Control system and method for laser Doppler vibrometer

Abstract

The present invention discloses a control system and method for laser Doppler vibration measuring equipment, which relates to the field of vibration measuring equipment control technology. The emergency processing strategy library outputs the corresponding emergency processing strategy, and the equipment is processed after the emergency processing strategy is executed; the operation value is constructed according to the feedback data of the first-level alarm instruction received, and if the operation value exceeds the expectation, a second-level alarm instruction is issued to the outside, and the key components of the equipment are maintained at the time node that meets the constraint conditions; the trained equipment fault diagnosis model is used to perform fault diagnosis, and if the equipment has an operation fault, the equipment fault maintenance knowledge map provides a maintenance plan according to the fault characteristics, the key components that cause the fault are automatically switched, and the measurement data is corrected, and the actual measurement value and the measurement correction value are output. On the basis of controlling the equipment, the existing faults are targeted for maintenance according to the maintenance plan.

Description

Technical Field

[0001] The invention relates to the technical field of vibration measuring equipment control, in particular to a control system and method for laser Doppler vibration measuring equipment. Background Art

[0002] Laser Doppler vibrometer is an advanced measuring instrument that integrates the laser Doppler effect and optical heterodyne interferometry technology. It uses a laser beam to illuminate the surface of the object to be measured, and detects the slight changes in the frequency of the reflected laser (i.e., Doppler frequency shift) to measure the object's dynamic parameters such as vibration velocity, displacement, and acceleration in a non-contact and high-precision manner.

[0003] The device works on the basis of the Doppler effect, which states that the frequency of light waves received by an observer changes when there is relative motion between the light source and the observer. In laser Doppler vibrometry, a laser is used as the light source, and the vibrations of the object are seen as tiny movements relative to the laser light source. By precisely measuring this frequency change, the device is able to calculate the vibration characteristics of the object.

[0004] Laser Doppler vibrometers have several significant advantages. First, they can measure at long distances without direct contact with the object being measured, thus avoiding the errors and interference that may be caused by traditional contact measurements. Second, the device has high spatial resolution and can accurately capture tiny vibrations on the surface of an object. In addition, its fast response characteristics and wide-band measurement capabilities enable the device to accurately capture the dynamic vibration information of an object in real time.

[0005] In the Chinese invention patent with application publication number CN117242412A, a fault diagnosis method and a fault diagnosis device are disclosed. The method includes: after the fault diagnosis device establishes a wireless connection with the first electrical device, the fault diagnosis device sends a first instruction to the first electrical device, and the first instruction is used to request to read the fault information of the first battery on the first electrical device; the fault diagnosis device receives the fault information of the first battery sent by the first electrical device based on the first instruction; the fault diagnosis device parses the fault information of the first battery, and controls the output device of the fault diagnosis device to output the current fault of the first battery to the diagnostic personnel. The method and fault diagnosis device provided in the embodiment of the present application can improve the convenience of the diagnostic personnel's operation, save fault diagnosis time, and improve efficiency.

[0006] Combined with the above application and the contents of the prior art:

[0007] When the laser Doppler vibrometer is running, it is necessary to monitor the operating status of the equipment in real time. When some control parameters are abnormal, the abnormal control parameters are optimized and adjusted through the control model to avoid equipment operation failures and enable the equipment to operate normally.

[0008] In the existing control method for laser Doppler vibrometer equipment, corresponding qualified intervals are usually set for various control parameters of the equipment, and the measurement results output by the equipment are used as feedback. When the reliability of the output measurement data is insufficient, the various control parameters of the equipment are corrected and controlled. However, when the operating environment of the equipment is poor or even abnormal, such as when there is a lot of dust, too strong or too weak light, the external operating conditions will cause great interference to the accuracy of the measurement. If this is used as feedback to control and adjust the equipment, or even to correct the measurement results, it may be difficult to achieve the expected effect due to poor targeting.

[0009] To this end, the present invention provides a control system and method for a laser Doppler vibrometer. Summary of the invention

[0010] 1. Technical issues to be resolved

[0011] In view of the deficiencies of the prior art, the present invention provides a control system and method for laser Doppler vibrometer equipment, wherein the emergency handling strategy library outputs the corresponding emergency handling strategy, and the equipment is processed after the emergency handling strategy is executed; an operating value is constructed based on the feedback data of the first-level alarm instruction received, and if the operating value exceeds the expectation, a second-level alarm instruction is issued to the outside, and the key components of the equipment are maintained at the time node that meets the constraint conditions; the trained equipment fault diagnosis model is used for fault diagnosis, and if the equipment has an operating fault, a maintenance plan is provided by the equipment fault maintenance knowledge map based on the fault characteristics, the key components that cause the fault are automatically switched, and the measurement data is corrected, and the actual measurement value and the measurement correction value are output. On the basis of controlling the equipment, the existing faults are targeted for maintenance according to the maintenance plan; thereby solving the technical problems recorded in the background technology.

[0012] (II) Technical solution

[0013] To achieve the above objectives, the present invention is implemented through the following technical solutions:

[0014] A control method for a laser Doppler vibrometer device includes taking environmental condition data in an operating area detected by a sensor network as input, adjusting the operating parameters of the device in real time, and generating a signal condition value from the signal quality data of the device. , if the signal condition value If the signal is lower than the threshold, a first-level alarm command will be issued to the outside;

[0015] If there are abnormal environmental conditions in the operating area, the emergency treatment strategy library will output the corresponding emergency treatment strategy based on the key factors affecting the operation of the equipment, and the equipment will be processed after the emergency treatment strategy is executed;

[0016] Build operating values ​​based on feedback data received from the first-level alarm command , if the running value If it exceeds expectations, a secondary alarm command will be issued to the outside, and key components of the equipment will be maintained at a time point that meets the constraints;

[0017] If the measurement value output by the equipment is not reliable enough, after recording the equipment's operating data in real time, use the trained equipment fault diagnosis model to perform fault diagnosis. If the equipment has an operating fault, extract the corresponding fault features.

[0018] The equipment fault maintenance knowledge graph provides a maintenance plan based on the fault characteristics, automatically switches the key components that cause the fault, and uses the trained measurement data correction model to correct the measurement data, and outputs the actual measurement value and the measurement correction value.

[0019] Furthermore, a sensor network is deployed in the operation area, and the sensor network detects and obtains corresponding environmental condition data, which are aggregated to generate an environmental condition data set; the real-time environmental condition data in the operation area is used as input, and the trained equipment automatic control model is used to adjust the equipment operation parameters in real time;

[0020] Obtain environmental condition parameters and corresponding signal quality data in the operating area, including signal-to-noise ratio, signal strength, frequency response, and phase accuracy, etc., summarize and generate a signal quality data set, and generate a signal condition value from the signal quality data set .

[0021] Furthermore, the signal condition value The generation method is as follows: , frequency response and phase accuracy Perform linear normalization and map the corresponding data values ​​to the interval Then follow the following method:

[0022] ;

[0023] in, , is the number of time nodes, For the The signal strength at each time node, For the The frequency response of each time node, For the Phase accuracy of time nodes; weight coefficient: , ,and , is the correction factor.

[0024] Furthermore, after receiving the first-level alarm command, the continuously acquired environmental condition data is used as input, and the trained abnormal data detection model is used to perform abnormality detection, and a reminder command is issued to the outside when the environmental condition is abnormal;

[0025] Through online retrieval or offline formulation, several equipment emergency response strategies for abnormal environmental conditions are collected in advance, and then an emergency response strategy library is generated.

[0026] Furthermore, historical environmental condition data and historical equipment operation data in the operating area are collected, and several key factors affecting equipment operation in environmental conditions are identified through principal component analysis;

[0027] After receiving the reminder instruction, the emergency handling strategy library outputs the corresponding emergency handling strategy according to the correspondence between the key factors and the emergency handling strategy.

[0028] Furthermore, when the number of consecutive alarm instructions received within the preset alarm period exceeds the expected number, the operating value is constructed based on the status data of the received first-level alarm instructions. , as follows:

[0029] ;

[0030] Where: is the conditional signal value at time t, The corresponding signal threshold, For time Conditional signal value on ; is the indicator function that the signal is below the qualified threshold, where , and is the weight coefficient, For time arrive time node.

[0031] Furthermore, after receiving the second-level alarm instruction, the key components of the equipment are regularly inspected and maintained within the pre-set prevention cycle, and the inspection and maintenance frequency is restricted. , perform maintenance on key components of the equipment at time nodes that meet the constraints. The constraints are as follows:

[0032] ;

[0033] in, is the number of times the secondary alarm command is received. It is Second level alarm command to The time interval, is the mean of the time interval; the weight coefficient, , ,and .

[0034] Furthermore, a plurality of continuous measurement values ​​are arranged along the time axis to generate a line graph, and a standard line graph is constructed after a standard measurement value is set for each measurement value;

[0035] Perform similarity analysis on the line graph and the standard line graph to obtain corresponding similarity data. If the similarity data is lower than expected, issue a real-time monitoring instruction to the outside.

[0036] Furthermore, after receiving the real-time monitoring instruction, the status monitoring network deployed on the equipment records the equipment's operating status, vibration characteristics and electrical parameters in real time, and summarizes them as a set of equipment operating data. The real-time operating data of the equipment is used as input, and fault diagnosis is performed using the trained equipment fault diagnosis model.

[0037] Furthermore, after recording all environmental condition data, operation data and alarm event data during equipment operation, a comprehensive equipment operation history database is compiled and established; using laser Doppler vibration measurement equipment fault maintenance as the target word, combined with the equipment operation history database, an equipment fault maintenance knowledge graph is pre-constructed.

[0038] Furthermore, the frequency of generating fault maintenance plans Constraints are imposed, and at the time nodes that meet the constraints, the maintenance plan is output to the outside. After selecting several module components with the highest failure rate based on the fault detection data, they are equipped with redundant components and automatically switched when key components fail. The maintenance plan is output to the outside at a certain frequency. The constraints are as follows:

[0039] ;

[0040] Where: is the abnormality of the environmental conditions in the operating area at time t, for to Moment within The running value on is the signal condition value at time t, is the cumulative distribution function of the standard normal distribution, with weight coefficients: , , .

[0041] Furthermore, the environmental condition data, equipment operation data, measurement data, and fault detection data in the operating area are summarized as a sample database, and some data are extracted from the sample database as sample data; a convolutional neural network is trained with the sample data to obtain a trained measurement data correction model, and after the equipment outputs the measurement data, the measured data is corrected using the trained measurement data correction model.

[0042] Furthermore, the total angle and single angle that the laser Doppler device needs to rotate and the acquisition period are preset;

[0043] After the command to start automatic acquisition is initiated, the trained automatic control model controls the horizontal and vertical rotating brackets to drive the laser Doppler device to rotate at a constant speed. When the laser Doppler device rotates to a preset single angle, the laser Doppler device is started to collect data.

[0044] After the laser Doppler device has collected vibration data of a detection point, the automatic control model controls the laser Doppler device to continue to operate, and the laser Doppler device collects vibration data within a preset total angle, at a preset and single angle, and at a collection cycle interval;

[0045] When the preset total angle is reached or the instruction to end the automatic collection is initiated in the middle of the operation, the automatic control model controls the automatic tracking pan-tilt head to stop rotating, and the laser Doppler device stops collecting.

[0046] A control system for laser Doppler vibrometer equipment includes a signal control unit that uses environmental condition data in an operating area detected by a sensor network as input, adjusts the operating parameters of the equipment in real time, and generates a signal condition value from the signal quality data of the equipment , if the signal condition value If the signal is lower than the threshold, a first-level alarm command will be issued to the outside;

[0047] Equipment emergency processing unit: if there are abnormal environmental conditions in the operating area, the emergency processing strategy library will output the corresponding emergency processing strategy based on the key factors affecting the operation of the equipment, and the equipment will be processed after the emergency processing strategy is executed;

[0048] Control maintenance unit, build operation value according to feedback data received from first level alarm command , if the running value If it exceeds expectations, a secondary alarm command will be issued to the outside, and key components of the equipment will be maintained at a time point that meets the constraints;

[0049] The fault diagnosis unit, if the reliability of the measured value output by the equipment is insufficient, records the equipment's operating data in real time and uses the trained equipment fault diagnosis model to perform fault diagnosis. If the equipment has an operating fault, the corresponding fault features are extracted;

[0050] The correction unit provides a maintenance plan based on the equipment fault maintenance knowledge graph according to the fault characteristics, automatically switches the key components that cause the fault, and uses the trained measurement data correction model to correct the measurement data, and outputs the actual measurement value and the measurement correction value.

[0051] (III) Beneficial effects

[0052] The present invention provides a control system and method for laser Doppler vibrometer equipment, which has the following beneficial effects:

[0053] 1. Use environmental condition data as feedback to adaptively control the control parameters of the equipment, so that the operating parameters of the equipment are adapted to the environmental conditions, and reduce the impact of environmental conditions on equipment operation and signal quality; record and count the signal quality of the equipment and construct the signal condition value accordingly. , evaluate the signal quality of the equipment, determine whether there is any abnormality in the signal quality of the equipment, adjust and control the equipment, and improve the reliability of the equipment.

[0054] 2. Verify whether there are any abnormalities in the current environmental conditions. If there are any abnormalities in the environmental conditions, issue reminders to the outside in real time to avoid further impact and interference of the environmental conditions on the equipment; provide corresponding emergency response strategies for the equipment based on the identification of key factors, and carry out targeted processing of the equipment when the environmental conditions have already caused a certain degree of impact on the equipment.

[0055] 3. After receiving the first-level alarm instruction and processing it, continue to build the operating value , to evaluate the status of the equipment. If the signal quality is continuously poor and the adjustment and control of the equipment fail to achieve the expected results, it can prevent the first-level alarm instruction from being ignored.

[0056] 4. As a response to the secondary alarm command, the key components of the equipment are selected and maintained to avoid the impact of failure or defects of key components on the equipment, and to improve the reliability of equipment operation and measurement when controlling and adjusting the equipment.

[0057] 5. After several consecutive tests, the reliability of the measurement results can be judged based on the measurement results, and the control and adjustment effect of the equipment can be verified; based on the collection of various operating data of the equipment, the equipment can be fault-detected, and if the equipment has an operating failure, it can be dealt with in a targeted manner.

[0058] 6. The equipment fault maintenance knowledge graph provides maintenance plans. On the basis of controlling the equipment, targeted maintenance can be performed on existing faults according to the maintenance plan. The maintenance frequency of the equipment can be constrained to prevent high-frequency maintenance from affecting the regular use of the equipment. By screening out key components that are more prone to failure and switching them, the equipment can continue to be used when a failure occurs.

[0059] 7. After the equipment completes the measurement, the measurement data is corrected to obtain the measurement correction value, so as to improve the measurement accuracy and reliability of the equipment while taking into account external interference. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 It is a schematic flow chart of a control method for a laser Doppler vibrometer device according to the present invention;

[0061] Figure 2 The figure is a schematic diagram of the control system structure of the laser Doppler vibrometer device used in the present invention. DETAILED DESCRIPTION

[0062] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0063] See also Figure 1 The present invention provides a control method for a laser Doppler vibrometer, comprising:

[0064] Step 1: Using the environmental condition data in the operating area detected by the sensor network as input, adjust the operating parameters of the equipment in real time, and generate the signal condition value from the signal quality data of the equipment , if the signal condition value If the signal is lower than the threshold, a first-level alarm command will be issued to the outside;

[0065] The step 1 includes the following contents:

[0066] Step 101: After determining the operating area of ​​the equipment, a sensor network is deployed in the operating area, such as a temperature and humidity sensor, a dust sensor, an illumination sensor, a wind speed and direction sensor, and a laser radar;

[0067] The corresponding detection data is obtained by the sensor network detection, such as using laser radar to perform high-precision monitoring of air flow and disturbance, etc., to obtain the following environmental condition data, including temperature data, humidity data, dust particle concentration, dust particle size, light intensity data, wind speed data, laser radar, coordinate data, elevation data and echo intensity data; the environmental condition data collected by the sensor network are aggregated to generate an environmental condition data set;

[0068] Step 102: Use the labeled sample data to train the machine learning algorithm to obtain a trained equipment automatic control model; use the real-time environmental condition data in the operating area as input, and use the trained equipment automatic control model to adjust the equipment's operating parameters in real time, such as filter parameter settings and laser power, etc.;

[0069] Obtain environmental condition parameters and corresponding signal quality data in the operating area, including signal-to-noise ratio, signal strength, frequency response, and phase accuracy, and generate a signal quality data set;

[0070] When in use, when the equipment is in operation within the operating area, the environmental condition data of the operating area is adjusted. When the environmental conditions within the operating area cannot be controlled, the environmental condition data is used as feedback to adaptively control the control parameters of the equipment, so that the operating parameters of the equipment are adapted to the environmental conditions, thereby reducing the impact of environmental conditions on equipment operation and signal quality.

[0071] Step 103: Generate a signal condition value from the data in the signal quality data set , for signal strength , frequency response and phase accuracy Perform linear normalization and map the corresponding data values ​​to the interval Then follow the following method:

[0072] ;

[0073] in, , is the number of time nodes, For the The signal strength at each time node, For the The frequency response of each time node, For the Phase accuracy of time nodes; weight coefficient: , ,and , is the correction coefficient, which ranges from 0.1 to 1;

[0074] Based on historical data and management expectations for device signal quality, signal thresholds are pre-set; if the signal condition value obtained is If the signal quality is lower than the signal threshold, it means that the current signal quality of the device is poor, which will interfere with and affect the accuracy of subsequent measurement results. At this time, the alarm is automatically triggered according to the changes in environmental parameters and measurement signals, and a first-level alarm instruction is issued to the outside.

[0075] When using, combine the contents in steps 101 to 103:

[0076] As a further content, based on the adaptive adjustment of the equipment operating parameters, the signal quality of the equipment is recorded and counted, and the signal condition value is constructed accordingly. At this time, considering the scenario of environmental interference, the signal quality of the device is evaluated to determine whether there is any abnormality in the signal quality of the device. If there is still an abnormality, the device needs to be adjusted and controlled to improve the reliability of the device.

[0077] In the existing control method for laser Doppler vibrometer equipment, corresponding qualified intervals are usually set for various control parameters of the equipment, and the measurement results output by the equipment are used as feedback. When the reliability of the output measurement data is insufficient, the various control parameters of the equipment are corrected and controlled. However, when the operating environment of the equipment is poor or even abnormal, such as when there is a lot of dust, too strong or too weak light, the external operating conditions will cause great interference to the accuracy of the measurement. If this is used as feedback to control and adjust the equipment, or even to correct the measurement results, it may be difficult to achieve the expected effect due to poor targeting.

[0078] Step 2: If there are abnormal environmental conditions in the operating area, the emergency treatment strategy library outputs the corresponding emergency treatment strategy based on the key factors affecting the operation of the equipment, and the equipment is processed after the emergency treatment strategy is executed;

[0079] The step 2 includes the following contents:

[0080] Step 201: train a machine learning model algorithm with the labeled sample data to obtain a trained abnormal data detection model; after receiving a first-level alarm instruction, use the continuously acquired environmental condition data as input, use the trained abnormal data detection model to perform abnormality detection, and issue a reminder instruction to the outside when the environmental condition is abnormal;

[0081] Through online retrieval or offline formulation, several equipment emergency response strategies for abnormal environmental conditions are collected in advance. For example, coherent detection and signal denoising technology are used to significantly improve the signal-to-noise ratio of the signal; a closed optical path system is designed, and sealed light pipes and high-performance dust covers are used to reduce interference from dust and ambient light;

[0082] Use high-performance optical coating materials (such as anti-reflection coatings) to reduce the impact of excess light reflection and scattering; implement laser frequency stabilization technology, such as optical comb technology or external cavity feedback control, to ensure high stability of laser output; equip with a real-time power adjustment system to dynamically adjust the laser power according to changes in external light intensity through a feedback control mechanism to ensure signal stability; generate an emergency treatment strategy library after summarizing several emergency treatment strategies;

[0083] When in use, as a response to abnormal device signals, the environmental conditions of the device are detected to verify whether there are any abnormalities in the environmental conditions. When there are abnormalities in the environmental conditions, real-time reminders are issued to the outside to avoid further impact and interference of the environmental conditions on the device.

[0084] Step 202: Collect historical environmental condition data and historical operation data of the equipment in the operation area, and identify several key factors in the environmental conditions that have a greater impact on the operation of the equipment through principal component analysis;

[0085] After receiving the reminder instruction, the emergency processing strategy library outputs the corresponding emergency processing strategy according to the correspondence between the key factors and the emergency processing strategy, and executes the emergency processing strategy to process the equipment. When the signal quality output by the equipment is poor, preliminary targeted processing can be carried out;

[0086] When using, combine the contents in steps 201 to 202:

[0087] When there are abnormal environmental conditions in the operating area, in this scenario, identify and screen out interference factors that have a greater impact on equipment operation within the abnormal environmental conditions, provide corresponding emergency processing strategies for the equipment based on the identified key factors, and carry out targeted processing on the equipment when the environmental conditions have already caused a certain degree of impact on the equipment.

[0088] Step 3: Build the operation value based on the feedback data received from the first-level alarm command , if the running value If it exceeds expectations, a secondary alarm command will be issued to the outside, and key components of the equipment will be maintained at a time point that meets the constraints;

[0089] The step three includes the following contents:

[0090] Step 301: If the number of consecutive alarm instructions received within the preset alarm period exceeds the expected number, the operation value is constructed based on the state data of the received first-level alarm instruction. , as follows:

[0091] ;

[0092] Where: is the conditional signal value at time t, The corresponding signal threshold, For time Conditional signal value on ; is the indicator function that the signal is below the qualified threshold, where , and is the weight coefficient, For time arrive Time point;

[0093] If the running value obtained If the preset operating threshold is exceeded, it means that the quality of the current measurement signal of the equipment has not been improved and further processing is required. At this time, a secondary alarm command is issued to the outside;

[0094] When used, as a further content, continue to build the running value after receiving the first-level alarm instruction and processing it. , evaluate the equipment status, if the signal quality is continuously poor, continue to issue the second-level alarm command, when the adjustment and control of the equipment fails to achieve the expected effect, prevent the first-level alarm command from being ignored;

[0095] Step 302: After receiving the secondary alarm command, regularly inspect and maintain key components of the equipment, such as optical elements, laser sources, and electronic modules, within a preset prevention period, and restrict the inspection and maintenance frequency. , perform maintenance on key components of the equipment at time nodes that meet the constraints. The constraints are as follows:

[0096] ;

[0097] in, is the number of times the secondary alarm command is received. It is Second level alarm command to The time interval, is the mean of the time interval; the weight coefficient, , ,and ;The weight coefficient is the same as the previous value;

[0098] When using, combine the contents in steps 301 to 302:

[0099] Considering that after emergency treatment of the equipment, the signal quality or even the detection results of the equipment are still abnormal or insufficient, as a response to the secondary alarm command, the key components of the equipment are selected and maintained. This can avoid the impact of failures or defects of key components on the equipment, and improve the reliability of equipment operation and measurement when controlling and adjusting the equipment.

[0100] Step 4: If the reliability of the measured value output by the equipment is insufficient, after recording the equipment's operating data in real time, use the trained equipment fault diagnosis model to perform fault diagnosis. If the equipment has an operating fault, extract the corresponding fault features.

[0101] The step 4 includes the following contents:

[0102] Step 401: after obtaining a plurality of measurement results of the continuous acquisition device, a plurality of continuous measurement values ​​are arranged along the time axis to generate a line graph, and a standard measurement value is set for each measurement value to construct a standard line graph;

[0103] Perform similarity analysis on the line graph and the standard line graph to obtain corresponding similarity data. If the similarity data is lower than expected, it means that the reliability of the equipment still fails to meet expectations after maintenance, and there may be certain operating failures. At this time, a real-time monitoring instruction is issued to the outside;

[0104] When in use, as a further content, after completing the shallow control and adjustment of the equipment, the equipment is put into the test use state. After several consecutive tests, the reliability of the measurement results can be judged based on the measurement results, and the control and adjustment effect of the equipment can be verified;

[0105] Step 402: After receiving the real-time monitoring instruction, the state monitoring network deployed on the equipment records the operation state, vibration characteristics and electrical parameters of the equipment in real time, such as temperature and current operation data, and summarizes them as a set of equipment operation data;

[0106] Use the labeled sample data to train the machine learning algorithm and obtain the trained equipment fault diagnosis model; use the real-time operation data of the equipment as input and use the trained equipment fault diagnosis model to perform fault diagnosis. If there is an operation fault, obtain the corresponding fault feature through feature extraction;

[0107] When using, combine the contents in steps 401 and 402:

[0108] As a further content, if the measurement data output by the equipment is difficult to achieve the expected accuracy or reliability, it may mean that the treatment measures taken previously have not fully achieved the expected results, indicating that the equipment may currently have certain operating failures and it is difficult to control the equipment effectively. Therefore, as a deep-level control of the equipment, on the basis of collecting various operating data of the equipment, the equipment is fault-detected and, if the equipment has an operating failure, it can be dealt with in a targeted manner.

[0109] Step 5: Provide a maintenance plan based on the fault characteristics using the equipment fault maintenance knowledge graph, automatically switch the key components that cause the fault, and use the trained measurement data correction model to correct the measurement data, and output the actual measurement value and the measurement correction value;

[0110] The step five includes the following contents:

[0111] Step 501, after recording all environmental condition data, operation data and alarm event data during equipment operation, a comprehensive equipment operation history database is established;

[0112] Taking laser Doppler vibration measurement equipment fault maintenance as the target word, combined with the equipment operation history database, after deep retrieval and entity relationship building, the equipment fault maintenance knowledge graph is pre-built; based on the correspondence between fault characteristics and maintenance plans, the equipment fault maintenance knowledge graph provides maintenance plans;

[0113] As a further content, when the equipment fails to operate, the equipment failure maintenance knowledge map provides a maintenance plan, which can be used to control the equipment and perform targeted maintenance on the existing faults according to the maintenance plan;

[0114] Step 502: Generate frequency of fault maintenance plan under dimensionless conditions Constraints are imposed, and at the time nodes that meet the constraints, the maintenance plan is output to the outside. The constraints are as follows:

[0115] ;

[0116] Where: is the abnormality of the environmental conditions in the operating area at time t, for to Moment within The running value on is the signal condition value at time t, is the cumulative distribution function of the standard normal distribution, with weight coefficients: , , ;

[0117] After selecting several module components with the highest failure rate based on the fault detection data, they are equipped with redundant components and automatically switched when key components fail;

[0118] During use, considering that the equipment may be in operation for a long time and is not convenient for real-time maintenance, the maintenance frequency of the equipment is restricted to prevent high-frequency maintenance from affecting the normal use of the equipment. As a supplement, by screening out key components that are more prone to failure and switching them, the equipment can continue to be used when a failure occurs.

[0119] The main body of the laser Doppler device, that is, the detection box is fixed to the fixed bracket on the top of the tracking gimbal through a supporting shaft, wherein the tracking gimbal includes an automatic tracking gimbal and an automatic tracking gimbal controller, the automatic tracking gimbal includes horizontal rotation and vertical rotation brackets and vertical motors and horizontal motors that have a transmission relationship with the vertical motors, respectively; secondly, the tracking gimbal is internally provided with a data communication module with the laser Doppler device, an image processing module, a laser Doppler device acquisition speed detection module, a laser Doppler device acquisition control module, a drive motor control module, an angle control module, a pause time control module, etc.

[0120] The laser Doppler device is connected to the data communication module of the intelligent tracking pan-tilt via the acquisition control line and the acquisition signal PC synchronization line, and feedback control is performed on each other; the automatic control model acquired by the laser Doppler device is written into the internal memory of the tracking pan-tilt, and the automatic control model controls the pause and rotation of the automatic tracking pan-tilt, and controls the laser Doppler device to detect and acquire data after a momentary pause according to the set acquisition angle, and controls the start rotation of the automatic tracking pan-tilt;

[0121] The total angle and single angle that the laser Doppler device needs to rotate and the collection cycle are preset; after the instruction to start automatic collection is started, the trained automatic control model controls the horizontal rotation and vertical rotation bracket to drive the laser Doppler device to rotate at a uniform speed, and when the laser Doppler device rotates to the preset single angle, the laser Doppler device is started to collect data; after the laser Doppler device collects the vibration data of a detection point, the automatic control model controls the laser Doppler device to continue to operate, and the laser Doppler device collects vibration data within the preset total angle at the preset and single angles and collection cycle intervals; when the preset total angle is reached or the instruction to end automatic collection is started in the middle of the operation, the automatic control model controls the automatic tracking pan-tilt to stop rotating, so that the laser Doppler device stops collecting; thereby, the real-time vibration state of the target to be measured can be obtained in a timely and accurate manner, thereby improving the measurement efficiency.

[0122] Step 503: The environmental condition data, equipment operation data, measurement data, and fault detection data in the operation area are aggregated as a sample database, and part of the data is extracted from the sample database as sample data; a convolutional neural network is trained with the sample data to obtain a trained measurement data correction model;

[0123] After the device outputs the measurement data, the measurement data is corrected using the trained measurement data correction model to obtain a measurement correction value, and the actual measurement value and the measurement correction value are output;

[0124] When using, combine the contents in steps 501 to 503:

[0125] As a further content, considering that the environment and related factors may cause a certain degree of interference to the output measurement data, as a coping method, after the device completes the measurement, the measurement data is corrected to obtain the measurement correction value, so as to improve the measurement accuracy and reliability of the device under the condition of taking into account the external interference.

[0126] The method of building the fault maintenance knowledge graph of laser Doppler vibrometer equipment is as follows:

[0127] Clarify the construction goal - First, it is necessary to clarify the construction goal of the knowledge graph, which is to achieve rapid fault diagnosis and maintenance of laser Doppler vibrometer equipment. This includes basic information about the equipment, common fault types, causes of faults, maintenance methods and other related knowledge.

[0128] Data collection and preprocessing - Data source determination: Collect technical manuals, maintenance records, failure cases, expert experience, etc. of laser Doppler vibrometer equipment as data sources. Data cleaning: Clean the collected data to remove duplicate, invalid or erroneous information. Data labeling: Label the cleaned data, including entity labeling (such as equipment name, fault type, etc.), relationship labeling (such as the relationship between the fault cause and the fault type), etc.

[0129] Knowledge graph construction - define data mode: Top-down: starting from the top-level concept of the equipment, gradually refine to the specific fault type and maintenance method, and form a structured data model. Bottom-up: starting from a specific fault case, summarize the general fault type and maintenance method, and then gradually abstract to form the upper-level concept. In actual operation, these two methods can be selected or combined according to the specific situation.

[0130] Entity and Relation Extraction - Extract entities and relations from text data using natural language processing (NLP) technology. For structured data (such as database records), entities and relations can be extracted directly.

[0131] Graph fusion - Integrate the knowledge extracted from different data sources to eliminate conflicts and duplications. Use ontology modeling tools (such as Protege, OntoStudio, etc.) to help formalize and integrate knowledge. Graph optimization: Verify and test the constructed knowledge graph to ensure the accuracy and completeness of the knowledge. Iterate and optimize the graph according to actual application needs.

[0132] Knowledge graph application - Fault diagnosis: When a device fails, you can retrieve possible fault causes and repair methods in the knowledge graph by entering the fault phenomenon or related parameters. Maintenance guidance: Provide targeted maintenance suggestions and guidance based on the equipment's operating conditions and maintenance history. Knowledge sharing and learning: Use the knowledge graph as a knowledge sharing and learning platform to facilitate communication and learning between maintenance personnel.

[0133] See also Figure 2 The present invention provides a control system for a laser Doppler vibrometer, comprising:

[0134] The signal control unit uses the environmental condition data in the operating area detected by the sensor network as input, adjusts the operating parameters of the equipment in real time, and generates a signal condition value from the signal quality data of the equipment , if the signal condition value If the signal is lower than the threshold, a first-level alarm command will be issued to the outside;

[0135] Equipment emergency processing unit: if there are abnormal environmental conditions in the operating area, the emergency processing strategy library will output the corresponding emergency processing strategy based on the key factors affecting the operation of the equipment, and the equipment will be processed after the emergency processing strategy is executed;

[0136] Control maintenance unit, build operation value according to feedback data received from first level alarm command , if the running value If it exceeds expectations, a secondary alarm command will be issued to the outside, and key components of the equipment will be maintained at a time point that meets the constraints;

[0137] The fault diagnosis unit, if the reliability of the measured value output by the equipment is insufficient, records the equipment's operating data in real time and uses the trained equipment fault diagnosis model to perform fault diagnosis. If the equipment has an operating fault, the corresponding fault features are extracted;

[0138] The correction unit provides a maintenance plan based on the equipment fault maintenance knowledge graph according to the fault characteristics, automatically switches the key components that cause the fault, and uses the trained measurement data correction model to correct the measurement data, and outputs the actual measurement value and the measurement correction value.

[0139] Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of this application.

[0140] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0141] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only some logical function divisions. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0142] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0143] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A control method for a laser Doppler vibrometer, characterized in that: include, The environmental condition data in the operating area of ​​the vibration measuring device detected by the sensor network is used as input to adjust the operating parameters of the device in real time, and the signal condition value is generated from the signal quality data of the device. , if the signal condition value If the signal is lower than the signal threshold, a first-level alarm command is issued to the outside; wherein, the environmental condition parameters and corresponding signal quality data in the operating area are obtained, including at least signal-to-noise ratio, signal strength, frequency response and phase accuracy, and a signal quality data set is generated by summarizing the signal quality data set, and a signal condition value is generated from the signal quality data set ;Signal condition value The generation method is as follows: , frequency response and phase accuracy Perform linear normalization and map the corresponding data values ​​to the interval Then follow the following method: ； in, , is the number of time nodes, For the The signal strength at each time node, For the The frequency response of each time node, For the Phase accuracy of time nodes; weight coefficient: , ,and , is the correction factor; If there are abnormal environmental conditions in the operating area, the emergency treatment strategy library will output the corresponding emergency treatment strategy based on the key factors affecting the operation of the equipment, and the equipment will be processed after the emergency treatment strategy is executed; Build operating values ​​based on feedback data received from the first-level alarm command , if the running value If it exceeds expectations, a secondary alarm command will be issued to the outside, and key components of the equipment will be maintained at a time point that meets the constraints; Among them, when the number of consecutive first-level alarm instructions received within the preset alarm period exceeds expectations, the operating value is constructed based on the status data of the first-level alarm instructions received. , as follows: ； Where: is the conditional signal value at time t, The corresponding signal threshold, For time Conditional signal value on ; is the indicator function that the signal is below the qualified threshold, where , and is the weight coefficient, For time arrive Time point; If the measurement value output by the equipment is not reliable enough, after recording the equipment's operating data in real time, use the trained equipment fault diagnosis model to perform fault diagnosis. If the equipment has an operating fault, extract the corresponding fault features. The equipment fault maintenance knowledge graph provides a maintenance plan based on the fault characteristics, automatically switches the key components that cause the fault, and uses the trained measurement data correction model to correct the measurement data, and outputs the actual measurement value and the measurement correction value.

2. The control method for a laser Doppler vibrometer according to claim 1, characterized in that: A sensor network is deployed in the operating area. The sensor network detects and obtains the corresponding environmental condition data, which are aggregated to generate an environmental condition data set. The real-time environmental condition data in the operating area is used as input, and the trained equipment automatic control model is used to adjust the equipment's operating parameters in real time.

3. The control method for a laser Doppler vibrometer according to claim 2, characterized in that: After receiving the first-level alarm command, the continuously acquired environmental condition data is used as input, and the trained abnormal data detection model is used to detect abnormalities, and a reminder command is issued to the outside when the environmental conditions are abnormal; several equipment emergency treatment strategies for abnormal environmental conditions are collected in advance, and then an emergency treatment strategy library is generated; Collect historical environmental condition data and historical operation data of equipment in the operating area, and identify several key factors in environmental conditions that affect equipment operation through principal component analysis; after receiving the reminder instruction, the emergency treatment strategy library outputs the corresponding emergency treatment strategy based on the correspondence between the key factors and the emergency treatment strategy.

4. The control method for a laser Doppler vibrometer according to claim 3, characterized in that: After receiving the second-level alarm instruction, regularly inspect and maintain the key components of the equipment within the pre-set prevention cycle, and limit the inspection and maintenance frequency , perform maintenance on key components of the equipment at time nodes that meet the constraints. The constraints are as follows: ； in, is the number of times the secondary alarm command is received. It is Second level alarm command to The time interval, is the mean of the time interval; the weight coefficient, , ,and .

5. The control method for a laser Doppler vibrometer according to claim 4, characterized in that: A line graph is generated by arranging several consecutive measurement values ​​along the time axis, and a standard line graph is constructed after setting a standard measurement value for each measurement value; a similarity analysis is performed on the line graph and the standard line graph to obtain the corresponding similarity data. If the similarity data is lower than expected, a real-time monitoring instruction is issued to the outside.

6. The control method for a laser Doppler vibrometer according to claim 5, characterized in that: After receiving the real-time monitoring instruction, the condition monitoring network deployed on the equipment records the equipment's operating status, vibration characteristics, and electrical parameters in real time, and summarizes them as a set of equipment operation data. The real-time operation data of the equipment is used as input, and fault diagnosis is performed using the trained equipment fault diagnosis model.

7. The control method for a laser Doppler vibrometer according to claim 6, characterized in that: After recording all environmental condition data, operation data and alarm event data during equipment operation, a comprehensive equipment operation history database is compiled and established; using laser Doppler vibration measurement equipment failure maintenance as the target word, combined with the equipment operation history database, a knowledge graph for equipment failure maintenance is pre-built.

8. The control method for a laser Doppler vibrometer according to claim 7, characterized in that: Frequency of generating maintenance plans for failures Constraints are imposed, and the maintenance plan is output to the outside at the time node that meets the constraints; After selecting several module components with the highest failure rate based on the fault detection data, they are equipped with redundant components and automatically switched when key components fail; the maintenance plan is output to the outside in frequency. The constraints are as follows: ； Where: is the abnormality of the environmental conditions in the operating area at time t, for to Moment within The running value on is the signal condition value at time t, is the cumulative distribution function of the standard normal distribution, with weight coefficients: , , .

9. The control method for a laser Doppler vibrometer according to claim 8, characterized in that: The total angle and single angle that the device needs to rotate as well as the collection cycle are preset; after the command to start automatic collection is started, the trained automatic control model controls the horizontal and vertical rotating brackets to drive the device to rotate at a constant speed. When the device rotates to the preset single angle, the device is started for data collection; After the device collects the vibration data of a detection measuring point, the automatic control model controls the device to continue running, and the device collects vibration data within the preset total angle, at the preset and single angles and at the collection cycle interval; when the preset total angle is reached or the instruction to end the automatic collection is started midway through the operation, the automatic control model controls the automatic tracking pan-tilt head to stop rotating, so that the device stops collecting.

10. The control method for a laser Doppler vibrometer according to claim 9, characterized in that: The environmental condition data, equipment operation data, measurement data, and fault detection data in the operation area are aggregated as a sample database, and part of the data is extracted from the sample database as sample data; The convolutional neural network is trained with sample data to obtain a trained measurement data correction model. After the device outputs the measurement data, the measured data is corrected using the trained measurement data correction model.

11. A control system for a laser Doppler vibrometer, applied to the control method according to any one of claims 1 to 10, characterized in that: include, The signal control unit uses the environmental condition data in the operating area detected by the sensor network as input, adjusts the operating parameters of the equipment in real time, and generates a signal condition value from the signal quality data of the equipment , if the signal condition value If the signal is lower than the threshold, a first-level alarm command will be issued to the outside; Equipment emergency processing unit: if there are abnormal environmental conditions in the operating area, the emergency processing strategy library will output the corresponding emergency processing strategy based on the key factors affecting the operation of the equipment, and the equipment will be processed after the emergency processing strategy is executed; Control maintenance unit, build operation value according to feedback data received from first level alarm command , if the running value If it exceeds expectations, a secondary alarm command will be issued to the outside, and key components of the equipment will be maintained at a time point that meets the constraints; The fault diagnosis unit, if the reliability of the measured value output by the equipment is insufficient, records the equipment's operating data in real time and uses the trained equipment fault diagnosis model to perform fault diagnosis. If the equipment has an operating fault, the corresponding fault features are extracted; The correction unit provides a maintenance plan based on the equipment fault maintenance knowledge graph according to the fault characteristics, automatically switches the key components that cause the fault, and uses the trained measurement data correction model to correct the measurement data, and outputs the actual measurement value and the measurement correction value.

Images