Abnormality cause estimation device of fluid machine, abnormality cause estimation method and abnormality cause estimation system of fluid machine
Patent Information
- Application Number
- JP2021200117
- Authority / Receiving Office
- JP · JP
- Patent Type
- Applications
- Current Assignee / Owner
- Filing Date
- 2021-12-09
- Publication Date
- 2024-11-22
- Estimated Expiration
- 2041-12-09
AI Technical Summary
Existing technologies face challenges in accurately estimating the cause of abnormalities in fluid machines like compressors due to the difficulty in directly measuring and monitoring the internal state, leading to inaccurate candidate narrowing and increased costs from high-frequency monitoring.
An abnormality cause estimation apparatus and method that utilizes sensors to input measured values, calculates performance, determines abnormalities, and estimates causes using a database, enabling accurate abnormality detection in fluid machines.
Enables easy and highly accurate estimation of abnormality causes in fluid machines by considering internal states indirectly measured, reducing costs and improving diagnostic accuracy.
Smart Images

Figure 00000000_0000_ABST
Abstract
Description
Technical Field
[0005]
[0001] The present invention relates to an abnormality cause estimation device for a fluid machine such as a compressor for estimating an abnormality cause of the fluid machine, an abnormality cause estimation method thereof, and an abnormality cause estimation system for a fluid machine.
Background Art
[0002] In recent years, in chemical plants and power plants, for the purpose of stable operation and rationalization of maintenance work, the introduction of condition monitoring and diagnostic techniques for plant equipment has been promoted.
[0003] When an abnormality occurs in plant equipment, by detecting the abnormality of the plant equipment at an early stage and taking countermeasures, it is possible to expect continuous operation and shortening of maintenance work time. In order to implement countermeasures more effectively, when an abnormality occurs in plant equipment, a diagnostic technique for estimating the cause of the abnormality is important.
[0004] As an example of a diagnostic technique for equipment, for example, in Patent Document 1, in order to more accurately diagnose equipment, a phenomenon pattern extraction unit that extracts a phenomenon pattern of past sensor signals of the equipment, a linking unit that links the sensor signals based on maintenance history information, a classification criterion creation unit that creates a classification criterion for classifying the phenomenon pattern based on the phenomenon pattern extracted by the phenomenon pattern extraction unit and a work keyword included in the maintenance history information that links the sensor signal that is the basis of the phenomenon pattern, a phenomenon pattern classification unit that classifies the phenomenon pattern based on the classification criterion created by the classification criterion creation unit, and a diagnostic model creation unit that creates a diagnostic model for estimating a work keyword to be presented to a maintenance worker based on the phenomenon pattern classified by the phenomenon pattern classification unit and the work keyword are described.
[0005] Furthermore, Patent Document 2 describes a fault diagnosis device for equipment, which includes at least one measuring means for measuring the operating state of a contact device, a plurality of fault diagnosis means for analyzing the measurement results of the measuring means and outputting a group of fault cause candidates including one or more sets of fault cause candidates and their confidence levels, and an integrating means for integrating the group of fault cause candidates output by each of the plurality of fault diagnosis means and outputting a total diagnosis result.
[0006] On the other hand, Patent Document 3 describes a vibration diagnostic device for rotating machinery that, in order to improve the confidence level of the diagnostic results for operators, service operators, or engineers of remote monitoring services when diagnosing vibrations of rotating machinery, includes a diagnostic processing means that takes in various data obtained from the rotating machinery and estimates the cause of vibration from a vibration causal matrix stored in a knowledge base by relating the characteristics of multiple vibration phenomena occurring in the rotating machinery with multiple causes; a memory that stores vibration analysis models corresponding to the diagnostic results in advance; a vibration analysis verification means that takes in the vibration cause estimated by the diagnostic processing means and retrieves an analysis model that matches this vibration cause from the memory for analysis; and a diagnostic result evaluation means that takes in the analysis verification results from the vibration analysis verification means and the diagnostic results from the diagnostic processing means and evaluates the occurrence of vibration phenomena. [Prior art documents] [Patent Documents]
[0007] [Patent Document 1] Japanese Patent Publication No. 2015-148867 [Patent Document 2] Japanese Patent Publication No. 2007-293489 [Patent Document 3] Japanese Patent Publication No. 2003-149043 [Overview of the project] [Problems that the invention aims to solve]
[0008] Incidentally, in fluid machinery such as compressors, the casing is designed to prevent fluid leakage to the outside of the casing by providing fluid flow channels. Therefore, it is difficult to install sensors inside the fluid machinery to directly measure and monitor its internal condition.
[0009] As described in the aforementioned Patent Documents 1-3, when attempting to estimate the cause of an anomaly based solely on measured values, it was not possible to accurately narrow down the possible causes of anomalies related to changes in the internal state of the fluid machine, since measured values from inside the fluid machine could not be obtained.
[0010] Furthermore, generally, extracting the characteristics of vibration phenomena as described in Patent Document 3 often requires frequency analysis, and estimating the cause of vibration based on a vibration causal matrix also assumes that the frequency components are known. For this purpose, it is necessary to acquire measured values at a high sampling frequency. Continuously monitoring measured values at a high sampling frequency may lead to increased costs for measurement systems and data storage systems.
[0011] The present invention has been made in view of the above-mentioned points, and its object is to provide a device for estimating the cause of abnormalities in fluid machinery, a method for estimating the cause of abnormalities in fluid machinery, and a system for estimating the cause of abnormalities in fluid machinery, which can estimate the cause of abnormalities in fluid machinery simply and with high accuracy. [Means for solving the problem]
[0012] The fluid machine abnormality cause estimation device of the present invention is a fluid machine abnormality cause estimation device that monitors the state of a fluid machine and estimates the cause of an abnormality when an abnormality is detected, in order to achieve the above objective, and is characterized by comprising: a measurement value input unit that receives measurement values obtained from sensors installed on the fluid machine; a performance calculation unit that calculates the performance of the fluid machine from the measurement values obtained from the measurement value input unit; an abnormality determination unit that determines whether or not there is an abnormality in the measurement values obtained from the measurement value input unit and the performance of the fluid machine calculated by the performance calculation unit; an abnormality cause estimation unit that estimates the cause of an abnormality in the fluid machine based on the determination result of whether or not there is an abnormality in the performance of the fluid machine determined by the abnormality determination unit and the cause of an abnormality of the fluid machine stored in advance in an abnormality cause database; and a result output unit that outputs the abnormality cause estimation result of the fluid machine estimated by the abnormality cause estimation unit.
[0013] Furthermore, the present invention provides a method for estimating the cause of an abnormality in a fluid machine, which, in order to achieve the above objective, monitors the state of the fluid machine and estimates the cause of an abnormality when one occurs, characterized in that: measured values obtained from sensors installed on the fluid machine are input to a measured value input unit; the performance of the fluid machine is calculated by a performance calculation unit from the measured values obtained by the measured value input unit; an abnormality determination unit determines whether there is an abnormality in the measured values obtained by the measured value input unit and the performance of the fluid machine calculated by the performance calculation unit; an abnormality cause estimation unit estimates the cause of the abnormality in the fluid machine based on the determination result of whether there is an abnormality in the performance of the fluid machine determined by the abnormality determination unit and the causes of the abnormality of the fluid machine stored in advance in an abnormality cause database; and thereafter, the abnormality cause estimation result of the fluid machine estimated by the abnormality cause estimation unit is output to a result output unit.
[0014] In addition, in order to achieve the above object, in the abnormal cause estimation system of the fluid machine of the present invention, the measured value of the fluid machine measured by a sensor installed at a predetermined location of the fluid machine is transmitted to an abnormal cause estimation device via a network, and the abnormal cause of the fluid machine estimated by the abnormal cause estimation device based on the measured value is transmitted to the owner of the fluid machine. The abnormal cause estimation system is characterized in that the abnormal cause estimation device is an abnormal cause estimation device for a fluid machine having the above configuration.
Effects of the Invention
[0015] According to the present invention, the abnormal cause of a fluid machine can be estimated simply and with high accuracy.
Brief Description of the Drawings
[0016] [Figure 1] It is a diagram showing the overall configuration of a centrifugal compressor to which the abnormal cause estimation device of the fluid machine of the present invention is applied. [Figure 2] It is a diagram showing Example 1 of the abnormal cause estimation device of the fluid machine of the present invention. [Figure 3] It is a diagram showing an example of measured values in Example 1 of the abnormal cause estimation device of the fluid machine of the present invention. [Figure 4] It is a diagram showing an example of the result of determining the presence or absence of an abnormality by an abnormality determination unit in Example 1 of the abnormal cause estimation device of the fluid machine of the present invention. [Figure 5] It is a diagram showing an example of an abnormal cause database in Example 1 of the abnormal cause estimation device of the fluid machine of the present invention. [Figure 6] It is a cross-sectional view showing an enlarged view of the periphery of the impeller of the centrifugal compressor shown in FIG. 1. [Figure 7] It is a diagram showing an example of a diagnostic result output and displayed by a result output unit in Example 1 of the abnormal cause estimation device of the fluid machine of the present invention.
Modes for Carrying Out the Invention
[0017] Hereinafter, an apparatus for estimating the cause of abnormality of a fluid machine, a method for estimating the cause of abnormality thereof, and a system for estimating the cause of abnormality of a fluid machine according to the present invention will be described based on the illustrated embodiments. In each figure, the same reference numerals are used for the same components.
Embodiment
[0018] Fig. 1 shows the overall configuration of a centrifugal compressor 201 to which the apparatus for estimating the cause of abnormality of a fluid machine according to the present invention is applied.
[0019] As shown in Fig. 1, the centrifugal compressor 201 is generally composed of a casing 202 that is usually formed in a cylindrical shape and serves as a stationary part, a rotor 205 disposed inside the casing 202, and journal bearings 203 and thrust bearings 204 that rotatably support the rotor 205. A plurality of impellers 206 are provided on the above-mentioned rotor 205, and the fluid flowing in the flow path is compressed by the rotation of the impellers 206.
[0020] Sensors for measuring the shaft vibration, suction temperature, discharge temperature, suction pressure, discharge pressure, rotational speed, etc. of the centrifugal compressor 201 are appropriately installed at locations suitable for each measurement on the centrifugal compressor 201 having the above configuration.
[0021] The sensor 207 shown in Fig. 1 is installed on the rotor 205 to measure the shaft vibration as an example. The measured value (shaft vibration) of the centrifugal compressor 201 measured by this sensor 207 is transmitted to a monitoring system 210 via a network such as a DCS (Distributed Control System) 208 and a LAN 209. When an abnormality is detected while monitoring the state of the centrifugal compressor 201 from the monitoring system 210, it is transmitted to an abnormality cause estimation device 1 that estimates the cause of the abnormality. The cause of the abnormality of the centrifugal compressor 201 estimated by the abnormality cause estimation device 1 based on the measured value measured by the sensor 207 constitutes an abnormality cause estimation system that is transmitted to the owner (contract user) 211 of the centrifugal compressor 201.
[0022] Next, the above-mentioned abnormality cause estimation device 1 will be described with reference to Fig. 2. Fig. 2 is a diagram showing Embodiment 1 of the apparatus for estimating the cause of abnormality of a fluid machine according to the present invention.
[0023] As shown in Figure 2, the abnormality cause estimation device 1 of this embodiment is generally composed of: a measurement value input unit 2 into which measurement values obtained from a sensor 207 installed on the centrifugal compressor 201 are input; a performance calculation unit 3 that calculates the performance of the centrifugal compressor 201 from the measurement values obtained from the measurement value input unit 2; an abnormality determination unit 4 that determines whether there is an abnormality in the performance of the centrifugal compressor 201 obtained from the measurement value input unit 2 and the performance calculation unit 3; an abnormality cause estimation unit 6 that estimates the abnormality cause of the centrifugal compressor 201 based on the determination result of whether there is an abnormality in the performance of the centrifugal compressor 201 determined by the abnormality determination unit 4 and the abnormality causes of the centrifugal compressor 201 stored in advance in the abnormality cause database 7; and a result output unit 8 that outputs the abnormality cause estimation result of the centrifugal compressor 201 estimated by the abnormality cause estimation unit 6. Reference numeral 5 denotes the normal measurement value database in which the measurement values and performance of the centrifugal compressor 201 under normal conditions are stored.
[0024] In Figure 2, the measurement input unit 2 receives measurement values such as shaft vibration, suction temperature, discharge temperature, suction pressure, discharge pressure, and rotational speed obtained from the sensor 207 installed on the centrifugal compressor 201 (Figure 3 shows an example of measurement values input to the measurement input unit 2, where measurement values obtained at each time point are input to the measurement input unit 2).
[0025] Furthermore, the performance calculation unit 3 calculates performance from a portion of the measured values obtained from the measured value input unit 2. In this embodiment, the polytropic efficiency of the centrifugal compressor 201 is calculated from the measured values of the suction temperature, suction pressure, discharge temperature, discharge pressure, and gas composition measured by the sensor 207.
[0026] In this embodiment, polytrope efficiency was used as the performance indicator for the centrifugal compressor 201, but this is not the only option. For example, similar effects can be obtained by using other indicators that represent the performance of the centrifugal compressor 201, such as polytrope head or shaft power, instead of polytrope efficiency.
[0027] Furthermore, the abnormality determination unit 4 has the function of determining whether or not there is an abnormality in the measured values obtained from the sensor 207 input to the measured value input unit 2 and the polytrope efficiency (performance) calculated by the performance calculation unit 3.
[0028] An abnormality in the measurement values and performance of sensor 207 means that the measurement values and performance of sensor 207 deviate from a predetermined normal range, which may impair or have already impaired the normal function of centrifugal compressor 201.
[0029] Furthermore, the abnormality determination unit 4 obtains the measured values and performance of the centrifugal compressor 201 under normal conditions from the normal measurement value database 5, which is stored in advance. Based on the measured values and performance under normal conditions obtained from the normal measurement value database 5, it determines whether or not there is an abnormality in the measured values measured by the sensor 207 and the polytrope efficiency (performance) calculated by the performance calculation unit 3.
[0030] For example, the Mahalanobis-Taguchi method can be used as the processing method for such anomaly detection unit 4, but it is not limited to this.
[0031] Figure 4 shows an example of the results of the abnormality detection unit 4 determining whether there is an abnormality in the centrifugal compressor 201. It shows a case where an abnormality is detected in the shaft vibration of the measured values input to the measured value input unit 2, and in the polytropic efficiency calculated by the performance calculation unit 3 from the measured values of suction temperature, suction pressure, discharge temperature, discharge pressure, and gas composition. The abnormality detection result determined by this abnormality detection unit 4 is output to the abnormality cause estimation unit 6.
[0032] Furthermore, the abnormality cause database 7 has pre-stored information on the relationship between the cause of an abnormality in the centrifugal compressor 201 and the measured values and performance at which the abnormality appears. On the other hand, the abnormality cause estimation unit 6 has the function of estimating the cause of an abnormality in the centrifugal compressor 201 based on the abnormality determination result received from the abnormality determination unit 4.
[0033] In other words, the abnormality cause estimation unit 6 estimates the cause of the abnormality in the centrifugal compressor 201 based on the abnormality determination result of the input abnormality determination unit 4 and the relationship between the abnormality cause of the centrifugal compressor 201 and the measured values and performance in which the abnormality appears, which are stored in advance in the abnormality cause database 7.
[0034] As a processing method for such anomaly cause estimation unit 6, for example, a method is used in which causal relationships are organized in a directed acyclic graph structure and the probability of occurrence of each anomaly cause is inferred from the presence or absence of anomalies, but it is not limited to this.
[0035] An example of the abnormal cause database 7 related to this embodiment is shown in Figure 5.
[0036] By using not only the measured values but also the presence or absence of abnormalities in the polytrope efficiency shown in Figure 5 to estimate the cause, it is possible to estimate the cause of the abnormality while considering the internal state of the centrifugal compressor 201, which cannot be directly measured by the sensor 207.
[0037] For example, if the shaft vibration and polytrope efficiency shown in Figure 5 are determined to be abnormal, it is suspected that dirt is adhering to a part of the flow path of the impeller 206. That is, as shown in Figure 6, which shows a magnified view of the area around the impeller 206, if the shaft vibration and polytrope efficiency are determined to be abnormal, it can be inferred that the flow path dirt 601 is adhering to the impeller 206, increasing the imbalance of the rotor 205, and causing abnormal shaft vibration.
[0038] Furthermore, if abnormalities are detected in the shaft movement, the temperature of the thrust bearing 204, and the polytrope efficiency as shown in Figure 5, it is suspected that the pressure balance inside the centrifugal compressor 201 has changed. In other words, if abnormalities are detected in the shaft movement, the temperature of the thrust bearing 204, and the polytrope efficiency, as shown in Figure 6, it is possible to estimate that the cause of the abnormality may be an increase in internal leakage due to deterioration of the eye seal portion 602 or interstage seal portion 603, or the balance drum seal portion (not shown), which prevents gas from flowing from the high-pressure side to the low-pressure side through the gap between the impeller 206 and the casing 202 and rotor 205.
[0039] Furthermore, the result output unit 8 receives the measured values from the abnormality cause estimation unit 6, the abnormality judgment result for performance, and the estimated cause of the abnormality, and outputs and displays graphs and tables as diagnostic results.
[0040] Figure 7 shows an example of the diagnostic results displayed in the results output unit 8. The abnormal parameters displayed are axial vibration and polytrope efficiency, and the cause of the abnormality is indicated as flow path contamination.
[0041] According to this embodiment, when the cause of an anomaly is estimated based solely on whether or not there is an anomaly in the measured value obtained from the sensor 207, it is not possible to separate candidate causes of an anomaly (by using not only the measured value but also whether or not there is an anomaly in the polytrope efficiency in the cause estimation, the cause of the anomaly is estimated considering the internal state of the centrifugal compressor 201, which cannot be directly measured by the sensor 207). However, by estimating the cause of an anomaly based on the anomaly in the measured value and the deterioration of performance (decrease in polytrope efficiency) calculated from the measured value, it becomes possible to narrow down candidate causes of an anomaly that take into account the deterioration of the internal state of the centrifugal compressor 201, and the cause of an anomaly in the centrifugal compressor 201 can be estimated simply and with higher accuracy.
[0042] It should be noted that the present invention is not limited to the embodiments described above, and various modifications are included. For example, the embodiments described above are described in detail for the purpose of explaining the present invention in an easy-to-understand manner, and are not necessarily limited to those having all the configurations described. Furthermore, some or all of the above configurations and functions may be realized by designing them, for example, with an integrated circuit. Furthermore, the above configurations and functions may be realized in software by having a processor interpret and execute a program that realizes each function. [Explanation of symbols]
[0043] 1...Anomaly cause estimation device, 2...Measurement value input unit, 3...Performance calculation unit, 4...Anomaly determination unit, 5...Normal measurement value database, 6...Anomaly cause estimation unit, 7...Anomaly cause database, 8...Result output unit, 201...Centrifugal compressor, 202...Casing, 203...Journal bearing, 204...Thrust bearing, 205...Rotor, 206...Impeller, 207...Sensor, 208...DCS (Distributed Control System), 209...LAN, 210...Monitoring system, 211...Owner of centrifugal compressor, 601...Flow path contamination, 602...Eye seal section, 603...Interstage seal section.
Claims
1. A device for estimating a cause of an abnormality in a fluid machine, which monitors a state of the fluid machine and, if an abnormality occurs, estimates a cause of the abnormality, the fluid machinery includes a measurement value input unit to which measurement values obtained from a sensor installed in the fluid machinery are input, a performance calculation unit to calculate the performance of the fluid machinery from the measurement values obtained by the measurement value input unit, an abnormality determination unit to determine whether or not there is an abnormality in the measurement values obtained by the measurement value input unit and the performance of the fluid machinery calculated by the performance calculation unit, an abnormality cause estimation unit to estimate a cause of an abnormality of the fluid machinery based on a determination result of whether or not there is an abnormality in the performance of the fluid machinery determined by the abnormality determination unit and on the cause of the abnormality of the fluid machinery estimated from a relationship between the measurement values and performance in which an abnormality appears and the cause of the abnormality of the fluid machinery stored in advance in an abnormality cause database, and a result output unit to output an abnormality cause estimation result of the fluid machinery estimated by the abnormality cause estimation unit, The abnormality determination unit acquires normal measurement values and performance of the fluid machinery that are stored in advance in a normal measurement value database, and determines whether or not there is an abnormality in the measurement values obtained by the measurement value input unit and the performance calculated by the performance calculation unit based on the normal measurement values and performance acquired from the normal measurement value database.
2. 2. The device for estimating a cause of an abnormality in a fluid machine according to claim 1, 2. A device for estimating the cause of an abnormality in a fluid machinery, comprising: a first index representing a performance of the fluid machinery;
3. 3. The abnormality cause estimating device for a fluid machinery according to claim 2, 13. The device for estimating the cause of an abnormality in a fluid machinery, wherein either a polytropic efficiency, a polytropic head or a shaft power is used as the efficiency of the fluid machinery.
4. The abnormality cause estimating device for a fluid machinery according to claim 3, The fluid machine is a centrifugal compressor including a cylindrical casing, a rotor disposed in the casing, and a journal bearing and a thrust bearing which rotatably support the rotor, the rotor being provided with a plurality of impellers, and a fluid flowing through a flow path is compressed by the rotation of the impellers, When an abnormality is detected in the axial vibration of the centrifugal compressor measured by the sensor and input to the measurement value input unit, and in the polytropic efficiency, the polytropic head, or the axial power calculated from the suction temperature, suction pressure, discharge temperature, discharge pressure, and gas composition of the centrifugal compressor by the performance calculation unit, the device estimates that an abnormality in the axial vibration is caused by dirt adhering to a part of the flow path of the impeller.
5. The abnormality cause estimating device for a fluid machinery according to claim 3, The fluid machine is a centrifugal compressor including a cylindrical casing, a rotor disposed in the casing, and a journal bearing and a thrust bearing which rotatably support the rotor, the rotor being provided with a plurality of impellers, and a fluid flowing through a flow path is compressed by the rotation of the impellers, When an abnormality is detected in the axial movement of the centrifugal compressor, the temperature of the thrust bearing, and the polytropic efficiency or the polytropic head or the axial power, an increase in internal leakage due to deterioration of an eye seal, an interstage seal, or a balance drum seal that prevents gas from flowing from the high pressure side to the low pressure side through gaps between the impeller and the casing and the rotor is estimated as a possible cause of the abnormality.
6. A method for estimating a cause of an abnormality in a fluid machine, which monitors a state of the fluid machine and, if an abnormality occurs, estimates a cause of the abnormality, comprising: a measurement value input unit inputs measurement values obtained from sensors installed in the fluid machine, a performance calculation unit calculates the performance of the fluid machine from the measurement values obtained by the measurement value input unit, an abnormality determination unit determines whether or not there is an abnormality in the measurement values obtained by the measurement value input unit and in the performance of the fluid machine calculated by the performance calculation unit, and an abnormality cause estimation unit estimates a cause of the abnormality of the fluid machine based on a determination result of whether or not there is an abnormality in the performance of the fluid machine determined by the abnormality determination unit and a cause of the abnormality of the fluid machine estimated from a relationship between the measurement values and performance in which an abnormality appears and the cause of the abnormality of the fluid machine stored in advance in an abnormality cause database, and then outputs the result of the estimation of the cause of the abnormality of the fluid machine estimated by the abnormality cause estimation unit to a result output unit, The method for estimating the cause of an abnormality in a fluid machinery is characterized in that the abnormality judgment unit acquires normal measurement values and performance of the fluid machinery that are previously stored in a normal measurement value database, and determines whether or not there is an abnormality in the measurement values obtained by the measurement value input unit and the performance calculated by the performance calculation unit based on the normal measurement values and performance acquired from the normal measurement value database.
7. A method for estimating a cause of an abnormality in a fluid machinery according to claim 6, comprising the steps of: A method for estimating the cause of an abnormality in a fluid machinery, comprising the steps of: (a) determining whether or not a fluid machinery is in an abnormal state;
8. A method for estimating a cause of an abnormality in a fluid machinery according to claim 7, A method for estimating the cause of an abnormality in a fluid machinery, comprising the steps of: determining whether or not the efficiency of the fluid machinery is a polytropic efficiency; determining whether or not the efficiency of the fluid machinery is a polytropic head; and determining whether or not the efficiency of the fluid machinery is a polytropic efficiency.
9. A system for estimating a cause of an abnormality in a fluid machine, in which measurement values of the fluid machine are measured by a sensor installed at a predetermined position of the fluid machine and transmitted to an abnormality cause estimation device via a network, and a cause of the abnormality in the fluid machine estimated by the abnormality cause estimation device based on the transmitted measurement values is transmitted to an owner of the fluid machine, 6. A system for estimating the cause of an abnormality in a fluid machine, comprising: the abnormality cause estimating device according to claim 1 ;
