Air leakage on-line detection and positioning method and device for pneumatic actuating mechanism

Abstract

The invention provides a pneumatic actuator air leakage on-line detection and positioning method and device, and the method comprises the steps: deploying high-temperature-resistant acoustic sensors at an air source inlet, an air cylinder sealing surface and a stuffing box, installing a miniature pressure and flow sensor at an air source pipeline, and transmitting data to an edge terminal through a shielding cable; environment noise is collected in a standby state of the mechanism, a baseline spectrogram is generated, and a dynamic noise threshold value is set; multi-source signals are synchronously collected in the action period of an executing mechanism, band-pass filtering is carried out on sound wave signals, environmental interference is eliminated, and smooth processing is carried out on pressure and flow signals; and air leakage is judged and graded by combining the sound signal amplitude exceeding threshold value, the pressure drop rate and the flow sudden change characteristic, and accurate leakage point positioning is realized by utilizing the multi-point sound signal arrival time difference and the pressure / flow characteristic points. According to the method, online accurate identification and positioning of the leakage point can be realized, and meanwhile, the method adapts to a high-temperature, high-humidity and strong-noise field environment.

Description

Technical Field

[0001] This invention belongs to the field of detection technology, specifically relating to a method and device for online detection and location of air leakage in pneumatic actuators. Background Technology

[0002] Pneumatic actuators are key control components in the turbine hall of thermal power plants, widely used in core operations such as valve switching and damper adjustment. Their operational reliability directly affects the safety, stability, and energy consumption of the unit. The turbine hall is a complex environment with high temperature (ambient temperature can reach 40-60℃), high humidity (relative humidity often exceeds 70%), high dust, and strong electromagnetic interference. Existing leak detection technologies have significant shortcomings: 1. Manual inspection relies on the soap and water smear method, which is inefficient (a single inspection takes more than 15 minutes), highly subjective, and difficult to detect minute leaks (leakage rate <5mL / min); 2. Traditional ultrasonic detectors are affected by steam flow noise and equipment vibration noise in the turbine hall, resulting in a low signal-to-noise ratio and a false alarm rate exceeding 30%; 3. Pressure drop detection methods require the actuator to be shut down, which cannot meet the "non-stop or minimal downtime" operation requirements of thermal power plants, and cannot locate the leak point. The aforementioned problems lead to the delayed detection of air leakage faults in pneumatic actuators, which not only wastes compressed air but may also cause valve jamming, reduced regulation accuracy, and in severe cases, reduced unit load or unplanned shutdown. Therefore, there is an urgent need for an efficient, accurate, and online-operable air leakage detection method that is suitable for the complex environment of the turbine hall. Summary of the Invention

[0003] The present invention aims to at least partially solve one of the technical problems in the related art.

[0004] Therefore, the first objective of this invention is to provide an online method for detecting and locating air leakage in pneumatic actuators.

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies in the environment of steam turbine rooms in thermal power plants, and to provide a method for detecting air leakage in pneumatic actuators in steam turbine rooms of thermal power plants, so as to achieve online accurate identification and location of leaks, while adapting to the on-site environment of high temperature, high humidity and strong noise.

[0006] The second objective of this invention is to provide an online detection and location device for air leakage in pneumatic actuators.

[0007] The third objective of this invention is to provide a computer device.

[0008] The fourth objective of this invention is to provide a non-transitory computer-readable storage medium.

[0009] To achieve the above objectives, a first aspect of the present invention provides a method for online detection and location of air leakage in a pneumatic actuator, comprising: S1, deploy a multi-source sensor system, install high-temperature resistant acoustic sensors on the air source inlet pipe section, cylinder end cover sealing surface and piston rod stuffing box of the pneumatic actuator, and install miniature pressure sensor and flow sensor on the air source pipeline near the actuator, and transmit sensor data to the edge computing terminal through electromagnetic interference shielded cable; S2, establish an environmental noise baseline, control the pneumatic actuator to be in a standby state without action, collect environmental noise signals and perform spectrum analysis, generate an environmental noise baseline spectrum, and set a dynamic noise threshold based on the baseline spectrum. S3 collects dynamic leakage signals and performs multi-dimensional filtering. When the pneumatic actuator completes a full action cycle, it simultaneously acquires sound wave signals, pressure signals, and flow signals. It performs bandpass filtering on the sound wave signals and combines them with the environmental noise baseline spectrum to remove interference signals. It also performs smoothing processing on the pressure and flow signals to eliminate instantaneous fluctuations. S4 identifies and locates leaks based on multi-source signal characteristics. By comparing the amplitude of the target acoustic signal with the dynamic noise threshold, analyzing the pressure drop rate of the pressure signal during the stagnant phase, and the abrupt change characteristics of the flow signal, it determines the existence of leaks and classifies the leak rate level. It also determines the location of the leak by combining the signal arrival time difference of acoustic sensors at different locations with the characteristic points of the pressure and flow curves.

[0010] In one embodiment of the present invention, S1 includes: S11, the operating temperature range of the high-temperature resistant acoustic sensor installed in the air source inlet pipe section, cylinder end cover sealing surface and piston rod stuffing box is -20℃ to 80℃. S12, the shielded cable meets the electromagnetic interference resistance level ≥ EMCEN 61000-6-2 standard.

[0011] In one embodiment of the present invention, S2 includes: S21, when the pneumatic actuator is in a standby state without action, the ambient noise signal is continuously collected for 30 seconds by an acoustic sensor, wherein the collection time window length is 30 seconds and the sampling frequency is 44.1 kHz. S22, extract the frequency distribution characteristics of the noise signal through Fourier transform. The frequency distribution characteristics are mainly concentrated in the 50-2000Hz frequency band, and set the dynamic noise threshold based on the maximum amplitude of the baseline spectrum × 1.2.

[0012] In one embodiment of the present invention, S3 further includes: S31, when performing bandpass filtering on acoustic signals, a filter with a passband frequency range of 2000-20000Hz is used, and the passband frequency range avoids the main frequency band of environmental noise, 50-2000Hz. S32, when smoothing the pressure signal and flow signal, a moving average algorithm is used. The window size of the moving average algorithm is 5 sampling points and the sampling interval is 10ms.

[0013] In one embodiment of the present invention, S4 includes: S41, when a leak is determined, the amplitude of the target acoustic signal S must be ≥ 2 times the noise threshold T, and the pressure drop rate of the pressure curve P(t) during the stagnant phase must be ≥ 0.5 kPa / min, while the flow rate of the flow curve Q(t) during this phase must be ≥ 3 L / min. S42, when classifying leakage rate levels, slight leakage is defined as Q<5L / min, moderate leakage is defined as 5L / min≤Q<15L / min, and severe leakage is defined as Q≥15L / min.

[0014] To achieve the above objectives, a second aspect of the present invention provides an online detection and location device for air leakage in a pneumatic actuator, comprising: The sensor deployment module is used to install high-temperature resistant acoustic sensors on the air source inlet pipe section, cylinder end cover sealing surface and piston rod stuffing box of the pneumatic actuator, and to install miniature pressure sensors and flow sensors on the air source pipe near the actuator. The sensor data is transmitted to the edge computing terminal through electromagnetic interference shielded cables. An environmental noise baseline establishment module is used to control the pneumatic actuator to be in a standby state without action, collect environmental noise signals and perform spectrum analysis to generate an environmental noise baseline spectrum, and set a dynamic noise threshold based on the baseline spectrum. The dynamic leakage signal acquisition and filtering module is used to synchronously acquire sound wave signals, pressure signals and flow signals when the pneumatic actuator completes a full action cycle. The sound wave signal is bandpass filtered and combined with the environmental noise baseline spectrum to remove interference signals. The pressure signal and flow signal are smoothed to eliminate instantaneous fluctuations. The leak detection and location module is used to determine the existence of leaks and classify the leak rate level by comparing the amplitude of the target acoustic signal with the dynamic noise threshold, analyzing the pressure drop rate of the pressure signal during the stagnant phase, and the sudden change characteristics of the flow signal. It also determines the location of the leak by combining the signal arrival time difference of the acoustic sensors at different locations with the characteristic points of the pressure and flow curves.

[0015] The present invention provides a method and apparatus for online detection and location of air leakage in a pneumatic actuator, which has the following advantages: 1. Strong environmental adaptability: By adopting high-temperature resistant sensors, electromagnetic interference-resistant shielded cables, and environmental noise baseline filtering technology, it can work stably in the turbine room environment of 40-60℃ high temperature, >70% high humidity and strong electromagnetic interference, effectively avoiding detection errors caused by environmental factors.

[0016] 2. Online accurate detection: Without stopping the operation of the pneumatic actuator, it can complete the identification and location of leaks under normal operating conditions of the unit. The leak rate detection range covers 0.5-50L / min, the leak location accuracy is ≤5cm, and the false alarm rate is <5%, which solves the problems of low efficiency, high false alarm rate and need to stop the machine in traditional methods.

[0017] 3. Real-time early warning and traceability: Automatically generates detection reports and triggers graded early warnings in conjunction with the DCS system, which can promptly remind maintenance personnel to handle air leakage faults. At the same time, it retains the original sensor data for subsequent fault tracing and analysis, reducing the waste of compressed air and the safety risks of the unit caused by air leakage.

[0018] To achieve the above objectives, a third aspect of this application provides a computer device, including a processor and a memory; wherein the processor reads executable program code stored in the memory to run a program corresponding to the executable program code, for implementing the online detection and location method for air leakage of pneumatic actuators as described in the first aspect embodiment.

[0019] To achieve the above objectives, a fourth aspect of this application provides a non-transitory computer-readable storage medium storing a computer program that, when executed by a processor, implements the online detection and location method for air leakage of pneumatic actuators as described in the first aspect embodiment.

[0020] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0021] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of an online detection and location method for air leakage in a basic pneumatic actuator according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the process for detecting air leakage of pneumatic actuators in the turbine hall environment of a thermal power plant according to an embodiment of the present invention; Figure 3 This is a schematic diagram of system deployment according to an embodiment of the present invention; Figure 4 This is a structural diagram of an online air leakage detection and positioning device for a pneumatic actuator according to an embodiment of the present invention; Figure 5 It is a computer device according to an embodiment of the present invention. Detailed Implementation

[0022] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0023] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0024] The following description, with reference to the accompanying drawings, describes an online detection and location method and apparatus for air leakage in a pneumatic actuator according to an embodiment of the present invention.

[0025] Example 1 Figure 1 This is a flowchart of an online detection and location method for air leakage in a pneumatic actuator according to an embodiment of the present invention, such as... Figure 1 As shown, it includes: S1 deploys a multi-source sensor system, installing high-temperature resistant acoustic sensors on the air source inlet pipe section of the pneumatic actuator, the sealing surface of the cylinder end cover, and the piston rod stuffing box. Miniature pressure sensors and flow sensors are also installed on the air source pipeline near the actuator. Sensor data is transmitted to the edge computing terminal via electromagnetic interference shielded cables.

[0026] In some implementations, this step first involves installing high-temperature resistant acoustic sensors at the air inlet pipe section, cylinder end cover sealing surface, and piston rod stuffing box of the pneumatic actuator. These locations are where leaks are most likely to occur. The acoustic sensors capture the high-frequency sound waves (typically in the 2000-20000Hz range) generated by the leak to initially identify the leak. The selected acoustic sensor must have a wide operating temperature range. To adapt to the high altitude inside the turbine hall The sensor is set to a high-temperature environment. The sampling frequency is set to 44.1kHz to ensure complete capture of high-frequency leakage signals.

[0027] In addition, miniature pressure and flow sensors are installed near the actuator in the gas supply pipeline to monitor pressure changes within the pipeline in real time (measurement accuracy is...). kPa) and gas flow rate (range: (L / min). These sensors, together with the acoustic sensors, constitute a multi-dimensional data acquisition system, providing multi-source verification evidence for leak detection.

[0028] For data transmission, all sensors are connected to the edge computing terminal via electromagnetic interference-resistant shielded cables. These cables must meet the EMC EN 61000-6-2 standard to ensure signal integrity and stability in environments with strong electromagnetic interference. The edge computing terminal has a built-in temperature compensation module to correct the impact of high-temperature environments on sensor output signals, thereby improving detection accuracy.

[0029] Furthermore, S1 includes: S11, the high-temperature resistant acoustic sensor installed in the air source inlet pipe section, cylinder end cover sealing surface and piston rod stuffing box has an operating temperature range of -20℃ to 80℃.

[0030] In some implementations, the selected acoustic sensor needs to have high temperature resistance, with an operating temperature range of [temperature range missing]. It can adapt to the high altitude inside the turbine hall of a thermal power plant. The ambient temperature is controlled. The sensor's sampling frequency is set to 44.1kHz, conforming to international standards for audio signal acquisition (such as IEC 61000-6-2), ensuring good capture capability for high-frequency sound waves (typically in the 2000-20000Hz range) generated by air leakage. For installation, the sensor is fixed to the target location using threads or snap-fit ​​structures, ensuring a tight fit with the equipment surface to improve sound wave reception efficiency.

[0031] Specifically, the sensor placement must cover critical areas such as the air source inlet, cylinder end cover sealing surface, and piston rod stuffing box, which are common sealing failure points in pneumatic actuators. Simultaneously, the sensor is connected to the edge computing terminal via a shielded cable (electromagnetic interference immunity level ≥ EMC EN 61000-6-2) to effectively suppress the impact of strong electromagnetic interference in the turbine hall on signal transmission. The terminal's built-in temperature compensation module can perform real-time correction of the sensor output signal, eliminating sensitivity drift caused by changes in ambient temperature.

[0032] S12, the shielded cable meets the electromagnetic interference resistance level ≥ EMCEN 61000-6-2 standard.

[0033] In some implementations, the shielded cables used in the steps of this invention must meet the electromagnetic interference immunity level ≥ EMCEN 61000-6-2 standard. This standard is an internationally recognized specification for testing the immunity of equipment in industrial environments and is applicable to the stable operation of industrial equipment in complex electromagnetic environments. The anti-interference capability of shielded cables is mainly reflected in their suppression effect on electromagnetic fields (EMF). By wrapping a metal shielding layer (such as aluminum foil or braided copper mesh) around the conductor and using a twisted-pair structure, the interference of external electromagnetic noise on signal transmission can be effectively reduced. In the turbine hall of thermal power plants, due to the presence of a large number of electromagnetic interference sources such as high-voltage equipment, frequency converters, and motors, ordinary cables are prone to signal distortion due to induced noise, affecting the accuracy of leak detection. Therefore, this invention uses shielded cables that conform to the EMCEN 61000-6-2 standard to ensure signal integrity is maintained even in industrial environments with interference intensity ≥ 30V / m.

[0034] Furthermore, the shielded cable is designed with an impedance matching of 50Ω to reduce signal reflection and transmission loss. Its transmission bandwidth covers 20Hz–20kHz, meeting the real-time data transmission requirements of the acoustic sensor at a sampling frequency of 44.1kHz. During cabling, the cable should be laid parallel to high-voltage lines as much as possible, and the shielding layer must be grounded at a single point at the edge computing terminal to prevent ground loop interference. In addition, the cable's temperature resistance must match the sensor, with an operating temperature range of [insert temperature range here]. To adapt to the high-temperature environment inside the turbine hall.

[0035] S2, establish an environmental noise baseline, control the pneumatic actuator to be in a standby state without action, collect environmental noise signals and perform spectrum analysis to generate an environmental noise baseline spectrum, and set a dynamic noise threshold based on the baseline spectrum.

[0036] This step involves controlling the pneumatic actuator to a standby, inactive state, ensuring its air supply pressure remains within the rated range (e.g., 0.4~0.6MPa), while the actuator refrains from any extension, retraction, or rotation operations to avoid introducing interference signals from non-environmental noise. In this state, high-temperature resistant acoustic sensors (operating temperature range: [insert temperature range here]) are deployed at key components of the actuator. The ambient noise signal was continuously collected for 30 seconds (with a sampling frequency of 44.1 kHz), and this initial noise signal was recorded. .

[0037] After the data collection is completed, the edge computing terminal performs the following steps: A Fast Fourier Transform (FFT) was performed to extract its frequency distribution characteristics. The ambient noise in the turbine hall is mainly concentrated in the 50–2000 Hz frequency band; therefore, a representative baseline ambient noise spectrum can be generated through spectral analysis. This spectrum reflects the noise distribution characteristics within the turbine hall, composed of steam flow, equipment vibration, and background noise, under leak-free conditions, providing a benchmark for subsequent leakage signal filtering.

[0038] Furthermore, based on the baseline spectrum, a dynamic noise threshold is set. This threshold is used to distinguish between environmental noise and characteristic acoustic signals generated by air leakage, thereby effectively filtering out interference in subsequent signal processing.

[0039] Furthermore, S2 includes: S21, when the pneumatic actuator is in a standby state without action, the ambient noise signal is continuously collected for 30 seconds by an acoustic sensor, wherein the collection time window length is 30 seconds and the sampling frequency is 44.1 kHz.

[0040] In some implementations, this step involves installing high-temperature resistant acoustic sensors at key locations such as the air inlet pipe section of the pneumatic actuator, the cylinder end cover sealing surface, and the piston rod stuffing box. The operating temperature range of these sensors is [insert temperature range here]. To meet the needs of steam turbine halls in thermal power plants The high-temperature environment requirements are met. The sampling frequency of the acoustic sensor is set to 44.1kHz, which conforms to the general standard for audio signal acquisition and can completely capture the high-frequency sound wave characteristics generated by air leakage.

[0041] In practice, the edge computing terminal controls the pneumatic actuator to remain inactive, ensuring the air pressure is maintained at the rated value and preventing additional noise from the actuator's movement. In this state, the acoustic sensor continuously collects ambient noise signals for 30 seconds, recording this as the initial noise signal. Subsequently, the edge computing terminal... A Fast Fourier Transform (FFT) is performed to extract the frequency distribution characteristics, which are typically concentrated in the 50-2000Hz range. Based on these spectral characteristics, the system generates a baseline spectrum of the environmental noise. And set the noise threshold accordingly. ,in This represents the maximum amplitude in the baseline spectrum. This threshold is used for subsequent filtering of the leak signal to remove environmental noise interference.

[0042] S22, extract the frequency distribution characteristics of the noise signal through Fourier transform. The frequency distribution characteristics are mainly concentrated in the 50-2000Hz frequency band, and set the dynamic noise threshold based on the maximum amplitude of the baseline spectrum × 1.2.

[0043] Specifically, this step first controls the pneumatic actuator to a standby state without operation, ensuring that the air source pressure is maintained within the rated range. At this time, the acoustic sensor collects 30 seconds of ambient noise signal at a sampling frequency of 44.1kHz, which is recorded as the initial noise signal. The sampling frequency satisfies the Nyquist sampling theorem and can completely cover the spectral range of the leaking sound wave signal.

[0044] Furthermore, the edge computing terminal collects... A Fast Fourier Transform (FFT) is performed to convert the time-domain signal into a frequency-domain signal, thereby extracting the frequency distribution characteristics of the noise signal. According to the experimental data in the technical disclosure document, the frequency spectrum of the turbine hall environmental noise is mainly concentrated in the 50-2000Hz frequency band, which significantly overlaps with the leakage sound wave signal. Therefore, it is necessary to clarify the noise baseline through spectral analysis. (Baseline spectrum diagram) In the process, the system identifies the point with the largest amplitude and sets a dynamic noise threshold based on that amplitude. This threshold is used to filter the amplitude of subsequent air leakage signals, effectively filtering out environmental noise interference.

[0045] S3 acquires dynamic leakage signals and performs multi-dimensional filtering. When the pneumatic actuator completes a full action cycle, it simultaneously acquires sound wave signals, pressure signals, and flow signals. It performs bandpass filtering on the sound wave signals and combines them with the environmental noise baseline spectrum to eliminate interference signals. It also smooths the pressure and flow signals to eliminate instantaneous fluctuations.

[0046] In some implementations, the system synchronously acquires signals through acoustic sensors, pressure sensors, and flow sensors, with a uniform sampling frequency of 10ms to ensure signal continuity and consistency over time. The acoustic sensors are positioned at key leakage risk points such as the air source inlet, cylinder end cover sealing surface, and piston rod stuffing box, and their operating temperature range is [insert temperature range here]. This ensures stable operation in the high-temperature environment of the turbine hall. The collected acoustic signals... First, the signal is preliminarily processed using a bandpass filter with a passband frequency set to 2000-20000Hz. This frequency band avoids the common environmental noise frequency range (50-2000Hz) found in turbine halls, thus effectively extracting the high-frequency characteristic sound waves generated by gas leakage. Further, the edge computing terminal compares the filtered signal with a pre-established environmental noise baseline spectrum. By comparing and eliminating interference signals with amplitudes less than or equal to the noise threshold, the target acoustic signal containing only air leakage characteristics is finally obtained. .

[0047] In pressure signal With flow signal In the processing, a moving average algorithm is used for smoothing, with a window size of 5 sampling points, to eliminate measurement errors caused by instantaneous fluctuations, thereby obtaining a more stable pressure curve. With flow curve This processing method conforms to the general standards for noise suppression and trend extraction in industrial signal processing.

[0048] Furthermore, S3 includes: S31, when performing bandpass filtering on acoustic signals, uses a filter with a passband frequency range of 2000-20000Hz, and the passband frequency range avoids the main frequency band of environmental noise, 50-2000Hz.

[0049] In some implementations, the passband frequency range of this filter is set to 2000-20000Hz. This range avoids the main noise frequency band (50-2000Hz) in the turbine hall environment, thereby significantly improving the signal-to-noise ratio (SNR) and providing high-quality acoustic data for subsequent leak identification and location. (Acoustic signal) Acoustic data is acquired using a high-temperature resistant sensor at a sampling frequency of 44.1 kHz, ensuring complete coverage of signals below 20,000 Hz. The bandpass filter can optionally employ an IIR (Infinite Impulse Response) or FIR (Finite Impulse Response) structure, and its design must comply with the electromagnetic interference immunity requirements of the IEC 61000-6-2 standard. In further implementation, the filter's transition band width should be controlled within 500 Hz to ensure accurate extraction of the target frequency band while avoiding excessive attenuation of adjacent frequency bands.

[0050] S32, when smoothing the pressure signal and flow signal, a moving average algorithm is used. The window size of the moving average algorithm is 5 sampling points and the sampling interval is 10ms.

[0051] In some implementations, the moving average algorithm generates a smoothed output signal by sliding a fixed-length window across the time series and averaging the signal values ​​within that window. Specifically, for any given time... The corresponding pressure signal and flow signal The smoothed values ​​can be expressed as follows:

[0052]

[0053] The window size is 5 sampling points, meaning two points are taken before and after the current sampling point, for a total of 5 points, which are then averaged. The sampling interval is 10ms, meaning data is collected every 10 milliseconds, and the signal time resolution is 100Hz. This parameter setting is well-adapted to the complex environment of the turbine room, ensuring real-time data transmission while avoiding signal distortion caused by excessively low sampling frequency.

[0054] Furthermore, the moving average algorithm is typically implemented using a circular buffer structure to reduce computational latency and improve processing efficiency. In edge computing terminals, this algorithm can be embedded into the data preprocessing module and run in parallel with the sensor acquisition module to ensure the real-time performance and stability of signal processing. This step complies with the real-time performance and reliability requirements for industrial control signal processing in the IEC 61131-3 standard.

[0055] S4 identifies and locates leaks based on multi-source signal characteristics. By comparing the amplitude of the target acoustic signal with the dynamic noise threshold, analyzing the pressure drop rate of the pressure signal during the stagnant phase, and the abrupt change characteristics of the flow signal, it determines the existence of leaks and classifies the leak rate level. It also determines the location of the leak by combining the signal arrival time difference of acoustic sensors at different locations with the characteristic points of the pressure and flow curves.

[0056] Specifically, in terms of leak detection, the system first processes the target acoustic signal after filtering. Perform amplitude analysis. If... The amplitude is consistently greater than or equal to (in If the maximum amplitude of the environmental noise baseline spectrum is multiplied by 1.2, it indicates the presence of abnormal acoustic characteristics, possibly indicating an air leak. Simultaneously, the system monitors the pressure signal of the actuator during the quiescent phase (i.e., when there is no extension or rotation). With flow signal .like voltage drop rate ,and numerical value This further confirms the existence of a leak. This method, through joint judgment of multiple signals, significantly reduces the false alarm rate and improves the robustness of leak detection.

[0057] In terms of leak location, the system utilizes data collected by multiple acoustic sensors. Location is determined by the Time Difference of Arrival (TDOA). Because the ultrasonic signals generated by leaks have directionality and propagation delay characteristics, the location of the sensor that first receives the signal can serve as a preliminary basis for locating the leak. Furthermore, this is combined with pressure curves... Pressure drop starting point and flow rate curve By identifying the mutation point, the system can accurately locate the leak, with a positioning accuracy of up to [missing information]. This method requires no shutdown and can be implemented under normal operating conditions in thermal power plants, meeting the needs of online monitoring.

[0058] Furthermore, S4 includes: S41, when a leak is determined, the amplitude of the target acoustic signal S must be ≥ 2 times the noise threshold T, and the pressure drop rate of the pressure curve P(t) during the stagnant phase must be ≥ 0.5 kPa / min, while the flow rate curve Q(t) during this phase must be ≥ 3 L / min.

[0059] Specifically, when the target acoustic signal The amplitude is continuously greater than or equal to twice the noise threshold over a continuous period of time. That is, satisfying This indicates the presence of significant ultrasonic leakage characteristics. This threshold... Based on the environmental noise baseline spectrum The maximum amplitude value is set by multiplying it by a factor of 1.2 to effectively suppress background noise interference and improve the signal-to-noise ratio.

[0060] Simultaneously, the system monitors the pressure curve. The pressure drop rate during the actuator's stationary phase. This phase is defined as the duration during which the actuator remains stationary after completing a full action (e.g., valve fully open / close). During this phase, if the pressure drop rate... This indicates a continuous gas leak, causing a slow drop in system pressure. This parameter is set based on the normal pressure maintenance capability of the pneumatic actuators in a non-operating state in a thermal power plant; exceeding this threshold indicates an abnormal gas leak.

[0061] In addition, the flow curve The values ​​during the motion stagnation phase must meet the following requirements. This further verifies the existence of leaks. The flow threshold considers the minimum leakage tolerance of the system in a non-operational state, ensuring sensitive detection of even minor leaks. Through joint analysis of acoustic, pressure, and flow signals, the system can achieve online, real-time leak identification, with a leak detection range covering [missing information]. The false alarm rate is less than 5%, which is significantly better than traditional methods.

[0062] S42, when classifying leakage rate levels, slight leakage is defined as Q<5L / min, moderate leakage is defined as 5L / min≤Q<15L / min, and severe leakage is defined as Q≥15L / min.

[0063] In some implementations, the leakage rate level is classified based on the range of the leakage flow rate Q measured by the flow sensor. Specifically, when Q < 5 L / min, it is considered a minor leak; when 5 L / min ≤ Q < 15 L / min, it is considered a moderate leak; and when Q ≥ 15 L / min, it is considered a severe leak. This classification standard is based on the actual energy consumption loss and operational risk assessment of pneumatic actuators in thermal power plants under different leakage levels, and has clear engineering significance. The flow sensor has a range of 0–50 L / min, a sampling interval of 10 ms, and a measurement accuracy of ±0.5 L / min, which can meet the requirements for identifying minor leaks.

[0064] Specifically, the leak rate thresholds (5 L / min and 15 L / min) set in this step are derived from the recommendations of ISO 12178 standard regarding the classification of leak levels in pneumatic systems, and optimized based on actual operating experience in thermal power plants. Simultaneously, this method uses a moving average algorithm to smooth Q(t) with a window size of 5 sampling points to eliminate the interference of instantaneous fluctuations on leak rate judgment, ensuring the stability and reliability of the data.

[0065] The present invention discloses an online method for detecting and locating air leaks in pneumatic actuators, which can realize online air leak detection and accurate location of pneumatic actuators in the high temperature, high humidity and strong electromagnetic interference environment of steam turbine room in thermal power plants. It has high detection sensitivity, low false alarm rate, and positioning accuracy ≤5cm, and does not require machine shutdown, which significantly improves the real-time performance and accuracy of air leak fault identification.

[0066] Example 2 The following describes in detail, with reference to the accompanying drawings, a method for online detection and location of air leakage in a pneumatic actuator according to an embodiment of the present invention.

[0067] A method for detecting air leakage in pneumatic actuators in a thermal power plant turbine hall, such as... Figure 2 As shown, it includes the following steps: S10: Detection System Deployment: Install adapter components at critical leakage risk points of pneumatic actuators, such as... Figure 3 As shown, it specifically includes: High-temperature resistant acoustic sensors (operating temperature range -20℃ to 80℃) are fixed at the air source inlet pipe section, cylinder end cover sealing surface, and piston rod stuffing box of the actuator, respectively. The sensor sampling frequency is set to 44.1kHz to collect the characteristic acoustic wave signal generated by air leakage. A miniature pressure sensor (measurement accuracy ±0.2kPa) and a flow sensor (range 0-50L / min) are installed near the actuator in the gas source pipeline to collect the pressure and flow values ​​in the pipeline in real time. All sensors are connected to the edge computing terminal via shielded cables (electromagnetic interference immunity level ≥ EMC EN 61000-6-2). The terminal has a built-in temperature compensation module to offset the impact of the high temperature in the turbine room on the sensor data. S20: Establishment of environmental noise baseline: S201. Control the pneumatic actuator to be in a standby state with no action (the air source pressure is maintained at the rated value, and the actuator has no extension or rotation movement). Collect the ambient noise signal of the turbine hall for 30 seconds through the sound sensor and record it as the initial noise signal S0. S202, the edge computing terminal performs Fourier transform on S0 to extract the frequency distribution characteristics of the noise signal (mainly concentrated in 50-2000Hz) and generate the environmental noise baseline spectrum B. S203. Based on the baseline spectrum B, set a noise threshold T (T = maximum amplitude of the baseline spectrum × 1.2) for subsequent filtering of environmental interference signals.

[0068] S30: Dynamic Leakage Signal Acquisition and Filtering S301. Control the pneumatic actuator to complete one complete action cycle according to the normal operating conditions of the thermal power plant (such as the valve going from fully closed to fully open and then back to fully closed). During this period, the acoustic sensor, pressure sensor and flow sensor collect data synchronously, with the collection interval set to 10ms, to obtain a dynamic signal group (acoustic signal S1, pressure signal P, flow signal Q). S302. Filter the acoustic signal S1: First, filter it with a bandpass filter (passband frequency 2000-20000Hz, avoiding the main frequency band of environmental noise), then compare the filtered signal with the environmental noise baseline spectrum B, remove signals with amplitude ≤ threshold T, and obtain the target acoustic signal S2 containing only air leakage characteristics. S303. Smooth the pressure signal P and the flow signal Q: Use the moving average algorithm (window size 5 sampling points) to eliminate instantaneous fluctuations and obtain stable pressure curve P(t) and flow curve Q(t).

[0069] S40: Leak Detection and Leak Location: Leak detection: If the amplitude of the target acoustic signal S2 is continuously ≥ 2 times the noise threshold T, and the pressure drop rate of the pressure curve P(t) during the actuator's stagnant phase (no extension / rotation) is ≥ 0.5 kPa / min, and the flow rate curve Q(t) during this phase is ≥ 3 L / min, then it is determined that there is air leakage in the pneumatic actuator. Based on the magnitude of S2, the pressure drop rate of P(t), and the value of Q(t), the leakage rate is classified into: slight leakage (Q < 5 L / min), moderate leakage (5 L / min ≤ Q < 15 L / min), and severe leakage (Q ≥ 15 L / min).

[0070] Leakage location: Compare the arrival times of the target acoustic signal S2 collected by acoustic sensors at different locations: the location of the sensor with the earliest arrival time is the core area of ​​the leak. By combining the pressure drop start position of the pressure curve P(t) with the abrupt change position of the flow curve Q(t), the leak range is further narrowed (positioning accuracy ≤ 5cm), and the leak location is obtained through the edge computing terminal. S50: Results output and early warning.

[0071] By applying this method to batch testing of pneumatic actuators in the turbine hall of Unit #1, the requirement for timely detection of air leakage faults can be met. It also overcomes noise interference, has the advantages of low false alarm rate, and requires no system shutdown for measurement. This achieves air leakage alarm upon actuator startup, ensuring safe system operation.

[0072] Example 3 To achieve the above embodiments, such as Figure 4 As shown, this embodiment also provides an online detection and location device 10 for air leakage of pneumatic actuators. The device 10 includes a sensor deployment module 100, an environmental noise baseline establishment module 200, a dynamic air leakage signal acquisition and filtering module 300, and an air leakage identification and location module 400.

[0073] The sensor deployment module 100 is used to install high-temperature resistant acoustic sensors on the air source inlet pipe section, cylinder end cover sealing surface and piston rod stuffing box of the pneumatic actuator, and to install miniature pressure sensors and flow sensors on the air source pipe near the actuator. The sensor data is transmitted to the edge computing terminal through electromagnetic interference shielded cables. The environmental noise baseline establishment module 200 is used to control the pneumatic actuator to be in a standby state without action, collect environmental noise signals and perform spectrum analysis to generate an environmental noise baseline spectrum, and set a dynamic noise threshold based on the baseline spectrum. The dynamic leakage signal acquisition and filtering module 300 is used to synchronously acquire sound wave signals, pressure signals and flow signals when the pneumatic actuator completes a full action cycle. It performs bandpass filtering on the sound wave signal and combines it with the environmental noise baseline spectrum to remove interference signals. It also performs smoothing processing on the pressure signal and flow signal to eliminate instantaneous fluctuations. The leak detection and location module 400 is used to determine the existence of leaks and classify the leak rate level by comparing the amplitude of the target acoustic signal with the dynamic noise threshold, analyzing the pressure drop rate of the pressure signal during the action stagnation phase and the sudden change characteristics of the flow signal, and combining the signal arrival time difference of acoustic sensors at different locations with the characteristic points of the pressure and flow curves to determine the location of the leak.

[0074] Furthermore, the aforementioned environmental noise baseline establishment module 100 is also used for: When the pneumatic actuator is in a standby state without any action, the ambient noise signal is continuously collected for 30 seconds by an acoustic sensor, with a collection time window length of 30 seconds and a sampling frequency of 44.1 kHz. The frequency distribution characteristics of the noise signal are extracted by Fourier transform. The frequency distribution characteristics are mainly concentrated in the 50-2000Hz frequency band, and the dynamic noise threshold is set based on 1.2 times the maximum amplitude of the baseline spectrum.

[0075] Furthermore, the aforementioned dynamic leakage signal acquisition and filtering module 200 is also used for: When performing bandpass filtering on acoustic signals, a filter with a passband frequency range of 2000-20000Hz is used, and the passband frequency range avoids the main frequency band of environmental noise, 50-2000Hz. When smoothing the pressure and flow signals, a moving average algorithm is used, with a window size of 5 sampling points and a sampling interval of 10ms.

[0076] Furthermore, the aforementioned dynamic leakage signal acquisition and filtering module 300 is also used for: When performing bandpass filtering on acoustic signals, a filter with a passband frequency range of 2000-20000Hz is used, and the passband frequency range avoids the main frequency band of environmental noise, 50-2000Hz. When smoothing the pressure and flow signals, a moving average algorithm is used, with a window size of 5 sampling points and a sampling interval of 10ms.

[0077] Furthermore, the aforementioned leak detection and location module 400 is also used for: When a leak is determined, the amplitude of the target acoustic signal S must be ≥ 2 times the noise threshold T, and the pressure drop rate of the pressure curve P(t) during the stagnant phase must be ≥ 0.5 kPa / min, while the flow rate of the flow curve Q(t) during this phase must be ≥ 3 L / min. When classifying leakage rates, a minor leak is defined as Q < 5 L / min, a moderate leak is defined as 5 L / min ≤ Q < 15 L / min, and a severe leak is defined as Q ≥ 15 L / min.

[0078] An online leak detection and location device for pneumatic actuators according to an embodiment of the present invention can realize online leak detection and accurate location of pneumatic actuators in the high temperature, high humidity and strong electromagnetic interference environment of steam turbine room in thermal power plants. It has high detection sensitivity, low false alarm rate, and positioning accuracy ≤5cm, and does not require shutdown, which significantly improves the real-time performance and accuracy of leak fault identification.

[0079] To implement the methods of the above embodiments, the present invention also provides a computer device, such as... Figure 5 As shown, the computer device 600 includes a memory 601 and a processor 602; wherein, the processor 602 reads executable program code stored in the memory 601 to run a program corresponding to the executable program code, so as to implement the various steps of the method described above.

[0080] To implement the above embodiments, this application also proposes a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements the method described in the foregoing embodiments.

[0081] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0082] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A method for online detection and location of air leakage in a pneumatic actuator, characterized in that, include: S1, deploy a multi-source sensor system, install high-temperature resistant acoustic sensors on the air source inlet pipe section, cylinder end cover sealing surface and piston rod stuffing box of the pneumatic actuator, and install miniature pressure sensor and flow sensor on the air source pipeline near the actuator, and transmit sensor data to the edge computing terminal through electromagnetic interference shielded cable; S2, establish an environmental noise baseline, control the pneumatic actuator to be in a standby state without action, collect environmental noise signals and perform spectrum analysis, generate an environmental noise baseline spectrum, and set a dynamic noise threshold based on the baseline spectrum. S3 collects dynamic leakage signals and performs multi-dimensional filtering. When the pneumatic actuator completes a full action cycle, it simultaneously acquires sound wave signals, pressure signals, and flow signals. It performs bandpass filtering on the sound wave signals and combines them with the environmental noise baseline spectrum to remove interference signals. It also performs smoothing processing on the pressure and flow signals to eliminate instantaneous fluctuations. S4 identifies and locates leaks based on multi-source signal characteristics. By comparing the amplitude of the target acoustic signal with the dynamic noise threshold, analyzing the pressure drop rate of the pressure signal during the stagnant phase, and the abrupt change characteristics of the flow signal, it determines the existence of leaks and classifies the leak rate level. It also determines the location of the leak by combining the signal arrival time difference of acoustic sensors at different locations with the characteristic points of the pressure and flow curves.

2. The method as described in claim 1, characterized in that, S1 includes: S11, the operating temperature range of the high-temperature resistant acoustic sensor installed in the air source inlet pipe section, cylinder end cover sealing surface and piston rod stuffing box is -20℃ to 80℃. S12, the shielded cable meets the electromagnetic interference resistance level ≥ EMCEN 61000-6-2 standard.

3. The method as described in claim 1, characterized in that, The S2 includes: S21, when the pneumatic actuator is in a standby state without action, the ambient noise signal is continuously collected for 30 seconds by an acoustic sensor, wherein the collection time window length is 30 seconds and the sampling frequency is 44.1 kHz. S22, extract the frequency distribution characteristics of the noise signal through Fourier transform. The frequency distribution characteristics are mainly concentrated in the 50-2000Hz frequency band, and set the dynamic noise threshold based on the maximum amplitude of the baseline spectrum × 1.

2.

4. The method as described in claim 1, characterized in that, The S3 further includes: S31, when performing bandpass filtering on acoustic signals, a filter with a passband frequency range of 2000-20000Hz is used, and the passband frequency range avoids the main frequency band of environmental noise, 50-2000Hz. S32, when smoothing the pressure signal and flow signal, a moving average algorithm is used. The window size of the moving average algorithm is 5 sampling points and the sampling interval is 10ms.

5. The method as described in claim 1, characterized in that, The S4 includes: S41, when a leak is determined, the amplitude of the target acoustic signal S must be ≥ 2 times the noise threshold T, and the pressure drop rate of the pressure curve P(t) during the stagnant phase must be ≥ 0.5 kPa / min, while the flow rate of the flow curve Q(t) during this phase must be ≥ 3 L / min. S42, when classifying leakage rate levels, slight leakage is defined as Q<5L / min, moderate leakage is defined as 5L / min≤Q<15L / min, and severe leakage is defined as Q≥15L / min.

6. A pneumatic actuator leakage online detection and positioning device, characterized in that, include: The sensor deployment module is used to install high-temperature resistant acoustic sensors on the air source inlet pipe section, cylinder end cover sealing surface and piston rod stuffing box of the pneumatic actuator, and to install miniature pressure sensors and flow sensors on the air source pipe near the actuator. The sensor data is transmitted to the edge computing terminal through electromagnetic interference shielded cables. An environmental noise baseline establishment module is used to control the pneumatic actuator to be in a standby state without action, collect environmental noise signals and perform spectrum analysis to generate an environmental noise baseline spectrum, and set a dynamic noise threshold based on the baseline spectrum. The dynamic leakage signal acquisition and filtering module is used to synchronously acquire sound wave signals, pressure signals and flow signals when the pneumatic actuator completes a full action cycle. The sound wave signal is bandpass filtered and combined with the environmental noise baseline spectrum to remove interference signals. The pressure signal and flow signal are smoothed to eliminate instantaneous fluctuations. The leak detection and location module is used to determine the existence of leaks and classify the leak rate level by comparing the amplitude of the target acoustic signal with the dynamic noise threshold, analyzing the pressure drop rate of the pressure signal during the stagnant phase, and the sudden change characteristics of the flow signal. It also determines the location of the leak by combining the signal arrival time difference of the acoustic sensors at different locations with the characteristic points of the pressure and flow curves.

7. The apparatus as claimed in claim 6, characterized in that, The environmental noise baseline establishment module is also used for: When the pneumatic actuator is in a standby state without any action, the ambient noise signal is continuously collected for 30 seconds by an acoustic sensor, with a collection time window length of 30 seconds and a sampling frequency of 44.1 kHz. The frequency distribution characteristics of the noise signal are extracted by Fourier transform. The frequency distribution characteristics are mainly concentrated in the 50-2000Hz frequency band, and the dynamic noise threshold is set based on 1.2 times the maximum amplitude of the baseline spectrum.

8. The apparatus as claimed in claim 6, characterized in that, The dynamic leakage signal acquisition and filtering module is also used for: When performing bandpass filtering on acoustic signals, a filter with a passband frequency range of 2000-20000Hz is used, and the passband frequency range avoids the main frequency band of environmental noise, 50-2000Hz. When smoothing the pressure and flow signals, a moving average algorithm is used, with a window size of 5 sampling points and a sampling interval of 10ms.

9. A computer device, characterized in that, Including processor and memory; The processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the online detection and location method for air leakage of pneumatic actuator as described in any one of claims 1-5.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements the online detection and location method for air leakage of pneumatic actuators as described in any one of claims 1-5.

Images