A chemical process machine pump fault early warning system

Abstract

The present application belongs to the technical field of chemical process pumps, and particularly relates to an early warning system for chemical process pump failures, comprising multiple sets of pumps equipped with sound collection devices, signal relay stations and a signal receiving central station; the sound collection device comprises a sound collection end, which is arranged in the pump; the sound collection device can convert real-time sound signals collected by the sound collection end into initial sound data and send them; the signal receiving central station comprises a signal receiver and an upper computer, which is provided with a data processing system and an early warning judgment system. Compared with the prior art, the present application has the following advantages: it can collect sound information emitted by bearings during operation in a directional manner, has accuracy and timeliness, and can provide data support and basis for predictive maintenance; by using the principle of frequency spectrum analysis and the early warning judgment system, it can provide early warning in the early stage of failure, quickly determine the failure type by analyzing the frequency spectrum state, shorten the maintenance time and reduce the maintenance cost.

Description

Technical Field

[0001] This invention belongs to the field of chemical process pump technology, specifically relating to an early warning system for chemical process pump failures. Background Technology

[0002] In large-scale industrial production fields such as petroleum and petrochemicals, metal smelting, and hydropower generation, pumps play a role in liquid transmission and pressurization. With the extensive use of pumps, it is crucial to monitor them to ensure their safe, long-term, and stable operation, which can effectively prevent production delays and major safety accidents.

[0003] For pump monitoring in the petrochemical industry, the main methods currently used are vibration and temperature to monitor whether the bearings are functioning properly. However, both of these methods are lagging, and vibration and temperature can usually only detect the wear when the bearings are severely worn. If the bearing wear is not detected and replaced in time, it will increase the maintenance cost and difficulty of the pumps, resulting in significant economic losses.

[0004] Existing technologies lack real-time sound detection and analysis systems for pump operation, relying mostly on handheld sound probes and other mechanical tools. During detection, the subjectivity of personnel leads to drawbacks such as inconsistent monitoring points, inconsistent distance measurements, and insensitivity to sound, preventing the formation of an effective data chain. Furthermore, data cannot be saved, hindering the establishment of an effective early warning system to support predictive maintenance. Application number 2021211495395 discloses a fault early warning device for thermal power plant pumps based on the Industrial Internet. This device monitors and alarms the internal temperature and volume of the pump by installing a housing outside the pump body and placing volume and temperature sensors inside. However, in practical applications, the large outer casing of pumps in large industrial production areas is unsuitable for a relatively enclosed "housing" environment, inevitably resulting in significant noise pollution. Therefore, simply detecting the volume level is insufficient for providing early warnings for the pump. Summary of the Invention

[0005] The purpose of this invention is to address the problem that the failure of bearings in existing large pumps and machinery often increases maintenance costs and difficulty during operation, and to provide an early warning system for pump and machinery failures in chemical processes.

[0006] This invention is achieved through the following technical solution: an early warning system for pump failure in a chemical process, comprising multiple pumps equipped with sound acquisition devices, a signal relay station, and a signal receiving station; the sound acquisition device includes a sound acquisition end, which is installed inside the pump; the sound acquisition device can convert the real-time sound signal acquired by the sound acquisition end into initial sound data and send it.

[0007] The signal relay station includes a signal relay module, which is used to receive the initial sound data sent by the sound acquisition device and send it to the signal receiving main station;

[0008] The signal receiving station includes a signal receiver and a host computer. The signal receiver receives the initial sound data transmitted by the signal relay station and transmits it to the host computer. The host computer is equipped with a data processing system and an early warning judgment system. The data processing system processes the initial sound data to obtain real-time noise data, stores it, and simultaneously inputs the real-time noise data into the early warning judgment system.

[0009] The real-time noise data here uses Hilbert transform to analyze the noise volume envelope and obtain a noise volume curve. When the pump is running normally, the noise volume curve is relatively stable and can be used as a benchmark. When the curve changes abnormally, a corresponding warning is given by comparing it with the benchmark. In addition, the fault analysis system is used to analyze the fault type.

[0010] Specifically, the pump includes a pump casing, a rotating shaft, an impeller, and a guide vane. The pump casing has a cavity, and a through-hole is provided on the side wall of the pump casing. The central axis of the through-hole is perpendicular to the axis of the rotating shaft.

[0011] The sound acquisition device includes a housing with a cover screwed onto it. A sound transmission tube is rotatably connected to the bottom of the housing. A probe is provided on one side of the lower end of the sound transmission tube. A sound sensing microphone, a signal processing and transmission module, and a power supply module are provided inside the housing. A signal receiver is provided outside the cover. The signal processing and transmission module is electrically connected to the sound sensing microphone and the signal receiver, respectively. The upper part of the sound transmission tube is snapped into a through-hole, and the lower part of the sound transmission tube and the probe are located in a cavity.

[0012] The sound transmission tube is rotatably connected to the housing and can be rotated to a predetermined position before being limited. The sound source has a certain directionality. Rotating the probe to a suitable angle can maximize the acquisition of the sound information to be detected, while effectively attenuating noise from other directions, ensuring the effectiveness of the sound acquisition device in acquiring sound and reducing the difficulty of data processing.

[0013] Specifically, the upper end of the sound propagation tube is equipped with a sound intensity control valve, which can adjust the inner diameter of the sound propagation tube and adjust the intensity of the propagated sound by setting the sound intensity control valve;

[0014] The wall thickness of the portion of the sound transmission tube located inside the through-hole is greater than the wall thickness of the lower part of the sound transmission tube. A sealing bushing is provided between the outer wall of the sound transmission tube and the inner wall of the through-hole to ensure a sealing effect and reduce the interference of pump body vibration on the sound inside the sound transmission tube.

[0015] Specifically, the housing is provided with a mounting bracket, on which a signal processing and transmission module and a power supply module are fixed, wherein the power supply module is used to supply power to the various electrical components of the sound acquisition device.

[0016] Specifically, the early warning judgment system compares real-time noise data with preset noise data. If the noise exceeds 20-60% of the preset noise data, an intermittent alarm is triggered; if it exceeds 60-120%, a continuous alarm is triggered. The thresholds for intermittent and continuous alarms can be adjusted according to actual needs. The alarm methods can be various, including but not limited to upper computer pop-up alarms, siren alarms, telephone and SMS alarm lights, etc.

[0017] Among them, preset noise data Real-time noise data The calculation formula is as follows:

[0018] (1)

[0019] (2)

[0020] in, The average value of the noise volume after integrating from t1 to t2 is taken for the part with good noise conditions. It is t n to t n+1 The average noise volume after integration over a time period, where t1 is the base sampling start time and t2 is the base sampling end time. n For real-time sampling start time, t n+1 This is the end time of real-time sampling. The sampling time interval is set manually.

[0021] Specifically, the host computer is also equipped with a fault analysis system. When the early warning judgment system detects an alarm signal, it identifies the system as being in an abnormal working state; if no alarm signal is detected, it identifies the system as being in a normal working state. The initial sound data collected during the abnormal working state is processed to obtain a fault generation spectrum, which is then compared with a fault analysis database to determine the fault type. The method for constructing the fault analysis database includes the following:

[0022] S1: Under normal working conditions, the initial sound data is randomly sampled as a reference sample, and the reference sample is decomposed to obtain the spectrum of the reference sample.

[0023] In abnormal working state, the initial sound data is randomly sampled in the early stage and the later stage to obtain the initial abnormal sample and the later abnormal sample, respectively. The initial abnormal sample and the later abnormal sample are decomposed to obtain the spectrum diagram of the initial abnormal sample and the spectrum diagram of the later abnormal sample.

[0024] S2: The frequencies of the spectrum diagrams of the initial abnormal samples and the spectrum diagrams of the reference samples are arithmetically subtracted to obtain the initial residual spectrum diagram, which is used as the fault generation spectrum diagram.

[0025] The frequency of the spectrum of the later abnormal sample and the spectrum of the reference sample are arithmetically subtracted to obtain the remaining spectrum of the later sample.

[0026] S3: Subtract the frequencies of the later remaining spectrum map and the initial remaining spectrum map arithmetically to obtain the external interference spectrum map;

[0027] S4: Input the fault generation spectrum diagram and corresponding faults into the fault analysis system to establish a fault analysis database.

[0028] Specifically, the decomposition of the baseline sample, initial abnormal sample, and later abnormal sample is based on a sampling frequency f = 1 / , The sampling time interval is set manually.

[0029] According to the Nyquist sampling theory, in order to ensure that the sound is not distorted, the sampling frequency should be around 40kHz. The sampling frequency refers to the number of times the sound wave amplitude is sampled per second when digitizing an analog sound wave.

[0030] The early warning and judgment system can issue alarms of different levels based on real-time noise data collected by the sound acquisition device, so as to facilitate early detection and troubleshooting.

[0031] The fault analysis system can establish a fault analysis database based on the spectrum. An abnormal sample spectrum diagram corresponds to a corresponding fault, which facilitates early detection and identification of fault types, reducing maintenance costs and difficulties.

[0032] The acquisition end of the sound acquisition device includes a sound transmission tube and a probe.

[0033] Each characteristic frequency of the signal has a corresponding rotating component. By analyzing the amplitude and phase information of each frequency of the signal, the operating status of the equipment can be determined. By comparing the changes in the values ​​of the same frequency components or the appearance of new frequency components, the type of equipment fault, the cause of the fault, and the degree of fault deterioration can be determined.

[0034] The present invention has the following advantages over the prior art:

[0035] (1) The sound acquisition end of the sound acquisition device is set inside the pump casing. The sound propagation tube drives the probe to rotate in a specific direction, which can collect the sound information emitted when the bearing is running in a specific direction. It has accuracy and timeliness, and can attenuate noise from other directions, reduce the difficulty of data processing, and provide data support and basis for predictive maintenance.

[0036] (2) It solves the problems of lag, large detection error, insensitive data and inability to effectively establish an early warning system in manual probe monitoring and vibration temperature detection; by cooperating with the host computer and multiple pumps equipped with sound acquisition devices, it is convenient to monitor the pump operation status in real time.

[0037] (3) Using the principle of spectrum analysis, the initial spectrum characteristics of the fault are extracted, recorded and a fault analysis database is created. Through the early warning judgment system, early warning can be given in the early stage of the fault, and the fault type can be quickly determined by analyzing the spectrum status, thus shortening the maintenance time and reducing the maintenance cost. Attached Figure Description

[0038] Figure 1 This is a schematic diagram of the structure of the present invention.

[0039] Figure 2 This is a schematic diagram of the installation structure of the sound acquisition device.

[0040] Figure 3 This is a cross-sectional view of the sound acquisition device.

[0041] Figure 4 This is a schematic diagram of the envelope of noise volume.

[0042] Figure 5 It is a noise volume curve.

[0043] Figure 6 This is a diagram illustrating the division between normal and abnormal states.

[0044] Figure 7 This is the spectrum of the benchmark sample, Pn.

[0045] Figure 8 This is the spectrum diagram Pmn of the initial abnormal sample.

[0046] Figure 9 It is the spectrum diagram Psn of the abnormal sample in the later stage.

[0047] Figure 10 This is the initial remaining spectrum diagram.

[0048] Figure 11 It is the remaining spectrum diagram in the later stage.

[0049] Figure 12 This is a spectrum diagram of external interference.

[0050] Among them, 1-pump, 11-pump casing, 12-cavity, 13-shaft, 14-through hole, 2-sound acquisition device, 21-sound acquisition end, 211-sound propagation tube, 2111-upper part, 2112-lower part, 212-probe, 213-sound intensity control valve, 22-shell, 23-cover, 24-mounting bracket, 25-signal processing and transmission module, 26-power supply module, 27-microphone, 28-signal transceiver, 3-signal relay station, 4-signal receiving station, 41-signal receiver, 42-host computer;

[0051] Figures 7-12 The vertical axis represents the energy values ​​at various frequencies after decomposition. Detailed Implementation

[0052] The invention will be further described below with reference to the accompanying drawings.

[0053] 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 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 are within the scope of protection of the present invention. Example 1

[0054] like Figure 1-3 As shown, an early warning system for pump failure in a chemical process includes multiple pumps 1 equipped with sound acquisition devices 2, a signal relay station 3, and a signal receiving station 4. The sound acquisition device 2 includes a sound acquisition end 21, which is disposed inside the pump 1. The sound acquisition device 2 can convert the real-time sound signal acquired by the sound acquisition end into initial sound data and send it. The pump 1 includes a pump casing 11, a rotating shaft 13, an impeller, and a guide vane. The pump casing 11 has a cavity 12, wherein a through-hole 14 is provided on the side wall of the pump casing 11, and the central axis of the through-hole 14 is perpendicular to the axis of the rotating shaft 13.

[0055] The sound acquisition device 2 includes a housing 22, with a cover 23 screwed onto the housing 22. A mounting bracket 24 (the specific structure is not limited here) is provided inside the housing 22. A signal processing and transmission module 25 and a power supply module 26 are fixed on the mounting bracket 24, wherein the power supply module 26 supplies power to the various electrical components of the sound acquisition device 3. A sound transmission tube 211 is rotatably connected to the bottom of the housing 22. A sound sensing microphone 27 is located near the top of the sound transmission tube 211 at the bottom of the housing. A probe 212 is located on one side of the lower end of the sound transmission tube 211. A signal transceiver 28 is located outside the cover 23. The signal processing and transmission module 25 is electrically connected to the sound sensing microphone 24 and the signal transceiver 28, respectively. The upper part 2111 of the sound transmission tube 211 is snapped onto... Within the through-hole 14, the lower part 2112 of the sound propagation tube 211 and the probe 212 are located within the cavity 12. A sound intensity control valve 213 is provided at the upper end of the sound propagation tube 211. This valve can be a pipe clamp, located outside the pump, and can adjust the inner diameter of the sound propagation tube 211 according to actual production needs, thereby adjusting the intensity of the propagated sound. The sound propagation tube 211 is rotatably connected to the housing 22, and can be rotated to a predetermined position for limiting. In practical applications, quick-release clamps used for pipe fittings can be used for limiting. The sound source has a certain directionality; rotating the probe 212 to a suitable angle allows for the maximum acquisition of the sound information to be detected, while effectively attenuating noise from other directions, ensuring the effectiveness of the sound acquisition device 2 in acquiring sound and reducing the difficulty of data processing.

[0056] The signal relay station 3 includes a signal relay module, which is used to receive the initial sound data sent by the sound acquisition device 2 and send it to the signal receiving station 4.

[0057] The signal receiving station 4 includes a signal receiver 41 and a host computer 42. The signal receiver 41 receives the initial sound data transmitted by the signal relay station 3 and transmits it to the host computer 42. The host computer 42 is equipped with a data processing system, an early warning judgment system and a fault analysis system. The data processing system performs calculations on the initial sound data to obtain real-time noise data and stores it. At the same time, it inputs the real-time noise data into the early warning judgment system.

[0058] The data noise processing system, such as Figure 4 As shown, the real-time noise data collected by the sound acquisition device 2 is analyzed using the Hilbert transform to extract the noise volume envelope. After calculation, as shown... Figure 5 The noise volume curve is shown below;

[0059] Figure 5 In the middle, preset noise data Real-time noise data The calculation formula is as follows:

[0060] (1)

[0061] (2)

[0062] in, The average value of the noise volume after integrating from t1 to t2 is taken for the part with good noise conditions. It is t n to t n+1 The average noise volume after integration over a time period, where t1 is the base sampling start time and t2 is the base sampling end time. n For real-time sampling start time, t n+1 This is the end time of real-time sampling. The sampling time interval is set manually; The vibration threshold value is set manually.

[0063] The content of the early warning judgment system is to process real-time noise data. Compared with preset noise data The comparison calculation yields the excess percentage B. If B > 30%, an intermittent alarm is triggered; if B > 100%, a continuous alarm is triggered. The percentage B value can be adjusted based on actual data statistics.

[0064] The formula for calculating the percentage exceeding B is:

[0065] (3)

[0066] The specific alarm methods can take many forms, including but not limited to pop-up alarms on the host computer, alarm sirens, telephone and SMS alarm lights, etc.; intermittent alarms indicate a relatively low alarm level, while continuous alarms indicate a higher alarm level, which can only be stopped after confirmation by staff.

[0067] When the real-time noise data input early warning judgment system determines that the data is abnormal, it can analyze the fault type according to the fault analysis system. The working principle of the fault analysis system is to process the initial sound data collected in the abnormal working state to obtain the fault generation spectrum, and compare it with the fault analysis database to obtain the fault type.

[0068] The method for constructing the fault analysis database includes the following:

[0069] S1: As Figure 6 As shown, under normal operating condition C1, this curve segment is stable. Random sampling of the initial sound data is used as a reference sample, and the reference sample is then decomposed to obtain the following... Figure 7 The spectrum of the reference sample Pn is shown, where the various frequencies after decomposition are marked as f on the coordinate line.n ;

[0070] During abnormal operating state C2, this section of the curve exhibits abnormal changes. Random sampling of the initial sound data was performed at the beginning and end of the process to obtain initial and later abnormal samples, respectively. These initial and later abnormal samples were then decomposed to obtain the following results: Figure 8 The spectrum of the initial abnormal sample Pmn and as shown Figure 9 The spectrum of the later abnormal sample is shown in Psn.

[0071] In the initial spectrum of the anomalous samples, the anomalous frequency lines are marked as f on the horizontal axis. mn In the spectrum diagram of the later abnormal samples, the abnormal frequency spectral lines are marked as f on the horizontal axis. sn ;

[0072] The decomposition of the baseline sample, initial outlier sample, and later outlier sample is based on the sampling frequency f = 1 / , The sampling time interval is set manually;

[0073] S2: Subtract the frequencies of the initial abnormal sample's spectrum from the reference sample's spectrum arithmetically, i.e., Pmn - Pn, to obtain the following... Figure 10 The initial remaining spectrum diagram shown is used as the spectrum diagram for fault occurrence.

[0074] The frequencies of the spectrum of the later abnormal sample and the spectrum of the reference sample are arithmetically subtracted, i.e., Psn - Pn, to obtain the following: Figure 11 The remaining spectrum diagram shown below;

[0075] S3: Subtract the frequencies of the later residual spectrum from the initial residual spectrum arithmetically, i.e., (Psn - Pn) - (Pmn - Pn), to obtain the following: Figure 12 The external interference spectrum diagram is shown below;

[0076] S4: Input the fault generation spectrum diagram and corresponding faults into the fault analysis system to establish a fault analysis database.

[0077] The working principle is as follows: the sound acquisition module is directly installed inside the pump cavity, providing first-hand data support for predictive maintenance; the spectrum of abnormal samples is analyzed to quickly determine pump failure; initial sound data is randomly sampled in the early and late stages because the pump's operating status will be affected for a period of time after a failure, which may lead to other noises and affect the initial spectrum judgment of abnormalities; therefore, by collecting a large amount of data to establish a fault analysis database, it has better timeliness and accuracy compared to existing temperature, vibration and other monitoring methods.

[0078] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above, and that the invention can be implemented in other specific forms without departing from its spirit or essential characteristics. Therefore, the embodiments should be considered illustrative and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be included within the scope of the invention. No reference numerals in the claims should be construed as limiting the scope of the claims.

[0079] Furthermore, it should be understood that although this specification describes embodiments, not every embodiment contains only one independent technical solution. This narrative style is merely for clarity. Those skilled in the art should consider the specification as a whole, and the technical solutions in each embodiment can also be appropriately combined to form other embodiments that can be understood by those skilled in the art.

Claims

1. An early warning system for pump failures in chemical processes, characterized in that, The system includes multiple pumps equipped with sound acquisition devices, signal relay stations, and signal receiving stations; the sound acquisition device includes a sound acquisition end, which is installed inside the pump; the sound acquisition device can convert the real-time sound signal acquired by the sound acquisition end into initial sound data and send it. The signal relay station includes a signal relay module, which is used to receive the initial sound data sent by the sound acquisition device and send it to the signal receiving main station; The signal receiving station includes a signal receiver and a host computer. The signal receiver receives the initial sound data transmitted by the signal relay station and transmits it to the host computer. The host computer is equipped with a data processing system and an early warning judgment system. The data processing system processes the initial sound data to obtain real-time noise data and stores it. At the same time, the real-time noise data is input into the early warning judgment system. The pump includes a pump casing, a rotating shaft, an impeller, and a guide vane. The pump casing has a cavity, and a through-hole is provided on the side wall of the pump casing. The central axis of the through-hole is perpendicular to the axis of the rotating shaft. The sound acquisition device includes a housing with a cover screwed onto it. A sound transmission tube is rotatably connected to the bottom of the housing. A probe is provided on one side of the lower end of the sound transmission tube. A sound sensing microphone, a signal processing and transmission module, and a power supply module are provided inside the housing. A signal receiver is provided outside the cover. The signal processing and transmission module is electrically connected to the sound sensing microphone and the signal receiver, respectively. The upper part of the sound transmission tube is snapped into a through-hole, and the lower part of the sound transmission tube and the probe are located in a cavity.

2. The early warning system for pump failure in a chemical process as described in claim 1, characterized in that, The sound transmission tube is equipped with a sound intensity control valve at its upper end.

3. The early warning system for pump failure in a chemical process as described in claim 1, characterized in that, The wall thickness of the portion of the sound transmission tube located inside the through-hole is greater than the wall thickness of the lower part of the sound transmission tube.

4. The early warning system for pump failure in a chemical process as described in claim 3, characterized in that, A sealing bushing is provided between the outer wall of the sound transmission tube and the inner wall of the through-hole.

5. The early warning system for pump failure in a chemical process as described in claim 1, characterized in that, The housing is equipped with a mounting bracket, on which the signal processing and transmission module and the power supply module are fixed.

6. The early warning system for chemical process pump failures as described in any one of claims 1-5, characterized in that, The warning judgment system compares real-time noise data with preset noise data. If the noise exceeds the preset noise data by 20-60%, an intermittent alarm is triggered; if it exceeds the preset noise data by 60-120%, a continuous alarm is triggered. Among them, preset noise data Real-time noise data The calculation formula is as follows: (2) in, The average value of the noise volume after integrating from t1 to t2 is taken for the part with good noise conditions. It is t n to t n+1 The average noise volume after integration over a time period, where t1 is the base sampling start time and t2 is the base sampling end time. n For real-time sampling start time, t n+1 The sampling end time is Δt, which is a manually set sampling time interval.

7. The early warning system for pump failure in a chemical process as described in claim 6, characterized in that, The host computer is also equipped with a fault analysis system. When the early warning judgment system detects an alarm signal, it identifies the system as being in an abnormal working state; if no alarm signal is detected, it identifies the system as being in a normal working state. The initial sound data collected during the abnormal working state is processed to obtain a fault generation spectrum, which is then compared with a fault analysis database to determine the fault type. The method for constructing the fault analysis database includes the following: S1: Under normal working conditions, the initial sound data is randomly sampled as a reference sample, and the reference sample is decomposed to obtain the spectrum of the reference sample. In abnormal working state, the initial sound data is randomly sampled in the early stage and the later stage to obtain the initial abnormal sample and the later abnormal sample, respectively. The initial abnormal sample and the later abnormal sample are decomposed to obtain the spectrum diagram of the initial abnormal sample and the spectrum diagram of the later abnormal sample. S2: The frequencies of the spectrum diagrams of the initial abnormal samples and the spectrum diagrams of the reference samples are arithmetically subtracted to obtain the initial residual spectrum diagram, which is used as the fault generation spectrum diagram. The frequency of the spectrum of the later abnormal sample and the spectrum of the reference sample are arithmetically subtracted to obtain the remaining spectrum of the later sample. S3: Subtract the frequencies of the later remaining spectrum map and the initial remaining spectrum map arithmetically to obtain the external interference spectrum map; S4: Input the fault generation spectrum diagram and corresponding faults into the fault analysis system to establish a fault analysis database.

8. The early warning system for pump failure in a chemical process as described in claim 7, characterized in that, The decomposition of the baseline sample, the initial abnormal sample, and the later abnormal sample is the sampling frequency f = 1 / Δt, where Δt is a manually set sampling time interval.

Images