A method for detecting abnormal particle parameters based on probe array

Abstract

This invention provides a method for detecting abnormal particle parameters based on probe arrays, applicable to fault monitoring and electrostatic detection in heavy-duty gas turbines. The method includes: uniformly arranging multiple electrostatic probe arrays on the inner wall of a metal gas pipeline, wherein each electrostatic probe array is connected to an electrostatic detection terminal; when particles are present in the metal gas pipeline, the electrostatic probe arrays capture the particles, and the time-domain electrostatic signals induced by the electrostatic probes in each array are transmitted to the electrostatic detection terminal for signal fusion and enhancement processing to obtain an enhanced signal dataset; based on the enhanced signal dataset, charge quantity fitting and velocity linear fitting are performed respectively to obtain linear fitting formulas for particle charge quantity and particle velocity, which can achieve real-time high-precision inversion of the charge quantity, velocity, and position of abnormally charged particles in the gas pipeline, and has the advantages of fast response, high accuracy, and strong anti-interference ability.

Description

Technical Field

[0001] This invention relates to the field of fault monitoring and electrostatic detection of heavy-duty gas turbines, and in particular to a method for detecting abnormal particle parameters based on a probe array. Background Technology

[0002] As a critical power equipment, the health status of heavy-duty gas turbines directly affects the safety and reliability of the power supply system. Among various fault symptoms in gas turbines, abnormal particles generated inside the gas pipeline (such as metal or non-metal debris generated due to rubbing, erosion, or burning of combustion chamber components) are one of the core indicators for early fault warning. These abnormal particles typically have a diameter of over 100 micrometers. In high-temperature, high-speed gas flow, they carry static charges due to friction and collision, causing significant changes in the electrostatic field level within the pipeline. Therefore, monitoring the electrostatic signals of the gas pipeline to detect abnormal particles has become an important technical approach in gas turbine health management systems (PHM).

[0003] The parameters of abnormal particles are helpful in determining the type or location of the current abnormal fault. However, there is no existing technology to detect the parameters of abnormal particles inside the pipe. Therefore, there is an urgent need for a solution to detect the parameters of abnormal particles inside the pipe. Summary of the Invention

[0004] To address the aforementioned problems, this invention provides a method for detecting abnormal particle parameters based on a probe array, used to monitor abnormal particles within the gas pipeline of a heavy-duty gas turbine, achieving early warning of faults. Specifically, it includes:

[0005] S1. Multiple electrostatic probe arrays are uniformly arranged on the inner wall of the metal gas pipeline, wherein each electrostatic probe array is connected to an electrostatic detection terminal.

[0006] S2. When there are particles in the metal gas pipeline, the electrostatic probe array captures the particles. The time-domain electrostatic signal generated by the electrostatic probe in each electrostatic probe array is transmitted to the electrostatic detection terminal for signal fusion and enhancement processing to obtain the enhanced signal dataset.

[0007] S3. Based on the enhanced signal dataset, perform charge fitting and velocity linear fitting respectively to obtain the particle charge linear fitting formula and the particle velocity linear fitting formula.

[0008] Optionally, the method of uniformly arranging multiple electrostatic probe arrays on the inner wall of the metal gas pipeline includes:

[0009] The number of electrostatic probe arrays must be at least four;

[0010] The electrostatic probes in the electrostatic probe array include a ceramic layer and a metal core;

[0011] The ceramic layer is made of alumina, zirconium oxide, silicon nitride, or aluminum nitride.

[0012] The ceramic layer is in physical contact or electrically insulated from the metal pipe, and the metal core is electrically insulated from the metal pipe.

[0013] Optionally, a plurality of the electrostatic probe arrays are uniformly mounted circumferentially on the inner wall of the metal gas pipeline;

[0014] The electrostatic probes in the electrostatic probe array are fixed to the inner wall of the metal gas pipeline by screws.

[0015] Optionally, when particles are present in the metal gas pipeline, the electrostatic probe array captures the particles by:

[0016] When particles are present in the metal gas pipeline, it is determined that there is a fault in the metal gas pipeline, and the electrostatic probe array captures the electrostatic signals of the particles.

[0017] Optionally, the time-domain electrostatic signal generated by the electrostatic probes in each electrostatic probe array is transmitted to the electrostatic detection terminal for signal fusion and enhancement processing, resulting in an enhanced signal dataset including:

[0018] The average value of all electrostatic probe signals is obtained by averaging the electrostatic signals detected by all electrostatic probes.

[0019] The standard deviation and bias of the signal are calculated based on the average value of all electrostatic probe signals.

[0020] The enhanced signal dataset is obtained based on the standard deviation and bias of the signal.

[0021] Optionally, the standard deviation of the signal is formulated as formula (1):

[0022] (1)

[0023] s1 is the standard deviation of the signal, and N is the number of all electrostatic probes; The average value of all electrostatic probe signals. Let be the electrostatic signal detected by the i-th electrostatic probe.

[0024] Optionally, the formula for the deviation of the signal is formula (2):

[0025] (2)

[0026] S2 represents the signal deviation, and j and k represent the probe numbers. Let be the potential of the j-th probe. Let be the potential of the k-th probe.

[0027] Optionally, the calculation formula for the data in the enhanced signal dataset is formula (3):

[0028] (3)

[0029] The linear regression data represents the data within the enhanced signal dataset, i.e., the data from the probe array; a and b are the regression coefficients determined through simulation, where a = 0.384 and b = -3.034.

[0030] 0.217.

[0031] The above technical solution has at least the following advantages compared with the existing technology:

[0032] By employing an array of more than two electrostatic probes circumferentially distributed along the pipe wall, this invention fundamentally overcomes the limitations of single-point monitoring. The array design achieves spatial coverage of the entire pipe cross-section, ensuring effective detection regardless of whether abnormal particles pass through the pipe's center or edge, significantly reducing the probability of missed detection. By introducing a linear regression algorithm to fuse multiple signals (calculating the mean, standard deviation, and bias), random noise and systematic errors caused by the randomness of particle positions are effectively suppressed, thereby significantly improving the signal-to-noise ratio and the accuracy of charge inversion. As shown in Table 1, the mean square error of the four-probe array is reduced by more than 93% compared to the two-probe array, demonstrating its high-precision detection capability under complex operating conditions.

[0033] This method can detect the charge and speed of abnormal particles. Based on these two parameters, technicians, combined with experience or other characteristics, can determine the material of the particles. Based on the material of the particles, those skilled in the art can determine the location of the abnormal particles within a wide range, and can troubleshoot the fault points within that range, effectively preventing the occurrence of larger faults.

[0034] This invention can realize real-time high-precision inversion of the charge, velocity and position of abnormally charged particles in gas pipelines. It has the advantages of fast response, high accuracy and strong anti-interference ability. It is suitable for fault early warning and health management in high temperature and high flow rate gas pipeline environments such as heavy gas turbines. Attached Figure Description

[0035] To more clearly illustrate the technical solutions in the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0036] Figure 1 This is a schematic diagram showing the distribution of four electrostatic probe arrays on the inner wall of a metal gas pipeline according to one embodiment of the present invention.

[0037] Figure 2 This is a schematic diagram showing the distribution of five electrostatic probe arrays on the inner wall of a metal gas pipeline according to one embodiment of the present invention.

[0038] Figure 3 This is a schematic diagram of the distribution of six electrostatic probe arrays on the inner wall of a metal gas pipeline according to one embodiment of the present invention.

[0039] Markings: 1. Ceramic isolation layer of electrostatic probe; 2. Metal electrode core of electrostatic probe; 3. Metal gas passage pipe wall. Detailed Implementation

[0040] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the described embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0041] Unless otherwise defined, the technical or scientific terms used in this invention shall have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms “first,” “second,” and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Similarly, the terms “an,” “a,” or “the,” and similar terms do not indicate a quantity limitation, but rather indicate the presence of at least one. The terms “comprising,” “including,” or “including,” and similar terms mean that the element or object preceding the word encompasses the element or object listed following the word and its equivalents, without excluding other elements or objects. The terms “connected,” “linked,” or “connected,” and similar terms are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect.

[0042] To achieve real-time, high-precision inversion of the charge, velocity, and position of abnormally charged particles in gas pipelines, this invention provides an abnormal particle parameter detection method based on a probe array. This method is used to monitor abnormal particles in the gas pipelines of heavy-duty gas turbines, enabling early warning of faults. Specifically, it includes:

[0043] S1. Multiple electrostatic probe arrays are uniformly arranged on the inner wall of the metal gas pipeline, wherein each electrostatic probe array is connected to an electrostatic detection terminal.

[0044] The number of electrostatic probe arrays must be at least four; the reasons for requiring at least four are as follows:

[0045] The MSE of different numbers of probe arrays were calculated based on simulation experiments, and the results are shown in Table 1 below:

[0046] Table 1

[0047]

[0048] In this embodiment, to quantitatively evaluate the impact of the number of probes on the performance of the probe array, the mean square error (MSE) is introduced as an evaluation index for the accuracy of the system's detection signal, as shown in the following formula:

[0049]

[0050] n is the total number of probes. Let be the average value of the potentials of n probes. represents the linear regression value of the probe array.

[0051] The results show that the MSE of the four-probe array is significantly lower than that of the three-probe array (from 1.55 to 0.34), while the signal stability and signal-to-noise ratio are greatly improved. In practical applications, the four-probe array can effectively distinguish complex situations such as multiple particles passing simultaneously and non-uniform particle flows. In more complex application scenarios, higher detection accuracy can be achieved by increasing the number of probes. This probe array is suitable for electrostatic detection inside pipes with higher flow rates and larger dimensions, such as fault monitoring of gas pipelines downstream of gas turbine combustion chambers. Figures 1 to 3 The diagrams show the distribution of 4, 5, and 6 electrostatic probe arrays within the pipe, respectively.

[0052] The electrostatic probe array comprises a ceramic layer and a metal core; the ceramic layer is made of alumina, zirconium oxide, silicon nitride, or aluminum nitride; the ceramic layer is in physical contact or electrically insulated from the metal pipe, and the metal core is electrically insulated from the metal pipe. Multiple electrostatic probe arrays are uniformly installed circumferentially on the inner wall of the metal gas pipe; the electrostatic probes within the electrostatic probe array are fixed to the inner wall of the metal gas pipe using screws.

[0053] In one specific implementation, the probe structure comprises a ceramic layer and an internal metal core. The ceramic layer material can be alumina, zirconium oxide, silicon nitride, aluminum nitride, or other high-temperature resistant insulating materials, or new materials composited with these materials. The metal core is a high-temperature resistant alloy composed of iron, chromium, nickel, tungsten, manganese, etc. The ceramic layer on the outside of the probe (electrostatic probe) can be in physical contact or electrically insulated from the metal pipe, while the metal pipe and the internal metal core of the probe must be electrically insulated. The internal metal core of the probe must be connected to an electrostatic detection terminal for detecting the electrostatic information of particles. The electrostatic sensor terminal includes an electrostatic sensor, signal processing circuitry, and host computer software. The ceramic layer material can be alumina, zirconium oxide, silicon nitride, aluminum nitride, or other high-temperature resistant insulating materials, or new materials composited with these materials. The metal core is a high-temperature resistant alloy composed of iron, chromium, nickel, tungsten, manganese, etc. The probes are circumferentially distributed on the metal pipe wall; the probes are fixed to the metal pipe wall using screws.

[0054] S2. When there are particles in the metal gas pipeline, the electrostatic probe array captures the particles. The time-domain electrostatic signal generated by the electrostatic probe in each electrostatic probe array is transmitted to the electrostatic detection terminal for signal fusion and enhancement processing to obtain the enhanced signal dataset.

[0055] When particles are present in the metal gas pipeline, a fault is determined to exist in the metal gas pipeline, and the electrostatic probe array captures the electrostatic signals of the particles. Generally, particles do not exist in pipelines; they only appear when a certain fault occurs, such as erosion, corrosion, or burning inside the combustion chamber. In such cases, particles will be present in the gas turbine pipeline. To roughly pinpoint the fault location, it is necessary to know the parameters of the particles and then compare their physical properties to roughly determine the type of particles. Since differences in particle size lead to different amounts of charge carried by the particles, causing changes in the electrostatic charge level in the gas pipeline, abnormal particulate matter can be detected based on changes in the electrostatic field of the gas pipeline. Therefore, this application uses an electrostatic probe array for detection.

[0056] The time-domain electrostatic signal generated by the electrostatic probes in each electrostatic probe array is transmitted to the electrostatic detection terminal for signal fusion and enhancement processing, resulting in an enhanced signal dataset including:

[0057] The average value of all electrostatic probe signals is obtained by averaging the electrostatic signals detected by all electrostatic probes.

[0058] The formula for calculating the average is: ;

[0059] N is the total number of electrostatic probes; The average value of all electrostatic probe signals. Let be the electrostatic signal detected by the i-th electrostatic probe.

[0060] The standard deviation and bias of the signal are calculated based on the average value of all electrostatic probe signals.

[0061] The formula for the standard deviation s1 of the signal is formula (1):

[0062] (1)

[0063] The formula for the deviation of the signal is formula (2):

[0064] (2)

[0065] S2 represents the signal deviation, and j and k represent the probe numbers. Let be the potential of the j-th probe. Let be the potential of the k-th probe.

[0066] The calculation formula for the data in the enhanced signal dataset is formula (3):

[0067] (3)

[0068] The linear regression data represents the data within the enhanced signal dataset, i.e., the data from the probe array; a and b are the regression coefficients determined through simulation, where a = 0.384 and b = -3.034. ;

[0069] S3. Based on the enhanced signal dataset, perform charge fitting and velocity linear fitting respectively to obtain the particle charge linear fitting formula and the particle velocity linear fitting formula:

[0070] For linear regression data The charge quantity is fitted, and the formula for linear fitting of charge quantity is: Where Q is the particle charge of the linearly fitted particle, The slope is fitted to the "output-charge" equation. The intercept for fitting the charge quantity;

[0071] For linear regression data Perform a Fast Fourier Transform (FFT) to transform the time-domain signal Convert the signal to the frequency domain and normalize it to obtain the normalized frequency domain signal. ;

[0072] Acquire signal 3dB bandwidth B w The linear fitting formula for velocity is: ;

[0073] Where v is the particle velocity of the linear fit, and k is the slope of the "signal 3dB bandwidth - velocity" fitting. The intercept for velocity fitting.

[0074] In one specific implementation, the probe array consists of four electrostatic probes that are evenly distributed circumferentially on the wall of the metal gas pipeline. The included angle between adjacent probes is 90°, ensuring good spatial coverage and symmetry of the electrostatic signal across the pipeline cross-section.

[0075] Each electrostatic probe is fixed to a pre-drilled mounting hole in the pipe wall via a threaded connection, ensuring mechanical strength and airtightness. The probe structure consists of an outer ceramic isolation layer and an inner metal electrode core. The ceramic isolation layer is made of high-purity alumina ceramic (…). It possesses excellent high-temperature insulation properties and mechanical strength, and can operate for extended periods in high-temperature environments above 800°C. The metal electrode core is made of a high-temperature alloy, primarily composed of nickel, chromium, and iron, exhibiting excellent high-temperature oxidation and corrosion resistance, making it suitable for the high-temperature, high-pressure gas flow environment of gas turbines.

[0076] The metal electrode core extends into the pipe, and its exposed length needs to be adjusted according to the actual pipe dimensions. A ceramic insulating layer ensures complete electrical insulation between the metal electrode core and the metal pipe wall, preventing grounding interference.

[0077] Four probes are connected to a multi-channel high-precision electrometer via shielded high-temperature signal cables. Each channel has a sampling rate of over 4kHz to meet the requirements for acquiring higher frequency signals. After pre-amplification and filtering, the signals are synchronously acquired by a high-speed data acquisition card and transmitted to the host computer for processing.

[0078] In this embodiment, the signals collected by the four probes are denoted as follows: , , , hereinafter referred to as The signal processing flow is as follows:

[0079] Calculate the average value of the four probe signals: ;

[0080] Calculate the standard deviation s1 and bias s2 of the four-probe array;

[0081] ;

[0082] ;

[0083] Perform linear regression to obtain the enhanced signal dataset. :

[0084] ;

[0085] Where the regression coefficient a 1−4 and b 1−4 Determined through preliminary calibration experiments;

[0086] right Perform linear fitting of charge: ;

[0087] Speed ​​inversion is still based on the 3dB bandwidth of the signal after FFT transformation. w : .

[0088] Based on the above calculations, the charge and velocity of the abnormal particles can be obtained, and the type of particles can be predicted based on the relevant parameters to pinpoint the range of the situation.

[0089] By employing an array of more than two electrostatic probes circumferentially distributed along the pipe wall, this invention fundamentally overcomes the limitations of single-point monitoring. The array design achieves spatial coverage of the entire pipe cross-section, ensuring effective detection regardless of whether abnormal particles pass through the pipe's center or edge, significantly reducing the probability of missed detection. By introducing a linear regression algorithm to fuse multiple signals (calculating the mean, standard deviation, and bias), random noise and systematic errors caused by the randomness of particle positions are effectively suppressed, thereby significantly improving the signal-to-noise ratio and the accuracy of charge inversion. As shown in Table 1, the mean square error of the four-probe array is reduced by more than 93% compared to the two-probe array, demonstrating its high-precision detection capability under complex operating conditions.

[0090] This method can detect the charge and speed of abnormal particles. Based on these two parameters, those skilled in the art, combined with experience or other characteristics, can determine the material of the particles. Based on the material of the particles, those skilled in the art can determine the location of the abnormal particles within a wide range, and can troubleshoot the fault points within that range, effectively preventing the occurrence of larger faults.

[0091] Furthermore, this invention proposes to use multiple algorithms such as probe array signal mean method, linear regression method, and nonlinear regression method to analyze the charge of charged particles and improve the calculation accuracy. It also proposes to use Fast Fourier Transform (FFT) and time domain method to solve the motion state and position of charged particles. This invention can realize real-time high-precision inversion of the charge, motion speed and position of abnormal charged particles in gas pipelines. It has the advantages of fast response, high accuracy and strong anti-interference ability. It is suitable for fault early warning and health management in high temperature and high flow rate gas pipeline environments such as heavy gas turbines.

[0092] The following points need to be explained:

[0093] (1) The accompanying drawings of the embodiments of the present invention only involve the structures involved in the embodiments of the present invention. Other structures can refer to the general design.

[0094] (2) For clarity, the thickness of layers or regions is enlarged or reduced in the drawings used to describe embodiments of the invention, i.e., these drawings are not drawn to scale. It is understood that when an element such as a layer, film, region or substrate is referred to as being “above” or “below” another element, the element may be “directly” located “above” or “below” the other element or there may be intermediate elements.

[0095] (3) Where there is no conflict, the embodiments of the present invention and the features in the embodiments can be combined with each other to obtain new embodiments.

[0096] The above are merely specific embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. The scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for detecting abnormal particle parameters based on a probe array, characterized in that, include: S1. Multiple electrostatic probe arrays are uniformly arranged on the inner wall of the metal gas pipeline, wherein each electrostatic probe array is connected to an electrostatic detection terminal. S2. When there are particles in the metal gas pipeline, the electrostatic probe array captures the particles. The time-domain electrostatic signal generated by the electrostatic probe in each electrostatic probe array is transmitted to the electrostatic detection terminal for signal fusion and enhancement processing to obtain the enhanced signal dataset. S3. Based on the enhanced signal dataset, perform charge fitting and velocity linear fitting respectively to obtain the particle charge linear fitting formula and the particle velocity linear fitting formula.

2. The method for detecting abnormal particle parameters based on a probe array according to claim 1, characterized in that, The method of uniformly arranging multiple electrostatic probe arrays on the inner wall of the metal gas pipeline includes: The number of electrostatic probe arrays must be at least four; The electrostatic probes in the electrostatic probe array include a ceramic layer and a metal core; The ceramic layer is made of alumina, zirconium oxide, silicon nitride, or aluminum nitride. The ceramic layer is in physical contact or electrically insulated from the metal pipe, and the metal core is electrically insulated from the metal pipe.

3. The method for detecting abnormal particle parameters based on a probe array according to claim 2, characterized in that, Multiple electrostatic probe arrays are uniformly installed circumferentially on the inner wall of the metal gas pipeline; The electrostatic probes in the electrostatic probe array are fixed to the inner wall of the metal gas pipeline by screws.

4. The method for detecting abnormal particle parameters based on a probe array according to claim 3, characterized in that, When particles are present in the metal gas pipeline, the electrostatic probe array captures the particles, including: When particles are present in the metal gas pipeline, it is determined that there is a fault in the metal gas pipeline, and the electrostatic probe array captures the electrostatic signals of the particles.

5. The method for detecting abnormal particle parameters based on a probe array according to claim 4, characterized in that, The time-domain electrostatic signal generated by the electrostatic probes in each electrostatic probe array is transmitted to the electrostatic detection terminal for signal fusion and enhancement processing, resulting in an enhanced signal dataset including: The average value of all electrostatic probe signals is obtained by averaging the electrostatic signals detected by all electrostatic probes. The standard deviation and bias of the signal are calculated based on the average value of all electrostatic probe signals. The enhanced signal dataset is obtained based on the standard deviation and bias of the signal.

6. The method for detecting abnormal particle parameters based on a probe array according to claim 5, characterized in that, The formula for the standard deviation of the signal is formula (1): ；（1） s1 is the standard deviation of the signal, and N is the number of all electrostatic probes; The average value of all electrostatic probe signals. Let be the electrostatic signal detected by the i-th electrostatic probe.

7. The method for detecting abnormal particle parameters based on a probe array according to claim 6, characterized in that, The formula for the deviation of the signal is formula (2): ；（2） S2 represents the signal deviation, and j and k represent the probe numbers. Let be the potential of the j-th probe. Let be the potential of the k-th probe.

8. The method for detecting abnormal particle parameters based on a probe array according to claim 7, characterized in that, The calculation formula for the data in the enhanced signal dataset is formula (3): ；（3） The linear regression data represents the data within the enhanced signal dataset, i.e., the data from the probe array; a and b are the regression coefficients determined through simulation, where a = 0.384 and b = -3.

034. =0.217.

Images