CN117370824A - Intelligent monitoring method and system for air inlet state of air compressor - Google Patents
Intelligent monitoring method and system for air inlet state of air compressor Download PDFInfo
- Publication number
- CN117370824A CN117370824A CN202311671501.8A CN202311671501A CN117370824A CN 117370824 A CN117370824 A CN 117370824A CN 202311671501 A CN202311671501 A CN 202311671501A CN 117370824 A CN117370824 A CN 117370824A
- Authority
- CN
- China
- Prior art keywords
- air inlet
- air compressor
- parameter
- actual
- preset
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 50
- 238000012544 monitoring process Methods 0.000 title claims abstract description 36
- 239000011159 matrix material Substances 0.000 claims abstract description 55
- 238000012545 processing Methods 0.000 claims abstract description 12
- 238000012423 maintenance Methods 0.000 claims description 60
- 230000002159 abnormal effect Effects 0.000 claims description 49
- 230000007704 transition Effects 0.000 claims description 24
- 238000007476 Maximum Likelihood Methods 0.000 claims description 6
- 238000013135 deep learning Methods 0.000 claims description 6
- 239000000428 dust Substances 0.000 claims description 6
- 238000005457 optimization Methods 0.000 claims description 6
- 238000012549 training Methods 0.000 claims description 6
- 238000009826 distribution Methods 0.000 claims description 4
- 238000004519 manufacturing process Methods 0.000 abstract description 9
- 238000010586 diagram Methods 0.000 description 14
- 238000003860 storage Methods 0.000 description 7
- 238000004088 simulation Methods 0.000 description 6
- 238000005516 engineering process Methods 0.000 description 4
- 230000005540 biological transmission Effects 0.000 description 3
- 238000012937 correction Methods 0.000 description 3
- 230000007547 defect Effects 0.000 description 3
- 230000000694 effects Effects 0.000 description 3
- 230000008569 process Effects 0.000 description 3
- 238000006243 chemical reaction Methods 0.000 description 2
- 238000005094 computer simulation Methods 0.000 description 2
- 238000001514 detection method Methods 0.000 description 2
- 238000011161 development Methods 0.000 description 2
- 238000003745 diagnosis Methods 0.000 description 2
- 238000007726 management method Methods 0.000 description 2
- 230000003287 optical effect Effects 0.000 description 2
- 238000004458 analytical method Methods 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000009776 industrial production Methods 0.000 description 1
- 238000013450 outlier detection Methods 0.000 description 1
- 238000005498 polishing Methods 0.000 description 1
- 238000007781 pre-processing Methods 0.000 description 1
- 238000010223 real-time analysis Methods 0.000 description 1
- 230000008439 repair process Effects 0.000 description 1
- 230000004044 response Effects 0.000 description 1
- 238000006467 substitution reaction Methods 0.000 description 1
- 238000012384 transportation and delivery Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
- G06F18/232—Non-hierarchical techniques
- G06F18/2321—Non-hierarchical techniques using statistics or function optimisation, e.g. modelling of probability density functions
-
- F—MECHANICAL ENGINEERING; LIGHTING; HEATING; WEAPONS; BLASTING
- F04—POSITIVE - DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS FOR LIQUIDS OR ELASTIC FLUIDS
- F04B—POSITIVE-DISPLACEMENT MACHINES FOR LIQUIDS; PUMPS
- F04B51/00—Testing machines, pumps, or pumping installations
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F17/00—Digital computing or data processing equipment or methods, specially adapted for specific functions
- G06F17/10—Complex mathematical operations
- G06F17/16—Matrix or vector computation, e.g. matrix-matrix or matrix-vector multiplication, matrix factorization
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/20—Administration of product repair or maintenance
Abstract
The invention relates to the technical field of compressor monitoring, in particular to an intelligent monitoring method and system for the air inlet state of an air compressor, which are used for processing the actual air inlet parameters of a target air compressor according to a fuzzy C-means clustering algorithm to obtain a plurality of membership matrixes; detecting and correcting each membership matrix through an LOF algorithm to obtain a plurality of corrected membership matrixes; obtaining actual air inlet parameters of different categories according to corresponding elements in each corrected membership matrix; comparing the same kind of actual air inlet parameters with preset air inlet parameters to judge whether the parameter states of the actual air inlet parameters in the target air compressor are normal or not; and judging the real-time fault type information of the target air compressor according to the parameter state of the actual air inlet parameter in the target air compressor. The method can rapidly analyze the fault state information of the air compressor, can reduce the shutdown time of equipment and improve the production efficiency and the equipment utilization rate.
Description
Technical Field
The invention relates to the technical field of compressor monitoring, in particular to an intelligent monitoring method and system for an air inlet state of an air compressor.
Background
Air compressors are equipment commonly used in industrial production for compressing air into high pressure gas for use in various process equipment. In order to ensure the normal operation and safety of the air compressor, it is important to intelligently monitor the air inlet state of the air compressor. At present, with the development of the Internet of things, a sensor technology and a data analysis technology, the development of an intelligent monitoring method for the air inlet state of an air compressor is attracting attention. In the existing method, when a large amount of data is transmitted to a cloud for real-time analysis and processing, the problems of abnormal data transmission, data loss, insufficient processing capacity and the like exist, when a prediction model of an air inlet state is established, the model precision is insufficient or the prediction capacity for the abnormal state is limited, so that the intelligent diagnosis effect is not ideal, the real-time performance and the accuracy of a monitoring result are influenced, the data processing algorithm is complex, and the system operation efficiency and the robustness are influenced.
Disclosure of Invention
The invention overcomes the defects of the prior art and provides an intelligent monitoring method and system for the air inlet state of an air compressor.
The technical scheme adopted by the invention for achieving the purpose is as follows:
the invention discloses an intelligent monitoring method for the air inlet state of an air compressor, which comprises the following steps:
Acquiring the actual running state of the target air compressor, and determining the preset air inlet parameter of the target air compressor according to the actual running state of the target air compressor;
acquiring actual air inlet parameters of a target air compressor within preset time, and processing the actual air inlet parameters of the target air compressor according to a fuzzy C-means clustering algorithm to obtain a plurality of membership matrixes;
introducing an LOF algorithm, detecting and correcting each membership matrix through the LOF algorithm, and obtaining a plurality of corrected membership matrices; obtaining actual air inlet parameters of different categories according to corresponding elements in each corrected membership matrix;
comparing the same kind of actual air inlet parameters with preset air inlet parameters to judge whether the parameter states of the actual air inlet parameters in the target air compressor are normal or not; and judging real-time fault type information of the target air compressor according to the parameter state of the actual air inlet parameter in the target air compressor, and generating a corresponding fault maintenance scheme.
Further, in a preferred embodiment of the present invention, an actual operation state of the target air compressor is obtained, and a preset air intake parameter of the target air compressor is determined according to the actual operation state of the target air compressor, which specifically includes:
Retrieving and obtaining standard air inlet parameters corresponding to various preset running states of the target air compressor based on a big data network;
establishing a prediction model based on a deep learning network, and importing standard air inlet parameters corresponding to various preset running states of a target air compressor into the prediction model for training to obtain a trained prediction model;
acquiring service life information, maintenance condition information and mechanical wear degree information of a target air compressor, and determining the actual running state of the target air compressor according to the service life information, the maintenance condition information and the mechanical wear degree information;
the actual running state of the target air compressor is led into the trained prediction model, and the preset air inlet parameters of the target air compressor under the current actual running state condition are predicted;
the air inlet parameters comprise air inlet temperature, air inlet pressure, air inlet flow rate, air inlet volume, air inlet humidity and air inlet dust concentration.
Further, in a preferred embodiment of the present invention, the actual air intake parameters of the target air compressor are processed according to a fuzzy C-means clustering algorithm to obtain a plurality of membership matrices, which are specifically:
The method comprises the steps of presetting the clustering quantity and the ambiguity parameters, and randomly initializing the weight of each actual air inlet parameter according to the preset clustering quantity and the ambiguity parameters; wherein the weight of each actual air intake parameter is a positive number, and the total weight is 1;
multiplying the value of each actual air intake parameter with the corresponding weight to obtain the weighted value of each actual air intake parameter, and dividing the weighted value of each actual air intake parameter by the preset weight to obtain the weighted average value of each actual air intake parameter;
presetting a plurality of threshold intervals, and distributing each actual air inlet parameter into a matched threshold interval according to the weighted average value of each actual air inlet parameter;
after distribution is completed, combining weighted average values of actual air inlet parameters in all threshold intervals to form a vector, and determining fuzzy clustering centroids of the actual air inlet parameters in all threshold intervals according to the formed vector;
calculating Euclidean distance between the actual air inlet parameter in each threshold interval and the fuzzy clustering centroid thereof, and calculating membership degree between the actual air inlet parameter in each threshold interval and the fuzzy clustering centroid thereof according to the Euclidean distance;
judging whether the membership degree between the actual air inlet parameter and the fuzzy clustering centroid in each threshold interval is smaller than the preset membership degree one by one, and if so, eliminating the corresponding actual air inlet parameter in the corresponding threshold interval;
Updating the residual actual air inlet parameters in each threshold interval, and constructing a plurality of membership matrixes according to membership between the residual actual air inlet parameters in each threshold interval and the fuzzy clustering centroid of the residual actual air inlet parameters.
Further, in a preferred embodiment of the present invention, an LOF algorithm is introduced, and detection and correction are performed on each membership matrix by the LOF algorithm, so as to obtain a plurality of corrected membership matrices, which are specifically:
for each membership degree matrix, calculating Manhattan distances between each element in the matrix and elements in a preset field, summing the Manhattan distances between each element and the elements in the preset field, and taking an average value to obtain local reachable densities of each element;
calculating a local outlier factor of each element according to the local reachable density, wherein the local outlier factor represents the outlier degree of the element relative to other elements in the neighborhood of the element; comparing the local outlier factor of each element of the membership matrix with a preset threshold;
taking the element with the local outlier greater than the preset threshold value as a clustering abnormal element, importing the clustering abnormal element into other membership matrixes, and judging whether the local outlier of the clustering abnormal element in the other membership matrixes is greater than the preset threshold value;
If the local outlier factors of the clustering abnormal elements in the rest membership matrixes are larger than a preset threshold value, the clustering abnormal elements are thoroughly removed; if the local outlier factor of the clustering abnormal element in the rest membership matrix is not greater than a preset threshold value, the clustering abnormal element is distributed to the membership matrix with the minimum local outlier factor;
and updating the position condition of each element in each membership matrix to obtain a corrected membership matrix.
Further, in a preferred embodiment of the present invention, the actual intake parameters of the same category are compared with the preset intake parameters to determine whether the parameter status of the actual intake parameters in the target air compressor is normal, specifically:
comparing the actual air inlet parameters of the same category with preset air inlet parameters to obtain parameter deviation values; comparing the parameter deviation value with a preset deviation value;
if the parameter deviation value is larger than a preset deviation value, marking the corresponding actual air inlet parameter in the target air compressor as an abnormal air inlet parameter;
and if the parameter deviation value is not greater than the preset deviation value, marking the corresponding actual air inlet parameter in the target air compressor as a normal air inlet parameter.
Further, in a preferred embodiment of the present invention, the real-time fault type information of the target air compressor is determined according to the parameter status of the actual air intake parameter in the target air compressor, and a corresponding fault maintenance scheme is generated, which specifically includes:
searching based on a big data network to obtain corresponding fault characteristic data when various fault types occur to a target air compressor, defining random variables according to the fault characteristic data, and determining target nodes of a Bayesian network structure according to the random variables;
constructing a Bayesian network based on the target node, performing iterative optimization on the Bayesian network based on a maximum likelihood estimation method until the iterative times reach a preset time, and outputting a conditional probability table; introducing a Markov chain, calculating the state transition probability of the target air compressor when various fault types occur under various preset abnormal air inlet parameter combination conditions through the Markov chain and combining the conditional probability table;
constructing a probability state transition matrix according to the state transition probability, and importing the state transition matrix into a Bayesian network for structure learning to obtain a Bayesian network model;
The parameter state is an abnormal air inlet parameter, the abnormal air inlet parameter is led into the Bayesian network to be deduced and predicted, the fault probability of various fault types of the target air compressor is obtained, the fault types with the fault probability larger than the preset fault probability are extracted and output, and the real-time fault type information of the target air compressor is obtained;
and generating a search tag according to the real-time fault type information of the target air compressor, searching a big data network based on the search tag to obtain a corresponding fault maintenance scheme, and pushing the fault maintenance scheme to a preset platform for display.
The invention discloses an intelligent monitoring system for the air inlet state of an air compressor, which comprises a memory and a processor, wherein an intelligent monitoring method program for the air inlet state of the air compressor is stored in the memory, and when the intelligent monitoring method program for the air inlet state of the air compressor is executed by the processor, the following steps are realized:
acquiring the actual running state of the target air compressor, and determining the preset air inlet parameter of the target air compressor according to the actual running state of the target air compressor;
Acquiring actual air inlet parameters of a target air compressor within preset time, and processing the actual air inlet parameters of the target air compressor according to a fuzzy C-means clustering algorithm to obtain a plurality of membership matrixes;
introducing an LOF algorithm, detecting and correcting each membership matrix through the LOF algorithm, and obtaining a plurality of corrected membership matrices; obtaining actual air inlet parameters of different categories according to corresponding elements in each corrected membership matrix;
comparing the same kind of actual air inlet parameters with preset air inlet parameters to judge whether the parameter states of the actual air inlet parameters in the target air compressor are normal or not; and judging real-time fault type information of the target air compressor according to the parameter state of the actual air inlet parameter in the target air compressor, and generating a corresponding fault maintenance scheme.
Further, in a preferred embodiment of the present invention, an actual operation state of the target air compressor is obtained, and a preset air intake parameter of the target air compressor is determined according to the actual operation state of the target air compressor, which specifically includes:
retrieving and obtaining standard air inlet parameters corresponding to various preset running states of the target air compressor based on a big data network;
Establishing a prediction model based on a deep learning network, and importing standard air inlet parameters corresponding to various preset running states of a target air compressor into the prediction model for training to obtain a trained prediction model;
acquiring service life information, maintenance condition information and mechanical wear degree information of a target air compressor, and determining the actual running state of the target air compressor according to the service life information, the maintenance condition information and the mechanical wear degree information;
the actual running state of the target air compressor is led into the trained prediction model, and the preset air inlet parameters of the target air compressor under the current actual running state condition are predicted;
the air inlet parameters comprise air inlet temperature, air inlet pressure, air inlet flow rate, air inlet volume, air inlet humidity and air inlet dust concentration.
Further, in a preferred embodiment of the present invention, the actual intake parameters of the same category are compared with the preset intake parameters to determine whether the parameter status of the actual intake parameters in the target air compressor is normal, specifically:
comparing the actual air inlet parameters of the same category with preset air inlet parameters to obtain parameter deviation values; comparing the parameter deviation value with a preset deviation value;
If the parameter deviation value is larger than a preset deviation value, marking the corresponding actual air inlet parameter in the target air compressor as an abnormal air inlet parameter;
and if the parameter deviation value is not greater than the preset deviation value, marking the corresponding actual air inlet parameter in the target air compressor as a normal air inlet parameter.
Further, in a preferred embodiment of the present invention, the real-time fault type information of the target air compressor is determined according to the parameter status of the actual air intake parameter in the target air compressor, and a corresponding fault maintenance scheme is generated, which specifically includes:
searching based on a big data network to obtain corresponding fault characteristic data when various fault types occur to a target air compressor, defining random variables according to the fault characteristic data, and determining target nodes of a Bayesian network structure according to the random variables;
constructing a Bayesian network based on the target node, performing iterative optimization on the Bayesian network based on a maximum likelihood estimation method until the iterative times reach a preset time, and outputting a conditional probability table; introducing a Markov chain, calculating the state transition probability of the target air compressor when various fault types occur under various preset abnormal air inlet parameter combination conditions through the Markov chain and combining the conditional probability table;
Constructing a probability state transition matrix according to the state transition probability, and importing the state transition matrix into a Bayesian network for structure learning to obtain a Bayesian network model;
the parameter state is an abnormal air inlet parameter, the abnormal air inlet parameter is led into the Bayesian network to be deduced and predicted, the fault probability of various fault types of the target air compressor is obtained, the fault types with the fault probability larger than the preset fault probability are extracted and output, and the real-time fault type information of the target air compressor is obtained;
and generating a search tag according to the real-time fault type information of the target air compressor, searching a big data network based on the search tag to obtain a corresponding fault maintenance scheme, and pushing the fault maintenance scheme to a preset platform for display.
The invention solves the technical defects existing in the background technology, and has the following beneficial effects: acquiring actual air inlet parameters of a target air compressor within preset time, and processing the actual air inlet parameters of the target air compressor according to a fuzzy C-means clustering algorithm to obtain a plurality of membership matrixes; introducing an LOF algorithm, detecting and correcting each membership matrix through the LOF algorithm, and obtaining a plurality of corrected membership matrices; obtaining actual air inlet parameters of different categories according to corresponding elements in each corrected membership matrix; comparing the same kind of actual air inlet parameters with preset air inlet parameters to judge whether the parameter states of the actual air inlet parameters in the target air compressor are normal or not; and judging real-time fault type information of the target air compressor according to the parameter state of the actual air inlet parameter in the target air compressor, and generating a corresponding fault maintenance scheme. The method can rapidly analyze the fault state information of the air compressor, and the monitoring result has higher timeliness and accuracy, can reduce the shutdown time of equipment, and improves the production efficiency and the equipment utilization rate.
Drawings
In order to more clearly illustrate the embodiments of the invention or the technical solutions in the prior art, the drawings that are required in the embodiments or the description of the prior art will be briefly described, it being obvious that the drawings in the following description are only some embodiments of the invention, and that other embodiments of the drawings can be obtained according to these drawings without inventive effort for a person skilled in the art.
FIG. 1 is a first method flow chart of an intelligent monitoring method for the air intake state of an air compressor;
FIG. 2 is a second method flow chart of a method for intelligently monitoring the intake status of an air compressor;
FIG. 3 is a third method flow chart of a method for intelligently monitoring the air inlet state of an air compressor;
fig. 4 is a system block diagram of an intelligent monitoring system for the intake status of an air compressor.
Detailed Description
In order that the above-recited objects, features and advantages of the present invention will be more clearly understood, a more particular description of the invention will be rendered by reference to the appended drawings and appended detailed description. It should be noted that, in the case of no conflict, the embodiments of the present application and the features in the embodiments may be combined with each other.
In the following description, numerous specific details are set forth in order to provide a thorough understanding of the present invention, however, the present invention may be practiced in other ways than those described herein, and therefore the scope of the present invention is not limited to the specific embodiments disclosed below.
As shown in fig. 1, the first aspect of the present invention discloses an intelligent monitoring method for an air intake state of an air compressor, comprising the following steps:
s102: acquiring the actual running state of the target air compressor, and determining the preset air inlet parameter of the target air compressor according to the actual running state of the target air compressor;
s104: acquiring actual air inlet parameters of a target air compressor within preset time, and processing the actual air inlet parameters of the target air compressor according to a fuzzy C-means clustering algorithm to obtain a plurality of membership matrixes;
s106: introducing an LOF algorithm, detecting and correcting each membership matrix through the LOF algorithm, and obtaining a plurality of corrected membership matrices; obtaining actual air inlet parameters of different categories according to corresponding elements in each corrected membership matrix;
s108: comparing the same kind of actual air inlet parameters with preset air inlet parameters to judge whether the parameter states of the actual air inlet parameters in the target air compressor are normal or not; and judging real-time fault type information of the target air compressor according to the parameter state of the actual air inlet parameter in the target air compressor, and generating a corresponding fault maintenance scheme.
As shown in fig. 2, in a preferred embodiment of the present invention, an actual operation state of the target air compressor is obtained, and a preset air intake parameter of the target air compressor is determined according to the actual operation state of the target air compressor, which is specifically:
s202: retrieving and obtaining standard air inlet parameters corresponding to various preset running states of the target air compressor based on a big data network;
s204: establishing a prediction model based on a deep learning network, and importing standard air inlet parameters corresponding to various preset running states of a target air compressor into the prediction model for training to obtain a trained prediction model;
s206: acquiring service life information, maintenance condition information and mechanical wear degree information of a target air compressor, and determining the actual running state of the target air compressor according to the service life information, the maintenance condition information and the mechanical wear degree information;
s208: the actual running state of the target air compressor is led into the trained prediction model, and the preset air inlet parameters of the target air compressor under the current actual running state condition are predicted;
the air inlet parameters comprise air inlet temperature, air inlet pressure, air inlet flow rate, air inlet volume, air inlet humidity and air inlet dust concentration.
It should be noted that, factors such as service life, maintenance condition, mechanical wear degree and the like relate to the operation state of the air compressor, and standard operation parameters of the air compressor are different in different operation states, for example, the internal temperature value of the air compressor in the initial wear stage is slightly higher, which is caused by the mutual wear of transmission parts and belongs to a normal phenomenon. The method can further predict and obtain the preset air inlet parameters of the air compressor, namely the standard air inlet parameters, according to the operation state of the standard air compressor, so as to judge whether each actual air inlet parameter of the air compressor is normal or not according to the predicted air inlet parameters obtained by prediction.
Further, in a preferred embodiment of the present invention, the actual air intake parameters of the target air compressor are processed according to a fuzzy C-means clustering algorithm to obtain a plurality of membership matrices, which are specifically:
the method comprises the steps of presetting the clustering quantity and the ambiguity parameters, and randomly initializing the weight of each actual air inlet parameter according to the preset clustering quantity and the ambiguity parameters; wherein the weight of each actual air intake parameter is a positive number, and the total weight is 1;
multiplying the value of each actual air intake parameter with the corresponding weight to obtain the weighted value of each actual air intake parameter, and dividing the weighted value of each actual air intake parameter by the preset weight to obtain the weighted average value of each actual air intake parameter;
Presetting a plurality of threshold intervals, and distributing each actual air inlet parameter into a matched threshold interval according to the weighted average value of each actual air inlet parameter;
after distribution is completed, combining weighted average values of actual air inlet parameters in all threshold intervals to form a vector, and determining fuzzy clustering centroids of the actual air inlet parameters in all threshold intervals according to the formed vector;
calculating Euclidean distance between the actual air inlet parameter in each threshold interval and the fuzzy clustering centroid thereof, and calculating membership degree between the actual air inlet parameter in each threshold interval and the fuzzy clustering centroid thereof according to the Euclidean distance;
judging whether the membership degree between the actual air inlet parameter and the fuzzy clustering centroid in each threshold interval is smaller than the preset membership degree one by one, and if so, eliminating the corresponding actual air inlet parameter in the corresponding threshold interval;
updating the residual actual air inlet parameters in each threshold interval, and constructing a plurality of membership matrixes according to membership between the residual actual air inlet parameters in each threshold interval and the fuzzy clustering centroid of the residual actual air inlet parameters.
It should be noted that, when the actual air intake parameters of the air compressor are acquired through a series of sensors inside the air compressor, the actual air intake parameters need to be classified at this time, so that the parameter types of each actual air intake parameter which are primarily screened, such as which are actual air intake temperature parameters and which are actual air intake humidity, are determined, so that the actual air intake parameters of the corresponding types are compared with the preset air intake parameters, and whether the actual air intake parameters of each type are normal is determined. Specifically, the number of clusters (k-value) and the ambiguity parameters (typically denoted by m) need to be determined first, and then the weights of each actual intake parameter are initialized randomly; wherein the weight of each actual air intake parameter is a positive number, and the total weight is 1; and calculating the mass center of the fuzzy cluster (also called as the fuzzy cluster center) by using a weighted average algorithm, and combining the weighted average values of each actual air intake parameter to form a vector, wherein the vector is the mass center point of the cluster, namely the cluster center. By calculating the cluster center using a weighted average method, the importance of different features or attributes can be taken into account, thereby more accurately representing the central location of the clusters. And then judging whether the membership degree between the actual air inlet parameter and the fuzzy clustering centroid in each threshold interval is smaller than the preset membership degree one by one, if so, rejecting the corresponding actual air inlet parameter in the corresponding threshold interval. The method can quickly perform preliminary clustering on the acquired actual air inlet parameters, and distinguish the actual air inlet parameters of different categories at a preliminary stage, so that a corresponding membership matrix is obtained, the algorithm is easy to realize, the system robustness can be improved, and the system operation efficiency and response rate are improved.
Further, in a preferred embodiment of the present invention, an LOF algorithm is introduced, and detection and correction are performed on each membership matrix by the LOF algorithm, so as to obtain a plurality of corrected membership matrices, which are specifically:
for each membership degree matrix, calculating Manhattan distances between each element in the matrix and elements in a preset field, summing the Manhattan distances between each element and the elements in the preset field, and taking an average value to obtain local reachable densities of each element;
calculating a local outlier factor of each element according to the local reachable density, wherein the local outlier factor represents the outlier degree of the element relative to other elements in the neighborhood of the element; comparing the local outlier factor of each element of the membership matrix with a preset threshold;
taking the element with the local outlier greater than the preset threshold value as a clustering abnormal element, importing the clustering abnormal element into other membership matrixes, and judging whether the local outlier of the clustering abnormal element in the other membership matrixes is greater than the preset threshold value;
if the local outlier factors of the clustering abnormal elements in the rest membership matrixes are larger than a preset threshold value, the clustering abnormal elements are thoroughly removed; if the local outlier factor of the clustering abnormal element in the rest membership matrix is not greater than a preset threshold value, the clustering abnormal element is distributed to the membership matrix with the minimum local outlier factor;
And updating the position condition of each element in each membership matrix to obtain a corrected membership matrix.
It should be noted that, when a large number of actual intake parameters are classified by the fuzzy C-means clustering algorithm, due to the defect of the fuzzy C-means clustering algorithm, a classification error condition is unavoidable in the classification process, such as classifying the intake temperature parameter at a certain moment into the intake humidity parameter. Therefore, after preliminary classification, each membership degree needs to be corrected, so that the data precision and reliability are improved, and the monitoring precision is further improved. The LOF algorithm is also called local outlier factor algorithm, and is based on the outlier detection method of the density, the outlier degree of each data point is determined by calculating the density of each data point relative to other data points in the neighborhood of the data point, abnormal points or outliers in a membership matrix are screened out through the LOF algorithm, the abnormal points possibly cause inaccuracy of a clustering result, and the stability and the accuracy of clustering can be improved through correction. And the redundant points with overlarge redundancy can be removed by correcting the membership matrix, so that the clustering result is more reasonable and reliable, the clustering effect and quality are improved, the data cleaning and preprocessing are facilitated, the redundant points with less influence on the clustering result are removed, and the quality and accuracy of the data are improved. In addition, the interference of the redundant points on the model can be reduced, and the stability and the robustness of the model are improved.
As shown in fig. 3, in a preferred embodiment of the present invention, the actual intake parameters of the same category are compared with the preset intake parameters to determine whether the parameter status of the actual intake parameters in the target air compressor is normal, specifically:
s302: comparing the actual air inlet parameters of the same category with preset air inlet parameters to obtain parameter deviation values; comparing the parameter deviation value with a preset deviation value;
s304: if the parameter deviation value is larger than a preset deviation value, marking the corresponding actual air inlet parameter in the target air compressor as an abnormal air inlet parameter;
s306: and if the parameter deviation value is not greater than the preset deviation value, marking the corresponding actual air inlet parameter in the target air compressor as a normal air inlet parameter.
It should be noted that, by the method, whether each actual air intake parameter of the target air compressor is determined can be rapidly.
Further, in a preferred embodiment of the present invention, the real-time fault type information of the target air compressor is determined according to the parameter status of the actual air intake parameter in the target air compressor, and a corresponding fault maintenance scheme is generated, which specifically includes:
Searching based on a big data network to obtain corresponding fault characteristic data when various fault types occur to a target air compressor, defining random variables according to the fault characteristic data, and determining target nodes of a Bayesian network structure according to the random variables;
constructing a Bayesian network based on the target node, performing iterative optimization on the Bayesian network based on a maximum likelihood estimation method until the iterative times reach a preset time, and outputting a conditional probability table; introducing a Markov chain, calculating the state transition probability of the target air compressor when various fault types occur under various preset abnormal air inlet parameter combination conditions through the Markov chain and combining the conditional probability table;
constructing a probability state transition matrix according to the state transition probability, and importing the state transition matrix into a Bayesian network for structure learning to obtain a Bayesian network model;
the parameter state is an abnormal air inlet parameter, the abnormal air inlet parameter is led into the Bayesian network to be deduced and predicted, the fault probability of various fault types of the target air compressor is obtained, the fault types with the fault probability larger than the preset fault probability are extracted and output, and the real-time fault type information of the target air compressor is obtained;
And generating a search tag according to the real-time fault type information of the target air compressor, searching a big data network based on the search tag to obtain a corresponding fault maintenance scheme, and pushing the fault maintenance scheme to a preset platform for display.
The Bayesian network is a probability graph model, is used for describing the dependency relationship and probability distribution among random variables based on Bayesian theorem and graph theory, predicts the fault probability through the Bayesian network, can utilize historical data and a probability inference method to carry out probabilistic prediction on equipment faults, and provides decision support for equipment maintenance and management. Specifically, potential problems can be found in advance through Bayesian network prediction of possible faults, corresponding maintenance and repair measures are adopted, serious influences on production and equipment caused by the faults are avoided, so that the state monitoring has higher timeliness, the outage time of the equipment is reduced, the production efficiency and the equipment utilization rate are improved, and the possible faults can be predicted to help engineers to quickly locate the problems, perform fault diagnosis and analysis, shorten the fault removal time and improve the reliability and the stability of the equipment.
In addition, a search tag is generated according to the real-time fault type information of the target air compressor, and the big data network is searched based on the search tag to obtain a corresponding fault maintenance scheme, and the method specifically comprises the following steps:
engineering drawing information of a target air compressor is obtained, and a three-dimensional structure model diagram of the target air compressor is constructed according to the engineering drawing information;
acquiring electrical parameter information of a target air compressor, performing format conversion on the electrical parameter information, and importing the electrical parameter information subjected to format conversion into a three-dimensional structure model diagram of the target air compressor to obtain a dynamic simulation three-dimensional model diagram;
generating a retrieval tag according to real-time fault type information of the target air compressor, and retrieving a big data network based on the retrieval tag to obtain a plurality of fault maintenance schemes;
each fault maintenance scheme is led into the dynamic simulation three-dimensional model diagram for simulation maintenance to obtain a simulation maintenance result, and simulation air inlet parameters after maintenance of the target air compressor are obtained based on the simulation maintenance result;
comparing the simulated air inlet parameters of the target air compressor subjected to simulated maintenance according to each fault maintenance scheme with preset air inlet parameters to obtain a plurality of simulated parameter difference values;
And sorting the plurality of simulation parameter differences based on the size to obtain a sorting result, extracting a fault maintenance scheme corresponding to the minimum simulation parameter difference according to the sorting result, and outputting the fault maintenance scheme.
It should be noted that, the fault maintenance scheme of group price can be screened out through the local side to recommend, thereby improving the rationality of maintenance scheme recommendation, improving the equipment operation effect after maintenance, reducing the maintenance times.
Furthermore, the method comprises the following steps:
acquiring the fault maintenance time required by a corresponding fault maintenance scheme, and determining the equipment downtime according to the fault maintenance time; predicting the lack yield of a production workshop in the maintenance process according to the equipment downtime;
obtaining the residual spare yield in a production workshop, and judging whether the missing yield is greater than the spare yield; if the characteristic data information is larger than the characteristic data information, acquiring the characteristic data information of the shutdown air compressor in the production workshop;
acquiring characteristic data information of the air compressor in an idle state in a production workshop, and calculating a hash value between the characteristic data information of the air compressor in the idle state and the characteristic data information of the air compressor in the idle state through a hash algorithm;
And sorting the hash values between the characteristic data information of the shutdown air compressor and the characteristic data information of the air compressor in the idle state to obtain a maximum hash value, and recommending the air compressor in the idle state corresponding to the maximum hash value as an alternative air compressor to a preset platform for display.
It should be noted that, when the air compressor with the fault is maintained, if the required maintenance time is too long, the delivery schedule of the product may be affected, so that the remaining air compressors capable of replacing the air compressor to be maintained to work need to be further screened out to ensure that the capacity schedule can be implemented smoothly. The method can screen out the matched air compressor, and recommend the air compressor, thereby improving the rationality of workshop production management.
Furthermore, the method comprises the following steps:
obtaining a standard characteristic three-dimensional model diagram of various components in the air compressor through a big data network, constructing a knowledge graph, and importing the standard characteristic three-dimensional model diagram of various components in the air compressor into the knowledge graph;
acquiring image information of a fault component in the air compressor, and constructing an actual three-dimensional model diagram of the fault component according to the image information;
Acquiring type information of fault components in the air compressor, and retrieving corresponding standard feature model diagrams from the knowledge graph according to the type information of the fault components;
comparing the standard characteristic model diagram obtained by retrieval with an actual three-dimensional model diagram to obtain a deviation model diagram;
calculating the volume value of the deviation model diagram, and judging whether the volume value is in a preset volume value range or not; if not, the faulty component is directly scrapped, and if so, the faulty component is repaired.
After the target air compressor is maintained, if a part of parts need to be replaced, such as a gear, a bearing, a transmission shaft and the like, the image information of the fault components can be acquired at the moment, so that an actual three-dimensional model diagram is constructed by utilizing an image reconstruction technology; if the volume value is not in the preset volume value range, the difficulty of polishing and repairing the fault components is high, the cost is high, and the scrapping treatment is directly carried out at the moment; if the volume value is in the preset volume value range, the fault components can be polished and repaired. The method can reasonably discard the fault component device and improve the resource utilization efficiency.
As shown in fig. 4, the second aspect of the present invention discloses an intelligent air compressor intake status monitoring system, which includes a memory 20 and a processor 30, wherein the memory 20 stores an intelligent air compressor intake status monitoring method program, and when the intelligent air compressor intake status monitoring method program is executed by the processor 30, the following steps are implemented:
acquiring the actual running state of the target air compressor, and determining the preset air inlet parameter of the target air compressor according to the actual running state of the target air compressor;
acquiring actual air inlet parameters of a target air compressor within preset time, and processing the actual air inlet parameters of the target air compressor according to a fuzzy C-means clustering algorithm to obtain a plurality of membership matrixes;
introducing an LOF algorithm, detecting and correcting each membership matrix through the LOF algorithm, and obtaining a plurality of corrected membership matrices; obtaining actual air inlet parameters of different categories according to corresponding elements in each corrected membership matrix;
comparing the same kind of actual air inlet parameters with preset air inlet parameters to judge whether the parameter states of the actual air inlet parameters in the target air compressor are normal or not; and judging real-time fault type information of the target air compressor according to the parameter state of the actual air inlet parameter in the target air compressor, and generating a corresponding fault maintenance scheme.
Further, in a preferred embodiment of the present invention, an actual operation state of the target air compressor is obtained, and a preset air intake parameter of the target air compressor is determined according to the actual operation state of the target air compressor, which specifically includes:
retrieving and obtaining standard air inlet parameters corresponding to various preset running states of the target air compressor based on a big data network;
establishing a prediction model based on a deep learning network, and importing standard air inlet parameters corresponding to various preset running states of a target air compressor into the prediction model for training to obtain a trained prediction model;
acquiring service life information, maintenance condition information and mechanical wear degree information of a target air compressor, and determining the actual running state of the target air compressor according to the service life information, the maintenance condition information and the mechanical wear degree information;
the actual running state of the target air compressor is led into the trained prediction model, and the preset air inlet parameters of the target air compressor under the current actual running state condition are predicted;
the air inlet parameters comprise air inlet temperature, air inlet pressure, air inlet flow rate, air inlet volume, air inlet humidity and air inlet dust concentration.
Further, in a preferred embodiment of the present invention, the actual intake parameters of the same category are compared with the preset intake parameters to determine whether the parameter status of the actual intake parameters in the target air compressor is normal, specifically:
comparing the actual air inlet parameters of the same category with preset air inlet parameters to obtain parameter deviation values; comparing the parameter deviation value with a preset deviation value;
if the parameter deviation value is larger than a preset deviation value, marking the corresponding actual air inlet parameter in the target air compressor as an abnormal air inlet parameter;
and if the parameter deviation value is not greater than the preset deviation value, marking the corresponding actual air inlet parameter in the target air compressor as a normal air inlet parameter.
Further, in a preferred embodiment of the present invention, the real-time fault type information of the target air compressor is determined according to the parameter status of the actual air intake parameter in the target air compressor, and a corresponding fault maintenance scheme is generated, which specifically includes:
searching based on a big data network to obtain corresponding fault characteristic data when various fault types occur to a target air compressor, defining random variables according to the fault characteristic data, and determining target nodes of a Bayesian network structure according to the random variables;
Constructing a Bayesian network based on the target node, performing iterative optimization on the Bayesian network based on a maximum likelihood estimation method until the iterative times reach a preset time, and outputting a conditional probability table; introducing a Markov chain, calculating the state transition probability of the target air compressor when various fault types occur under various preset abnormal air inlet parameter combination conditions through the Markov chain and combining the conditional probability table;
constructing a probability state transition matrix according to the state transition probability, and importing the state transition matrix into a Bayesian network for structure learning to obtain a Bayesian network model;
the parameter state is an abnormal air inlet parameter, the abnormal air inlet parameter is led into the Bayesian network to be deduced and predicted, the fault probability of various fault types of the target air compressor is obtained, the fault types with the fault probability larger than the preset fault probability are extracted and output, and the real-time fault type information of the target air compressor is obtained;
and generating a search tag according to the real-time fault type information of the target air compressor, searching a big data network based on the search tag to obtain a corresponding fault maintenance scheme, and pushing the fault maintenance scheme to a preset platform for display.
In the several embodiments provided in this application, it should be understood that the disclosed apparatus and method may be implemented in other ways. The above described device embodiments are only illustrative, e.g. the division of the units is only one logical function division, and there may be other divisions in practice, such as: multiple units or components may be combined or may be integrated into another system, or some features may be omitted, or not performed. In addition, the various components shown or discussed may be coupled or directly coupled or communicatively coupled to each other via some interface, whether indirectly coupled or communicatively coupled to devices or units, whether electrically, mechanically, or otherwise.
The units described above as separate components may or may not be physically separate, and components shown as units may or may not be physical units; can be located in one place or distributed to a plurality of network units; some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.
In addition, each functional unit in each embodiment of the present invention may be integrated in one processing unit, or each unit may be separately used as one unit, or two or more units may be integrated in one unit; the integrated units may be implemented in hardware or in hardware plus software functional units.
Those of ordinary skill in the art will appreciate that: all or part of the steps for implementing the above method embodiments may be implemented by hardware related to program instructions, and the foregoing program may be stored in a computer readable storage medium, where the program, when executed, performs steps including the above method embodiments; and the aforementioned storage medium includes: a mobile storage device, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk or an optical disk, or the like, which can store program codes.
Alternatively, the above-described integrated units of the present invention may be stored in a computer-readable storage medium if implemented in the form of software functional modules and sold or used as separate products. Based on such understanding, the technical solutions of the embodiments of the present invention may be embodied in essence or a part contributing to the prior art in the form of a software product stored in a storage medium, including several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the methods of the embodiments of the present invention. And the aforementioned storage medium includes: a removable storage device, ROM, RAM, magnetic or optical disk, or other medium capable of storing program code.
The foregoing is merely illustrative embodiments of the present invention, but the scope of the present invention is not limited thereto, and any person skilled in the art can easily think about variations or substitutions within the technical scope of the present invention, and the invention should be covered. Therefore, the protection scope of the invention is subject to the protection scope of the claims.
Claims (10)
1. The intelligent monitoring method for the air inlet state of the air compressor is characterized by comprising the following steps of:
acquiring the actual running state of the target air compressor, and determining the preset air inlet parameter of the target air compressor according to the actual running state of the target air compressor;
acquiring actual air inlet parameters of a target air compressor within preset time, and processing the actual air inlet parameters of the target air compressor according to a fuzzy C-means clustering algorithm to obtain a plurality of membership matrixes;
introducing an LOF algorithm, detecting and correcting each membership matrix through the LOF algorithm, and obtaining a plurality of corrected membership matrices; obtaining actual air inlet parameters of different categories according to corresponding elements in each corrected membership matrix;
comparing the same kind of actual air inlet parameters with preset air inlet parameters to judge whether the parameter states of the actual air inlet parameters in the target air compressor are normal or not; and judging real-time fault type information of the target air compressor according to the parameter state of the actual air inlet parameter in the target air compressor, and generating a corresponding fault maintenance scheme.
2. The intelligent monitoring method for air compressor air intake state according to claim 1, wherein the actual operation state of the target air compressor is obtained, and the preset air intake parameter of the target air compressor is determined according to the actual operation state of the target air compressor, specifically:
retrieving and obtaining standard air inlet parameters corresponding to various preset running states of the target air compressor based on a big data network;
establishing a prediction model based on a deep learning network, and importing standard air inlet parameters corresponding to various preset running states of a target air compressor into the prediction model for training to obtain a trained prediction model;
acquiring service life information, maintenance condition information and mechanical wear degree information of a target air compressor, and determining the actual running state of the target air compressor according to the service life information, the maintenance condition information and the mechanical wear degree information;
the actual running state of the target air compressor is led into the trained prediction model, and the preset air inlet parameters of the target air compressor under the current actual running state condition are predicted;
The air inlet parameters comprise air inlet temperature, air inlet pressure, air inlet flow rate, air inlet volume, air inlet humidity and air inlet dust concentration.
3. The intelligent monitoring method of the air compressor air intake state according to claim 1, wherein the actual air intake parameters of the target air compressor are processed according to a fuzzy C-means clustering algorithm to obtain a plurality of membership matrices, specifically:
the method comprises the steps of presetting the clustering quantity and the ambiguity parameters, and randomly initializing the weight of each actual air inlet parameter according to the preset clustering quantity and the ambiguity parameters; wherein the weight of each actual air intake parameter is a positive number, and the total weight is 1;
multiplying the value of each actual air intake parameter with the corresponding weight to obtain the weighted value of each actual air intake parameter, and dividing the weighted value of each actual air intake parameter by the preset weight to obtain the weighted average value of each actual air intake parameter;
presetting a plurality of threshold intervals, and distributing each actual air inlet parameter into a matched threshold interval according to the weighted average value of each actual air inlet parameter;
after distribution is completed, combining weighted average values of actual air inlet parameters in all threshold intervals to form a vector, and determining fuzzy clustering centroids of the actual air inlet parameters in all threshold intervals according to the formed vector;
Calculating Euclidean distance between the actual air inlet parameter in each threshold interval and the fuzzy clustering centroid thereof, and calculating membership degree between the actual air inlet parameter in each threshold interval and the fuzzy clustering centroid thereof according to the Euclidean distance;
judging whether the membership degree between the actual air inlet parameter and the fuzzy clustering centroid in each threshold interval is smaller than the preset membership degree one by one, and if so, eliminating the corresponding actual air inlet parameter in the corresponding threshold interval;
updating the residual actual air inlet parameters in each threshold interval, and constructing a plurality of membership matrixes according to membership between the residual actual air inlet parameters in each threshold interval and the fuzzy clustering centroid of the residual actual air inlet parameters.
4. The intelligent monitoring method of the air inlet state of the air compressor according to claim 1, wherein an LOF algorithm is introduced, and each membership matrix is detected and corrected by the LOF algorithm to obtain a plurality of corrected membership matrices, specifically:
for each membership degree matrix, calculating Manhattan distances between each element in the matrix and elements in a preset field, summing the Manhattan distances between each element and the elements in the preset field, and taking an average value to obtain local reachable densities of each element;
Calculating a local outlier factor of each element according to the local reachable density, wherein the local outlier factor represents the outlier degree of the element relative to other elements in the neighborhood of the element; comparing the local outlier factor of each element of the membership matrix with a preset threshold;
taking the element with the local outlier greater than the preset threshold value as a clustering abnormal element, importing the clustering abnormal element into other membership matrixes, and judging whether the local outlier of the clustering abnormal element in the other membership matrixes is greater than the preset threshold value;
if the local outlier factors of the clustering abnormal elements in the rest membership matrixes are larger than a preset threshold value, the clustering abnormal elements are thoroughly removed; if the local outlier factor of the clustering abnormal element in the rest membership matrix is not greater than a preset threshold value, the clustering abnormal element is distributed to the membership matrix with the minimum local outlier factor;
and updating the position condition of each element in each membership matrix to obtain a corrected membership matrix.
5. The intelligent monitoring method for air intake state of air compressor according to claim 1, wherein the comparison between the actual intake parameters of the same category and the preset intake parameters is performed to determine whether the parameter states of the actual intake parameters in the target air compressor are normal, specifically:
Comparing the actual air inlet parameters of the same category with preset air inlet parameters to obtain parameter deviation values; comparing the parameter deviation value with a preset deviation value;
if the parameter deviation value is larger than a preset deviation value, marking the corresponding actual air inlet parameter in the target air compressor as an abnormal air inlet parameter;
and if the parameter deviation value is not greater than the preset deviation value, marking the corresponding actual air inlet parameter in the target air compressor as a normal air inlet parameter.
6. The intelligent monitoring method for air compressor air intake state according to claim 1, wherein the real-time fault type information of the target air compressor is determined according to the parameter state of the actual air intake parameter in the target air compressor, and a corresponding fault maintenance scheme is generated, specifically:
searching based on a big data network to obtain corresponding fault characteristic data when various fault types occur to a target air compressor, defining random variables according to the fault characteristic data, and determining target nodes of a Bayesian network structure according to the random variables;
constructing a Bayesian network based on the target node, performing iterative optimization on the Bayesian network based on a maximum likelihood estimation method until the iterative times reach a preset time, and outputting a conditional probability table; introducing a Markov chain, calculating the state transition probability of the target air compressor when various fault types occur under various preset abnormal air inlet parameter combination conditions through the Markov chain and combining the conditional probability table;
Constructing a probability state transition matrix according to the state transition probability, and importing the state transition matrix into a Bayesian network for structure learning to obtain a Bayesian network model;
the parameter state is an abnormal air inlet parameter, the abnormal air inlet parameter is led into the Bayesian network to be deduced and predicted, the fault probability of various fault types of the target air compressor is obtained, the fault types with the fault probability larger than the preset fault probability are extracted and output, and the real-time fault type information of the target air compressor is obtained;
and generating a search tag according to the real-time fault type information of the target air compressor, searching a big data network based on the search tag to obtain a corresponding fault maintenance scheme, and pushing the fault maintenance scheme to a preset platform for display.
7. The intelligent monitoring system for the air inlet state of the air compressor is characterized by comprising a memory and a processor, wherein an intelligent monitoring method program for the air inlet state of the air compressor is stored in the memory, and when the intelligent monitoring method program for the air inlet state of the air compressor is executed by the processor, the following steps are realized:
Acquiring the actual running state of the target air compressor, and determining the preset air inlet parameter of the target air compressor according to the actual running state of the target air compressor;
acquiring actual air inlet parameters of a target air compressor within preset time, and processing the actual air inlet parameters of the target air compressor according to a fuzzy C-means clustering algorithm to obtain a plurality of membership matrixes;
introducing an LOF algorithm, detecting and correcting each membership matrix through the LOF algorithm, and obtaining a plurality of corrected membership matrices; obtaining actual air inlet parameters of different categories according to corresponding elements in each corrected membership matrix;
comparing the same kind of actual air inlet parameters with preset air inlet parameters to judge whether the parameter states of the actual air inlet parameters in the target air compressor are normal or not; and judging real-time fault type information of the target air compressor according to the parameter state of the actual air inlet parameter in the target air compressor, and generating a corresponding fault maintenance scheme.
8. The intelligent monitoring system for air compressor intake status according to claim 7, wherein the actual operation status of the target air compressor is obtained, and the preset intake parameter of the target air compressor is determined according to the actual operation status of the target air compressor, specifically:
Retrieving and obtaining standard air inlet parameters corresponding to various preset running states of the target air compressor based on a big data network;
establishing a prediction model based on a deep learning network, and importing standard air inlet parameters corresponding to various preset running states of a target air compressor into the prediction model for training to obtain a trained prediction model;
acquiring service life information, maintenance condition information and mechanical wear degree information of a target air compressor, and determining the actual running state of the target air compressor according to the service life information, the maintenance condition information and the mechanical wear degree information;
the actual running state of the target air compressor is led into the trained prediction model, and the preset air inlet parameters of the target air compressor under the current actual running state condition are predicted;
the air inlet parameters comprise air inlet temperature, air inlet pressure, air inlet flow rate, air inlet volume, air inlet humidity and air inlet dust concentration.
9. The intelligent monitoring system for air compressor intake status according to claim 7, wherein the comparison between the actual intake parameters of the same category and the preset intake parameters is performed to determine whether the parameter status of the actual intake parameters in the target air compressor is normal, specifically:
Comparing the actual air inlet parameters of the same category with preset air inlet parameters to obtain parameter deviation values; comparing the parameter deviation value with a preset deviation value;
if the parameter deviation value is larger than a preset deviation value, marking the corresponding actual air inlet parameter in the target air compressor as an abnormal air inlet parameter;
and if the parameter deviation value is not greater than the preset deviation value, marking the corresponding actual air inlet parameter in the target air compressor as a normal air inlet parameter.
10. The intelligent monitoring system for air compressor intake status according to claim 7, wherein the real-time fault type information of the target air compressor is determined according to the parameter status of the actual intake parameter in the target air compressor, and a corresponding fault maintenance scheme is generated, specifically:
searching based on a big data network to obtain corresponding fault characteristic data when various fault types occur to a target air compressor, defining random variables according to the fault characteristic data, and determining target nodes of a Bayesian network structure according to the random variables;
constructing a Bayesian network based on the target node, performing iterative optimization on the Bayesian network based on a maximum likelihood estimation method until the iterative times reach a preset time, and outputting a conditional probability table; introducing a Markov chain, calculating the state transition probability of the target air compressor when various fault types occur under various preset abnormal air inlet parameter combination conditions through the Markov chain and combining the conditional probability table;
Constructing a probability state transition matrix according to the state transition probability, and importing the state transition matrix into a Bayesian network for structure learning to obtain a Bayesian network model;
the parameter state is an abnormal air inlet parameter, the abnormal air inlet parameter is led into the Bayesian network to be deduced and predicted, the fault probability of various fault types of the target air compressor is obtained, the fault types with the fault probability larger than the preset fault probability are extracted and output, and the real-time fault type information of the target air compressor is obtained;
and generating a search tag according to the real-time fault type information of the target air compressor, searching a big data network based on the search tag to obtain a corresponding fault maintenance scheme, and pushing the fault maintenance scheme to a preset platform for display.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311671501.8A CN117370824B (en) | 2023-12-07 | 2023-12-07 | Intelligent monitoring method and system for air inlet state of air compressor |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311671501.8A CN117370824B (en) | 2023-12-07 | 2023-12-07 | Intelligent monitoring method and system for air inlet state of air compressor |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117370824A true CN117370824A (en) | 2024-01-09 |
| CN117370824B CN117370824B (en) | 2024-03-22 |
Family
ID=89393290
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311671501.8A Active CN117370824B (en) | 2023-12-07 | 2023-12-07 | Intelligent monitoring method and system for air inlet state of air compressor |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117370824B (en) |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080015246A (en) * | 2006-08-14 | 2008-02-19 | 현대자동차주식회사 | Purge system for engine |
| CN101446283A (en) * | 2008-12-17 | 2009-06-03 | 西安交通大学 | Evolutionary computation-based method for diagnosing compressor fault |
| JP2009137527A (en) * | 2007-12-10 | 2009-06-25 | Sanden Corp | Compressor torque estimation device |
| CN109779894A (en) * | 2018-12-29 | 2019-05-21 | 西安交通大学 | A kind of reciprocating compressor fault diagnosis system and method based on neural network algorithm |
| CN113435453A (en) * | 2021-06-11 | 2021-09-24 | 西安交通大学 | Reciprocating compressor fault diagnosis method and system based on convolutional neural network |
| US20210355932A1 (en) * | 2019-02-12 | 2021-11-18 | Hang Zhou Zeta Technology Co., Lts | Analysis method of absolute energy efficiency and relative energy efficiency of compressed air system |
| CN116703485A (en) * | 2023-08-04 | 2023-09-05 | 山东创亿智慧信息科技发展有限责任公司 | Advertisement accurate marketing method and system based on big data |
-
2023
- 2023-12-07 CN CN202311671501.8A patent/CN117370824B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20080015246A (en) * | 2006-08-14 | 2008-02-19 | 현대자동차주식회사 | Purge system for engine |
| JP2009137527A (en) * | 2007-12-10 | 2009-06-25 | Sanden Corp | Compressor torque estimation device |
| CN101446283A (en) * | 2008-12-17 | 2009-06-03 | 西安交通大学 | Evolutionary computation-based method for diagnosing compressor fault |
| CN109779894A (en) * | 2018-12-29 | 2019-05-21 | 西安交通大学 | A kind of reciprocating compressor fault diagnosis system and method based on neural network algorithm |
| US20210355932A1 (en) * | 2019-02-12 | 2021-11-18 | Hang Zhou Zeta Technology Co., Lts | Analysis method of absolute energy efficiency and relative energy efficiency of compressed air system |
| CN113435453A (en) * | 2021-06-11 | 2021-09-24 | 西安交通大学 | Reciprocating compressor fault diagnosis method and system based on convolutional neural network |
| CN116703485A (en) * | 2023-08-04 | 2023-09-05 | 山东创亿智慧信息科技发展有限责任公司 | Advertisement accurate marketing method and system based on big data |
Non-Patent Citations (1)
| Title |
|---|
| 杨清宇;孙凤伟;张曌，张迪，庄健: "利用测地线距离的改进谱聚类算法", 西安交通大学学报, no. 08, 10 August 2012 (2012-08-10), pages 1 - 7 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117370824B (en) | 2024-03-22 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN111047082B (en) | Early warning method and device of equipment, storage medium and electronic device | |
| US20220283576A1 (en) | Automatic diagnosis method, system and storage medium for equipment | |
| CN115578015B (en) | Sewage treatment whole process supervision method, system and storage medium based on Internet of things | |
| US11669083B2 (en) | System and method for proactive repair of sub optimal operation of a machine | |
| CN113570138B (en) | Method and device for predicting residual service life of equipment of time convolution network | |
| CN113454552A (en) | Sensor metering data integration | |
| CN115358281B (en) | Machine learning-based cold and hot all-in-one machine monitoring control method and system | |
| CN111190349A (en) | Method, system and medium for monitoring state and diagnosing fault of ship engine room equipment | |
| Calabrese et al. | An event based machine learning framework for predictive maintenance in industry 4.0 | |
| CN116862081A (en) | Operation and maintenance method and system for pollution treatment equipment | |
| CN116882790B (en) | Carbon emission equipment management method and system for mine ecological restoration area | |
| CN116956639B (en) | High-heat-dissipation storage battery assessment method and system based on data analysis | |
| CN114138601A (en) | Service alarm method, device, equipment and storage medium | |
| CN114202179A (en) | Target enterprise identification method and device | |
| CN117370824B (en) | Intelligent monitoring method and system for air inlet state of air compressor | |
| CN115964470B (en) | Method and system for predicting service life of motorcycle accessories | |
| CN116737510A (en) | Data analysis-based intelligent keyboard monitoring method and system | |
| CN110489852A (en) | Improve the method and device of the wind power system quality of data | |
| CN115684792A (en) | Electrical automation equipment detecting system based on artificial intelligence | |
| CN115424106A (en) | Trolley detection method and device, electronic equipment and storage medium | |
| CN116681266B (en) | Production scheduling method and system of mirror surface electric discharge machine | |
| CN117494072B (en) | Method and system for monitoring operation state of battery-changing cabinet based on data fusion | |
| CN114692743A (en) | Failure prediction method, failure prediction device and storage medium | |
| CN115587717A (en) | Data quality detection method, device, storage medium and equipment | |
| CN117478549A (en) | Fault perception method and device for program analysis cloud support platform |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |