CN117932349A - Multi-parameter fusion health diagnosis and fault prediction method, instrument and system for machine - Google Patents

Abstract

The invention provides a method, an instrument and a system for machine multi-parameter fusion health diagnosis and fault prediction, wherein the method comprises the following steps: the method comprises the steps of obtaining fault cause parameters in a machine, wherein the fault cause parameters comprise online oil parameters, online vibration parameters, online temperature parameters and electrical signal parameters, processing fault cause parameter data, comparing the processed fault cause parameter data with corresponding data in a database, outputting a fault diagnosis report and a fault prediction report of the machine based on comparison results, and generating machine operation and maintenance suggestions based on report results. According to the invention, the health state of the machine is accurately judged through a plurality of key operation parameters, the machine fault is early warned, an optimization scheme of machine operation and maintenance is provided, and the fault rate is reduced or the fault is avoided.

Description

Multi-parameter fusion health diagnosis and fault prediction method, instrument and system for machine

Technical Field

The invention mainly relates to the technical field related to fault monitoring of machine equipment, in particular to a method, an instrument and a system for multi-parameter fusion health diagnosis and fault prediction of a machine.

Background

The fault diagnosis device can diagnose and early warn faults of machine equipment through analysis of operation data of lubricating systems such as gearbox driving systems, turbine systems and the like in industries such as mines, coal mines, thermal power, traffic, wind power, chemical industry, cement, steel, military industry and the like.

The existing methods for machine health diagnosis and the corresponding equipment conditions are as follows:

1) And an online oil liquid sensor is independently used for detecting metal abrasion particles, oil quality and pollution. The oil sensor can find out the abrasion fault of an oil lubrication part in the equipment earlier, but the specific fault point is difficult to locate; for faults unrelated to oil lubrication, such as: misalignment, loose foundations, unbalance, electrical failure, etc., which are not possible for the oil sensor to detect.

2) The on-line vibration sensor is independently used, and the vibration sensor is very sensitive to misalignment, foundation looseness, unbalance, electrical faults and the like; wear and loosening faults of parts requiring lubrication can be found: such as bearings, gears, shafting, etc.; but lags behind the oil wear sensor when a failure is found that requires lubrication of the components.

3) There are also techniques for comprehensively diagnosing machine faults using a handheld offline vibration analysis system and an offline oil analysis system. The offline oil analysis needs to be sampled by a person, and the samples are sent to a laboratory for detection, so that the period is long and the automation cannot be realized.

Some existing technologies put the objective on diagnosing faults, do not study which parameters are the potential causes of machine faults (indirect fault causes), namely, find the changes of those parameters, and take measures to avoid faults.

Disclosure of Invention

In order to solve the defects of the prior art, the invention combines the prior art, and provides a method, an instrument and a system for multi-parameter fusion health diagnosis and fault prediction of a machine from practical application, wherein the health state of the machine is accurately judged, the fault of the machine is early warned, an optimization scheme for machine operation and maintenance is provided, and the fault rate is reduced or the fault is avoided.

The technical scheme of the invention is as follows:

According to an invention, there is provided a machine multi-parameter fusion health diagnosis and fault prediction method comprising the steps of: the method comprises the steps of obtaining fault cause parameters in a machine, wherein the fault cause parameters comprise online oil parameters, online vibration parameters, online temperature parameters and electrical signal parameters, processing fault cause parameter data, comparing the processed fault cause parameter data with corresponding data in a database, outputting a fault diagnosis report and a fault prediction report of the machine based on comparison results, and generating machine operation and maintenance suggestions based on report results.

Further, the online oil parameters include wear value, pollution degree, kinematic viscosity, moisture and aging degree of the oil, the online vibration parameters include misalignment, looseness, run-out, meshing degree, bearing failure frequency and vibration strength of the machine, the online temperature parameters include operation temperature of the machine, and the electrical signal parameters include voltage, current, torque and power of the machine.

Further, the method further comprises the steps of dividing the fault cause parameters into direct fault cause parameters and indirect fault cause parameters based on the influence degree of the fault cause parameters on the machine, realizing fault diagnosis of the machine based on the direct cause fault parameters, and realizing fault prediction of the machine based on the indirect fault cause parameters.

Further, in the database, each direct fault cause parameter comprises four levels of fault state limit values of normal, warning, serious warning and shutdown inspection, the collected direct cause fault parameters and the four fault state limit values are compared, then the fault state warning corresponding to the corresponding parameters is given, and the corresponding fault prediction report is given according to the occurrence frequency of the collected indirect fault cause parameters.

Further, the direct cause fault parameters comprise wear values of online oil parameters, online vibration parameters and electrical signal parameters, and the indirect cause fault parameters comprise pollution degree, kinematic viscosity, moisture, aging degree and online temperature parameters of the online oil parameters.

According to another aspect of the present invention, there is provided a machine multi-parameter fusion health diagnosis and fault prediction apparatus for implementing the machine multi-parameter fusion health diagnosis and fault prediction method described above, comprising: the system comprises an online oil sensor, an online vibration sensor, an online temperature sensor, an electrical signal sensor and an industrial personal computer;

the online oil sensor is used for acquiring online oil parameters of the machine, and data acquired by the online oil sensor enters the industrial personal computer after being processed by the controller;

The on-line vibration sensor is used for collecting on-line vibration parameters of the machine, the data collected by the on-line vibration sensor enters the industrial personal computer after being processed by the vibration signal collecting card,

The online temperature sensor is used for acquiring online temperature parameters of the machine, and data acquired by the online temperature sensor enters the industrial personal computer after being processed by the temperature signal acquisition card;

The electric signal sensor is used for acquiring electric signal parameters of the machine, and data acquired by the electric signal sensor enters the industrial personal computer after being processed by the electric signal acquisition card;

The industrial personal computer processes based on the uploaded data and compares the data with the data in a built-in database, outputs a fault diagnosis report and a fault prediction report of the machine based on the comparison result, and generates machine operation and maintenance advice based on the report result.

Further, the online oil liquid sensor comprises a ferromagnetic particle sensor and/or a metal particle sensor, a kinematic viscosity sensor, a pollution degree sensor, a moisture sensor and an oil quality sensor;

the on-line vibration sensor comprises a displacement sensor, a speed sensor and an acceleration sensor;

The electrical signal sensor comprises a voltage transformer and a current transformer.

Further, the on-line vibration sensor and the on-line temperature sensor are of an integrated structure or a split structure.

According to still another aspect of the present invention, there is provided a machine multi-parameter fusion health diagnosis and fault prediction system for implementing the machine multi-parameter fusion health diagnosis and fault prediction method described above, including:

The acquisition module is configured to acquire fault cause parameters in the machine, wherein the fault cause parameters comprise online oil parameters, online vibration parameters, online temperature parameters and electrical signal parameters;

the transmission module is configured to send the fault cause parameters acquired by the acquisition module to the processing terminal;

The processing terminal is configured to process the data of the fault cause parameters uploaded by the transmission module, compare the fault cause parameters with corresponding data in a built-in database, output a fault diagnosis report and a fault prediction report of the machine based on the comparison result, and generate machine operation and maintenance suggestions based on the report result.

Further, the cloud server and the processing terminal realize data interaction, and the user mobile terminal and the cloud server realize data interaction.

The invention has the beneficial effects that:

1. The invention can diagnose and predict the machine health state through the multi-parameter fusion of the online oil sensor, the online vibration sensor, the temperature sensor and the electric signal sensor, improve the analysis speed and the accuracy, realize the automation on line, judge the machine health state more accurately, early warn the machine fault, give out the optimization proposal of the machine operation and maintenance, reduce the fault rate or avoid the fault occurrence, reduce the machine abrasion and prolong the service life of the machine. The machine fault can be early warned and judged 1-3 months in advance.

2. According to the invention, the collected data is compared with the data prestored in the database, so that the diagnosis and early warning of the machine health state and faults can be more accurately carried out, and meanwhile, the database can be continuously optimized through the collected data, so that the diagnosis and prediction accuracy is further improved.

3. According to the invention, the parameters are divided into the direct fault cause parameters and the indirect fault cause parameters through reasonable classification of the parameters, so that the machine fault diagnosis and prediction can be more accurate and quicker.

Drawings

FIG. 1 is a flow chart of example 1;

FIG. 2 is a diagram of the structure of embodiment 2;

Fig. 3 is a structural diagram of embodiment 3.

Detailed Description

The application will be further described with reference to the accompanying drawings and specific embodiments. It is to be understood that these examples are illustrative of the present application and are not intended to limit the scope of the present application. Further, it will be understood that various changes and modifications may be made by those skilled in the art after reading the teachings of the application, and equivalents thereof fall within the scope of the application as defined by the claims.

Example 1:

the embodiment provides a machine multi-parameter fusion health diagnosis and fault prediction method.

The method mainly comprises the steps of merging a plurality of key operation parameters to judge the health state of the machine more accurately, early warning the machine fault, providing an optimization scheme for machine operation and maintenance, and reducing the fault rate or avoiding the fault.

Referring to fig. 1, a related flow chart of the present method is shown. The method for machine multi-parameter fusion health diagnosis and fault prediction mainly comprises the following steps: the method comprises the steps of obtaining fault cause parameters in a machine, wherein the fault cause parameters comprise online oil parameters, online vibration parameters, online temperature parameters and electrical signal parameters, processing fault cause parameter data, comparing the processed fault cause parameter data with corresponding data in a database, outputting a fault diagnosis report and a fault prediction report of the machine based on comparison results, and generating machine operation and maintenance suggestions based on report results.

During actual use of the machine, it is mainly the gearbox drive system, turbine, motor, etc. that need lubrication or rotation that need to be monitored. The cause of machine failure is a variety of parameters including design, assembly, installation, material, operating conditions, operating errors, looseness, misalignment, unbalance, etc., poor lubrication (lubricant failure, lubrication contamination), wear, component failure, high temperature, overload, etc. Problems such as design, assembly, installation, and materials are generally exposed at the initial stage of machine operation, so that the failure cause parameters related to long-term operation of the machine are focused on in the scheme of the embodiment.

In this embodiment, the oil parameters monitored online mainly include wear, pollution level, kinematic viscosity, moisture, aging level, etc., and the vibration parameters monitored online are: failure characteristics and vibration intensity such as misalignment, looseness, meshing degree, bearing failure frequency and the like, and on-line monitoring temperature parameters are as follows: the electrical signal parameters of the temperature on-line monitoring are as follows: voltage, current, torque, power. The oil parameters can mainly raise the condition of lubricating oil, further judge the running condition of equipment, the temperature is an important parameter for machine running, early warning of machine faults can be achieved, and voltage, current and power signals can reflect the electrical performance, turn-to-turn short circuit, stator looseness, uneven air gap, machine locked-rotor, frequent start-stop, illegal operation and the like of a motor. And comprehensively analyzing and judging the fault condition of the machine equipment in the multi-parameter fusion mode.

In the process of analyzing and processing each parameter, the embodiment is firstly divided into a direct fault reason parameter and an indirect fault reason parameter according to the direct association degree of the fault reason parameter to the machine fault. The specific classification is as follows: the direct cause fault parameters comprise wear values, online vibration parameters and electrical signal parameters of online oil parameters, and the indirect cause fault parameters comprise pollution degree, kinematic viscosity, moisture, aging degree and online temperature parameters of the online oil parameters. For the failure of a machine caused by a direct-cause failure parameter, the degradation of the direct-cause failure parameter is accelerated when the occurrence frequency of an indirect-cause failure parameter is high, and then the failure of the machine is generated, so that the failure of the machine is predicted and diagnosed, and the failure cause parameters need to be comprehensively analyzed.

The machine fault diagnosis method based on the above classified parameters is as follows: the direct fault cause parameters determine the diagnosis of the severity of the fault. In the database of the industrial personal computer, each direct fault cause parameter has four levels of fault state limit values of normal, warning, serious warning and stop inspection. Software in the industrial personal computer can compare the collected direct fault cause parameter monitoring value with four fault state limit values, and then gives out a fault state warning of the corresponding parameter. The machine fault is predicted as follows: and early warning faults according to the indirect fault cause parameters. The indirect fault causes include: the viscosity of the oil motion, the attenuation of oil wear-resistant additives (oil aging), the pollution degree, overload, illegal operation and the like. Such as reduced kinematic viscosity, thinning of the oil film, increased chance of contact of the friction pair metal, increased wear, such as frequent locked-up starts (illegal operation), and failure of machine parts (tooth breakage, or abnormal wear), resulting in corresponding vibrations. These parameters do not directly lead to machine failure, are potential factors for machine failure, and if they occur more frequently, can facilitate direct failure cause parameter generation.

In the above method provided in this embodiment, the calculation method may not only be limited to calculating with a single parameter, but also include related algorithms such as DS evidence theory, neural network, fuzzy decision theory, etc. that take as input all direct fault cause and indirect fault cause parameters, and take as output fault prediction, fault severity, maintenance advice, and operation optimization scheme.

Example 2

The embodiment provides a machine multi-parameter fusion health diagnosis and fault prediction instrument, which is used for more accurately judging the health state of a machine, early warning the fault of the machine and giving an optimization scheme of machine operation and maintenance by fusing a plurality of key operation parameters through configuring the related method of the embodiment 1, so that the fault rate is reduced or the fault is avoided.

Referring to fig. 2, the present apparatus mainly includes a plurality of sensors, a controller, an industrial personal computer, and the like.

The sensor comprises an online oil sensor, an online vibration sensor, an online temperature sensor and an electrical signal sensor, wherein the online oil sensor is used for collecting online oil parameters of a machine, data collected by the online oil sensor are processed by a controller and then enter an industrial personal computer, and the online oil sensor comprises a ferromagnetic particle sensor and/or a metal particle sensor, a kinematic viscosity sensor, a pollution degree sensor, a moisture sensor and an oil quality sensor and is mainly used for collecting abrasion, pollution degree, kinematic viscosity, moisture, aging degree and the like in oil, and the signals enter the controller and then enter the industrial personal computer from the controller. The on-line vibration sensor is used for collecting on-line vibration parameters of the machine and comprises a displacement sensor, a speed sensor and an acceleration sensor, signals of the sensors enter a vibration signal collecting card and then enter an industrial personal computer, software in the industrial personal computer processes vibration time domain signals, and vibration intensity and characteristic frequency signals are extracted for later diagnosis processing. Component failure may be monitored, for example, by vibration analysis. The on-line temperature sensor is used for collecting on-line temperature parameters of the machine, the temperature signal is an important parameter of machine operation, machine faults can be early warned, data collected by the on-line temperature sensor enters the industrial personal computer after being processed by the temperature signal collecting card, and different temperature sensors can be used for temperature monitoring according to the monitoring temperature range. In a preferred embodiment provided in this embodiment, the online temperature sensor and the online vibration sensor may be integrally provided. The electric signal sensor is used for collecting electric signal parameters of the machine, and comprises a voltage transformer and a current transformer, wherein signals of the voltage transformer and the current transformer enter the collecting card to calculate signals such as voltage, current, torque and power, and the signals enter the industrial personal computer to be analyzed; the voltage, current and power signals can reflect the electrical performance, turn-to-turn short circuit, stator looseness, uneven air gap, machine locked-rotor, frequent start-stop and illegal operation and the like of the motor.

In this embodiment, the industrial personal computer is mainly used for analyzing and processing the uploaded data, a database is built in the industrial personal computer, the multi-parameter signals are comprehensively processed, analyzed and stored in the industrial personal computer, and are subjected to fusion processing with the machine health characteristic limit value and the operation rule database to generate a machine health state diagnosis report, and meanwhile, the parameters are returned to the database for storage, and a self-learning optimization database can be also arranged; maintenance recommendations are generated based on the diagnostic report to normalize the operation and guide the maintenance operations.

Example 3:

the embodiment provides a machine multi-parameter fusion health diagnosis and fault prediction system, which reduces the fault rate or avoids fault occurrence by configuring the related method of the embodiment 1, and by fusing a plurality of key operation parameters to more accurately judge the health state of a machine, early warn the fault of the machine, and provide an optimization scheme for machine operation and maintenance.

As described with reference to fig. 3, the present embodiment mainly includes an acquisition module, a transmission module, a processing terminal, a cloud server, a user mobile terminal, and the like. The acquisition module is mainly used for acquiring fault cause parameters in the machine, wherein the fault cause parameters comprise online oil parameters, online vibration parameters, online temperature parameters and electrical signal parameters, and particularly comprises corresponding sensors, acquisition cards and the like. The transmission module is mainly used for sending the fault reason parameters acquired by the acquisition module to the processing terminal in a wired or wireless mode, the processing terminal is mainly used for carrying out data processing on the fault reason parameters uploaded by the transmission module, comparing the fault reason parameters with corresponding data in a built-in database, outputting a fault diagnosis report and a fault prediction report of the machine based on the comparison result, and generating machine operation and maintenance advice based on the report result.

Furthermore, in this embodiment, a cloud server and a user mobile terminal are further provided, data interaction is achieved between the cloud server and the processing terminal through a wireless transmission mode, data storage and analysis processing can be achieved in the cloud server, and a user can interact with the cloud server through the user mobile terminal, so that a fault diagnosis report, a fault prediction report, operation and maintenance advice and the like of a machine can be queried through a mobile phone.

Claims (10)

1. The multi-parameter fusion health diagnosis and fault prediction method for the machine is characterized by comprising the following steps of: the method comprises the steps of obtaining fault cause parameters in a machine, wherein the fault cause parameters comprise online oil parameters, online vibration parameters, online temperature parameters and electrical signal parameters, processing fault cause parameter data, comparing the processed fault cause parameter data with corresponding data in a database, outputting a fault diagnosis report and a fault prediction report of the machine based on comparison results, and generating machine operation and maintenance suggestions based on report results.

2. The machine multi-parameter fusion health diagnostic and fault prediction method of claim 1, wherein the on-line oil parameters include oil wear value, pollution level, kinematic viscosity, moisture, aging level, the on-line vibration parameters include machine misalignment, looseness, run-out, meshing level, bearing failure frequency, and vibration strength, the on-line temperature parameters include machine operating temperature, and the electrical signal parameters include machine voltage, current, torque, and power.

3. The machine multi-parameter fusion health diagnosis and fault prediction method according to claim 1 or 2, further comprising classifying the fault cause parameters into direct fault cause parameters and indirect fault cause parameters based on the degree of influence of the fault cause parameters on the machine, implementing fault diagnosis of the machine based on the direct cause fault parameters, and implementing fault prediction of the machine based on the indirect fault cause parameters.

4. A machine multi-parameter fusion health diagnosis and fault prediction method according to claim 3, wherein each direct fault cause parameter comprises four levels of fault state limit values of normal, warning, serious warning and shutdown inspection in a database, the collected direct cause fault parameters and the four fault state limit values are compared, then the fault state warning corresponding to the corresponding parameters is given, and the corresponding fault prediction report is given according to the occurrence frequency of the collected indirect fault cause parameters.

5. The machine multi-parameter fusion health diagnosis and fault prediction method according to claim 4, wherein the direct cause fault parameters comprise wear values of on-line oil parameters, on-line vibration parameters, electrical signal parameters, and the indirect cause fault parameters comprise pollution degree, kinematic viscosity, moisture, aging degree and on-line temperature parameters of the on-line oil parameters.

6. A machine multi-parameter fusion health diagnosis and fault prediction apparatus for implementing the machine multi-parameter fusion health diagnosis and fault prediction method according to any one of claims 1 to 5, characterized by comprising: the system comprises an online oil sensor, an online vibration sensor, an online temperature sensor, an electrical signal sensor and an industrial personal computer;

the online oil sensor is used for acquiring online oil parameters of the machine, and data acquired by the online oil sensor enters the industrial personal computer after being processed by the controller;

The on-line vibration sensor is used for collecting on-line vibration parameters of the machine, the data collected by the on-line vibration sensor enters the industrial personal computer after being processed by the vibration signal collecting card,

The online temperature sensor is used for acquiring online temperature parameters of the machine, and data acquired by the online temperature sensor enters the industrial personal computer after being processed by the temperature signal acquisition card;

The electric signal sensor is used for acquiring electric signal parameters of the machine, and data acquired by the electric signal sensor enters the industrial personal computer after being processed by the electric signal acquisition card;

The industrial personal computer processes based on the uploaded data and compares the data with the data in a built-in database, outputs a fault diagnosis report and a fault prediction report of the machine based on the comparison result, and generates machine operation and maintenance advice based on the report result.

7. The machine multi-parameter fusion health diagnostic and fault prediction instrument of claim 6, wherein the online oil sensor comprises a ferromagnetic particle sensor and/or a metallic particle sensor, a kinematic viscosity sensor, a contamination level sensor, a moisture sensor, an oil quality sensor;

the on-line vibration sensor comprises a displacement sensor, a speed sensor and an acceleration sensor;

The electrical signal sensor comprises a voltage transformer and a current transformer.

8. The machine multi-parameter fusion health diagnostic and fault prediction instrument of claim 6, wherein the on-line vibration sensor and the on-line temperature sensor are in a one-piece integrated structure or a split structure.

9. A machine multi-parameter fusion health diagnosis and fault prediction system for implementing the machine multi-parameter fusion health diagnosis and fault prediction method according to any one of claims 1 to 5, characterized by comprising:

The acquisition module is configured to acquire fault cause parameters in the machine, wherein the fault cause parameters comprise online oil parameters, online vibration parameters, online temperature parameters and electrical signal parameters;

the transmission module is configured to send the fault cause parameters acquired by the acquisition module to the processing terminal;

The processing terminal is configured to process the data of the fault cause parameters uploaded by the transmission module, compare the fault cause parameters with corresponding data in a built-in database, output a fault diagnosis report and a fault prediction report of the machine based on the comparison result, and generate machine operation and maintenance suggestions based on the report result.

10. The machine multi-parameter fusion health diagnosis and fault prediction system of claim 9 further comprising a cloud server and a user mobile terminal, the cloud server implementing data interactions with the processing terminal, the user mobile terminal implementing data interactions with the cloud server.