Centrifugal pump operation monitoring method and system

Abstract

The invention provides a centrifugal pump operation monitoring method and system, and the method comprises the following steps: collecting the operation parameters of a centrifugal pump in real time through a plurality of sensors, including inlet flow, inlet and outlet pressure and pressure difference, a main shaft vibration signal, a main shaft rotation speed, and the temperature of a pump body and a medium; denoising, abnormal value elimination and normalization preprocessing are carried out on the collected data; establishing a centrifugal pump spindle radial hydraulic exciting force model; calculating the alternating stress of the main shaft according to the hydraulic exciting force and predicting the fatigue life; and a fault early warning mechanism is designed based on the fatigue life prediction result, and real-time monitoring and fault early warning of the running state of the centrifugal pump are achieved. Through combination of data acquisition and processing, life prediction and a fault early warning mechanism, real-time monitoring of the running state of the centrifugal pump spindle and residual life prediction are achieved, and the running reliability and safety of equipment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of centrifugal pumps, and in particular to a centrifugal pump operation monitoring method and system. Background Art

[0002] As a key equipment in industrial production, the operation status of centrifugal pumps directly affects production efficiency and equipment safety. However, in actual operation, centrifugal pumps may be affected by a variety of factors such as vibration signals, temperature, pressure, speed changes, and flow fluctuations, leading to equipment failure or even shutdown. The existing centrifugal pump operation monitoring methods have the following technical problems: Incomplete or inaccurate monitoring data: Single data (such as pressure or flow) collected by traditional monitoring methods cannot fully reflect the operating status of the centrifugal pump.

[0003] Lack of real-time performance: Data collection and analysis are delayed, and potential faults cannot be discovered in time.

[0004] Insufficient data analysis capabilities: There is no effective algorithm to conduct comprehensive analysis of multi-dimensional data, making it difficult to accurately predict failure trends.

[0005] High maintenance cost: Blind maintenance or excessive maintenance leads to increased operating costs.

[0006] Therefore, it is necessary to design a centrifugal pump operation monitoring method and system to solve the above technical problems. Summary of the invention

[0007] The present invention provides a centrifugal pump operation monitoring method and system, which are used to solve the technical problems mentioned in the above background technology.

[0008] In order to solve the above problems, the present invention provides the following technical solution: a centrifugal pump operation monitoring method, comprising the following steps: S100 uses multiple sensors to obtain the operating parameters of the centrifugal pump in real time, including: Flow rate Q: flow rate at the inlet of the centrifugal pump; Pressure P: includes the inlet pressure P_in of the centrifugal pump and the outlet pressure P_out of the centrifugal pump; Vibration signal a: vibration signal of the spindle; Speed ​​n: spindle speed; Temperature T: includes pump body temperature T_pump and medium temperature T_fluid; S200, performing denoising, outlier elimination and normalization processing on the collected operating parameters; S300, establishing a radial hydraulic excitation force model of the centrifugal pump main shaft; S400, calculates the alternating stress of the main shaft according to the radial hydraulic excitation force, calculates the fatigue damage value D based on Miner's fatigue cumulative damage theory, and predicts its fatigue life; S500, design a corresponding fault warning mechanism based on the fatigue damage value D, design a corresponding fault warning mechanism to perform real-time monitoring and fault warning on the operation status of the centrifugal pump.

[0009] Furthermore, in step S200, it includes: S201, denoising the vibration signal using wavelet transform; S202, smooth the flow and pressure data using the sliding average method to eliminate instantaneous fluctuations; S203, using the 3σ criterion to eliminate abnormal values ​​in the operating parameters; S204, normalize the flow, pressure, and vibration signals to the [0,1] interval.

[0010] Furthermore, in step S300, it includes: S301, calculate the pressure difference: ΔP = P_in - P_out; S302, calculate radial hydraulic excitation force: ; Among them, C1 is the fluid mechanics coefficient related to the pump structure, Q is the flow rate at the inlet of the centrifugal pump, ΔP is the pressure difference, and n is the spindle speed.

[0011] Furthermore, in step S400, it includes: S401, calculation of the main shaft alternating stress based on the radial hydraulic excitation force: σ = (K · F) / A, Among them, K is the structural coefficient, A is the cross-sectional area of ​​the principal axis; S402, calculate the fatigue damage value D based on Miner fatigue cumulative damage theory: D = Σ(n_i / N_i), Among them, n_i is the number of cycles of a certain stress amplitude, and N_i is the number of fatigue limit cycles under the stress amplitude; S403, predict the remaining life of the spindle based on the fatigue damage value: L_remain = L_total · (1 - D) Among them, L_total is the total design life of the spindle, and L_remain is the remaining life.

[0012] Furthermore, in step S500, it includes: S501, design the corresponding fault warning mechanism based on the fatigue damage value D: Level 1 warning: D ≥ 0.8, reminding users to plan maintenance; Level 2 warning: D ≥ 0.9, it is recommended that the user stop the machine for inspection; Level 3 warning: D = 1, immediate shutdown; S502, predicting the future damage value by real-time monitoring of the fatigue damage change rate dD / dt; S503, when the predicted damage value exceeds the warning threshold at a certain time point in the future, an alarm signal is triggered in advance and corresponding maintenance suggestions are provided.

[0013] Furthermore, the fault warning mechanism also includes a temperature abnormality monitoring step: S5011, monitor the temperature difference ΔT between the pump body temperature T_pump and the medium temperature T_fluid. When the temperature difference exceeds the preset threshold ΔT_threshold, a temperature warning is triggered; S5012, monitoring the temperature change rate dT / dt, when dT / dt exceeds the preset threshold dT / dt_threshold, prompting an overheating fault; S5013, combined with the fatigue damage change rate dD / dt and temperature anomalies, comprehensively evaluate the operating status of the centrifugal pump and optimize maintenance recommendations.

[0014] The present invention also discloses a centrifugal pump operation monitoring system, comprising: The data acquisition module obtains the operating parameters of the centrifugal pump in real time through multiple sensors, including: Flow rate Q: flow rate at the inlet of the centrifugal pump; Pressure P: includes the inlet pressure P_in of the centrifugal pump and the outlet pressure P_out of the centrifugal pump, and calculates the pressure difference ΔP; Vibration signal a: vibration signal of the spindle; Speed ​​n: spindle speed; Temperature T: includes pump body temperature T_pump and medium temperature T_fluid; The data processing module is used to perform denoising, outlier removal and normalization on the collected operating parameters; The radial hydraulic excitation force modeling module is used to establish the radial hydraulic excitation force model of the centrifugal pump main shaft; The main shaft fatigue prediction module calculates the alternating stress of the main shaft according to the radial hydraulic excitation force, calculates the fatigue damage value D based on Miner's fatigue cumulative damage theory, and predicts its fatigue life; The fault warning module designs the corresponding fault warning mechanism based on the fatigue damage value D, and performs real-time monitoring and fault warning on the operation status of the centrifugal pump.

[0015] Compared with the prior art, the present invention has at least the following beneficial effects: the present invention realizes real-time monitoring of the operating status of the centrifugal pump main shaft and prediction of the remaining life by real-time collection of data such as flow, pressure, vibration signal, speed and temperature during the operation of the centrifugal pump, combined with data preprocessing, main shaft radial hydraulic excitation force modeling, fatigue life prediction and fault warning mechanism. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0017] Figure 1 is a flow chart of the steps of the centrifugal pump operation monitoring method in an embodiment of the present invention; Figure 2 Schematic diagram of the structure of the centrifugal pump operation monitoring system in an embodiment of the present invention. DETAILED DESCRIPTION

[0018] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work belong to the protection scope of the present invention. It is understood that the drawings are only provided for reference and explanation, and are not used to limit the present invention.

[0019] See also Figure 1 , The present invention discloses a centrifugal pump operation monitoring method, which realizes real-time monitoring of the operation state of the centrifugal pump main shaft and prediction of the remaining life by real-time collection of flow, pressure, vibration signal, rotation speed, temperature and other data during the operation of the centrifugal pump, combined with data preprocessing, main shaft radial hydraulic excitation force modeling, fatigue life prediction and fault warning mechanism, and specifically includes the following steps: Step S100: Data collection The following operating parameters are collected in real time using sensors installed on the centrifugal pump: 1. Flow Q: The inlet flow is collected by installing a flow meter in the pump inlet pipe. The flow meter is generally designed in a straight pipe section to avoid fluid turbulence affecting the collection accuracy.

[0020] Example data: Assume that the collected pump inlet flow rate is Q = 20.55 m³ / h.

[0021] 2. Pressure P: The centrifugal pump inlet pressure P_in and outlet pressure P_out are measured by high-precision pressure sensors. The pressure sensors are installed at the pump inlet and outlet flange connection points. The centrifugal pump inlet pressure P_in refers to the pressure in the pipe before the liquid enters the pump body. The centrifugal pump outlet pressure refers to the pressure P_out when the liquid is discharged from the pump body. The pressure difference is calculated: ΔP = P_in - P_out.

[0022] Sample data: The inlet and outlet pressures collected by the pressure sensor are P_in = 1.45 MPa, P_out = 1.15 MPa, and the calculation results are: ΔP = 1.45 MPa - 1.15 MPa = 0.3 MPa.

[0023] 3. Vibration signal a: A triaxial vibration sensor was installed near the spindle bearing seat, and the sampling frequency was set to 10 kHz to capture high-frequency vibration signals.

[0024] Sample data: The vibration signal collected by the vibration sensor is a.

[0025] 4. Speed ​​n: The photoelectric speed sensor is installed at the end of the spindle to measure the spindle speed.

[0026] Sample data: Assume that the acquired spindle speed is n = 2950 rpm.

[0027] 5. Temperature T: The temperature of the pump body and the medium is measured by thermocouples, which are installed on the surface of the pump body and in the outlet pipe respectively.

[0028] Example data: Pump temperature is T_pump = 75 °C, medium temperature is T_fluid = 80 °C.

[0029] The collected data is transmitted wirelessly (such as ZigBee or LoRa) to a central monitoring system, or wired to a data storage unit.

[0030] The data acquisition frequency is 1 Hz, and high-frequency vibration signals are stored independently to meet subsequent analysis needs.

[0031] Purpose: To achieve multi-dimensional monitoring of pump operating parameters and provide high-quality data support for subsequent modeling and analysis. Through real-time data collection, problems can be discovered and measures can be taken in time during abnormal operation to reduce the risk of sudden equipment failure.

[0032] Step S200, data preprocessing: The original collected data may contain noise interference, missing values ​​or outliers. It is necessary to denoise, remove outliers and normalize the collected original data, including: S201, vibration signal denoising: Wavelet transform is used to denoise the vibration signal a, and Daubechies basis function is selected for three-layer decomposition to filter out high-frequency noise.

[0033] Example data: The denoised signal is {a_filtered}, which retains the main features of the vibration signal.

[0034] S202, pressure and flow signal smoothing: The pressure and flow data are smoothed using the sliding average method to eliminate instantaneous fluctuations.

[0035] Example data: The original traffic data is {20.3, 20.6, 20.2, 20.9, 20.1}, and the average value is obtained after smoothing: The sliding window size (k) represents the average value of several adjacent data points each time; In this example, assume that the sliding window size is 3 (i.e., an average is calculated for every 3 consecutive data points) Move the sliding window step by step, and calculate the simple arithmetic mean of the data in the window each time: Step 1 (window range: 1st, 2nd, 3rd data points): (20.3 + 20.6 + 20.2) / 3 = 20.37.

[0036] Step 2 (window range: 2nd, 3rd, 4th data points): (20.6 + 20.2 + 20.9) / 3 = 20.57.

[0037] Step 3 (window range: 3rd, 4th, 5th data points): (20.2 + 20.9 + 20.1) / 3 = 20.40.

[0038] The flow data after sliding average is: {20.37, 20.57, 20.40}.

[0039] Then take the average value of the sliding average data: (20.37 + 20.57 + 20.40) / 3 = 20.45 ≈ 20.5 m³ / h.

[0040] Therefore, Q_smooth = 20.5 m³ / h.

[0041] In this embodiment, the pressure difference after smoothing is ΔP_smooth = 0.3 MPa.

[0042] S203, outlier removal: The 3σ criterion is used to detect outliers and eliminate abnormal peaks in operating parameters that exceed three times the standard deviation of the mean.

[0043] Sample data: The instantaneous peak value in a vibration signal is 50g, which is judged as an abnormal value and is removed.

[0044] S204, normalization processing: Normalize the flow, pressure, and vibration signals to the [0, 1] interval using the following formula: Normalized value = (current value - minimum value) / (maximum value - minimum value).

[0045] Example data: The flow range is [10, 25] m³ / h, the current flow is Q_smooth = 20.5 m³ / h, and the normalized result is: Normalized value = (20.5 - 10) / (25 - 10) = 0.7.

[0046] Purpose: Data preprocessing effectively improves the accuracy and robustness of subsequent modeling calculations, avoids misjudgment caused by noise or outliers, and further enhances the accuracy of fatigue analysis.

[0047] Step S300, modeling of the main shaft radial hydraulic excitation force: Based on the flow rate Q, pressure difference ΔP and speed n, the radial hydraulic excitation force acting on the main shaft is calculated, and combined with the denoised vibration signal {a_filtered}, the dynamic response of the main shaft is analyzed as follows: S301, calculate the pressure difference: ΔP = P_in - P_out.

[0048] Example data: ΔP = 1.45 MPa - 1.15 MPa = 0.3 MPa.

[0049] S302, Excitation force calculation: The radial hydraulic excitation force is calculated using the formula: , Wherein, C1 is the fluid mechanics coefficient. In this embodiment, the centrifugal pump is a common closed impeller single-stage pump, and the operating condition is stable, and the value C1 is 0.04.

[0050] Example data: Q = 20.5 m³ / h, ΔP = 0.3 MPa, n = 2950 rpm, calculated as: .

[0051] Objective: To quantify the radial hydraulic excitation effect during the operation of a centrifugal pump by modeling the radial hydraulic excitation force, and to provide a basis for fatigue life prediction.

[0052] Step S400: Spindle fatigue life prediction Combined with the radial hydraulic excitation force, the alternating stress of the main shaft is calculated and the fatigue life of the main shaft is predicted, including: S401, Alternating Stress Calculation: σ = (K · F) / A Example data: Assuming the cross-sectional area of ​​the main shaft is A = 0.02 m², the structural coefficient K is 1-2, in this embodiment K = 1.2, and the radial hydraulic excitation force F = 1482.9 N, then: σ = (1.2 × 1482.9) / 0.02 ≈ 88.97 kPa, S402, fatigue damage value calculation: The fatigue damage value D is calculated based on Miner's fatigue cumulative damage theory: D = Σ (n_i / N_i); Among them, n_i is the number of cycles of a certain stress amplitude, and N_i is the number of fatigue limit cycles under this stress amplitude.

[0053] In this embodiment: the main shaft alternating stress σ = 88.97 kPa, the corresponding fatigue life is 5800 cycles, and the current number of cycles is 1200 times, then: D = 1200 / 5800 ≈ 0.207.

[0054] S403, prediction of remaining life of centrifugal pump main shaft: Assuming the total designed spindle life is 10,000 hours, the remaining spindle life is: L_remain = L_total × (1 - D).

[0055] Example data: L_remain = 10000 × (1 - 0.207) ≈ 7930 hours.

[0056] Purpose: To estimate the remaining service life of the spindle through fatigue life prediction and provide a basis for maintenance planning.

[0057] Step S500: Fault warning mechanism Design the corresponding fault warning mechanism based on the fatigue damage value D: S501, warning level: Level 1 warning: D ≥ 0.8, reminding users to plan maintenance; Level 2 warning: D ≥ 0.9, it is recommended that the user stop the machine for inspection; Level 3 warning: D = 1, shut down immediately.

[0058] S502, prediction of future damage value: Monitor fatigue damage change rate dD / dt in real time and predict future damage value.

[0059] The current fatigue damage change rate is dD / dt = 0.01 / hour, and the damage value predicted for the next 60 hours is: For example, future damage value = 0.207 + (0.01 × 60) = 0.807.

[0060] S503, trigger alarm signal When the predicted damage value exceeds the warning threshold at a certain point in the future, an alarm signal is triggered in advance and corresponding maintenance suggestions are provided.

[0061] Conclusion: Since the damage value after 60 hours is 0.807, the system enters the first-level warning state after 60 hours to remind the user to plan maintenance.

[0062] Purpose: Through the real-time fault warning mechanism, potential faults can be discovered in advance and the risk of sudden failure of centrifugal pumps can be reduced.

[0063] The present invention also includes a temperature anomaly monitoring step, specifically: Step S5011, monitoring the temperature difference ΔT between the pump body temperature T_pump and the medium temperature T_fluid, when the temperature difference exceeds a preset threshold ΔT_threshold, triggering a temperature warning.

[0064] If the temperature difference ΔT = T_fluid - T_pump> ΔT_threshold (set threshold), a thermal failure risk is indicated; Step S5012, if the temperature change rate dT / dt>dT / dt_threshold, it is indicated that an overheating fault exists; Step S5013: Combine the fatigue damage change rate dD / dt and the temperature anomaly to comprehensively evaluate the operation status of the centrifugal pump and optimize the maintenance suggestions.

[0065] The present invention realizes intelligent, real-time and precise operation monitoring of centrifugal pumps by combining data collection, preprocessing, modeling and life prediction, effectively improving the safety of equipment operation and the accuracy of life prediction.

[0066] See also Figure 2 The present invention also discloses a centrifugal pump operation monitoring system, comprising: The data acquisition module 11 obtains the operating parameters of the centrifugal pump in real time through multiple sensors, including: Flow rate Q: flow rate at the inlet of the centrifugal pump; Pressure P: includes the inlet pressure P_in of the centrifugal pump and the outlet pressure P_out of the centrifugal pump; Vibration signal a: vibration signal of the spindle; Speed ​​n: spindle speed; Temperature T: includes pump body temperature T_pump and medium temperature T_fluid; The data processing module 12 is used to perform denoising, outlier removal and normalization processing on the collected operating parameters; A radial hydraulic excitation force modeling module 13 is used to establish a radial hydraulic excitation force model of a centrifugal pump main shaft; The main shaft fatigue prediction module 14 calculates the alternating stress of the main shaft according to the radial hydraulic excitation force and predicts its fatigue life; The fault warning module 15 designs a fault warning mechanism based on the fatigue life prediction result to perform real-time monitoring and fault warning on the operation status of the centrifugal pump.

[0067] In the specification and claims of the present application, the words "include / comprise" and the words "have / include" and their variations are used to specify the existence of stated features, values, steps or components, but do not exclude the existence or addition of one or more other features, values, steps, components or combinations thereof.

[0068] Some features of the present invention are described in different embodiments for clarity of description, however, these features may also be described in combination in a single embodiment. Conversely, some features of the present invention are described in only a single embodiment for brevity, however, these features may also be described in different embodiments alone or in any suitable combination.

[0069] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions and improvements made within the spirit and principles of the present invention should be included in the protection scope of the present invention.

Claims

1. A centrifugal pump operation monitoring method, characterized in that: The following steps are involved: S100 uses multiple sensors to obtain the operating parameters of the centrifugal pump in real time, including: Flow rate Q: flow rate at the inlet of the centrifugal pump; Pressure P: includes the inlet pressure P_in of the centrifugal pump and the outlet pressure P_out of the centrifugal pump; Vibration signal a: vibration signal of the spindle; Speed ​​n: spindle speed; Temperature T: includes pump body temperature T_pump and medium temperature T_fluid; S200, performing denoising, outlier elimination and normalization processing on the collected operating parameters; S300, establishing a radial hydraulic excitation force model of the centrifugal pump main shaft; S400, calculates the alternating stress of the main shaft according to the radial hydraulic excitation force, calculates the fatigue damage value D based on Miner's fatigue cumulative damage theory, and predicts the remaining life of the main shaft; S500, based on the fatigue damage value D, designs the corresponding fault warning mechanism to conduct real-time monitoring and fault warning of the centrifugal pump operation status.

2. The centrifugal pump operation monitoring method according to claim 1, characterized in that: In step S200, it includes: S201, denoising the vibration signal using wavelet transform; S202, smooth the flow and pressure data using the sliding average method to eliminate instantaneous fluctuations; S203, using the 3σ criterion to eliminate abnormal values ​​in the operating parameters; S204, normalize the flow, pressure, and vibration signals to the [0,1] interval.

3. The centrifugal pump operation monitoring method according to claim 2, characterized in that: In step S300, it includes: S301, calculate the pressure difference: ΔP = P_in - P_out; S302, calculate radial hydraulic excitation force: ; Among them, C1 is the fluid mechanics coefficient related to the pump structure, Q is the flow rate at the inlet of the centrifugal pump, ΔP is the pressure difference, and n is the spindle speed.

4. The centrifugal pump operation monitoring method according to claim 3, characterized in that: In step S400, it includes: S401, calculation of the main shaft alternating stress based on the radial hydraulic excitation force: σ = (K · F) / A, Among them, K is the structural coefficient, A is the cross-sectional area of ​​the principal axis; S402, calculate the fatigue damage value D based on Miner fatigue cumulative damage theory: D = Σ(n_i / N_i), Among them, n_i is the number of cycles of a certain stress amplitude, and N_i is the number of fatigue limit cycles under the stress amplitude; S403, predict the remaining life of the spindle based on the fatigue damage value: L_remain = L_total · (1 - D) Among them, L_total is the total design life of the spindle, and L_remain is the remaining life.

5. The centrifugal pump operation monitoring method according to claim 4, characterized in that: In step S500, it includes: S501, design the corresponding fault warning mechanism based on the fatigue damage value D: Level 1 warning: D ≥ 0.8, reminding users to plan maintenance; Level 2 warning: D ≥ 0.9, it is recommended that the user stop the machine for inspection; Level 3 warning: D = 1, immediate shutdown; S502, predicting the future damage value by real-time monitoring of the fatigue damage change rate dD / dt; S503, when the predicted damage value exceeds the warning threshold at a certain time point in the future, an alarm signal is triggered in advance and corresponding maintenance suggestions are provided.

6. The centrifugal pump operation monitoring method according to claim 5, characterized in that: The fault warning mechanism also includes a temperature abnormality monitoring step: S5011, monitor the temperature difference ΔT between the pump body temperature T_pump and the medium temperature T_fluid. When the temperature difference exceeds the preset threshold ΔT_threshold, a temperature warning is triggered; S5012, monitor the temperature change rate dT / dt, when dT / dt exceeds the preset threshold dT / dt_threshold, prompt overheating fault; S5013, combined with the fatigue damage change rate dD / dt and temperature anomalies, comprehensively evaluate the operating status of the centrifugal pump and optimize maintenance recommendations.

7. A centrifugal pump operation monitoring system, characterized in that: include: The data acquisition module obtains the operating parameters of the centrifugal pump in real time through multiple sensors, including: Flow rate Q: flow rate at the inlet of the centrifugal pump; Pressure P: includes the inlet pressure P_in of the centrifugal pump and the outlet pressure P_out of the centrifugal pump, and calculates the pressure difference ΔP; Vibration signal a: vibration signal of the spindle; Speed ​​n: spindle speed; Temperature T: includes pump body temperature T_pump and medium temperature T_fluid; The data processing module is used to perform denoising, outlier removal and normalization on the collected operating parameters; The radial hydraulic excitation force modeling module is used to establish the radial hydraulic excitation force model of the centrifugal pump main shaft; The main shaft fatigue prediction module calculates the alternating stress of the main shaft according to the radial hydraulic excitation force, calculates the fatigue damage value D based on Miner's fatigue cumulative damage theory, and predicts the remaining life of the main shaft; The fault warning module designs the corresponding fault warning mechanism based on the fatigue damage value D, and performs real-time monitoring and fault warning on the operation status of the centrifugal pump.

Images