Tool pouch tightness detection method and device of machine tool magazine and computer program product

Abstract

The invention discloses a tool pouch tightness detection method and device of a machine tool magazine and a computer program product. The method comprises the steps that multiple kinds of tightness detection data between a tool and a tool pouch are obtained, and the tightness detection data are data collected by a sensor deployed in the tool pouch in the tightness detection process after the tool enters the tool pouch; feature extraction is carried out on the multiple kinds of tightness detection data, and target feature data corresponding to each kind of tightness detection data is obtained; inputting the target feature data into a tightness state determination model, and processing the target feature data by using the tightness state determination model to obtain the tightness degree between the tool pouch and the tool; and determining the tightness state between the cutter sleeve and the cutter according to the tightness threshold range of the tightness degree. According to the invention, the technical problems of low detection efficiency and detection precision limitation caused by the fact that a traditional tool tightness detection method mainly depends on manual detection or a simple mechanical device in the prior art are solved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent manufacturing and automated control technology, and in particular to a method and device for detecting the tightness of a tool sleeve of a machine tool magazine, and a computer program product. Background Art

[0002] In modern manufacturing, machine tool tool magazines, as important tool management systems, are widely used in automated machining processes. Tool magazines are responsible for storing and managing different types of cutting tools. The tightness of the tools directly affects the machining accuracy, efficiency and safety of machine tools. However, traditional tool tightness detection methods mainly rely on manual inspection or simple mechanical devices. These methods have some limitations, including: low efficiency of manual inspection, limited mechanical detection accuracy, poor real-time performance, and potential safety hazards.

[0003] In view of the problem that the traditional tool tightness detection method in the above-mentioned related technologies mainly relies on manual inspection or simple mechanical devices, resulting in low inspection efficiency and limited detection accuracy, no effective solution has been proposed yet. Summary of the invention

[0004] The embodiments of the present invention provide a method and device for detecting the tightness of a tool sleeve of a machine tool magazine, and a computer program product, so as to at least solve the technical problem that the traditional tool tightness detection method in the related art mainly relies on manual inspection or simple mechanical devices, resulting in low inspection efficiency and limited detection accuracy.

[0005] According to one aspect of an embodiment of the present invention, a method for detecting the tightness of a tool sleeve of a machine tool magazine is provided, comprising: acquiring a plurality of tightness detection data between a tool and a tool sleeve, wherein the tightness detection data is data collected by a sensor deployed in the tool sleeve during a tightness detection process after the tool enters the tool sleeve; performing feature extraction on the plurality of tightness detection data to obtain target feature data corresponding to each of the tightness detection data; inputting the target feature data into a tightness state determination model to process the target feature data using the tightness state determination model to obtain the tightness between the tool sleeve and the tool, wherein the tightness state determination model is trained by deep learning using multiple sets of training data, each of the multiple sets of training data comprising: sample target feature data, and a sample tightness corresponding to the sample target feature data; determining the tightness between the tool sleeve and the tool according to a tightness threshold range in which the tightness is located.

[0006] Optionally, obtaining the tightness detection data between the tool and the tool sleeve includes at least one of the following: obtaining the spacing distance between the tool and the bottom of the tool sleeve detected by a photoelectric sensor; obtaining the clamping force between the tool sleeve and the tool detected by a force sensor, wherein the clamping force is used to reflect the degree of fixation between the tool sleeve and the tool; obtaining the vibration frequency of the tool sleeve and the tool during machine tool processing detected by a vibration sensor.

[0007] Optionally, obtaining the tightness detection data between the tool and the tool holder includes one of the following: when the current moment reaches a predetermined moment, receiving the tightness detection data transmitted by the sensor based on a communication channel, wherein the predetermined moment is a pre-set collection moment, and the communication channel is used for data transmission with the sensor; when a tightness detection signal is received, sending a collection instruction to the sensor to control the sensor to collect the current tightness detection data, and receiving the tightness detection data through the communication channel.

[0008] Optionally, extracting features from the tightness detection data to obtain target feature data includes: extracting features from the tightness detection data using a first formula to obtain the target feature data, wherein the first formula is: i (t) = f i [W i *Z i (t)+b i ], i represents the number of the sensor, t represents the current time, f[] represents the activation function, W represents the convolution kernel, b represents the bias, Z i (t) represents the tightness detection data collected by the i-th sensor at the current time t, M i (t) represents the difference with Z i (t) The corresponding target feature data.

[0009] Optionally, before the target feature data is input into the tightness determination model so as to process the target feature data using the tightness determination model to obtain the tightness between the tool sleeve and the tool, the tool sleeve tightness detection method for the machine tool magazine further includes: iteratively training the convolutional neural network model using multiple groups of sample training data collected by each sensor in a historical time period until an iteration termination condition is reached to obtain an initial tightness determination sub-model corresponding to the sensor, wherein each of the multiple groups of sample training data includes: the sample target feature data, and a sample sub-tightness corresponding to the sample target feature data, and the iteration termination condition is at least one of the following: the number of iterations reaches a preset number of iterations, and the error between the predicted value and the true value of the convolutional neural network model is less than an error threshold; respectively, the multiple groups of sample training data corresponding to each sensor are used to verify the accuracy of the corresponding initial tightness determination sub-model in a cross-validation manner to obtain the accuracy of each of the tightness determination sub-models; when the accuracy is greater than the accuracy threshold, the initial tightness determination sub-model is determined to be the tightness determination sub-model.

[0010] Optionally, the tightness determination model includes a plurality of tightness determination sub-models, and the target feature data is input into the tightness determination model so as to process the target feature data using the tightness determination model to obtain the tightness between the tool sleeve and the tool, including: inputting the target feature data corresponding to each of the sensors into the tightness determination sub-model respectively so as to process the target feature data using the tightness determination sub-model to obtain the sub-tightness corresponding to the target feature data, wherein each of the tightness determination sub-models is used to process the data collected by a corresponding sensor; and determining the attention weight of each sub-tightness according to the accuracy of each of the tightness determination sub-models using a second formula, wherein the second formula is: h i represents the accuracy of the tension state determination sub-model corresponding to the i-th sensor, n represents the total number of sensors, s i The attention weight of the sub-tightness corresponding to the i-th sensor is represented, and the attention weight refers to the weight ratio of the sub-tightness in the tightness; the tightness between the tool sleeve and the tool is determined by a third formula according to the plurality of sub-tightnesses and the attention weight corresponding to each sub-tightness, wherein the third formula is: P(t) represents the tightness between the tool sleeve and the tool at the current moment.

[0011] Optionally, the tightness threshold range includes: a first tightness threshold range, a second tightness threshold range, a third tightness threshold range, a fourth tightness threshold range and a fifth tightness threshold range, and the tightness state between the tool sleeve and the tool is determined according to the tightness threshold range in which the tightness is located, including: when the tightness is within the first tightness threshold range, determining the tightness state as not loose, wherein the first tightness threshold range is a range in which the tightness is less than or equal to the first tightness threshold; when the tightness is within the second tightness threshold range, determining the tightness state as slightly loose, wherein the second tightness threshold range is a range in which the tightness is greater than the first tightness threshold and less than or equal to the second tightness threshold, and the second tightness threshold is greater than the first tightness threshold; When the tightness is within the third tightness threshold range, the tightness state is determined to be slightly loose, wherein the third tightness threshold range is a range in which the tightness is greater than the second tightness threshold and is less than or equal to the third tightness threshold, and the third tightness threshold is greater than the second tightness threshold; when the tightness is within the fourth tightness threshold range, the tightness state is determined to be slightly loose, wherein the fourth tightness threshold range is a range in which the tightness is greater than the third tightness threshold and is less than or equal to the fourth tightness threshold, and the fourth tightness threshold is greater than the third tightness threshold; when the tightness is within the fifth tightness threshold range, the tightness state is determined to be not loose, wherein the fifth tightness threshold range is a range in which the tightness is greater than the fourth tightness threshold.

[0012] According to another aspect of an embodiment of the present invention, a tool sleeve tightness detection device for a machine tool tool magazine is provided, comprising: a first acquisition unit, for acquiring a plurality of tightness detection data between a tool and a tool sleeve, wherein the tightness detection data is data collected by a sensor deployed in the tool sleeve during a tightness detection process after the tool enters the tool sleeve; a second acquisition unit, for performing feature extraction on the plurality of tightness detection data to obtain target feature data corresponding to each of the tightness detection data; a third acquisition unit, for inputting the target feature data into a tightness state determination model, so as to process the target feature data using the tightness state determination model to obtain the tightness degree between the tool sleeve and the tool, wherein the tightness state determination model is obtained by training with a plurality of sets of training data by a deep learning method, and each of the plurality of sets of training data comprises: sample target feature data and a sample tightness degree corresponding to the sample target feature data; a first determination unit, for determining the tightness state between the tool sleeve and the tool according to a tightness threshold range in which the tightness degree is located.

[0013] Optionally, the first acquisition unit includes at least one of the following: a first acquisition module, used to acquire the spacing distance between the tool and the bottom of the tool sleeve detected by a photoelectric sensor; a second acquisition module, used to acquire the clamping force between the tool sleeve and the tool detected by a force sensor, wherein the clamping force is used to reflect the degree of fixation between the tool sleeve and the tool; a third acquisition module, used to acquire the vibration frequency of the tool sleeve and the tool during machine tool processing detected by a vibration sensor.

[0014] Optionally, the first acquisition unit includes one of the following: a first receiving module, used to receive the tightness detection data transmitted by the sensor based on the communication channel when the current time reaches a predetermined time, wherein the predetermined time is a pre-set collection time, and the communication channel is used to transmit data with the sensor; a second receiving module, used to send a collection instruction to the sensor when receiving a tightness detection signal, so as to control the sensor to collect the current tightness detection data, and receive the tightness detection data through the communication channel.

[0015] Optionally, the second acquisition unit includes: a fourth acquisition module, configured to extract features from the tightness detection data using a first formula to obtain the target feature data, wherein the first formula is: i (t) = f i [W i *Z i (t)+b i ], i represents the number of the sensor, t represents the current time, f[] represents the activation function, W represents the convolution kernel, b represents the bias, Z i (t) represents the tightness detection data collected by the i-th sensor at the current time t, M i (t) represents the difference with Z i (t) The corresponding target feature data.

[0016] Optionally, the tool sleeve tightness detection device of the machine tool magazine further includes: a fourth acquisition unit, which is used to iteratively train the convolutional neural network model using multiple groups of sample training data collected by each sensor in a historical time period before inputting the target feature data into the tightness determination model to process the target feature data using the tightness determination model to obtain the tightness between the tool sleeve and the tool, until an iteration termination condition is reached, thereby obtaining an initial tightness determination sub-model corresponding to the sensor, wherein each of the multiple groups of sample training data includes: the sample target feature data, The sample sub-tightness corresponding to the sample target feature data, the iteration termination condition is at least one of the following: the number of iterations reaches a preset number of iterations, and the error between the predicted value and the true value of the convolutional neural network model is less than an error threshold; a fifth acquisition unit is used to use the multiple groups of sample training data corresponding to each of the sensors to respectively verify the accuracy of the corresponding initial tightness state determination sub-model by cross-validation to obtain the accuracy of each of the tightness state determination sub-model; a second determination unit is used to determine that the initial tightness state determination sub-model is a tightness state determination sub-model when the accuracy is greater than the accuracy threshold.

[0017] Optionally, the tension state determination model includes a plurality of tension state determination sub-models, and the third acquisition unit includes: a fifth acquisition module, used to input the target feature data corresponding to each of the sensors into the tension state determination sub-model respectively, so as to process the target feature data using the tension state determination sub-model to obtain a sub-tension degree corresponding to the target feature data, wherein each tension state determination sub-model is used to process data collected by a corresponding sensor; a first determination module, used to determine the attention weight of each sub-tension degree using a second formula according to the accuracy of each tension state determination sub-model, wherein the second formula is: h i represents the accuracy of the tension state determination sub-model corresponding to the i-th sensor, n represents the total number of sensors, s i The attention weight of the sub-tightness corresponding to the i-th sensor is represented, and the attention weight refers to the weight ratio of the sub-tightness in the tightness; a second determination module is used to determine the tightness between the tool sleeve and the tool according to a plurality of sub-tightnesses and the attention weight corresponding to each sub-tightness using a third formula, wherein the third formula is: P(t) represents the tightness between the tool sleeve and the tool at the current moment.

[0018] Optionally, the tightness threshold range includes: a first tightness threshold range, a second tightness threshold range, a third tightness threshold range, a fourth tightness threshold range and a fifth tightness threshold range, and the first determining unit includes: a third determining module, used to determine that the tightness state is not loose when the tightness is within the first tightness threshold range, wherein the first tightness threshold range is a range in which the tightness is less than or equal to the first tightness threshold; a fourth determining module, used to determine that the tightness state is slightly loose when the tightness is within the second tightness threshold range, wherein the second tightness threshold range is a range in which the tightness is greater than the first tightness threshold and less than or equal to the second tightness threshold, and the second tightness threshold is greater than the first tightness threshold; a fifth determining module, used to determine that the tightness state is slightly loose when the tightness is within the second tightness threshold range. A sixth determination module is used to determine that the tightness state is slightly loose when the tightness is within the third tightness threshold range, wherein the third tightness threshold range is a range in which the tightness is greater than the second tightness threshold and is less than or equal to the third tightness threshold, and the third tightness threshold is greater than the second tightness threshold; a sixth determination module is used to determine that the tightness state is slightly loose when the tightness is within the fourth tightness threshold range, wherein the fourth tightness threshold range is a range in which the tightness is greater than the third tightness threshold and is less than or equal to the fourth tightness threshold, and the fourth tightness threshold is greater than the third tightness threshold; a seventh determination module is used to determine that the tightness state is not loose when the tightness is within the fifth tightness threshold range, wherein the fifth tightness threshold range is a range in which the tightness is greater than the fourth tightness threshold.

[0019] According to another aspect of an embodiment of the present invention, a tool sleeve tightness detection system for a machine tool tool magazine is provided. The tool sleeve tightness detection system for a machine tool tool magazine uses any of the above-mentioned tool sleeve tightness detection methods for a machine tool tool magazine.

[0020] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is further provided, the computer-readable storage medium comprising a stored program, wherein the program executes any one of the above-mentioned methods for detecting the tightness of a tool sleeve of a machine tool magazine.

[0021] According to another aspect of an embodiment of the present invention, a processor is further provided, the processor being used to run a program, wherein when the program is run, any one of the above-mentioned methods for detecting tightness of a tool sleeve of a machine tool magazine is executed.

[0022] According to another aspect of an embodiment of the present invention, a computer program product is provided, comprising computer instructions, wherein when the computer instructions are executed by a processor, any one of the above-mentioned methods for detecting the tightness of a tool sleeve of a machine tool magazine is executed.

[0023] In an embodiment of the present invention, a variety of tightness detection data between the tool and the tool sleeve can be obtained, wherein the tightness detection data is data collected by a sensor deployed in the tool sleeve during a tightness detection process after the tool enters the tool sleeve; feature extraction is performed on the various tightness detection data to obtain target feature data corresponding to each tightness detection data; the target feature data is input into a tightness state determination model to process the target feature data using the tightness state determination model to obtain the tightness between the tool sleeve and the tool, wherein the tightness state determination model is trained by deep learning using multiple sets of training data, and each of the multiple sets of training data includes: sample target feature data, and a sample tightness corresponding to the sample target feature data; the tightness between the tool sleeve and the tool is determined according to the tightness threshold range in which the tightness is located. Through the above technical scheme, the goal of collecting the tightness detection data between the tool sleeve and the tool during the tightness detection process by using sensors, processing it with the trained deep learning model to obtain the tightness between the tool sleeve and the tool, and finally determining the tightness state of the tool sleeve according to the threshold range of the tightness is achieved. The technical effect of collecting and analyzing data through an automated system and determining the tightness state between the tool sleeve and the tool by intelligent analysis is achieved, which improves the detection efficiency and accuracy, and further solves the technical problem that the traditional tool tightness detection method in the related technology mainly relies on manual inspection or simple mechanical devices, resulting in low inspection efficiency and limited detection accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] The drawings described herein are used to provide a further understanding of the present invention and constitute a part of this application. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:

[0025] Figure 1 It is a hardware structure block diagram of a mobile terminal of a method for detecting the tightness of a tool sleeve of a machine tool magazine according to an embodiment of the present invention;

[0026] Figure 2 is a flow chart of a method for detecting tightness of a tool sleeve of a machine tool magazine according to an embodiment of the present invention;

[0027] Figure 3 is a schematic diagram of a tool sleeve tightness detection system for a machine tool tool magazine according to an embodiment of the present invention;

[0028] FIG4( a ) is a schematic diagram of a cannon-type knife sleeve before improvement according to an embodiment of the present invention;

[0029] FIG4( b ) is a schematic diagram of an improved barrel-type knife sleeve according to an embodiment of the present invention;

[0030] Figure 5is a flow chart at a terminal node according to an embodiment of the present invention;

[0031] Figure 6 is a flow chart at a coordinator node according to an embodiment of the present invention;

[0032] Figure 7 Schematic diagram of a tool sleeve tightness detection device for a machine tool tool magazine according to an embodiment of the present invention.

[0033] The above drawings include the following reference numerals:

[0034] 102. Processor; 104. Memory; 106. Transmission device; 108. Input / output device; A1. Force sensor; A2. Photoelectric sensor; A3. Vibration sensor; B1. Information sent regularly by the sensor; B2. Query request sent by the coordinator node; C1. Query instruction sent from the computer serial port; C2. Wireless data sent regularly by the terminal node. DETAILED DESCRIPTION

[0035] In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work should fall within the scope of protection of the present invention.

[0036] It should be noted that the terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish similar objects, and are not necessarily used to describe a specific order or sequence. It should be understood that the data used in this way can be interchanged where appropriate, so that the embodiments of the present invention described herein can be implemented in an order other than those illustrated or described herein. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions, for example, a process, method, system, product or device that includes a series of steps or units is not necessarily limited to those steps or units that are clearly listed, but may include other steps or units that are not clearly listed or inherent to these processes, methods, products or devices.

[0037] As introduced in the background technology, the traditional tool tightness detection method in the related art mainly relies on manual inspection or simple mechanical devices, resulting in low inspection efficiency and limited detection accuracy. In view of the above defects, in the embodiments of the present invention, a tool sleeve tightness detection method and device for a machine tool magazine and a computer program product are provided.

[0038] The technical solutions in the embodiments of the present invention will be described clearly and completely below in conjunction with the accompanying drawings in the embodiments of the present invention.

[0039] The method embodiments provided in the embodiments of the present invention can be executed in a mobile terminal, a computer terminal or a similar computing device. Taking running on a mobile terminal as an example, Figure 1 1 is a hardware structure block diagram of a mobile terminal for detecting the tightness of a tool sleeve of a machine tool magazine according to an embodiment of the present invention. Figure 1 As shown, the mobile terminal may include one or more ( Figure 1 Only one is shown in the figure) a processor 102 (the processor 102 may include but is not limited to a processing device such as a microprocessor MCU or a programmable logic device FPGA) and a memory 104 for storing data, wherein the mobile terminal may also include a transmission device 106 and an input / output device 108 for communication functions. It can be understood by those skilled in the art that Figure 1 The structure shown is only for illustration and does not limit the structure of the mobile terminal. Figure 1 More or fewer components as shown, or with Figure 1 Different configurations are shown.

[0040] The memory 104 can be used to store computer programs, for example, software programs and modules of application software, such as the computer program corresponding to the tool sleeve looseness detection method of the machine tool magazine in the embodiment of the present invention. The processor 102 executes various functional applications and data processing by running the computer program stored in the memory 104, that is, to implement the above method. The memory 104 may include a high-speed random access memory, and may also include a non-volatile memory, such as one or more magnetic storage devices, flash memory, or other non-volatile solid-state memory. In some examples, the memory 104 may further include a memory remotely arranged relative to the processor 102, and these remote memories can be connected to the mobile terminal via a network. Examples of the above network include but are not limited to the Internet, an intranet, a local area network, a mobile communication network, and a combination thereof. The transmission device 106 is used to receive or send data via a network. The above network specific examples may include a wireless network provided by a communication provider of the mobile terminal. In one example, the transmission device 106 includes a network adapter (Network Interface Controller, referred to as NIC), which can be connected to other network devices through a base station so as to communicate with the Internet. In one example, the transmission device 106 may be a radio frequency (RF) module, which is used to communicate with the Internet wirelessly.

[0041] According to an embodiment of the present invention, a method embodiment of a method for detecting the tightness of a tool sleeve of a machine tool magazine is provided. It should be noted that the steps shown in the flowchart of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and although a logical order is shown in the flowchart, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0042] Figure 2 : is a flow chart of a method for detecting the tightness of a tool sleeve of a machine tool magazine according to an embodiment of the present invention. Figure 2 As shown, the method comprises the following steps:

[0043] Step S202, obtaining a variety of tightness detection data between the tool and the tool sheath, wherein the tightness detection data is data collected by a sensor deployed in the tool sheath during a tightness detection process after the tool enters the tool sheath.

[0044] In this embodiment, multiple sensors can be deployed in the tool sheath to collect corresponding data during the tightness detection process after the tool enters the tool sheath, so as to provide a data basis for the subsequent analysis of the tightness status between the tool and the tool sheath. The tightness detection process here can be understood as the process of starting the detection after the tool enters the tool sheath.

[0045] Combine the following Figure 3 The above embodiments of the present invention are described in detail. Figure 3 Schematic diagram of a tool sleeve tightness detection system for a machine tool tool magazine according to an embodiment of the present invention. Figure 3 As shown, the tool sleeve tightness detection system can use the tool sleeve tightness detection method of the machine tool magazine based on deep learning provided by the embodiment of the present invention, which mainly includes the following modules: data collection module, data processing module, model training module, model verification module and real-time detection module; its overall implementation process is: the data collection module transmits all sensor information on the tool sleeve to the computer in real time through the serial port UART, and the remaining modules are completed on the computer.

[0046] According to the above embodiment of the present invention, in the above step S202, the tightness detection data between the tool and the tool sleeve is obtained, including at least one of the following: obtaining the spacing distance between the tool and the bottom of the tool sleeve detected by the photoelectric sensor; obtaining the clamping force between the tool sleeve and the tool detected by the force sensor, wherein the clamping force is used to reflect the degree of fixation between the tool sleeve and the tool; obtaining the vibration frequency of the tool sleeve and the tool during the machine tool processing process detected by the vibration sensor.

[0047] The above embodiment of the present invention is described in detail below in conjunction with Figures 4(a) and 4(b). Figure 4(a) is a schematic diagram of the barrel-type knife sleeve before improvement according to the embodiment of the present invention, and Figure 4(b) is a schematic diagram of the barrel-type knife sleeve after improvement according to the embodiment of the present invention; the improved knife sleeve structure shown in Figure 4(b) adds three sensors compared to the knife sleeve section before improvement shown in Figure 4(a), wherein A1 is a force sensor, A2 is a photoelectric sensor, and A3 is a vibration sensor. The photoelectric sensor A2 can detect the distance from the tool to the bottom of the knife sleeve, the force sensor A1 can detect the clamping force between the knife sleeve and the tool, and the vibration sensor A3 can detect the vibration frequency of the knife sleeve and the tool in the working state. The corresponding data are collected during the tightness detection process by these sensors, aiming to effectively improve the accuracy of the knife sleeve tightness detection, real-time monitoring capability, data analysis capability and overall efficiency.

[0048] According to the above embodiment of the present invention, in the above step S202, the tightness detection data between the tool and the tool sleeve is obtained, including one of the following: when the current moment reaches the predetermined moment, the tightness detection data transmitted by the sensor based on the communication channel is received, wherein the predetermined moment is a pre-set collection moment, and the communication channel is used for data transmission with the sensor; when a tightness detection signal is received, a collection instruction is sent to the sensor to control the sensor to collect the current tightness detection data, and the tightness detection data is received through the communication channel.

[0049] Specifically, in Figure 3 The data collection module shown mainly involves the deployment of sensors, the hardware design of terminal nodes and coordinator nodes, and the software design of terminal nodes and coordinator nodes; the information collected by each sensor deployed in the knife sleeve can be transmitted to an independent terminal node through a wire. The hardware design of this node is equivalent to the integration of a main control chip, a wireless communication module, a sensor module, and a voltage regulator chip. Moreover, the main control chip generally needs to have certain computing power, A / D conversion functions, etc., in order to control the wireless communication module, the sensor module, and the voltage regulator chip.

[0050] Combine the following Figure 5 and Figure 6 The above embodiments of the present invention are described in detail. Figure 5 is a flow chart at a terminal node according to an embodiment of the present invention, Figure 6 is a flow chart at a coordinator node according to an embodiment of the present invention.

[0051] In the embodiment provided by the present invention, ZigBee technology can be used to realize wireless communication between terminal nodes and coordinator nodes, that is, the signals sent by the terminal nodes are received and sorted by the coordinator nodes, and then transmitted to the computer using the serial port UART. Therefore, the hardware design of the coordinator node is similar to the hardware design of the terminal node, but does not include the connection of the sensor module. Moreover, if the interface between the computer and the coordinator node is inconsistent, it is necessary to add an interface conversion module to the hardware design of the coordinator node. For the software design of the terminal node and the coordinator node, since it involves wireless communication of multiple sensors, it is necessary to design a communication protocol, and then perform sensor data acquisition and terminal node information sending program design according to the timing or receipt of a query request sent from the coordinator node, and perform the coordinator node information sending program design to the computer according to the situation where the computer serial port sends a query instruction or the terminal node sends information at regular intervals. Specifically, Figure 5 As shown, the terminal node can not only receive the information B1 transmitted regularly by each sensor, but also control each sensor to collect data when receiving the query request B2 sent by the coordinator node, so as to obtain the corresponding tightness detection data; Figure 6 As shown, the coordination node can send a query request to the terminal node according to the query instruction C2 sent from the computer serial port to control the sensor to collect data, and then receive the tightness detection data sent by the terminal node. It can also receive the wireless data C2 sent by the terminal node at a fixed time to obtain the corresponding tightness detection data.

[0052] Step S204, extracting features from the various types of tightness detection data to obtain target feature data corresponding to each type of tightness detection data.

[0053] In this embodiment, you can use Figure 3 The data processing module shown performs feature extraction on the tightness detection data collected by multiple sensors to obtain target feature data corresponding to each tightness detection data.

[0054] According to the above embodiment of the present invention, in the above step S204, feature extraction is performed on the tightness detection data to obtain target feature data, including: using a first formula to extract features from the tightness detection data to obtain target feature data, wherein the first formula is: i (t) = f i [W i *Z i (t)+b i ], i represents the sensor number, t represents the current time, f[] represents the activation function, W represents the convolution kernel, b represents the bias, and Z i (t) represents the tightness detection data collected by the i-th sensor at the current time t, M i(t) represents the difference with Z i (t) Corresponding target feature data.

[0055] Specifically, the data processing module mainly includes data preprocessing and feature extraction operations. Data preprocessing mainly involves denoising and normalizing the signals collected by force, photoelectric, and vibration sensors; and feature extraction operations can use convolutional neural networks to extract statistical features such as peak value, mean, and variance from force sensor data, extract features such as light intensity change rate and displacement from photoelectric sensors, and extract features such as vibration frequency and amplitude from vibration sensors. In addition, by collecting and processing sensor data multiple times, a corresponding data set is formed to provide a basis for subsequent model training and verification; assuming that the input sensor signal is (i=1, 2, 3, 1 represents the force sensor signal, 2 represents the optical sensor electrical signal, and 3 represents the vibration sensor signal, and the i mentioned later all express the above meaning), a simple low-pass filter can be used for denoising, and the filtered signal can be expressed as Y i (t) = X i (t)*H i (t), where H i (t) is the impulse response function of the low-pass filter; the normalization process is as follows: in, and are the maximum and minimum values ​​of the filtered signal, respectively. i (t) is the normalized signal, that is, the tightness detection data collected by the i-th sensor at the current time t after processing; the convolutional neural network is used to extract features, which can be expressed as M i (t) = f i [W i *Z i (t)+b i ], where M i (t) represents the eigenvector, which is related to Z i (t) The corresponding target feature data, W i is the convolution kernel, b i is the bias, * indicates the convolution operation, f i The function represents an activation function, usually a ReLU (rectified linear unit) function, through which the features of the corresponding sensor data can be extracted.

[0056] Step S206, input the target feature data into the tightness determination model, so as to process the target feature data using the tightness determination model to obtain the tightness between the tool sleeve and the tool, wherein the tightness determination model is trained by deep learning using multiple sets of training data, and each of the multiple sets of training data includes: sample target feature data, and sample tightness corresponding to the sample target feature data.

[0057] In this embodiment, you can use Figure 3 The tightness state determination module of the real-time detection module shown processes the target feature data obtained by the data processing module to analyze the degree of analysis between the tool sleeve and the tool.

[0058] According to the above embodiment of the invention, before the above step S206, that is, before inputting the target feature data into the tightness determination model to process the target feature data using the tightness determination model to obtain the tightness between the tool sleeve and the tool, the tool sleeve tightness detection method of the machine tool magazine also includes: iteratively training the convolutional neural network model using multiple groups of sample training data collected by each sensor in a historical time period, until the iteration termination condition is reached, and an initial tightness determination sub-model corresponding to the sensor is obtained, wherein each of the multiple groups of sample training data includes: sample target feature data, sample sub-tightness corresponding to the sample target feature data, and the iteration termination condition is at least one of the following: the number of iterations reaches a preset number of iterations, and the error between the predicted value and the true value of the convolutional neural network model is less than the error threshold; respectively, using the multiple groups of sample training data corresponding to each sensor, the corresponding initial tightness determination sub-model is cross-validated to verify the accuracy, and obtain the accuracy of each tightness determination sub-model; when the accuracy is greater than the accuracy threshold, the initial tightness determination sub-model is determined to be the tightness determination sub-model.

[0059] Specifically, Figure 3The module training module shown mainly involves the construction, training and parameter optimization of the deep learning model. First, in order to improve the accuracy of the detection results, the present invention selects to train the corresponding convolutional neural network (CNN) model for the data of each sensor respectively. This method can make full use of the unique signal characteristics of each sensor, thereby improving the overall detection performance. At this stage, an independent model is established for each sensor data, so that the model can more accurately capture the key features in different types of data; then, the cross-validation method is used to evaluate the model performance to prevent overfitting and ensure the generalization ability of the model. The data set can also be divided into multiple subsets, and some of them are used alternately for training, and the rest are used for verification, which can comprehensively test the performance of the model on unseen data, thereby improving its reliability and practicality; then, the features extracted by each model are fused using the attention mechanism, and the attention mechanism can help the model focus on more important features, thereby improving the effective utilization of information; the fused features will be input into the fully connected neural network for the final task training, and this step ensures that the information obtained from multiple sensors can be comprehensively utilized to achieve more accurate prediction results; finally, the entire model is optimized and fine-tuned by the back propagation algorithm. Back propagation can effectively adjust model parameters to reduce the loss function, thereby gradually improving the accuracy of predictions. After this series of training and optimization processes, the model can not only effectively identify and predict various states, but also maintain stable performance in a changing environment, which provides strong support for practical applications.

[0060] When training the model, for each sensor’s preprocessed data M i (t), establish an independent convolutional neural network model CNN i , and train each model: N i (t) = CNN i [M i (t)], that is, training the data collected by each sensor can obtain a corresponding loose and tight state determination sub-model. The training goal of each sensor is to minimize a loss function that measures the difference between the predicted result and the actual label: L i =￡[N i (t), M i (t)], where N i (t) is the output predicted by the convolutional neural network, M i (t) is the true label. If the loss function is the mean square error loss, then Here, n represents the number of samples.

[0061] When conducting preliminary verification of the model, the cross-validation method can be used. Assume that the data set D i Divided into K subsets (Di1 ,D i2 ,……,D iK ), each time a subset is selected as the validation set, and the remaining (K-1) subsets are used as training sets. The goal of cross-validation is to average the loss of each validation, that is, to obtain the average loss It can be expressed by the following formula:

[0062] In addition, you can also use Figure 3 The model validation module shown evaluates the performance of the fusion model on the validation set, compares the effects of different fusion strategies, and selects the best solution. In addition, it is necessary to regularly monitor the performance of the model on new data and retrain or adjust the fusion strategy when necessary to adapt to data changes. During the monitoring process, if it is found that the performance of the model on the new data has dropped significantly, retraining needs to be considered, which will involve recollecting and labeling data to ensure that the training set is representative. In addition, hyperparameter optimization techniques (such as grid search or Bayesian optimization) can be used to fine-tune the existing model to adapt to new data characteristics.

[0063] In a specific embodiment of the present invention, the tightness determination model includes a plurality of tightness determination sub-models, and the target feature data is input into the tightness determination model so as to process the target feature data using the tightness determination model to obtain the tightness between the tool sleeve and the tool, including: inputting the target feature data corresponding to each sensor into the tightness determination sub-model respectively so as to process the target feature data using the tightness determination sub-model to obtain the sub-tightness corresponding to the target feature data, wherein each tightness determination sub-model is used to process the data collected by a corresponding sensor; and determining the attention weight of each sub-tightness according to the accuracy of each tightness determination sub-model using the second formula, wherein the second formula is: h i represents the accuracy of the tightness state determination sub-model corresponding to the i-th sensor, n represents the total number of sensors, s i The attention weight of the sub-tightness corresponding to the i-th sensor is represented, and the attention weight refers to the weight proportion of the sub-tightness in the tightness; the tightness between the tool sleeve and the tool is determined by the third formula according to the multiple sub-tightnesses and the attention weight corresponding to each sub-tightness, wherein the third formula is: P(t) represents the tightness between the tool sleeve and the tool at the current moment.

[0064] Specifically, the core of the attention mechanism is to calculate the importance weight of each sensor feature and then perform weighted fusion. After each sensor model (i.e., the sub-model for determining the tightness and looseness corresponding to each sensor) is trained, the model is evaluated using the validation set, and the importance of each sensor is measured based on the accuracy of the model. The conversion formula is as follows: Among them, h i represents the accuracy of the tightness state determination sub-model corresponding to the i-th sensor, n represents the total number of sensors, s i represents the attention weight of the sub-tightness corresponding to the i-th sensor. The attention weight refers to the weight ratio of the sub-tightness in the tightness. Before inputting into the fully connected neural network, the attention weight is used to weight the features of each sensor into a new feature vector. P(t) represents the tightness between the tool sleeve and the tool at the current moment.

[0065] Step S208, determining the tightness state between the tool sleeve and the tool according to the tightness threshold range.

[0066] In this embodiment, the specific tightness state between the tool sleeve and the tool can be determined based on the tightness threshold range of the tightness obtained by analyzing the above steps.

[0067] According to the above embodiment of the present invention, in the above step S208, the tightness threshold range includes: a first tightness threshold range, a second tightness threshold range, a third tightness threshold range, a fourth tightness threshold range and a fifth tightness threshold range, and the tightness state between the tool sleeve and the tool is determined according to the tightness threshold range in which the tightness is located, including: when the tightness is in the first tightness threshold range, determining the tightness state as not loose, wherein the first tightness threshold range is a range in which the tightness is less than or equal to the first tightness threshold; when the tightness is in the second tightness threshold range, determining the tightness state as slightly loose, wherein the second tightness threshold range is a range in which the tightness is greater than the first tightness threshold and less than or equal to the second tightness threshold, and the second The tightness threshold is greater than the first tightness threshold; when the tightness is within the range of the third tightness threshold, the tightness state is determined to be slightly loose, wherein the range of the third tightness threshold is a range in which the tightness is greater than the second tightness threshold and is less than or equal to the third tightness threshold, and the third tightness threshold is greater than the second tightness threshold; when the tightness is within the range of the fourth tightness threshold, the tightness state is determined to be slightly loose, wherein the range of the fourth tightness threshold is a range in which the tightness is greater than the third tightness threshold and is less than or equal to the fourth tightness threshold, and the fourth tightness threshold is greater than the third tightness threshold; when the tightness is within the range of the fifth tightness threshold, the tightness state is determined to be not loose, wherein the range of the fifth tightness threshold is a range in which the tightness is greater than the fourth tightness threshold.

[0068] Specifically, five tightness threshold ranges can be roughly divided (of course, other numbers of threshold ranges can also be divided according to actual conditions, and there is no specific restriction here), which correspond one by one to several specific states usually included in the tool management of the machine tool magazine: 1) The first tightness threshold range corresponds to the normal state: the tool is firmly clamped by the tool sleeve, and there is no sign of looseness. At this time, the contact force, vibration and position between the tool and the tool sleeve are all within the preset normal range; 2) The second tightness threshold range corresponds to a slightly loose state: the contact force or vibration between the tool and the tool sleeve slightly exceeds the normal range, but has not yet caused a significant impact on the processing accuracy or equipment safety. This state may indicate that the tool is about to loosen and requires close attention from the operator; 3) The third tightness threshold range corresponds to a moderately loose state: the tool clamping force decreases significantly or the vibration increases significantly, and the processing accuracy begins to be affected. , there are certain safety hazards. At this time, the tool should be checked and adjusted immediately to prevent further loosening; 4) The fourth tightness threshold range corresponds to a severe loose state: the tool is almost completely loose, and the processing cannot be guaranteed. It may pose a serious threat to the equipment and operators. The processing must be stopped immediately for emergency treatment and maintenance; 5) The fifth tightness threshold range corresponds to a fault state: the tool and the tool sleeve cannot work normally due to excessive wear, damage or other reasons. The tool or tool sleeve needs to be replaced for maintenance; through real-time monitoring and accurate identification of these states, accidents during processing can be effectively prevented, processing efficiency can be improved, and the safety of equipment and operators can be guaranteed. In actual applications, through real-time data acquisition by sensors and analysis of deep learning models, it is possible to quickly determine which of the above states the tool is in, and take corresponding measures.

[0069] In addition, the real-time detection module also includes the design of a visual interface and the output of a detection report. The visual interface should display the data transmitted from each sensor in real time, including information such as temperature, pressure, and vibration, and present it in the form of a chart or dashboard so that users can quickly understand the current equipment status; at the same time, a reasonable data refresh frequency will ensure that information is updated in a timely manner to help users make quick decisions. In addition, the interface should clearly display the tightness between the tool sleeve and the tool, using color coding or graphic identification, such as green for normal status, yellow for warning, and red for fault. The relative position of the tool and tool sleeve can even be intuitively displayed through 3D models or animation effects. In terms of the output of inspection reports, the system should be able to automatically generate real-time reports containing inspection time, tool sleeve status and sensor data, and support multiple formats (such as PDF, Excel) to meet the needs of different users; the report should include data trend analysis, using a line graph to show the changing trend of the tool tightness status, and using a variety of visual chart forms such as pie charts and bar charts to enhance the readability of the report; in addition, users should be able to customize the content and format of the report as needed, select the required data items, and provide the function of automatically generating reports on a regular basis, such as daily, weekly or monthly reports, to help users systematically summarize and analyze equipment status.

[0070] As can be seen from the above, through the technical solution provided by the above embodiments of the present invention, it is possible to obtain a variety of tightness detection data between the tool and the tool sheath, wherein the tightness detection data is data collected by the sensor deployed in the tool sheath during the tightness detection process after the tool enters the tool sheath; perform feature extraction on the various tightness detection data to obtain target feature data corresponding to each tightness detection data; input the target feature data into a tightness state determination model, so as to process the target feature data using the tightness state determination model to obtain the tightness between the tool sheath and the tool, wherein the tightness state determination model is trained by deep learning using multiple sets of training data, and the multiple sets of training data are used to perform feature extraction on the target feature data. Each group of data includes: sample target feature data, sample tightness corresponding to the sample target feature data; determining the tightness state between the tool sleeve and the tool according to the tightness threshold range of the tightness, thereby achieving the goal of collecting the tightness detection data between the tool sleeve and the tool during the tightness detection process by using the sensor, and processing it with the trained deep learning model to obtain the tightness between the tool sleeve and the tool, and finally determining the tightness state of the tool sleeve according to the threshold range of the tightness, thereby achieving the technical effect of using an automated system to collect and analyze data and determine the tightness state between the tool sleeve and the tool through intelligent analysis, thereby improving detection efficiency and accuracy.

[0071] Therefore, the technical solution provided by the above-mentioned embodiment of the present invention solves the technical problem that the traditional tool tightness detection method in the related art mainly relies on manual inspection or simple mechanical devices, resulting in low inspection efficiency and limited detection accuracy.

[0072] It should be noted that, for the aforementioned method embodiments, for the sake of simplicity, they are all expressed as a series of action combinations, but those skilled in the art should be aware that the present application is not limited by the described order of actions, because according to the present application, certain steps can be performed in other orders or simultaneously. Secondly, those skilled in the art should also be aware that the embodiments described in the specification are all preferred embodiments, and the actions and modules involved are not necessarily required by the present application.

[0073] Through the description of the above implementation methods, those skilled in the art can clearly understand that the method according to the above embodiment can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk), and includes a number of instructions for a terminal device (which can be a mobile phone, computer, server, or network device, etc.) to execute the methods described in each embodiment of the present application.

[0074] According to an embodiment of the present invention, a tool sleeve looseness detection device for a machine tool tool magazine is provided for implementing the tool sleeve looseness detection method for a machine tool tool magazine. Figure 7 Schematic diagram of a tool sleeve tightness detection device for a machine tool tool magazine according to an embodiment of the present invention. Figure 7 As shown, the device comprises: a first acquisition unit 71, a second acquisition unit 73, a third acquisition unit 75 and a first determination unit 77. The tool sleeve tightness detection device for a machine tool magazine is described in detail below.

[0075] The first acquisition unit 71 is used to acquire a variety of tightness detection data between the tool and the tool sleeve, wherein the tightness detection data is data collected by a sensor deployed in the tool sleeve during a tightness detection process after the tool enters the tool sleeve.

[0076] The second acquisition unit 73 is used to extract features from the various types of tightness detection data to obtain target feature data corresponding to each type of tightness detection data.

[0077] The third acquisition unit 75 is used to input the target feature data into the tightness determination model, so as to process the target feature data using the tightness determination model to obtain the tightness between the tool sleeve and the tool, wherein the tightness determination model is trained by deep learning using multiple sets of training data, and each of the multiple sets of training data includes: sample target feature data and sample tightness corresponding to the sample target feature data.

[0078] The first determining unit 77 is used to determine the tightness state between the tool sleeve and the tool according to the tightness threshold range of the tightness.

[0079] It should be noted here that the above-mentioned first acquisition unit 71, second acquisition unit 73, third acquisition unit 75 and first determination unit 77 correspond to steps S202 to S208 in the above-mentioned embodiments, and the four units are the same as the instances and application scenarios implemented by the corresponding steps, but are not limited to the contents disclosed in the above-mentioned embodiments.

[0080] As can be seen from the above, in the scheme recorded in the above-mentioned embodiment of the present invention, a first acquisition unit can be used to acquire a variety of tightness detection data between the tool and the tool sheath, wherein the tightness detection data is data collected by the sensor deployed in the tool sheath during the tightness detection process after the tool enters the tool sheath; then a second acquisition unit is used to perform feature extraction on the various tightness detection data to obtain target feature data corresponding to each tightness detection data; then a third acquisition unit is used to input the target feature data into a tightness state determination model, so as to process the target feature data using the tightness state determination model to obtain the tightness between the tool sheath and the tool, wherein the tightness state determination model is trained by deep learning using multiple sets of training data. The training data is obtained by training, and each of the multiple groups of training data includes: sample target feature data, and sample tightness corresponding to the sample target feature data; finally, the first determination unit is used to determine the tightness state between the tool sleeve and the tool according to the tightness threshold range of the tightness, thereby achieving the goal of collecting the tightness detection data between the tool sleeve and the tool during the tightness detection process by using the sensor, and processing it by using the trained deep learning model to obtain the tightness between the tool sleeve and the tool, and finally determining the tightness state of the tool sleeve according to the threshold range of the tightness, thereby achieving the technical effect of collecting and analyzing data through an automated system to determine the tightness state between the tool sleeve and the tool through intelligent analysis, thereby improving the detection efficiency and accuracy.

[0081] Therefore, the technical solution provided by the above-mentioned embodiment of the present invention solves the technical problem that the traditional tool tightness detection method in the related art mainly relies on manual inspection or simple mechanical devices, resulting in low inspection efficiency and limited detection accuracy.

[0082] In an optional embodiment of the present invention, the first acquisition unit includes at least one of the following: a first acquisition module for acquiring the spacing distance between the tool and the bottom of the tool sleeve detected by the photoelectric sensor; a second acquisition module for acquiring the clamping force between the tool sleeve and the tool detected by the force sensor, wherein the clamping force is used to reflect the degree of fixation between the tool sleeve and the tool; a third acquisition module for acquiring the vibration frequency of the tool sleeve and the tool during machine tool processing detected by the vibration sensor.

[0083] In an optional embodiment of the present invention, the first acquisition unit includes one of the following: a first receiving module, used to receive the tightness detection data transmitted by the sensor based on the communication channel when the current time reaches the predetermined time, wherein the predetermined time is a pre-set collection time, and the communication channel is used to transmit data with the sensor; a second receiving module, used to send a collection instruction to the sensor when receiving the tightness detection signal, so as to control the sensor to collect the current tightness detection data, and receive the tightness detection data through the communication channel.

[0084] In an optional embodiment of the present invention, the second acquisition unit includes: a fourth acquisition module, which is used to extract features from the tightness detection data using the first formula to obtain target feature data, wherein the first formula is: i (t) = f i [W i *Z i (t)+b i ], i represents the sensor number, t represents the current time, f[] represents the activation function, W represents the convolution kernel, b represents the bias, and Z i (t) represents the tightness detection data collected by the i-th sensor at the current time t, M i (t) represents the difference with Z i (t) Corresponding target feature data.

[0085] In an optional embodiment of the present invention, the tool sleeve tightness detection device of the machine tool magazine further includes: a fourth acquisition unit, which is used to iteratively train the convolutional neural network model using multiple groups of sample training data collected by each sensor in a historical time period before inputting the target feature data into the tightness determination model to process the target feature data using the tightness determination model to obtain the tightness between the tool sleeve and the tool, until the iteration termination condition is reached, and the initial tightness determination sub-model corresponding to the sensor is obtained, wherein each of the multiple groups of sample training data includes: sample target feature data, sample sub-tightness corresponding to the sample target feature data, and the iteration termination condition is at least one of the following: the number of iterations reaches a preset number of iterations, and the error between the predicted value and the true value of the convolutional neural network model is less than the error threshold; a fifth acquisition unit, which is used to respectively use the multiple groups of sample training data corresponding to each sensor to verify the accuracy of the corresponding initial tightness determination sub-model in a cross-validation manner to obtain the accuracy of each tightness determination sub-model; a second determination unit, which is used to determine that the initial tightness determination sub-model is the tightness determination sub-model when the accuracy is greater than the accuracy threshold.

[0086] In an optional embodiment of the present invention, the tension state determination model includes multiple tension state determination sub-models, and the third acquisition unit includes: a fifth acquisition module, which is used to input the target feature data corresponding to each sensor into the tension state determination sub-model respectively, so as to process the target feature data using the tension state determination sub-model to obtain the sub-tension degree corresponding to the target feature data, wherein each tension state determination sub-model is used to process the data collected by a corresponding sensor; a first determination module, which is used to determine the attention weight of each sub-tension degree using the second formula according to the accuracy of each tension state determination sub-model, wherein the second formula is: h i represents the accuracy of the tightness state determination sub-model corresponding to the i-th sensor, n represents the total number of sensors, s i represents the attention weight of the sub-tightness corresponding to the i-th sensor, where the attention weight refers to the weight proportion of the sub-tightness in the tightness; a second determination module is used to determine the tightness between the tool sleeve and the tool using a third formula according to the multiple sub-tightnesses and the attention weight corresponding to each sub-tightness, wherein the third formula is: P(t) represents the tightness between the tool sleeve and the tool at the current moment.

[0087] In an optional embodiment of the present invention, the tightness threshold range includes: a first tightness threshold range, a second tightness threshold range, a third tightness threshold range, a fourth tightness threshold range and a fifth tightness threshold range, and the first determination unit includes: a third determination module, used to determine that the tightness state is not loose when the tightness is within the first tightness threshold range, wherein the first tightness threshold range is a range in which the tightness is less than or equal to the first tightness threshold; a fourth determination module, used to determine that the tightness state is slightly loose when the tightness is within the second tightness threshold range, wherein the second tightness threshold range is a range in which the tightness is greater than the first tightness threshold and less than or equal to the second tightness threshold, and the second tightness threshold is greater than the first tightness threshold; a fifth determination module, Used to determine that the tightness state is slightly loose when the tightness is within a third tightness threshold range, wherein the third tightness threshold range is a range in which the tightness is greater than the second tightness threshold and is less than or equal to the third tightness threshold, and the third tightness threshold is greater than the second tightness threshold; a sixth determination module, used to determine that the tightness state is slightly loose when the tightness is within a fourth tightness threshold range, wherein the fourth tightness threshold range is a range in which the tightness is greater than the third tightness threshold and is less than or equal to the fourth tightness threshold, and the fourth tightness threshold is greater than the third tightness threshold; a seventh determination module, used to determine that the tightness state is not loose when the tightness is within a fifth tightness threshold range, wherein the fifth tightness threshold range is a range in which the tightness is greater than the fourth tightness threshold.

[0088] According to another aspect of an embodiment of the present invention, a tool sleeve looseness detection system for a machine tool tool magazine is provided. The tool sleeve looseness detection system for a machine tool tool magazine uses any of the above-mentioned tool sleeve looseness detection methods for a machine tool tool magazine.

[0089] According to another aspect of an embodiment of the present invention, a computer-readable storage medium is further provided, the computer-readable storage medium comprising a stored program, wherein the program executes any one of the above-mentioned methods for detecting the tightness of a tool sleeve of a machine tool magazine.

[0090] Optionally, in this embodiment, the computer-readable storage medium may be located in any one of the computer terminals in a computer terminal group in a computer network, or in any one of the communication devices in a communication device group.

[0091] Optionally, in this embodiment, the computer-readable storage medium is configured to store program codes for executing the following steps: obtaining a variety of tightness detection data between the tool and the tool sleeve, wherein the tightness detection data is data collected by a sensor deployed in the tool sleeve during a tightness detection process after the tool enters the tool sleeve; performing feature extraction on the multiple tightness detection data to obtain target feature data corresponding to each tightness detection data; inputting the target feature data into a tightness state determination model to process the target feature data using the tightness state determination model to obtain the tightness between the tool sleeve and the tool, wherein the tightness state determination model is trained by deep learning using multiple sets of training data, and each set of the multiple sets of training data includes: sample target feature data, and a sample tightness corresponding to the sample target feature data; determining the tightness between the tool sleeve and the tool according to a tightness threshold range in which the tightness is located.

[0092] Optionally, in this embodiment, the computer-readable storage medium is configured to store program codes for executing the following steps: obtaining the spacing distance between the tool and the bottom of the tool sleeve detected by the photoelectric sensor; obtaining the clamping force between the tool sleeve and the tool detected by the force sensor, wherein the clamping force is used to reflect the degree of fixation between the tool sleeve and the tool; obtaining the vibration frequency of the tool sleeve and the tool during machine tool processing detected by the vibration sensor.

[0093] Optionally, in this embodiment, the computer-readable storage medium is configured to store program codes for executing the following steps: when the current moment reaches a predetermined moment, receiving tightness detection data transmitted by the sensor based on a communication channel, wherein the predetermined moment is a pre-set collection moment, and the communication channel is used for data transmission with the sensor; when a tightness detection signal is received, sending a collection instruction to the sensor to control the sensor to collect the current tightness detection data, and receiving the tightness detection data through the communication channel.

[0094] Optionally, in this embodiment, the computer-readable storage medium is configured to store a program code for executing the following steps: extracting features from the tightness detection data using a first formula to obtain target feature data, wherein the first formula is: i (t) = f i [W i *Z i (t)+b i ], i represents the sensor number, t represents the current time, f[] represents the activation function, W represents the convolution kernel, b represents the bias, and Z i (t) represents the tightness detection data collected by the i-th sensor at the current time t, M i (t) represents the difference with Z i (t) Corresponding target feature data.

[0095] Optionally, in this embodiment, the computer-readable storage medium is configured to store program code for executing the following steps: iteratively training the convolutional neural network model using multiple groups of sample training data collected by each sensor in a historical time period until an iteration termination condition is reached to obtain an initial tightness state determination submodel corresponding to the sensor, wherein each of the multiple groups of sample training data includes: sample target feature data, sample sub-tightness corresponding to the sample target feature data, and the iteration termination condition is at least one of the following: the number of iterations reaches a preset number of iterations, and the error between the predicted value and the true value of the convolutional neural network model is less than an error threshold; respectively, using multiple groups of sample training data corresponding to each sensor, a cross-validation method is used to verify the accuracy of the corresponding initial tightness state determination submodel to obtain the accuracy of each tightness state determination submodel; when the accuracy is greater than the accuracy threshold, the initial tightness state determination submodel is determined to be the tightness state determination submodel.

[0096] Optionally, in this embodiment, the computer-readable storage medium is configured to store a program code for executing the following steps: respectively inputting the target feature data corresponding to each sensor into the tension state determination sub-model, so as to process the target feature data using the tension state determination sub-model to obtain a sub-tightness corresponding to the target feature data, wherein each tension state determination sub-model is used to process data collected by a corresponding sensor; and determining the attention weight of each sub-tightness according to the accuracy of each tension state determination sub-model using a second formula, wherein the second formula is: h i represents the accuracy of the tightness state determination sub-model corresponding to the i-th sensor, n represents the total number of sensors, s i The attention weight of the sub-tightness corresponding to the i-th sensor is represented, and the attention weight refers to the weight proportion of the sub-tightness in the tightness; the tightness between the tool sleeve and the tool is determined by the third formula according to the multiple sub-tightnesses and the attention weight corresponding to each sub-tightness, wherein the third formula is: P(t) represents the tightness between the tool sleeve and the tool at the current moment.

[0097] Optionally, in this embodiment, the computer-readable storage medium is configured to store program codes for executing the following steps: when the tightness is within a first tightness threshold range, determining the tightness state as not loose, wherein the first tightness threshold range is a range in which the tightness is less than or equal to the first tightness threshold; when the tightness is within a second tightness threshold range, determining the tightness state as slightly loose, wherein the second tightness threshold range is a range in which the tightness is greater than the first tightness threshold and less than or equal to the second tightness threshold, and the second tightness threshold is greater than the first tightness threshold; when the tightness is within a third tightness threshold range, determining The tightness state is slightly loose, wherein the third tightness threshold range is a range in which the tightness is greater than the second tightness threshold and is less than or equal to the third tightness threshold, and the third tightness threshold is greater than the second tightness threshold; when the tightness is in the fourth tightness threshold range, the tightness state is determined to be slightly loose, wherein the fourth tightness threshold range is a range in which the tightness is greater than the third tightness threshold and is less than or equal to the fourth tightness threshold, and the fourth tightness threshold is greater than the third tightness threshold; when the tightness is in the fifth tightness threshold range, the tightness state is determined to be not loose, wherein the fifth tightness threshold range is a range in which the tightness is greater than the fourth tightness threshold.

[0098] According to another aspect of an embodiment of the present invention, a processor is further provided, and the processor is used to run a program, wherein when the program is run, any one of the above-mentioned methods for detecting the tightness of a tool sleeve of a machine tool magazine is executed.

[0099] According to another aspect of an embodiment of the present invention, a computer program product is provided, comprising computer instructions, which, when executed by a processor, execute any one of the above-mentioned methods for detecting the tightness of a tool sleeve of a machine tool magazine.

[0100] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0101] In the above embodiments of the present invention, the description of each embodiment has its own emphasis. For parts that are not described in detail in a certain embodiment, reference can be made to the relevant descriptions of other embodiments.

[0102] In the several embodiments provided in this application, it should be understood that the disclosed technical content can be implemented in other ways. Among them, the device embodiments described above are only schematic. For example, the division of the units can be a logical function division. There may be other division methods in actual implementation. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of units or modules, which can be electrical or other forms.

[0103] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple units. Some or all of the units may be selected according to actual needs to achieve the purpose of the present embodiment.

[0104] In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit. The above-mentioned integrated unit may be implemented in the form of hardware or in the form of software functional units.

[0105] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium, including a number of instructions for a computer device (which can be a personal computer, a server or a network device, etc.) to perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), mobile hard disk, magnetic disk or optical disk and other media that can store program codes.

[0106] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the principle of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for detecting the tightness of a tool sleeve of a machine tool magazine, characterized in that: include: Acquire a variety of tightness detection data between the tool and the tool sheath, wherein the tightness detection data is data collected by a sensor deployed in the tool sheath during a tightness detection process after the tool enters the tool sheath; Performing feature extraction on the multiple types of tightness detection data to obtain target feature data corresponding to each type of tightness detection data; The target feature data is input into a tightness determination model, so as to process the target feature data using the tightness determination model to obtain the tightness between the tool sleeve and the tool, wherein the tightness determination model is trained by deep learning using multiple sets of training data, and each of the multiple sets of training data includes: sample target feature data and sample tightness corresponding to the sample target feature data; The tightness state between the tool sleeve and the tool is determined according to a tightness threshold range of the tightness.

2. The method for detecting the tightness of a tool sleeve of a machine tool magazine according to claim 1, characterized in that: Obtain the tightness detection data between the tool and the tool holder, including one of the following: When the current time reaches the predetermined time, receiving the tightness detection data transmitted by the sensor based on the communication channel, wherein the predetermined time is a pre-set collection time, and the communication channel is used for data transmission with the sensor; When receiving the tightness detection signal, a collection instruction is sent to the sensor to control the sensor to collect the current tightness detection data, and the tightness detection data is received through the communication channel.

3. The method for detecting the tightness of a tool sleeve of a machine tool magazine according to claim 1, characterized in that: Obtaining the tightness detection data between the tool and the tool holder, including at least one of the following: Acquire the spacing distance between the tool and the bottom of the tool sleeve detected by the photoelectric sensor; Acquiring a clamping force between the tool sleeve and the tool detected by a force sensor, wherein the clamping force is used to reflect the degree of fixation between the tool sleeve and the tool; The vibration frequency of the tool sleeve and the tool during the machining process of the machine tool detected by the vibration sensor is obtained.

4. The method for detecting the tightness of a tool sleeve of a machine tool magazine according to claim 1, characterized in that: Feature extraction is performed on the tightness detection data to obtain target feature data, including: The first formula is used to extract the features of the tightness detection data to obtain the target feature data, wherein the first formula is: i (t) = f i [W i *Z i (t)+b i ], i represents the number of the sensor, t represents the current time, f[] represents the activation function, W represents the convolution kernel, b represents the bias, Z i (t) represents the tightness detection data collected by the i-th sensor at the current time t, M i (t) represents the difference with Z i (t) The corresponding target feature data.

5. The method for detecting the tightness of a tool sleeve of a machine tool magazine according to claim 1, characterized in that: Before inputting the target feature data into the tightness determination model to process the target feature data using the tightness determination model to obtain the tightness between the tool sleeve and the tool, the method further includes: The convolutional neural network model is iteratively trained using multiple groups of sample training data collected by each sensor in a historical time period, until an iteration termination condition is reached, and an initial tightness state determination sub-model corresponding to the sensor is obtained, wherein each of the multiple groups of sample training data includes: the sample target feature data, and a sample sub-tightness corresponding to the sample target feature data, and the iteration termination condition is at least one of the following: the number of iterations reaches a preset number of iterations, and the error between the predicted value of the convolutional neural network model and the true value is less than an error threshold; Using the multiple groups of sample training data corresponding to each sensor respectively, a cross-validation method is used to verify the accuracy of the corresponding initial tightness state determination sub-model, so as to obtain the accuracy of each tightness state determination sub-model; In a case where the accuracy is greater than an accuracy threshold, the initial tightness state determination sub-model is determined as the tightness state determination sub-model.

6. The method for detecting the tightness of a tool sleeve of a machine tool magazine according to claim 1, characterized in that: The tightness determination model includes a plurality of tightness determination sub-models, and the target feature data is input into the tightness determination model, so as to process the target feature data using the tightness determination model to obtain the tightness between the tool sleeve and the tool, including: Inputting the target feature data corresponding to each of the sensors into the tension state determination sub-model respectively, so as to process the target feature data using the tension state determination sub-model to obtain the sub-tightness corresponding to the target feature data, wherein each of the tension state determination sub-models is used to process the data collected by a corresponding one of the sensors; The attention weight of each sub-tightness is determined by using a second formula according to the accuracy of each of the tightness states to determine the sub-model, wherein the second formula is: h i represents the accuracy of the tension state determination sub-model corresponding to the i-th sensor, n represents the total number of sensors, s i The attention weight representing the sub-tightness corresponding to the i-th sensor, wherein the attention weight refers to the weight ratio of the sub-tightness in the tightness; The degree of tightness between the tool sleeve and the tool is determined using a third formula according to the plurality of sub-tightnesses and the attention weight corresponding to each sub-tightness, wherein the third formula is: P(t) represents the tightness between the tool sleeve and the tool at the current moment.

7. The method for detecting the tightness of a tool sleeve of a machine tool magazine according to claim 1, characterized in that: The tightness threshold range includes: a first tightness threshold range, a second tightness threshold range, a third tightness threshold range, a fourth tightness threshold range and a fifth tightness threshold range. The tightness state between the tool sleeve and the tool is determined according to the tightness threshold range in which the tightness is located, including: When the tightness is within the first tightness threshold range, determining that the tightness state is not loose, wherein the first tightness threshold range is a range in which the tightness is less than or equal to the first tightness threshold; When the tightness is within the second tightness threshold range, determining that the tightness state is slightly loose, wherein the second tightness threshold range is a range in which the tightness is greater than the first tightness threshold and less than or equal to the second tightness threshold, and the second tightness threshold is greater than the first tightness threshold; When the tightness is within the third tightness threshold range, determining that the tightness state is slightly loose, wherein the third tightness threshold range is a range in which the tightness is greater than the second tightness threshold and less than or equal to the third tightness threshold, and the third tightness threshold is greater than the second tightness threshold; When the tightness is within the fourth tightness threshold range, the tightness state is determined to be slightly loose, wherein the fourth tightness threshold range is a range in which the tightness is greater than the third tightness threshold and less than or equal to the fourth tightness threshold, and the fourth tightness threshold is greater than the third tightness threshold; When the tightness is within the fifth tightness threshold range, the tightness state is determined to be not loose, wherein the fifth tightness threshold range is a range in which the tightness is greater than the fourth tightness threshold.

8. A tool sleeve tightness detection device for a machine tool magazine, characterized in that: include: A first acquisition unit is used to acquire a variety of tightness detection data between the tool and the tool sleeve, wherein the tightness detection data is data collected by a sensor deployed in the tool sleeve during a tightness detection process after the tool enters the tool sleeve; A second acquisition unit is used to extract features from the multiple types of tightness detection data to obtain target feature data corresponding to each type of tightness detection data; a third acquisition unit, configured to input the target feature data into a tightness determination model, so as to process the target feature data using the tightness determination model to obtain the tightness between the tool sleeve and the tool, wherein the tightness determination model is obtained by training with multiple sets of training data by deep learning, and each of the multiple sets of training data includes: sample target feature data and sample tightness corresponding to the sample target feature data; The first determining unit is used to determine the tightness state between the tool sleeve and the tool according to a tightness threshold range of the tightness.

9. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored program, wherein the program executes the method for detecting the tightness of a tool sleeve of a machine tool magazine according to any one of claims 1 to 7.

10. A computer program product comprising computer instructions, characterized in that: When the computer instructions are executed by the processor, the method for detecting the tightness of a tool sleeve of a machine tool magazine according to any one of claims 1 to 7 is executed.

Images