CN203643114U - An Intelligent Fault Diagnosis System Based on Immune Detector - Google Patents

Abstract

The utility model provides an intelligent fault diagnosis system based on an immune detector comprising a vibration acquisition device. A plurality of vibration sensors are respectively disposed on corresponding parts of a to-be-detected set. Each of the vibration sensors comprises a permanent magnet pedestal, a vibration cavity disposed in the permanent magnet pedestal, a vibration sensing unit disposed on the bottom part of the vibration cavity, and a point positioning device used for locating the vibration sensor. A processing device is used for acquiring signals acquired by the vibration acquisition device, and can be used for the corresponding control over the set to be detected.

Description

An Intelligent Fault Diagnosis System Based on Immune Detector

technical field

The utility model relates to an intelligent fault diagnosis system based on an immune detector.

Background technique

At present, it is difficult to extract and classify fault features and online diagnosis in the existing diagnosis systems of mechanical fault time domain and frequency domain at home and abroad, and cause safety accidents and heavy losses due to equipment faults.

Contents of the invention

The technical problem to be solved by the utility model is to provide an intelligent fault diagnosis system based on an immune detector, which can effectively solve the problems of inaccurate fault signal classification and long diagnosis time in existing diagnostic devices. The utility model has the function of effectively detecting single failure, composite failure and early failure of industrial units, with an accuracy rate of more than 98%, and can simultaneously detect failures of multiple units.

In this regard, the utility model provides an intelligent fault diagnosis system based on an immune detector, including: a vibration acquisition device, including: a plurality of vibration sensors respectively arranged on corresponding parts of the unit to be detected; the vibration sensors include: a permanent magnet base , a vibration chamber arranged on the base of the permanent magnet, a vibration sensing unit arranged at the bottom of the vibration chamber, and a point positioning device for positioning the vibration sensor; a processing device for collecting the vibration collection The signal collected by the device is used to control the unit to be detected accordingly.

Wherein, the point positioning device refers to, for example, adopting a bolt, the position of the bolt point is fixed, and the vibration sensor can rotate around this point.

The vibration cavity refers to a housing with an inner space, which can not only accommodate the vibration sensing unit, but also effectively collect vibration signals.

Correspondingly, the utility model also provides a fault diagnosis method based on an immune detector-based intelligent fault diagnosis system, including: a vibration signal collection step, wherein the vibration sensor is arranged at a corresponding part of the unit to be detected for collecting vibration information; The feature extraction step is to analyze and process the vibration signal collected by the vibration sensor to obtain fault characteristic parameters; the matching judgment step is to match the fault characteristic parameters in the feature extraction step with the fault data stored in advance in the server; Store in the knowledge base step, store the fault characteristic parameters that fail to match in the matching judgment step into the knowledge base; the immune detector update step, store the newly stored fault characteristic parameters in the described step of storing in the knowledge base into the immune detector , to update the immune detector; the step of diagnosing faults and types, displaying the successful matching results in the matching judging step in the monitoring computer.

Based on the immune detector, it refers to the interrelated machine parts that are widely used in the operation table of the production workshop. The evolution to obtain the optimal antibody, get the new feature index of the best recognition ability, and solve the problem of insufficient classification ability.

As the core of the immune detector, the utility model mainly acquires various fault characteristic information of interrelated machine parts widely used in the operation desks of the production workshop. How to effectively obtain various fault feature information is one of the keys to the effectiveness of the intelligent fault diagnosis system based on immune detectors.

Generally speaking, whether the machine is faulty or not is highly related to the vibration of its various parts. The measurement parameter set by the utility model is vibration. This measurement requires the selection of the most sensitive diagnostic parameters for fault reflection. This parameter is called "sensitive factor", which means that when the machine state changes slightly, the characteristic parameters change greatly.

The processing device can construct an efficient and fast immune detector conforming to the principle of diversity and distribution of the artificial immune system, obtain the optimal antibody through the evolution of the antibody set, and obtain the new characteristic index of the best recognition ability, Solve the problem of insufficient classification ability.

The system can be used by embedding it on the operation table of the production workshop, which is convenient and easy to implement, and can realize the monitoring and diagnosis of multiple industrial units at the same time. From the perspective of system operation, it can effectively monitor and diagnose unit failures and ensure the normal operation of equipment.

Requirements for selecting and arranging measuring points: ①The points sensitive to vibration response should be as close as possible to the vibration source, avoiding or reducing the interface, cavity or spacer (such as sealing packing, etc.) of the signal on the transmission channel, and it is best to let the signal propagate in a straight line , to reduce the energy loss of the signal during transmission. ②Select the part where the vibration signal is more concentrated, so as to obtain more status information. ③ The selected measuring points should be subject to the purpose of diagnosis. If the purpose of diagnosis is different, the positions of the measuring points should be changed accordingly. ④The measuring point must have enough space to place the sensor and ensure good contact. The measuring point should also have sufficient rigidity. ⑤Because the on-site vibration measurement is carried out when the equipment is running, the safety of personnel and equipment must be ensured when installing the sensor. Bearings are usually the preferred measurement points. In addition, the foundations, casings, cylinders, inlet and outlet pipes, valves, foundations and other parts of the equipment are also permanent measurement points for vibration.

By adopting the above technical solution, the permanent magnet base can be conveniently connected with the measured part, and the installation of the point-of-use positioning device is also very simple and convenient for installation. More importantly, experiments have shown that using a permanent magnet base to connect the measured part is conducive to more effective measurement of vibration. However, in a long-term vibration environment, the permanent magnet base will be displaced, which will affect the monitoring accuracy. Therefore, a point positioning device is added at the same time, which can make the vibration sensor not deviate too far within a certain range, reducing maintenance costs. At the same time, the range also becomes a monitoring parameter that can reflect the working condition of the machine. Tests have shown that better detection results can be obtained by using the vibrating cavity.

Preferably, the vibration sensor is an acceleration sensor.

Preferably, the frequency range of the acceleration sensor is 10 to 100 KHz.

With the above technical solution, since the vibration measurement requires the selection of the most sensitive diagnostic parameters for fault reflection, this parameter is called "sensitivity factor", that is, when the machine state changes slightly, the characteristic parameters change greatly. . Therefore it is impossible to determine a sensitivity factor for each fault signal. The diagnostic parameters are selected according to the frequency characteristics of the vibration signal of the diagnostic object. Commonly used vibration measurement parameters include acceleration, velocity, and displacement. Generally, they are selected according to the following principles: low-frequency vibration <100Hz adopts displacement; medium-frequency vibration 10-1000Hz adopts velocity; high-frequency vibration> 1000Hz uses acceleration. The frequency response range of the acceleration sensor is relatively wide at 10KHz, so when there is no special requirement for on-site measurement, it is often used to measure the three parameters of displacement, velocity and acceleration at the same time, which can basically meet the requirements.

Preferably, the processing device includes: a server connected to the vibration collection device for storing fault data collected by the vibration collection device; and a monitoring computer connected to the server for displaying the vibration data collected by the vibration collection device Collected failure data.

By adopting the above-mentioned technical scheme, the system has effective online detection and online diagnosis of single faults, complex faults and early faults in industrial units, and effectively solves the problem of long diagnosis time in the existing diagnostic system. the

Compared with the prior art, the utility model has the advantage that the utility model has the functions of online monitoring and online diagnosis, and can effectively monitor and diagnose the single failure, compound failure and early failure of the unit, and the diagnostic accuracy rate is 98%. % or more, and has the characteristics of online, real-time, and rapidity; the system can be used by embedding it on the operating table of the production workshop, which is convenient and easy to implement, and can monitor and diagnose multiple industrial units at the same time. From the perspective of system operation, it can effectively monitor and diagnose unit failures and ensure the normal operation of equipment.

Description of drawings

Fig. 1 is the schematic diagram of an embodiment of the intelligent fault diagnosis system based on the immune detector of the present invention;

Fig. 2 is the schematic diagram of another embodiment of the intelligent fault diagnosis system based on the immune detector of the present invention;

Fig. 3 is the schematic diagram of another embodiment of the intelligent fault diagnosis system based on the immune detector of the present invention;

Fig. 4 is a schematic diagram of an embodiment of the intelligent fault diagnosis method based on the immune detector of the present invention.

Detailed ways

Below in conjunction with accompanying drawing, preferred embodiment of the present utility model is described in further detail:

As shown in Figures 1 and 2, the vibration acquisition device 1 includes: a plurality of vibration sensors 11 that are respectively arranged on the corresponding parts of the unit to be detected; Vibration cavity 112 on the top, vibration sensing unit 113 arranged at the bottom of the vibration cavity 112, and a point positioning device 114 for positioning the vibration sensor 11; processing device 2, used to collect the vibration collection device 1 The collected signals are used to control the unit to be detected accordingly.

Wherein, the point positioning device 114 refers to, for example, adopting a bolt, the position of the bolt point is fixed, and the vibration sensor can rotate around the point.

The vibration cavity 112 refers to a housing with an inner space, which can not only accommodate the vibration sensing unit 113, but also effectively collect vibration signals.

Based on the immune detector, it refers to the interrelated machine parts that are widely used in the operation table of the production workshop. The evolution to obtain the optimal antibody, get the new feature index of the best recognition ability, and solve the problem of insufficient classification ability.

As the core of the immune detector, the utility model mainly acquires various fault characteristic information of interrelated machine parts widely used in the operation desks of the production workshop. How to effectively obtain various fault feature information is one of the keys to the effectiveness of the intelligent fault diagnosis system based on immune detectors.

Generally speaking, whether the machine is faulty or not is highly related to the vibration of its various parts. The measurement parameter set by the utility model is vibration. This measurement requires the selection of the most sensitive diagnostic parameters for fault reflection. This parameter is called "sensitive factor", which means that when the machine state changes slightly, the characteristic parameters change greatly.

The processing device 2 can construct an efficient and fast immune detector conforming to the principle of diversity and distribution of the artificial immune system, obtain the optimal antibody through the evolution of the antibody set, and obtain new characteristic indicators of the best recognition ability , to solve the problem of insufficient classification ability.

The system can be used by embedding it on the operation table of the production workshop, which is convenient and easy to implement, and can realize the monitoring and diagnosis of multiple industrial units at the same time. From the perspective of system operation, it can effectively monitor and diagnose unit failures and ensure the normal operation of equipment.

Requirements for selecting and arranging measuring points: ①The points sensitive to vibration response should be as close as possible to the vibration source, avoiding or reducing the interface, cavity or spacer (such as sealing packing, etc.) of the signal on the transmission channel, and it is best to let the signal propagate in a straight line , to reduce the energy loss of the signal during transmission. ②Select the part where the vibration signal is more concentrated, so as to obtain more status information. ③ The selected measuring points should be subject to the purpose of diagnosis. If the purpose of diagnosis is different, the positions of the measuring points should be changed accordingly. ④The measuring point must have enough space to place the sensor and ensure good contact. The measuring point should also have sufficient rigidity. ⑤Because the on-site vibration measurement is carried out when the equipment is running, the safety of personnel and equipment must be ensured when installing the sensor. Bearings are usually the preferred measurement points. In addition, the foundations, casings, cylinders, inlet and outlet pipes, valves, foundations and other parts of the equipment are also permanent measurement points for vibration.

By adopting the above technical solution, since the permanent magnet base 111 can be conveniently connected with the measured part, and the installation of the use point positioning device 114 is also very simple and convenient for installation. More importantly, experiments have shown that using the permanent magnet base 111 to connect the parts to be measured is conducive to more effective vibration measurement. However, in the environment of long-term vibration, the permanent magnet base 111 will produce displacement, which will affect the monitoring accuracy. Therefore, a point positioning device 114 is added at the same time, so that the vibration sensor 11 will not deviate too far within a certain range. Maintenance cost, at the same time, the range also becomes a monitoring parameter, which can reflect the working condition of the machine. Experiments have shown that using the vibrating cavity 112 can obtain better detection results.

Preferably, the vibration sensor 11 is an acceleration sensor.

Preferably, the frequency range of the acceleration sensor is 10 to 100 KHz.

With the above technical solution, since the vibration measurement requires the selection of the most sensitive diagnostic parameters for fault reflection, this parameter is called "sensitivity factor", that is, when the machine state changes slightly, the characteristic parameters change greatly. . Therefore it is impossible to determine a sensitivity factor for each fault signal. The diagnostic parameters are selected according to the frequency characteristics of the vibration signal of the diagnostic object. Commonly used vibration measurement parameters include acceleration, velocity, and displacement. Generally, they are selected according to the following principles: low-frequency vibration <100Hz adopts displacement; medium-frequency vibration 10-1000Hz adopts velocity; high-frequency vibration> 1000Hz uses acceleration. The frequency response range of the acceleration sensor is relatively wide at 10KHz, so when there is no special requirement for on-site measurement, it is often used to measure the three parameters of displacement, velocity and acceleration at the same time, which can basically meet the requirements.

As shown in Figure 3, described processing device 2 comprises: server 21, is connected with described vibration acquisition device 1, is used for storing the failure data that described vibration acquisition device 1 collects; And monitoring computer 22, is connected with described server 21 , for displaying the fault data collected by the vibration collection device 1 .

By adopting the above-mentioned technical scheme, the system has effective online detection and online diagnosis of single faults, complex faults and early faults in industrial units, and effectively solves the problem of long diagnosis time in the existing diagnostic system.

As shown in Figure 4, adopt the fault diagnosis method of described intelligent fault diagnosis system based on immune detector, comprise:

Vibration signal collection step, the vibration sensor 11 is arranged on the corresponding part of the unit to be detected, for collecting vibration information;

The feature extraction step is to analyze and process the vibration signal collected by the vibration sensor 11 to obtain fault characteristic parameters;

Matching judging step, matching the fault characteristic parameter in the feature extraction step with the fault data pre-stored in the server 21;

Store in the knowledge base step, store the fault characteristic parameters of the matching failure in the matching judgment step into the knowledge base;

The immune detector update step is to store the newly stored fault characteristic parameters in the step of storing in the knowledge base to the immune detector to update the immune detector; the step of diagnosing faults and types is to store the successful matching results in the matching judgment step Displayed in the monitoring computer 22.

The above content is a further detailed description of the utility model in combination with specific preferred embodiments, and it cannot be assumed that the specific implementation of the utility model is only limited to these descriptions. For a person of ordinary skill in the technical field to which the utility model belongs, without departing from the concept of the utility model, some simple deduction or substitutions can also be made, which should be regarded as belonging to the protection scope of the utility model.

Claims (4)

1. the Intelligent Fault Diagnose Systems based on immune detectors, is characterized in that, comprising:

Vibration acquisition device (1), comprising: multiple vibration transducers (11) that are separately positioned on unit corresponding site to be detected; Described vibration transducer (11) comprising: permanent magnet base (111), is arranged on vibration cavity (112) on described permanent magnet base (111), is arranged on the vibrating sensing unit (113) of described vibration cavity (112) bottom, and for locating the location device (114) of this vibration transducer (11);

Treating apparatus (2), for collecting the collected signal of described vibration acquisition device (1), and controls unit to be detected accordingly accordingly.

2. the Intelligent Fault Diagnose Systems based on immune detectors as claimed in claim 1, is characterized in that, described vibration transducer (11) adopts acceleration transducer.

3. the Intelligent Fault Diagnose Systems based on immune detectors as claimed in claim 2, is characterized in that, described acceleration transducer frequency range is 10 to 100KHz.

4. the Intelligent Fault Diagnose Systems based on immune detectors as claimed in claim 1, it is characterized in that, described treating apparatus (2) comprising: server (21), is connected the fault data gathering for storing described vibration acquisition device (1) with described vibration acquisition device (1); And supervisory control comuter (22), be connected with described server (21), for the fault data that shows that described vibration acquisition device (1) gathers.

Images