Compressor fault judgment method and device

Abstract

The invention discloses a compressor fault judgment method and device, and the method comprises the steps: obtaining the operation parameters of a compressor at a plurality of time points in a preset time period, and obtaining a pressure-time corresponding relation, obtaining a division relationship of a plurality of work cycles in a predetermined time period; according to the pressure-time corresponding relation and the division relation of the multiple work periods, the actual indicator diagram of each work period of the compressor in the preset time period is determined; the actual indicator diagram of each work period is compared with a reference indicator diagram obtained in advance, the state of the compressor is judged according to the comparison result, the state comprises a normal type or a fault type, and the actual indicator diagram of the compressor can be obtained through compressor data obtained in real time; and the real-time working condition of compression is judged by comparing the real-time indicator diagram with the reference indicator diagram, so that the production loss caused by the bumping of the compressor is avoided, the operation cycle of the compressor is prolonged, and the maintenance cost of the compressor is reduced.

Description

Technical Field

[0001] This document relates to the field of compressors, and particularly to a method and device for judging compressor faults. Background Art

[0002] Reciprocating compressors are widely used in industrial production. Their main function is to provide gas compression services for various equipment and machines. For example, in the petrochemical industry, reciprocating compressors are usually used for the compression of gases such as natural gas, liquefied gas, methane, and ethylene, facilitating the storage and transportation of these gases, or compressing these gases to the high pressure required for reactions. Thus, it can be seen that the equipment performance of reciprocating compressors is related to the production quality and production efficiency of all links in industrial production. Once a fault occurs, it has a great impact on the safety and reliability of production. Therefore, it is necessary to judge in real time whether the reciprocating compressor is faulty and its fault type during the production process.

[0003] In the existing related technologies, most rely on relevant staff to monitor faults. However, this method depends on work experience and the dedication and concentration of the staff, with poor reliability and high costs; or early warnings are made by monitoring certain parameters. For example, when parameters such as temperature and pressure exceed the normal working threshold range, an alarm is given. However, this early warning method cannot determine the fault type of the compressor, and further fault troubleshooting is still required by maintenance personnel. Therefore, there is an urgent need for a compressor fault judgment method with high reliability, low cost, and that can provide reference and guidance for operation and maintenance personnel. Summary of the Invention

[0004] This application provides a method and device for judging compressor faults, which can detect in real time whether the compressor is faulty and its fault type, with high reliability, rapid response, low cost, facilitating fault location under the condition of accurately analyzing the fault type, helping operation and maintenance personnel to timely discover the cause of the fault, saving fault repair and recovery time, and thus ensuring the stable operation of the compressor.

[0005] On the one hand, an embodiment of this application provides a method for judging compressor faults, including:

[0006] Obtaining the operating parameters of the compressor at multiple time points within a predetermined time period, where the operating parameters include working pressure, to obtain a pressure-time correspondence relationship; obtaining the division relationship of multiple working cycles within the predetermined time period;

[0007] Determining the actual indicator diagram of each working cycle of the compressor within the predetermined time period according to the pressure-time correspondence relationship and the division relationship of the multiple working cycles;

[0008] Compare the actual indicator diagram of each working cycle with the pre-obtained reference indicator diagram respectively, and judge the state of the compressor according to the comparison result, where the state includes normal or fault types;

[0009] Among them, the reference indicator diagram includes the standard indicator diagram of the compressor in the normal working state and the fault indicator diagrams respectively corresponding to different fault types in the working state.

[0010] Optionally, the obtaining of the division relationship of multiple working cycles within the predetermined time period includes:

[0011] Obtain the moment when the zero point mark on the compressor flywheel returns to the preset mark point each time within the predetermined time period, where the preset mark point is located outside the compressor flywheel and is used to combine with the zero point mark to determine the working cycle of the compressor; among them, one working cycle of the compressor is the time period between any two adjacent moments when returning to the preset mark point.

[0012] Optionally, the determining of the actual indicator diagram of the compressor within the predetermined time period according to the pressure-time correspondence and the division relationship of the multiple working cycles includes:

[0013] Group the pressure-time correspondence, and respectively correspond the multiple time points to each working cycle to obtain the time groups corresponding to each working cycle;

[0014] Perform the following operations on each time point respectively: obtain the time difference between the time point and the starting moment of the working cycle corresponding to it; determine the cylinder volume corresponding to the time point according to the time difference, the total volume of the compressor cylinder and the clearance volume; obtain the working pressure corresponding to the time point according to the pressure-time correspondence;

[0015] Perform the following operations on each working cycle respectively: obtain the working pressure and the cylinder volume corresponding to each time point in the time group corresponding to the working cycle; use the cylinder volume as the abscissa and the working pressure as the ordinate to determine the actual indicator diagram of the working cycle.

[0016] Optionally, before comparing the actual indicator diagram of each working cycle with the pre-obtained reference indicator diagram respectively, it further includes:

[0017] Obtain a data set containing historical indicator diagrams, where the data set includes a labeled data set and an unlabeled data set, and the labeled data set includes a standard data set and a fault data set; the label of the indicator diagram in the standard data set is the normal working state; the label of the indicator diagram in the fault data set is the corresponding fault type;

[0018] Cluster the historical indicator diagram dataset based on a semi-supervised clustering algorithm to obtain the category clusters corresponding to the compressor in the normal working state and different fault type states respectively;

[0019] Determine that the target historical indicator diagrams corresponding to the center points of each category cluster are the standard indicator diagram and the fault indicator diagram respectively.

[0020] Optionally, comparing the actual indicator diagram of each working cycle with a pre-obtained reference indicator diagram respectively, and judging the state of the compressor according to the comparison result, where the state includes normal or fault type, including performing the following operations on the actual indicator diagram of each working cycle respectively:

[0021] Determine the first similarity between the actual indicator diagram of this working cycle and the standard indicator diagram. When the first similarity is greater than or equal to the first preset threshold, determine that the state is normal;

[0022] When the first similarity is less than the first preset threshold, determine the second similarity between the actual indicator diagram of this working cycle and the fault indicator diagrams of each different fault type, and select the maximum value as the maximum second similarity, and determine the fault type of the compressor in this working cycle as the fault type corresponding to the fault indicator diagram used to obtain the maximum second similarity.

[0023] Optionally, comparing the actual indicator diagram of each working cycle with a pre-obtained reference indicator diagram respectively includes performing one or more of the following operations on the actual indicator diagram of each working cycle respectively:

[0024] Comparing the overall area of the figure enclosed by the closed line segments in the indicator diagram;

[0025] Comparing the line types corresponding to each stroke;

[0026] Comparing the lengths corresponding to each stroke;

[0027] Comparing the starting and ending point positions corresponding to each stroke.

[0028] Optionally, comparing the actual indicator diagram of each working cycle with a pre-obtained reference indicator diagram respectively, and judging the state of the compressor according to the comparison result, where the state includes normal or fault type, including performing the following operations on the actual indicator diagram of each working cycle respectively:

[0029] Use the actual indicator diagram of this working cycle as an input item of a preset classification model, output the probabilities of the actual indicator diagram belonging to various classification states, and determine the maximum probability; the preset classification model is used to output the classification state to which the input actual indicator diagram belongs, and the classification states include a normal state and different fault type states; the preset classification model is trained based on historical indicator diagrams marked with corresponding classification states.

[0030] Determine the classification state to which the actual indicator diagram of this working cycle belongs according to the maximum probability, and judge the state of the compressor within this working cycle according to this classification state, where the state includes normal or a fault type.

[0031] Optionally, the operating parameters further include working temperature, working flow rate, and working frequency. After comparing the actual indicator diagram of each working cycle with a pre-obtained reference indicator diagram respectively and judging the state of the compressor according to the comparison result, where the state includes normal or a fault type, the following operations are respectively performed on the judgment results of each working cycle:

[0032] When the judgment result of this working cycle is one of the different fault types, obtain the fault parameter state corresponding to this fault in a preset relationship, and the fault parameter state includes at least one of a fault temperature state, a fault flow rate state, and a fault frequency state;

[0033] Determine the type of operating parameter to be verified in this working cycle according to the type of the obtained fault parameter state, and the operating parameter to be verified includes at least one of the working temperature, the working flow rate, and the working frequency;

[0034] Verify the operating parameter to be verified in this working cycle according to the fault parameter state, and output a warning message when the verification is passed;

[0035] Among them, the preset relationship includes at least one of the fault temperature state, the fault flow rate state, and the fault frequency state respectively corresponding to the different fault types. The fault temperature state includes rising, falling, or fluctuating. The fault flow rate state includes rising, falling, or fluctuating. The fault frequency state includes rising, falling, or fluctuating;

[0036] Among them, passing the verification means that the state of the operating parameter to be verified is the same as the fault parameter state.

[0037] Optionally, the different fault types include at least one of the following:

[0038] Suction valve leakage fault, gas valve flow area too small fault, valve disc mass too heavy fault, spring too hard fault, spring too soft fault, valve disc flutter fault, lift height too high fault, and air flow pressure pulsation fault.

[0039] On the other hand, the embodiment of the present application also provides a compressor fault judgment device, including a processor and a memory storing a computer program, characterized in that when the computer program is executed by the processor, the compressor fault judgment method as described in the above embodiment can be realized.

[0040] Compared with the related technology, the compressor fault judgment method and device of the embodiment of the present application can detect whether the compressor fails and the type of the fault in real time, with high reliability, rapid response, low cost, facilitating fault location under the condition of accurately analyzing the fault type, helping the operation and maintenance personnel to discover the cause of the fault in time, saving the fault repair and recovery time, and thus ensuring the stable operation of the compressor.

[0041] Other features and advantages of the present application will be described in the following specification, and part of them will be obvious from the specification, or will be understood by implementing the present application. Other advantages of the present application can be realized and obtained through the solutions described in the specification and the drawings. Description of the Drawings

[0042] The drawings are used to provide an understanding of the technical solutions of the present application, and constitute a part of the specification. They are used together with the embodiments of the present application to explain the technical solutions of the present application, and do not constitute a limitation to the technical solutions of the present application.

[0043] Figure 1 is the flowchart of the compressor fault judgment method of the embodiment of the present application;

[0044] Figure 2 is the schematic diagram of the division relationship of multiple working cycles of the embodiment of the present application;

[0045] Figure 3 is the standard indicator diagram of the compressor in the normal working state of the embodiment of the present application;

[0046] Figure 4a is the fault indicator diagram corresponding to the suction valve leakage fault of the embodiment of the present application;

[0047] Figure 4b is the fault indicator diagram corresponding to the gas valve flow area too small fault of the embodiment of the present application;

[0048] Figure 4c is the fault indicator diagram corresponding to the valve disc mass too heavy fault of the embodiment of the present application;

[0049] Figure 4dIt is the fault indicator diagram corresponding to the spring being too hard in the embodiment of the present application;

[0050] Figure 4e It is the fault indicator diagram corresponding to the spring being too soft in the embodiment of the present application;

[0051] Figure 4f It is the fault indicator diagram corresponding to the valve plate flutter fault in the embodiment of the present application;

[0052] Figure 4g It is the fault indicator diagram corresponding to the lift height being too high in the embodiment of the present application;

[0053] Figure 4h It is the fault indicator diagram corresponding to the air flow pressure pulsation fault in the embodiment of the present application;

[0054] Figure 5 It is the comparison diagram between the fault indicator diagram of the suction valve leakage fault in the embodiment of the present application and the standard indicator diagram under the normal working state;

[0055] Figure 6 It is the schematic diagram of the compressor fault judgment device in the embodiment of the present application;

[0056] Among them, Figure 3 、 Figures 4a - 4h 、 Figure 5 The reference numerals are as follows:

[0057] P 1 ——The theoretical suction pressure of the compressor, unit Pa; P 2 ——The theoretical discharge pressure of the compressor, unit Pa;

[0058] V 1 ——The clearance volume of the compressor, unit L; V 2 ——The total cylinder volume of the compressor, unit L;

[0059] The ab section represents the expansion stroke; the bc section represents the suction stroke; the cd section represents the compression stroke; the da section represents the discharge stroke;

[0060] Among them, Figure 5 The solid line in it is the fault indicator diagram of the suction valve leakage fault, and the dotted line is the standard indicator diagram. Specific implementation mode

[0061] This application describes multiple embodiments, but the description is exemplary rather than restrictive, and it will be apparent to those of ordinary skill in the art that there can be more embodiments and implementation solutions within the scope encompassed by the embodiments described in this application. Although many possible feature combinations are shown in the drawings and discussed in the detailed description, many other combinations of the disclosed features are also possible. Unless specifically restricted, any feature or element of any embodiment can be used in combination with any other feature or element in any other embodiment, or can replace any other feature or element in any other embodiment.

[0062] This application includes and contemplates combinations with features and elements known to those of ordinary skill in the art. The embodiments, features, and elements disclosed in this application can also be combined with any conventional features or elements to form unique inventive solutions defined by the claims. Any feature or element of any embodiment can also be combined with features or elements from other inventive solutions to form another unique inventive solution defined by the claims. Therefore, it should be understood that any feature shown and / or discussed in this application can be implemented alone or in any suitable combination. Thus, the embodiments are not subject to other limitations except those made in accordance with the appended claims and their equivalents. In addition, various modifications and changes can be made within the scope of the appended claims.

[0063] Furthermore, in describing representative embodiments, the specification may have presented the method and / or process as a particular sequence of steps. However, to the extent that the method or process does not depend on the particular order of the steps described herein, the method or process should not be limited to the particular order of steps described. As will be understood by those of ordinary skill in the art, other step sequences are possible. Therefore, the particular order of steps set forth in the specification should not be construed as a limitation on the claims. In addition, the claims directed to the method and / or process should not be limited to performing their steps in the order written, as those skilled in the art can readily understand that these orders can vary and still remain within the spirit and scope of the embodiments of this application.

[0064] A reciprocating compressor is a type of compressor that makes the volume of the cylinder change periodically through the reciprocating motion of the piston in the cylinder, and realizes the pressurization and transportation of gas. It belongs to the positive displacement compressor. When the piston is at the top dead center, the volume of the cylinder is the smallest, corresponding to the clearance volume. When the piston is at the bottom dead center, the volume of the cylinder is the largest, corresponding to the total volume of the cylinder. The volume between the top dead center and the bottom dead center is the stroke volume of the piston. A complete working cycle is for the piston to move from the top dead center to the bottom dead center and then return to the top dead center. A working cycle can be divided into four processes: expansion, suction, compression, and exhaust, which are called four strokes. During the expansion process, the volume of the cylinder of the compressor increases, and the pressure in the cylinder decreases. The residual gas originally remaining in the cylinder expands continuously. When the pressure in the cylinder drops below the gas pressure in the intake pipe, the gas in the intake pipe pushes open the suction valve and enters the cylinder, that is, enters the suction process. During this process, the volume of the cylinder continues to increase until the piston reaches the bottom dead center. When the piston moves in the reverse direction, the volume of the cylinder gradually decreases, and the compression process begins. The pressure in the cylinder increases. When the pressure in the cylinder increases above the gas pressure in the exhaust pipe, the exhaust valve is pushed open by the gas in the cylinder and exhausts outward, that is, the exhaust process, until the piston returns to the top dead center again. The pressure-cylinder volume relationship diagram of the compressor in a complete working cycle is called the indicator diagram, which can reflect the working state of the compressor.

[0065] The embodiment of the present application provides a method for judging the fault of a compressor, as Figure 1 shown, including steps S100 - S300:

[0066] S100: Obtain the operating parameters of the compressor at multiple time points within a predetermined time period. The operating parameters include the working pressure to obtain the pressure-time correspondence; obtain the division relationship of multiple working cycles within the predetermined time period;

[0067] S200: Determine the actual indicator diagram of each working cycle of the compressor within the predetermined time period according to the pressure-time correspondence and the division relationship of the multiple working cycles;

[0068] S300: Compare the actual indicator diagram of each working cycle with the pre-obtained reference indicator diagram respectively, and judge the state of the compressor according to the comparison result. The state includes normal or fault type.

[0069] In this embodiment, the reference indicator diagram includes the standard indicator diagram of the compressor in the normal working state and the fault indicator diagrams corresponding to different fault types in the working state respectively.

[0070] In this embodiment, a pressure sensor is arranged in the cylinder of the compressor, which can collect the working pressure in the cylinder of the compressor in real time.

[0071] In this embodiment, the compressor drives the rotating shaft through the flywheel to push the piston to work. Usually, the rotational speed of the flywheel is several hundred revolutions per minute, or even thousands of revolutions per minute (commonly 300 rpm to 1200 rpm). This means that hundreds or even thousands of actual indicator diagrams can be generated per minute, which can very sensitively and accurately reflect the operating conditions of the compressor. Each rotation of the flywheel represents a working cycle of the compressor. By installing a position switch on the flywheel, the division relationship of multiple working cycles of the compressor can be monitored.

[0072] The compressor fault judgment method of this embodiment can detect in real time whether the compressor is faulty and the type of fault. It has high reliability, rapid response, low cost, is convenient for fault location under the condition of accurately analyzing the fault type, helps the operation and maintenance personnel to find the cause of the fault in time, saves the fault repair and recovery time, and thus ensures the stable operation of the compressor.

[0073] In an exemplary embodiment, obtaining the division relationship of multiple working cycles within the predetermined time period in step S100 may include:

[0074] Obtaining the moment when the zero point mark on the compressor flywheel returns to the preset mark point each time within the predetermined time period, where the preset mark point is located outside the compressor flywheel and is used to combine with the zero point mark to determine the working cycle of the compressor; and a working cycle of the compressor is the time period between any two adjacent moments when returning to the preset mark point.

[0075] In this embodiment, a position switch can be installed at the zero point mark of the flywheel, and the position switch can be selected but not limited to a Hall position switch, a capacitive position switch, an ultrasonic position switch, or a high-frequency oscillation position switch.

[0076] In an implementation manner of this embodiment, the position switch adopts a Hall position switch. When the flywheel rotates and the zero point mark coincides with the preset mark point, the Hall position switch will generate a signal and be detected by an external control circuit. This moment represents the start of a new working cycle of the compressor and also represents the end of the previous working cycle. By obtaining the signals generated by the Hall position switch within the predetermined time period, the division relationship of multiple working cycles within the predetermined time period is obtained.

[0077] In this embodiment, for example, the predetermined time period is T 0 moment to T m moment, the position switch is respectively at T 1 , T 2 , T 3 , T 4 , T 5 , T 6 , T 7 , T8 , T 9 These nine moment output signals, then the division relationship of multiple working cycles is as Figure 2 and shown in Table 1. From time T 0 moment to time T m moment includes eight working cycles.

[0078] Table 1 Division relationship table of multiple working cycles

[0079] Starting moment Ending moment First cycle <![CDATA[T 1 > <![CDATA[T 2 > Second cycle <![CDATA[T 2 > <![CDATA[T 3 > Third cycle <![CDATA[T 3 > <![CDATA[T 4 > Fourth cycle <![CDATA[T 4 > <![CDATA[T 5 > Fifth cycle <![CDATA[T 5 > <![CDATA[T 6 > Sixth cycle <![CDATA[T 6 > <![CDATA[T 7 > Seventh cycle <![CDATA[T 7 > <![CDATA[T 8 > Eighth cycle <![CDATA[T 8 > <![CDATA[T 9 >

[0080] In an exemplary embodiment, step S200 may include steps S210 - S230:

[0081] S210: Group the pressure - time correspondence relationship, and respectively correspond the multiple time points to each of the working cycles to obtain the time groups corresponding to each of the working cycles;

[0082] S220: Perform the following operations on each time point respectively: Obtain the time difference between this time point and the starting moment of the working cycle corresponding to it; Determine the cylinder volume corresponding to this time point according to the time difference, the total volume of the compressor cylinder, and the clearance volume; Obtain the working pressure corresponding to this time point according to the pressure - time correspondence relationship;

[0083] S230: Perform the following operations on each of the working cycles respectively: Obtain the working pressure and the cylinder volume corresponding to each time point in the time group corresponding to this working cycle; Use the cylinder volume as the abscissa and the working pressure as the ordinate to determine the actual indicator diagram of this working cycle.

[0084] In this embodiment, the specific process of step S220 is exemplified as follows. For example: The total volume of the compressor cylinder is 120 liters, and the clearance volume is 20 liters. Then the volume between the top dead center and the bottom dead center is 100 liters; The time point T = 10.05s is between time T 3 = 10s and time T 4 = 10.2s, that is, the time point T is in the third cycle; During the period from T 3 to T 4 , the position change of the piston is: top dead center - bottom dead center - top dead center, and the change of the cylinder volume is: 20 liters - 120 liters - 20 liters; The time difference between the time point T and the time T 3 moment is 0.05s. At this time, the position of the piston is in the middle position between the top dead center and the bottom dead center, that is, the volume from the piston position to the top dead center is 50 liters, and the cylinder volume is the sum of 50 liters and the clearance volume, that is, 70 liters.

[0085] In an exemplary embodiment, steps S270 - S290 are further included before step S300:

[0086] S270: Obtain a data set containing historical indicator diagrams, where the data set includes a labeled data set and an unlabeled data set. The labeled data set includes a standard data set and a fault data set. The label of the indicator diagram in the standard data set is the normal working state, and the label of the indicator diagram in the fault data set is the corresponding fault type.

[0087] S280: Cluster the historical indicator diagram data set based on a semi-supervised clustering algorithm to obtain the category clusters corresponding to the compressor in the normal working state and different fault type states respectively.

[0088] S290: Determine that the target historical indicator diagrams corresponding to the center points of each category cluster are the standard indicator diagram and the fault indicator diagram respectively.

[0089] In this embodiment, the semi-supervised clustering algorithm can be, but is not limited to, the label propagation algorithm, the adaptive semi-supervised clustering algorithm, and the semi-supervised K-means active learning clustering algorithm.

[0090] In this embodiment, the standard indicator diagram represents the indicator diagram of the compressor in the normal working state. As Figure 3 shown, there are several fault indicator diagrams, which respectively correspond to the indicator diagrams in different fault type states. As Figures 4a - 4h , Figure 4a is the fault indicator diagram corresponding to the suction valve leakage fault, Figure 4b is the fault indicator diagram corresponding to the too small flow area of the valve, Figure 4c is the fault indicator diagram corresponding to the too heavy mass of the valve disc, Figure 4d is the fault indicator diagram corresponding to the too hard spring, Figure 4e is the fault indicator diagram corresponding to the too soft spring, Figure 4f is the fault indicator diagram corresponding to the valve disc flutter fault, Figure 4g is the fault indicator diagram corresponding to the too high lift height, Figure 4h is the fault indicator diagram corresponding to the air flow pressure pulsation fault; where P 1 represents the theoretical suction pressure of the compressor, with the unit of Pa; P 2 represents the theoretical discharge pressure of the compressor, with the unit of Pa; V 1 represents the clearance volume of the compressor, with the unit of L; V 2 represents the total cylinder volume of the compressor, with the unit of L; the ab segment represents the expansion stroke; the bc segment represents the suction stroke; the cd segment represents the compression stroke; the da segment represents the discharge stroke.

[0091] In an exemplary embodiment, step S300 performs the following operations on the actual indicator diagram of each working cycle, which may include steps S310 - S320:

[0092] S310: Determine a first similarity between the actual indicator diagram of this working cycle and the standard indicator diagram. When the first similarity is greater than or equal to a first preset threshold, determine that the state is normal; when the first similarity is less than the first preset threshold, execute step S320;

[0093] S320: Determine a second similarity between the actual indicator diagram of this working cycle and the fault indicator diagrams of each different fault type, and select the maximum value among them as the maximum second similarity. Determine the fault type of the compressor in this working cycle as the fault type corresponding to the fault indicator diagram used to obtain the maximum second similarity.

[0094] In one implementation manner of this embodiment, in the chronological order of the working cycles within a predetermined time period, step S310 can be sequentially executed for each actual indicator diagram. If the first similarity is greater than or equal to the first preset threshold, it indicates that the actual indicator diagram of this working cycle has a large similarity with the standard indicator diagram, and it is determined that the compressor has no fault in this working cycle and is in a normal working state; S310 is executed in a loop until the first similarity of the actual indicator diagram of a certain determined working cycle is less than the first preset threshold, indicating that the compressor has a fault, and step S320 is executed to determine the fault type corresponding to this working cycle.

[0095] In another implementation manner of this embodiment, step S310 can be simultaneously executed for the actual indicator diagrams corresponding to all working cycles within a predetermined time period, and one or more actual indicator diagrams with a first similarity less than the first preset threshold are selected. The selected actual indicator diagrams are simultaneously executed for step S320 respectively to determine the fault type corresponding to each actual indicator diagram.

[0096] In an exemplary embodiment, comparing the actual indicator diagram with the pre-obtained reference indicator diagram in step S300 includes performing one or more of the following operations on the actual indicator diagram of each working cycle respectively:

[0097] Comparing the overall area of the figure enclosed by the closed line segments in the indicator diagram;

[0098] Comparing the line types corresponding to each stroke;

[0099] Comparing the lengths corresponding to each stroke;

[0100] Comparing the start and end positions corresponding to each stroke.

[0101] In this embodiment, in addition to the method for calculating the first similarity and the second similarity used in the above embodiment, the comparison method in this embodiment can also be adopted. Since the indicator diagram of the compressor is a closed curve composed of four sides, the area enclosed by the closed curve of the actual indicator diagram, the standard area enclosed by the closed curve of the standard indicator diagram, and the fault areas enclosed by the closed curves of various types of fault indicator diagrams can be calculated. By comparing the sizes of the areas, it can be analyzed whether the compressor has a fault and the type of the fault.

[0102] In this embodiment, since each working cycle of the compressor is divided into four strokes: expansion, suction, compression, and exhaust, which respectively correspond to the four sides of the indicator diagram, the line types, lengths, and starting and ending point positions of each stroke of the actual indicator diagram can be compared with the standard indicator diagram and various types of fault indicator diagrams. For example, if the line type of a certain stroke in the standard indicator diagram is a straight line, while the line type of this stroke in the actual indicator diagram is a curve, and the length of this stroke in the actual indicator diagram is shorter than that in the standard indicator diagram, it can be determined that the compressor has a fault. Further, the line type, length, and starting and ending point positions of this stroke are compared with those of this stroke in various types of fault indicator diagrams, and the fault corresponding to the fault indicator diagram that is most consistent with the line type, length, and starting and ending point positions of this stroke in the actual indicator diagram is selected as the fault type of the compressor.

[0103] In this embodiment, for each of the four sides representing the four strokes in the actual indicator diagram, if the length of each side increases or decreases, the line type of each side changes, the starting and ending point positions of the working pressure of each side change, or the starting and ending point positions of the cylinder volume of each side change, it can clearly indicate the fault type and the faulty component of the compressor. Based on this, the fault type and the faulty component can be quickly located, which reduces the work pressure of the operation and maintenance personnel and saves the time for finding the fault point.

[0104] In an exemplary embodiment, in step S300, the following operations are respectively performed on the actual indicator diagram of each working cycle, including steps S330 - S340:

[0105] S330: Take the actual indicator diagram of this working cycle as an input item of a preset classification model, output the probabilities of the actual indicator diagram belonging to various classification states, and determine the maximum probability; the preset classification model is used to output the classification state to which the input actual indicator diagram belongs according to the input actual indicator diagram, and the classification states include a normal state and different fault type states; the preset classification model is trained based on historical indicator diagrams marked with corresponding classification states.

[0106] S340: Determine the classification state to which the actual indicator diagram of this working cycle belongs according to the maximum probability, and judge the state of the compressor within this working cycle according to this classification state, where the state includes normal or a fault type.

[0107] This embodiment is different from the aforementioned semi-supervised clustering algorithm. In this embodiment, it is necessary to pre-construct the preset classification model. The preset classification model can adopt but is not limited to Bayesian classification model, decision tree classification model, and support vector machine classification model. The normal state of the preset classification model has a corresponding standard indicator diagram, and different fault states have corresponding fault indicator diagrams of various types.

[0108] In one implementation manner of this embodiment, the actual indicator diagrams can be successively executed step S330 in the chronological order of the working cycles within a predetermined time period. After inputting the actual indicator diagram of a certain working cycle into the preset classification model, the maximum probability is determined, and then step S340 is executed. When the classification corresponding to the maximum probability is the normal state, it can be determined that the compressor state of this working cycle is normal; when the classification corresponding to the maximum probability is the fault state of a certain fault type, it can be determined that the compressor state of this working cycle is the corresponding fault type.

[0109] In another implementation manner of this embodiment, the actual indicator diagrams corresponding to all working cycles within a predetermined time period can be simultaneously executed step S330 respectively to determine the maximum probability corresponding to each actual indicator diagram, and then step S340 is simultaneously executed for all actual indicator diagrams respectively to determine that the state corresponding to each actual indicator diagram is normal or a certain fault type.

[0110] In an exemplary embodiment, the operating parameters further include working temperature, working flow rate, and working frequency. After step S300, for the judgment result of each working cycle, steps S410 - S430 can further be included:

[0111] S410: When the judgment result of this working cycle is one of the different fault types, obtain the fault parameter state corresponding to this fault in the preset relationship. The fault parameter state includes at least one of the fault temperature state, fault flow rate state, and fault frequency state;

[0112] S420: Determine the type of operating parameter that needs to be verified for this working cycle according to the type of the obtained fault parameter state. The operating parameter that needs to be verified includes at least one of the working temperature, working flow rate, and working frequency;

[0113] S430: Verify the operating parameter that needs to be verified in this working cycle according to the fault parameter state. When the verification is passed, an early warning message is output.

[0114] In this embodiment, the preset relationship may include at least one of the fault temperature state, the fault flow state, and the fault frequency state respectively corresponding to different fault types. The fault temperature state includes rising, falling, or fluctuating. The fault flow state includes rising, falling, or fluctuating. The fault frequency state includes rising, falling, or fluctuating.

[0115] In this embodiment, the verification being passed may include: the state of the operating parameter to be verified being the same as the fault parameter state.

[0116] In this embodiment, the warning information may be directly output after step S300 is executed, or the warning information may be output after steps S410 - S430 are executed. Steps S410 - S430 are methods for further verifying the judgment result of step S300.

[0117] In this embodiment, a temperature sensor may be provided on the outer wall of the cylinder of the compressor to collect the working temperature of the compressor in real time and obtain the temperature - time correspondence relationship within a predetermined time period. A flow sensor may be provided at the exhaust valve position of the compressor to collect and detect the working flow of the compressor and obtain the flow - time correspondence relationship. A rotational speed sensor may be provided on the flywheel of the compressor to collect the rotational speed of the flywheel in real time, and the working frequency of the compressor may be obtained based on the rotational speed of the flywheel to obtain the frequency - time correspondence relationship. The collection time points of the temperature - time correspondence relationship, the flow - time correspondence relationship, and the frequency - time correspondence relationship are the same as those of the pressure - time correspondence relationship in step S100.

[0118] In this embodiment, the preset relationship includes at least one of the fault temperature state, fault flow state, or fault frequency state of four strokes of different fault types. The fault temperature state is the comparison result with the temperature of the compressor in the normal working state, the fault flow state is the comparison result with the flow rate of the compressor in the normal working state, and the fault frequency state is the comparison result with the frequency of the compressor in the normal working state. For example, for the expansion, suction, compression, and exhaust strokes of fault type A, only the fault temperature state exists in the preset relationship, which are rising, falling, fluctuating, and fluctuating respectively. For the expansion, suction, compression, and exhaust strokes of fault type B, there are not only the fault temperature states, which are falling, fluctuating, rising, and fluctuating respectively, but also the fault flow states, which are rising, falling, fluctuating, and fluctuating respectively. When step S300 determines that the fault type is A, it can be determined that the operating parameter to be verified is the working temperature. Since the acquisition time points of the temperature-time correspondence relationship and the pressure-time correspondence relationship are the same, the temperature-time correspondence relationship can be marked on the actual indicator diagram, and then whether the working temperatures of the four strokes are the same as the fault temperature states of the four strokes of fault type A can be verified respectively. If they are the same, it means the verification is passed and a warning message is output. Similarly, when step S300 determines that the fault type is B, it can be determined that the operating parameters to be verified are the working temperature and the working flow rate. The temperature-time correspondence relationship and the flow rate-time correspondence relationship can be marked on the actual indicator diagram respectively, and then whether the working temperatures of the four strokes are the same as the fault temperature states of the four strokes of fault type B can be verified respectively, and whether the working flow rates of the four strokes are the same as the fault flow states of the four strokes of fault type B can be verified respectively. When the verification is passed, a warning message is output.

[0119] In an exemplary embodiment, the different fault types may include at least one of the following:

[0120] Suction valve leakage fault, too small valve flow area fault, too heavy valve disc mass fault, too hard spring fault, too soft spring fault, valve disc flutter fault, too high lift height fault, and gas flow pressure pulsation fault.

[0121] In this embodiment, taking the suction valve leakage fault as an example, as Figure 5 shown, Figure 5 is Figure 4a for Figure 3Comparison diagram. The fault indicator diagram of the suction valve leakage fault is compared with the standard indicator diagram in the normal working state. The dotted line is the standard indicator diagram, and the solid line is the fault indicator diagram of the suction valve leakage fault. When the suction valve leaks, during the expansion process, corresponding to the ab section of the indicator diagram, since the suction valve is in the leakage state, the residual gas originally remaining in the cylinder pushes the piston to move and leaks out at the same time, resulting in insufficient residual gas. The volume of the cylinder required for the pressure to drop to the suction pressure becomes smaller. Reflected on the actual indicator diagram, the expansion process is faster than the standard indicator diagram; during the suction process, corresponding to the bc section of the indicator diagram, since the distance that the piston needs to move during expansion becomes smaller and the position of the bottom dead center remains unchanged, the volume of suction increases. Reflected on the actual indicator diagram, the suction process becomes longer; during the compression process, corresponding to the cd section of the indicator diagram, since the suction valve continuously leaks out high-pressure gas, a part of the effective gas will be lost when the cylinder reaches the exhaust pressure, and the compression distance required to reach the exhaust pressure increases. After the above process, the exhaust distance decreases, that is, the da section of the indicator diagram becomes shorter.

[0122] The embodiment of the present application also provides a compressor fault judgment device, as Figure 6 shown, including a processor and a memory storing a computer program. When the computer program is executed by the processor, the compressor fault judgment method described in the above embodiment can be implemented.

[0123] Those of ordinary skill in the art will understand that all or some of the steps in the methods disclosed above, and the functional modules / units in the systems and devices, can be implemented as software, firmware, hardware, and appropriate combinations thereof. In the hardware implementation, the division of the functional modules / units mentioned above does not necessarily correspond to the division of physical components; for example, one physical component may have multiple functions, or one function or step may be executed by several physical components in cooperation. Some or all components may be implemented as software executed by a processor, such as a digital signal processor or a microprocessor, or may be implemented as hardware, or may be implemented as an integrated circuit, such as an application specific integrated circuit. Such software can be distributed on a computer-readable medium, which can include a computer storage medium (or non-transitory medium) and a communication medium (or transitory medium). As is well known to those of ordinary skill in the art, the term computer storage medium includes volatile and non-volatile, removable and non-removable media implemented in any method or technology for storing information, such as computer-readable instructions, data structures, program modules, or other data. Computer storage media includes, but is not limited to, RAM, ROM, EEPROM, flash memory or other memory technologies, CD-ROM, digital versatile disks (DVDs) or other optical disk storage, magnetic cassettes, tapes, magnetic disk storage or other magnetic storage devices, or any other medium that can be used to store the desired information and can be accessed by a computer. In addition, it is well known to those of ordinary skill in the art that communication media typically contains computer-readable instructions, data structures, program modules, or other data in a modulated data signal such as a carrier wave or other transmission mechanism, and can include any information delivery medium.

Claims

1. A method for judging compressor faults, characterized in that, it includes: Obtain the operating parameters of the compressor at multiple time points within a predetermined time period, where the operating parameters include the working pressure, and obtain the pressure-time correspondence; Obtain the division relationship of multiple working cycles within the predetermined time period; Determine the actual indicator diagram of the compressor for each working cycle within the predetermined time period according to the pressure-time correspondence and the division relationship of the multiple working cycles; Compare the actual indicator diagram of each working cycle with the pre-obtained reference indicator diagram respectively, and judge the state of the compressor according to the comparison result, where the state includes normal or fault type; Among them, the reference indicator diagram includes the standard indicator diagram of the compressor in the normal working state and the fault indicator diagrams corresponding to different fault types in the working state.

2. The compressor fault judgment method according to claim 1, characterized in that, The obtaining the division relationship of multiple working cycles within the predetermined time period includes: Obtain the moment when the zero point mark on the compressor flywheel returns to the preset mark point each time within the predetermined time period, where the preset mark point is located outside the compressor flywheel and is used to combine with the zero point mark to determine the working cycle of the compressor; where, one working cycle of the compressor is the time period between any two adjacent moments when returning to the preset mark point.

3. The compressor fault judgment method according to claim 2, characterized in that, The determining the actual indicator diagram of the compressor within the predetermined time period according to the pressure-time correspondence and the division relationship of the multiple working cycles includes: Group the pressure-time correspondence, and respectively correspond the multiple time points to each working cycle to obtain the time group corresponding to each working cycle; Perform the following operations on each time point respectively: obtain the time difference between the time point and the starting moment of the working cycle corresponding to it; determine the cylinder volume corresponding to the time point according to the time difference, the total cylinder volume and the clearance volume of the compressor; obtain the working pressure corresponding to the time point according to the pressure-time correspondence; Perform the following operations on each working cycle respectively: obtain the working pressure and the cylinder volume corresponding to each time point in the time group corresponding to the working cycle; use the cylinder volume as the abscissa and the working pressure as the ordinate to determine the actual indicator diagram of the working cycle.

4. The compressor fault judgment method according to claim 1, characterized in that, Before comparing the actual indicator diagram of each working cycle with the pre-obtained reference indicator diagram respectively, it further includes: Obtain a data set containing historical indicator diagrams, where the data set includes a labeled data set and an unlabeled data set, and the labeled data set includes a standard data set and a fault data set; the label of the indicator diagram in the standard data set is the normal working state; the label of the indicator diagram in the fault data set is the corresponding fault type; Cluster the historical indicator diagram dataset based on a semi-supervised clustering algorithm to obtain the category clusters corresponding to the compressor in the normal working state and different fault type states respectively; Determine that the target historical indicator diagrams corresponding to the center points of each category cluster are the standard indicator diagram and the fault indicator diagram respectively.

5. The compressor fault judgment method according to claim 4, characterized in that, in the step of comparing the actual indicator diagram of each working cycle with the pre-obtained reference indicator diagram respectively and judging the state of the compressor according to the comparison result, the state including normal or fault type, the following operations are respectively performed on the actual indicator diagram of each working cycle: Determine the first similarity between the actual indicator diagram of this working cycle and the standard indicator diagram. If the first similarity is greater than or equal to the first preset threshold, determine that the state is normal; If the first similarity is less than the first preset threshold, determine the second similarity between the actual indicator diagram of this working cycle and the fault indicator diagrams of each different fault type, and select the maximum value as the maximum second similarity, and determine that the fault type of the compressor in this working cycle is the fault type corresponding to the fault indicator diagram used to obtain the maximum second similarity.

6. The compressor fault judgment method according to claim 4, characterized in that, in the step of comparing the actual indicator diagram of each working cycle with the pre-obtained reference indicator diagram respectively, the following one or more operations are respectively performed on the actual indicator diagram of each working cycle: Compare the overall area of the figure enclosed by the closed line segments in the indicator diagram; Compare the line types corresponding to each stroke; Compare the lengths corresponding to each stroke; Compare the starting and ending point positions corresponding to each stroke.

7. The compressor fault judgment method according to claim 1, characterized in that, in the step of comparing the actual indicator diagram of each working cycle with the pre-obtained reference indicator diagram respectively and judging the state of the compressor according to the comparison result, the state including normal or fault type, the following operations are respectively performed on the actual indicator diagram of each working cycle: Take the actual indicator diagram of this working cycle as the input item of a preset classification model, output the probabilities of the actual indicator diagram belonging to various classification states, and determine the maximum probability; The preset classification model is used to output the classification state to which the input actual indicator diagram belongs, and the classification states include a normal state and different fault type states; The preset classification model is trained according to the historical indicator diagrams marked with corresponding classification states; Determine the classification state to which the actual indicator diagram of this working cycle belongs according to the maximum probability, and judge the state of the compressor in this working cycle according to this classification state, and the state includes normal or fault type.

8. The compressor fault judgment method according to any one of claims 5-7, characterized in that, The operating parameters further include the working temperature, the working flow rate, and the working frequency. After comparing the actual indicator diagram of each working cycle with the pre-obtained reference indicator diagram respectively and judging the state of the compressor according to the comparison result, where the state includes normal or fault types, the following operations are respectively performed on the judgment results of each working cycle: When the judgment result of this working cycle is one of the different fault types, obtain the fault parameter state corresponding to this fault in the preset relationship, and the fault parameter state includes at least one of the fault temperature state, the fault flow state, and the fault frequency state; Determine the type of operating parameter to be verified in this working cycle according to the type of the obtained fault parameter state, and the operating parameter to be verified includes at least one of the working temperature, the working flow rate, and the working frequency; Verify the operating parameter to be verified in this working cycle according to the fault parameter state, and output a warning message when the verification is passed; Wherein, the preset relationship includes at least one of the fault temperature state, the fault flow state, and the fault frequency state respectively corresponding to the different fault types, the fault temperature state includes rising, falling, or fluctuating, the fault flow state includes rising, falling, or fluctuating, and the fault frequency state includes rising, falling, or fluctuating; Wherein, the verification being passed includes: the state of the operating parameter to be verified is the same as the fault parameter state.

9. The compressor fault judgment method according to claim 1, characterized in that, the different fault types include at least one of the following: suction valve leakage fault, too small valve flow area fault, too heavy valve disc mass fault, too hard spring fault, too soft spring fault, valve disc flutter fault, too high lift height fault, and air flow pressure pulsation fault.

10. A compressor fault judgment device, characterized in that, it includes a processor and a memory storing a computer program, and is characterized in that when the computer program is executed by the processor, the compressor fault judgment method according to any one of claims 1-9 can be implemented.

Images