CN109933004B - Machine tool fault diagnosis and prediction method and system based on edge computing and cloud cooperation - Google Patents
Machine tool fault diagnosis and prediction method and system based on edge computing and cloud cooperation Download PDFInfo
- Publication number
- CN109933004B CN109933004B CN201910235933.1A CN201910235933A CN109933004B CN 109933004 B CN109933004 B CN 109933004B CN 201910235933 A CN201910235933 A CN 201910235933A CN 109933004 B CN109933004 B CN 109933004B
- Authority
- CN
- China
- Prior art keywords
- characteristic data
- fault diagnosis
- numerical control
- machine tool
- data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
Images
Classifications
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02P—CLIMATE CHANGE MITIGATION TECHNOLOGIES IN THE PRODUCTION OR PROCESSING OF GOODS
- Y02P90/00—Enabling technologies with a potential contribution to greenhouse gas [GHG] emissions mitigation
- Y02P90/02—Total factory control, e.g. smart factories, flexible manufacturing systems [FMS] or integrated manufacturing systems [IMS]
Landscapes
- Testing And Monitoring For Control Systems (AREA)
- Numerical Control (AREA)
Abstract
The invention discloses a method and a system for diagnosing and predicting machine tool faults based on edge computing and cloud cooperation, wherein the method comprises the following steps: 1) the method comprises the steps that a sensing node collects state data of a mechanical system and extracts characteristics to obtain mechanical characteristic data; the gateway node acquires state data of the numerical control system and extracts features to obtain numerical control feature data, and sends mechanical feature data and numerical control feature data to the cloud; 2) the cloud end carries out fault diagnosis and prediction on the mechanical characteristic data and the numerical control characteristic data by using a preset fault diagnosis and prediction model, and sends fault diagnosis and prediction results to the client end; 3) the client receives the feedback result of the machine tool expert on the fault diagnosis and prediction result and sends the feedback result to the cloud; 4) and the cloud end marks the mechanical characteristic data and the numerical control characteristic data according to the feedback result, trains and updates the fault diagnosis and prediction model, and returns to execute the step 2). By applying the embodiment of the invention, the fault diagnosis and prediction accuracy of the machine tool are improved.
Description
Technical Field
The invention relates to a machine tool fault diagnosis and prediction method and system, in particular to a machine tool fault diagnosis and prediction method and system based on edge computing and cloud cooperation.
Background
With the continuous development of modern industry and the continuous improvement of technology level, the electromechanical device is also developed to the direction of high speed, continuity, centralization, automation and precision, and further the composition and structure of the electromechanical device are more and more complicated, which directly leads to the increase of the failure rate and the increase of the diagnosis difficulty of the electromechanical device. For example, subtle damaging faults or abnormalities of critical components (e.g., rolling bearings, gears, ball screws, etc.) in a machine tool may cause the entire system to fail, or even cause catastrophic results if not detected and eliminated in time. Therefore, it is necessary to analyze and predict the failure of the electromechanical device. By means of predictive fault diagnosis, the occurrence of workpiece rejection and machine tool damage can be reduced or avoided to varying degrees. The machine tool as a typical mechatronic product can be divided into three subsystems as a whole: numerical control systems, electrical systems, and mechanical systems that are most susceptible to irreversible damage. Mechanical system failure is a necessary phenomenon in production processes. Conventionally, the prevention and treatment method is to check the quality of the workpiece processed by the machine tool by an operator, or to judge the health condition of the machine tool by a technician periodically or aperiodically based on vibration, noise, etc. of the machine tool. The two modes are easily influenced by subjective factors such as professional ability and the like, accuracy cannot be guaranteed, reliability is poor, and efficiency is low. At present, the online detection method for the health of the machine tool mainly comprises a chart method and an indirect method. The graph detection method is to process the collected data by using some algorithms, then generate a time domain or frequency domain graph, and judge by comparing images manually or by machine learning. The indirect detection method is used for detecting stress, spindle current, torque, vibration, sound, temperature, hydraulic pressure and the like so as to deduce the health degree of the machine tool. The graph detection method needs to compare images for judgment, brings great pressure to network flow and a server, is low in efficiency, and often cannot find problems in time; the indirect detection method involves too many factors and much interference, and the accuracy cannot be guaranteed.
In the prior art, only numerical control system data is used or only mechanical system data is collected, and fault diagnosis and prediction are performed according to the data, but the numerical control system and the mechanical system of the machine tool are in causal data connection in the process of machine tool fault occurrence, and the numerical control system and the mechanical system are not combined in the prior art, so that the technical problem that the accuracy of diagnosis and prediction is not high exists in the prior art.
Disclosure of Invention
The invention aims to provide a machine tool fault diagnosis and prediction method and system based on edge computing and cloud cooperation so as to improve the detection precision of an online detection method based on an industrial internet platform.
The invention solves the technical problems through the following technical scheme:
the invention provides a machine tool fault diagnosis and prediction method based on edge computing and cloud cooperation, which comprises the following steps:
1) the sensing node acquires mechanical state data of the machine tool component according to a preset first acquisition strategy and extracts characteristics to obtain mechanical characteristic data, and the mechanical characteristic data is uploaded to the gateway node; the gateway node acquires working state data of the numerical control system according to a preset second acquisition strategy and extracts characteristics to obtain numerical control characteristic data, and sends the mechanical characteristic data and the numerical control characteristic data to a cloud terminal;
2) the cloud receives the mechanical characteristic data and the numerical control characteristic data, performs fault diagnosis and prediction by using a preset fault diagnosis and prediction model, and sends a fault diagnosis and prediction result to the client;
3) the client displays the fault diagnosis and prediction result to a machine tool expert, receives a feedback result of the machine tool expert on the fault diagnosis and prediction result, and sends the feedback result to the cloud end;
4) the cloud receives the feedback result, and marks the mechanical characteristic data and the numerical control characteristic data to obtain sample data; and training the fault diagnosis and prediction model according to the sample data, and returning to execute the step 2).
Optionally, the gateway node is used as a root node, and the sensing node completes time synchronization through the FTSP based on the wireless sensing ad hoc network; the numerical control system completes time synchronization through a field bus NCUC-bus based on a wired network.
Optionally, the method further includes:
and the gateway node receives an instruction sent by the client, wherein the instruction comprises but is not limited to adjusting the characteristic data acquisition frequency, adjusting the characteristic data acquisition duration, adjusting the type of the characteristic data, updating a characteristic extraction algorithm library, and restarting or closing the sensing node or the gateway node.
The invention provides a machine tool fault diagnosis and prediction system based on edge computing and cloud coordination, which comprises an edge end and a cloud end, wherein,
the edge end comprises a gateway node and a sensing node, wherein the sensing node is arranged on a part of the machine tool, and is used for acquiring mechanical state data of the part according to a preset first acquisition strategy and extracting characteristics to obtain mechanical characteristic data;
the gateway node is used for acquiring working state data of the numerical control system according to a preset second strategy and extracting characteristics to obtain numerical control characteristic data, and sending the mechanical characteristic data and the numerical control characteristic data to the cloud end;
the gateway node sends data to the cloud, wherein the mechanical characteristic data includes, but is not limited to, one or a combination of temperature, humidity, rotating speed, torque, amplitude value, form factor, crest factor and kurtosis factor; the numerical control characteristic data comprises but is not limited to one or a combination of an instruction code, an alarm code, a spindle rotating speed, a moving direction and a coordinate;
the cloud end comprises a communication unit, a computing unit, a storage unit, a service unit and a training unit, wherein the communication unit is used for receiving the mechanical characteristic data and the numerical control characteristic data uploaded by the gateway node, writing the mechanical characteristic data and the numerical control characteristic data into the storage unit and forwarding the mechanical characteristic data and the numerical control characteristic data to the computing unit;
the computing unit is used for carrying out fault diagnosis and prediction on the mechanical characteristic data and the numerical control characteristic data and writing fault diagnosis and prediction results into the storage unit;
the storage unit is used for storing the mechanical characteristic data, the numerical control characteristic data, the feedback result of a machine tool expert and various service data;
the business unit is used for displaying the fault diagnosis and prediction result to a machine tool expert and receiving a feedback result of the machine tool expert on the fault diagnosis and prediction result;
and the training unit is used for training a fault diagnosis and prediction model according to the feedback result, the mechanical characteristic data and the numerical control characteristic data so as to enable the result of the fault diagnosis and prediction model and the feedback result and feed back model parameters to the calculating unit.
Optionally, the service unit is configured to issue an instruction to an edge, where the instruction includes, but is not limited to, adjusting a collection frequency or duration of data, a type of the collected data, updating a feature extraction algorithm library, and restarting or closing the sensing node or the gateway node.
Optionally, the client is further configured to receive an operation instruction of a user, and the client includes a Web application and a mobile phone App.
Optionally, the gateway node is configured to complete time synchronization of the gateway node through an NTP network, and serve as a basis for edge time synchronization.
Optionally, the gateway node is configured to determine whether the machine tool is in a non-shutdown and non-standby state, and if so, send the mechanical characteristic data and the numerical control characteristic data to the cloud.
Optionally, the system further includes a client configured to display the fault diagnosis and prediction result to a machine tool expert, and receive a feedback result of the machine tool expert on the diagnosis and prediction result.
Compared with the prior art, the invention has the following advantages:
by applying the embodiment of the invention, based on the cooperation of edge calculation and a cloud, the mechanical state data of a machine tool mechanical system is acquired and the characteristics are extracted at the edge by using the sensing node, the mechanical characteristic data is acquired, the working state data of the machine tool numerical control system is acquired and the characteristics are extracted at the gateway node, the numerical control characteristic data is acquired, then the mechanical characteristic data and the numerical control characteristic data are subjected to fault diagnosis and prediction at the cloud, then the mechanical characteristic data and the numerical control characteristic data are labeled by using expert feedback to acquire sample data, then the fault diagnosis and the prediction model are updated by using the sample data, and then the cycle processing is performed for multiple times, so that the most accurate fault diagnosis and prediction model is acquired, and the fault diagnosis and prediction accuracy are further improved.
Drawings
Fig. 1 is a schematic flowchart of a method for diagnosing and predicting a machine tool fault based on edge computing and cloud coordination according to an embodiment of the present invention;
fig. 2 is a schematic structural diagram of a machine tool fault diagnosis and prediction system based on edge computing and cloud coordination according to an embodiment of the present invention;
fig. 3 is a schematic data processing diagram of an edge terminal in a machine tool fault diagnosis and prediction system based on edge computing and cloud coordination according to an embodiment of the present invention;
fig. 4 is a schematic diagram of a process of issuing an instruction in a machine tool fault diagnosis and prediction system based on edge computing and cloud coordination according to an embodiment of the present invention;
fig. 5 is a schematic diagram illustrating a first process of time synchronization in a machine tool fault diagnosis and prediction system based on edge computing and cloud coordination according to an embodiment of the present invention;
fig. 6 is a schematic diagram of a second process of time synchronization in a machine tool fault diagnosis and prediction system based on edge computing and cloud coordination according to an embodiment of the present invention;
fig. 7 is a schematic diagram of a third process of time synchronization in a machine tool fault diagnosis and prediction system based on edge computing and cloud coordination according to an embodiment of the present invention.
Detailed Description
The following examples are given for the detailed implementation and specific operation of the present invention, but the scope of the present invention is not limited to the following examples.
The embodiment of the invention provides a method and a system for diagnosing and predicting machine tool faults based on edge computing and cloud coordination.
Example 1
Fig. 1 is a schematic flowchart of a method for diagnosing and predicting a machine tool fault based on edge computing and cloud coordination according to an embodiment of the present invention, where as shown in fig. 1, embodiment 1 of the present invention includes:
s101: the method comprises the steps that a sensing node collects mechanical state data of a machine tool component according to a preset first collection strategy and extracts characteristics to obtain mechanical characteristic data, and the mechanical characteristic data are uploaded to a gateway node; and the gateway node acquires the working state data of the numerical control system according to a preset second acquisition strategy, extracts the characteristics, obtains numerical control characteristic data, and sends the mechanical characteristic data and the numerical control characteristic data to the cloud.
In an exemplary manner, the first and second electrodes are,
the ball screw is the most commonly used transmission element on tool machinery and precision machinery, and has the main function of converting rotary motion into linear motion or converting torque into axial repeated acting force, and has the characteristics of high precision, reversibility and high efficiency. Ball screws are widely used in various industrial equipments and precision instruments due to their small frictional resistance. Overall, the ball screw has a complex structure, high precision, high movement efficiency but poor shock resistance, and high maintenance cost. The ball screw is operated most frequently in a numerical control machine tool, and each part often generates mechanical abrasion and poor lubrication, so that the faults of lowered positioning precision, overlarge reverse clearance, mechanical crawling, serious screw abrasion, overlarge noise and the like often occur.
In the embodiment, the data of the numerical control system and the data collected by the sensing nodes arranged on the key parts of the ball screw are combined to diagnose and predict the faults of the ball screw in the machine tool.
In the practical application of the fault diagnosis of the ball screw, the time domain data of the vibration speed and the acceleration of the key part has the characteristics of intuition and good real-time performance, some faults can directly reflect on the temperature, and the state code of the numerical control system has a certain direction guiding effect on a part of faults. The vibration time domain feature data in the mechanical state data related in the embodiment is as follows:
1. amplitude value
The peak value in the amplitude value reflects the maximum value of the amplitude at a certain time, and is suitable for diagnosing faults with instantaneous impact, such as surface pitting damage. The mean value of the amplitude values has substantially the same diagnostic effect as the peak value, and has the advantage that the detection value is more stable than the peak value. The RMS value of the amplitude value is averaged over time and is suitable for diagnosing faults such as wear, surface cracks, etc. where the amplitude value changes slowly over time.
2. Form factor
The form factor is defined as the ratio of the peak value to the mean value. When the waveform factor value is too large, the lead screw or the ball bearing can be corroded by points; when the form factor value is too small, the wear of the screw or the ball is indicated. The change time of the form factor is shorter, which indicates that the ball is more likely to have faults; and the longer the time of the form factor change, the lead screw or the ball is in failure.
3. Crest factor
The crest factor is defined as the ratio of the peak value to the root mean square value. The method is suitable for diagnosing the pitting type faults. By monitoring the change trend of the crest factor value along with time, early prediction can be effectively carried out, and the development change trend of faults can be reflected. When the rolling screw has no fault, the rolling screw is a small stable value; once the lead screw is damaged, an impact signal is generated, the vibration peak value is obviously increased, but the root mean square value is not obviously increased at the moment, so that the crest factor value is increased; when the fault is continuously expanded, the root mean square value starts to increase after the peak value gradually reaches the limit value, and the crest factor value gradually decreases until the value is recovered to the value without the fault.
4. Kurtosis coefficient
The calculation formula is as follows:
wherein the content of the first and second substances,
β2is a kurtosis coefficient; i XiL is the instantaneous amplitude at the ith moment;
is the amplitude average; σ is the standard deviation; n is the number of time instants.
When the amplitude of the faultless screw or ball meets the normal distribution rule, the kurtosis coefficient value is about 3. The kurtosis coefficient value has a similar trend of variation as the form factor as faults occur and progress. The method is suitable for diagnosing the surface flaws of the screw or the ball, especially early faults.
Since the stroke of the screw is much larger than the stroke of the balls, more data is required to diagnose and predict the health condition of the ball screw, and thus the health condition of the balls is diagnosed first.
The network is not laid in a workshop where the numerical control machine tool is located, gateway nodes are arranged near the numerical control machine tool and connected with the numerical control machine tool through a wired network, and a sensing node is arranged on each of bases at two ends of a ball screw pair and a screw. The method mainly comprises the following steps:
the gateway node completes time synchronization through NTP based on the wireless network; the gateway node is used as a root node, and the sensing node completes time synchronization through FTSP based on the wireless sensing ad hoc network; the numerical control system takes the gateway node as a main station to complete time synchronization through a field bus NCUC-bus.
Firstly, simultaneously acquiring data by a numerical control system, a sensing node and a gateway node according to a set acquisition strategy, wherein the duration time lengths are the same; then the numerical control system uploads the working state data to the gateway node, the sensing node extracts mechanical characteristic data and uploads the mechanical characteristic data to the gateway node, and the gateway node extracts numerical control characteristic data; and finally, the gateway node judges whether the machine tool is in a non-shutdown and non-standby state by using the numerical control characteristic data, and if so, the gateway node sends the mechanical characteristic data and the numerical control characteristic data to the cloud.
The acquisition frequency of the vibration acceleration is usually over 10kHz, so that the transmitted original data can seriously consume network flow, and through two-layer characteristic extraction of the sensing node and the gateway node, the network flow is greatly reduced, and the cloud centralized operation load is effectively reduced. Based on the method, the coordinated data mining of the numerical control system data and the mechanical system data of the machine tool in the fault diagnosis of the machine tool is realized for the first time on the basis of high-precision time synchronization, and the fault diagnosis precision can be effectively improved.
S102: the cloud receives the mechanical characteristic data and the numerical control characteristic data, performs fault diagnosis and prediction by using a preset fault diagnosis and prediction model, and sends a fault diagnosis and prediction result to the client.
At the cloud end, firstly, a communication unit receives mechanical characteristic data and numerical control characteristic data of a ball screw, writes the mechanical characteristic data and the numerical control characteristic data into a storage unit and forwards the mechanical characteristic data and the numerical control characteristic data to a computing unit; and then the computing unit carries out fault diagnosis and prediction on the mechanical characteristic data and the numerical control characteristic data of the ball screw, and writes the fault diagnosis and prediction results of the ball screw into the storage unit.
It should be noted that the predetermined fault diagnosis and prediction model includes, but is not limited to, a neural network model. In addition, the fault diagnosis and the training of the prediction model are the prior art, and are not described in detail here.
S103: the client displays the fault diagnosis and prediction results to the machine tool experts, receives the feedback results of the machine tool experts on the fault diagnosis and prediction results, and sends the feedback results to the cloud;
firstly, a service unit pushes the fault diagnosis and prediction results of a ball screw to a client; then the client displays the fault diagnosis and prediction results of the ball screw to a machine tool expert, the machine tool expert judges the fault diagnosis and prediction results of the ball screw and feeds back the judgment results; and finally, the cloud service unit receives a feedback result of the machine tool expert and writes the feedback result into the storage unit.
S104: the cloud receives the feedback result, and marks the mechanical characteristic data and the numerical control characteristic data to obtain sample data; and training a fault diagnosis and prediction model according to the sample data, and returning to execute the step S102.
Then, using a machine tool expert to perform fault diagnosis on the ball screw and a feedback result of a prediction result to finish the marking of mechanical characteristic data and numerical control characteristic data of the ball screw; then, screening the marked data to obtain sample data; and finally, correcting parameters in the fault diagnosis and prediction model of the ball screw by using machine learning according to the sample data, and feeding back the parameters to the computing unit so as to ensure that the computing unit processes subsequent mechanical characteristic data and numerical control characteristic data by using the updated model.
By applying the embodiment of the invention shown in fig. 1, based on the cooperation of edge calculation and a cloud, the invention acquires mechanical state data of a machine tool mechanical system and extracts characteristics at an edge end by using a sensing node to obtain mechanical characteristic data, acquires working state data of a machine tool numerical control system and extracts characteristics at a gateway node to obtain numerical control characteristic data, then carries out fault diagnosis and prediction on the mechanical characteristic data and the numerical control characteristic data at the cloud, then uses experts to feedback and mark the mechanical characteristic data and the numerical control characteristic data to obtain sample data, then uses the sample data to carry out fault diagnosis and model prediction updating, and then carries out multiple times of cyclic processing to obtain more accurate fault diagnosis and prediction models, thereby improving the accuracy of fault diagnosis and prediction.
In a specific implementation manner of the embodiment of the invention, a gateway node is used as a root node, and a sensing node completes time synchronization through FTSP based on a wireless sensing ad hoc network; the numerical control system completes time synchronization through a field bus NCUC-bus based on a wired network.
In a specific implementation manner of the embodiment of the present invention, the method further includes:
the gateway node receives an instruction sent by the client, wherein the instruction comprises but is not limited to adjusting the characteristic data acquisition frequency, adjusting the characteristic data acquisition duration, adjusting the type of the characteristic data, updating the characteristic extraction algorithm library, and restarting or closing the sensing node or the gateway node.
Firstly, a user issues an instruction to a machine tool at a client; then, at the cloud end, the business unit receives the instruction, writes the instruction into the storage unit, and issues the instruction to the gateway node of the corresponding edge end through the communication unit; and finally, the gateway node receives the instruction, if the target node is the gateway node, the instruction is directly executed, and if not, the instruction is forwarded to the sensing node for execution.
According to the embodiment of the invention, a large amount of synchronous real-time edge calculation replaces large-scale centralized calculation, small-scale feature data is uploaded instead of large-scale data uploading, a machine learning fault diagnosis and prediction model which can be continuously improved replaces a solidified fault knowledge base expert system, and the cloud end is used for adjusting the edge calculation to improve the expansion capability of the system, so that a fault diagnosis and prediction method which is low in flow, high in efficiency and continuously improved in diagnosis precision is realized along with continuous accumulation of real-time feature data, and the method has the advantages of better input-output ratio, better industrial universality and higher popularization value.
Example 2
Fig. 2 is a schematic structural diagram of a machine tool fault diagnosis and prediction system based on edge computing and cloud coordination according to an embodiment of the present invention, and fig. 3 is a schematic data processing diagram of an edge terminal in the machine tool fault diagnosis and prediction system based on edge computing and cloud coordination according to the embodiment of the present invention; as shown in fig. 2 and 3, the system for diagnosing and predicting machine tool faults based on edge computing and cloud coordination comprises an edge terminal and a cloud terminal, wherein,
and the edge end comprises a gateway node and a sensing node. The sensor nodes are disposed on a component of the machine tool and the gateway node may be deployed near the machine tool, for example, near a workstation of the machine tool. The gateway node is communicated with the numerical control system through a wired network, communicated with the sensing node through a wireless sensing ad hoc network and communicated with the cloud through a 4G network or a wired network. The gateway node is responsible for data acquisition, feature extraction and data uploading to the cloud; the sensing node is responsible for data acquisition, feature extraction and data uploading to the gateway node.
The sensing node is used for acquiring the mechanical state data of the component according to a preset first acquisition strategy, extracting the characteristics, acquiring mechanical characteristic data and uploading the characteristic data to the gateway node.
The gateway node is used for acquiring working state data of the numerical control system according to a preset second strategy and extracting characteristics to obtain numerical control characteristic data, and sending the mechanical characteristic data and the numerical control characteristic data to the cloud end;
the gateway node sends data to the cloud, and the mechanical characteristic data comprises but is not limited to one or a combination of temperature, humidity, rotating speed, torque, an amplitude value, a form factor, a crest factor, a kurtosis coefficient and the like; numerical control characteristic data comprises one or a combination of instruction codes, alarm codes, spindle rotating speed, moving direction, coordinates and the like;
the cloud end comprises a communication unit, a computing unit, a storage unit, a business unit and a training unit, wherein the communication unit is used for receiving the mechanical characteristic data and the numerical control characteristic data, writing the mechanical characteristic data and the numerical control characteristic data into the storage unit and forwarding the mechanical characteristic data and the numerical control characteristic data to the computing unit; the computing unit is used for carrying out fault diagnosis and prediction on the mechanical characteristic data and the numerical control characteristic data and writing the fault diagnosis and prediction results into the storage unit; the storage unit is used for storing mechanical characteristic data, numerical control characteristic data, feedback results of machine tool experts and various service data; the service unit is used for displaying the fault diagnosis and prediction results to the machine tool expert and receiving the feedback results of the machine tool expert on the fault diagnosis and prediction results; the training unit is used for training the fault diagnosis and prediction model according to the feedback result, the mechanical characteristic data and the numerical control characteristic data so that the result of the fault diagnosis and prediction model is the same as the feedback result of the machine tool expert, and the model parameters are fed back to the calculation unit.
At the cloud end, firstly, the communication unit receives mechanical characteristic data and numerical control characteristic data, writes the mechanical characteristic data and the numerical control characteristic data into the storage unit and forwards the mechanical characteristic data and the numerical control characteristic data to the computing unit; and then the computing unit carries out fault diagnosis and prediction on the mechanical characteristic data and the numerical control characteristic data, and writes the fault diagnosis and prediction results into the storage unit.
At the cloud end, firstly, a preset fault diagnosis and prediction model is used for carrying out fault diagnosis and prediction on received mechanical characteristic data and numerical control characteristic data; then displaying the fault diagnosis and prediction results to a machine tool expert, judging whether the fault diagnosis and prediction results are correct and feeding back the results by the machine tool expert according to the actual state of the machine tool, and marking mechanical characteristic data and numerical control characteristic data according to the feedback results to obtain sample data; and finally, correcting and evolving the fault diagnosis and prediction model by using machine learning, and feeding back to the calculation unit so as to ensure that the calculation unit processes subsequent mechanical characteristic data and numerical control characteristic data by using the adjusted model.
By applying the embodiment of the invention shown in fig. 2, based on the cooperation of edge calculation and a cloud, the invention acquires mechanical state data of a machine tool mechanical system and extracts characteristics by using a sensing node at an edge end to obtain mechanical characteristic data, acquires working state data of a machine tool numerical control system and extracts characteristics by using a gateway node to obtain numerical control characteristic data, performs fault diagnosis and prediction on the mechanical characteristic data and the numerical control characteristic data at the cloud, marks the mechanical characteristic data and the numerical control characteristic data by using a feedback result of a machine tool expert to obtain sample data, updates a fault diagnosis and prediction model by using the sample data, and performs multiple times of cyclic processing to obtain the most accurate fault diagnosis and prediction model, thereby improving the accuracy of fault diagnosis and prediction.
In addition, the embodiment of the invention also has flexible edge computing capability, abundant data types and more comprehensive and accurate fault diagnosis and prediction capability; the cloud-based storage and processing capacity is better in expansibility; the machine tool is not limited to fixed brands or models, and has better applicability; based on the cooperation of edge calculation and cloud, the transmitted data volume is greatly reduced, the fault diagnosis efficiency is effectively improved, potential risks are found in advance, and the method has good universality; based on the bidirectional cooperation of the edge calculation and the cloud, the edge calculation is adjusted remotely, so that the cost is effectively reduced, the income is improved, and the economic value is higher; based on a diagnosis/prediction-feedback mechanism, the method can quickly finish data annotation and promote model training, and has extremely high technical value.
According to the embodiment of the invention, the communication data volume is greatly reduced on the premise of keeping the core characteristics of data through cooperative edge computing, and finally, a closed loop cooperative diagnosis method comprising data uploading and instruction downlink is formed with the cloud, so that the causal connection of a machine tool numerical control system and a mechanical system in the machine tool fault is effectively utilized, a flexible data acquisition and characteristic data extraction mechanism and a mechanism that the cloud adjusts edge computing as required are realized, and the large-scale application of the wireless Internet of things in the machine tool fault diagnosis and prediction becomes possible.
Example 3
Fig. 4 is a schematic diagram of a process of issuing an instruction in a machine tool fault diagnosis and prediction system based on edge computing and cloud coordination according to an embodiment of the present invention, as shown in fig. 4, a difference between embodiment 3 of the present invention and embodiment 2 of the present invention is that a service unit is configured to issue an instruction to an edge terminal, where the instruction includes, but is not limited to, one or a combination of adjusting a data acquisition frequency or duration, an acquired data type, updating a feature extraction algorithm library, and restarting or closing a sensor node or a gateway node.
Firstly, a user issues an instruction to an edge end at a client, wherein the instruction comprises but is not limited to adjusting the acquisition frequency or duration of data, the type of the acquired data, updating a feature extraction algorithm library, restarting or closing a sensing node or a gateway node and the like; then the cloud service unit receives the instruction, writes the instruction into the storage unit and forwards the instruction to the communication unit; then the cloud communication unit receives the instruction and issues the instruction to the gateway node of the corresponding edge end; and finally, the gateway node receives the instruction, if the target node is the gateway node, the instruction is directly executed, and if not, the instruction is forwarded to the sensing node for execution.
By applying the embodiment of the invention, the flexible control of the cloud end to the edge end can be realized.
Example 4
The difference between embodiment 4 of the present invention and embodiment 2 of the present invention is that the client is further configured to receive an operation instruction of a user, and the client includes a Web application and a mobile phone App.
In practical application, firstly, a cloud service unit pushes a fault diagnosis and prediction result to a client; then the client displays the fault diagnosis and prediction results to a machine tool expert, and the machine tool expert judges the fault diagnosis and prediction results and feeds back the judgment results; and finally, the cloud service unit receives a feedback result of the machine tool expert and writes the feedback result into the storage unit.
By applying the embodiment of the invention, the expert can conveniently carry out fault diagnosis and judgment of the prediction result and give feedback in time.
Example 5
Fig. 5 is a schematic diagram of a first process of time synchronization in a system for diagnosing and predicting a fault of a machine tool based on edge computing and cloud coordination according to an embodiment of the present invention; fig. 6 is a schematic diagram of a second process of time synchronization in a machine tool fault diagnosis and prediction system based on edge computing and cloud coordination according to an embodiment of the present invention; fig. 7 is a schematic diagram of a third process of time synchronization in a machine tool fault diagnosis and prediction system based on edge computing and cloud coordination according to an embodiment of the present invention; as shown in fig. 5 to 6, embodiment 5 of the present invention is different from embodiment 2 of the present invention in that the gateway node is configured to complete time synchronization of the gateway node through the NTP network, and serve as a basis for edge time synchronization.
In practical application, the requirement of synchronous real-time data acquisition of the edge end on the time accuracy is high, so that the higher the accuracy of the time synchronization of each node is, the better the accuracy is.
When the industrial Ethernet supports Time synchronization, the gateway node completes Time synchronization (the precision is generally less than 50ms) through NTP (Network Time Protocol) based on the wired Network, otherwise, the gateway node completes Time synchronization through NTP based on the wireless Network.
The gateway node is used as a root node, and the sensing node completes Time Synchronization (with the precision less than 100 mus) through an FTSP (Flooding Time Synchronization Protocol) based on the wireless sensing ad hoc network.
When the industrial Ethernet supports time synchronization, the numerical control system completes time synchronization (the precision is less than 100ns) through a Field Bus NCUC-Bus (NC Union of China Field Bus, numerical control alliance Bus) based on a wired network, otherwise, the numerical control system completes time synchronization through the Field Bus NCUC-Bus by taking a gateway node as a main station.
Further, as shown in fig. 7, 1 gateway node and 8 sensing nodes may be ad hoc network, and 3R sensing nodes become sensing relay nodes; the two sensing nodes on the left side are communicated with the sensing node on the right side through the sensing relay node and further communicated to the gateway node GW. At the moment, FTSP is used, the gateway node is a root node, and all sensing node time synchronization is completed in three batches through the sensing relay nodes. Firstly, time synchronization is carried out on a sensing relay node R3 and sensing nodes S2 and S4 through a gateway node; then the sensing relay nodes R1 and R5 and the sensing node S6 are time-synchronized through the synchronized sensing relay node R3; finally, the sensing nodes S7 and S9 are time-synchronized by the already synchronized sensing relay node R5.
By applying the embodiment of the invention, high-precision time synchronization can be realized, thereby being beneficial to effectively exploring the cause and effect relationship of machine tool numerical control system and mechanical system data in machine tool faults;
example 6
The difference between embodiment 6 of the present invention and embodiment 2 of the present invention is that the gateway node is configured to determine whether the machine tool is in an un-powered off and un-standby state, and if so, send the mechanical feature data and the numerical control feature data to the cloud.
When data acquisition is carried out, firstly, a numerical control system, a sensing node and a gateway node respectively acquire data simultaneously according to a set acquisition strategy, and the duration time lengths are the same; then the numerical control system uploads the data to the gateway node, and the sensing node extracts characteristic data and uploads the characteristic data to the gateway node; and finally, the gateway node extracts feature data from the data acquired by the numerical control system, and judges whether the machine tool is in a non-shutdown and non-standby state, namely whether the machine tool is in a normal working state, if the machine tool is in the normal working state, the gateway node collects the feature data and uploads the feature data to the cloud, otherwise, the feature data is not uploaded.
The present invention is not limited to the above preferred embodiments, and any modifications, equivalent substitutions and improvements made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (7)
1. A machine tool fault diagnosis and prediction method based on edge computing and cloud coordination is characterized by comprising the following steps:
1) the sensing node acquires mechanical state data of the machine tool component according to a preset first acquisition strategy and extracts characteristics to obtain mechanical characteristic data, and the mechanical characteristic data is uploaded to the gateway node; the gateway node acquires working state data of the numerical control system according to a preset second acquisition strategy and extracts characteristics to obtain numerical control characteristic data, and sends the mechanical characteristic data and the numerical control characteristic data to a cloud terminal;
2) the cloud receives the mechanical characteristic data and the numerical control characteristic data, performs fault diagnosis and prediction by using a preset fault diagnosis and prediction model, and sends a fault diagnosis and prediction result to the client;
3) the client displays the fault diagnosis and prediction result to a machine tool expert, receives a feedback result of the machine tool expert on the fault diagnosis and prediction result, and sends the feedback result to the cloud end;
4) the cloud receives the feedback result, and marks the mechanical characteristic data and the numerical control characteristic data to obtain sample data; training the fault diagnosis and prediction model according to the sample data, and returning to execute the step 2);
the gateway node is used as a root node, and the sensing node completes time synchronization through FTSP based on the wireless sensing ad hoc network; the numerical control system completes time synchronization through a field bus NCUC-bus based on a wired network, 1 gateway node and 8 sensing nodes are self-organized into a network, and 3R sensing nodes are used as sensing relay nodes; the two sensing nodes on the left side are communicated with the sensing node on the right side through the sensing relay node and further communicated to the gateway node GW.
2. The edge computing and cloud coordination based machine tool fault diagnosis and prediction method according to claim 1, characterized in that the method further comprises:
and the gateway node receives instructions aiming at the sensing nodes and the gateway node, which are sent by the client, wherein the instructions include but are not limited to adjusting the characteristic data acquisition frequency, adjusting the characteristic data acquisition duration, adjusting the type of the characteristic data, updating a characteristic extraction algorithm library, and restarting or closing the sensing nodes or the gateway node.
3. The system for diagnosing and predicting the machine tool fault based on edge computing and cloud coordination is characterized by comprising an edge end and a cloud end, wherein,
the edge end comprises a gateway node and a sensing node, wherein the sensing node is arranged on a part of the machine tool, and is used for acquiring mechanical state data of the part according to a preset first acquisition strategy and extracting characteristics to obtain mechanical characteristic data;
the gateway node is used for acquiring working state data of the numerical control system according to a preset second strategy and extracting characteristics to obtain numerical control characteristic data, and sending the mechanical characteristic data and the numerical control characteristic data to the cloud end;
the gateway node is used for sending the mechanical characteristic data and the numerical control characteristic data to a cloud end, wherein the mechanical characteristic data comprises but is not limited to one or a combination of temperature, humidity, rotating speed, torque, amplitude value, form factor, crest factor and kurtosis factor; the numerical control characteristic data comprises but is not limited to one or a combination of an instruction code, an alarm code, a spindle rotating speed, a moving direction and a coordinate; the gateway node is used for completing time synchronization of the gateway node through NTP and serving as a basis for edge end time synchronization;
the cloud end comprises a communication unit, a computing unit, a storage unit, a service unit and a training unit, wherein the communication unit is used for receiving the mechanical characteristic data and the numerical control characteristic data uploaded by the gateway node, writing the mechanical characteristic data and the numerical control characteristic data into the storage unit and forwarding the mechanical characteristic data and the numerical control characteristic data to the computing unit;
the computing unit is used for carrying out fault diagnosis and prediction on the mechanical characteristic data and the numerical control characteristic data and writing fault diagnosis and prediction results into the storage unit;
the storage unit is used for storing the mechanical characteristic data, the numerical control characteristic data, the feedback result of a machine tool expert and various service data;
the business unit is used for displaying the fault diagnosis and prediction result to a machine tool expert and receiving a feedback result of the machine tool expert on the fault diagnosis and prediction result;
and the training unit is used for training a fault diagnosis and prediction model according to the feedback result, the mechanical characteristic data and the numerical control characteristic data so as to enable the result of the fault diagnosis and prediction model to be the same as the feedback result, and feeding back model parameters to the calculating unit.
4. The system of claim 3, wherein the business unit is configured to issue instructions to the edge terminal, wherein the instructions include, but are not limited to, adjusting the frequency or duration of data collection, the type of data collected, updating the feature extraction algorithm library, and restarting or shutting down one or a combination of the sensor node or the gateway node.
5. The system of claim 3, wherein the gateway node is configured to determine whether a machine tool is not powered off and in a standby state, and if so, send the mechanical feature data and the numerical control feature data to the cloud.
6. The edge computing and cloud coordination based machine tool fault diagnosis and prediction system according to claim 3, further comprising a client for displaying the fault diagnosis and prediction results to a machine tool expert and receiving feedback results of the machine tool expert on the fault diagnosis and prediction results.
7. The system of claim 3, wherein the client is further configured to receive an operation instruction of a user, and the client comprises a Web application and a cell phone App.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910235933.1A CN109933004B (en) | 2019-03-27 | 2019-03-27 | Machine tool fault diagnosis and prediction method and system based on edge computing and cloud cooperation |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910235933.1A CN109933004B (en) | 2019-03-27 | 2019-03-27 | Machine tool fault diagnosis and prediction method and system based on edge computing and cloud cooperation |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109933004A CN109933004A (en) | 2019-06-25 |
| CN109933004B true CN109933004B (en) | 2022-05-24 |
Family
ID=66988473
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910235933.1A Active CN109933004B (en) | 2019-03-27 | 2019-03-27 | Machine tool fault diagnosis and prediction method and system based on edge computing and cloud cooperation |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109933004B (en) |
Families Citing this family (22)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| TW202105100A (en) * | 2019-07-16 | 2021-02-01 | 神通資訊科技股份有限公司 | Abnormality detecting system on automatic processing machine and the method thereof |
| CN110749462B (en) * | 2019-07-19 | 2021-05-07 | 华瑞新智科技（北京）有限公司 | Industrial equipment fault detection method and system based on edge calculation |
| CN112394703B (en) * | 2019-08-14 | 2022-06-10 | 中车时代电动汽车股份有限公司 | Vehicle fault management system |
| CN110394688B (en) * | 2019-09-02 | 2021-02-12 | 太原科技大学 | Machine tool state monitoring method based on edge calculation |
| CN112486136B (en) * | 2019-09-11 | 2022-03-15 | 中科云谷科技有限公司 | Fault early warning system and method |
| CN110554657B (en) * | 2019-10-16 | 2023-03-03 | 河北工业大学 | Health diagnosis system and diagnosis method for operation state of numerical control machine tool |
| CN110781612A (en) * | 2019-11-14 | 2020-02-11 | 哈工大机器人（山东）智能装备研究院 | Fault diagnosis method and device for ball screw, computer device and storage medium |
| CN111507489B (en) * | 2020-04-20 | 2023-04-18 | 电子科技大学中山学院 | Cloud-edge-coordinated amusement equipment fault prediction and health management system and method |
| CN111594201B (en) * | 2020-05-28 | 2021-08-13 | 中铁工程装备集团有限公司 | TBM key parameter intelligent control system and method |
| CN111596629B (en) * | 2020-06-02 | 2021-08-03 | 曲阜师范大学 | Cloud-edge-collaborative industrial data fusion method and security controller |
| CN111737061B (en) * | 2020-06-11 | 2021-01-12 | 深圳千通科技有限公司 | Data processing method based on edge computing and 5G communication and central cloud server |
| CN111928334A (en) * | 2020-07-30 | 2020-11-13 | 瑞纳智能设备股份有限公司 | Heat supply control method and system based on edge cloud cooperative framework and computer equipment |
| CN112216085B (en) * | 2020-09-15 | 2022-05-10 | 青岛科技大学 | Equipment key load-bearing structural member health monitoring system based on edge calculation and online update sample intelligent identification |
| CN112051800A (en) * | 2020-09-17 | 2020-12-08 | 珠海格力智能装备有限公司 | Method and device for processing abnormality of machine tool, and computer-readable storage medium |
| CN112101532B (en) * | 2020-11-18 | 2021-02-12 | 天津开发区精诺瀚海数据科技有限公司 | Self-adaptive multi-model driving equipment fault diagnosis method based on edge cloud cooperation |
| CN113268047B (en) * | 2021-04-21 | 2022-09-02 | 深圳市道通科技股份有限公司 | Automobile diagnosis system and method and cloud server |
| CN112947290B (en) * | 2021-05-16 | 2021-08-20 | 北京赛博联物科技有限公司 | Edge cloud cooperation-based equipment state monitoring method and system and storage medium |
| CN113253037B (en) * | 2021-06-22 | 2021-10-08 | 北京赛博联物科技有限公司 | Current ripple-based edge cloud cooperative equipment state monitoring method and system and medium |
| CN113742855B (en) * | 2021-07-27 | 2022-03-18 | 清华大学 | Fault prediction method, system, electronic equipment and readable storage medium |
| CN113759854B (en) * | 2021-09-18 | 2023-09-05 | 富联裕展科技(深圳)有限公司 | Intelligent factory management and control system and method based on edge calculation |
| CN114273979B (en) * | 2021-12-10 | 2024-02-02 | 无锡市信昌电子科技有限公司 | Online monitoring system for processing engraving, milling, drilling and tapping numerical control system |
| CN115185234A (en) * | 2022-09-09 | 2022-10-14 | 中科航迈数控软件(深圳)有限公司 | Adaptive fault monitoring method and system for numerical control machine tool |
Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101118437A (en) * | 2007-09-03 | 2008-02-06 | 石毅 | New style numerically controlled machine remote condition monitoring and failure diagnosis system realizing method |
| CN207408818U (en) * | 2017-08-10 | 2018-05-25 | 黄河交通学院 | A kind of numerically-controlled machine tool with failure predication and diagnostic function |
| CN108363382A (en) * | 2018-02-09 | 2018-08-03 | 哈尔滨工业大学（威海） | A kind of complex equipment fault diagnosis method and system |
| CN108591104A (en) * | 2018-04-18 | 2018-09-28 | 广东寰球智能科技有限公司 | A kind of Research on Fan Fault Forecasting based on cloud platform and health management system arranged, method |
| CN108667725A (en) * | 2018-04-25 | 2018-10-16 | 哈尔滨电气股份有限公司 | A kind of industrial AnyRouter and implementation method based on a variety of accesses and edge calculations |
| CN108710362A (en) * | 2018-06-25 | 2018-10-26 | 济南浪潮高新科技投资发展有限公司 | A kind of CNC Machine Network diagnostic method based on industry internet platform |
| CN108805300A (en) * | 2018-05-30 | 2018-11-13 | 上海理工大学 | Numerically controlled machine remote live collaboration fault diagnosis and maintenance system |
| CN108828409A (en) * | 2018-08-03 | 2018-11-16 | 南方电网科学研究院有限责任公司 | A kind of fault detection system based on edge calculations |
| CN109118732A (en) * | 2018-08-27 | 2019-01-01 | 中国科学院上海微系统与信息技术研究所 | Sensor perception event judgement and alarm system and method based on marginal information |
| CN109460003A (en) * | 2018-10-17 | 2019-03-12 | 南京凯奥思数据技术有限公司 | Vehicle trouble predicts modeling method and system |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4228513A (en) * | 1978-11-13 | 1980-10-14 | The B. F. Goodrich Company | Monitor system and method for detecting sequential events in a cyclical process |
| EP3370125A1 (en) * | 2017-03-01 | 2018-09-05 | PLETHORA IIoT, S.L. | Device and system including multiple devices for supervision and control of machines in industrial installation |
| CN107160834B (en) * | 2017-07-14 | 2020-03-27 | 武汉红金龙印务股份有限公司 | Method for preventing moire in printing process during screen printing of flat paper |
| US10433726B2 (en) * | 2017-08-14 | 2019-10-08 | Amrita Vishwa Vidyapeetham | Systems, methods, and devices for remote health monitoring and management using internet of things sensors |
| CN108768843A (en) * | 2018-06-27 | 2018-11-06 | 安徽国讯芯微科技有限公司 | A kind of multi-internet integration edge calculations gateway |
-
2019
- 2019-03-27 CN CN201910235933.1A patent/CN109933004B/en active Active
Patent Citations (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN101118437A (en) * | 2007-09-03 | 2008-02-06 | 石毅 | New style numerically controlled machine remote condition monitoring and failure diagnosis system realizing method |
| CN207408818U (en) * | 2017-08-10 | 2018-05-25 | 黄河交通学院 | A kind of numerically-controlled machine tool with failure predication and diagnostic function |
| CN108363382A (en) * | 2018-02-09 | 2018-08-03 | 哈尔滨工业大学（威海） | A kind of complex equipment fault diagnosis method and system |
| CN108591104A (en) * | 2018-04-18 | 2018-09-28 | 广东寰球智能科技有限公司 | A kind of Research on Fan Fault Forecasting based on cloud platform and health management system arranged, method |
| CN108667725A (en) * | 2018-04-25 | 2018-10-16 | 哈尔滨电气股份有限公司 | A kind of industrial AnyRouter and implementation method based on a variety of accesses and edge calculations |
| CN108805300A (en) * | 2018-05-30 | 2018-11-13 | 上海理工大学 | Numerically controlled machine remote live collaboration fault diagnosis and maintenance system |
| CN108710362A (en) * | 2018-06-25 | 2018-10-26 | 济南浪潮高新科技投资发展有限公司 | A kind of CNC Machine Network diagnostic method based on industry internet platform |
| CN108828409A (en) * | 2018-08-03 | 2018-11-16 | 南方电网科学研究院有限责任公司 | A kind of fault detection system based on edge calculations |
| CN109118732A (en) * | 2018-08-27 | 2019-01-01 | 中国科学院上海微系统与信息技术研究所 | Sensor perception event judgement and alarm system and method based on marginal information |
| CN109460003A (en) * | 2018-10-17 | 2019-03-12 | 南京凯奥思数据技术有限公司 | Vehicle trouble predicts modeling method and system |
Also Published As
| Publication number | Publication date |
|---|---|
| CN109933004A (en) | 2019-06-25 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109933004B (en) | Machine tool fault diagnosis and prediction method and system based on edge computing and cloud cooperation | |
| CN101870075B (en) | Method for predicting service life of screw pair of numerical control machine on basis of performance degradation model | |
| CN101870076B (en) | Method for predicting service life of guide pair of numerical control machine on basis of performance degradation model | |
| CN113569475B (en) | Subway axle box bearing fault diagnosis system based on digital twin technology | |
| JP2018153901A (en) | Machine learning device and thermal displacement correction device | |
| US8639454B2 (en) | Method of efficacy anticipation and failure examination for an apparatus | |
| CN108620950B (en) | Method and system for monitoring machining state of turning tool | |
| CN108181105B (en) | Rolling bearing fault pre-diagnosis method and system based on logistic regression and J divergence | |
| CN115099633A (en) | Bearing processing equipment operation quality detection system based on data analysis | |
| CN112947290B (en) | Edge cloud cooperation-based equipment state monitoring method and system and storage medium | |
| CN111160652B (en) | Knowledge-sensing-based equipment abnormal state comprehensive judgment and operation and maintenance method | |
| CN114077919A (en) | System for predicting machining anomalies | |
| CN114326593B (en) | Cutter life prediction system and method | |
| Li et al. | A wavelet and neural networks based on fault diagnosis for HAGC system of strip rolling mill | |
| Albano et al. | Sensors: the enablers for proactive maintenance in the real world | |
| CN113632025A (en) | Methods, systems, and computer program products for assessing energy consumption in an industrial environment | |
| CN108537174B (en) | Online monitoring method and system for running state of rotating machinery under time-varying working condition | |
| Stoop et al. | Fleet learning of thermal error compensation in machine tools | |
| D'Emilia et al. | Prediction of the remaining useful life of mechatronic systems, using internal sensors | |
| CN109697271B (en) | Rolling bearing health assessment and performance prediction method based on short-time energy change ratio and nuclear extreme learning machine | |
| CN109632308B (en) | Deep groove ball rolling bearing inner ring fault degradation state identification method and system | |
| CN112990524A (en) | Residual error correction-based residual life prediction method for rolling bearing | |
| CN113290063B (en) | Monitoring and analyzing system and monitoring and analyzing method for sheet cold rolling mill | |
| CN113172931B (en) | Remote monitoring type intelligent control system and method for hydraulic single-row punching machine | |
| Chen et al. | Dynamic Drive Chain Error Analysis of Industrial Robots with Cyber Sensing Technology |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |