signal is calculated to obtain the average energy characteristic representing a continuous leak; simultaneously, the maximum peak value of the signal is detected to identify transient leak events. Synchronous analysis of pressure parameters and ultrasonic energy characteristics can distinguish between pressure drops caused by actual leaks and pressure changes caused by temperature drift. For example, when pressure changes are accompanied by a significant increase in ultrasonic energy characteristics, it is determined to be a true leak, while pressure changes without changes in energy characteristics are determined to be temperature interference.

[0096] Compared to existing technologies, traditional sealing test methods rely solely on single-parameter analysis of pressure sensors, failing to distinguish between actual leaks and pressure changes caused by temperature fluctuations or cavity deformation. This solution introduces energy characteristic analysis of ultrasonic signals, combined with bandpass filtering and time-domain energy calculation, to form a pressure-sound dual-dimensional detection mechanism. Existing technologies using fixed-frequency filtering cannot adapt to the frequency domain characteristics of different leakage modes. This solution, however, divides time windows based on the phase of the dynamic pressure curve, enabling time-series correlation analysis between leakage characteristics and pressure conditions, effectively improving the targeting of feature extraction.

[0097] Through the above technical solution, this application solves the problem of misjudgment caused by a single pressure parameter. By utilizing the synergistic analysis of ultrasonic energy characteristics and pressure parameters, it accurately distinguishes between pressure changes caused by actual leakage and non-leakage factors. Bandpass filtering eliminates environmental noise interference, time-domain energy calculation quantifies leakage intensity, and combined with the time window division of the dynamic pressure test phase, it achieves accurate extraction of leakage characteristics. The fusion judgment mechanism of pressure parameters and ultrasonic energy effectively reduces false positive alarms caused by temperature drift or system deformation, improving the reliability of sealing detection.

[0098] The leakage rate characteristic value refers to a quantitative index of gas leakage rate calculated from net pressure change data. Specifically, it can be implemented using the pressure decay rate or leakage volume flow rate formula, reflecting the overall sealing performance of the tested joint under dynamic pressure. The ultrasonic energy characteristic value refers to the acoustic emission energy index extracted after bandpass filtering of the ultrasonic signal. Specifically, it can be implemented using the root mean square value or peak value calculation method, characterizing the intensity of high-frequency sound waves generated by gas turbulence when leakage occurs. The rule base's judgment logic refers to classification rules established based on the correlation between leakage rate and ultrasonic energy. Specifically, it can be implemented using threshold comparison and state combination judgment, eliminating the risk of misjudgment based on a single parameter through multi-parameter cross-validation.

[0099] Specifically, during the dynamic pressure test, the synchronously acquired pressure signal is compensated for temperature to generate net pressure change data, and the leakage rate characteristic value is calculated using a leakage rate model. Simultaneously, an ultrasonic sensor captures high-frequency sound waves, which are filtered and then have their energy characteristic value extracted. When the leakage rate characteristic value does not exceed a first threshold and the ultrasonic energy characteristic value does not exceed a second threshold, it is considered no leakage. When only the leakage rate characteristic value exceeds the threshold, it indicates a slow leak without significant turbulence, and is considered a micro-leak. When only the ultrasonic energy characteristic value exceeds the threshold, it indicates external vibration or electromagnetic noise interference, and is considered an abnormal test environment. When both exceed the threshold, it confirms high-speed gas leakage accompanied by turbulent acoustic emission, and is considered a serious leak.

[0100] Compared to existing technologies, traditional static pressure holding methods rely solely on pressure as a single parameter, failing to distinguish between temperature drift and actual leakage. This solution, however, utilizes a dual-parameter fusion analysis of pressure and ultrasound to establish a correlation rule between the physical phenomena of leakage and acoustic emission characteristics. In existing technologies using a single threshold for judgment, pressure fluctuations and temperature changes can easily lead to false positive alarms. This solution, by introducing an ultrasonic energy threshold, can effectively identify misjudgments caused by environmental interference.

[0101] Through the above technical solution, this application solves the problem of misjudgment caused by single parameters in traditional testing, can clearly distinguish between real leakage and system noise or environmental interference, and can identify different defect types of micro-leakage and severe leakage, thus improving the accuracy and reliability of sealing performance judgment.

[0102] Among them, the algorithm model based on machine learning or fuzzy logic refers to a mathematical model that fuses multi-source sensor signals through nonlinear mathematical relationships. Specifically, it can be implemented using neural networks, support vector machines, or fuzzy inference systems, establishing a mapping relationship between input features and output scores through training or rule definition. The leakage rate characteristic value is a quantitative index of the leakage rate calculated from pressure change data after temperature compensation and baseline subtraction. It can be implemented through integral calculations or fitting algorithms, reflecting the overall leakage degree of the tested joint. The ultrasonic energy characteristic value refers to the energy intensity of the acoustic signal in a specific frequency band after bandpass filtering. It can be implemented using root mean square (RMS) or peak value detection algorithms, characterizing the intensity of turbulence or cavitation effects caused by leakage. The continuous sealing quality score is a comprehensive evaluation result of sealing performance expressed numerically, specifically implemented using a 0-100 score system or probability values, used to quantitatively evaluate the sealing status of the joint.

[0103] Specifically, during dynamic pressure testing, the synchronously acquired pressure signals are processed through temperature compensation and baseline subtraction to generate net pressure change data. Based on this data, a leakage rate characteristic value is calculated. Ultrasonic signals undergo bandpass filtering to extract energy characteristic values ​​for specific frequency bands. These two characteristic values ​​are input into a pre-trained machine learning model or used for inference calculations via a fuzzy logic system. The machine learning model establishes a mapping function from input features to output score by analyzing the complex nonlinear relationship between feature combinations and leakage levels in historical test data. The fuzzy logic system, by defining a membership function between leakage rate and ultrasonic energy and combining it with expert experience rules, performs comprehensive inference. Finally, a continuously quantified score is output; for example, a score below 60 is considered unqualified, 60-80 is a critical state requiring re-inspection, and above 80 is considered qualified. This scoring system can distinguish subtle leakage differences; for example, at the same leakage rate, a joint with higher ultrasonic energy may receive a lower score due to the presence of local turbulence.

[0104] Compared to existing technologies, traditional static pressure holding methods rely solely on a single pressure parameter for binary compliance determination, failing to address the coupling effect of temperature interference and dynamic leakage. While rule-based methods using fixed thresholds introduce multiple parameters, their linear logic struggles to capture the phase correlation between pressure fluctuations and acoustic signals in transient leaks. However, machine learning or fuzzy logic algorithms can automatically learn the nonlinear interaction between pressure and ultrasonic signals under complex operating conditions. For example, during periods of rapid pressure change, the ultrasonic signal may exhibit transient peaks while the pressure change is insignificant. In such cases, the algorithm can identify a dynamic leakage pattern and adjust the scoring weights accordingly.

[0105] Through the above technical solution, this application solves the problem of misjudgment caused by the single parameter interpretation in traditional methods, and reduces the interference of temperature fluctuations and dynamic pressure changes on test results. It achieves the differentiation and identification of minute leaks and interference signals, such as effectively distinguishing pressure fluctuations caused by material elastic deformation from actual leaks. A continuous scoring mechanism provides a more refined quality assessment, supporting graded processing decisions for critical state joints and avoiding the problems of over- or under-quality assessment caused by traditional binary judgment.

[0106] Example 2:

[0107] This invention also proposes a pneumatic joint sealing performance testing system, comprising:

[0108] Air circuit and tooling module: Provides a clean and stable air source, including a precision filter, pressure regulating valve, air tank, high-speed pneumatic control valve (proportional valve or on / off valve assembly), a precisely calibrated test chamber, and tooling mechanism for quick clamping of the test connector.

[0109] Sensing module:

[0110] High-precision pressure sensor: installed on the test chamber, the measurement range covers the test pressure, the accuracy is better than ±0.1%FS, and the response time is <1ms.

[0111] High-response temperature sensor: PT100 or precision NTC, deeply inserted into the test chamber, directly in contact with the gas, with an accuracy of ±0.1℃, used for real-time temperature monitoring.

[0112] Ultrasonic sensor: Frequency response range covers 20kHz-200kHz, with a certain directionality, to be aimed at the key sealing parts of the connector being tested.

[0113] Control and data acquisition module: High-speed synchronous data acquisition card (DAQ): at least 3 analog input channels, 16-bit or higher resolution, sampling rate ≥1kS / s, supporting multi-channel synchronous sampling.

[0114] Main controller: Adopting an industrial PC (IPC) or "PLC + host computer" architecture, it is responsible for controlling the action of pneumatic valves to generate dynamic pressure spectrum and receiving sensor data uploaded by DAQ card.

[0115] Software module: Runs on the main controller and includes:

[0116] Human-computer interface (HMI) is used for parameter setting, process control, and data display.

[0117] The core algorithm program performs signal filtering, temperature compensation, baseline subtraction, feature extraction, and intelligent decision-making.

[0118] The database function is used to store test parameters, raw data, compensated data, and judgment results, and can generate and export test reports.

[0119] The following is a detailed description of this embodiment:

[0120] This application further proposes a pneumatic joint sealing performance testing system, including a pneumatic circuit and tooling module, a sensing module, and a control and data acquisition module. The pneumatic circuit and tooling module includes a test chamber and a high-speed pneumatic control valve for connecting the joint under test and generating a dynamic pressure change curve; the sensing module includes a pressure sensor, a temperature sensor, and an ultrasonic sensor for acquiring pressure, temperature, and ultrasonic signals within the test chamber; the control and data acquisition module includes a high-speed synchronous data acquisition card and a main controller for synchronously acquiring sensor signals and performing data analysis and judgment.

[0121] The test cavity refers to the cavity structure used to accommodate the pneumatic connector under test and form a sealed test space. Specifically, it can be implemented using a modular design with replaceable volumes to adapt to the testing requirements of connectors of different sizes. The high-speed pneumatic control valve is an actuator capable of rapidly adjusting gas flow to generate a dynamic pressure curve. Specifically, it can be implemented using a high-speed proportional valve or a combination of dual high-speed switching solenoid valves. The valve opening is controlled by a preset program to simulate pressure fluctuations in actual working conditions. The ultrasonic sensor is a detection device capable of capturing high-frequency sound wave signals. Specifically, it can be implemented using a piezoelectric ceramic transducer. By receiving the high-frequency sound wave signals generated by leaking airflow, it can locate and detect the leak point. The high-speed synchronous data acquisition card is a signal acquisition device with multi-channel synchronous sampling capability. Specifically, it can be implemented using an FPGA architecture acquisition card to ensure the timing consistency of pressure, temperature, and ultrasonic signals and eliminate errors caused by signal acquisition delays.

[0122] Specifically, the testing system generates periodic pressure change curves through a high-speed pneumatic control valve in the gas path and tooling module, causing the gas pressure inside the test chamber to dynamically fluctuate according to a preset pattern, simulating the alternating stress environment of the pneumatic system during actual operation. Pressure sensors in the sensing module monitor chamber pressure changes in real time, temperature sensors synchronously collect gas temperature data, and ultrasonic sensors capture high-frequency sound waves generated by leakage. The control and data acquisition module simultaneously records pressure, temperature, and ultrasonic signals through a high-speed synchronous data acquisition card. The main controller performs temperature compensation calculations on the pressure data to eliminate pressure change interference caused by temperature fluctuations and compares the compensated net pressure change data with the system baseline. Simultaneously, the ultrasonic signal is processed by bandpass filtering to extract energy characteristic values, which are then combined with pressure leakage rate characteristic values ​​and input into a comprehensive judgment model to achieve multi-parameter fusion analysis.

[0123] Compared to existing technologies, traditional static pressure holding test systems rely solely on a single pressure sensor for static pressure monitoring, failing to eliminate temperature interference and exhibiting low testing efficiency. This solution, through dynamic pressure excitation and simultaneous acquisition of multiple parameters, not only simulates the alternating pressure environment under actual operating conditions but also eliminates the impact of ambient temperature fluctuations on pressure data through a temperature compensation algorithm. Existing technologies do not employ ultrasonic sensors, making it impossible to distinguish between actual leaks and system noise; however, this solution achieves precise leak location through ultrasonic energy characteristic analysis.

[0124] Through the above technical solution, this application solves the problem of misjudgment caused by temperature interference in traditional testing methods. It detects transient leaks through dynamic pressure testing and reduces the misjudgment rate of single parameters by utilizing multi-sensor data fusion. The system improves detection sensitivity while maintaining testing efficiency through high-speed synchronous acquisition and real-time temperature compensation. The introduction of ultrasonic signals further enables the location and identification of the leak source.

[0125] Among them, the high-speed proportional valve refers to a pneumatic control valve that continuously adjusts the valve opening through analog signals. Specifically, it can be implemented using a servo proportional valve with valve core position feedback, which achieves continuous and precise adjustment of the pressure curve through closed-loop control. The combination valve refers to a valve group consisting of two independently controlled high-speed switching solenoid valves that control intake and exhaust. Specifically, it can be implemented using a high-speed switching valve controlled by a PWM pulse width modulation signal, which achieves equivalent flow control by alternately opening and closing the two valves.

[0126] Specifically, in test scenarios requiring the generation of continuously varying pressure curves such as sine or triangular waves, the high-speed proportional valve's spool position feedback mechanism can adjust the opening in real time, ensuring that pressure changes within the test chamber precisely follow the preset curve. In test scenarios requiring the generation of square or step pressure curves, the combination valve achieves instantaneous response to sudden pressure changes by controlling the rapid switching between the inlet and outlet valves. The choice between the two valve types depends on the characteristics of the test waveform and cost requirements; the high-speed proportional valve is suitable for high-precision laboratory environments, while the combination valve is suitable for large-scale testing in industrial settings.

[0127] Compared to existing technologies, traditional pneumatic testing systems typically employ a single type of control valve, such as a standard solenoid valve or a proportional valve with low response speed, resulting in a trade-off between the accuracy of dynamic pressure curve generation and response speed. This solution, through the parallel arrangement of two valve architectures, satisfies both the precise control requirements of continuous pressure curves and the rapid generation of step pressure changes, while simultaneously reducing the hardware cost of high-frequency response valve assemblies.

[0128] Through the above technical solution, this application effectively solves the contradiction between the response speed of the control valve and the accuracy of the pressure curve in dynamic pressure testing. It can flexibly select the valve group structure according to the test waveform type and adapt to the economic requirements of different test scenarios while ensuring the accuracy of dynamic pressure generation.

[0129] The ability to precisely measure and replace the test chamber volume means that the volume of the test chamber can be changed according to the diameter of the pneumatic connector being tested. This can be achieved using modular chamber assemblies in conjunction with a volume calibration device. Each chamber assembly must be calibrated before installation, and the calibration value must be recorded. This design avoids problems such as the dilution of minute leakage signals due to an overly large chamber or excessive pressure fluctuations due to an underly small chamber by matching the optimal chamber volume.

[0130] The use of a PT100 platinum resistance thermometer or an NTC thermistor as the temperature sensor indicates the adoption of a sensor type with high-precision temperature measurement characteristics. This is achieved by directly inserting the sensor probe into the gas environment inside the test chamber. This design eliminates the thermal conduction hysteresis effect of traditional external temperature sensors, ensuring real-time capture of the true temperature changes of the gas.

[0131] Specifically, when testing pneumatic fittings of different diameters, the volume of the test chamber is matched to the diameter of the fitting by replacing the pre-calibrated cavity assembly. For example, a larger cavity assembly is used for fittings with larger diameters to reduce pressure fluctuations; a smaller cavity assembly is used for fittings with smaller diameters to improve leak detection sensitivity. Simultaneously, the temperature sensor probe extends directly into the test chamber, ensuring full contact with the gas and monitoring gas temperature changes in real time. During dynamic pressure testing, the data collected by the temperature sensor is used to compensate for pressure signals in real time, thereby eliminating the influence of temperature fluctuations on the pressure measurement results.

[0132] In some specific implementations, the cavity assembly can be connected to the testing system via a quick-release interface, and the outer surface of each cavity assembly is marked with a precisely measured volume value. The temperature sensor probe can be installed in a pre-drilled hole in the inner wall of the cavity assembly, and the airtightness of the testing chamber is ensured by a sealing structure.

[0133] Compared to existing technologies, traditional testing systems typically use a fixed-volume test cavity, leading to significant differences in testing sensitivity for connectors of different sizes. Furthermore, temperature sensors are often installed externally to the cavity or within the piping, failing to accurately reflect the actual temperature of the gas inside. This solution, through the coordinated design of replaceable cavity components and a built-in temperature sensor, achieves universal compatibility with different connector specifications while improving the accuracy of temperature compensation.

[0134] Through the above technical solution, this application solves the measurement error problem caused by the inability of the test cavity to adapt to connectors of different sizes, and also eliminates compensation inaccuracies caused by improper installation of the temperature sensor. The replaceable volume design of the test cavity ensures the consistency of test sensitivity for connectors of different diameters, while the built-in installation of the temperature sensor improves the real-time performance and accuracy of temperature data, thereby enhancing the reliability of the leakage rate calculation results.

[0135] As used in the specification and claims, certain terms refer to specific components. Those skilled in the art will understand that hardware manufacturers may use different names to refer to the same component. This specification and claims do not distinguish components based on differences in name, but rather on differences in function. The term "comprising" throughout the specification and claims is an open-ended term and should be interpreted as "comprising but not limited to." "Approximately" means that within an acceptable margin of error, those skilled in the art can solve the technical problem and substantially achieve the technical effect within a certain margin of error.

[0136] The foregoing description illustrates and describes several preferred embodiments of the present invention. However, as previously stated, it should be understood that the present invention is not limited to the forms disclosed herein and should not be construed as excluding other embodiments. It can be used in various other combinations, modifications, and environments, and can be altered within the scope of the inventive concept described herein through the foregoing teachings or techniques or knowledge in related fields. Any modifications and variations made by those skilled in the art that do not depart from the spirit and scope of the present invention should be within the protection scope of the appended claims.

Claims

1. A method for testing the sealing performance of a pneumatic joint, characterized in that, Includes the following steps: System baseline self-learning steps: Perform dynamic pressure testing using a standard leak-free workpiece to acquire and store the pressure baseline data of the test system itself; Dynamic pressure test procedure: Apply a preset dynamic pressure change curve to the test chamber containing the pneumatic connector under test. Multi-parameter synchronous acquisition step: During the dynamic pressure test, the pressure signal, temperature signal and ultrasonic signal in the test chamber are acquired synchronously and at high speed. Data fusion analysis steps: Based on the temperature signal, real-time temperature compensation is performed on the pressure signal. This real-time temperature compensation, according to the ideal gas law, compensates the original pressure value Praw collected at each moment to a fixed reference temperature T. ref The calculation formula is: Where T(t) is the real-time temperature value acquired synchronously, the pressure data after temperature compensation is obtained; the pressure data after temperature compensation is compared and subtracted from the pressure baseline data to obtain the net pressure change data; at the same time, the energy characteristics of the ultrasonic signal are analyzed. After bandpass filtering the acquired raw ultrasonic signal, the root mean square value or peak value within a specific time window is calculated as the characteristic value. Intelligent comprehensive judgment step: Based on the leakage rate characteristic value calculated from the net pressure change data and the energy characteristic value of the ultrasonic signal, an evaluation result of the sealing performance of the pneumatic joint is output through a comprehensive judgment model. The comprehensive judgment model in the intelligent comprehensive judgment step is a judgment logic based on a rule base. If the leakage rate characteristic value is less than the first threshold and the ultrasonic energy characteristic value is less than the second threshold, it is considered qualified. If the leakage rate characteristic value is greater than the first threshold but the ultrasonic energy characteristic value is less than the second threshold, it is determined to be a micro-leakage; If the leakage rate characteristic value is less than the first threshold but the ultrasonic energy characteristic value is greater than the second threshold, it is determined that there is interference. If the leakage rate characteristic value is greater than the first threshold and the ultrasonic energy characteristic value is greater than the second threshold, it is determined to be a serious leakage.

2. The method according to claim 1, characterized in that, The dynamic pressure change curve is a periodically changing waveform, including square wave, sine wave, triangle wave, or a custom pressure-time curve simulating actual working conditions.

3. The method according to claim 1, characterized in that, The comprehensive judgment model in the intelligent comprehensive judgment step is an algorithm model based on machine learning or fuzzy logic, which takes leakage rate feature value and ultrasonic energy feature value as input and outputs a continuous sealing quality score.

4. A pneumatic joint sealing performance testing system for implementing the method according to any one of claims 1-3, characterized in that, include: The air circuit and tooling module includes a test chamber and a high-speed pneumatic control valve, which are used to connect the connector under test and generate the dynamic pressure change curve. The sensing module includes a pressure sensor for detecting the pressure inside the test chamber, a temperature sensor for detecting the temperature of the gas inside the test chamber, and an ultrasonic sensor for capturing high-frequency sound waves generated by the leak. The control and data acquisition module includes a high-speed synchronous data acquisition card and a main controller. The data acquisition card is connected to the sensing module for synchronously acquiring sensor signals. The main controller is connected to the data acquisition card and the pneumatic control valve for controlling the generation of pressure curves and executing the data fusion analysis and intelligent comprehensive judgment steps.

5. The system according to claim 4, characterized in that, The high-speed pneumatic control valve is a high-speed proportional valve or a combination valve consisting of two high-speed switching solenoid valves.

6. The system according to claim 4, characterized in that, The volume of the test chamber is precisely measurable and replaceable to accommodate pneumatic connectors of different diameters; the temperature sensor is a PT100 platinum resistance thermometer or an NTC thermistor, and its probe extends directly into the test chamber to contact the gas.

7. A computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the steps of the method as described in any one of claims 1 to 3.

Images