Multifunctional test platform for respiratory valve and automated detection method

Abstract

The application discloses a multifunctional test platform and an automatic detection method for a breathing valve, and relates to the field of automatic control, which comprises the following steps: injecting a test medium and calibrating a reference pressure parameter, performing positive and negative pressure sealing detection and pressure change cycle test under different pressure grades, recording pressure change rate and leakage to form a data set, adjusting medium flow to measure ventilation resistance, generating a flow-resistance characteristic curve, repeatedly detecting sealing performance and resistance change under simulated actual working conditions, identifying abnormal data and correcting by using machine learning and comparing with a preset threshold, integrating sealing performance, ventilation resistance, flow characteristics and durability data, outputting qualified, unqualified and re-inspection determination results, discharging the medium after detection and completing equipment self-inspection. The application has the advantages that multi-dimensional detection and complex working condition simulation are covered, intelligent abnormal processing is combined to guarantee data accuracy, standard determination results are efficiently output, and the detection comprehensiveness, reliability and standardization are considered.

Description

Technical Field

[0001] This invention relates to the field of automation control, and in particular to a multifunctional test platform for breather valves and an automated testing method. Background Technology

[0002] Breathing valves are critical safety devices in storage tank systems used in petroleum, chemical, and pharmaceutical industries. Their sealing performance, ventilation capabilities, and opening / closing pressure accuracy directly impact production safety and environmental compliance. Traditional testing methods rely heavily on manual operation and decentralized equipment, resulting in low efficiency, poor data consistency, and significant human error, failing to meet the demands of modern industry for high-precision, high-frequency testing. With the widespread adoption of intelligent manufacturing and automation technologies, the industry urgently needs integrated and intelligent testing solutions.

[0003] The core disadvantages of current breathing valve testing methods lie in their lack of automation, comprehensiveness, and intelligence. They largely rely on manual operation, such as manually hoisting and fixing the breathing valve, and manually adjusting the valve and pressure. Low automation leads to low testing efficiency, and manual pressure adjustment is prone to overpressure and recording errors, introducing human error. Integrated testing is insufficient; various performance tests are often conducted separately, and frequent disassembly and reassembly affect the accuracy of results. Some methods even fail to cover negative pressure testing, posing safety hazards. The ability to simulate operating conditions is weak; many methods cannot automatically adapt to real-world operating temperature and humidity, or have limited adjustment ranges, making it difficult to reproduce complex operating conditions such as temperature cycles and pressure pulses, resulting in test results that are out of touch with actual applications. Furthermore, most methods use single-sensor testing, lacking data optimization and intelligent anomaly identification capabilities, making it impossible to trace the cause of anomalies. They also lack a complete system self-testing process, making it difficult to guarantee the accuracy of batch testing and the long-term stability of the equipment. Summary of the Invention

[0004] To improve existing systems and methods, a multi-functional testing platform and automated testing method for breathing valves are provided. This method achieves fully automated operation, covers multi-dimensional testing and complex working condition simulation, combines intelligent anomaly handling to ensure data accuracy, efficiently outputs standardized judgment results, and takes into account the comprehensiveness, reliability and standardization of testing.

[0005] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0006] Automated testing methods for breathing valves include:

[0007] Inject a pre-set type of test medium into the breather valve test line, adjust the pressure in the line to the sealing performance test reference pressure, and collect the pressure change data in the line for reference parameter calibration.

[0008] Based on the calibrated baseline parameters, the pressure in the test pipeline is adjusted to different test levels in sequence to perform positive pressure sealing test, negative pressure sealing test and variable pressure cyclic sealing test. The pressure change rate and leakage amount at each test level are collected in real time to form a sealing performance test dataset.

[0009] Based on the preset flow gradient, the flow rate of the test medium is adjusted. The pressure difference between the inlet and outlet of the breathing valve under different flow conditions is collected by the pressure sensing component. The ventilation resistance value corresponding to each flow is obtained. Based on the resistance value and flow data, the flow-resistance characteristic curve of the breathing valve is generated.

[0010] Based on the complex operating conditions data of the breather valve in actual application, the simulation of temperature cycle change, pressure pulse impact and medium corrosion environment is carried out. Through the automatic repeated sealing performance test and ventilation resistance test process, the sealing performance and ventilation resistance change data are collected.

[0011] Based on the preset normal parameter threshold range, combined with machine learning algorithms, anomalies are identified in the collected changing data, the causes of the detected abnormal data are determined, and compensation and correction are performed.

[0012] The system integrates the effective data from the testing process to generate test results that include sealing performance, ventilation resistance, flow characteristics, and durability. It also judges each indicator and outputs three judgment results: qualified, unqualified, and pending retest.

[0013] After the test is completed, the test medium in the test pipeline is discharged, the sample is reset, and the equipment self-test process is started to test the performance status of each component.

[0014] Preferably, the step of injecting a preset type of test medium into the breather valve detection pipeline, adjusting the pressure in the pipeline to the sealing performance testing reference pressure, and collecting pressure change data in the pipeline for reference parameter calibration specifically includes:

[0015] Inject the medium into the test pipeline according to the preset medium type, and use a gradient pressure increase and decrease method to adjust the pressure in the pipeline to the preset sealing performance test reference pressure;

[0016] Once the pressure reaches the baseline value, the pressure data in the pipeline is collected through the pressure sensing component, the average and maximum pressure fluctuations are calculated, and the baseline parameter calibration is completed by comparing them with the preset allowable fluctuation range.

[0017] If the pressure fluctuation exceeds the allowable range, the pipeline sealing self-test process is triggered, the two ends of the breather valve are closed, the pipeline is subjected to segmented pressure testing, and the leak point is located.

[0018] Preferably, the step of sequentially adjusting the pressure in the detection pipeline to different detection levels based on the calibrated reference parameters to perform positive pressure sealing tests, negative pressure sealing tests, and variable pressure cyclic sealing tests, and collecting the pressure change rate and leakage amount at each detection level in real time to form a sealing performance test dataset specifically includes:

[0019] Based on the calibrated baseline pressure parameters, specific parameter thresholds are generated for three detection levels: positive pressure, negative pressure, and variable pressure cycle.

[0020] The pressure in the test pipeline is increased to the preset positive pressure sealing test range, stabilized for 60 seconds, and then the pressure is uniformly reduced to the negative pressure test threshold. After stabilization, it enters the pressure transformation cycle stage and cycles 10 times according to the positive pressure, normal pressure and negative pressure cycle.

[0021] Ten sets of pipeline pressure data are collected by a pressure sensing component to calculate the real-time pressure change rate, and the leakage amount at each detection stage is recorded by a pipeline end leakage detection device.

[0022] The pressure change rate and leakage parameters are integrated into a sealing performance test dataset after noise reduction processing.

[0023] Preferably, the step of adjusting the flow rate of the test medium based on a preset flow gradient, collecting the pressure difference between the inlet and outlet of the breathing valve under different flow conditions through a pressure sensing component, obtaining the ventilation resistance value corresponding to each flow rate, and generating the breathing valve flow-resistance characteristic curve based on the resistance value and flow data specifically includes:

[0024] Based on the model and size data of the breathing valve to be tested, the test condition parameters are matched with the preset test standard library. The test condition parameters include the medium type, initial pressure, temperature range, test duration and number of cycles.

[0025] A flow regulation scheme is generated based on the matched operating parameters, including the flow range, residence time and regulation rate of each gradient.

[0026] Based on the preset scheme, the flow rate of the test medium is adjusted. During the stabilization stage of each flow gradient, pressure data is collected synchronously through the pressure sensing components at the inlet and outlet of the breather valve, and the pressure difference between the inlet and outlet is calculated.

[0027] Based on the collected pressure difference and corresponding flow rate, the ventilation resistance under each flow rate condition is calculated. Based on real-time monitoring of the pressure change inflection point, the opening pressure, fully opening pressure and closing pressure of the breathing valve are obtained.

[0028] The flow rate, resistance, and characteristic parameters of each gradient are correlated and integrated to generate a flow rate-resistance characteristic curve.

[0029] Preferably, the simulation of temperature cycling, pressure pulse impact, and media corrosion environment based on complex operating condition data in actual applications of the breather valve, and the collection of sealing performance and ventilation resistance change data through an automatic repeated sealing performance test and ventilation resistance test process, specifically includes:

[0030] Based on various working conditions in the actual application of breather valves, a multi-working-condition collaborative simulation scheme is generated. Through the synergistic effect of temperature gradient rise and fall, pressure pulse impact and continuous wetting by corrosive media, the actual values ​​of parameters for each working condition are obtained.

[0031] The detection process is triggered based on a preset time interval, and the sealing performance test process and the ventilation resistance test process are repeated.

[0032] Throughout the entire test period, high-frequency data on sealing performance and ventilation resistance changes were collected and stored in a time-stamped manner.

[0033] Preferably, the step of identifying anomalies in the collected change data based on a preset normal parameter threshold range, combined with machine learning algorithms, and determining the cause of the detected abnormal data and performing compensation and correction specifically includes:

[0034] The data on sealing performance and ventilation resistance changes are denoised and smoothed, invalid collection points are removed, and a standardized analysis dataset is formed.

[0035] A machine learning classification algorithm based on random forest is used to perform time series analysis on the standardized analysis dataset. By comparing it with threshold ranges and historical qualified data samples, abnormal data points such as pressure mutations, excessive leakage, and abnormal resistance fluctuations are identified.

[0036] For the identified abnormal data, the corresponding timestamps of the detection conditions, equipment operating status, and pipeline sealing status are traced back.

[0037] If the error is determined to be due to equipment error, the parameter calibration and correction process is initiated, and a compensation coefficient is calculated based on the error difference for real-time compensation. If the error is determined to be due to a defect in the breathing valve itself, the defect type and corresponding detection node are marked.

[0038] Preferably, the process of integrating effective data from the testing process to generate test results including sealing performance, ventilation resistance, flow characteristics, and durability, and judging each indicator and outputting three judgment results: qualified, unqualified, and pending re-inspection, specifically includes:

[0039] Obtain valid detection data after anomaly correction, and integrate time-series data and feature parameters under the same detection dimension according to the detection stage to form a standardized data matrix;

[0040] Extract the core test indicators of the standardized data matrix, including the maximum leakage and pressure stabilization time in sealing performance, the average gradient resistance in ventilation resistance, the opening and closing pressure threshold and flow-resistance curve characteristic parameters in flow characteristics, and the cycle tolerance number and parameter decay rate in durability.

[0041] Based on the preset pass / fail criteria, each extracted indicator is compared with the standard threshold one by one, and the judgment is automatically made by adopting the judgment rule that all items pass the test, a single item exceeds the standard, and edge fluctuations indicate that the test needs to be re-inspected.

[0042] Preferably, after the test is completed, the test medium in the test pipeline is discharged, the sample is reset, and the equipment self-test process is started to test the performance status of each component. This specifically includes:

[0043] After the test is completed, open the discharge valve to discharge the test medium in the test pipeline, drive the clamping assembly to release the breather valve, and drive the positioning mechanism to reset to the initial standby position.

[0044] The equipment self-test program is started, and the performance of key components of the testing platform is tested in sequence, including: the action response speed test of the sensing components and the adjustment module, and the static sealing test of the sealing components.

[0045] After all components have completed their self-inspection, the self-inspection data is summarized and a self-inspection report is generated.

[0046] Furthermore, a multifunctional test platform for breathing valves is proposed, including:

[0047] Test medium control module: Injects test medium according to preset type, adjusts pipeline pressure to the detection reference value, and realizes gradient pressure increase and decrease control;

[0048] Sealing performance testing module: Performs positive pressure, negative pressure and variable pressure cycle sealing tests, collects pressure change rate and leakage data in real time, and generates standardized sealing performance dataset;

[0049] Flow-resistance characteristic detection module: Based on the flow gradient, the medium flow rate is adjusted, the pressure difference between the inlet and outlet of the breathing valve is measured, the ventilation resistance is calculated, and the flow-resistance characteristic curve is generated;

[0050] Complex operating condition simulation module: Simulates actual operating conditions through temperature cycling, pressure pulses, and corrosive media, triggering repeated tests to collect durability data;

[0051] Intelligent analysis and anomaly handling module: uses machine learning algorithms to identify abnormal data, correlates with operating parameters to determine the cause of the anomaly, and realizes real-time compensation or defect marking;

[0052] Result determination module: integrates test data, compares it with threshold standards, outputs the pass / fail determination result, and generates a test report;

[0053] Equipment self-test module: After completing the test, the medium is emptied, the sample is reset, and the sensors and sealing components are self-tested;

[0054] Processor: The processor is used to handle the calculation process of each formula and the construction calculation process of each model.

[0055] Compared with the prior art, the advantages of the present invention are:

[0056] Through fully automated operation, the system covers multiple dimensions of testing, including sealing performance, flow-resistance characteristics, and durability under complex operating conditions. This eliminates the need for extensive manual intervention, significantly improving testing efficiency and reducing human error. Utilizing benchmark parameter calibration and machine learning algorithms, it achieves accurate identification, cause tracing, and compensation correction of abnormal data, ensuring the reliability of test data. By simulating real-world complex operating conditions such as temperature cycles and pressure pulses, the test results are made more closely aligned with actual application scenarios. Simultaneously, it integrates all effective data from the entire process to form standardized reports, clearly defining the pass / fail determination results. Coupled with post-test media discharge, sample resetting, and equipment self-inspection procedures, it ensures testing standardization and equipment stability, providing comprehensive and scientific technical support for the quality assessment of breather valves. Attached Figure Description

[0057] Figure 1 This is a schematic diagram of the method proposed in this invention;

[0058] Figure 2 This is a schematic diagram of the calibration of the reference parameters proposed in this invention;

[0059] Figure 3 This is a schematic diagram of the sealing performance testing proposed in this invention;

[0060] Figure 4 This is a schematic diagram of the ventilation resistance detection proposed in this invention;

[0061] Figure 5 This is a schematic diagram of the multi-condition simulation proposed in this invention;

[0062] Figure 6 This is a schematic diagram of the anomaly identification and data calibration proposed in this invention;

[0063] Figure 7 This is a schematic diagram illustrating the integration and judgment of detection results proposed in this invention;

[0064] Figure 8 This is a schematic diagram of the sample reset and device self-test proposed in this invention. Detailed Implementation

[0065] The following description is intended to disclose the invention and enable those skilled in the art to implement it. The preferred embodiments described below are merely examples, and other obvious variations will occur to those skilled in the art.

[0066] A multi-functional test platform for breather valves includes:

[0067] Test medium control module: Injects test medium according to preset type, adjusts pipeline pressure to the detection reference value, and realizes gradient pressure increase and decrease control;

[0068] Sealing performance testing module: Performs positive pressure, negative pressure and variable pressure cycle sealing tests, collects pressure change rate and leakage data in real time, and generates standardized sealing performance dataset;

[0069] Flow-resistance characteristic detection module: Based on the flow gradient, the medium flow rate is adjusted, the pressure difference between the inlet and outlet of the breathing valve is measured, the ventilation resistance is calculated, and the flow-resistance characteristic curve is generated;

[0070] Complex operating condition simulation module: Simulates actual operating conditions through temperature cycling, pressure pulses, and corrosive media, triggering repeated tests to collect durability data;

[0071] Intelligent analysis and anomaly handling module: uses machine learning algorithms to identify abnormal data, correlates with operating parameters to determine the cause of the anomaly, and realizes real-time compensation or defect marking;

[0072] Result determination module: integrates test data, compares it with threshold standards, outputs the pass / fail determination result, and generates a test report;

[0073] Equipment self-test module: After completing the test, the medium is emptied, the sample is reset, and the sensors and sealing components are self-tested;

[0074] Processor: The processor is used to handle the calculation process of each formula and the construction calculation process of each model.

[0075] See Figure 1 As shown, the automated detection method for breathing valves includes:

[0076] Step 1: Inject a pre-set type of test medium into the breather valve test line, adjust the pressure in the line to the reference pressure for sealing performance testing, and collect the pressure change data in the line for reference parameter calibration.

[0077] Step 2: Based on the calibrated reference parameters, adjust the pressure in the test pipeline to different test levels in sequence to perform positive pressure sealing test, negative pressure sealing test and variable pressure circulation sealing test. Collect the pressure change rate and leakage amount at each test level in real time to form a sealing performance test dataset.

[0078] Step 3: Adjust the flow rate of the test medium based on the preset flow gradient, collect the pressure difference between the inlet and outlet of the breathing valve under different flow conditions through the pressure sensing component, obtain the ventilation resistance value corresponding to each flow, and generate the flow-resistance characteristic curve of the breathing valve based on the resistance value and flow data.

[0079] Step 4: Based on the complex operating conditions data in the actual application of the breather valve, simulate temperature cycle changes, pressure pulse impacts and media corrosion environment, and collect sealing performance and ventilation resistance change data through automatic repeated sealing performance testing and ventilation resistance testing process.

[0080] Step 5: Based on the preset normal parameter threshold range, combine machine learning algorithms to identify anomalies in the collected change data, determine the cause of the detected abnormal data, and make compensation and correction.

[0081] Step Six: Integrate the valid data in the testing process to generate test results including sealing performance, air resistance, flow characteristics and durability, and judge each indicator to output three judgment results: qualified, unqualified and pending retest.

[0082] Step 7: After the test is completed, drain the test medium from the test pipeline, reset the sample, and start the equipment self-test process to test the performance status of each component.

[0083] See Figure 2 As shown, a preset type of test medium is injected into the breather valve test line, the pressure in the line is adjusted to the reference pressure for sealing performance testing, and the pressure change data in the line is collected for reference parameter calibration. Specifically, this includes:

[0084] Inject the medium into the test pipeline according to the preset medium type, and use a gradient pressure increase and decrease method to adjust the pressure in the pipeline to the preset sealing performance test reference pressure;

[0085] Once the pressure reaches the baseline value, the pressure data in the pipeline is collected through the pressure sensing component, the average and maximum pressure fluctuations are calculated, and the baseline parameter calibration is completed by comparing them with the preset allowable fluctuation range.

[0086] If the pressure fluctuation exceeds the allowable range, the pipeline sealing self-test process is triggered, the two ends of the breather valve are closed, the pipeline is subjected to segmented pressure testing, and the leak point is located.

[0087] Specifically, the test medium type is determined according to the actual application scenario of the breather valve. If the breather valve is used in a gas delivery system, dry nitrogen is selected as the test medium. If it is used in a liquid medium system, anti-wear hydraulic oil that matches the actual working conditions is selected. If it is used in a multiphase flow scenario, a gas-liquid mixture medium is configured.

[0088] By compensating for pressure fluctuations in real time, the reference pressure fluctuation range is kept within ±0.005MPa and maintained in this stable state for 20 minutes. During this period, high-precision pressure sensing components arranged in a distributed manner are used to collect pressure data in the pipeline.

[0089] See Figure 3 As shown, based on the calibrated reference parameters, the pressure in the test pipeline is sequentially adjusted to different test levels to perform positive pressure sealing tests, negative pressure sealing tests, and variable pressure cyclic sealing tests. The pressure change rate and leakage amount at each test level are collected in real time to form a sealing performance test dataset, which specifically includes:

[0090] Based on the calibrated baseline pressure parameters, specific parameter thresholds are generated for three detection levels: positive pressure, negative pressure, and variable pressure cycle.

[0091] The pressure in the test pipeline is increased to the preset positive pressure sealing test range, stabilized for 60 seconds, and then the pressure is uniformly reduced to the negative pressure test threshold. After stabilization, it enters the pressure transformation cycle stage and cycles 10 times according to the positive pressure, normal pressure and negative pressure cycle.

[0092] Ten sets of pipeline pressure data are collected by a pressure sensing component to calculate the real-time pressure change rate, and the leakage amount at each detection stage is recorded by a pipeline end leakage detection device.

[0093] The pressure change rate and leakage parameters are integrated into a sealing performance test dataset after noise reduction processing.

[0094] Specifically, based on the calibrated reference parameters, five levels of detection pressure are set according to the nominal pressure rating of the breathing valve, covering 50%, 80%, 100%, 120%, and 150% of the rated working pressure, with a preset stabilization time of 10 minutes for each level; the pressure is adjusted by a precision servo valve, with the adjustment accuracy controlled within ±0.003MPa;

[0095] For the first detection level, the pressure is steadily increased to the set pressure at a rate of 0.05 MPa / s. During the 10-minute pressure stabilization period, pressure data is collected by two sets of high-precision pressure sensors located on the outlet side of the breather valve and in the middle section of the pipeline. The laser leak detection device is activated to scan and detect key sealing parts such as the breather valve seat and sealing gasket. The pressure change curve and leakage data are recorded in real time. If the pressure change rate is ≤0.001 MPa / min and the leakage is ≤0.001 mL / s within 10 minutes, the positive pressure seal is deemed qualified for this level; otherwise, it is marked as abnormal and the start time of the abnormality is recorded.

[0096] After completing the full-level positive pressure test, switch to the negative pressure sealing test mode: use a vacuum generator to evacuate at a rate of 0.03MPa / s to sequentially reach the negative pressure value corresponding to each test level, and stabilize the pressure for 8 minutes at each level; use a differential pressure sensor to monitor pipeline pressure changes, and at the same time use the sealing cavity negative pressure adsorption detection method to assist in verifying the leakage situation, that is, indirectly judge the leakage status by the pressure change of the sealing cavity outside the breather valve, and collect and record the negative pressure value, pressure change rate and leakage amount data;

[0097] Conduct variable pressure cycle seal test: Set the pressure change cycle to 60 seconds (30 seconds of pressure increase, 10 seconds of pressure stabilization, and 20 seconds of pressure release), pressure rise and fall rate to 0.1 MPa / s, cycle 15 times, and the cycle pressure range covers the maximum and minimum values ​​of each test level; during the cycle, the pressure sensor and leakage detection device work continuously, focusing on collecting pressure change points and leakage peak data during the pressure increase and pressure release phases; after the cycle, stabilize the pressure for an additional 15 minutes to verify the stability of the sealing performance.

[0098] See Figure 4 As shown, the flow rate of the test medium is adjusted based on a preset flow gradient. The pressure difference between the inlet and outlet of the breathing valve is collected through a pressure sensing component under different flow conditions to obtain the ventilation resistance value corresponding to each flow rate. Based on the resistance value and flow data, the flow-resistance characteristic curve of the breathing valve is generated, specifically including:

[0099] Based on the model and size data of the breathing valve to be tested, the test condition parameters are matched with the preset test standard library. The test condition parameters include the medium type, initial pressure, temperature range, test duration and number of cycles.

[0100] A flow regulation scheme is generated based on the matched operating parameters, including the flow range, residence time and regulation rate of each gradient.

[0101] Based on the preset scheme, the flow rate of the test medium is adjusted. During the stabilization stage of each flow gradient, pressure data is collected synchronously through the pressure sensing components at the inlet and outlet of the breather valve, and the pressure difference between the inlet and outlet is calculated.

[0102] Based on the collected pressure difference and corresponding flow rate, the ventilation resistance under each flow rate condition is calculated. Based on real-time monitoring of the pressure change inflection point, the opening pressure, fully opening pressure and closing pressure of the breathing valve are obtained.

[0103] The flow rate, resistance, and characteristic parameters of each gradient are correlated and integrated to generate a flow rate-resistance characteristic curve.

[0104] Specifically, keep the breather valve detection line sealed, close the pressure regulating unit, open the flow regulating loop valve, and set 13 continuous flow gradients according to the breather valve's rated flow parameters, with intervals of 10% of the rated flow, covering the entire flow range; after each flow gradient reaches the set value, maintain a stable flow state for 5 minutes.

[0105] Using a differential pressure sensor, 300 sets of inlet and outlet pressure difference data are continuously collected for each flow gradient, while simultaneously recording the corresponding flow rate and medium temperature data. A 5-point moving average method is used to remove flow noise interference, and the 3σ criterion is used to eliminate abnormal data caused by instantaneous fluctuations in the sensor. The average value of the remaining valid data is taken as the final pressure difference value under that flow condition, which directly corresponds to the ventilation resistance value of the breathing valve. The repeatability of each set of flow-resistance data is verified. If the resistance value of three consecutive measurements under the same flow gradient deviates by ≤2%, the gradient detection is repeated.

[0106] The flow rate values ​​corresponding to all flow rate gradients are correlated with the ventilation resistance values. A quadratic polynomial fitting using the least squares method is performed to generate a smooth flow-resistance characteristic curve. The curve is labeled with the resistance value corresponding to the rated flow rate point, the maximum allowable resistance threshold line, and the critical flow rate.

[0107] See Figure 5 As shown, based on complex operating conditions data from actual applications of breather valves, simulations of temperature cycling, pressure pulse impacts, and media corrosion environments are performed. Through an automated, repetitive sealing performance testing and ventilation resistance testing process, data on sealing performance and ventilation resistance changes are collected, specifically including:

[0108] Based on various working conditions in the actual application of breather valves, a multi-working-condition collaborative simulation scheme is generated. Through the synergistic effect of temperature gradient rise and fall, pressure pulse impact and continuous wetting by corrosive media, the actual values ​​of parameters for each working condition are obtained.

[0109] The detection process is triggered based on a preset time interval, and the sealing performance test process and the ventilation resistance test process are repeated.

[0110] Throughout the entire test period, high-frequency data on sealing performance and ventilation resistance changes were collected and stored in a time-stamped manner.

[0111] Specifically, based on the actual application scenarios of the breather valve, the following composite operating parameters are set: the temperature cycle adopts a wide temperature range of -40 to 120℃, the heating and cooling rate is 2℃ / min, the single cycle cycle is 60 minutes, and the total number of cycles is 30; the pressure pulse impact is set with an amplitude range of 0-1.5 times the rated pressure of the breather valve, a pulse frequency of 0.5Hz, 500 impacts, and a sine wave pulse waveform; the corrosive medium environment is selected using 3.5% sodium chloride solution or 5% sulfuric acid solution.

[0112] A temperature-pressure-corrosion composite working condition is constructed according to the set parameters. A pressure pulse impact is carried out simultaneously every 5 temperature cycles. The benchmark calibration, sealing performance test and ventilation resistance test are automatically repeated in each working condition cycle stage. The data acquisition frequency is increased to 25Hz, focusing on recording the decay trend of sealing performance parameters and ventilation resistance values, marking the performance change inflection point of each cycle, and forming a durability test time series dataset.

[0113] See Figure 6 As shown, based on a preset normal parameter threshold range, and combined with machine learning algorithms, anomaly identification is performed on the collected changing data. The causes of the detected anomalies are determined, and compensation and correction are implemented. Specifically, this includes:

[0114] The data on sealing performance and ventilation resistance changes are denoised and smoothed, invalid collection points are removed, and a standardized analysis dataset is formed.

[0115] A machine learning classification algorithm based on random forest is used to perform time series analysis on the standardized analysis dataset. By comparing it with threshold ranges and historical qualified data samples, abnormal data points such as pressure mutations, excessive leakage, and abnormal resistance fluctuations are identified.

[0116] For the identified abnormal data, the corresponding timestamps of the detection conditions, equipment operating status, and pipeline sealing status are traced back.

[0117] If the error is determined to be due to equipment error, the parameter calibration and correction process is initiated, and a compensation coefficient is calculated based on the error difference for real-time compensation. If the error is determined to be due to a defect in the breathing valve itself, the defect type and corresponding detection node are marked.

[0118] Specifically, a machine learning classification model based on random forest is launched. The standardized dataset is divided into time-series data segments in 10-second time windows and input into the model for window-by-window analysis. The model calls a preset normal parameter threshold library and historical qualified data samples. By comparing the trend consistency and parameter deviation between the time-series data segments and qualified samples, abnormal data points are accurately identified: pressure mutation is determined as a pressure change exceeding 0.05MPa within 1 second; leakage exceeding the standard is determined as 3 consecutive data points exceeding the 0.001mL / s threshold; and abnormal resistance fluctuation is determined as a resistance value change exceeding 10% of the rated value within 5 seconds. After identification, the timestamp, the detection item, and the abnormality type of the abnormal data are marked.

[0119] Based on the timestamps of abnormal data, a traceability chain is constructed by retrieving all related data from the same period: First, the operating parameters are checked to verify the deviation between the set values ​​and actual values ​​of temperature, pressure, and flow at the corresponding time; second, the equipment operating status is examined by extracting the stability of sensor output signals and the response speed of pressure and flow regulation units; third, the pipeline sealing status is checked. By cross-comparing the traceability data, the root cause of the anomaly is determined: if the parameter deviation is synchronized with the fluctuations in equipment operation and there are no abnormalities in the pipeline sealing, it is determined to be an equipment error; if only a single sample shows an anomaly, and the pipeline sealing is good and the equipment parameters are stable, it is determined to be a defect in the breather valve itself.

[0120] If the error is due to equipment, immediately initiate the parameter calibration and correction process: pause the current test, switch to the backup sensor to verify the data, calculate the compensation coefficient based on the error difference between the abnormal data and the standard data, and combine it with the equipment's historical calibration records, and perform real-time compensation for similar test data; after compensation, restart the corresponding test item, collect 3 sets of parallel data, and if the deviation is ≤2%, the correction is deemed effective; if the defect is due to the breathing valve itself, locate the defect location, mark the defect type and corresponding test node, and generate a defect tracing report.

[0121] See Figure 7 As shown, the effective data in the testing process is integrated to generate test results including sealing performance, ventilation resistance, flow characteristics, and durability. Each indicator is then judged, and three judgment results are output: qualified, unqualified, and pending retest. Specifically, these include:

[0122] Obtain valid detection data after anomaly correction, and integrate time-series data and feature parameters under the same detection dimension according to the detection stage to form a standardized data matrix;

[0123] Extract the core test indicators of the standardized data matrix, including the maximum leakage and pressure stabilization time in sealing performance, the average gradient resistance in ventilation resistance, the opening and closing pressure threshold and flow-resistance curve characteristic parameters in flow characteristics, and the cycle tolerance number and parameter decay rate in durability.

[0124] Based on the preset pass / fail criteria, each extracted indicator is compared with the standard threshold one by one, and the judgment is automatically made by adopting the judgment rule that all items pass the test, a single item exceeds the standard, and edge fluctuations indicate that the test needs to be re-inspected.

[0125] Specifically, corrected and valid data are retrieved, including sealing performance data, ventilation resistance data, flow characteristic data, and durability data. Based on a preset judgment standard system, a tiered judgment is performed. The first tier is for individual indicator judgment: sealing performance must meet the following requirements: pressure change rate ≤ 0.001 MPa / min and leakage ≤ 0.001 mL / s for each test level; ventilation resistance must have a deviation ≤ ±5% at rated flow; the flow characteristic curve must have no abnormal inflection point; and the durability decay rate must be ≤ 10%. If an individual indicator does not meet these requirements, it is marked as "single item unqualified". The second tier is for comprehensive judgment: if all individual indicators are qualified, it is judged as "qualified"; if two or more individual indicators are unqualified, it is directly judged as "unqualified"; if only one individual indicator is at the critical threshold or the data shows slight fluctuations but the cause of the abnormality is not clear, it is judged as "pending re-inspection".

[0126] See Figure 8 As shown, after the test is completed, the test medium in the test pipeline is discharged, the sample is reset, and the equipment self-test process is started to test the performance status of each component, specifically including:

[0127] After the test is completed, open the discharge valve to discharge the test medium in the test pipeline, drive the clamping assembly to release the breather valve, and drive the positioning mechanism to reset to the initial standby position.

[0128] The equipment self-test program is started, and the performance of key components of the testing platform is tested in sequence, including: the action response speed test of the sensing components and the adjustment module, and the static sealing test of the sealing components.

[0129] After all components have completed their self-inspection, the self-inspection data is summarized and a self-inspection report is generated.

[0130] It should be noted that the order of the above embodiments of the present invention is merely for descriptive purposes and does not represent the superiority or inferiority of the embodiments. Furthermore, the above description focuses on specific embodiments of this specification. Additionally, the processes depicted in the accompanying drawings do not necessarily require a specific or sequential order to achieve the desired results. In some embodiments, multitasking and parallel processing are possible or may be advantageous.

[0131] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0132] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the principles of the present invention should be included within the protection scope of the present invention.

Claims

1. An automated detection method for a breathing valve, characterized in that, include: Inject a pre-set type of test medium into the breather valve test line, adjust the pressure in the line to the sealing performance test reference pressure, and collect the pressure change data in the line for reference parameter calibration. Based on the calibrated baseline parameters, the pressure in the test pipeline is adjusted to different test levels in sequence to perform positive pressure sealing test, negative pressure sealing test and variable pressure cyclic sealing test. The pressure change rate and leakage amount at each test level are collected in real time to form a sealing performance test dataset. Based on the preset flow gradient, the flow rate of the test medium is adjusted. The pressure difference between the inlet and outlet of the breathing valve under different flow conditions is collected by the pressure sensing component. The ventilation resistance value corresponding to each flow is obtained. Based on the resistance value and flow data, the flow-resistance characteristic curve of the breathing valve is generated. Based on the complex operating conditions data of the breather valve in actual application, the simulation of temperature cycle change, pressure pulse impact and medium corrosion environment is carried out. Through the automatic repeated sealing performance test and ventilation resistance test process, the sealing performance and ventilation resistance change data are collected. Based on the preset normal parameter threshold range, combined with machine learning algorithms, anomalies are identified in the collected changing data, the causes of the detected abnormal data are determined, and compensation and correction are performed. The system integrates the effective data from the testing process to generate test results that include sealing performance, ventilation resistance, flow characteristics, and durability. It also judges each indicator and outputs three judgment results: qualified, unqualified, and pending retest. After the test is completed, the test medium in the test pipeline is discharged, the sample is reset, and the equipment self-test process is started to test the performance status of each component.

2. The automated detection method for a breathing valve according to claim 1, characterized in that, The process of injecting a preset type of test medium into the breather valve detection pipeline, adjusting the pressure in the pipeline to the sealing performance test reference pressure, and collecting pressure change data in the pipeline for reference parameter calibration specifically includes: Inject the medium into the test pipeline according to the preset medium type, and use a gradient pressure increase and decrease method to adjust the pressure in the pipeline to the preset sealing performance test reference pressure; Once the pressure reaches the baseline value, the pressure data in the pipeline is collected through the pressure sensing component, the average and maximum pressure fluctuations are calculated, and the baseline parameter calibration is completed by comparing them with the preset allowable fluctuation range. If the pressure fluctuation exceeds the allowable range, the pipeline sealing self-test process is triggered, the two ends of the breather valve are closed, the pipeline is subjected to segmented pressure testing, and the leak point is located.

3. The automated detection method for a breathing valve according to claim 1, characterized in that, Based on the calibrated reference parameters, the pressure in the detection pipeline is sequentially adjusted to different detection levels to perform positive pressure sealing tests, negative pressure sealing tests, and variable pressure cyclic sealing tests. The pressure change rate and leakage amount at each detection level are collected in real time to form a sealing performance test dataset. Specifically, this includes: Based on the calibrated baseline pressure parameters, specific parameter thresholds are generated for three detection levels: positive pressure, negative pressure, and variable pressure cycle. The pressure in the test pipeline is increased to the preset positive pressure sealing test range, stabilized for 60 seconds, and then the pressure is uniformly reduced to the negative pressure test threshold. After stabilization, it enters the pressure transformation cycle stage and cycles 10 times according to the positive pressure, normal pressure and negative pressure cycle. Ten sets of pipeline pressure data are collected by a pressure sensing component to calculate the real-time pressure change rate, and the leakage amount at each detection stage is recorded by a pipeline end leakage detection device. The pressure change rate and leakage parameters are integrated into a sealing performance test dataset after noise reduction processing.

4. The automated detection method for a breathing valve according to claim 1, characterized in that, The process of adjusting the flow rate of the test medium based on a preset flow gradient, collecting the pressure difference between the inlet and outlet of the breathing valve under different flow conditions through a pressure sensing component, obtaining the ventilation resistance value corresponding to each flow rate, and generating the breathing valve flow-resistance characteristic curve based on the resistance value and flow data specifically includes: Based on the model and size data of the breathing valve to be tested, the test condition parameters are matched with the preset test standard library. The test condition parameters include the medium type, initial pressure, temperature range, test duration and number of cycles. A flow regulation scheme is generated based on the matched operating parameters, including the flow range, residence time and regulation rate of each gradient. Based on the preset scheme, the flow rate of the test medium is adjusted. During the stabilization stage of each flow gradient, pressure data is collected synchronously through the pressure sensing components at the inlet and outlet of the breather valve, and the pressure difference between the inlet and outlet is calculated. Based on the collected pressure difference and corresponding flow rate, the ventilation resistance under each flow rate condition is calculated. Based on real-time monitoring of the pressure change inflection point, the opening pressure, fully opening pressure and closing pressure of the breathing valve are obtained. The flow rate, resistance, and characteristic parameters of each gradient are correlated and integrated to generate a flow rate-resistance characteristic curve.

5. The automated detection method for a breathing valve according to claim 1, characterized in that, Based on complex operating conditions data from actual applications of breather valves, simulations of temperature cycling, pressure pulse impacts, and media corrosion environments are performed. Through an automated, repetitive sealing performance testing and ventilation resistance testing process, data on sealing performance and ventilation resistance changes are collected, specifically including: Based on various working conditions in the actual application of breather valves, a multi-working-condition collaborative simulation scheme is generated. Through the synergistic effect of temperature gradient rise and fall, pressure pulse impact and continuous wetting by corrosive media, the actual values ​​of parameters for each working condition are obtained. The detection process is triggered based on a preset time interval, and the sealing performance test process and the ventilation resistance test process are repeated. Throughout the entire test period, high-frequency data on sealing performance and ventilation resistance changes were collected and stored in a time-stamped manner.

6. The automated detection method for a breathing valve according to claim 1, characterized in that, The process of identifying anomalies in the collected data based on a preset normal parameter threshold range, combined with machine learning algorithms, and determining the cause of the detected anomalies and performing compensation and correction specifically includes: The data on sealing performance and ventilation resistance changes are denoised and smoothed, invalid collection points are removed, and a standardized analysis dataset is formed. A machine learning classification algorithm based on random forest is used to perform time series analysis on the standardized analysis dataset. By comparing it with threshold ranges and historical qualified data samples, abnormal data points such as pressure mutations, excessive leakage, and abnormal resistance fluctuations are identified. For the identified abnormal data, the corresponding timestamps of the detection conditions, equipment operating status, and pipeline sealing status are traced back. If the error is determined to be due to equipment error, the parameter calibration and correction process is initiated, and a compensation coefficient is calculated based on the error difference for real-time compensation. If the error is determined to be due to a defect in the breathing valve itself, the defect type and corresponding detection node are marked.

7. The automated detection method for a breathing valve according to claim 1, characterized in that, The process integrates effective data from the testing procedure to generate test results that include sealing performance, ventilation resistance, flow characteristics, and durability. Each indicator is then judged, and three judgment results are output: qualified, unqualified, and pending retest. Specifically, this includes: Obtain valid detection data after anomaly correction, and integrate time-series data and feature parameters under the same detection dimension according to the detection stage to form a standardized data matrix; Extract core testing indicators from the standardized data matrix, including maximum leakage and pressure stabilization time in sealing performance, average gradient resistance in ventilation resistance, opening and closing pressure thresholds and characteristic parameters of the flow-resistance curve in flow characteristics, and cycle tolerance number and parameter decay rate in durability. Based on the preset pass / fail criteria, each extracted indicator is compared with the standard threshold one by one, and the judgment is automatically made by adopting the judgment rule that all items pass the test, a single item exceeds the standard, and edge fluctuations indicate that the test needs to be re-inspected.

8. The automated detection method for a breathing valve according to claim 1, characterized in that, After the test is completed, the test medium in the test pipeline is discharged, the sample is reset, and the equipment self-test process is started to test the performance status of each component. Specifically, this includes: After the test is completed, open the discharge valve to discharge the test medium in the test pipeline, drive the clamping assembly to release the breather valve, and drive the positioning mechanism to reset to the initial standby position. The equipment self-test program is started, and the performance of key components of the testing platform is tested in sequence, including: the action response speed test of the sensing components and the adjustment module, and the static sealing test of the sealing components. After all components have completed their self-inspection, the self-inspection data is summarized and a self-inspection report is generated.

9. A multifunctional testing platform for breathing valves, used to implement the automated testing method for breathing valves as described in any one of claims 1-8, characterized in that, include: Test medium control module: Injects test medium according to preset type, adjusts pipeline pressure to the detection reference value, and realizes gradient pressure increase and decrease control; Sealing performance testing module: Performs positive pressure, negative pressure and variable pressure cycle sealing tests, collects pressure change rate and leakage data in real time, and generates standardized sealing performance dataset; Flow-resistance characteristic detection module: Based on the flow gradient, the medium flow rate is adjusted, the pressure difference between the inlet and outlet of the breathing valve is measured, the ventilation resistance is calculated, and the flow-resistance characteristic curve is generated; Complex operating condition simulation module: Simulates actual operating conditions through temperature cycling, pressure pulses, and corrosive media, triggering repeated tests to collect durability data; Intelligent analysis and anomaly handling module: uses machine learning algorithms to identify abnormal data, correlates with operating parameters to determine the cause of the anomaly, and realizes real-time compensation or defect marking; Result determination module: integrates test data, compares it with threshold standards, outputs the pass / fail determination result, and generates a test report; Equipment self-test module: After completing the test, the medium is emptied, the sample is reset, and the sensors and sealing components are self-tested; Processor: The processor is used to handle the calculation process of each formula and the construction calculation process of each model.

Images