Integrated thermal displacement and temperature rise compensation electric spindle multi-dimensional state monitoring system

Abstract

This invention discloses a multi-dimensional condition monitoring system for an electric spindle integrating thermal displacement and temperature rise compensation. The system includes a multi-dimensional sensing module, a signal conditioning module, a data acquisition module, a central processing module, a thermal compensation execution module, a fault early warning module, a display and interaction module, and a data storage module. These modules are electrically connected sequentially to form a closed-loop monitoring and compensation system. The multi-dimensional sensing module is used to collect multi-dimensional state parameters during the operation of the electric spindle, including thermal displacement parameters, temperature rise parameters, vibration parameters, rotational speed parameters, and load parameters. This multi-dimensional condition monitoring system for an electric spindle integrating thermal displacement and temperature rise compensation integrates the monitoring of multi-dimensional state parameters such as thermal displacement, temperature rise, vibration, rotational speed, and load. It can be expanded to monitor hydraulic and gas pressure, flow rate, and current. Through optimized sensor layout and multi-channel synchronous acquisition, it collects triaxial vibration data, comprehensively capturing the operating status of the electric spindle.

Description

Technical Field

[0001] This invention relates to the field of electric spindle monitoring technology, specifically to a multi-dimensional condition monitoring system for electric spindles that integrates thermal displacement and temperature rise compensation. Background Technology

[0002] As the "heart" of CNC machine tools, the electric spindle's operating status directly determines machining accuracy, production efficiency, and equipment lifespan. It is widely used in aerospace, high-end manufacturing, and precision machining. During high-speed operation, the electric spindle's motor coils and bearings generate heat due to friction, leading to increased spindle temperature and thermal displacement deformation, which is one of the main factors affecting machining accuracy. At the same time, the electric spindle may also experience abnormal vibration, speed fluctuations, and load imbalances during operation. If these issues are not monitored and warned of in a timely manner, they can easily lead to spindle wear, bearing damage, or even machine collisions, causing serious economic losses.

[0003] Currently, existing electric spindle monitoring systems suffer from problems such as limited functionality, insufficient monitoring dimensions, low compensation accuracy, and poor installation adaptability. On the one hand, most monitoring systems only monitor a single parameter (such as temperature or vibration), failing to achieve coordinated monitoring of multiple dimensions such as electric spindle thermal displacement, temperature rise, vibration, speed, and load, making it difficult to comprehensively reflect the operational health of the electric spindle. On the other hand, existing thermal compensation technologies are mostly based on linear fitting models of temperature and thermal displacement, failing to consider the impact of uneven temperature field distribution and nonlinear changes in thermal displacement under complex working conditions (such as changes in cutting speed, depth of cut, and feed rate). This results in limited compensation accuracy, and thermal displacement monitoring and temperature rise compensation are independent of each other, making it impossible to achieve data linkage and real-time dynamic compensation, which is insufficient to meet the precision requirements of high-end precision machining for electric spindles.

[0004] Furthermore, existing monitoring systems mostly employ traditional filtering and threshold judgment methods for signal processing, resulting in weak identification capabilities for weak fault signals and a tendency for false alarms and missed alarms. Simultaneously, these systems lack in-depth analysis and trend prediction capabilities for monitoring data, making it impossible to anticipate potential faults in the electric spindle and hindering predictive maintenance. Moreover, some systems are bulky and complex to install, unsuitable for retrofitting old machine tools. Existing vibration monitoring solutions often use IEPE analog signals, which are susceptible to noise interference and require the development of custom PLC, HMI, and IPC programs, leading to high application development and hardware costs. Typically, they only trigger a shutdown when the spindle momentarily jams, limiting their practical value to end users. For example… For example, some existing technologies only monitor the temperature rise of the electric spindle using temperature sensors and then issue simple alarms based on preset temperature thresholds, without combining thermal displacement data for compensation. This results in errors caused by thermal deformation during machining that cannot be effectively corrected. Other technologies, while achieving thermal displacement monitoring, do not link it with temperature rise data, leading to poor robustness of the compensation model, significant influence from ambient temperature and cutting process conditions, and unsatisfactory practical application results. Therefore, developing a multi-dimensional status monitoring system that integrates thermal displacement and temperature rise compensation, is easy to install, highly adaptable, and provides accurate data, to achieve comprehensive monitoring, precise compensation, and intelligent early warning of the electric spindle's operating status, has significant engineering application value and practical significance. Summary of the Invention

[0005] The purpose of this invention is to provide a multi-dimensional condition monitoring system for electric spindles that integrates thermal displacement and temperature rise compensation, in order to solve the problems mentioned in the background art, such as the single dimension of existing monitoring systems, low thermal compensation accuracy, weak fault identification capability, inability to achieve predictive maintenance, poor installation adaptability, and high application cost.

[0006] To achieve the above objectives, the present invention provides the following technical solution: a multi-dimensional status monitoring system for an electric spindle integrating thermal displacement and temperature rise compensation, comprising a multi-dimensional sensing module, a signal conditioning module, a data acquisition module, a central processing module, a thermal compensation execution module, a fault early warning module, a display and interaction module, and a data storage module. These modules are electrically connected sequentially to form a closed-loop monitoring and compensation system. The multi-dimensional sensing module is used to collect multi-dimensional status parameters during the operation of the electric spindle, including thermal displacement parameters, temperature rise parameters, vibration parameters, speed parameters, and load parameters. The multi-dimensional sensing module includes a thermal displacement sensing unit, a temperature rise sensing unit, a vibration sensing unit, a speed sensing unit, and a load sensing unit. All sensors are designed to be waterproof, dustproof, and anti-interference, with a protection level of not less than IP66. Simultaneously, the sensor installation positions avoid interference with the moving parts of the electric spindle, do not affect the normal operation of the electric spindle, and are adaptable to the installation requirements of both new and old machine tools. The thermal displacement sensing unit employs an eddy current displacement sensor with a resolution of no less than 1 / 80000 of full scale, a measurement range of 0.1mm to 0.5mm, and a sampling time of no more than 1 second. It is fixedly mounted on the bottom of the rotating part of the electric spindle via a bracket, with the sensing head facing the axial plane of the bottom of the rotating part of the electric spindle. This is used to collect axial and radial thermal displacement data of the electric spindle. The bracket has air nozzles fixedly arranged around the sensor head. An internal gas channel is provided within the bracket, with one end connected to the air nozzle and the other end connected to an air source. During operation, airflow is sprayed to blow away cutting fluid and chips around the sensing head, preventing interference with measurement accuracy. When retrofitting old machine tools, the sensor can be adsorbed using a magnetic base accessory. The installation is convenient, either at the nose of the electric spindle or by drilling and locking. The temperature rise sensing unit uses a platinum resistance thermometer (Pt100) with a measurement accuracy of ±0.1℃. It is embedded in the stator, rotor, and bearings of the electric spindle. Temperature-sensitive points are selected using k-means clustering and Pearson correlation analysis to collect temperature data from key parts of the electric spindle and calculate temperature rise parameters. It is also compatible with the Kty84 temperature sensor, with 3 PT100 input channels and 1 Kty84 input channel to adapt to different temperature monitoring needs. The temperature measurement range covers -40℃ to 120℃, with a temperature resolution of 0.1℃. The vibration sensing unit uses a MEMS triaxial sensor. The accelerometer (preferably a six-axis accelerometer) is compact, with the square flat shape measuring only 16×20×7mm and the cylindrical shape measuring M10×35mm, making it the smallest of its kind globally. Its installation position allows it to be close to the vibration source, reducing noise interference and improving data accuracy. The sensor has a bandwidth of up to 6000Hz, an oversampling rate of up to 26,700 times per second, a vibration measurement sensitivity of up to 0.000488g, a maximum measurement efficiency of 24,000 times per second, and a measurement range adjustable in four increments of ±2 / ±4 / ±8 / ±16g. Mounted on the electric spindle housing, it collects vibration acceleration signals in the X, Y, and Z directions of the electric spindle, detecting issues such as bearing wear and rotor imbalance. Fault characteristics; Built-in AI noise reduction function can reduce calibration and installation time; Digital signal transmission is used, and its anti-interference ability is far superior to traditional IEPE analog signals; The speed sensing unit adopts a photoelectric encoder, which is coaxially connected to the output shaft of the electric spindle to collect real-time speed data of the electric spindle. The speed measurement error does not exceed ±0.20 rpm, and it can cover a speed range of 3000~60000 r / min, supporting real-time transmission and analysis of speed signals; The load sensing unit adopts a torque sensor, which is connected in series in the electric spindle transmission link to collect real-time load torque data of the electric spindle; At the same time, the system can be expanded to access parameters such as hydraulic and gas pressure, flow rate, and current monitoring to adapt to multiple monitoring needs; The signal conditioning module is used to preprocess the raw signals acquired by the multi-dimensional sensing module, eliminate noise interference, and improve signal quality. The signal conditioning module includes a filter circuit, an amplifier circuit, and an analog-to-digital conversion circuit, which are integrated into the multi-functional acquisition unit to simplify the system structure. The filtering circuit employs a second-order active low-pass filter with an adjustable cutoff frequency, used to filter out high-frequency noise and interference signals in the original signal; the amplification circuit uses an instrumentation amplifier with an adjustable amplification factor (1~100 times), used to amplify weak sensing signals to a range suitable for data acquisition; the analog-to-digital conversion circuit uses a 16-bit high-speed ADC converter with a sampling frequency of not less than 10kHz, used to convert analog signals into digital signals and transmit them to the data acquisition module; the vibration analog output error does not exceed ±0.1V, the position signal true value error does not exceed ±0.1V, the temperature true value error does not exceed ±1.17℃, and the temperature operating cycle time error does not exceed ±2.40s; The data acquisition module employs a high-speed data acquisition card (integrated into the multi-functional acquisition unit), electrically connected to the signal conditioning module and the central processing module. It receives digital signals output from the signal conditioning module, enabling synchronous acquisition and buffering of multi-dimensional data such as thermal displacement, temperature rise, vibration, rotational speed, and load. The data acquisition module supports multi-channel synchronous acquisition, with a minimum of 8 channels and a buffer capacity of at least 16GB. The acquisition frequency (adjustable from 1 to 10kHz) can be set according to monitoring requirements. Acquired data is transmitted to the central processing module via a high-speed bus (PCIe) to ensure real-time and complete data transmission. Simultaneously, it supports one encoder input, one position sensor input, and two 4-20mA analog inputs, enabling synchronous acquisition of multiple signal types. The multi-functional data collector measures W90×H70×D25mm, uses an embedded operating system, has a startup time of no more than 5s, a power supply voltage of DC24V, uses RS485 for communication with a communication distance of up to 20m, has a collision response time of no more than 0.1ms (actual response speed depends on the CNC controller processing speed), has a collision data storage capacity of no less than 1000 records, a feature data storage capacity of no less than 2 hours, an operating temperature range of -25℃ to 70℃, and is installed using a threaded fixing method, with an adapter for rear-mounted electric spindle. The central processing module, as the core control unit of the system, adopts an industrial-grade embedded processor (such as ARM Cortex-A9) and integrates data processing, compensation control, fault diagnosis and logic control functions. The central processing module includes a data processing unit, a compensation control unit and a fault diagnosis unit, and also integrates a time synchronization unit and a microcomputer. It supports Wi-Fi wireless connection and eliminates the need to purchase additional human-machine interface and IPC hardware, thus reducing application costs. The time synchronization unit uses GPS or network time synchronization to ensure synchronized data acquisition and processing across modules, avoiding data misalignment and improving monitoring and compensation accuracy. The data processing unit analyzes and processes the multidimensional data transmitted by the data acquisition module, including data denoising, outlier removal, and feature extraction. Specifically, a wavelet threshold denoising algorithm is used to denoise the vibration signal, combined with the AI ​​denoising function of the MEMS sensor to further improve signal quality. The 3σ criterion is used to remove outlier data, and the peak value, RMS value, kurtosis, and waveform factor of the vibration signal are extracted. Characteristic parameters, such as the rate of change and steady-state value of temperature rise data, and the amplitude and trend of thermal displacement data, provide data support for fault diagnosis and thermal compensation. In random vibration testing, the RMS extreme value error does not exceed ±0.19g, and the statistical time error does not exceed ±5s. The compensation control unit is used to construct a coupled thermal displacement-temperature rise compensation model based on thermal displacement and temperature rise data to achieve real-time dynamic compensation of the electric spindle's thermal error. The coupled thermal displacement-temperature rise compensation model adopts a BiLSTM network model optimized by the CPO-IGWO dual-algorithm strategy, after screening... Temperature rise and thermal displacement data from temperature-sensitive points are used as inputs. A three-dimensional thermal model of the electric spindle is established using Abaqus software to simulate the temperature and displacement fields during high-speed operation. The model is trained using experimentally collected data and outputs thermal compensation. The compensation control unit converts the thermal compensation into control signals and transmits them to the thermal compensation execution module to achieve closed-loop compensation of thermal errors. Simultaneously, the compensation control unit can dynamically adjust compensation parameters according to the speed and load changes of the electric spindle to improve compensation accuracy. The fault diagnosis unit is used to identify and diagnose common faults of the electric spindle based on multi-dimensional feature parameters and an improved BP neural network algorithm. Preset fault types include bearing wear, rotor imbalance, stator overheating, load overload, abnormal thermal displacement, and collision. By inputting the extracted feature parameters into the trained BP neural network model, the fault type, fault level, and fault location are output, and the fault development trend is calculated to provide a basis for fault early warning. The improved BP neural network improves fault identification accuracy and convergence speed by adding a momentum term and adaptive learning rate. The fault identification accuracy is not less than 96%, which can effectively identify weak fault signals and reduce false alarms and missed alarms. The thermal compensation execution module is electrically connected to the central processing module and is used to receive the compensation control signal output by the central processing module and execute thermal displacement compensation actions. The thermal compensation execution module includes a piezoelectric ceramic micro-displacement actuator and a displacement feedback sensor. The piezoelectric ceramic micro-displacement actuator is installed on the feed mechanism of the electric spindle, with a resolution of not less than 0.1μm and a response time of not more than 10ms. It outputs corresponding micro-displacements according to the compensation control signal to achieve accurate compensation of the thermal displacement of the electric spindle. The displacement feedback sensor is used in conjunction with the piezoelectric ceramic micro-displacement actuator to collect the actual displacement data after compensation and feed it back to the central processing module to form a closed-loop compensation control to ensure compensation accuracy. It also has a vibration suppression function, which can maximize the usable tool life, accurately predict the remaining tool life, and adapt to the needs of unmanned intelligent factories to replace tools in advance. The fault early warning module is electrically connected to the central processing module and is used to realize graded early warning of electric spindle faults based on the fault diagnosis results and data trends output by the central processing module. The fault early warning module includes an audible and visual alarm unit and a communication alarm unit. The audible and visual alarm unit uses LED indicator lights and a buzzer. The communication alarm unit uses RS485, Ethernet or 5G communication modules to transmit fault information (fault type, fault level, occurrence time) to the field monitoring terminal and remote management platform in real time. It also supports SMS alarm function to promptly notify maintenance personnel for handling, reducing the fault false alarm rate and the missed alarm rate. It has a two-level critical value alarm function and is equipped with two DO outputs, which can be directly connected to the CNC controller IO terminal (PLC IO terminal) to realize emergency stop control. The machine tool manufacturer can set the maximum vibration limit of the "machine table", and the end user can adjust the maximum vibration limit of the "workpiece" according to the processing conditions. The display and interaction module uses a 10-inch industrial touch screen, which is electrically connected to the central processing module. It is used to display the multi-dimensional status parameters (thermal displacement, temperature rise, vibration, speed, load), thermal compensation, fault information, and operating status of the electric spindle in real time. It also supports manual interaction, allowing users to set monitoring parameter thresholds, compensation parameters, alarm parameters, query historical data and fault records, and achieve flexible configuration of system parameters. It also supports wireless connection to a tablet, facilitating on-site debugging and operation without the need for complex programming. The data storage module combines solid-state drives (SSDs) and cloud storage, and is electrically connected to the central processing module. It stores real-time monitoring data, historical data, fault records, compensation parameters, etc., of the electric spindle. The SSD has a storage capacity of at least 128GB for local data storage, ensuring data integrity even after power failure. Cloud storage connects via Ethernet or 5G networks, enabling remote data backup and sharing. It supports historical data queries, trend analysis, and data export, providing data support for predictive maintenance and performance optimization of the electric spindle. The data storage frequency can be specified between 0.1 and 5 seconds, with a minimum of 0.1 seconds for collecting and storing the peak monitoring parameters within that 0.1-second time interval. The data is recorded and stored cyclically for over 30 days. It supports exporting data logs and spindle diagnostic reports (PDF format), which include current configuration information, spindle continuous runtime statistics, operation statistics, vibration statistics, and collision events.

[0007] Preferably, the integrated thermal displacement and temperature rise compensation electric spindle multi-dimensional condition monitoring system further includes a power supply module to provide stable operating power for each module. The power supply module adopts a switching power supply with an input voltage of AC220V and an output voltage of DC5V, DC12V, and DC24V. It has overvoltage, overcurrent, and short-circuit protection functions to ensure stable system operation. The multi-functional data acquisition unit can be directly connected to a 24V power supply to meet the power supply needs of industrial sites.

[0008] Preferably, the integrated thermal displacement and temperature rise compensation electric spindle multi-dimensional condition monitoring system supports two installation modes to adapt to different machine tool scenarios: New machine tool integration: The spindle is equipped with a digital vibration sensor, the smart box (integrating signal conditioning, data acquisition, and central processing functions) is connected to a 24V power supply and the CNC controller IO terminal, and the tablet is wirelessly connected to the smart box to check the machine tool vibration and set the maximum vibration limit of the machine tool; Old machine tool retrofit: The vibration sensor is attached to the nose of the electric spindle with a magnetic base accessory, or a hole is drilled for locking, the smart box is connected to a 24V power supply and the CNC controller IO terminal, and the tablet is wirelessly connected to the smart box to complete parameter setting and debugging.

[0009] Compared with the prior art, the beneficial effects of the present invention are: (1) The electric spindle multi-dimensional status monitoring system integrating thermal displacement and temperature rise compensation has comprehensive monitoring dimensions and realizes multi-dimensional collaborative monitoring: The present invention integrates the monitoring of multi-dimensional status parameters such as thermal displacement, temperature rise, vibration, speed, and load, and can be expanded to monitor liquid pressure, flow rate, and current. Through optimized sensor layout and multi-channel synchronous acquisition, triaxial vibration data is collected to fully capture the operating status of the electric spindle, solve the problem of the existing system having a single monitoring dimension and being unable to fully reflect the health status of the spindle, provide comprehensive data support for fault diagnosis and thermal compensation, and adapt to the monitoring needs of high-end rotating machinery and ordinary CNC machine tools; (2) The integrated thermal displacement and temperature rise compensation electric spindle multi-dimensional state monitoring system has high thermal compensation accuracy and realizes dynamic closed-loop compensation: The present invention constructs a thermal displacement-temperature rise coupled compensation model, adopts a BiLSTM network model optimized by CPO-IGWO dual algorithm strategy, and combines electric spindle three-dimensional thermal simulation and experimental data training to consider the influence of complex working conditions such as speed and load, so as to realize the linkage compensation of thermal displacement and temperature rise data; at the same time, through closed-loop feedback control, the compensation parameters are dynamically adjusted to significantly improve the thermal compensation accuracy and effectively correct the machining error caused by electric spindle thermal deformation. Combined with the vibration suppression function and tool life prediction, the machining defect rate can be reduced to meet the needs of high-end precision machining. (3) The electric spindle multi-dimensional condition monitoring system integrating thermal displacement and temperature rise compensation has accurate fault identification and realizes graded early warning and predictive maintenance: The present invention adopts an improved BP neural network algorithm and combines multi-dimensional feature parameters for fault diagnosis, which improves the fault identification accuracy and enhances the ability to identify weak fault signals; it integrates collision monitoring function, and the collision response time does not exceed 0.1ms, which can effectively avoid the expansion of faults; at the same time, through fault trend analysis, it realizes graded early warning of faults, supports audible and visual alarms, remote communication alarms and SMS alarms, and promptly notifies maintenance personnel to handle the situation, reducing equipment wear and downtime losses, promoting the transformation of electric spindle maintenance mode from periodic inspection to predictive maintenance, and adapting to the intelligent needs of unmanned intelligent factories; (4) The integrated thermal displacement and temperature rise compensation electric spindle multi-dimensional status monitoring system has strong system stability, is suitable for harsh working environments and is easy to install: the multi-dimensional sensing module adopts waterproof, dustproof and anti-interference design, and the protection level is not lower than IP66. The power module has overvoltage, overcurrent and short circuit protection functions. The central processing module adopts industrial-grade embedded processor and the multi-functional data acquisition unit adopts industrial-grade protection design to ensure long-term stable operation of the system in the harsh working environment of CNC machine tools. At the same time, it supports two installation modes: new machine tool integration and old machine tool modification. Old machine tools can be quickly installed through magnetic base without complicated modification. The installation is convenient and adaptable to various machine tool scenarios. It adopts a combination of local storage and cloud storage to ensure data security and facilitate subsequent data traceability and performance optimization. (5) The integrated thermal displacement and temperature rise compensation electric spindle multi-dimensional condition monitoring system is interactive, practical, and low in application cost: parameter setting, data display and operation control are realized through industrial touch screen and tablet wireless connection. The interface is intuitive and easy to operate. CNC technicians can complete the debugging with one click and there is no learning threshold. It supports remote communication and data sharing, which is convenient for operation and maintenance personnel to remotely monitor and manage. The system can flexibly adjust the monitoring parameters and compensation parameters according to different models and working conditions of electric spindles. It has strong adaptability and can be widely used in electric spindle monitoring and compensation of various CNC machine tools and machining centers. It integrates a microcomputer, so there is no need to purchase additional human-machine interface and IPC hardware. The standard version is free of R&D and integration, and the professional version provides a standard communication interface, which greatly reduces the application development cost and hardware cost. Compared with traditional vibration sensing solutions, it has significant advantages in size, anti-interference ability and application convenience. Detailed Implementation

[0010] The technical solutions in the embodiments of the present invention will be clearly and completely described below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0011] This invention provides a technical solution: a multi-dimensional status monitoring system for an electric spindle integrating thermal displacement and temperature rise compensation, comprising a multi-dimensional sensing module, a signal conditioning module, a data acquisition module, a central processing module, a thermal compensation execution module, a fault early warning module, a display and interaction module, and a data storage module. These modules are electrically connected sequentially to form a closed-loop monitoring and compensation system. The multi-dimensional sensing module is used to collect multi-dimensional status parameters during the operation of the electric spindle, including thermal displacement parameters, temperature rise parameters, vibration parameters, speed parameters, and load parameters. The multi-dimensional sensing module includes a thermal displacement sensing unit, a temperature rise sensing unit, a vibration sensing unit, a speed sensing unit, and a load sensing unit. All sensors are designed to be waterproof, dustproof, and anti-interference, with a protection level of no less than IP66, adapting to the harsh working environment of CNC machine tools and ensuring long-term stable operation of the sensors. Simultaneously, the sensor installation positions avoid interference with the moving parts of the electric spindle, do not affect the normal operation of the electric spindle, and are compatible with the installation requirements of both new and old machine tools. The thermal displacement sensing unit uses an eddy current displacement sensor with a resolution of no less than 1 / 80000 of full scale, a measurement range of 0.1mm to 0.5mm, and a sampling time of no more than 1 second. It is fixedly mounted on the bottom of the rotating part of the electric spindle via a bracket, with the sensing head facing the axial plane of the bottom of the rotating part of the electric spindle. It is used to collect axial and radial thermal displacement data of the electric spindle. Air nozzles are fixedly arranged around the sensor head on the bracket. A gas flow channel is provided inside the bracket, with one end connected to the air nozzle and the other end connected to an air source. During operation, airflow is sprayed to blow away cutting fluid and chips around the sensing head, preventing them from affecting measurement accuracy. When retrofitting old machine tools, the sensor can be attracted using a magnetic base accessory. Attached to the nose of the electric spindle or secured by drilling, installation is convenient. The temperature rise sensing unit uses a platinum resistance thermometer (Pt100) with a measurement accuracy of ±0.1℃. It is embedded in the stator, rotor, and bearings of the electric spindle, and temperature-sensitive points are selected using k-means clustering and Pearson correlation analysis to collect temperature data from key parts of the electric spindle and calculate temperature rise parameters. It is also compatible with the Kty84 temperature sensor, with 3 PT100 input channels and 1 Kty84 input channel to adapt to different temperature monitoring needs. The temperature measurement range covers -40℃ to 120℃, with a temperature resolution of 0.1℃. The vibration sensing unit uses a MEMS triaxial sensor. The speed sensor (preferably a six-axis accelerometer) is compact, with the square flat shape measuring only 16×20×7mm and the cylindrical shape measuring M10×35mm, making it the smallest of its kind globally. Its installation position allows it to be close to the vibration source, reducing noise interference and improving data accuracy. The sensor has a bandwidth of up to 6000Hz, an oversampling rate of up to 26,700 times per second, a vibration measurement sensitivity of up to 0.000488g, a maximum measurement efficiency of 24,000 times per second, and a measurement range adjustable in four increments of ±2 / ±4 / ±8 / ±16g. Mounted on the electric spindle housing, it collects vibration acceleration signals in the X, Y, and Z directions of the electric spindle, detecting bearing wear and rotor unevenness. The system features fault characteristics such as imbalance; it incorporates AI noise reduction to reduce calibration and installation time; it uses digital signal transmission, offering significantly better anti-interference capabilities than traditional IEPE analog signals; the speed sensing unit employs a photoelectric encoder, coaxially connected to the electric spindle output shaft, to collect real-time speed data of the electric spindle, with a speed measurement error not exceeding ±0.20 rpm, covering a speed range of 3000~60000 r / min, and supporting real-time transmission and analysis of speed signals; the load sensing unit uses a torque sensor, connected in series in the electric spindle drive link, to collect real-time load torque data of the electric spindle; simultaneously, the system can be expanded to monitor parameters such as hydraulic and gas pressure, flow rate, and current, adapting to monitoring needs in multiple scenarios; The signal conditioning module is used to preprocess the raw signals acquired by the multi-dimensional sensing module, eliminate noise interference, and improve signal quality. The signal conditioning module includes a filter circuit, an amplifier circuit, and an analog-to-digital conversion circuit, which are integrated into the multi-functional acquisition unit to simplify the system structure. The filtering circuit employs a second-order active low-pass filter with an adjustable cutoff frequency, used to filter out high-frequency noise and interference signals in the original signal; the amplification circuit uses an instrumentation amplifier with an adjustable amplification factor (1~100 times), used to amplify weak sensing signals to a range suitable for data acquisition; the analog-to-digital conversion circuit uses a 16-bit high-speed ADC converter with a sampling frequency of not less than 10kHz, used to convert analog signals into digital signals and transmit them to the data acquisition module; the vibration analog output error does not exceed ±0.1V, the position signal true value error does not exceed ±0.1V, the temperature true value error does not exceed ±1.17℃, and the temperature operating cycle time error does not exceed ±2.40s; The data acquisition module uses a high-speed data acquisition card (integrated into the multi-functional acquisition unit), which is electrically connected to the signal conditioning module and the central processing module. It is used to receive digital signals output by the signal conditioning module to achieve synchronous acquisition and buffering of multi-dimensional data such as thermal displacement, temperature rise, vibration, speed, and load. The data acquisition module supports multi-channel synchronous acquisition, with no less than 8 channels and a buffer capacity of no less than 16GB. The acquisition frequency can be set according to monitoring needs (adjustable from 1 to 10kHz). The acquired data is transmitted to the central processing module through a high-speed bus (PCIe) to ensure the real-time performance and integrity of data transmission. It also supports 1 encoder input, 1 position sensor input, and 2 4-20mA analog inputs to achieve synchronous acquisition of multiple types of signals. The multi-functional data acquisition unit measures W90×H70×D25mm, uses an embedded operating system, has a startup time of no more than 5 seconds, is powered by DC24V, communicates via RS485 with a communication distance of up to 20m, has a collision response time of no more than 0.1ms (actual response speed depends on the CNC controller processing speed), has a collision data storage capacity of no less than 1000 records, a feature data storage capacity of no less than 2 hours, an operating temperature range of -25℃ to 70℃, and is installed using a threaded fixing method, compatible with rear-mounted electric spindles. The central processing module, as the core control unit of the system, adopts an industrial-grade embedded processor (such as ARM Cortex-A9) and integrates data processing, compensation control, fault diagnosis and logic control functions. The central processing module includes a data processing unit, a compensation control unit and a fault diagnosis unit, and also integrates a time synchronization unit and a microcomputer. It supports Wi-Fi wireless connection and eliminates the need to purchase additional human-machine interface and IPC hardware, thus reducing application costs. The time synchronization unit uses GPS or network time synchronization to ensure synchronized data acquisition and processing across modules, avoiding data misalignment and improving monitoring and compensation accuracy. The data processing unit analyzes and processes the multidimensional data transmitted from the data acquisition modules, including data denoising, outlier removal, and feature extraction. Specifically, a wavelet threshold denoising algorithm is used to denoise the vibration signal, combined with the AI ​​denoising function of the MEMS sensor to further improve signal quality. The 3σ criterion is used to remove outlier data. By extracting characteristic parameters such as peak value, RMS value, kurtosis, and waveform factor of the vibration signal, the rate of change and steady-state value of temperature rise data, and the amplitude and trend of thermal displacement data, data is provided for fault diagnosis and thermal compensation. According to the support; in random vibration testing, the RMS extreme value error does not exceed ±0.19g, and the statistical time error does not exceed ±5s; the compensation control unit is used to construct a thermal displacement-temperature rise coupled compensation model based on thermal displacement and temperature rise data to realize real-time dynamic compensation of the thermal error of the electric spindle; the thermal displacement-temperature rise coupled compensation model adopts a BiLSTM network model optimized by the CPO-IGWO dual algorithm strategy, using the temperature rise data and thermal displacement data of the selected temperature sensitive points as input, and establishing a three-dimensional thermal model of the electric spindle through Abaqus software to simulate the temperature field and displacement field during high-speed operation, and training the model with experimentally collected data to output the thermal compensation amount; the compensation control unit converts the thermal compensation amount into a control signal and transmits it to the thermal compensation execution unit. The module enables closed-loop compensation for thermal errors. Simultaneously, the compensation control unit dynamically adjusts compensation parameters based on changes in the electric spindle's speed and load, improving compensation accuracy. During the first piece processing, the system automatically sets the maximum vibration limit for the workpiece using machine learning vibration curves. Subsequent processing automatically compares this limit, and if it exceeds the limit, the machine stops for further processing, reducing the defect rate to 0%. The fault diagnosis unit uses multi-dimensional feature parameters combined with an improved BP neural network algorithm to identify and diagnose common electric spindle faults. Preset fault types include bearing wear, rotor imbalance, stator overheating, load overload, abnormal thermal displacement, and collisions. By inputting the extracted feature parameters into the trained BP neural network model, the system outputs the fault type, fault level, and fault location. The system calculates the fault development trend to provide a basis for fault early warning. The improved BP neural network enhances fault identification accuracy and convergence speed by adding a momentum term and adaptive learning rate, achieving a fault identification accuracy of no less than 96%. It can effectively identify weak fault signals and reduce false alarms and missed alarms. In collision tests, the system can accurately record collision events. It can completely record 6 impact events in 6 impact tests, with a collision extreme value error of no more than ±2g. The first trigger threshold time is 0ms. The alarm signal response time is the time from when the acquisition device captures the collision signal to when the fault early warning module issues an alarm signal. The maximum value is no more than 0.20ms, and the minimum alarm signal duration is no less than 100ms, which can effectively prevent the collision fault from escalating. The thermal compensation execution module is electrically connected to the central processing module and is used to receive the compensation control signal output by the central processing module to execute thermal displacement compensation actions. The thermal compensation execution module includes a piezoelectric ceramic micro-displacement actuator and a displacement feedback sensor. The piezoelectric ceramic micro-displacement actuator is installed on the feed mechanism of the electric spindle, with a resolution of not less than 0.1μm and a response time of not more than 10ms. It outputs the corresponding micro-displacement according to the compensation control signal to achieve accurate compensation of the thermal displacement of the electric spindle. The displacement feedback sensor is used in conjunction with the piezoelectric ceramic micro-displacement actuator to collect the actual displacement data after compensation and feed it back to the central processing module to form a closed-loop compensation control to ensure compensation accuracy. It also has a vibration suppression function, which can maximize the usable tool life, accurately predict the remaining tool life, and adapt to the needs of unmanned intelligent factories to replace tools in advance. The fault early warning module is electrically connected to the central processing module and is used to realize graded early warning of electric spindle faults based on the fault diagnosis results and data trends output by the central processing module. The fault early warning module includes an audible and visual alarm unit and a communication alarm unit. The audible and visual alarm unit uses LED indicator lights and buzzers to emit different colored lights and different frequency alarm sounds according to the fault level (minor, general, severe). Minor faults emit yellow lights and low-frequency buzzers, general faults emit orange lights and medium-frequency buzzers, and severe faults emit red lights and high-frequency buzzers. The communication alarm unit uses RS485, Ethernet, or 5G communication modules to transmit fault information (fault type, fault level, occurrence time) to the field monitoring terminal and remote management platform in real time. It also supports SMS alarm function to promptly notify maintenance personnel for handling, reducing the fault missed rate and false alarm rate. It has a two-level critical value alarm function and is equipped with two DO outputs, which can be directly connected to the CNC controller IO terminal (PLC IO terminal) to realize emergency stop control. The machine tool manufacturer can set the maximum vibration limit of the "machine table", and the end user can adjust the maximum vibration limit of the "workpiece" according to the processing conditions. The display and interaction module uses a 10-inch industrial touchscreen, electrically connected to the central processing module, to display the multi-dimensional status parameters (thermal displacement, temperature rise, vibration, speed, load), thermal compensation, fault information, and operating status of the electric spindle in real time. It also supports manual interaction, allowing users to set monitoring parameter thresholds, compensation parameters, alarm parameters, query historical data and fault records, and flexibly configure system parameters. Wireless tablet connectivity facilitates on-site debugging and operation without complex programming. CNC operators only need to press a one-click learning button when making the first piece to complete vibration learning for all 64 tools. Each tool also provides a two-stage alarm, with no learning curve. The professional version provides a standard communication interface (such as EtherCAT), making it easy for machine tool manufacturers to develop and add value themselves. The standard version includes a user-end application APP, enabling intelligent upgrades of combat-ready machine tool models without R&D or integration requirements. The data storage module combines solid-state drives (SSDs) and cloud storage, electrically connected to the central processing module. It stores real-time monitoring data, historical data, fault records, compensation parameters, etc., of the electric spindle. The SSD storage capacity is no less than 128GB for local data storage, ensuring data integrity even after power failure. Cloud storage connects via Ethernet or 5G networks, enabling remote data backup and sharing. It supports historical data queries, trend analysis, and data export, providing data support for predictive maintenance and performance optimization of the electric spindle. The data storage frequency can be specified between 0.1 and 5 seconds, with a minimum of 0.1 seconds for collecting and storing the peak monitoring parameters within that 0.1-second time interval. The data is recorded and stored cyclically for over 30 days. It supports data log export and spindle diagnostic report export (PDF format). The report includes current configuration information, spindle continuous runtime statistics, operation statistics, vibration statistics, and collision events. Furthermore, an integrated multi-dimensional condition monitoring system for electric spindles with thermal displacement and temperature rise compensation also includes a power supply module to provide stable operating power to each module. The power supply module adopts a switching power supply with an input voltage of AC220V and an output voltage of DC5V, DC12V, and DC24V. It has overvoltage, overcurrent, and short-circuit protection functions to ensure stable system operation. The multi-functional data acquisition unit can be directly connected to a 24V power supply to meet the power supply needs of industrial sites. Furthermore, an integrated electric spindle multi-dimensional condition monitoring system with thermal displacement and temperature rise compensation supports two installation modes to adapt to different machine tool scenarios: New machine tool integration: The spindle is equipped with a digital vibration sensor, and the smart box (integrating signal conditioning, data acquisition, and central processing functions) is connected to a 24V power supply and the CNC controller's IO terminals. The tablet wirelessly connects to the smart box to check the machine tool vibration and set the maximum vibration limit. Old machine tool retrofitting: The vibration sensor is attached to the nose of the electric spindle using a magnetic base accessory, or secured by drilling. The smart box is connected to a 24V power supply and the CNC controller's IO terminals, and the tablet wirelessly connects to the smart box to complete parameter settings and debugging. All content not described in detail in this specification belongs to existing technology known to those skilled in the art.

[0012] Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-dimensional condition monitoring system for an electric spindle integrating thermal displacement and temperature rise compensation, comprising a multi-dimensional sensing module, a signal conditioning module, a data acquisition module, a central processing module, a thermal compensation execution module, a fault early warning module, a display and interaction module, and a data storage module, wherein each module is electrically connected in sequence to form a closed-loop monitoring and compensation system, characterized in that: The multidimensional sensing module is used to collect multidimensional state parameters during the operation of the electric spindle, including thermal displacement parameters, temperature rise parameters, vibration parameters, speed parameters, and load parameters; the multidimensional sensing module includes a thermal displacement sensing unit, a temperature rise sensing unit, a vibration sensing unit, a speed sensing unit, and a load sensing unit. The thermal displacement sensing unit uses an eddy current displacement sensor with a resolution of no less than 1 / 80000 of full scale, a measurement range of 0.1mm to 0.5mm, and a sampling time of no more than 1 second. It is fixedly mounted on the bottom of the rotating part of the electric spindle via a bracket, with the sensing head facing the axial plane of the bottom of the rotating part of the electric spindle. It is used to collect axial and radial thermal displacement data of the electric spindle. The bracket has air nozzles fixedly arranged around the sensor head. The bracket has a gas flow channel inside, with one end connected to the air nozzle and the other end connected to an air source. The air source is dry compressed air with a pressure range of 0.3 to 0.6 MPa. During operation, the airflow is sprayed to blow away cutting fluid and chips around the sensing head, avoiding affecting measurement accuracy. When retrofitting old machine tools, the sensor can be magnetically attached to the nose of the electric spindle using a magnetic base accessory, or fixed by drilling. The temperature rise sensing unit uses a platinum resistance thermometer (Pt100) with a measurement accuracy of ±0.1℃. It is embedded in the stator, rotor, and bearing of the electric spindle and arranged with temperature sensitive points selected by k-means clustering algorithm and Pearson correlation analysis. It is also compatible with Kty84 temperature sensor and has 3 PT100 input channels and 1 Kty84 input channel. The vibration sensing unit uses a MEMS triaxial accelerometer (preferably a six-axis accelerometer), with a flat square shape of only 16×20×7mm and a cylindrical shape of M10×35mm. The installation position can be close to the machining vibration source and installed on the electric spindle housing. It is used to collect vibration acceleration signals in the X, Y, and Z directions of the electric spindle and capture fault characteristics such as bearing wear and rotor imbalance. Built-in AI noise reduction function can reduce calibration and installation time; adopts digital signal transmission, and its anti-interference ability is far superior to traditional IEPE analog signals. The speed sensing unit uses a photoelectric encoder, which is coaxially connected to the output shaft of the electric spindle. It is used to collect real-time speed data of the electric spindle. The speed measurement error does not exceed ±0.20 rpm, and it can cover a speed range of 3000~60000 r / min. It supports real-time transmission and analysis of speed signals. The load sensing unit uses a torque sensor connected in series in the electric spindle drive link to collect real-time load torque data of the electric spindle; at the same time, the system can be expanded to access parameters such as hydraulic and gas pressure, flow rate, and current monitoring to adapt to monitoring needs in multiple scenarios. The signal conditioning module is used to preprocess the raw signals acquired by the multi-dimensional sensing module, eliminate noise interference, and improve signal quality; the signal conditioning module includes a filtering circuit, an amplification circuit, and an analog-to-digital conversion circuit, which are integrated into the multi-functional acquisition unit; The filtering circuit adopts a second-order active low-pass filter circuit with an adjustable cutoff frequency; the amplification circuit adopts an instrumentation amplifier with an adjustable amplification factor (1~100 times); the analog-to-digital conversion circuit adopts a 16-bit high-speed ADC converter with a sampling frequency of not less than 10kHz, used to convert analog signals into digital signals and transmit them to the data acquisition module; the vibration analog output error does not exceed ±0.1V, the position signal true value error does not exceed ±0.1V, the temperature true value error does not exceed ±1.17℃, and the temperature working cycle time error does not exceed ±2.40s; The data acquisition module employs a high-speed data acquisition card (integrated into the multi-functional acquisition unit), electrically connected to the signal conditioning module and the central processing module. It receives digital signals output from the signal conditioning module, enabling synchronous acquisition and buffering of multi-dimensional data such as thermal displacement, temperature rise, vibration, rotational speed, and load. The data acquisition module supports multi-channel synchronous acquisition, with a minimum of 8 channels and a buffer capacity of at least 16GB. The acquisition frequency (adjustable from 1 to 10kHz) can be set according to monitoring requirements. Acquired data is transmitted to the central processing module via a high-speed bus (PCIe) to ensure real-time and complete data transmission. Simultaneously, it supports one encoder input, one position sensor input, and two 4-20mA analog inputs, enabling synchronous acquisition of multiple signal types. The multi-functional data acquisition unit measures W90×H70×D25mm, uses an embedded operating system, has a startup time of no more than 5s, is powered by DC24V, communicates via RS485 with a communication distance of up to 20m, and has a collision response time of no more than 0.1ms from the occurrence of a collision to the acquisition unit capturing the collision signal (actual response speed depends on the CNC controller processing speed). It has a collision data storage capacity of no less than 1000 records, a feature data storage capacity of no less than 2 hours, an operating temperature range of -25℃ to 70℃, and is installed using a threaded fixing method, with the adapter mounted at the rear of the electric spindle. The central processing module, as the core control unit of the system, adopts an industrial-grade embedded processor (such as ARM Cortex-A9) and integrates data processing, compensation control, fault diagnosis and logic control functions. The central processing module includes a data processing unit, a compensation control unit and a fault diagnosis unit, and also integrates a time synchronization unit and a microcomputer, supporting Wi-Fi wireless connection. The time synchronization unit uses GPS or network time synchronization to ensure synchronized data acquisition and processing of each module. The data processing unit is used to analyze and process the multidimensional data transmitted by the data acquisition module, including data denoising, outlier removal, and feature extraction. Among them, wavelet threshold denoising algorithm is used to denoise the vibration signal, and combined with the AI ​​denoising function of MEMS sensor, the signal quality is further improved. The 3σ criterion is used to remove outlier data. By extracting characteristic parameters such as peak value, effective value, kurtosis, and waveform factor of vibration signal, characteristic parameters such as rate of change and steady-state value of temperature rise data, and characteristic parameters such as amplitude and trend of thermal displacement data, data support is provided for fault diagnosis and thermal compensation. In random vibration test, RMS extreme value error does not exceed ±0.19g, and statistical time error does not exceed ±5s. The compensation control unit is used to construct a coupled thermal displacement-temperature rise compensation model based on thermal displacement and temperature rise data. This model employs a BiLSTM network optimized using a CPO-IGWO dual-algorithm strategy. It takes filtered temperature rise and thermal displacement data from temperature-sensitive points as input, and uses Abaqus software to establish a three-dimensional thermal model of the electric spindle, simulating the temperature and displacement fields during high-speed operation. The model is trained using experimentally acquired data and outputs the thermal compensation amount. The compensation control unit converts the thermal compensation amount into a control signal and transmits it to the thermal compensation execution module to achieve closed-loop compensation of thermal errors. Simultaneously, the compensation control unit can dynamically adjust the compensation parameters according to changes in the electric spindle's speed and load, improving compensation accuracy. The fault diagnosis unit is used to identify and diagnose common electric spindle faults based on multi-dimensional feature parameters and an improved BP neural network algorithm. The thermal compensation execution module is electrically connected to the central processing module and is used to receive the compensation control signal output by the central processing module and execute thermal displacement compensation actions. The thermal compensation execution module includes a piezoelectric ceramic micro-displacement actuator and a displacement feedback sensor. The piezoelectric ceramic micro-displacement actuator is installed on the feed mechanism of the electric spindle, with a resolution of not less than 0.1μm and a response time of not more than 10ms. It outputs corresponding micro-displacements according to the compensation control signal to achieve accurate compensation of the thermal displacement of the electric spindle. The displacement feedback sensor is used in conjunction with the piezoelectric ceramic micro-displacement actuator to collect the actual displacement data after compensation and feed it back to the central processing module to form a closed-loop compensation control to ensure compensation accuracy. It also has a vibration suppression function, which can maximize the usable tool life, accurately predict the remaining tool life, and adapt to the needs of unmanned intelligent factories to replace tools in advance. The fault early warning module is electrically connected to the central processing module and is used to realize graded early warning of electric spindle faults based on the fault diagnosis results and data trends output by the central processing module. The fault early warning module includes an audible and visual alarm unit and a communication alarm unit. The audible and visual alarm unit uses LED indicator lights and a buzzer. The communication alarm unit uses RS485, Ethernet or 5G communication modules. The display interaction module uses a 10-inch industrial touch screen, which is electrically connected to the central processing module. It also supports manual interaction, allowing users to set monitoring parameter thresholds, compensation parameters, alarm parameters, query historical data and fault records, and achieve flexible configuration of system parameters. It also supports wireless connection to a tablet, facilitating on-site debugging and operation. The data storage module combines solid-state drives (SSDs) and cloud storage, and is electrically connected to the central processing module. It stores real-time monitoring data, historical data, fault records, compensation parameters, etc., of the electric spindle. The SSD storage capacity is no less than 128GB. The cloud storage is connected via Ethernet or 5G network to achieve remote data backup and sharing, supporting historical data query, trend analysis, and data export, providing data support for predictive maintenance and performance optimization of the electric spindle. The data storage frequency can be specified between 0.1 and 5 seconds, with a minimum of 0.1 seconds for collecting and storing the peak monitoring parameters within that 0.1-second time interval. The data is recorded and stored cyclically for more than 30 days. It supports data log export and spindle diagnostic report export (PDF format), which includes current configuration information, spindle continuous running time statistics, running statistics, vibration statistics, and collision events.

2. The multi-dimensional condition monitoring system for an electric spindle integrating thermal displacement and temperature rise compensation according to claim 1, characterized in that: It also includes a power supply module to provide a stable operating power for each module. The power supply module adopts a switching power supply with an input voltage of AC220V and an output voltage of DC5V, DC12V, and DC24V. It has overvoltage, overcurrent, and short circuit protection functions to ensure stable system operation. The multi-functional data acquisition unit can be directly connected to a 24V power supply to meet the power supply needs of industrial sites.

3. The multi-dimensional condition monitoring system for an electric spindle integrating thermal displacement and temperature rise compensation according to claim 1, characterized in that: It supports two installation modes to adapt to different machine tool scenarios: New machine tool integration: The spindle is equipped with a digital vibration sensor, the smart box (integrating signal conditioning, data acquisition, and central processing functions) is connected to a 24V power supply and the CNC controller IO terminal, and the tablet is wirelessly connected to the smart box to check the machine tool vibration and set the maximum vibration limit of the machine tool; Old machine tool retrofit: The vibration sensor is attached to the nose of the electric spindle with a magnetic base accessory, or a hole is drilled for locking, the smart box is connected to a 24V power supply and the CNC controller IO terminal, and the tablet is wirelessly connected to the smart box to complete parameter setting and debugging.