Reciprocating compressor fault diagnosis method based on multi-source vibration angular domain-working characteristic conjoint analysis

Abstract

The invention discloses a reciprocating compressor fault diagnosis method based on multi-source vibration angular domain-working characteristic conjoint analysis. The method comprises the following steps: firstly, synchronously collecting a vibration signal of a certain same key part of each cylinder and a key phase signal of a crankshaft; secondly, adaptively extracting a key phase pulse trigger position, dividing a rotation period and constructing an angular domain abscissa sequence; envelope processing and key phase truncation are carried out on the vibration signals, a phase compensation angle is determined according to the number of the air cylinder, and phase compensation is carried out on the vibration envelope signals of each channel; and finally, drawing a multi-source angular domain envelope-working characteristic joint diagram after phase compensation, and marking a component action state corresponding to the impact moment. Integrated display and synchronous analysis of the multi-source vibration signals of the compressor are achieved, vibration impact characteristics are associated with key component states, and the efficiency and accuracy of fault diagnosis of the reciprocating compressor are effectively improved.

Description

Technical Field

[0001] This invention belongs to the field of reciprocating compressor fault diagnosis technology, specifically relating to a reciprocating compressor fault diagnosis method based on multi-source vibration angular domain-operating characteristic joint analysis. Background Technology

[0002] Reciprocating compressors are general-purpose machines that compress gaseous media by changing the volume of their working chamber, and are widely used in industries such as petroleum, chemical, and energy. Due to their long-term operation in high-temperature, high-pressure, and corrosive environments, and the numerous moving parts and complex excitation transmission paths, malfunctions can pose a serious threat to enterprise production efficiency and personal safety. Vibration monitoring of reciprocating compressors using data acquisition equipment allows for the analysis and evaluation of the unit's operating status. The vibration signals of reciprocating compressors have the following characteristics: Reciprocating compressors have numerous vibration excitation sources, including gas force, inertial force, mechanical impact, and friction, resulting in significant nonlinear and non-stationary impact characteristics. Vibration impact signals generated by faults (such as broken piston rings or loose bolts) are superimposed on normal signals in both the time and frequency domains, making feature extraction difficult and distinguishing fault features from normal impacts. Therefore, in environments with multiple interferences, how to quickly and accurately extract and identify the vibration impact characteristics of key components has become a major technical challenge for this type of equipment.

[0003] Currently, conventional methods for diagnosing vibration faults in reciprocating compressors mainly revolve around the acquisition, processing, and feature extraction of vibration signals from the equipment surface (Yang Changhua, Zhang Dunhui, Teng Yang. Analysis of Monitoring and Fault Diagnosis Technology for Reciprocating Compressors in Shale Gas Fields [J]. China Equipment Engineering, 2021, (S1): 50-53.). The main steps are: first, acquiring vibration acceleration signals from key components such as cylinders and crossheads through a compressor status parameter monitoring system; then, processing individual measurement point signals using time-frequency analysis, wavelet packet analysis, and HHT methods to separate and extract vibration features related to the fault; finally, comprehensively judging the compressor's health status by monitoring the changing trends of vibration spectra or sensitive fault characteristic parameters. However, conventional methods have the following obvious limitations: on the one hand, they are only suitable for processing vibration signals from a single measurement point, performing vibration analysis on a single component in isolation, and cannot simultaneously analyze multi-source vibration signals from multiple cylinders, resulting in low analysis efficiency; on the other hand, they fail to fully integrate vibration impact characteristics with the working characteristics of key components, leading to weak targeting in fault feature extraction and identification, making it difficult to distinguish between normal impacts and fault impacts.

[0004] To address the shortcomings of the aforementioned methods, the key issue in improving the efficiency and accuracy of fault diagnosis is how to construct an efficient vibration signal analysis method for reciprocating compressors that fully utilizes multi-source vibration signals to achieve rapid and accurate assessment of the working status of critical components. Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, the present invention aims to provide a fault diagnosis method for reciprocating compressors based on the joint analysis of multi-source vibration angular domain and operating characteristics. By constructing a joint graph of multi-source vibration angular domain and operating characteristics, the method realizes the integrated display and synchronous analysis of multi-source vibration signals of the compressor, and effectively correlates vibration and impact characteristics with the status of key components, thereby achieving rapid and accurate fault diagnosis of reciprocating compressors.

[0006] To achieve the above objectives, the technical solution adopted by the present invention is as follows:

[0007] A fault diagnosis method for reciprocating compressors based on multi-source vibration angular domain-operating characteristic joint analysis includes the following steps:

[0008] 1) Set the signal sampling frequency to f s The vibration signals from a specific key component of each cylinder in a reciprocating compressor and the key phase signal from the crankshaft are collected synchronously and stored in matrix A. M×N In the middle, the first M-1 rows of signals are all vibration signals, and the Mth row of signals are key phase signals, where: M represents the channel number and N represents the signal sampling length;

[0009] 2) Take matrix A M×N For the Mth row of signals, the position corresponding to the trigger time of all pulses in the key phase signal is obtained by using the adaptive extraction method of the initial phase point of the key phase signal. The result is stored in the array O[k], k = 1, 2, 3, ..., C, where C represents the total number of pulse triggers.

[0010] 3) Construct the sequence of x-coordinates of the angular domain The process is as follows: Take matrix A M×N The signal in the Mth row is divided into rotation periods by the boundaries of adjacent pulse trigger positions, and the number of points in the kth rotation period is defined as L. k =O[k+1]-O[k]+1, k=1,2,3,...,C-1; for each rotation cycle, its angle sequence Through formula We obtain: By adding the corresponding offset of 360°×(k-1) to each periodic angle sequence, we construct a complete angular domain abscissa sequence.

[0011] 4) For matrix A M×NThe signals in the first M-1 rows are processed to obtain A′, and all envelope signals are truncated by key phase. The result is stored in submatrix B. The specific operation is to extract all rows from matrix A′, and the data from column O[1] to column O[C] are used to form submatrix B.

[0012] 5) Determine the phase compensation angle of each row of signals and perform compensation processing. The result is stored in matrix B′. The process is as follows: According to the cylinder number U[i] to which each row of signals belongs, i=1,2,3,...,M-1, determine the corresponding phase compensation angle and store it in vector W[i]. Perform phase compensation processing on each row of data in submatrix B according to the compensation angle listed in vector W[i] to obtain the compensated matrix B′.

[0013] 6) Draw the combined diagram of multi-source vibration angular domain and operating characteristics after phase compensation. The specific process is as follows: First, using the angular domain... Establish a coordinate system with the horizontal axis as the x-axis and the signal envelope amplitude as the y-axis; then, plot the waveforms of each row of signals in matrix B′ sequentially from top to bottom along the y-axis, with each signal arranged at equal intervals at the center of the horizontal axis, and the interval being P; finally, plot the waveforms in the angular domain. Plot the waveforms of each signal on the horizontal axis to obtain the combined diagram of multi-source vibration angular domain and working characteristics after phase compensation, and mark the component action state corresponding to the impact moment in the diagram;

[0014] 7) Using the multi-source vibration angular domain-operating characteristic combined diagram, compare and analyze the vibration and impact amplitudes of the same component under the same action, identify abnormal impacts, locate the fault location based on the angular domain phase characteristics of the abnormal impact, and analyze and evaluate the health status of the compressor.

[0015] The key components in step 1) include the crosshead slide housing, cylinder and valve cover; the crankshaft key phase signal is based on the piston of cylinder U[1] being at the cylinder inner dead center position.

[0016] The determination of the phase compensation angle W[i] in step 5) follows the following principle: the phase compensation angle W[i] is determined based on the relative position relationship between its corresponding crank and the crank of cylinder U[1], and is determined by consulting the equipment technical manual.

[0017] The formula for calculating the spacing P in step 6) is as follows:

[0018] P = ceiling(0.8 × max(B′) / 100) × 100, where the function ceiling(x) represents rounding x up.

[0019] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0020] This invention combines multi-channel vibration angular domain information with compressor operating characteristics to construct a multi-source vibration angular domain-operating characteristic joint diagram, and marks the component action state corresponding to each impact moment in the diagram, thereby establishing a precise mapping between vibration impact and the operating state of key compressor components. This invention realizes the integrated display and synchronous analysis of multi-source vibration signals of the compressor, and can quickly and accurately correlate vibration impact characteristics with the state of key components, significantly improving the efficiency and accuracy of data analysis, and providing effective technical support for reciprocating compressor fault diagnosis. Attached Figure Description

[0021] Figure 1 This is a flowchart of an embodiment.

[0022] Figure 2 This is a diagram showing the basic structure of the compressor and the arrangement of sensor measuring points in an embodiment.

[0023] Figure 3 This is a combined diagram of the multi-source vibration angular domain and working characteristics of the horizontal measuring point of the crosshead in the embodiment.

[0024] Figure 4 This is a combined diagram of the multi-source vibration angular domain and working characteristics of the vertical measuring point of the crosshead in the embodiment. Detailed Implementation

[0025] The present invention will be further described in detail below with reference to embodiments and accompanying drawings: The embodiments involve a JGC / 6 reciprocating compressor manufactured by Airel Corporation of the United States. This compressor is currently used in the pressurization stage of a natural gas extraction plant, and its simplified structural diagram is shown below. Figure 2 As shown: The main structure of this compressor consists of 6 compression cylinders, 6 crosshead housings, and 1 crankcase. The cylinders are arranged symmetrically at both ends. Each compression cylinder is equipped with 4 intake valves and 4 exhaust valves, with valve covers distributed on both sides of the cylinder. The measuring point arrangement rules are as follows: one vibration measuring point is arranged in the horizontal and vertical directions of the crosshead housing, numbered STH and STV respectively; one vibration measuring point is arranged in the horizontal radial and axial directions of the cylinders, numbered GTH and GTA respectively; one vibration measuring point is arranged at the center of each intake valve cover, and the vibration is measured from the air... The cylinder crankshaft side (inner dead center) to the cylinder head side (outer dead center) are numbered J1 to J4 sequentially. Similarly, the exhaust valve cover is numbered P1 to P4. The vibration acceleration sensor used is model CTCAC192-1D. The key phase measurement point is based on the piston of cylinder #1 being at inner dead center. Reflective tape is pasted on the surface of the crankshaft flywheel, and the photoelectric sensor is aligned with the reflective tape to collect the key phase signal. It is numbered JX, and the photoelectric sensor used is model ROS-P. During the data acquisition process, the compressor running speed is stable at 850 rpm.

[0026] Reference Figure 1A fault diagnosis method for reciprocating compressors based on a multi-source vibration angular domain-operating characteristic joint diagram includes the following steps:

[0027] 1) Set the signal sampling frequency to f s =25600Hz, synchronously acquiring vibration acceleration signals from six crosshead boxes of the reciprocating compressor and key phase signals from the crankshaft. The measuring points of the crosshead boxes are arranged into two groups: horizontal and vertical. The horizontal measuring points are numbered sequentially as 1#STH, 2#STH, 3#STH, 4#STH, 5#STH, and 6#STH, and the vertical measuring points are numbered sequentially as 1#STV, 2#STV, 3#STV, 4#STV, 5#STV, and 6#STV. The key phase is numbered JX. The signals are stored in matrix A. M×N The specific arrangement is as follows: the first 6 rows of the matrix are the horizontal vibration signals of each measuring point, the 7th to 12th rows are the vertical vibration signals of each measuring point, and the 13th row is the key phase signal. The signal sampling length is N = 51200.

[0028] 2) Take matrix A M×N The signal in the 13th row of the key phase signal is obtained by using the adaptive extraction method of the initial phase point of the key phase signal to obtain the position corresponding to the trigger time of all pulses in the key phase signal. The result is stored in the array O[k], k = 1, 2, 3, ..., 29. The result is shown in Table 1.

[0029] Table 1. Locations corresponding to pulse trigger times.

[0030] k 1 2 3 … 28 29 O[k] 733 2500 4267 … 48485 50253

[0031] 3) Construct the sequence of x-coordinates of the angular domain The process is as follows: Take matrix A M×N The signal in line 13 is divided into rotation periods by the boundaries of adjacent pulse trigger positions. Taking the first period as an example, when k=1, the number of points in this period is L1=2500-733+1=1768, and the corresponding angular domain sequence of the first rotation period is... By iterating through all k values ​​using this method, the angle sequence for each period can be obtained. Add the corresponding offset to each periodic sequence to form a complete angular domain abscissa sequence.

[0032] 4) Take matrix A M×N The signals in the first 12 rows are enveloped to obtain A′, and all envelope signals are key-phase truncated. The result is stored in matrix B. Specifically, all rows and columns 733 to 50253 are extracted from matrix A′ to form submatrix B.

[0033] 5) Determine the phase compensation angle of each row of signals and perform compensation processing. The results are stored in matrix B′. The process is as follows: According to the cylinder number U[i] to which each row of signals belongs, consult the equipment technical manual to determine the corresponding phase compensation angle W[i]. The results are shown in Tables 2 and 3 below. Perform phase compensation processing on each row of data in submatrix B according to the compensation angle listed in vector W[i] to obtain the compensated matrix B′.

[0034] Table 2. Phase compensation values ​​for each row of signals at horizontal measuring points.

[0035] i 1 2 3 4 5 6 U[i] 1 2 3 4 5 6 W[i] 0° 0° 240° 240° 120° 120°

[0036] Table 3. Phase compensation values ​​for each row of signals at vertical measurement points.

[0037] i 6 7 8 9 10 11 U[i] 1 2 3 4 5 6 W[i] 0° 0° 240° 240° 120° 120°

[0038] 6) Draw the combined diagram of multi-source vibration angular domain and operating characteristics after phase compensation. The specific process is as follows: First, using the angular domain... Establish a coordinate system with the horizontal axis as the x-axis and the signal envelope amplitude as the y-axis; then, plot the waveforms of each row of signals in matrix B′ sequentially from top to bottom along the y-axis, with the centers of each signal on the horizontal axis kept at equal intervals, with a spacing P = 200 m / s. 2 Finally, with the corner domain Plot the waveforms of each signal on the horizontal axis to obtain the combined diagram of multi-source vibration angular domain and working characteristics after phase compensation, and mark the component action state corresponding to the impact moment in the diagram; Figure 3 , Figure 4The graph displays the envelope signals of the crosshead measuring points in the horizontal and vertical directions for six cylinders within the 0–720° working cycle. The horizontal axis 0° corresponds to the piston of cylinder #1 being at inside dead center. Dashed event labels are used to mark the action states, for example: “OSC” (Outboard Stroke Crossover) indicates the crosshead moving towards the cylinder head; “ISC” (Inboard Stroke Crossover) indicates the crosshead moving towards the crankshaft; “ISVO” (Inner Suction Valve Open) indicates the crankshaft-side intake valve is open; “IDVO” (Inner Discharge Valve Open) indicates the crankshaft-side exhaust valve is open; “OSVO” (Outer Suction Valve Open) indicates the cylinder head-side intake valve is open; and “ODVO” (Outer Discharge Valve Open) indicates the cylinder head-side intake valve is open. (Open) indicates that the exhaust valve on the cylinder head side is open; it can be seen that within a complete working cycle, the envelope signal of each cylinder crosshead measuring point exhibits significant impact characteristics, especially during the intake and exhaust strokes; after phase compensation, the vibration signal of each channel has a unified physical reference (0° position represents the piston being at the inner dead center position), which means that at any angle position of each line signal, the corresponding cylinder working state is also consistent, providing convenience for quickly comparing the vibration impact of each cylinder;

[0039] 7) Utilize the multi-source vibration angular domain-operating characteristic combined diagram to compare and analyze the vibration and impact amplitudes of the same component during the same action, and identify abnormal impacts: such as... Figure 3 As shown, the horizontal measuring point (6#STH) of the crosshead showed a significant impact at a crankshaft rotation angle of 275°, with a value reaching 238.226 m / s. 2 The amplitude of the impact was more than twice that of other crossheads under the same conditions, indicating a significant abnormal impact. At the horizontal measuring point (4#STH) of crosshead #4, a high-amplitude impact of 153.734 m / s was observed at the reversing point. 2 The vibration amplitude is more than twice that of other crossheads under the same conditions, indicating that the gap between the crosshead bearing surface and the slide may have increased due to wear. This combined diagram realizes the integrated display and synchronous analysis of multi-source vibration signals of the compressor, and correlates the vibration impact characteristics with the status of key components, effectively improving the efficiency and accuracy of reciprocating compressor fault diagnosis.

Claims

1. A fault diagnosis method for reciprocating compressors based on joint analysis of multi-source vibration angular domain and operating characteristics, characterized in that, Includes the following steps: 1) Set the signal sampling frequency to f s The vibration signals from a specific key component of each cylinder in a reciprocating compressor and the key phase signal from the crankshaft are collected synchronously and stored in matrix A. M×N In the middle, the first M-1 rows of signals are all vibration signals, and the Mth row of signals are key phase signals, where: M represents the channel number and N represents the signal sampling length; 2) Take matrix A M×N For the Mth row of signals, the position corresponding to the trigger time of all pulses in the key phase signal is obtained by using the adaptive extraction method of the initial phase point of the key phase signal. The result is stored in the array O[k], k = 1, 2, 3, ..., C, where C represents the total number of pulse triggers. 3) Construct the sequence of x-coordinates of the angular domain The process is as follows: Take matrix A M×N The signal in the Mth row is divided into rotation periods by the boundaries of adjacent pulse trigger positions, and the number of points in the kth rotation period is defined as L. k =O[k+1]-O[k]+1, k=1,2,3,...,C-1; for each rotation cycle, its angle sequence Through formula We obtain: By adding the corresponding offset of 360°×(k-1) to each periodic angle sequence, we construct a complete angular domain abscissa sequence. 4) For matrix A M×N The envelope of the first M-1 rows of signals is processed to obtain A. ′ And perform key phase truncation on all envelope signals, and store the results in submatrix B. The specific operation is to extract all rows from matrix A′, and the data from column O[1] to column O[C] to form submatrix B. 5) Determine the phase compensation angle of each row of signals and perform compensation processing. The result is stored in matrix B′. The process is as follows: According to the cylinder number U[i] to which each row of signals belongs, i=1,2,3,...,M-1, determine the corresponding phase compensation angle and store it in vector W[i]. Perform phase compensation processing on each row of data in submatrix B according to the compensation angle listed in vector W[i] to obtain the compensated matrix B′. 6) Draw the combined diagram of multi-source vibration angular domain and operating characteristics after phase compensation. The specific process is as follows: First, using the angular domain... Establish a coordinate system with the horizontal axis as the x-axis and the signal envelope amplitude as the y-axis; then, plot the waveforms of each row of signals in matrix B′ sequentially from top to bottom along the y-axis, with each signal arranged at equal intervals at the center of the horizontal axis, and the interval being P; finally, plot the waveforms in the angular domain. Plot the waveforms of each signal on the horizontal axis to obtain the combined diagram of multi-source vibration angular domain and working characteristics after phase compensation, and mark the component action state corresponding to the impact moment in the diagram; 7) Using the multi-source vibration angular domain-operating characteristic combined diagram, compare and analyze the vibration and impact amplitudes of the same component under the same action, identify abnormal impacts, locate the fault location based on the angular domain phase characteristics of the abnormal impact, and analyze and evaluate the health status of the compressor.

2. The method according to claim 1, characterized in that: The key components in step 1) include the crosshead slide housing, cylinder and valve cover; the crankshaft key phase signal is based on the piston of cylinder U[1] being at the cylinder inner dead center position.

3. The method according to claim 1, characterized in that, The determination of the phase compensation angle W[i] in step 5) follows the following principle: the phase compensation angle W[i] is determined based on the relative position relationship between its corresponding crank and the crank of cylinder U[1], and is determined by consulting the equipment technical manual.

4. The method according to claim 1, characterized in that, The formula for calculating the spacing P in step 6) is as follows: P = ceiling(0.8 × max(B′) / 100) × 100, where the function ceiling(x) represents rounding x up.

Images