US20190390665A1

US20190390665A1 - Methods and systems for diagnosing valve defects in reciprocating compressors

Info

Publication number: US20190390665A1
Application number: US16/017,425
Authority: US
Inventors: Jeffrey Jacob Bizub
Original assignee: Al Alpine Us Bidco Inc; AI Alpine US Bidco Inc
Current assignee: Al Alpine Us Bidco Inc; AI Alpine US Bidco Inc
Priority date: 2018-06-25
Filing date: 2018-06-25
Publication date: 2019-12-26
Also published as: WO2020003290A1

Abstract

A diagnostic system for use with a compressor. The diagnostic system may include a sensor configured to measure a vibration signal of a compressor cylinder; and a controller having a hardware processor and a machine-readable storage medium on which is stored instructions that cause the hardware processor to execute a diagnostic process. The diagnostic process may include the steps of: storing a signature vibration data; receiving a sample vibration data from the sensor, wherein the sample vibration data is representative of the vibration signal measured by the sensor; comparing the sample vibration data to the signature vibration data to determine a similarity therebetween; and based on the determined similarity, diagnosing that the compressor cylinder comprises a condition. The signature vibration data may include resonance bands clustered within resonance band groups, each of the resonance band groups including a primary resonance band adjacent to a secondary resonance band.

Description

BACKGROUND OF THE INVENTION

The present invention relates to the diagnosis of malfunctioning components in reciprocating compressors via the detection of particular vibration signals. More specifically, but not by way of limitation, the present invention relates to the use of knock sensors for detecting certain vibrational responses in reciprocating compressors for diagnosing particular operating conditions or malfunctioning components.
Reciprocating compressors are used in many industries, including oil refineries, gas pipelines, chemical plants, natural gas processing plants and refrigeration plants. For example, to transport natural gas from production sites to consumers, pipeline operators install large compressors at transport stations along the pipelines. Natural gas pipeline networks connect production operations with local distribution companies through thousands of miles of gas transmission lines. Typically, reciprocating gas compressors are used as the prime mover for pipeline transport operations because of the relatively high pressure ratio required. Reciprocating compressors compress fluid using a piston in a cylinder connected to a crankshaft 13. Crankshaft 13 may be driven by a motor or an engine. A suction valve in the compressor cylinder receives input gas, which is then compressed by the piston and discharged through a discharge valve.
A specific challenge when using high-horsepower, high-speed, variable-speed compressors is failure of the compressor valves. A common type of valve used for reciprocating compressors is a plate-type compressor valve (or “plate valve”). These valves experience high plate impact velocities that often result in fatigue or crack failures and a short operating life, leading to frequent valve replacement. As discussed more below, such valves may exhibit a “flutter” during operation, which negatively impacts performance and may be indicative of defective components within the valve. It would be beneficial to accurately detect such malfunctions and defects so that valve components could be repaired or replaced to avoid costly failures and/or improve compressor performance. For example, a cost-effective way to detect valve flutter in reciprocating compressors may be leverage to improve performance, lower maintenance costs, and extend machine life.

BRIEF DESCRIPTION OF THE INVENTION

The present application describes a diagnostic system for use with a compressor system having a compressor cylinder. The diagnostic system may include a sensor configured to measure a vibration signal of the compressor cylinder; and a controller having a hardware processor and a machine-readable storage medium on which is stored instructions that cause the hardware processor to execute a diagnostic process. The diagnostic process may include the steps of: storing a signature vibration data; receiving a sample vibration data from the sensor, wherein the sample vibration data is representative of the vibration signal measured by the sensor over an operating period; comparing the sample vibration data to the signature vibration data to determine a similarity therebetween; and based on the determined similarity, diagnosing that the compressor cylinder has a condition. The signature vibration data may include resonance bands clustered within resonance band groups, each of the resonance band groups including a primary resonance band adjacent to a secondary resonance band.
These and other features of the present application will become apparent upon review of the following detailed description of the preferred embodiments when taken in conjunction with the drawings and the appended claims.

BRIEF DESCRIPTION OF THE DRAWINGS

These and other features of this invention will be more completely understood and appreciated by careful study of the following more detailed description of exemplary embodiments of the invention taken in conjunction with the accompanying drawings, in which:

FIG. 1 is a block diagram of a reciprocating gas compressor system with controller in accordance with the present disclosure;

FIG. 2 illustrates an exemplary plate valve;

FIG. 3 is an exemplary plot of vibration data measured by a sensor in accordance with embodiments of the present disclosure;

FIG. 4 is an exemplary plot of vibration data measured by a sensor showing a signature of a fluttering valve in accordance with embodiments of the present disclosure; and

FIG. 5 is a flow chart of an exemplary embodiment of a diagnostic process suitable for analyzing vibration data for the detection of a malfunctioning plate valve.

DETAILED DESCRIPTION OF THE INVENTION

One or more specific embodiments of the present invention will be described below. In an effort to provide a concise description of these embodiments, all features of an actual implementation may not be described in the specification. It should be appreciated that in the development of any such actual implementation, as in any engineering or design project, numerous implementation-specific decisions must be made to achieve the developers' specific goals, such as compliance with system-related and business-related constraints, which may vary from one implementation to another. Moreover, it should be appreciated that such a development effort might be complex and time consuming, but would nevertheless be a routine undertaking of design, fabrication, and manufacture for those of ordinary skill having the benefit of this disclosure. As will be seen, the techniques described herein include systems and methods in which knock sensors and the like are used to detect a dynamic response of engine components during the operations of a reciprocating compressor. For example, signals from a knock sensor may be collected and analyzed for diagnosing health and operational fitness related plate valves.
Turning to the drawings, FIG. 1 is a block diagram of the basic elements of a reciprocating gas compressor system (or “compressor system”) 10. The elements of compressor system 10 are depicted as those of a typical system, and include an engine or driver 11, compressor cylinders 12, filter bottles 18, suction and discharge piping connections, controller 14, and sensor 19. In the example of FIG. 1, compressor system 10 has three compressor cylinders 12. In practice, compressor system 10 may have fewer or more such cylinders 12. Compressor valves (not explicitly visible in FIG. 1) are installed within each cylinder 12 to permit one-way flow into or out of the cylinder volume. Driver 11, for example, may be an internal combustion engine. Filter bottles 18 are placed between the compressor and the lateral piping, on the suction or discharge side or on both sides. In the example of FIG. 1, a suction side filter bottle 18 a and a discharge side filter bottle 18 b are included. As will be appreciated, filter bottles, such as those shown, are installed as a common method for pulsation control.
Controller 14 is used to monitor and control operating parameters affecting compressor load and capacity, as well as other functions that will be discussed more below. As indicated, controller 14 may be equipped with a hardware processor 15 and memory 16 (e.g., non-transitory computer readable medium), as well as appropriate input and output devices, and a user interface. Controller 14 may be programmed to perform the various control tasks and deliver control parameters to compressor system 10. Given appropriate input data, output specifications, and control objectives, algorithms for programming controller 14 may be developed and executed. Controller 14 may electronically communicate with sensor 19 and receive data therefrom.
According to exemplary embodiments, sensor 19 is one suitable for detecting noise or vibrational response within compressor system 10 and/or compressor cylinder 12 during operation. For example, sensor 19 may be a conventional knock sensor. More generally, sensor 19 may be any other conventional sensor configured to sense vibration, sound, acceleration, and/or movement caused by the operation of compressor cylinder 12, for example, a Piezo-electric accelerometer, microelectromechanical system sensor, Hall effect sensor, magnetostrictive sensor, etc. As will be seen, sensor 19 may be used to detect a variety of noise or vibration signals for comparison to a signature of vibration data (or “signature vibration data”) that, in accordance with embodiments of the present invention, has been found to correlate to particular valve condition or defect. For example, in one embodiment, a current vibration signal or data sample may be analyzed to determine if it contains the signature vibration data, which then is used to determine whether the corresponding defect or malfunction is likely present within compressor cylinder 12. To do this, sensor 19 may be disposed within compressor system 10. For example, according to preferred embodiments, sensor 19 is attached to compressor cylinder 12. Because of the percussive nature of valve operation within compressor cylinder 12, sensor 19 may be capable of efficiently detecting and measuring vibration signals when positioned in this way. In some embodiments, a single sensor 19 may be used, whereas, in other embodiments, each compressor cylinder 12 may include one or more sensors 19. Sensor 19 is shown communicatively coupled to controller 14. During operations, vibration data representative of the vibration signal measured by sensor 19 is communicated to controller 14 for analysis thereby.
FIG. 2 illustrates a conventional plate valve 20 that is commonly used in the compressor cylinders of reciprocating compressors. In general, plate valves include a plate that moves between a seat and guard, which causes channels to alternatively open and close. This movement generally results in high impact velocities between the plate and seat or plate and guard, and this may cause fatigue or crack failures in the plate and, thereby, shorten life and lead to frequent valve replacement. As will be discussed more below, the plate of such valves may “float” or “flutter” at is opens or closes. This defect in operation compromises performance and/or signals a degraded component within the valve. It would be beneficial to accurately detect such malfunction so that valve components could be repaired or replaced to improve performance and avoid component failure. As will be seen, the present disclosure includes a cost-effective way to accurately detect valve flutter during operation of reciprocating compressors.
In the example of FIG. 2, plate valve 20 includes a valve seat 21 and guard 22, which each has passage channels, and one or more plates, which are moveable between seat 21 and guard 22 to control flow through the valve. Though other configurations are possible, in the example shown, valve 20 has two plates: a sealing plate 23; and an optional damping plate 26. Each of the sealing plate 23 and damping plate 26 may include passage channels that are opposed by corresponding sealing surfaces and passages on valve seat 21 and guard 22, respectively, so that when valve 20 operates, fluid will flow or not flow through valve 20 depending on the position of sealing plate 23. For example, when valve 20 is closed, the passages through sealing plate 23 is sealed or closed by a corresponding sealing surface on seat 21. Springs 24 are used to hold sealing plate 23 in a closed position, and fluid flow direction determines whether fluid will flow or not flow through the valve. As shown, a stem may be used to move and/or guide sealing plate 23 against the closure force of springs 24 to open valve 20.
In an example of operation, plate valve 20 begins in a closed state. In this state, fluid flow in one direction is prevented as sealing plate 23 is held by springs 24 against sealing surfaces of seat 21. When the differential of pressures acting upon the two sides of sealing plate 23 overcomes the force of springs 24 holding it against seat 21, valve 20 begins to open as sealing plate 23 dislodges from seat 21, which allows gas to start flowing through the passages of sealing plate 23. Opening further, sealing plate 23 continue its movement away from seat 21 until such movement is prevented when sealing plate 23 impacts guard 22. It is desirable for sealing plate 23 to settle upon guard 22 so that valve 20 quickly attains a fully open state. However, for a variety of reasons, the impact with guard 22 may result in sealing plate 23 bouncing or fluttering so that repeated impacts, which generally decrease in size, occur before sealing plate 23 settles against guard 22 and the fully open state is achieved. The process then reverses when the differential of pressures acting upon the two sides of sealing plate 23 is overcome by the force of springs 24. When this happens, sealing plate 23 dislodges and moves away from guard 22. This movement is then finally arrested when sealing plate 23 impacts seat 21. As before, it is desirable for sealing plate 23 to quickly settle against seat 21 so that the fully closed state is quickly attained. However, the impact against seat 21 also may result in sealing plate 23 bouncing or fluttering before properly coming to rest.
In this way, sealing plate and the associated springs within a plate valve constitute a potentially oscillating system, with the excitation necessary to give rise to such oscillations being provided by the interaction of this system with the flow and the alternating impacts occurring between the sealing plate and the seat/guard components. When the oscillation of the sealing plate is excessive, it is referred to as “valve flutter” or, simply, “flutter”. Flutter is an undesirable condition as it can lead to premature valve plate failures, degraded seals, and poor performance. Further, flutter often is an indicator that components within the plate seal are not functioning properly due to some defect or degradation. For example, flutter within such valves can indicate that springs are worn or not sized properly or that the seal plate is cracked or excessively fatigued. As will be appreciated, early diagnosis of such flutter is advantageous in that corrective action can be taken to improve performance and avoided component failure.
Turning now to FIG. 3, a plot 80 is shown, which may be derived by controller 14 using vibration data measured by sensor 19 mounted to cylinder 12 of reciprocating compressor system 10. Specifically, plot 80 includes an amplitude curve 84 showing vibration data representative of signal amplitude measured via sensor 19, in which x-axis 81 is time and y-axis 82 is vibration amplitude. In general, amplitude curve 84 shows a substantially regular signal amplitude over the time period of the plot. Further included within plot 80 is a plate position curve 86, which indicates actual movement or displacement of sealing plate 23 during operation, for example, as sealing plate 23 moves from a closed state 88, which is represented by the substantially flat lower portions of plate position curve 86, to an open state 89, which is represented by the raised, slightly rounded portions of plate position curve 86. In general, as will be appreciated, the depicted movement of sealing plate 23 in curve 86 is characteristic of a properly functioning valve. That is, plate position curve 86 shows that sealing plate 23 generally comes to rest quickly after the impacts associated with transitions between closed and open states. In other words, sealing plate 23 is not bouncing or fluttering on impact.
In regard to FIG. 3, it is noteworthy that amplitude curve 84 remain substantially regular as plate valve transitions between closed and open states. In accordance with the present disclosure, it has been found that this type of regular amplitude data reliably correlates to a plate valve that is properly functioning, for example, not exhibiting flutter. As provided herein, amplitude curve 94, thus, represents a test result that provides information as to the current health and performance of a plate valve. According to exemplary embodiments, such results may be cost-effectively obtained via a single knock sensor sampling vibrational data while the compressor operates. Thus, the testing and diagnostic systems and methods described herein do not require compressor shut down to function.
Referring now to FIG. 4, a plot 90 includes another amplitude curve—an amplitude curve 94—that, in accordance with the present disclosure, includes an exemplary signature vibration data of a malfunctioning or “fluttering” sealing plate. Similar to the above, amplitude 94 is a plot of vibration data measured by sensor 19, in which x-axis 91 is time and y-axis 92 is amplitude. Superimposed on amplitude curve 94 is a plate position curve 96 showing actual valve movement, for example, as sealing plate 23 moves from a closed state 98, which is represented by the substantially flat, lower portions of plate position curve 96, to an open state 99, which is represented by the raised, jagged portions of plate position curve 96. As can be seen, unlike the curve above, open state 99 of plate position curve 96 is not smooth, but jagged and highly variable. Specifically, instead of being smoothly rounded as before, each open state 99 within plate position curve 96 shows that sealing plate 23 is bouncing between elevations as plate valve 20 opens. As will be appreciated, this type of movement is indicative of valve flutter. That is, sealing plate 23 is not desirably coming to rest upon impacting guard 22, but instead, is bouncing or fluttering before finally coming to rest. As already described, such sealing plate flutter negatively impacts the performance of the plate valve and/or signals that the plate valve may have a defective component, such as a cracked sealing plate or degraded springs.
According to the present disclosure, certain aspects of amplitude curve 94 of FIG. 4 can be used to diagnose a malfunctioning plate valve, in particularly, whether the sealing plate of a compressor valve is fluttering. For example, characteristics of the amplitude curve 94 may be used as a signature that, if present in samples of current operation, indicate likely flutter within the plate valve. The characteristics of this signature (or “signature vibration data”) will now be discussed.
A first of these characteristics is the presence of what will be referred to herein as resonance bands. Instead of the more regular amplitude data shown in FIG. 3, the amplitude data of FIG. 4 shows pronounced resonance bands 100, which occur at regular intervals. As used herein, such resonance bands are defined as a period within the plot of increased amplitude, for example, where vibration amplitude increases rapidly to an approximate peak and then falls away sharply. The signature vibration data of the present disclosure may include multiple resonance bands 100, which may be spaced and grouped within the curve in particular ways. For example, the signature vibration data may include resonance bands 100 clustered within a grouping of such resonance bands (or “resonance band group”) 104. The signature vibration data may include multiple, regularly spaced resonance band groups, where each of the resonance band groups 104 include a primary resonance band adjacent to a secondary resonance band. According to preferred embodiments, as the example of FIG. 4 shows, each of resonance band groups 104 may include three such resonance bands 100. In such cases, a center resonance band 101 is flanked by two lesser or secondary resonance bands, which may be specifically referred to as a leading resonance band 102 and trailing resonance band 103. The resonance bands within each resonance band group 104 may be positioned next or adjacent to each other. Additionally, within each grouping, center resonance band 101 generally has a greater amplitude than both leading and trailing resonance band 102, 103, whereas the amplitudes of leading and trailing resonance bands 102, 103 may be approximately similar. According to the present disclosure, these types of resonance bands 100, including the arrangement or grouping thereof, have been found to correlate to (and provide a data signature for) flutter occurring with a plate valve of a reciprocating compressor.
Further, between occurrences of resonance band groups 104, the signature vibration data may have an intervening signal 95, which is characterized by a more regular amplitude, which is significantly lower than an amplitudes of the included resonance bands 100. That is, between resonance bands groups 104, the amplitude of the vibration data may return to a more regular pattern, such as that shown in FIG. 3.
When amplitude curve 94 is compared to plate position curve 96, it is seen that resonance bands 100 generally coincide with the transition from closed state 98 to open state 99 and that resonance bands 100 occur primarily during open state 99. It should be appreciated that, depending on the behavior of the system, a sealing plate could exhibit fluttering when transitioning from open state 99 to closed state 98 (i.e., when the sealing plate impacts the seat during closing, instead of the guard during opening). In such cases, resonance bands 100 and resonance band groups 104 would coincide with this transition and, thus, would occur primarily within closed state 98 of the valve. It should be understood that the techniques described herein are applicable to diagnosing flutter in either situation.
FIG. 5 is a flow chart depicting a diagnostic process 110 suitable for gathering and analyzing vibrational data to diagnose compressor valve flutter, or, at least, determine that an increased likelihood of valve flutter is present. Diagnostic process 110, for example, may be implemented via a computer system that includes a hardware processor and machine-readable storage medium, such as controller 14. The storage medium may include instructions that cause the hardware processor to execute diagnostic process 110. Further, a sensor, such as sensor 19, may be configured to measure a vibration signal of a compressor cylinder and transmit such data to the controller.
For example, according to an embodiment of the present disclosure, diagnostic process 110 may include an initial step 115 at which a signature vibration data is stored, for example, within memory 16 of controller 14. The signature vibration data, for example, may include resonance bands clustered within resonance band groups, as already described. At a step 120, process 110 may include receiving, for example, at controller 14, a sample vibration data from sensor 19, where the sample vibration data is representative of the vibration signal measured by sensor 19 over a particular operating period. At a step 125, diagnostic process 110 may include comparing the sample vibration data against the signature vibration data to determine a similarity therebetween. Then, at a step 130, based on the determined similarity, diagnostic process 110 may include diagnosing that the compressor cylinder does or does not have a particularly condition. As described in detail above, the condition may be valve flutter or flutter, e.g., the bouncing or fluttering of a sealing plate within a plate valve. Alternatively, the diagnosed condition may be that there is an increased likelihood that the sealing plate is fluttering within the plate valve.
At a step 135, diagnostic process 110 may further include providing an alert regarding the diagnosed condition. For example, controller 14 may prepare an electronic communication regarding valve flutter or increased likelihood of valve flutter and the particular compressor cylinder for which the test results apply. The electronic communication, such as email, text, and the like, may then be transmitted to a user device (e.g., computer, phone, tablet) of a specified person, such as an operator, associated with the maintenance of the compressor system.
Technical effects of the invention include detecting vibrations via certain sensors, such as knock sensors, within compressor cylinders and diagnosing valve defects and actions related to such diagnoses, i.e., the sending of alerts. It should be understood that the many varying features and configurations described above in relation to the several exemplary embodiments may be further selectively applied to form other possible embodiments of the present invention. For the sake of brevity and taking into account the abilities of one of ordinary skill in the art, all possible iterations are not provided or discussed in detail, though all combinations and possible embodiments embraced by the several claims below or otherwise are intended to be part of the present application. For instance, features illustrated or described as part of one embodiment may be used on another embodiment to yield a still further embodiment. In addition, from the above description of several exemplary embodiments of the invention, those skilled in the art will perceive improvements, changes and modifications. Such improvements, changes and modifications within the skill of the art are intended to be covered by the appended claims. Numerous changes and modifications may be made to the exemplary embodiments provided herein without departing from the spirit and scope of the application as defined by the following claims and the equivalents thereof.

Claims

That which is claimed:

1. A diagnostic system for use with a compressor system having a compressor cylinder, the diagnostic system comprising:

a sensor configured to measure a vibration signal of the compressor cylinder; and

a controller comprising a hardware processor and a machine-readable storage medium on which is stored instructions that cause the hardware processor to execute a diagnostic process, wherein the diagnostic process includes the steps of:

storing a signature vibration data;

receiving a sample vibration data from the sensor, wherein the sample vibration data is representative of the vibration signal measured by the sensor over an operating period of the compressor cylinder;

comparing the sample vibration data to the signature vibration data to determine a similarity therebetween; and

based on the determined similarity, diagnosing that the compressor cylinder comprises a condition;

wherein the signature vibration data comprises resonance bands clustered within resonance band groups, each of the resonance band groups including a primary resonance band adjacent to a secondary resonance band.

2. The diagnostic system according to claim 1, wherein the signature vibration data comprises multiple ones of the resonance band groups occurring at substantially regular intervals; and

wherein each of the sample vibration data and the signature vibration data comprises vibration amplitude plotted against time.

3. The diagnostic system according to claim 2, wherein the sensor comprises a knock sensor.

4. The diagnostic system according to claim 3, wherein the sensor is attached to the compressor cylinder; and

wherein the compressor cylinder is diagnosed as comprising the defect when the degree of similarity is determined to be beyond a predetermined threshold.

5. The diagnostic system according to claim 3, wherein, between occurrences of the resonance band groups, the signature vibration data comprises an intervening signal, the intervening signal comprising a substantially regular amplitude that is significantly lower than an amplitude of each of the resonance bands within the resonance band groups; and

wherein the resonance bands are each defined as a period in which the vibration amplitude increases rapidly to a peak and then falls away sharply.

6. The diagnostic system according to claim 5, wherein each of the resonance band groups include a one of the primary resonance band positioned adjacent to and between ones of the secondary resonance bands such that:

a leading secondary resonance band leads the primary resonance band; and

a trailing secondary resonance band trails the primary resonance band.

7. The diagnostic system according to claim 6, wherein the compressor cylinder comprises a valve;

wherein the condition comprises flutter within the valve.

8. The diagnostic system according to claim 6, wherein the compressor cylinder comprises a plate valve, the plate valve including a sealing plate moveable between a seat and guard for controlling a flow through the plate valve.

9. The diagnostic system according to claim 8, wherein the condition comprises an increased likelihood that the sealing plate is fluttering.

10. The diagnostic system according to claim 6, wherein the diagnostic process further comprises the steps of:

preparing an electronic communication regarding the diagnosis of the condition; and

transmitting the electronic communication as an alert to a user device of an operator of the compressor system.

11. A computer-implemented method related to diagnosing a condition within a compressor system having a compressor cylinder, wherein a sensor is configured to measure a vibration signal of the compressor cylinder, the method comprising:

comparing the sample vibration data to a signature vibration data to determine a similarity therebetween; and

12. The method according to claim 11, wherein the signature vibration data comprises multiple ones of the resonance band groups occurring at substantially regular intervals; and

13. The method according to claim 12, wherein the sensor comprises a knock sensor.

14. The method according to claim 13, wherein the sensor is attached to the compressor cylinder; and

15. The method according to claim 13, wherein, between occurrences of the resonance band groups, the signature vibration data comprises an intervening signal, the intervening signal comprising a substantially regular amplitude that is significantly lower than an amplitude of each of the resonance bands within the resonance band groups; and

16. The method according to claim 15, wherein each of the resonance band groups include a one of the primary resonance band positioned adjacent to and between ones of the secondary resonance bands such that:

a leading secondary resonance band leads the primary resonance band; and

a trailing secondary resonance band trails the primary resonance band.

17. The method according to claim 16, wherein the compressor cylinder comprises a valve;

wherein the condition comprises flutter within the valve.

18. The method according to claim 16, wherein the compressor cylinder comprises a plate valve, the plate valve including a sealing plate moveable between a seat and guard for controlling a flow through the plate valve.

19. The method according to claim 18, wherein the condition comprises an increased likelihood that the sealing plate is fluttering.

20. The method according to claim 16, further comprising the steps of: