Automobile production line intelligent clamp monitoring method and system based on Internet of Things

Abstract

The invention relates to the technical field of intelligent manufacturing and Internet of Things monitoring, and discloses an intelligent fixture monitoring method and system for an automobile production line based on the Internet of Things, and the method comprises the steps: dividing a calculation region based on a Voronoi diagram; the state of the clamp is monitored in real time; screening key data based on the data importance index; constructing a triangulation network at the cloud; and carrying out position offset prediction monitoring by using a long-short-term memory network. Compared with the prior art that data processing depends on cloud computing, so that network bandwidth occupation is high, data transmission delay is large, and especially under the conditions that a production line runs at a high speed and the monitoring data volume is large, low-delay fixture state monitoring cannot be achieved, edge computing and an intelligent screening mechanism are introduced, so that the fixture state monitoring accuracy is improved. Cooperative processing at the local computing device and the cloud is realized, the problems of too high bandwidth occupation and untimely data processing are avoided, and the real-time performance of data processing and the overall efficiency of the system are improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of intelligent manufacturing and Internet of Things monitoring, and particularly relates to an intelligent fixture monitoring method and system for an automobile production line based on the Internet of Things. Background Art

[0002] Currently, in the intelligent manufacturing of automobile production lines, the status monitoring of fixtures is crucial for ensuring production accuracy and process stability. However, existing intelligent fixture monitoring systems still have many deficiencies. For example, traditional monitoring systems usually rely on centralized cloud computing for data analysis and processing. However, in a high-speed production line, this architecture has the following problems: High network bandwidth occupancy: All monitoring data is uploaded to the cloud for processing, resulting in a huge data flow, increasing communication costs, and potentially causing data transmission delays under high load; Large data processing delay: Since cloud computing involves remote data transmission and centralized processing, the system cannot respond in real time to key events such as fixture offset and vibration anomalies on the production line, resulting in detection and adjustment lags; Lack of intelligent data screening mechanism: Current systems often upload data in full volume. Even if some data has no practical analysis value, it still occupies cloud storage and computing resources, resulting in reduced computing efficiency; Inability to achieve accurate prediction: Existing monitoring systems usually make anomaly judgments based on fixed thresholds and lack intelligent prediction models, resulting in the inability to intervene in advance before a failure occurs and affecting production continuity. Therefore, there is an urgent need for an intelligent fixture monitoring method that combines edge computing, intelligent data screening, geometric optimization calculation, and machine learning prediction, so that in a high-data flow environment, it is still possible to achieve efficient and low-latency fixture status monitoring, and improve the fault prediction ability and the overall operation efficiency of the production line. Summary of the Invention

[0003] Aiming at the above-mentioned existing technical deficiencies, the purpose of the present invention is to propose an intelligent fixture monitoring method for an automobile production line based on the Internet of Things, aiming to solve the technical problem that data processing in the prior art relies on cloud computing, resulting in high network bandwidth occupancy and large data transmission delays, especially in the case of high-speed operation of the production line and a large amount of monitoring data, and being unable to achieve low-latency fixture status monitoring.

[0004] To solve the above technical problems, the present invention adopts the following technical solutions: The present invention provides an intelligent fixture monitoring method for an automobile production line based on the Internet of Things,

[0005] The intelligent fixture monitoring method for an automobile production line based on the Internet of Things includes:

[0006] Step S10: Obtain the basic load information of the monitoring node i of the intelligent fixture of the automobile production line, and calculate the real-time load balance score L according to the basic load information i , when L iWhen it is greater than the preset real-time load balancing sub-threshold, transfer the current computing task to the adjacent monitoring node, record the transfer flow of the computing task in real time, and set the computing area E to which the monitoring node i belongs using a Voronoi diagram;

[0007] Step S20: Obtain the set of monitoring points within the computing area E, collect the real-time status information of the intelligent fixture at time t within the set of monitoring points, and calculate the data importance index;

[0008] Step S30: Determine whether to upload the real-time status information of the intelligent fixture in the computing area E to the cloud for processing according to the data importance index;

[0009] Step S40: Receive the real-time status information of the intelligent fixture in the cloud and construct a triangular data network using the triangulation algorithm accelerated by the K-dimensional binary tree, and output a fused data set;

[0010] Step S50: Use the fused data set as input and predict the predicted value P of the position offset of the intelligent fixture using a long short-term memory network fault , if P fault is greater than the preset offset threshold, trigger a monitoring alarm instruction.

[0011] Preferably, in step S10, the basic load information includes the initial position (x i , y i ), CPU utilization rate C i , memory occupancy rate M i and network bandwidth utilization rate B i ; the calculation formula for the real-time load balancing score L i is where β is the preset adjustment coefficient for the urgency of the monitoring point.

[0012] Preferably, in step S20, the real-time status information of the intelligent fixture includes the position data P t , alignment error A t and vibration value V t , and the calculation formula for the data importance index DII is where P ref is the standard position of the preset fixture, A ref is the maximum allowable alignment error of the preset fixture, V ref is the normal range of vibration of the preset fixture, w 1 , w 2 , w 3 are data weight factors.

[0013] Preferably, in step S30, the step of determining whether to upload the real-time status information of the intelligent fixture in the calculation area E to the cloud for processing according to the data importance index specifically includes: presetting a data importance index threshold θ, and if DII>θ, marking the real-time status information of the intelligent fixture as critical data D key ; if DII≤θ, storing the real-time status information of the intelligent fixture in the local edge computing device; where the local edge computing device includes an embedded development board inside the production line intelligent fixture and stores JSON format data using a hierarchical storage architecture.

[0014] Preferably, in step S40, the step of receiving the real-time status information of the intelligent fixture in the cloud and constructing a triangular data network using the triangulation algorithm accelerated by the K-dimensional binary tree specifically includes: introducing the K-dimensional binary tree data structure and using the K-dimensional binary tree data structure to accelerate the processing of the critical data D key and finding the nearest neighbor of each point in the time complexity of O(logn), and then constructing a triangular data network.

[0015] Preferably, in step S10, in the step of setting the calculation area E to which the monitoring node i belongs using the Voronoi diagram, the calculation area E satisfies the condition d(x i ,y i ,E)≤d(x i ,y i ,E * ), where E * is any calculation area other than E, d(x,y,E) is the Euclidean distance from the position (x i ,y i ) of the monitoring node to the center of the area E, and d(x i ,y i ,E * ) is the Euclidean distance from the position (x i ,y i ) of the monitoring node to the center of the calculation area E * .

[0016] Preferably, in step S50, the step of predicting the predicted value P fault of the position offset of the intelligent fixture using the long short-term memory network specifically includes:

[0017] Cleaning the critical data D mergef fused from step S40 to obtain a standardized data set D norm , constructing a long short-term memory network, where the input of the long short-term memory network is the standardized data set D norm , and the output is the fixture position offset value at a future moment;

[0018] Use the preset historical data to train the long short - term memory network. The training objective is to optimize the parameters of the long short - term memory network by minimizing the mean - square error loss function.

[0019] Apply the trained long short - term memory network to real - time data to predict the position offset of the intelligent fixture and obtain the predicted value P of the position offset of the intelligent fixture. fault 。

[0020] The present invention also provides an intelligent fixture monitoring system for an automobile production line based on the Internet of Things, including:

[0021] A task allocation module, which is used to obtain the basic load information of the monitoring node i of the intelligent fixture on the automobile production line, calculate the real - time load balance score L. i When L i is greater than the preset real - time load balance score threshold, transfer the current calculation task to the adjacent monitoring node, record the flow direction of the calculation task transfer in real - time, and set the calculation area E to which the monitoring node i belongs using a Voronoi diagram.

[0022] A data importance calculation module, which is used to obtain the set of monitoring points within the calculation area E, collect the real - time status information of the intelligent fixture at time t within the set of monitoring points, and calculate the data importance index.

[0023] A data upload decision module, which is used to decide whether to upload the real - time status information of the intelligent fixture in the calculation area E to the cloud for processing according to the data importance index.

[0024] A cloud data processing module, which is used to receive the real - time status information of the intelligent fixture in the cloud and construct a triangular data network using the triangulation algorithm accelerated by a K - dimensional binary tree, and output a fused data set.

[0025] An intelligent monitoring and warning module, which is used to take the fused data set as input, and use the long short - term memory network to predict the predicted value P of the position offset of the intelligent fixture. fault If P fault is greater than the preset offset threshold, trigger a monitoring alarm instruction.

[0026] The present invention also provides a computer program product, including an intelligent fixture monitoring program for an automobile production line based on the Internet of Things. When the intelligent fixture monitoring program for an automobile production line based on the Internet of Things is executed by a processor, it implements the above - mentioned intelligent fixture monitoring method for an automobile production line based on the Internet of Things.

[0027] The beneficial effects of the present invention are as follows: Compared with the prior art where data processing relies on cloud computing, resulting in high network bandwidth occupancy and large data transmission delays, especially under the conditions of high-speed operation of the production line and a large amount of monitoring data, it is impossible to achieve low-latency fixture status monitoring. By introducing edge computing and an intelligent screening mechanism, the present invention realizes collaborative processing between local computing devices and the cloud, avoiding problems such as excessive bandwidth occupancy and untimely data processing, and improving the real-time performance of data processing and the overall efficiency of the system.

[0028] Through the LSTM network model, the present invention can predict future fixture position offsets based on historical data, and perform real-time monitoring and adjustment. During the actual production process, if the predicted value is greater than the set threshold, the alarm mechanism can be triggered to ensure the accuracy and stability of the production line. BRIEF DESCRIPTION OF THE DRAWINGS

[0029] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can also be obtained based on these drawings.

[0030] Figure 1 It is a schematic flowchart of the first embodiment of an intelligent fixture monitoring method for an automotive production line based on the Internet of Things according to the present invention.

[0031] Figure 2 It is a schematic diagram showing the relationship between load balancing and the utilization rate of each resource in the first embodiment.

[0032] Figure 3 It is a schematic diagram of a heat map showing the correlation between resource utilization rate and load balancing score in the first embodiment.

[0033] Figure 4 It is a schematic diagram showing the division of the node calculation area in the first embodiment.

[0034] Figure 5 It is a schematic diagram of a triangular mesh and nearest neighbor analysis in the first embodiment.

[0035] Figure 6 It is a schematic diagram of LSTM predicting fixture position offset and an alarm mechanism in the first embodiment.

[0036] Figure 7 It is a schematic diagram of the device of an intelligent fixture monitoring method for an automotive production line based on the Internet of Things according to the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0037] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0038] Embodiment 1: As Figure 1 shown, it is a schematic flowchart of the first embodiment of the intelligent fixture monitoring method for an automobile production line based on the Internet of Things of the present invention, and the first embodiment of the intelligent fixture monitoring method for an automobile production line based on the Internet of Things of the present invention is proposed.

[0039] In the first embodiment, the intelligent fixture monitoring method for an automobile production line based on the Internet of Things includes:

[0040] Step S10: Obtain the basic load information of the monitoring node i of the intelligent fixture on the automobile production line, and calculate the real-time load balance score L i , when L i is greater than the preset real-time load balance score threshold, transfer the current calculation task to the adjacent monitoring node, record the transfer flow of the calculation task in real time, and set the calculation area E to which the monitoring node i belongs using a Voronoi diagram;

[0041] It should be noted that in step S10, the basic load information includes the initial position (x i , y i ), CPU utilization rate C i , memory occupancy rate M i and network bandwidth utilization rate B i ; the calculation formula of the real-time load balance score L i is where β is a preset adjustment coefficient for the urgency of the monitoring point. In step S10, in the step of setting the calculation area E to which the monitoring node i belongs using a Voronoi diagram, the calculation area E satisfies the condition d(x i , y i , E) ≤ d(x i , y i , E * ), where E * is any calculation area other than E, d(x, y, E) is the Euclidean distance from the position (x i , y i ) of the monitoring node to the center of the area E, and d(x i , y i , E * ) is the Euclidean distance from the position (x i , y i ) of the monitoring node to the calculation area E *Euclidean distance to the center.

[0042] It should be understood that by introducing the load balancing score calculation formula, the load conditions of each monitoring node can be accurately evaluated. In the case of excessive load, tasks will be transferred to nodes with lighter loads, thus achieving more efficient computing resource allocation. Using the Voronoi diagram for computing area division ensures that each monitoring node divides the most suitable computing area according to its physical location and load status. This method can reasonably avoid task conflicts between nodes and improve the allocation efficiency of computing tasks.

[0043] For example, there are 5 monitoring nodes, labeled A, B, C, D, and E respectively. The CPU utilization rate, memory occupancy rate, bandwidth utilization rate, and load balancing score of each node are recorded as follows: {"Node": "Node A", "CPU Utilization Rate (%)": 80, "Memory Occupancy Rate (%)": 70, "Bandwidth Utilization Rate (%)": 85, "Load Balancing Score": 0.75}, {"Node": "Node B", "CPU Utilization Rate (%)": 90, "Memory Occupancy Rate (%)": 80, "Bandwidth Utilization Rate (%)": 95, "Load Balancing Score": 0.80}, {"Node": "Node C", "CPU Utilization Rate (%)": 60, "Memory Occupancy Rate (%)": 50, "Bandwidth Utilization Rate (%)": 70, "Load Balancing Score": 0.60}, {"Node": "Node D", "CPU Utilization Rate (%)": 65, "Memory Occupancy Rate (%)": 55, "Bandwidth Utilization Rate (%)": 80, "Load Balancing Score": 0.70}, {"Node": "Node E", "CPU Utilization Rate (%)": 85, "Memory Occupancy Rate (%)": 75, "Bandwidth Utilization Rate (%)": 90, "Load Balancing Score": 0.77}, as Figure 2 shown, through the scatter plot, it can be intuitively seen that the data is relatively concentrated around 0.775, and the real-time load balancing score threshold refers to this value; as Figure 3 shown, the heat map shows the correlation between the CPU utilization rate, memory occupancy rate, bandwidth utilization rate, and load balancing score. By observing the shade of the color, it can be intuitively understood that the correlation between resource usages is relatively strong; as Figure 4 shown, by using the Voronoi diagram for computing area division, different computing areas can be obtained.

[0044] The solution of the present invention has higher dynamic scheduling ability compared with the traditional technology. It can not only respond to the resource usage situation in real time but also perform intelligent resource adjustment and task migration when the load is too heavy.

[0045] Step S20: Obtain the set of monitoring points within the computing area E, collect the real-time status information of the intelligent fixture at time t within the set of monitoring points, and calculate the data importance index;

[0046] It should be noted that in step S20, the real-time status information of the intelligent fixture includes the position data P t , the alignment error A t and the vibration value V t . The calculation formula of the data importance index DII is where P ref is the standard position of the preset fixture, A ref is the maximum allowable alignment error of the preset fixture, V ref is the normal range of the preset fixture vibration, w 1 , w 2 , w 3 are data weight factors.

[0047] It should be understood that the design concept of this formula comes from the method of multi-factor comprehensive evaluation, which can simultaneously consider the position error, position deviation and vibration error of the fixture, and obtain the data importance index (DII) through weighted calculation, so as to comprehensively reflect the working state of the fixture. This formula evaluates the current state of the intelligent fixture by calculating the relative errors of the current position, position deviation and vibration value of the intelligent fixture from the preset state. The larger the calculated DII value is, the more the real-time state of the fixture deviates from the preset standard state, which means that more resources are needed or adjustments are required.

[0048] Step S30: Determine whether to upload the real-time status information of the intelligent fixture in the calculation area E to the cloud for processing according to the data importance index;

[0049] It should be noted that in step S30, the step of determining whether to upload the real-time status information of the intelligent fixture in the calculation area E to the cloud for processing according to the data importance index specifically includes: presetting a data importance index threshold θ. If DII > θ, the real-time status information of the intelligent fixture is marked as key data D key ; if DII ≤ θ, the real-time status information of the intelligent fixture is stored in the local edge computing device; among them, the local edge computing device includes an embedded development board inside the production line intelligent fixture, and stores JSON format data using a hierarchical storage architecture.

[0050] It should be understood that according to the value of DII, it is possible to determine in real time whether to upload data or save data locally. This dynamic decision-making mechanism can effectively reduce the pressure on the bandwidth and optimize the utilization rate of network resources. For the real-time status information that does not need to be processed immediately, it is saved on the edge computing device, thereby reducing the burden on the cloud and enhancing the response speed.

[0051] Step S40: Receive the real-time status information of the intelligent fixture in the cloud and construct a triangular data network using a triangulation algorithm accelerated by a K-dimensional binary tree, and output a fused data set;

[0052] It should be noted that in step S40, the steps of receiving the real-time status information of the intelligent fixture in the cloud and constructing a triangular data network using a triangulation algorithm accelerated by a K-dimensional binary tree specifically include: introducing a K-dimensional binary tree data structure, using the K-dimensional binary tree data structure to accelerate the processing of key data D key Perform acceleration processing, find the nearest neighbor of each point in O(logn) time complexity, and then construct a triangular data network.

[0053] It can be understood that a K-dimensional binary tree is a data structure used in multi-dimensional space, especially suitable for solving the "nearest neighbor search" problem, and is used to quickly find the closest node in the triangulation network. It recursively divides the data points into several regions to form a binary tree structure. When querying, only the relevant part of the K-d tree needs to be traversed, so as to find the target data within the time complexity of O(logn).

[0054] It should be understood that by using a K-d Tree, the system can query the nearest neighbor of data points with a relatively low time complexity (O(logn)), thus accelerating the construction of the triangular network. This technology can significantly improve the efficiency of real-time data processing, especially suitable for the production line environment with dynamic changes and large data volume.

[0055] For example, as Figure 5 shown, it shows how to divide a point set into several triangular regions through triangulation. By using a K-dimensional binary tree and a Delaunay triangular network, we can not only efficiently query the nearest neighbor data points, but also construct a clear data network, which has an important optimization effect on the real-time monitoring and data processing of intelligent fixtures on the production line.

[0056] Step S50: Use the fused data set as input and adopt a long short-term memory network to predict the predicted value P of the position offset of the intelligent fixture fault , if P fault is greater than the preset offset threshold, trigger a monitoring alarm instruction.

[0057] It should be noted that in step S50, the steps of using a long short-term memory network to predict the predicted value P of the position offset of the intelligent fixture fault specifically include:

[0058] Clean the key data D fused from step S40 to obtain a standardized data set D merged , construct a long short-term memory network, and the input of the long short-term memory network is the standardized data set D norm , norm, the output is the fixture position offset value at a future moment;

[0059] Use the preset historical data for the training of the long short-term memory network. The training objective is to optimize the parameters of the long short-term memory network by minimizing the mean square error loss function.

[0060] Apply the trained long short-term memory network to the real-time data to predict the intelligent fixture position offset and obtain the predicted value P of the intelligent fixture position offset. fault .

[0061] It should be understood that through the LSTM network model, the future fixture position offset can be predicted based on the historical data, and real-time monitoring and adjustment can be carried out. In the actual production process, if the predicted value is greater than the set threshold, the alarm mechanism can be triggered to ensure the accuracy and stability of the production line.

[0062] For example, the following data exists in the input fusion dataset:

[0063]

[0064] Such as Figure 6 As shown, through the LSTM network model, the prediction effect of the LSTM model on the fixture position offset can be intuitively seen. The display of the alarm point can clearly indicate the situation when the predicted value exceeds the set threshold, thus prompting the system to trigger the alarm mechanism to ensure the accuracy and stability of the production line.

[0065] Embodiment 2: In addition, an intelligent fixture monitoring system for an automobile production line based on the Internet of Things provided by the present invention adopts an intelligent fixture monitoring method for an automobile production line based on the Internet of Things in the above embodiment, and can solve the technical problem of intelligent fixture monitoring for an automobile production line based on the Internet of Things. Compared with the prior art, the beneficial effects of the intelligent fixture monitoring system for an automobile production line based on the Internet of Things provided by the present invention are the same as those of the intelligent fixture monitoring method for an automobile production line based on the Internet of Things provided in the above embodiment, and other technical features in the intelligent fixture monitoring system for an automobile production line based on the Internet of Things are the same as the features disclosed in the above embodiment method, and will not be elaborated here.

[0066] Embodiment 3: The present invention provides an intelligent fixture monitoring device for an automobile production line based on the Internet of Things. Please refer to Figure 7, An intelligent fixture monitoring device for an automotive production line based on the Internet of Things includes: at least one processor; and a memory communicatively connected to the at least one processor; wherein the memory stores instructions executable by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute a method for monitoring intelligent fixtures on an automotive production line based on the Internet of Things in the first embodiment above. The intelligent fixture monitoring device for an automotive production line based on the Internet of Things in the embodiments of the present invention may include, but is not limited to, mobile terminals such as mobile phones, laptop computers, digital broadcast receivers, PDAs (Personal Digital Assistant), PADs (Portable Application Description), PMPs (Portable Media Player), in-vehicle terminals (such as in-vehicle navigation terminals), etc., and fixed terminals such as digital TVs, desktop computers, etc. The intelligent fixture monitoring device for an automotive production line based on the Internet of Things is merely an example and should not impose any limitations on the functions and scope of use of the embodiments of the present invention. The intelligent fixture monitoring device for an automotive production line based on the Internet of Things may include a processing device 1001 (such as a central processing unit, a graphics processing unit, etc.), which may perform various appropriate actions and processes according to a program stored in a read-only memory (ROM: Read Only Memory) 1002 or a program loaded from a storage device 1003 into a random access memory (RAM: Random Access Memory) 1004. In the RAM 1004, various programs and data required for the operation of the intelligent fixture monitoring device for an automotive production line based on the Internet of Things are also stored. The processing device 1001, the ROM 1002, and the RAM 1004 are connected to each other through a bus 1005. An input / output (I / O) interface 1006 is also connected to the bus. Generally, the following systems may be connected to the I / O interface 1006: an input device 1007 including, for example, a touch screen, a touchpad, a keyboard, a mouse, an image sensor, a microphone, an accelerometer, a gyroscope, etc.; an output device 1008 including, for example, a liquid crystal display (LCD: Liquid Crystal Display), a speaker, a vibrator, etc.; a storage device 1003 including, for example, a magnetic tape, a hard disk, etc.; and a communication device 1009. The communication device 1009 may allow the intelligent fixture monitoring device for an automotive production line based on the Internet of Things to communicate with other devices wirelessly or wiredly to exchange data. Although the figure shows an intelligent fixture monitoring device for an automotive production line based on the Internet of Things having various systems, it should be understood that it is not required to implement or have all the systems shown. More or fewer systems may be implemented or had alternatively.

[0067] Embodiment 4: The present invention further provides a computer program product, including a computer program which, when executed by a processor, implements the steps of a method for intelligent fixture monitoring of an automotive production line based on the Internet of Things as described above. The computer program product provided by the present invention can solve the technical problem of intelligent fixture monitoring of an automotive production line based on the Internet of Things. Compared with the prior art, the beneficial effects of the computer program product provided by the present invention are the same as those of the method for intelligent fixture monitoring of an automotive production line based on the Internet of Things provided in the above embodiment, and will not be elaborated here.

[0068] Specifically, according to the embodiments disclosed by the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, the embodiments disclosed by the present invention include a computer program product which includes a computer program carried on a computer-readable medium, and the computer program contains program codes for executing the methods shown in the flowcharts. In such an embodiment, the computer program can be downloaded and installed from the network through a communication device, or installed from a storage device 1003, or installed from a ROM 1002. When the computer program is executed by a processing device 1001, it executes the above functions defined in the methods of the embodiments disclosed by the present invention.

[0069] It should be understood that the various parts disclosed by the present invention can be implemented by hardware, software, firmware or a combination thereof. In the description of the above embodiments, specific features, structures, materials or characteristics can be combined in a suitable manner in any one or more embodiments or examples.

[0070] Obviously, those skilled in the art can make various modifications and variations to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalent technologies, the present invention also intends to include these modifications and variations.

Claims

1. An intelligent fixture monitoring method for automobile production line based on Internet of Things, characterized in that: Methods include: Step S10: Obtain basic load information of monitoring node i of the intelligent fixture of the automobile production line, and calculate the real-time load balancing distribution L according to the basic load information. i , when L i When it is greater than the preset real-time load balancing threshold, the current computing task is transferred to the adjacent monitoring node, the computing task transfer flow is recorded in real time, and the Voronoi diagram is used to set the computing area E to which the monitoring node i belongs; Step S20: obtaining a set of monitoring points in the calculation area E, collecting the real-time status information of the intelligent fixture at time t in the set of monitoring points, and calculating the data importance index; Step S30: Determine whether to upload the real-time status information of the smart fixture in the calculation area E to the cloud for processing according to the data importance index; Step S40: receiving the real-time status information of the intelligent fixture in the cloud and constructing a triangulated data network using a triangulation algorithm accelerated by a K-dimensional binary tree, and outputting a fused data set; Step S50: Taking the fused data set as input, using the long short-term memory network to predict the predicted value P of the intelligent fixture position offset fault , if P fault If it is greater than the preset deviation threshold, the monitoring alarm command is triggered.

2. The method for monitoring intelligent fixtures of automobile production lines based on the Internet of Things according to claim 1, characterized in that: In step S10, the basic load information includes the initial position (x i ,y i ), CPU utilization C i 、Memory usage M i and network bandwidth utilization B i ; Real-time load balancing i The calculation formula is Among them, β is the preset monitoring point emergency adjustment coefficient.

3. The method for monitoring intelligent fixtures of automobile production lines based on the Internet of Things as claimed in claim 2, characterized in that: In step S20, the real-time status information of the intelligent fixture includes position data P t , Alignment error A t and vibration value V t , the calculation formula of data importance index DII is Among them, P ref is the preset standard position of the fixture, A ref is the maximum alignment error allowed by the preset fixture, V ref is the preset normal range of fixture vibration, w1, w2, w3 are data weight factors.

4. The method for monitoring intelligent fixtures of automobile production lines based on the Internet of Things as claimed in claim 3, characterized in that: In step S30, the step of determining whether to upload the real-time status information of the intelligent fixture in the calculation area E to the cloud for processing according to the data importance index specifically includes: presetting a data importance index threshold θ, and if DII>θ, marking the real-time status information of the intelligent fixture as key data D key ; If DII≤θ, the real-time status information of the smart fixture is stored in the local edge computing device; wherein the local edge computing device includes an embedded development board inside the smart fixture of the production line, and a hierarchical storage architecture is used to store JSON format data.

5. The method for monitoring intelligent fixtures of automobile production lines based on the Internet of Things as claimed in claim 4, characterized in that: In step S40, the steps of receiving the real-time status information of the intelligent fixture in the cloud and using the triangulation algorithm accelerated by the K-dimensional binary tree to construct the triangulation data network specifically include: introducing the K-dimensional binary tree data structure, using the K-dimensional binary tree data structure to triangulate the key data D key Accelerate the processing and find the nearest neighbor of each point in O(logn) time complexity to build a triangulated data network.

6. The method for monitoring intelligent fixtures of automobile production lines based on the Internet of Things as claimed in claim 1, characterized in that: In step S10, in the step of using the Voronoi diagram to set the calculation area E to which the monitoring node i belongs, the calculation area E satisfies the condition d(x i ,y i ,E)≤d(x i ,y i ,E * ), where E * is any calculation area outside E, d(x,y,E) is the location of the monitoring node (x i ,y i ) to the center of region E, d(x i ,y i ,E * ) is the location of the monitoring node (x i ,y i ) to the calculation area E * Euclidean distance from the center.

7. The method for monitoring intelligent fixtures of automobile production lines based on the Internet of Things as claimed in claim 1, characterized in that: In step S50, the long short-term memory network is used to predict the predicted value P of the intelligent fixture position offset. fault The steps include: The key data D after fusion from step S40 merged Clean and get the standardized data set D norm , construct a long short-term memory network, the input of the long short-term memory network is the standardized data set D norm , the output is the fixture position offset value at a certain moment in the future; Use the preset historical data to train the LSTM network. The training goal is to optimize the parameters of the LSTM network by minimizing the mean square error loss function. The trained LSTM network is applied to real-time data to predict the position offset of the smart fixture and obtain the predicted value P of the smart fixture position offset. fault .

8. An intelligent fixture monitoring system for automobile production lines based on the Internet of Things, characterized in that: The automobile production line intelligent fixture monitoring system based on the Internet of Things includes: The task allocation module is used to obtain the basic load information of the monitoring node i of the intelligent fixture of the automobile production line, and calculate the real-time load balancing distribution L according to the basic load information. i , when L i When it is greater than the preset real-time load balancing threshold, the current computing task is transferred to the adjacent monitoring node, the computing task transfer flow is recorded in real time, and the Voronoi diagram is used to set the computing area E to which the monitoring node i belongs; The data importance calculation module is used to obtain the monitoring point set in the calculation area E, collect the real-time status information of the intelligent fixture at time t in the monitoring point set, and calculate the data importance index; The data upload decision module is used to decide whether to upload the real-time status information of the intelligent fixture in the calculation area E to the cloud for processing according to the data importance index; The cloud data processing module is used to receive the real-time status information of the intelligent fixture in the cloud and use the triangulation algorithm accelerated by the K-dimensional binary tree to construct a triangulated data network and output a fused data set; The intelligent monitoring and early warning module is used to take the fused data set as input and use the long short-term memory network to predict the predicted value P of the intelligent fixture position offset. fault , if P fault If it is greater than the preset deviation threshold, the monitoring alarm command is triggered.

9. An intelligent fixture monitoring device for automobile production line based on Internet of Things, characterized in that: The IoT-based automobile production line intelligent fixture monitoring device comprises: a memory, a processor, and an IoT-based automobile production line intelligent fixture monitoring program stored in the memory and executable on the processor. When the IoT-based automobile production line intelligent fixture monitoring program is executed by the processor, the IoT-based automobile production line intelligent fixture monitoring method according to any one of claims 1 to 7 is implemented.

10. A computer program product, characterized in that The computer program product includes an intelligent fixture monitoring program for an automobile production line based on the Internet of Things. When the intelligent fixture monitoring program for an automobile production line based on the Internet of Things is executed by a processor, the intelligent fixture monitoring method for an automobile production line based on the Internet of Things described in any one of Claims 1 to 7 is implemented.

Images