A method and system for monitoring compressor operation status

Abstract

The present invention relates to the field of intelligent control technology, and in particular to a method and system for monitoring the operating status of a compressor. The method includes: obtaining pressure sequences and temperature sequences at historical moments of the compressor, as well as an ideal compression cycle; obtaining multiple pressure analysis groups of the pressure sequence arranged in time sequence; determining the first, second, and third abnormal values ​​of the pressure analysis group; recording the average of the first, second, and third abnormal values ​​as the total abnormal value of the target analysis group; determining the abnormal pressure group; determining the attention of the abnormal pressure group; obtaining the final penalty parameter; and training the SVM model with the final penalty parameter to realize the monitoring of the operating status of the compressor. The present invention can specifically mine the abnormal conditions in the compressor operating status data by determining three groups of abnormal values ​​of the pressure analysis group and judging whether it is an abnormal pressure group based on the total abnormal value, thereby avoiding the problem of abnormal data being ignored due to data imbalance.

Description

Technical Field

[0001] The present invention relates to the field of intelligent control technology, and in particular to a compressor operating status monitoring method and system. Background Art

[0002] A compressor is a device used to compress various gases. The working principle of a reciprocating piston compressor is to suck gas into the cylinder through the reciprocating piston, and then compress the gas and discharge it, so as to increase the gas pressure. It is usually composed of cylinders, pistons, crankshafts, connecting rods, valves and other components. These components work together to achieve gas compression. Since the compressor is in a state of high temperature, high pressure and high speed for a long time, it is prone to mechanical wear or failure, so it is necessary to monitor the operating status of the compressor. Support Vector Machine (SVM) is a powerful supervised learning algorithm that can be used to monitor the operating status of the compressor. Its core idea is to distinguish different categories of data points by finding an optimal hyperplane and maximize the interval between different categories; among them, the penalty parameter C of SVM is used to control the degree of penalty for misclassification. A larger C value will make the SVM model pay more attention to the misclassified samples, while a smaller C value will make the SVM model pay more attention to the smoothness of the decision boundary.

[0003] The patent document with publication number CN112733603B discloses a variable frequency scroll compressor fault diagnosis method based on improved VMD and SVM. The method processes and analyzes the vibration signal of the variable frequency scroll compressor from three aspects: signal processing, feature extraction and classification recognition. First, a vibration test experiment of the variable frequency scroll compressor is carried out, and vibration acceleration signals are collected respectively for the normal state, scroll plate fault, bearing fault and crankshaft fault of the variable frequency scroll compressor to obtain data samples of different fault types; using the envelope entropy-correlation index as the fitness function, the sparrow search algorithm (SSA) is used to optimize the variational mode decomposition (VMD) algorithm to process the vibration signal of the variable frequency scroll compressor to obtain the intrinsic mode functions of different scales; the multi-scale permutation entropy of different intrinsic mode functions is calculated to form a feature vector, which is input into a classifier established based on a support vector machine (SVM) for training and prediction classification to determine the fault type of the variable frequency scroll compressor.

[0004] However, the above patent document does not solve the problem of data imbalance in the compressor operating status data in an actual industrial environment, that is, the abnormal data (minority class) is far less than the normal data (majority class). This data imbalance will affect the training of the SVM model, making it biased towards the majority class (normal data), thereby reducing the detection ability of the minority class (abnormal data). Summary of the invention

[0005] In order to solve the problem of imbalance in compressor operating status data, which affects the training of the SVM model and reduces the ability to detect abnormal data, the present invention provides a compressor operating status monitoring method and system.

[0006] In a first aspect, the present invention provides a method for monitoring the operating status of a compressor, which adopts the following technical solution: A compressor operation status monitoring method comprises: obtaining a pressure sequence, a temperature sequence and an ideal compression cycle of the compressor at historical moments; obtaining a plurality of pressure analysis groups of the pressure sequence arranged in time sequence, wherein the two ends of the pressure analysis group are respectively a maximum point in the pressure sequence and a minimum point adjacent to the maximum point; for a target analysis group in the pressure analysis group, according to the moments corresponding to the extreme points in the target analysis group and the next pressure analysis group, and the ideal compression cycle, determining a first abnormal value used to reflect the periodic characteristics of the target analysis group at the moment; according to the pressure values ​​of the extreme points in the target analysis group and the next pressure analysis group, determining a first abnormal value used to reflect the amplitude characteristics of the pressure data of the target analysis group the second abnormal value; according to the time corresponding to the extreme point in the target analysis group and the pressure value of the extreme point in the target analysis group, determine the third abnormal value used to reflect the target analysis group as a whole; record the average of the first, second and third abnormal values ​​as the total abnormal value of the target analysis group; in response to the total abnormal value being greater than the abnormal threshold, identify the target analysis group as an abnormal pressure group; perform a linear fit on the temperature sequence corresponding to the abnormal pressure group to obtain the fitting slope of the abnormal pressure group, and determine the degree of attention used to reflect the influence of temperature on the abnormal pressure group according to the size of the fitting slope; perform a linear mapping on the preset penalty parameter according to the degree of attention to obtain the final penalty parameter; train the final penalty parameter Model is used to monitor the operating status of the compressor.

[0007] The beneficial effects are as follows: by introducing three abnormal values ​​(the first, second, and third abnormal values) of the pressure analysis group and calculating their mean, the abnormal conditions of the compressor during operation can be identified more comprehensively and accurately; by fitting the temperature sequence corresponding to the abnormal pressure group, calculating its fitting slope, and then determining the attention of the abnormal group, the SVM model can focus on the moments and states that have a greater impact on the abnormal conditions during the training process, thereby improving the sensitivity and accuracy of abnormality detection; there is often a situation in the compressor operation data that the amount of normal state data is much larger than the abnormal state data, which makes the model prone to normal state during training. By linearly mapping the penalty parameters based on the attention, the attention to the abnormal data is dynamically adjusted, so that the SVM model can better adapt to unbalanced data during the training process, thereby enhancing its detection ability under abnormal conditions; through the above-mentioned targeted data processing methods, the SVM model can more accurately identify and judge abnormal conditions in complex compressor operation conditions, thereby improving its stability and prediction accuracy.

[0008] Furthermore, the pressure sequence and temperature sequence are pressure sequences and temperature sequences that have been subjected to filtering, denoising, and normalization processing.

[0009] Furthermore, the first abnormal value satisfies: ; In the formula, For the The first outlier in the stress analysis group, and Respectively The time corresponding to the maximum and minimum points in the pressure analysis group, and Respectively The time corresponding to the maximum and minimum points in the pressure analysis group, is the ideal compression cycle, is the standard normalization function, is the absolute value symbol.

[0010] The beneficial effects are: by calculating the difference between the extreme point moments of the current and next pressure analysis groups, abnormal fluctuations in the compressor pressure sequence can be effectively identified; the ideal compression cycle is introduced to ensure that the abnormal value judgment can adapt to the difference between the ideal working state and the actual pressure change, thereby improving the accuracy of the judgment; combining the extreme point information of the previous and next pressure groups, the system's ability to cope with complex fluctuations is enhanced, misjudgment is reduced, and monitoring accuracy is improved.

[0011] Furthermore, the second abnormal value satisfies: ; In the formula, For the The second outlier in the stress analysis group, and Respectively The pressure values ​​of the maximum and minimum points in the pressure analysis group, and Respectively The pressure values ​​of the maximum and minimum points in the pressure analysis group, is the standard normalization function, is the absolute value symbol.

[0012] The beneficial effects are: by comparing the pressure difference of the extreme points of the current and next pressure analysis groups, more complex abnormal fluctuations can be effectively identified; the normalization process eliminates the influence of irregular fluctuations and noise, and improves the accuracy and stability of outlier detection.

[0013] Furthermore, the third abnormal value satisfies: ; In the formula, For the The third outlier in the stress analysis group, and Respectively The stress analysis group and The moment corresponding to the maximum value in the pressure analysis group, and Respectively The stress analysis group and The maximum pressure value in the pressure analysis group, is the mean of twice the time difference between the maximum and minimum points of all pressure analysis groups, is the mean of the pressure value difference between the maximum and minimum points of all pressure analysis groups, is the standard normalization function, is the absolute value symbol.

[0014] The beneficial effects are: by combining the time difference and the pressure difference, the anomalies in the pressure change can be identified more comprehensively; by comparing with the global mean, the anomalies that deviate from the normal fluctuation can be effectively discovered, thereby improving the accuracy of detection; by introducing the mean parameter, the system can adapt to the anomaly threshold under different states and enhance flexibility; by comprehensively considering the time and pressure differences, the accuracy of anomaly detection and the robustness of the system are improved.

[0015] Furthermore, the attention level of the abnormal pressure group satisfies: ; In the formula, For the The attention of abnormal pressure groups, For the The fitted slope of the abnormal pressure group, and are respectively a preset first threshold and a preset second threshold, For the Total abnormal values ​​of abnormal pressure groups, For the The variance of the temperature values ​​in the temperature series of the time period corresponding to the abnormal pressure group, For the The mean of the slopes of all two adjacent temperature values ​​in the temperature series of the time period corresponding to the abnormal pressure group, is the standard normalization function, The mathematical symbol for "or".

[0016] The beneficial effects are: combining temperature variance and slope mean, considering the impact of temperature changes on abnormal pressure, and improving the accuracy of attention calculation; dynamically selecting variance or slope according to different values ​​to calculate attention, so that the system can flexibly respond to different abnormal situations, and when the fitting slope deviates from the normal range, relying on the slope to identify extreme anomalies and enhance the sensitivity of detection; using variance within the normal range to reduce the impact of small fluctuations on attention and improve system stability.

[0017] Further, the monitoring of the compressor operation status is realized by using the trained The model identifies anomalies in the compressor's current pressure and temperature data and completes compressor operating status monitoring.

[0018] In a second aspect, the present invention provides a compressor operation status monitoring system, which adopts the following technical solution: A compressor operation status monitoring system comprises: a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, the above-mentioned compressor operation status monitoring method is implemented.

[0019] By adopting the above technical solution, the above-mentioned compressor operation status monitoring method is generated into a computer program and stored in a memory so as to be loaded and executed by a processor, thereby making a terminal device based on the memory and the processor for easy use.

[0020] The present invention has the following technical effects: By determining the three groups of abnormal values ​​in the pressure analysis group and judging whether it is an abnormal pressure group based on the total abnormal value, the abnormal conditions in the compressor operation status data can be specifically mined, avoiding the problem of abnormal data being ignored due to data imbalance, so that the model can pay more attention to abnormal data and improve the detection ability of abnormal data; not only can the abnormal pressure group be identified, but also the temperature series corresponding to the abnormal pressure group can be fitted to obtain the fitting slope, so as to determine the attention of the abnormal pressure group, which can more carefully measure the severity or development trend of the abnormality and provide a more accurate basis for subsequent processing; according to the attention, the preset penalty parameter is linearly mapped to obtain the final penalty parameter for training SV The M model enables the model to dynamically adjust the degree of emphasis on different samples according to the actual abnormal situation during the training process, making the training of the SVM model more reasonable and effective, thereby improving the accuracy and reliability of the model in monitoring the compressor operating status and better adapting to the characteristics of the compressor operating status data and actual monitoring needs; it comprehensively considers the information of the pressure sequence and temperature sequence, analyzes and judges the compressor operating status from multiple angles, and can more comprehensively reflect the operating status of the compressor, reduce misjudgments and missed judgments, and improve the performance and stability of the entire compressor operating status monitoring system, which helps to timely discover potential problems in the operation of the compressor and ensure the safe and stable operation of the compressor. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] By reading the detailed description below with reference to the accompanying drawings, the above and other purposes, features and advantages of the exemplary embodiments of the present invention will become readily understood. In the accompanying drawings, several embodiments of the present invention are shown in an exemplary and non-restrictive manner, and the same or corresponding reference numerals represent the same or corresponding parts.

[0022] Figure 1 It is a flow chart of a method for monitoring the operating status of a compressor according to an embodiment of the present invention.

[0023] Figure 2 It is a schematic diagram of a pressure analysis group in a compressor operation status monitoring method according to an embodiment of the present invention. DETAILED DESCRIPTION

[0024] 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 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 those skilled in the art without creative work are within the scope of protection of the present invention.

[0025] It should be understood that when the terms "first", "second", etc. are used in the claims, descriptions, and drawings of the present invention, they are only used to distinguish different objects, rather than to describe a specific order. The terms "include" and "comprise" used in the description and claims of the present invention indicate the presence of the described features, wholes, steps, operations, elements, and / or components, but do not exclude the presence or addition of one or more other features, wholes, steps, operations, elements, components, and / or their collections.

[0026] The embodiment of the present invention discloses a method for monitoring the operating status of a compressor, referring to Figure 1 , comprising steps S1 to S8: S1: Obtain the pressure sequence and temperature sequence of the compressor at historical moments, as well as the ideal compression cycle.

[0027] A temperature sensor is installed on the outer wall of each cylinder of the compressor to monitor the operating temperature of the cylinder, helping to identify changes in the heat load in the cylinder, the efficiency of the cooling system, and potential overheating problems. A pressure sensor is installed at the exhaust port of the compressor to monitor the exhaust pressure of the compressor, reflecting the load of the compressor and potential exhaust system problems, such as exhaust valve failure. The ideal compression cycle is the time required to complete a complete compression process.

[0028] Implementers can set the collection frequency according to specific implementation conditions, for example, 1s / time.

[0029] It should be noted that in an industrial environment, the data collected by sensors are usually interfered by various noises, which may come from mechanical vibration, electromagnetic interference, environmental noise, etc. In order to improve the quality of the data, the data collected by temperature sensors and pressure sensors need to be preprocessed.

[0030] Specifically, the preprocessing process includes: performing Kalman filter algorithm to denoise the data collected by the temperature sensor and the pressure sensor, and normalizing the results to obtain a pressure sequence and a temperature sequence.

[0031] S2: Obtain multiple time-sequentially arranged pressure analysis groups of the pressure sequence.

[0032] The two ends of the pressure analysis group are respectively the maximum point in the pressure sequence and the minimum point adjacent to the maximum point; Figure 2 As shown, the horizontal axis is time, vertical axis is the pressure value of the pressure series, the central axis It is a compressed equilibrium state (the initial position of the piston rod is in the middle of the air chamber), the center axis The upper part indicates that the pressure data is greater than 0, indicating compression (the piston rod is close to the leftmost side of the air cavity); the middle axis The pressure data below is less than 0, indicating that it is stretched (the piston rod is close to the rightmost side of the air cavity); starting from the initial moment, each pair of adjacent maximum and minimum points is divided into a pressure analysis group, and each pressure analysis group is not repeated.

[0033] Specifically, the maximum point and the minimum point satisfy: ; In the formula, is an extreme point, when is the maximum point, when When is the minimum point, For the The value of the pressure sequence at a certain moment, For the The value of the pressure sequence at a certain moment, For the The value of the pressure sequence at a certain moment, The mathematical symbol for "and".

[0034] Among them, when At that time, Pressure data at each moment Greater than Pressure data at all times , and is greater than Pressure data at all times hour, Assign a value of 1, reflecting the Pressure data at each moment is the maximum value point in the pressure data; when At that time, Pressure data at each moment Less than Pressure data at all times , and less than Pressure data at each moment hour, Assign a value of 0 to reflect the Pressure data at each moment is the minimum point in the pressure data.

[0035] S3: Determine the first, second, and third outliers of the stress analysis group.

[0036] It should be noted that under normal conditions, the temperature of the outer wall of the cylinder will gradually increase with the operation of the compressor and reach a stable temperature range. The temperature change should be slow and stable without sharp fluctuations. Under normal conditions, the exhaust port pressure will change periodically with the operation cycle of the compressor, but generally remain within the design pressure range. The pressure change should be smooth without abnormal high or low pressure. The relationship between temperature and pressure in the compressor: (1) Temperature increase may cause gas expansion, thereby increasing the exhaust port pressure; conversely, abnormal pressure will also affect temperature changes; for example, excessive pressure may cause cylinder overheating, and there is a positive correlation between temperature and pressure. (2) The reciprocating piston compressor realizes gas compression and exhaust through the reciprocating motion of the piston in the cylinder; specifically, the suction stage: the piston moves downward and the gas is sucked into the cylinder; the compression stage: the piston moves upward and the gas is compressed; the exhaust stage: the piston continues to move upward and the compressed gas is discharged from the cylinder; the exhaust end stage: the piston reaches the top dead center and the exhaust ends. Since the reciprocating motion of the piston is periodic, the pressure at the exhaust port will also show periodic changes. From the perspective of time, if there is no abnormality in the two pressure analysis groups, their periods should be the same; and the period difference can be obtained by the time difference corresponding to the two maximum points and the time difference corresponding to the two minimum points in the two pressure analysis groups ( ); At the same time, when there is no anomaly, the cycle obtained based on the extreme point ( ) should be closer to 2 times the ideal duty cycle of the compressor .

[0037] For a target analysis group in the pressure analysis group, a first abnormal value reflecting the periodic characteristics of the target analysis group at the time is determined according to the time corresponding to the extreme point in the target analysis group and the next pressure analysis group, and the ideal compression cycle.

[0038] Specifically, the first abnormal value satisfies: ; In the formula, For the The first outlier in the stress analysis group, and Respectively The time corresponding to the maximum and minimum points in the pressure analysis group, and Respectively The time corresponding to the maximum and minimum points in the pressure analysis group, is the ideal compression cycle, is the standard normalization function, is the absolute value symbol.

[0039] Among them, the first abnormal value reflects the abnormal degree of the pressure analysis group in the time period characteristics; when the time difference corresponding to the two maximum points in the two pressure analysis groups The difference in time corresponding to the two minimum points The smaller the difference is, the smaller the cycle obtained according to the extreme point Twice the ideal duty cycle of the compressor The closer the ratio is to 1, that is, the smaller the difference between this ratio and 1, the The smaller the first outlier value of the stress analysis group, the smaller the first outlier value of the stress analysis group, and vice versa; The two obtained periods are used. If the period is more accurate, the two should be closer and the difference will be smaller. The closer the obtained cycle is to the ideal cycle, the smaller the difference from 1 is, and the abnormal situation of the cycle characteristics at the moment is comprehensively determined from two aspects.

[0040] It should be noted that, from the perspective of pressure data (amplitude), if there is no abnormality in the pressure analysis group, then in each pressure analysis group, the difference between the maximum point and the minimum point is the same, and in the next adjacent pressure analysis group, the fluctuation degree of the maximum point and the fluctuation degree of the minimum point are both small. The fluctuation degree of the maximum point and the fluctuation degree of the minimum point can directly reflect the changes in the maximum and minimum values ​​of the pressure data in the pressure analysis group.

[0041] According to the pressure values ​​of the extreme points in the target analysis group and the next pressure analysis group, a second abnormal value for reflecting the amplitude characteristics of the pressure data of the target analysis group is determined.

[0042] Specifically, the second abnormal value satisfies: ; In the formula, For the The second outlier in the stress analysis group, and Respectively The pressure values ​​of the maximum and minimum points in the pressure analysis group, and Respectively The pressure values ​​of the maximum and minimum points in the pressure analysis group, is the standard normalization function, is the absolute value symbol.

[0043] Among them, the second outlier reflects the degree of abnormality of the pressure data amplitude characteristics of the pressure analysis group; when the difference between the maximum and minimum points of the two pressure analysis groups is The smaller the value, the greater the difference between the maximum value points in the adjacent next pressure analysis group. And the difference of the minimum The difference can reflect the fluctuation degree of the pressure data between the maximum and minimum values ​​in the single pressure analysis group. The smaller the second outlier value of the stress analysis group, the smaller the second outlier value of the stress analysis group, and vice versa.

[0044] It should be noted that, combined with the amplitude difference of the time and pressure data, the analysis is performed from an overall perspective: if the abnormality of the pressure analysis group is small, the average time difference and the average amplitude difference between this pressure analysis group and all pressure analysis groups are small.

[0045] According to the time corresponding to the extreme point in the target analysis group and the pressure value of the extreme point in the target analysis group, a third abnormal value reflecting the target analysis group as a whole is determined.

[0046] Specifically, the third abnormal value satisfies: ; In the formula, For the The third outlier in the stress analysis group, and Respectively The stress analysis group and The moment corresponding to the maximum value in the pressure analysis group, and Respectively The stress analysis group and The maximum pressure value in the pressure analysis group, is the mean of twice the time difference between the maximum and minimum points of all pressure analysis groups, is the mean of the pressure value difference between the maximum and minimum points of all pressure analysis groups, is the standard normalization function, is the absolute value symbol.

[0047] Among them, the third abnormal value can reflect the overall abnormal situation of the pressure analysis group in terms of time and amplitude; Twice the time difference between the maximum and minimum points in the pressure analysis group , that is, the cycle time of the first pressure analysis group, and the average time difference of all pressure analysis groups The smaller the difference, the The difference in pressure data between the maximum and minimum points in the pressure analysis group , that is, the amplitude difference of the first pressure analysis group and the average amplitude difference of all pressure analysis groups The smaller the difference, the The smaller the third outlier value of the first stress analysis group, the smaller the third outlier value, and vice versa.

[0048] S4: Record the mean of the first, second, and third outliers as the total outlier value of the target analysis group.

[0049] Among them, the first, second and third abnormal values ​​reflect the abnormal conditions of the pressure analysis group in terms of time, pressure data (amplitude) and overall respectively; when the first, second and third abnormal values ​​of the pressure analysis group are larger, the total abnormal value is larger, and vice versa.

[0050] S5: Identify abnormal pressure groups.

[0051] In response to the total abnormal value being greater than the abnormal threshold, the target analysis group is identified as an abnormal pressure group.

[0052] Implementers can set the abnormal threshold according to specific implementation conditions, for example, 0.7.

[0053] S6: Determine the attention level of the abnormal stress group.

[0054] It should be noted that the temperature data will exist in two states, one is the heating state, and the fluctuation of the fitting slope between the temperature data at adjacent moments is small; the other is that the temperature reaches a stable state, that is, the slope between the temperature data at adjacent moments is approximately 0, and the fluctuation is small.

[0055] A straight-line fitting is performed on the temperature sequence corresponding to the abnormal pressure group to obtain a fitting slope of the abnormal pressure group, and the degree of attention reflecting the influence of temperature on the abnormal pressure group is determined according to the size of the fitting slope.

[0056] Specifically, the attention level of the abnormal pressure group meets the following requirements: ; In the formula, For the The attention of abnormal pressure groups, For the The fitted slope of the abnormal pressure group, and are respectively a preset first threshold and a preset second threshold, For the Total abnormal values ​​of abnormal pressure groups, For the The variance of the temperature values ​​in the temperature series of the time period corresponding to the abnormal pressure group, For the The mean of the slopes of all two adjacent temperature values ​​in the temperature series of the time period corresponding to the abnormal pressure group, is the standard normalization function, The mathematical symbol for "or".

[0057] Implementers may set the first threshold and the second threshold according to specific implementation conditions. For example, the first threshold is 0.7 and the second threshold is 0.3.

[0058] Among them, if , it can be considered that The fitting slope of the temperature data corresponding to the abnormal pressure group is approximately 0, so The more abnormal pressure groups are likely to be in a temperature stable state, the more likely they are to be in a temperature stable state. Abnormality level of abnormal stress group The larger the value, the corresponding variance of the temperature data The larger the The more abnormal pressure groups there are, the greater the corresponding attention, and vice versa; if , it can be considered that If the fitting slope of the temperature data corresponding to the abnormal pressure group is not approximately 0, then The more abnormal pressure groups are likely to be in a temperature rise state, the more likely they are to be in a temperature rise state. Abnormality level of abnormal stress group The larger the value, the higher the average slope between the corresponding pairwise temperature data. The larger the The higher the abnormal pressure group, the greater the corresponding attention, and vice versa.

[0059] S7: Obtain the final penalty parameter.

[0060] It should be noted that in the case of data imbalance, you can try to increase the penalty parameter, that is, the C value, so that the model pays more attention to the classification of the minority class (abnormal data), thereby improving the ability to detect abnormal data. Specifically, you can set different C values ​​for cross-validation and select the C value that performs best on the validation set. The attention level indicates the degree of abnormality and importance of each abnormal pressure group. The higher the value, the more attention the model should pay to the data of the abnormal pressure group.

[0061] The preset penalty parameter is linearly mapped according to the attention degree to obtain a final penalty parameter.

[0062] Specifically, the Abnormal stress group needs attention For example, use linear mapping to transform Abnormal stress group needs attention Convert to The final penalty parameter for the abnormal pressure group ; Linear mapping: , is the minimum penalty parameter, set to a small positive number, for example, ; is the maximum penalty parameter, exemplarily, ;in, Should follow increases with the increase of yes is a monotonically increasing function of .

[0063] S8: Training by the final penalty parameter Model is used to monitor the operating status of the compressor.

[0064] Specifically, the monitoring of the compressor operation status includes: Using trained The model identifies anomalies in the compressor's current pressure and temperature data and completes compressor operating status monitoring.

[0065] An embodiment of the present invention further discloses a compressor operation status monitoring system, including a processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a compressor operation status monitoring method according to the present invention is implemented.

[0066] The above system also includes other components well known to those skilled in the art, such as a communication bus and a communication interface, and their configuration and functions are known in the art, so they will not be described in detail here.

[0067] In the present invention, the aforementioned memory may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, a computer-readable storage medium may be any suitable magnetic storage medium or magneto-optical storage medium, or any other medium that can be used to store the required information and can be accessed by an application, module, or both. Any such computer storage medium may be part of the device or accessible or connectable to the device.

[0068] Although this specification has shown and described a number of embodiments of the present invention, it will be apparent to those skilled in the art that such embodiments are provided by way of example only. Those skilled in the art will conceive of many modifications, changes and alternatives without departing from the ideas and spirit of the present invention. It should be understood that in the practice of the present invention, various alternatives to the embodiments of the present invention described herein may be employed.

[0069] The above are all preferred embodiments of the present invention, and are not intended to limit the protection scope of the present invention. Therefore, any equivalent changes made based on the structure, shape, and principle of the present invention should be included in the protection scope of the present invention.

Claims

1. A method for monitoring the operating status of a compressor, characterized in that: include: Obtain the pressure sequence, temperature sequence and ideal compression cycle of the compressor at historical moments; Acquire multiple pressure analysis groups of the pressure sequence arranged in time sequence, wherein two ends of the pressure analysis group are respectively a maximum value point in the pressure sequence and a minimum value point adjacent to the maximum value point; For a target analysis group in the pressure analysis group, according to the time corresponding to the extreme value point in the target analysis group and the next pressure analysis group, and the ideal compression cycle, determine a first abnormal value for reflecting the periodic characteristics of the target analysis group at the time; According to the pressure values ​​of the extreme points in the target analysis group and the next pressure analysis group, a second abnormal value for reflecting the amplitude characteristics of the pressure data of the target analysis group is determined; according to the time corresponding to the extreme point in the target analysis group and the pressure value of the extreme point in the target analysis group, a third abnormal value for reflecting the target analysis group as a whole is determined; the average of the first, second and third abnormal values ​​is recorded as the total abnormal value of the target analysis group; in response to the total abnormal value being greater than the abnormal threshold, the target analysis group is identified as an abnormal pressure group; Performing a straight-line fitting on the temperature sequence corresponding to the abnormal pressure group to obtain a fitting slope of the abnormal pressure group, and determining the degree of attention for reflecting the influence of temperature on the abnormal pressure group according to the size of the fitting slope; Linearly mapping the preset penalty parameter according to the attention degree to obtain a final penalty parameter; The final penalty parameter is trained by Model is used to monitor the operating status of the compressor.

2. A compressor operating status monitoring method according to claim 1, characterized in that: The pressure sequence and temperature sequence are pressure sequences and temperature sequences that have been subjected to filtering, denoising, and normalization processing.

3. A compressor operating status monitoring method according to claim 1, characterized in that: The first abnormal value satisfies: ； In the formula, For the The first outlier in the stress analysis group, and Respectively The time corresponding to the maximum and minimum points in the pressure analysis group, and Respectively The time corresponding to the maximum and minimum points in the pressure analysis group, is the ideal compression cycle, is the standard normalization function, is the absolute value symbol.

4. A compressor operating status monitoring method according to claim 1, characterized in that: The second outlier satisfies: ； In the formula, For the The second outlier in the stress analysis group, and Respectively The pressure values ​​of the maximum and minimum points in the pressure analysis group, and Respectively The pressure values ​​of the maximum and minimum points in the pressure analysis group, is the standard normalization function, is the absolute value symbol.

5. A compressor operating status monitoring method according to claim 1, characterized in that: The third abnormal value satisfies: ； In the formula, For the The third outlier in the stress analysis group, and Respectively The stress analysis group and The moment corresponding to the maximum value in the pressure analysis group, and Respectively The stress analysis group and The maximum pressure value in the pressure analysis group, is the mean of twice the time difference between the maximum and minimum points of all pressure analysis groups, is the mean of the pressure value difference between the maximum and minimum points of all pressure analysis groups, is the standard normalization function, is the absolute value symbol.

6. A compressor operating status monitoring method according to claim 1, characterized in that: The attention level of the abnormal pressure group meets the following requirements: ； In the formula, For the The attention of abnormal pressure groups, For the The fitted slope of the abnormal pressure group, and are respectively a preset first threshold and a preset second threshold, For the Total abnormal values ​​of abnormal pressure groups, For the The variance of the temperature values ​​in the temperature series of the time period corresponding to the abnormal pressure group, For the The mean of the slopes of all two adjacent temperature values ​​in the temperature series of the time period corresponding to the abnormal pressure group, is the standard normalization function, The mathematical symbol for "or".

7. A compressor operating status monitoring method according to claim 1, characterized in that: The monitoring of the compressor operation status is realized, including: Using trained The model identifies anomalies in the compressor's current pressure and temperature data and completes compressor operating status monitoring.

8. A compressor operation status monitoring system, characterized in that: include: A processor and a memory, wherein the memory stores computer program instructions, and when the computer program instructions are executed by the processor, a compressor operating status monitoring method according to any one of claims 1-7 is implemented.

Images