Intelligent monitoring method and system for seal performance of liquid-driven compressor

Abstract

The application provides an intelligent monitoring method and system for sealing performance of a liquid-driven compressor, wherein the method comprises: collecting high-frequency pressure data, circulating liquid flow data and friction sound signals of a sealing interface of a sealing cavity of the liquid-driven compressor; constructing a pressure-flow covariant relationship based on the high-frequency pressure data and the circulating liquid flow data; performing directional enhancement processing on the friction sound signals to form sound energy intensity distribution information; performing adaptive clustering processing on the sound energy intensity distribution information by using a DIANA algorithm to generate a sealing abnormality index; jointly analyzing the pressure-flow covariant relationship and the sealing abnormality index to generate a monitoring result representing the sealing performance of the liquid-driven compressor; and triggering a sealing failure monitoring alarm when the monitoring result exceeds a preset warning value. The application improves the timeliness and reliability of sealing performance monitoring.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of compressor sealing performance monitoring, in particular to an intelligent monitoring method and system for sealing performance of a liquid-driven compressor. BACKGROUND

[0002] Liquid-driven compressors are widely used in the fields of petrochemical industry and gas transportation, and their sealing performance directly affects the efficiency and safety of equipment operation. Since the compressor piston rod is in a high-speed reciprocating motion state for a long time, micro-leakage may occur at the sealing interface, and traditional monitoring methods are difficult to detect early failure in a timely manner. Therefore, an intelligent monitoring method is urgently needed to capture dynamic sealing performance changes in real time and accurately warn of leakage risks.

[0003] At present, a monitoring scheme based on vibration signal analysis is used for sealing performance evaluation of liquid-driven compressors. This scheme arranges vibration sensors outside the sealing cavity, collects vibration signals caused by the movement of the piston rod, identifies abnormal vibration patterns by combining frequency domain feature extraction methods, and then determines the sealing state by threshold comparison. Some improved schemes also introduce machine learning algorithms to train classification models using historical data, thereby improving the accuracy of anomaly detection.

[0004] Vibration signals are easily disturbed by mechanical noise, especially under complex working conditions, and micro-leakage characteristics are easily covered by background vibration, leading to false positives or false negatives. At the same time, vibration analysis is difficult to distinguish between sealing failure and other mechanical failures (such as bearing wear), reducing the relevance of the monitoring results. In addition, relying on a single vibration parameter and lacking multi-physical quantity collaborative analysis limits the detection ability of early weak leakage. SUMMARY

[0005] The present application provides an intelligent monitoring method and system for sealing performance of a liquid-driven compressor to solve the problems of poor timeliness and low reliability of sealing performance monitoring in the prior art.

[0006] In a first aspect, the present application provides an intelligent monitoring method for sealing performance of a liquid-driven compressor, comprising:

[0007] Collecting high-frequency pressure data, circulating liquid flow data, and friction sound signals of the sealing interface of the liquid-driven compressor sealing cavity;

[0008] Based on the high-frequency pressure data and the circulating liquid flow data, a pressure-flow covariant relationship is constructed;

[0009] The friction sound signals are subjected to directional enhancement processing to form sound energy intensity distribution information;

[0010] The DIANA algorithm is used to perform adaptive clustering processing on the sound energy intensity distribution information to generate a sealing abnormality index;

[0011] Jointly analyze the pressure flow covariation relationship and the sealing anomaly index to generate a monitoring result representing the sealing performance of the liquid driving compressor, and when the monitoring result exceeds a preset warning value, a sealing failure monitoring warning is triggered.

[0012] Optionally, the sealing anomaly index is generated by adaptively clustering the sound energy intensity distribution information using the DIANA algorithm, including:

[0013] Taking the spatial grid of the sealing interface as a sample point, an initial clustering cluster is constructed;

[0014] The dispersion degree of the sound energy intensity value in the initial clustering cluster is calculated, and when the dispersion degree of the sound energy intensity value in the initial clustering cluster is greater than a preset splitting threshold, the initial clustering cluster is divided into two sub-clusters;

[0015] Each of the sub-clusters is taken as a corresponding intermediate clustering cluster, and the calculation operation and the splitting operation are iteratively performed until the dispersion degree of the sound energy intensity value in all intermediate clustering clusters is less than or equal to the preset splitting threshold, and a target clustering cluster is obtained;

[0016] The number of spatial grids with a sound energy intensity value exceeding a preset abnormal boundary in the target clustering cluster is counted to generate a sealing anomaly index.

[0017] Optionally, the dispersion degree of the sound energy intensity value in the initial clustering cluster is calculated, including:

[0018] Based on the sound energy intensity values corresponding to all spatial grids in the initial clustering cluster, a sound energy intensity value set is formed;

[0019] An outlier correction process is performed on the sound energy intensity value set;

[0020] Based on the corrected sound energy intensity value set, the dispersion degree of the sound energy intensity value in the initial clustering cluster is determined using a standard deviation calculation formula.

[0021] Optionally, the initial clustering cluster is divided into two sub-clusters, including:

[0022] A sound energy intensity value spatial gradient field of the initial clustering cluster is constructed;

[0023] In the sound energy intensity value spatial gradient field, a gradient vector with the maximum modulus is determined;

[0024] A segmentation plane is established in the direction of the gradient vector;

[0025] The initial clustering cluster is divided into two sub-clusters with the segmentation plane as a boundary.

[0026] Optionally, the pressure flow covariation relationship is constructed based on the high-frequency pressure data and the circulating liquid flow data, including:

[0027] dividing the high-frequency pressure data into a plurality of discrete pressure segments according to a single reciprocating motion cycle of a piston rod of the liquid-driven compressor;

[0028] extracting a pressure peak time sequence from each of the pressure segments;

[0029] synchronously dividing the cyclic liquid flow rate data into a plurality of flow fluctuation segments;

[0030] calculating an instantaneous flow rate change rate of each flow fluctuation segment;

[0031] establishing a dynamic correlation between the pressure peak time sequence and the instantaneous flow rate change rate within the same motion cycle;

[0032] based on the dynamic correlation of the continuous N motion cycles, constructing a pressure-flow covariation relationship, wherein N is greater than or equal to 3.

[0033] Optionally, the directional enhancement processing of the friction sound signal to form sound energy intensity distribution information comprises:

[0034] spatially selectively strengthening the sound energy components of the friction sound signal through sound wave guidance;

[0035] performing frequency domain analysis on the strengthened sound energy components to identify a specific frequency band range related to high-frequency leakage;

[0036] obtaining sound energy intensity data in the specific frequency band range;

[0037] mapping the sound energy intensity data to a sealing interface spatial coordinate system to form a two-dimensional sound energy intensity distribution map;

[0038] dividing the two-dimensional sound energy intensity distribution map into a plurality of equally spaced spatial grids, calculating the cumulative value of sound energy intensity in each spatial grid, and generating sound energy intensity distribution information based on the cumulative value of sound energy intensity in each spatial grid.

[0039] Optionally, the joint analysis of the pressure-flow covariation relationship and the sealing abnormality indicator to generate a monitoring result representing the sealing performance of the liquid-driven compressor comprises:

[0040] calculating a pressure-flow state scalar based on the pressure-flow covariation relationship;

[0041] converting the sealing abnormality indicator into an acoustic abnormality scalar;

[0042] performing weighted fusion calculation on the pressure-flow state scalar and the acoustic abnormality scalar, and taking the weighted fusion calculation result as the monitoring result representing the sealing performance of the liquid-driven compressor.

[0043] In a second aspect, the application provides an intelligent monitoring system for sealing performance of a liquid-driven compressor, comprising:

[0044] a collection module configured to collect high-frequency pressure data, circulating liquid flow data, and friction sound signals of a sealing interface of a sealing cavity of the liquid-driven compressor;

[0045] a construction module configured to construct a pressure-flow covariant relationship based on the high-frequency pressure data and the circulating liquid flow data;

[0046] an enhancement module configured to perform directional enhancement processing on the friction sound signals to form sound energy intensity distribution information;

[0047] a generation module configured to perform adaptive clustering processing on the sound energy intensity distribution information using a DIANA algorithm to generate a sealing abnormality index;

[0048] a triggering module configured to jointly analyze the pressure-flow covariant relationship and the sealing abnormality index to generate a monitoring result representing the sealing performance of the liquid-driven compressor, and trigger a sealing failure monitoring alarm when the monitoring result exceeds a preset warning value.

[0049] In a third aspect, the application provides a computing device comprising a processor and a memory, wherein the memory stores a computer program, and the processor is configured to execute the computer program to perform the intelligent monitoring method for sealing performance of a liquid-driven compressor according to any one of the first aspect.

[0050] In a fourth aspect, the application provides a computer storage medium storing computer program instructions, wherein the computer program instructions are executed by a processor to implement the intelligent monitoring method for sealing performance of a liquid-driven compressor according to any one of the first aspect.

[0051] In the application, an intelligent monitoring method for sealing performance of a liquid-driven compressor is provided, which comprises the following steps: collecting high-frequency pressure data, circulating liquid flow data, and friction sound signals of a sealing interface of a sealing cavity of the liquid-driven compressor; constructing a pressure-flow covariant relationship based on the high-frequency pressure data and the circulating liquid flow data; performing directional enhancement processing on the friction sound signals to form sound energy intensity distribution information; performing adaptive clustering processing on the sound energy intensity distribution information using a DIANA algorithm to generate a sealing abnormality index; jointly analyzing the pressure-flow covariant relationship and the sealing abnormality index to generate a monitoring result representing the sealing performance of the liquid-driven compressor; and triggering a sealing failure monitoring alarm when the monitoring result exceeds a preset warning value.

[0052] The technical solution provided by the application has the following beneficial effects:

[0053] The application realizes multi-source data synchronous collection, provides basic data support for comprehensive evaluation of sealing performance, establishes a dynamic correlation model of pressure and flow, accurately reflects the fluid state change characteristics in the sealing cavity, effectively extracts the micro leakage characteristic signal of the sealing interface, improves the detection sensitivity of weak acoustic signals, realizes adaptive partitioning of acoustic intensity distribution, accurately identifies the abnormal energy aggregation area, and realizes multi-dimensional collaborative evaluation of the sealing state by comprehensively considering the pressure flow parameters and acoustic characteristics.

[0054] Further, the application further divides the sealing interface into spatial grid elements as sample points, first constructs an initial cluster containing all grids, calculates the dispersion degree of the acoustic intensity value in the cluster, splits the initial cluster into two sub-clusters along the maximum change direction when the dispersion degree exceeds the set threshold, iteratively executes the splitting process until the dispersion degree of all sub-clusters meets the convergence condition, and finally counts the number of grids exceeding the abnormal boundary in each cluster to generate a quantitative sealing abnormality index.

[0055] Moreover, the scheme realizes intelligent partition detection of acoustic energy distribution of the sealing interface, accurately identifies the abnormal area through an adaptive hierarchical splitting algorithm, overcomes the limitations of the traditional fixed threshold method, and improves the accuracy and early warning capability of micro leakage detection.

[0056] These aspects or other aspects of the application will be more apparent in the following description of the embodiments. BRIEF DESCRIPTION OF DRAWINGS

[0057] In order to more clearly illustrate the technical solutions in the embodiments of the application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiment or prior art description. Obviously, the drawings in the following description are some embodiments of the application, and those skilled in the art can also obtain other drawings according to these drawings without creative labor.

[0058] Figure 1 A flow chart of an intelligent monitoring method for sealing performance of a liquid-driven compressor provided by an embodiment of the application;

[0059] Figure 2 A structural schematic diagram of an intelligent monitoring system for sealing performance of a liquid-driven compressor provided by an embodiment of the application;

[0060] Figure 3 A structural schematic diagram of a computing device provided by an embodiment of the application. DETAILED DESCRIPTION

[0061] In order to enable personnel in the art to better understand the application scheme, the technical solutions in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application.

[0062] In some of the flowcharts described in the description and claims of the present application and in the above description of the drawings, a plurality of operations are included which occur in a particular order, but it should be clearly understood that the operations can be performed in an order other than that in which they appear or in parallel, and the serial numbers of the operations, such as 101, 102, etc., are only used to distinguish different operations, and the serial numbers themselves do not represent any execution order. In addition, the flowcharts can include more or fewer operations, and the operations can be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this paper are used to distinguish different messages, devices, modules, etc., and do not represent the order and do not limit that "first" and "second" are different types.

[0063] In the field of liquid-driven compressor sealing performance monitoring, the existing scheme based on vibration signal analysis has limitations: vibration signals are easily disturbed by mechanical noise, causing the micro-leakage characteristics to be covered by background vibration, resulting in insufficient reliability of the detection results; at the same time, this method is difficult to distinguish between sealing failure and other mechanical failures, and the specificity of fault identification is poor; in addition, single vibration parameter analysis lacks multi-physical quantity cooperation, and the detection sensitivity of early weak leakage is insufficient. These problems are essentially due to the singleness of the monitoring dimension and the extensive nature of the signal processing method, making it difficult to meet the high-precision sealing state monitoring demand.

[0064] In view of the above defects, the present application proposes an intelligent monitoring method for sealing performance of a liquid-driven compressor, which synchronously collects high-frequency pressure data, circulating liquid flow data and friction sound signals, constructs a pressure-flow dynamic covariant model, and realizes comprehensive evaluation of the sealing state by combining adaptive clustering analysis of sound energy intensity distribution. Specifically, the sound wave guiding technology is used to enhance the micro-leakage characteristics in the friction sound signal, the DIANA algorithm is used to perform hierarchical split clustering on the sound energy distribution, and the abnormal energy aggregation area is accurately identified; at the same time, the pressure-flow covariant parameters and acoustic anomaly indicators are fused to establish a multi-dimensional sealing performance evaluation system. This method breaks through the limitations of single vibration monitoring, effectively suppresses noise interference through multi-source signal collaborative analysis and feature fusion, improves the sensitivity of micro-leakage detection and the accuracy of fault identification, and provides reliable early warning for the degradation of compressor sealing performance.

[0065] The technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application. Obviously, the described embodiments are only a part of the embodiments of the present application, not all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor are within the scope of protection of the present application.

[0066] Figure 1 A flowchart of an intelligent monitoring method for sealing performance of a liquid-driven compressor provided by the embodiments of the present application is shown in Figure 1As shown, the method comprises:

[0067] Step 101: Collecting high-frequency pressure data of the sealing cavity of the liquid drive compressor, circulating liquid flow data, and friction sound signals of the sealing interface.

[0068] In step 101, the high-frequency pressure data represents the rapid change data of the fluid pressure in the sealing cavity, reflecting the dynamic pressure fluctuation characteristics in the sealing cavity when the piston rod moves. The circulating liquid flow data represents the flow rate data of the working medium in the compressor circulating system, characterizing the liquid exchange state of the inlet and outlet of the sealing cavity. The friction sound signal represents the sound signal generated by the relative movement of the sealing interface, containing the microscopic feature information of the sealing contact state.

[0069] In the embodiments of the present application, the pressure fluctuation signal is collected in real time by the pressure sensor installed in the sealing cavity, the liquid flow rate in the circulating pipeline is recorded synchronously using the flowmeter, and the friction sound wave is captured by arranging the acoustic sensor array outside the sealing ring. The pressure sensor continuously records the sealing cavity pressure value at a very short interval, the flowmeter measures the flow rate of the liquid through a specific cross section, and the acoustic sensor array uses a specific structure to enhance the sound wave receiving ability in the normal direction of the sealing interface. The three data collection devices realize data synchronization through a unified clock signal.

[0070] For example, in the test of a certain type of liquid drive compressor, an A brand pressure sensor is used to collect the sealing cavity pressure at a frequency of 10,000 times per second, a B type flowmeter is used to record the circulating pipeline flow change, and a C series acoustic sensor array is arranged around the sealing ring. The pressure sensor is installed in the middle position of the sealing cavity, the flowmeter is installed in the straight pipe section of the circulating pipeline, and the acoustic sensor is distributed at a uniform interval. The three devices ensure that the data collection time is aligned through a synchronous controller, the pressure data reflects the periodic pressure fluctuation caused by the reciprocating motion of the piston rod, the flow data shows the corresponding relationship with the pressure change, and the sound signal presents the friction characteristics of the sealing interface.

[0071] Step 102: Based on the high-frequency pressure data and the circulating liquid flow data, a pressure-flow covariation relationship is constructed.

[0072] In step 102, the pressure-flow covariation relationship represents a dynamic correlation model between pressure and flow parameters, characterizing the cooperative characteristics of the fluid state change in the sealing system.

[0073] In the embodiments of the present application, the collected high-frequency pressure data is divided into several time periods according to the piston rod movement period, and the pressure peak sequence in each time period is extracted. The flow data of the corresponding time period is processed synchronously, and the flow change rate is calculated. The corresponding relationship between the pressure peak value and the flow change is established, and the cooperative change law of the two in multiple movement periods is analyzed by statistical analysis, and a covariation relationship model reflecting the dynamic characteristics of the sealing system is constructed.

[0074] For example, the test obtained pressure data is segmented according to the 120 millisecond motion cycle of the piston rod 120, and 12 pressure extreme points in each segment are extracted to form a sequence. The instantaneous change amount of the flow data corresponding to the time period is calculated every millisecond interval. Analysis of three sets of complete motion cycle data shows that the flow change begins to increase at a certain time after the pressure peak value appears, the correlation coefficient reaches 0.82, and a dynamic correlation model between the two is established. The correlation coefficient calculation formula is wherein is the pressure peak value, is the corresponding flow rate change rate, and are the mean values, respectively.

[0075] Step 103: Perform directional enhancement processing on the friction sound signal to form sound energy intensity distribution information.

[0076] In step 103, the sound energy intensity distribution information represents the spatial distribution characteristic data of the sound wave energy of each position of the sealing interface.

[0077] In the embodiments of the present application, the collected friction sound signal is subjected to spatial filtering processing to enhance the sound wave component in the normal direction of the sealing interface. The enhanced signal is subjected to frequency spectrum analysis to identify the effective component in a specific frequency band. The intensity characteristic of the signal in the frequency band is extracted, and is mapped to the two-dimensional coordinate system of the sealing ring according to the position of the sound sensor to form a distribution map reflecting the sound wave energy of each position.

[0078] For example, in the test, the sound guide structure is used to enhance the sound signal in the normal direction of the sealing ring, and frequency spectrum analysis is performed to determine that 8500 to 11500 Hz is the characteristic frequency band. The root mean square value of the signal in the frequency band is extracted as the intensity index, and is mapped to the coordinate system of the sealing ring surface according to the position of the sensor to form a sound energy distribution map of 40 mm square. The intensity calculation formula is wherein is the discrete sampling value, is the number of sampling points.

[0079] Step 104: Perform adaptive clustering processing on the sound energy intensity distribution information using the DIANA algorithm to generate a sealing abnormality index.

[0080] In step 104, the sealing abnormality index represents a comprehensive parameter quantitatively reflecting the abnormality degree of the sealing performance.

[0081] In the embodiments of the present application, the sound energy distribution map is divided into a plurality of grid elements as sample points, and all the grids are initially classified into one cluster. The dispersion degree of the sound energy intensity of each point in the cluster is calculated, and when it exceeds a set threshold, the cluster is divided into two sub-clusters along the direction of the greatest change in sound energy. The splitting process is iteratively performed until the dispersion degree of all sub-clusters meets the requirements, and finally the proportion of grids exceeding the abnormal threshold in each cluster is counted to generate the sealing abnormality index.

[0082] For example, the sound energy distribution is divided into 400 grids, and the standard deviation of the initial clustering cluster is calculated to be 0.85. When the threshold value 0.8 is exceeded, the direction is deflected by 15 degrees along the axial direction, and the two sub-clusters are split. After iterative splitting, 7 stable clusters are obtained, and the grid ratio of the intensity exceeding 5 is 23% as an abnormal index. The threshold value 0.8 is determined by historical data analysis.

[0083] Step 105: jointly analyzing the pressure flow covariant relationship and the sealing abnormal index to generate a monitoring result representing the sealing performance of the liquid-driven compressor, and triggering a sealing failure monitoring alarm when the monitoring result exceeds a preset warning value.

[0084] In step 105, the monitoring result represents a quantitative parameter for comprehensively evaluating the sealing performance state.

[0085] In the embodiments of the present application, the pressure flow covariant parameter is converted into a state scalar, and the acoustic abnormal index is converted into an abnormal scalar. The two scalars are weighted and fused to obtain a monitoring value that comprehensively reflects the sealing performance. When the monitoring value continuously exceeds the warning line, a warning signal is triggered.

[0086] For example, the pressure flow correlation coefficient 0.82 is taken as the state scalar, and the abnormal index 23% is converted into 76.7 points. The monitoring value 0.79 is obtained by weighting and fusing according to the weights 0.6 and 0.4. When it exceeds 0.75 for three consecutive times, an alarm is triggered. The warning value 0.75 is determined based on historical normal data statistics.

[0087] The method realizes comprehensive monitoring of the sealing performance of the liquid-driven compressor through synchronous acquisition and fusion analysis of multiple source data. The pressure flow covariant relationship accurately reflects the fluid state of the sealing system, the sound energy distribution analysis effectively identifies the micro-leakage characteristics, the adaptive clustering algorithm improves the abnormal detection accuracy, and the multi-parameter fusion evaluation ensures the warning reliability, thereby providing an effective technical means for compressor sealing maintenance.

[0088] In order to solve the problem of insufficient micro-leakage feature recognition accuracy in the sealing performance monitoring of the liquid-driven compressor, in some embodiments, step 104: the DIANA algorithm is used to perform adaptive clustering processing on the sound energy intensity distribution information to generate a sealing abnormal index, including:

[0089] Step 201: taking the spatial grid of the sealing interface as a sample point, an initial clustering cluster is constructed.

[0090] In step 201, the initial clustering cluster refers to a collection of all grid elements as initial analysis objects.

[0091] In the embodiments of the present application, first, the surface of the sealing ring is divided into a regular array of square grids according to the geometric size of the sealing ring, ensuring that each grid has the same area. The sound energy intensity value collected by each grid unit is taken as a sample point attribute, and all the grids collectively constitute an initial clustering cluster, providing a basic data set for subsequent analysis.

[0092] Step 202: Calculate the dispersion degree of the sound energy intensity values in the initial clustering cluster. When the dispersion degree of the sound energy intensity values in the initial clustering cluster is greater than a preset splitting threshold, the initial clustering cluster is divided into two sub-clusters.

[0093] In step 202, the dispersion degree refers to the distribution dispersion degree of the sound energy intensity values in the clustering cluster, reflecting the uniformity of sound energy distribution. The splitting threshold is a critical value for determining whether the clustering cluster needs to be divided, which is determined according to historical normal data statistical analysis. The sub-cluster is a new clustering unit obtained by the division operation.

[0094] In the embodiments of the present application, the standard deviation of all grid sound energy intensity values in the initial clustering cluster is calculated as the dispersion degree index. When this value exceeds the preset splitting threshold, the spatial variation trend of the sound energy intensity of each grid point is analyzed to determine the direction of the greatest sound energy variation. The initial clustering cluster is divided into two new sub-clusters along this direction, ensuring that the sound energy distribution inside each sub-cluster is relatively uniform.

[0095] Step 203: Take each of the sub-clusters as a corresponding intermediate clustering cluster, and iteratively perform the calculation operation and the splitting operation until the dispersion degree of the sound energy intensity values in all intermediate clustering clusters is less than or equal to the preset splitting threshold, obtaining a target clustering cluster.

[0096] In step 203, the calculation operation is followed by the calculation of the dispersion degree of the sound energy spatial distribution in the intermediate clustering cluster. The intermediate clustering cluster refers to the sub-cluster set to be analyzed in the iteration process. The target clustering cluster refers to the final stable clustering unit that meets the dispersion degree requirement.

[0097] In the embodiments of the present application, the sub-clusters obtained in step 202 are taken as new analysis objects, and the process of calculating the dispersion degree and judging the splitting condition is repeated. The sub-clusters that do not meet the dispersion requirement are continuously split along the direction of the greatest variation until the dispersion degree of all sub-clusters is lower than the splitting threshold, obtaining the final target clustering cluster set.

[0098] Step 204: Count the number of spatial grids in the target clustering cluster whose sound energy intensity value exceeds a preset abnormal boundary to generate a sealing abnormality index.

[0099] In step 204, the abnormal boundary refers to the critical value for determining whether the sound energy intensity of a grid unit is abnormal, which is determined according to the statistical distribution of sound energy intensity under normal working conditions.

[0100] In the embodiments of the present application, in all target clustering clusters, the number of grid cells with sound energy intensity exceeding the abnormal boundary is counted, and the proportion of the number in the total number of grids is calculated as the sealing abnormality index. The index quantitatively reflects the abnormality degree of the sealing interface.

[0101] The following is a specific example:

[0102] In the sealing performance test of the liquid-driven compressor, based on the collected pressure, flow and sound signal data, when the sound energy intensity distribution information is adaptively clustered, first, the sound energy distribution map of the sealing ring surface of 40 mm square is divided into 400 square grid cells with a side length of 0.2 mm. The sound energy intensity value E of each grid is calculated by the formula , wherein M=2000 is the number of sampling points in a 20 ms period; all grids are taken as initial clustering clusters, the standard deviation σ=0.85 of the 400 intensity values is calculated, and the value is obtained by the formula , wherein is the intensity value of the i-th grid, μ is the average value, and N=400; when σ exceeds the preset splitting threshold value 0.8, the spatial variation gradient of the sound energy intensity of each grid point is calculated, it is determined that the maximum variation direction is along the axial deflection of 15 degrees of the piston rod, and the initial cluster is divided into two sub-clusters along this direction, which respectively contain 158 and 242 grid cells; the above calculation process is repeated for the two sub-clusters, and the splitting is stopped when the standard deviations of all sub-clusters are all ≤0.8. Finally, 5 stable clustering clusters are obtained; the number of grids with intensity exceeding 5 is counted, which is 89, accounting for 22.25% of the total number of grids, as the sealing abnormality index. The splitting threshold value 0.8 and the abnormal boundary 5 are both determined by analyzing 100 groups of normal working condition data of the compressor, and the 95th and 99th percentile values are taken respectively to ensure the reliability of the detection result.

[0103] In the embodiments of the present application, the method realizes intelligent partition detection of the sound energy distribution of the sealing interface through adaptive clustering analysis, accurately identifies the abnormal energy aggregation area, overcomes the limitations of the fixed threshold method, improves the accuracy and early warning capability of micro-leakage detection, and provides a reliable basis for the sealing state evaluation of the compressor.

[0104] In order to further improve the accuracy of the calculation of the dispersion degree of the sound energy intensity, in some embodiments, step 202: the calculation of the dispersion degree of the sound energy intensity values in the initial clustering cluster comprises:

[0105] Step 301: forming a sound energy intensity value set based on the sound energy intensity values corresponding to all spatial grids in the initial clustering cluster.

[0106] In step 301, the sound energy intensity value set refers to a set composed of sound energy intensity data corresponding to all spatial grid cells in the initial clustering cluster. Each data point represents a sound energy intensity measurement value of a grid cell.

[0107] In the embodiment of the present application, first, all the spatial grid cells contained in the initial clustering cluster are traversed, the sound energy intensity values recorded by each cell are extracted, these values are arranged in order of grid number to form a complete sound energy intensity value dataset, which provides basic data for subsequent analysis.

[0108] Step 302: performing outlier correction processing on the sound energy intensity value set.

[0109] In step 302, the outlier correction processing refers to the process of identifying and adjusting extreme values that deviate significantly from the normal range in the sound energy intensity value set, aiming to eliminate the influence of measurement errors or local interference.

[0110] In the embodiment of the present application, the box plot method is used to identify outliers. First, the quartiles and interquartile range of the sound energy intensity value set are calculated. Values that exceed the upper quartile plus 1.5 times the interquartile range or are lower than the lower quartile minus 1.5 times the interquartile range are determined as outliers, and these outliers are replaced with the median value of adjacent grid cells to ensure data quality.

[0111] Step 303: based on the corrected sound energy intensity value set, using the standard deviation calculation formula to determine the dispersion degree of the sound energy intensity values in the initial clustering cluster.

[0112] In step 303, the standard deviation calculation formula refers to the mathematical expression used to quantify the dispersion degree of sound energy intensity values, reflecting the distribution of data points around the mean value.

[0113] In the embodiment of the present application, for the corrected sound energy intensity value set, first, the mean value of all values is calculated, then the sum of squares of the difference between each value and the mean value is calculated, and then the result is divided by the total number of data points and square rooted to finally obtain the standard deviation result representing the dispersion degree. This standard deviation will be used to determine whether the clustering cluster needs to be further split.

[0114] The following is a specific example:

[0115] In the liquid-driven compressor sealing performance test, based on the 400 grid unit sound energy intensity value set obtained by the right 2 embodiment, first, the box plot method is used for outlier correction processing, the specific process is to calculate the upper quartile Q3 of the data set as 3.8, and the lower quartile Q1 as 2.4, and the interquartile range IQR=Q3-Q1=1.4 is obtained. The values greater than Q3+1.5IQR, that is, 5.9 or less than Q1-1.5IQR, that is, 0.3, are determined as outliers, and a total of 8 abnormal grid points are identified. These outliers are replaced by the median value of the adjacent 8 grids, and the replacement value is determined by sorting the intermediate value; after the correction, the dispersion degree is calculated based on the corrected sound energy intensity value set. First, the average value μ=3.2 of all grid intensity values is calculated, and then the standard deviation formula The calculation obtains σ=0.83, wherein N=400 is the total number of grids.

[0116] In the embodiments of the present application, the method combines outlier correction and standard deviation calculation, effectively improves the accuracy of sound energy intensity dispersion degree evaluation, eliminates the influence of local interference on clustering analysis, lays a data foundation for reliable generation of subsequent sealing abnormality indicators, and enhances the stability and anti-interference ability of the entire monitoring system.

[0117] In order to further improve the accuracy of clustering cluster division, in some embodiments, step 202: the initial clustering cluster is divided into two sub-clusters, including:

[0118] Step 401: constructing a sound energy intensity value space gradient field of the initial clustering cluster.

[0119] In step 401, the sound energy intensity value space gradient field refers to a vector field reflecting the spatial variation rate of the sound energy intensity of the sealing ring surface. Each grid point corresponds to a gradient vector, which represents the change direction and amplitude of the sound energy intensity in space.

[0120] In the embodiments of the present application, first, the sound energy intensity variation rate between each unit and the adjacent eight units contained in the initial clustering cluster is calculated. The central difference method is used to determine the intensity variation components in the axial and radial directions of the grid point, and the two components are combined to form the gradient vector of the point. The gradient vectors of all grid points together form the space gradient field.

[0121] Step 402: determining the gradient vector with the maximum modulus in the sound energy intensity value space gradient field.

[0122] In step 402, the maximum modulus refers to the maximum value obtained after comparing the lengths (modulus) of all gradient vectors in the sound energy intensity density gradient field, and the direction of the gradient vector is the spatial orientation with the most drastic change in sound energy intensity density, and the modulus value is calculated by the square root of the sum of the partial derivatives of the sound energy intensity density with respect to the spatial coordinates. The gradient vector refers to the longest vector in the gradient field, and its direction represents the spatial orientation with the most drastic change in sound energy intensity, and the modulus value reflects the change amplitude.

[0123] In the embodiments of the present application, the gradient vectors of all grid points in the gradient field are traversed, the lengths of each vector are calculated, the vector with the maximum length is found by comparison, and the direction angle and modulus value of the vector are recorded as the basis for subsequent segmentation.

[0124] Step 403: Establish a segmentation plane in the direction of the gradient vector.

[0125] In step 403, the segmentation plane refers to a virtual plane perpendicular to the direction of the maximum gradient vector, which is used to divide the clustering cluster into two parts, and the plane position is determined by an optimization algorithm.

[0126] In the embodiments of the present application, an infinitely extended virtual plane is established in the direction of the maximum gradient vector, and the plane position is adjusted to minimize the internal sound energy intensity difference of the sub-clusters on both sides of the plane, ensuring that the two sub-clusters after segmentation are as uniform as possible.

[0127] Step 404: Divide the initial clustering cluster into two sub-clusters with the segmentation plane as the boundary.

[0128] In the embodiments of the present application, according to the determined segmentation plane position, all grid elements are divided into two sets on both sides of the plane, the statistical characteristics of the sound energy intensity of each set are calculated, and whether the dispersion degree meets the requirements is verified.

[0129] The following is a specific example:

[0130] In the liquid-driven compressor sealing performance test, based on the obtained 400 grid element sound energy intensity data, when the initial clustering cluster standard deviation σ = 0.85 exceeds the threshold value 0.8, the following division operation is performed: first, construct the sound energy intensity gradient field, and for each grid point, use the central difference formula and The gradient component is calculated, where Δx=Δy=0.2 mm is the grid spacing, and E(x,y) represents the grid sound intensity value at coordinates (x,y); all grid point gradient vectors are traversed, and it is found that the grid point at coordinates (12,18) has the maximum gradient modulus value of 14.3, and the direction is an axial deflection of 16 degrees; a segmentation plane is established along the direction, and the plane position should be 6.8 mm away from the center line of the sealing ring through the optimization algorithm, at which time the sound intensity standard deviations of the two sub-clusters on both sides of the plane are 0.61 and 0.59, respectively; finally, the initial clustering cluster is divided into two sub-clusters along the plane, which respectively contain 165 and 235 grid cells, and the segmentation position optimization adopts the criterion of minimizing the sum of the variances of the sub-clusters, and the specific expression is wherein and E(x,y) and E(x,y) respectively represent the sound intensity values of the two sub-clusters, and E(x,y) are the average values of the corresponding sub-clusters.

[0131] In the embodiments of the present application, the method realizes accurate division of the clustering cluster by constructing a sound intensity gradient field and optimizing a segmentation plane, ensures the uniformity of the sound energy distribution in the sub-clusters, provides a reliable basis for subsequent abnormal area identification, and effectively improves the accuracy of the sealing performance detection.

[0132] In order to further improve the accuracy of the construction of the pressure-flow covariant relationship, in some embodiments, step 102: the pressure-flow covariant relationship is constructed based on the high-frequency pressure data and the circulating liquid flow data, comprising:

[0133] Step 501: the high-frequency pressure data is divided into a plurality of discrete pressure segments according to the single reciprocating motion period of the piston rod of the liquid-driven compressor.

[0134] In step 501, the single reciprocating motion period of the piston rod refers to the duration of the entire process of the extension and retraction of the piston rod of the liquid-driven compressor, which is determined by the mechanical motion signal measured by the piston rod displacement sensor or the crank angle encoder, and the time length is directly determined by the compressor speed. The relationship between the piston rod and the liquid-driven compressor: the piston rod is the core moving part of the liquid-driven compressor, one end of which is connected to the hydraulic drive system, and the other end drives the compressor piston to reciprocate in the sealed cavity, and the movement state directly determines the volume change and pressure fluctuation of the sealed cavity. The discrete pressure segment refers to the time period data obtained by intercepting the continuously collected high-frequency pressure data according to the complete reciprocating motion period of the piston rod, and each segment contains the pressure change characteristics of a complete motion period.

[0135] In the embodiment of the present application, first, the start and end time points of a single reciprocating motion are determined by the piston rod displacement sensor, and the continuously collected high-frequency pressure data is divided into multiple equal-length data segments according to the time points, and each data segment corresponds to a complete piston motion cycle, thereby providing basic time period data for subsequent feature extraction.

[0136] Step 502: Extracting a pressure peak time sequence from each pressure segment.

[0137] In step 502, the pressure peak time sequence refers to a set of extreme value points arranged in time sequence in each pressure segment, reflecting the key features of pressure fluctuations in the cycle.

[0138] In the embodiment of the present application, for each segmented pressure segment data, all peak points are identified by an extreme value detection algorithm, arranged in time sequence to form a peak sequence, and the time sequence characteristics of pressure changes are retained.

[0139] Step 503: Synchronously dividing the circulating liquid flow data into multiple flow fluctuation segments.

[0140] In step 503, the flow fluctuation segment refers to a flow data segment divided synchronously with the pressure segment, representing the dynamic flow state of the fluid in the same motion cycle.

[0141] In the embodiment of the present application, according to the time nodes of the pressure data division, the flow data collected by the flow sensor is synchronously intercepted, ensuring that each flow segment is completely consistent with the time range of the corresponding pressure segment.

[0142] Step 504: Calculating the instantaneous flow rate of change of each flow fluctuation segment.

[0143] In step 504, the instantaneous flow rate of change refers to the flow rate of change between adjacent sampling points in the flow fluctuation segment, reflecting the dynamic response characteristics of the flow.

[0144] In the embodiment of the present application, for each continuous sampling point of the flow segment, the difference between the flow values of the adjacent two points is divided by the sampling time interval to obtain the instantaneous change rate at that time point, forming a flow rate of change curve.

[0145] Step 505: Establishing a dynamic correlation between the pressure peak time sequence and the instantaneous flow rate of change in the same motion cycle.

[0146] In step 505, the same motion cycle refers to the fact that the pressure data segment and the flow data segment strictly correspond to the same mechanical motion stage of the piston rod (such as both being the Nth extension process), and the data segmentation matching is realized by time domain alignment of the crank angle signal or the piston rod displacement signal. The dynamic correlation refers to the coordinated change rule of the pressure peak value and the flow rate of change in time sequence, representing the coupling characteristics of the fluid state in the sealing cavity.

[0147] In the embodiment of the application, the time points of the pressure peak value sequence in the same motion cycle are aligned with the flow rate change rate curve, the response mode and delay time of the flow rate change rate after the peak value appears are analyzed, and a timing correlation model between them is established.

[0148] In step 506, based on the dynamic correlation relationship of the consecutive N motion cycles, a pressure-flow covariant relationship is constructed, wherein N is greater than or equal to 3.

[0149] In step 506, the consecutive judgment needs to meet two conditions (1) the cycle interval time is equal to an integer multiple of the piston rod motion cycle; and (2) there is no missing data in adjacent cycles. The minimum number is determined by the characteristics of the compressor, and usually more than or equal to 3 cycles can establish an effective statistical rule.

[0150] In the embodiment of the application, the pressure-flow dynamic correlation data of multiple motion cycles are continuously analyzed, the consistency index of the correlation characteristics of each cycle is calculated, and when the fluctuation of the index of the consecutive three cycles is less than a set threshold, it is confirmed that the covariant relationship is stably established.

[0151] The following is a specific example:

[0152] In the sealing performance test of a certain type of liquid-driven compressor, based on the collected synchronous data, the specific implementation process of constructing the pressure-flow covariant relationship is as follows: first, the high-frequency pressure data collected at 10,000 times per second are divided according to the motion cycle of 120 milliseconds of the piston rod, and 12 pressure peak points are extracted in each cycle segment to form a timing sequence through the extreme value detection algorithm; the flow rate data are processed synchronously, the flow rate value in the same time period is calculated as the instantaneous change rate at a millisecond interval, and the formula is , wherein represents the flow rate value at t time, and Δt=1 millisecond is the sampling interval; the data of the consecutive three motion cycles are selected, the pressure peak value occurrence time in each cycle is matched with the flow rate change rate at the corresponding time, the correlation coefficient formula is used to calculate the correlation between them; the correlation coefficients of the three cycles are 0.81, 0.83 and 0.82 respectively, and the average value is 0.82, indicating that the flow rate change rate starts to rise synchronously about 2 milliseconds after the pressure peak value appears, and a stable pressure-flow covariant relationship model is established.

[0153] In the embodiment of the application, the method accurately captures the cooperative change rule of pressure and flow rate through timing alignment and dynamic correlation analysis, and the covariant relationship established can effectively reflect the fluid state change in the sealing cavity, thereby providing a reliable fluid dynamic characteristic basis for sealing performance evaluation.

[0154] To further improve the accuracy of friction acoustic signal processing, in some embodiments, step 103: the directional enhancement processing of the friction acoustic signal is performed to form acoustic energy intensity distribution information, comprising:

[0155] Step 601: The spatially selective strengthening of the acoustic energy component of the friction acoustic signal is performed by acoustic wave guiding.

[0156] In step 601, acoustic wave guiding refers to an acoustic guiding device with a specific geometric structure, which can selectively enhance the acoustic signal component in the normal direction of the sealing interface and suppress the interference noise in other directions. The acoustic energy component is the specific direction (sealing interface normal direction) acoustic energy component extracted after the spatially selective strengthening of the friction acoustic signal by the acoustic wave guiding structure, which is obtained by directional filtering of the original acoustic signal by the physical guiding structure.

[0157] In the embodiments of the present application, the sound guiding structure arranged in a ring array is installed around the sealing ring, and the internal cavity design makes the acoustic waves in the vertical direction of the sealing interface be focused and enhanced, while the obliquely incident acoustic waves are attenuated, thereby improving the signal-to-noise ratio of the effective signal.

[0158] Step 602: Frequency domain analysis is performed on the strengthened acoustic energy component to identify a specific frequency band range related to high-frequency leakage.

[0159] In step 602, frequency domain analysis refers to converting time domain acoustic signals into frequency distribution representation, and identifying characteristic frequency bands containing leakage information through spectral features. High-frequency leakage refers to the acoustic signal characteristics in the frequency band above 8 kHz generated when the liquid-driven compressor seal fails, and the basis for differentiation is that in the sealing performance monitoring scene, "high frequency" specifically refers to the 8 kHz-12 kHz frequency band (the energy in this frequency band is increased by more than 6 dB when leakage occurs, which is measured through experiments), and "low frequency" refers to the frequency band below 1 kHz (mainly containing mechanical vibration noise); The mechanism of generating high-frequency leakage signal is the high-frequency acoustic wave excited when micron-level sealing gap forms turbulent flow, while low-frequency signal is mainly generated by mechanical vibration of piston rod and other components. After separation by Fourier transform, only the high-frequency component is used for leakage diagnosis. The specific frequency band range refers to the continuous frequency interval highly related to the sealing leakage characteristics determined by frequency domain analysis. The acoustic signal energy in this frequency band is higher than the background noise and has a specific frequency spectrum feature mode, which is used to extract effective leakage signal components.

[0160] In the embodiments of the present application, the enhanced acoustic signal is subjected to fast Fourier transform, and after generating a frequency spectrum, the energy distribution of each frequency band is analyzed, and a continuous frequency band with energy higher than the background noise and matching the leakage characteristics is selected as the effective analysis range.

[0161] Step 603: Acoustic energy intensity data is obtained in the specific frequency band range.

[0162] In step 603, the acoustic energy intensity data refers to the quantitative representation of the sound wave energy in the characteristic frequency band, reflecting the intensity characteristics of the signal in a specific frequency range.

[0163] In the embodiments of the present application, the time-domain waveform of the characteristic frequency band signal is extracted, and the root mean square value thereof is calculated as the intensity index, which comprehensively reflects the energy size of the sound wave in the frequency band.

[0164] Step 604: Map the acoustic energy intensity data to the sealed interface space coordinate system to form a two-dimensional acoustic energy intensity distribution map.

[0165] In step 604, the sealed interface space coordinate is a two-dimensional coordinate system established with the piston rod seal ring contact surface of the liquid-driven compressor as the reference, with the x-axis along the circumferential direction of the piston rod and the y-axis along the axial direction, and the coordinate origin fixed at the geometric center of the seal ring. The coordinate system is determined in advance through mechanical design drawings. The two-dimensional acoustic energy intensity distribution map refers to the visualization of the energy distribution formed by mapping the acoustic energy intensity values of each measurement point to the two-dimensional plane of the seal ring according to the measurement position.

[0166] In the embodiments of the present application, the acoustic energy intensity values of the corresponding points measured by the acoustic sensor array are marked on the seal ring plane projection map according to the spatial position coordinates of each probe, and a continuous two-dimensional distribution map is generated through an interpolation algorithm.

[0167] Step 605: Divide the two-dimensional acoustic energy intensity distribution map into multiple equally spaced spatial grids, calculate the cumulative value of the acoustic energy intensity in each spatial grid, and generate acoustic energy intensity distribution information based on the cumulative value of the acoustic energy intensity in each spatial grid.

[0168] In step 605, the spatial grid refers to dividing the two-dimensional plane of the seal ring into several regular small areas (such as squares or hexagons) with equal areas, each grid element corresponding to an independent data acquisition and analysis element for quantifying the spatial distribution characteristics of acoustic energy intensity. The acoustic energy intensity refers to the sound wave energy per unit area per unit time, which is derived from the square quantity processing of the sound pressure signal in a specific frequency band (high-frequency leakage characteristic frequency band) and converted into a standard energy unit (μJ / mm²) through the calibration coefficient of the acoustic sensor. The cumulative value refers to the sum obtained by adding or integrating the acoustic energy intensity of all sampling time points in each spatial grid element, reflecting the total acoustic energy received in the grid area during the analysis period.

[0169] In the embodiments of the present application, the two-dimensional distribution map is divided into regular grids, and the intensity values in each grid area are time-integrated and added to obtain a grid-based acoustic energy intensity data set, which is convenient for computer algorithm processing.

[0170] The following is a specific example:

[0171] In the sealing performance test of a certain type of liquid drive compressor, based on the collected acoustic signal data, the specific implementation process of directional enhancement processing is as follows: first, the spatial filtering of the friction sound signal is performed through the annular arrangement of 8 unit sound guide structures, which makes the sound wave intensity in the normal direction of the sealing interface increase to 3 times of the original signal, and the oblique incident interference noise is attenuated by more than 50%; the enhanced signal is subjected to frequency spectrum analysis, and the fast Fourier transform is used to convert the time domain signal into frequency domain representation, and it is found that the signal energy in the frequency band of 8600 to 11400 Hz reaches 8 times of the background noise, and the frequency band range is determined through analysis of 100 groups of historical leakage data, which can effectively represent the micro-leakage characteristics; the root mean square value of the signal in the frequency band is extracted as the intensity index, and the calculation formula is , wherein M=2000 is the number of sampling points in a 20 millisecond period; the intensity values of each sensor measuring point are mapped to the two-dimensional coordinate system of the sealing ring according to the installation position, and the sound energy intensity distribution map of the 40 millimeter*40 millimeter area is generated through bilinear interpolation; the distribution map is divided into 6400 grid units of 0.5 millimeter*0.5 millimeter, and the sound energy intensity of 60 milliseconds accumulated in each unit in 3 motion cycles is integrated and added, wherein the cumulative value of the kth grid , represents the intensity value of the grid in the tth millisecond; the finally generated sound energy intensity distribution information shows that there is a high energy area in the 4 o'clock direction of the sealing ring, and the average cumulative value of the grid is 12.5 microjoules, which is 3.8 times that of other areas.

[0172] In the embodiments of the present application, the method effectively highlights the acoustic characteristics related to leakage through directional enhancement and feature band extraction, and the distribution information after grid processing not only retains spatial details but also facilitates quantitative analysis, thereby providing a high-quality data basis for accurately identifying abnormal sealing areas.

[0173] In order to further improve the comprehensiveness and accuracy of the sealing performance evaluation, in some embodiments, step 105: the joint analysis of the pressure-flow covariation relationship and the sealing abnormality index generates a monitoring result representing the sealing performance of the liquid drive compressor, including:

[0174] Step 701: based on the pressure-flow covariation relationship, a pressure-flow state scalar is calculated.

[0175] In step 701, the pressure-flow state scalar refers to quantifying the dynamic covariation relationship between pressure and flow into a single numerical index, reflecting the comprehensive characteristics of the fluid state of the sealing system.

[0176] In the embodiments of the present application, the key feature parameters in the pressure-flow covariation relationship are extracted, including correlation coefficient, response delay time, etc., and through normalization processing and multi-parameter fusion algorithm, these feature parameters are converted into scalar values ranging from 0 to 1, and the larger the value, the higher the risk of fluid state abnormality.

[0177] Step 702: converting the sealing abnormality index into an acoustic abnormality scalar.

[0178] In step 702, the acoustic abnormality scalar refers to converting the sealing abnormality index into a numerical value with a unified dimension for collaborative analysis with the fluid state scalar.

[0179] In the embodiments of the present application, according to the statistical distribution of historical data, the sealing abnormality index is mapped to a scale of 0 to 100 points according to the percentile, and the higher the score, the more serious the acoustic abnormality, which is convenient for weighted calculation with fluid parameters.

[0180] Step 703: weighted fusion calculation of the pressure flow state scalar and the acoustic abnormality scalar, and taking the weighted fusion calculation result as the monitoring result representing the sealing performance of the liquid drive compressor.

[0181] In step 703, the weighted fusion calculation refers to giving different weights according to the reliability of different monitoring parameters, and generating a comprehensive evaluation value through linear combination.

[0182] In the embodiments of the present application, the pressure flow state scalar and the acoustic abnormality scalar are linearly combined according to the preset weight coefficient, wherein the fluid state weight is slightly higher than the acoustic parameter, reflecting its higher sensitivity to early leakage, and the finally generated monitoring result value is used to trigger the graded early warning.

[0183] The following is a specific example:

[0184] In the sealing performance test of a certain type of liquid drive compressor, based on the obtained analysis data, the specific implementation process of joint evaluation is as follows: first, the key parameters are extracted from the established pressure flow covariant relationship, including the correlation coefficient 0.82 and the peak response delay 2 milliseconds, and the pressure flow state scalar is calculated by the formula , wherein r is the correlation coefficient, Δt is the delay time, and τ=5 milliseconds is the maximum delay threshold allowed by the system; at the same time, the sealing abnormality index 23% is converted into the acoustic abnormality scalar 73.5 points by the linear mapping formula , wherein A is the current abnormality index value, and are the typical values of historical normal and serious leakage respectively; finally, the two scalars are weighted and fused, and the comprehensive monitoring value is calculated by the formula , wherein the weight coefficient is determined by analyzing 50 sets of experimental data, reflecting the more sensitive characteristics of the pressure flow parameter to early leakage; when the monitoring value of three consecutive motion cycles exceeds the preset warning line 75 points, the early warning signal is triggered, and the warning value is determined according to the 99% confidence upper limit of the normal operation data of the same type of equipment per thousand hours.

[0185] In the embodiment of the present application, the method realizes comprehensive evaluation of fluid dynamic characteristics and acoustic characteristics through multi-parameter weighted fusion, which not only utilizes the sensitivity of pressure and flow parameters to early leakage, but also combines the positioning ability of acoustic detection to small defects, thereby improving the accuracy and reliability of the sealing state monitoring.

[0186] Figure 2 A structural schematic diagram of an intelligent monitoring system for sealing performance of a liquid-driven compressor provided in the embodiment of the present application is shown in Figure 2 The system comprises:

[0187] The acquisition module 21 is configured to acquire high-frequency pressure data, circulating liquid flow data and friction sound signals of a sealing interface of the sealing cavity of the liquid-driven compressor.

[0188] The construction module 22 is configured to construct a pressure-flow covariant relationship based on the high-frequency pressure data and the circulating liquid flow data.

[0189] The enhancement module 23 is configured to perform directional enhancement processing on the friction sound signals to form sound energy intensity distribution information.

[0190] The generation module 24 is configured to perform adaptive clustering processing on the sound energy intensity distribution information by using a DIANA algorithm to generate a sealing abnormality index.

[0191] The triggering module 25 is configured to jointly analyze the pressure-flow covariant relationship and the sealing abnormality index to generate a monitoring result representing the sealing performance of the liquid-driven compressor, and when the monitoring result exceeds a preset warning value, a sealing failure monitoring alarm is triggered.

[0192] Figure 2 The intelligent monitoring system for sealing performance of a liquid-driven compressor can perform Figure 1 The implementation principle and technical effects of the intelligent monitoring method for sealing performance of a liquid-driven compressor in the embodiment shown in the above are not described again. The specific operation manner of each module and unit of the intelligent monitoring system for sealing performance of a liquid-driven compressor in the above embodiment has been described in detail in the embodiment related to the method, and will not be described in detail here.

[0193] In one possible design, Figure 2 The intelligent monitoring system for sealing performance of a liquid-driven compressor in the embodiment shown in the above can be implemented as a computing device, as shown in Figure 3 The computing device can comprise a storage component 31 and a processing component 32.

[0194] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 32.

[0195] The processing component 32 is configured to execute methods described above. Figure 1 The embodiment is an intelligent monitoring method for sealing performance of a liquid-driven compressor.

[0196] The processing component 32 can include one or more processors to execute computer instructions to complete all or part of the steps in the above methods. Of course, the processing component can also be one or more Application-Specific Integrated Circuit (ASIC), Digital Signal Process (DSP), Digital Signal Process Device (DSPD), Programmable Logic Device (PLD), Field Programmable Gate Array (FPGA), controller, microcontroller, microprocessor or other electronic component, for executing the above methods.

[0197] The storage component 31 is configured to store various types of data to support the operation of the terminal. The storage component can be implemented by any type of volatile or non-volatile storage devices or their combinations, such as Static Random-Access Memory (SRAM), Electrically Erasable Programmable Read Only Memory (EEPROM), Erasable Programmable Read Only Memory (EPROM), Programmable Read Only Memory (PROM), Read Only Memory (ROM), magnetic storage, flash memory, magnetic disk or optical disk.

[0198] Of course, the computing device can also include other components, such as input / output interface, display component, communication component, etc.

[0199] The input / output interface provides an interface between the processing component and peripheral interface modules, which can be output devices, input devices, etc.

[0200] The communication component is configured to facilitate wired or wireless communication between the computing device and other devices, etc.

[0201] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform, and the computing device can be a cloud server, and the processing component and the storage component can be basic server resources rented or purchased from the cloud computing platform.

[0202] The embodiment of the application further provides a computer storage medium storing a computer program, and the computer program can realize the method when executed by a computer. Figure 1 The embodiment of the application further provides a computer storage medium storing a computer program, and the computer program can realize the method when executed by a computer.

[0203] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the system, the device and the unit described above can refer to the corresponding processes in the foregoing method embodiments, and will not be described here.

[0204] The device embodiments described above are only schematic, and the units described as separate components can or can not be physically separated, and the components displayed as units can or can not be physical units, that is, they can be located in one place, or can be distributed on multiple network units. Part or all of the modules can be selected to achieve the purpose of the embodiment scheme according to actual needs. Those skilled in the art can understand and implement without creative labor.

[0205] Through the foregoing description of the embodiments, those skilled in the art can clearly understand that the embodiments can be realized by means of software and the necessary general hardware platform, and of course, can also be realized by hardware. Based on such understanding, the foregoing technical solutions can be embodied in the form of a software product, which can be stored in a computer readable storage medium, such as a ROM / RAM, a magnetic disk, an optical disk, etc., and includes a plurality of instructions to make a computer device (which can be a personal computer, a server, or a network device, etc.) execute the methods described in the embodiments or some parts of the embodiments.

[0206] Finally, it should be noted that: the foregoing embodiments are only used to illustrate the technical solutions of the application, and not to limit them; although the application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing embodiments, or make equivalent replacement to some technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the application.

Claims

1. A method for intelligent monitoring of the sealing performance of a liquid-driven compressor, characterized in that, include: Collect high-frequency pressure data, circulating fluid flow data, and friction sound signals at the sealing interface of the liquid-driven compressor sealing cavity; Based on the high-frequency pressure data and the circulating fluid flow rate data, a pressure-flow covariance relationship is constructed; The frictional sound signal is subjected to directional enhancement processing to form sound energy intensity distribution information; The DIANA algorithm is used to adaptively cluster the acoustic energy intensity distribution information to generate sealing anomaly indicators; By jointly analyzing the pressure-flow covariance relationship and the sealing anomaly index, monitoring results characterizing the sealing performance of the liquid-driven compressor are generated. When the monitoring results exceed the preset warning value, a sealing failure monitoring alarm is triggered. The step of constructing a pressure-flow covariance relationship based on the high-frequency pressure data and the circulating fluid flow rate data includes: The high-frequency pressure data is divided into multiple discrete pressure segments according to the single reciprocating motion cycle of the piston rod of the hydraulic compressor. Extract the time sequence of pressure peaks from each of the pressure segments; The circulating fluid flow data is simultaneously divided into multiple flow fluctuation segments; Calculate the instantaneous rate of change of flow for each flow fluctuation segment; Establish a dynamic correlation between the time sequence of the pressure peak and the instantaneous flow rate change within the same motion cycle; Based on the dynamic correlation over N consecutive motion cycles, a pressure-flow covariance relationship is constructed, where N is greater than or equal to 3.

2. The method according to claim 1, characterized in that, The step of using the DIANA algorithm to adaptively cluster the acoustic energy intensity distribution information to generate sealing anomaly indicators includes: Using the spatial grid of the sealed interface as sample points, an initial cluster is constructed; Calculate the dispersion of the sound energy intensity values ​​within the initial cluster. If the dispersion of the sound energy intensity values ​​within the initial cluster is greater than a preset splitting threshold, divide the initial cluster into two sub-clusters. Each of the subclusters is taken as a corresponding intermediate cluster, and the calculation and splitting operations are performed iteratively until the dispersion of the acoustic energy intensity values ​​in all intermediate clusters is less than or equal to the preset splitting threshold, thus obtaining the target cluster. The number of spatial grids in the target cluster whose acoustic energy intensity values ​​exceed the preset anomaly boundary is counted to generate a sealing anomaly index.

3. The method according to claim 2, characterized in that, The calculation of the dispersion of the acoustic energy intensity values ​​within the initial cluster includes: Based on the acoustic energy intensity values ​​corresponding to all spatial grids within the initial cluster, a set of acoustic energy intensity values ​​is formed. The set of sound energy intensity values ​​is subjected to outlier correction processing; Based on the corrected set of acoustic energy intensity values, the dispersion of acoustic energy intensity values ​​within the initial cluster is determined using the standard deviation calculation formula.

4. The method according to claim 2, characterized in that, The process of dividing the initial cluster into two sub-clusters includes: Construct the spatial gradient field of the acoustic energy intensity values ​​of the initial cluster; In the spatial gradient field of the acoustic energy intensity value, determine the gradient vector with the largest modulus; Establish a segmentation plane along the direction of the gradient vector; Using the segmentation plane as the boundary, the initial cluster is divided into two sub-clusters.

5. The method according to claim 1, characterized in that, The directional enhancement processing of the frictional sound signal to form sound energy intensity distribution information includes: The acoustic energy components of the frictional sound signal are spatially selectively amplified by acoustic wave guidance; Frequency domain analysis was performed on the enhanced acoustic energy components to identify specific frequency bands associated with high-frequency leakage; Within the specified frequency band, sound energy intensity data is obtained; The acoustic energy intensity data is mapped to the spatial coordinate system of the sealed interface to form a two-dimensional acoustic energy intensity distribution map. The two-dimensional acoustic energy intensity distribution map is divided into multiple equally spaced spatial grids. The cumulative value of acoustic energy intensity in each spatial grid is calculated. Based on the cumulative value of acoustic energy intensity in each spatial grid, acoustic energy intensity distribution information is generated.

6. The method according to claim 1, characterized in that, The joint analysis of the pressure-flow covariance relationship and the sealing anomaly index generates monitoring results characterizing the sealing performance of the liquid-driven compressor, including: Based on the aforementioned pressure-flow covariance relationship, calculate the pressure-flow state scalar; The sealing anomaly index is converted into an acoustic anomaly scalar; The pressure and flow state scalar and the acoustic anomaly scalar are weighted and fused together, and the weighted fusion calculation result is used as the monitoring result characterizing the sealing performance of the liquid-driven compressor.

7. An intelligent monitoring system for the sealing performance of a liquid-driven compressor, characterized in that, include: The acquisition module is used to acquire high-frequency pressure data, circulating fluid flow data, and friction sound signals at the sealing interface of the liquid-driven compressor sealing cavity. The module is used to construct a pressure-flow covariance relationship based on the high-frequency pressure data and the circulating fluid flow data. An enhancement module is used to perform directional enhancement processing on the frictional sound signal to form sound energy intensity distribution information; The generation module is used to perform adaptive clustering processing on the acoustic energy intensity distribution information using the DIANA algorithm to generate sealing anomaly indicators. The triggering module is used to jointly analyze the pressure-flow covariance relationship and the sealing anomaly index to generate monitoring results characterizing the sealing performance of the liquid-driven compressor. When the monitoring results exceed the preset warning value, a sealing failure monitoring alarm is triggered. The step of constructing a pressure-flow covariance relationship based on the high-frequency pressure data and the circulating fluid flow rate data includes: The high-frequency pressure data is divided into multiple discrete pressure segments according to the single reciprocating motion cycle of the piston rod of the hydraulic compressor. Extract the time sequence of pressure peaks from each of the pressure segments; The circulating fluid flow data is simultaneously divided into multiple flow fluctuation segments; Calculate the instantaneous rate of change of flow for each flow fluctuation segment; Establish a dynamic correlation between the time sequence of the pressure peak and the instantaneous flow rate change within the same motion cycle; Based on the dynamic correlation over N consecutive motion cycles, a pressure-flow covariance relationship is constructed, where N is greater than or equal to 3.

8. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement an intelligent monitoring method for the sealing performance of a liquid-driven compressor as described in any one of claims 1 to 6.

9. A computer storage medium, characterized in that, The system contains a computer program that, when executed by a computer, implements an intelligent monitoring method for the sealing performance of a liquid-driven compressor as described in any one of claims 1 to 6.

Images