A compressor health diagnosis system

Abstract

The present invention relates to the field of signal processing technology, and discloses a compressor health diagnosis system, including sensors, data acquisition modules, and data processing and analysis modules connected in sequence; the sensors are arranged at key parts of the compressor to ensure that various operating data can be accurately obtained; the data acquisition module is used to collect key operating parameters of the compressor; the data processing and analysis module is used to process the key operating parameters using a moving average method to achieve abnormality detection. The present invention can provide faster processing speed and better real-time performance through optimization algorithms and hardware implementations, has higher flexibility, can be quickly adjusted and optimized according to different application requirements, adopts more advanced noise suppression technology, can better suppress noise and interference, can reduce resource consumption through algorithm optimization and dedicated hardware design, and can better adapt to different working conditions and environmental changes.

Description

Technical Field

[0001] The present invention relates to the technical field of signal processing, and in particular to a compressor health diagnosis system. Background Art

[0002] The diaphragm compressor is a special volumetric compressor. Its core component is the diaphragm, which completely isolates the compression system from the hydraulic oil system to ensure the purity of the compressed gas. The diaphragm reciprocates in the cylinder and is driven by the pressure changes in the hydraulic oil system to achieve gas compression and transportation. During the compression process, the reciprocating motion of the diaphragm causes the cylinder volume to change alternately, and the gas is compressed and discharged in the cylinder. Since the diaphragm is not in direct contact with the gas, lubrication is not required, which reduces the introduction of pollutants. It is suitable for compressing special gases such as high-purity, flammable and explosive, toxic and harmful gases.

[0003] The compressor health diagnosis system uses low-pass filtering to remove high-frequency noise in the signal, thereby improving the accuracy of the diagnosis. The compressor health diagnosis system generates a large amount of vibration signal data during operation, which usually contains important information about the compressor's operating status, such as mechanical failures, wear and tear, etc. However, as the complexity of the diagnostic system increases, the surge in data volume poses a challenge to real-time data processing. Massive amounts of data require more efficient algorithms and more powerful computing power for rapid analysis to achieve real-time monitoring and fault warning.

[0004] Traditional filtering methods usually require large amounts of computation and long processing times, making them unsuitable for real-time signal processing. They are complex to design and implement, difficult to adjust and optimize, and may not be effective enough when dealing with complex noise and interference, requiring more hardware resources such as computing power and storage space. At the same time, their performance is greatly affected by device parameters and they have poor adaptability to the environment. Summary of the invention

[0005] The present invention provides a compressor health diagnosis system, which can provide faster processing speed and better real-time performance through optimized algorithm and hardware implementation, has higher flexibility, can be quickly adjusted and optimized according to different application requirements, adopts more advanced noise suppression technology, can better suppress noise and interference, and can reduce resource consumption through algorithm optimization and dedicated hardware design, and can better adapt to different working conditions and environmental changes.

[0006] The present invention provides a compressor health diagnosis system, comprising a sensor, a data acquisition module, and a data processing and analysis module connected in sequence;

[0007] The sensors are arranged at key positions of the compressor to ensure that various operating data can be accurately obtained;

[0008] The data acquisition module is used to collect key operating parameters of the compressor; wherein the operating parameters include temperature, pressure, vibration, flow, speed, power consumption, and exhaust volume;

[0009] The data processing and analysis module is used to process the key operating parameters using a moving average method to achieve abnormality detection.

[0010] Furthermore, the data processing and analysis module includes a real-time data monitoring unit, a moving average method application unit and an anomaly detection unit. The real-time data monitoring unit is used to monitor the collected key operating parameters in real time to observe their changing trends; the moving average method application unit is used to process the key operating parameters using the moving average method to smooth data fluctuations and determine the trend or periodicity of the data; the anomaly detection unit is used to compare the real-time operating parameters with preset thresholds or historical data to identify abnormal data points.

[0011] Furthermore, in the moving average method application unit, the calculation formula of the simple moving average method is:

[0012]

[0013] Among them, F t is the moving average value at time t, A i is the i-th data point in the time series; n is the number of moving average items;

[0014] The weighted moving average method is:

[0015]

[0016] Among them, wi is the weight of the i-th data point.

[0017] Furthermore, the data processing steps of the moving average method application unit include:

[0018] Collect real-time sensor data of the compressor, including but not limited to temperature, vibration, and pressure signals;

[0019] Preprocess the collected data to remove obvious outliers and noise;

[0020] The preprocessed data is input into the moving average algorithm for smoothing to reduce data noise;

[0021] Perform feature extraction on the smoothed data to identify possible failure modes;

[0022] The fault mode is input into the diagnosis model to generate the fault diagnosis conclusion of the compressor.

[0023] Furthermore, the step of preprocessing the collected data includes:

[0024] The sensor data is denoised and the three-point median filter method is used to remove obvious abnormal data;

[0025] The data is normalized to ensure that the units and magnitudes of data from different sensors are consistent, which facilitates subsequent moving average processing.

[0026] Furthermore, the step of inputting the pre-processed data into a moving average algorithm for smoothing to reduce data noise includes:

[0027] For the time series data of each sensor, set a sliding window with a window size of M;

[0028] For the data in each sliding window, calculate its arithmetic mean, the formula is as follows:

[0029]

[0030] Wherein, x[k] represents the sample value of index position k in the signal sequence.

[0031] Furthermore, the step of extracting features from the smoothed data and identifying possible failure modes includes:

[0032] Perform frequency domain analysis on vibration data and extract spectrum features using Fast Fourier Transform (FFT);

[0033] Perform trend analysis on temperature data and extract the temperature change slope;

[0034] Perform statistical analysis on pressure data and extract its mean and variance;

[0035] The extracted features are input into a fault diagnosis model, which is a machine learning classifier or a rule-based system;

[0036] The diagnostic model first classifies the input features and matches them with the features in the preset fault mode library;

[0037] If the matching degree is higher than the threshold, the system generates a fault alarm and indicates the potential fault type.

[0038] Furthermore, it also includes a fault repair and optimization module, which is used to repair the fault in a timely manner according to the fault type and location in the diagnosis conclusion; perform performance tests on the repaired compressor to ensure that the fault has been completely eliminated and restored to the optimal operating state, analyze the cause of the fault, and make necessary optimization adjustments to the compressor.

[0039] Furthermore, it also includes an operating parameter adjustment module, which is used to adjust the operating parameters according to the actual operating conditions and load requirements of the compressor to optimize the operating efficiency of the compressor; and set the operating parameter range to ensure that the compressor operates in an efficient and stable state.

[0040] Furthermore, it also includes a preventive maintenance plan formulation module, which is used to formulate a preventive maintenance plan based on the diagnosis results and the historical operation data of the compressor, including regular inspection, maintenance, and replacement of wearing parts to prevent the occurrence of potential failures.

[0041] The present invention also provides a computer device, comprising a memory and a processor, wherein the memory stores a computer program, and the processor implements the steps of the above method when executing the computer program.

[0042] The present invention also provides a computer-readable storage medium on which a computer program is stored, and the computer program implements the steps of the above method when executed by a processor.

[0043] The beneficial effects of the present invention are:

[0044] The compressor health diagnosis system provided by the present invention is a monitoring information processing process, which realizes real-time online monitoring of equipment status parameters such as vibration, temperature, displacement, pressure, and speed of the unit. It can determine the fault location and diagnose the cause of the fault through analysis methods such as indicator diagrams, frequency spectra, and trend analysis. At the same time, the status data is saved in a database, necessary information is provided for equipment maintenance, predictive maintenance is realized, and the compressor unit can operate safely and stably for a long period of time. Therefore, it is of great significance in the application of reciprocating compressors. BRIEF DESCRIPTION OF THE DRAWINGS

[0045] Figure 1 It is a structural schematic diagram of the compressor health diagnosis system of the present invention.

[0046] Figure 2 Schematic diagram of data after noise processing in the present invention.

[0047] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0048] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0049] The fast low-pass filtering algorithm removes high-frequency noise from the signal collected by the sensor, thereby extracting key low-frequency information that reflects the compressor status. The moving average method is a classic low-pass filtering technique that achieves filtering by calculating the average value in the signal sequence. This method is simple and efficient, suitable for real-time data processing, and can effectively smooth signals, reduce the impact of noise, and improve diagnostic accuracy. In compressor fault diagnosis, the mean shift method helps to identify low-frequency features related to mechanical fatigue, abnormal vibration, etc., which is particularly critical for early fault detection and health status assessment.

[0050] The low-pass filter allows low-frequency signals to pass through while suppressing high-frequency signals above a certain cutoff frequency. This can effectively filter out noise that may be generated by mechanical vibration, electrical interference, etc., and retain low-frequency signal components related to the health status of the compressor. In this way, the diagnostic system can more accurately analyze the working status of the compressor, detect potential faults in a timely manner, and ensure the stable operation of the compressor and extend its service life.

[0051] Vibration signals can reflect the dynamic characteristics and operating status of equipment, and they have a certain indicative significance for the degree of damage and location of equipment. By analyzing vibration signals, the type and location of equipment failure can be judged more accurately. Vibration signals have high frequency response and good dynamic performance. They are good at capturing early subtle fault characteristics and can detect faults earlier than other commonly used signals to achieve predictive maintenance. The installation method of vibration sensors is friendly and non-invasive, with little impact on the equipment itself, no impact on on-site production, and easy to monitor without interrupting production. The vibration monitoring system can monitor the operating status of the equipment in real time, detect and diagnose equipment failures in a timely manner, thereby ensuring the continuous, reliable, safe and efficient operation of the equipment.

[0052] The present invention belongs to the technical field of signal processing and is mainly used for equipment health monitoring and fault diagnosis in the machinery manufacturing industry.

[0053] like Figure 1 As shown, the present invention provides a compressor health diagnosis system, including a sensor, a data acquisition module, a data processing and analysis module, a fault repair and optimization module, an operating parameter adjustment module, and a preventive maintenance plan formulation module connected in sequence.

[0054] The sensors are arranged at key positions of the compressor to ensure that various operating data can be accurately obtained; the data acquisition module is used to collect key operating parameters of the compressor; wherein the operating parameters include temperature, pressure, vibration, flow, speed, power consumption, and exhaust volume.

[0055] Data processing and analysis module:

[0056] The data processing and analysis module is used to process the key operating parameters using a moving average method to achieve abnormality detection, and includes a real-time data monitoring unit, a moving average method application unit and an abnormality detection unit.

[0057] Real-time data monitoring unit: The real-time data monitoring unit is used to monitor the collected key operating parameters in real time to observe their changing trends;

[0058] Moving average method application unit: The moving average method application unit is used to process the key operating parameters using the moving average method to smooth data fluctuations and determine the trend or periodicity of the data;

[0059] The moving average method is used to process the collected signal and can realize the calculation of high-speed low-pass filtering.

[0060] Moving average (MA), also known as rolling average or sliding average, is a calculation method in statistics that analyzes data points by creating a series of averages of different subsets of the entire data set. It is also a type of finite impulse response filter.

[0061] The moving average method is to move the data in the time series item by item, calculate the average of a certain number of items, and form a new sequence to reflect the trend or periodicity of the original sequence. Depending on the number of moving items, the moving average method can be divided into a simple moving average method and a weighted moving average method. The moving average sequence is usually smoother than the original data sequence, which helps to observe the long-term trend and periodic fluctuations of the data.

[0062] The calculation formula of the simple moving average method is:

[0063]

[0064] Among them, F t is the moving average value at time t, A i is the i-th data point in the time series; n is the number of moving average items; Example

[0065] Suppose there is a set of compressor energy consumption data (unit: kWh), and you want to analyze its trend using the simple moving average method. The data is: [100, 105, 110, 108, 102, 107], and n=3.

[0066] For F3 (i.e. the moving average after the third data point):

[0067]

[0068] Similarly, moving averages for other time points can be calculated.

[0069] The weighted moving average method is based on the simple moving average method, which assigns different weights to data at different time points. Generally, the closer the data is to the current time point, the greater the weight is. The calculation formula is:

[0070]

[0071] Among them, wi is the weight of the i-th data point.

[0072] Example

[0073] Continue to use the energy consumption data from above, but this time use the weighted moving average method with weights [1,2,3] (i.e. the most recent data point has the largest weight).

[0074] For F3:

[0075]

[0076] In compressors, the moving average method can be used to: monitor the energy consumption changes of the compressor and identify abnormal energy consumption points; analyze the operating efficiency of the compressor and predict future operating trends; and combine other data (such as temperature, pressure, etc.) to conduct fault warning and diagnosis. The moving average method can provide a clearer understanding of the operating status of the compressor and provide strong support for equipment maintenance and management.

[0077] The data processing steps of the moving average method application unit include:

[0078] S1. Collect real-time sensor data of the compressor, including but not limited to temperature, vibration, and pressure signals;

[0079] During the operation of the compressor, multiple sensor data are collected, including but not limited to: vibration sensor (sampling frequency 1kHz), temperature sensor (sampling frequency 10Hz), pressure sensor (sampling frequency 50Hz); the data of each sensor is converted into a digital signal through an A / D converter and then enters the data acquisition module.

[0080] S2, preprocessing the collected data to remove obvious outliers and noise;

[0081] The sensor data is denoised and the three-point median filter method is used to remove obvious abnormal data;

[0082] The data is normalized to ensure that the units and magnitudes of data from different sensors are consistent, which facilitates subsequent moving average processing.

[0083] S3, inputting the preprocessed data into a moving average algorithm for smoothing to reduce data noise;

[0084] For the normalized data, the moving average method is used for smoothing. The specific steps are as follows:

[0085] For the time series data of each sensor, a sliding window is set with a window size of M (e.g., M=10);

[0086] For the data in each sliding window, calculate its arithmetic mean, the formula is as follows:

[0087]

[0088] Wherein, x[k] represents the sample value of index position k in the signal sequence.

[0089] The smoothed data is used to reduce the noise in the signal, making the subsequent feature extraction more accurate, such as Figure 2 As shown, the broken line part is the data before noise processing, and the curved line part is the data after noise processing; the data after moving average processing will be used as the input for subsequent fault feature extraction.

[0090] S4. Extract features from the smoothed sensor data to identify possible failure modes; specifically, including:

[0091] S41, performing frequency domain analysis on the vibration data, and extracting spectrum features using fast Fourier transform (FFT);

[0092] S42, performing trend analysis on the temperature data to extract the temperature change slope;

[0093] S43, performing statistical analysis on the pressure data to extract its mean and variance;

[0094] S44, inputting the extracted features into a fault diagnosis model, where the diagnosis model is based on a machine learning classifier (e.g., a support vector machine, a decision tree) or a rule-based system;

[0095] S45, the diagnostic model first classifies the input features and matches them with the features in the preset fault mode library;

[0096] If the match is higher than the threshold, the system generates a fault alarm and indicates the potential fault type (such as bearing wear, overheating, etc.).

[0097] S5. Input the fault mode into the diagnosis model to generate a fault diagnosis conclusion of the compressor.

[0098] Anomaly detection unit: The anomaly detection unit is used to compare real-time operating parameters with preset thresholds or historical data to identify abnormal data points.

[0099] Fault repair and optimization module:

[0100] The fault repair and optimization module is used to repair the fault in a timely manner according to the fault type and location in the diagnosis conclusion; perform performance tests on the repaired compressor to ensure that the fault has been completely eliminated and restored to the optimal operating state, analyze the cause of the fault, and make necessary optimization adjustments to the compressor to prevent similar faults from occurring again.

[0101] Run the parameter adjustment module:

[0102] The operating parameter adjustment module is used to adjust the operating parameters (such as speed, pressure, temperature, etc.) according to the actual operating conditions and load requirements of the compressor to optimize the operating efficiency of the compressor; and set the operating parameter range to ensure that the compressor operates in an efficient and stable state.

[0103] Preventive maintenance planning module:

[0104] The preventive maintenance plan formulation module is used to formulate a preventive maintenance plan based on the diagnosis results and the historical operation data of the compressor, including regular inspection, maintenance, and replacement of wearing parts to prevent the occurrence of potential failures.

[0105] The compressor health diagnosis system provided by the present invention is a monitoring information processing process, which realizes real-time online monitoring of equipment status parameters such as vibration, temperature, displacement, pressure, and speed of the unit. It can determine the fault location and diagnose the cause of the fault through analysis methods such as indicator diagrams, frequency spectra, and trend analysis. At the same time, the status data is saved in a database, necessary information is provided for equipment maintenance, predictive maintenance is realized, and the compressor unit can operate safely and stably for a long period of time. Therefore, it is of great significance in the application of reciprocating compressors.

[0106] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, device, article or method including a series of elements includes not only those elements, but also includes other elements not explicitly listed, or also includes elements inherent to such process, device, article or method. In the absence of further restrictions, an element defined by the sentence "includes a ..." does not exclude the presence of other identical elements in the process, device, article or method including the element.

[0107] The above description is only a preferred embodiment of the present invention, and does not limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made by using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A compressor health diagnosis system, characterized in that: It includes sensors, data acquisition modules, and data processing and analysis modules connected in sequence; The sensors are arranged at key positions of the compressor to ensure that various operating data can be accurately obtained; The data acquisition module is used to collect key operating parameters of the compressor; wherein the operating parameters include temperature, pressure, vibration, flow, speed, power consumption, and exhaust volume; The data processing and analysis module is used to process the key operating parameters using a moving average method to achieve abnormality detection.

2. The compressor health diagnosis system according to claim 1, characterized in that: The data processing and analysis module includes a real-time data monitoring unit, a moving average method application unit and an anomaly detection unit. The real-time data monitoring unit is used to monitor the collected key operating parameters in real time to observe their changing trends; The moving average method application unit is used to process the key operating parameters using the moving average method to smooth data fluctuations and determine the trend or periodicity of the data; The anomaly detection unit is used to compare the real-time operating parameters with preset thresholds or historical data to identify abnormal data points.

3. The compressor health diagnosis system according to claim 2, characterized in that: In the moving average method application unit, the calculation formula of the simple moving average method is: Among them, F t is the moving average value at time t, A i is the i-th data point in the time series; n is the number of moving average items; The weighted moving average method is: Among them, wi is the weight of the i-th data point.

4. The compressor health diagnosis system according to claim 3, characterized in that: The data processing steps of the moving average method application unit include: Collect real-time sensor data of the compressor, including but not limited to temperature, vibration, and pressure signals; Preprocess the collected data to remove obvious outliers and noise; The preprocessed data is input into the moving average algorithm for smoothing to reduce data noise; Perform feature extraction on the smoothed data to identify possible failure modes; The fault mode is input into the diagnosis model to generate the fault diagnosis conclusion of the compressor.

5. The compressor health diagnosis system according to claim 4, characterized in that: The step of preprocessing the collected data includes: The sensor data is denoised and the three-point median filter method is used to remove obvious abnormal data; The data is normalized to ensure that the units and magnitudes of data from different sensors are consistent, which facilitates subsequent moving average processing.

6. The compressor health diagnosis system according to claim 5, characterized in that: The step of inputting the pre-processed data into a moving average algorithm and performing smoothing processing to reduce data noise includes: For the time series data of each sensor, set a sliding window with a window size of M; For the data in each sliding window, calculate its arithmetic mean, the formula is as follows: Wherein, x[k] represents the sample value of index position k in the signal sequence.

7. The compressor health diagnosis system according to claim 6, characterized in that: The step of extracting features from the smoothed data and identifying possible failure modes includes: Perform frequency domain analysis on vibration data and extract spectrum features using Fast Fourier Transform (FFT); Perform trend analysis on temperature data and extract the temperature change slope; Perform statistical analysis on pressure data and extract its mean and variance; The extracted features are input into a fault diagnosis model, which is a machine learning classifier or a rule-based system; The diagnostic model first classifies the input features and matches them with the features in the preset fault mode library; If the matching degree is higher than the threshold, the system generates a fault alarm and indicates the potential fault type.

8. The compressor health diagnosis system according to claim 1, characterized in that: It also includes a fault repair and optimization module, which is used to repair the fault in a timely manner according to the fault type and location in the diagnosis conclusion; perform performance tests on the repaired compressor to ensure that the fault has been completely eliminated and restored to the optimal operating state, analyze the cause of the fault, and make necessary optimization adjustments to the compressor.

9. The compressor health diagnosis system according to claim 1, characterized in that: It also includes an operating parameter adjustment module, which is used to adjust the operating parameters according to the actual operating conditions and load requirements of the compressor to optimize the operating efficiency of the compressor; and set the operating parameter range to ensure that the compressor operates in an efficient and stable state.

10. The compressor health diagnosis system according to claim 1, characterized in that: It also includes a preventive maintenance plan making module, which is used to make a preventive maintenance plan based on the diagnosis results and the historical operation data of the compressor, including regular inspection, maintenance, and replacement of wearing parts to prevent the occurrence of potential failures.

Images