Main shaft detection method and device based on array eddy current, voltage and current

Abstract

The invention belongs to the technical field of industrial detection, and particularly relates to a main shaft detection method and device based on array eddy current, voltage and current. The method comprises the following steps: S1, acquiring a voltage signal, a current signal and an eddy current signal of a main shaft in an operation process; s2, determining a voltage unbalance degree, a current harmonic distortion rate and a power factor according to the voltage signal and the current signal of the main shaft, and calculating a signal amplitude change rate and a phase offset according to the eddy current signal; and S3, the voltage unbalance degree, the current harmonic distortion rate, the power factor, the signal amplitude change rate and the phase offset form feature vectors, fault types are determined based on a pre-trained intelligent diagnosis model, and the fault types comprise surface cracks, holes, abrasion, electrical short circuit, electrical open circuit and electrical overload of different severity degrees. According to the invention, omnibearing, automatic and high-precision detection of the main shaft is realized, and the detection efficiency and accuracy are improved.

Description

Technical Field

[0001] The present application belongs to the field of industrial detection technology, and in particular relates to a spindle detection method and device based on array eddy current, voltage, and current. Background Art

[0002] The spindle is a key part of mechanical equipment, and its quality and reliability directly affect the performance and safe operation of the entire equipment. During the production and manufacturing process of the spindle and after long-term use, defects such as cracks, sand holes, and wear may appear on the surface and near the surface, and its quality directly affects the processing accuracy, stability, and service life of the equipment.

[0003] Traditional spindle inspection methods mainly use a single inspection method. For example, simple eddy current inspection can only detect defects on the spindle surface and near the surface, and cannot fully reflect the overall condition of the spindle; while conventional electrical inspection only focuses on the basic parameters of voltage and current, making it difficult to deeply analyze potential electrical faults. In addition, most of these inspection methods rely on manual experience for judgment, with low inspection efficiency and difficult to ensure accuracy. They are difficult to meet the needs of modern industrial production for high-precision and high-reliability spindle inspections, and cannot adapt to the inspection requirements of different types of spindles and complex working conditions. Summary of the invention

[0004] In order to solve the above problems, the present application provides a spindle detection method and device based on array eddy current, voltage and current to provide reliable spindle quality assurance for industrial production.

[0005] The first aspect of the present application provides a spindle detection method based on array eddy current, voltage and current, mainly comprising: Step S1, obtaining a voltage signal, a current signal and an eddy current signal of the spindle during operation; Step S2, determining the voltage imbalance, current harmonic distortion rate and power factor according to the voltage signal and current signal of the main shaft, and calculating the signal amplitude change rate and phase offset according to the eddy current signal; Step S3, the voltage imbalance, current harmonic distortion rate, power factor, signal amplitude change rate and phase offset are combined into a feature vector, and the fault type is determined based on a pre-trained intelligent diagnosis model. The fault type includes surface cracks, holes, wear, electrical short circuit, electrical open circuit and electrical overload of different severity.

[0006] Preferably, in step S1, a voltage transformer is coupled to the electrical connection part of the spindle to collect the voltage signal of the spindle during operation; a current transformer is connected in series to the power supply circuit of the spindle to measure the current signal of the spindle during operation; an array structure composed of a plurality of eddy current sensors arranged around the spindle is moved along the surface of the spindle to collect the eddy current signal of the spindle during operation.

[0007] Preferably, step S1 further comprises: Adopt adaptive filtering algorithm to reduce noise of voltage and current signals; The wavelet transform algorithm is used to reduce the noise of the collected eddy current signal.

[0008] Preferably, step S3 further comprises: Step S31, obtaining attribute parameters and / or operation parameters of the spindle; Step S32: Calling a corresponding intelligent diagnosis model to predict the fault type according to the attribute parameters and / or operation parameters of the spindle.

[0009] Preferably, after step S3, the method further comprises: Step S4: storing the confirmed fault prediction results and the corresponding model input information in a database as new sample data for the intelligent diagnosis model to learn.

[0010] The second aspect of the present application provides a spindle detection device based on array eddy current, voltage and current, mainly comprising: A signal acquisition module is used to obtain the voltage signal, current signal and eddy current signal of the spindle during operation; A data preprocessing module is used to determine the voltage imbalance, current harmonic distortion rate and power factor according to the voltage signal and current signal of the main shaft, and calculate the signal amplitude change rate and phase offset according to the eddy current signal; The fault type prediction module is used to form a feature vector consisting of voltage imbalance, current harmonic distortion rate, power factor, signal amplitude change rate and phase offset, and determine the fault type based on a pre-trained intelligent diagnosis model. The fault types include surface cracks of varying severity, holes, wear, electrical short circuit, electrical open circuit and electrical overload.

[0011] Preferably, in the signal acquisition module, a voltage transformer is coupled with the electrical connection part of the spindle to collect the voltage signal of the spindle during operation; a current transformer is connected in series to the power supply circuit of the spindle to measure the current signal of the spindle during operation; an array structure composed of multiple eddy current sensors arranged around the spindle moves along the surface of the spindle to collect the eddy current signal of the spindle during operation.

[0012] Preferably, the signal acquisition module includes: A voltage and current signal noise reduction unit, used to perform noise reduction processing on voltage and current signals using an adaptive filtering algorithm; The eddy current signal noise reduction unit is used to perform noise reduction processing on the collected eddy current signal by adopting a wavelet transform algorithm.

[0013] Preferably, the fault type prediction module includes: A spindle parameter acquisition unit, used to acquire attribute parameters and / or operation parameters of the spindle; The intelligent diagnosis model selection unit is used to call the corresponding intelligent diagnosis model to predict the fault type according to the attribute parameters and / or operation parameters of the spindle.

[0014] Preferably, the spindle detection device further comprises: The model optimization module is used to store the confirmed fault prediction results and the corresponding model input information in the database as new sample data for the intelligent diagnosis model to learn.

[0015] This application realizes all-round, automated, and high-precision detection of the spindle, thereby improving detection efficiency and accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] Figure 1 It is a flow chart of a preferred embodiment of the main shaft detection method based on array eddy current, voltage and current of the present application.

[0017] Figure 2 It is a schematic diagram of the spindle detection mechanism. DETAILED DESCRIPTION

[0018] In order to make the purpose, technical scheme and advantages of the implementation of this application clearer, the technical scheme in the implementation of this application will be described in more detail in combination with the drawings in the implementation of this application. In the drawings, the same or similar reference numerals throughout represent the same or similar elements or elements with the same or similar functions. The described implementation is a part of the implementation of this application, not all of the implementations. The implementation described below with reference to the drawings is exemplary and is intended to be used to explain this application, and cannot be understood as a limitation on this application. Based on the implementation in this application, all other implementations obtained by ordinary technicians in this field without making creative work are within the scope of protection of this application. The implementation of this application is described in detail below in combination with the drawings.

[0019] The first aspect of the present application provides a spindle detection method based on array eddy current, voltage, and current, such as Figure 1 As shown, it mainly includes: Step S1, obtaining a voltage signal, a current signal and an eddy current signal of the spindle during operation; Step S2, determining the voltage imbalance, current harmonic distortion rate and power factor according to the voltage signal and current signal of the main shaft, and calculating the signal amplitude change rate and phase offset according to the eddy current signal; Step S3, the voltage imbalance, current harmonic distortion rate, power factor, signal amplitude change rate and phase offset are combined into a feature vector, and the fault type is determined based on a pre-trained intelligent diagnosis model. The fault type includes surface cracks, holes, wear, electrical short circuit, electrical open circuit and electrical overload of different severity.

[0020] The present application collects voltage signals, current signals and eddy current signals through step S1, and then pre-processes these signals in step S2 to form parameters that can comprehensively reflect the state of the spindle. Finally, in step S3, these parameters are combined into feature vectors and input into the pre-trained intelligent diagnosis model, and the fault type of the spindle is output through the intelligent diagnosis model. Among them, the intelligent diagnosis model is constructed based on machine learning algorithms, including but not limited to neural networks, support vector machines and other algorithms. The model is continuously optimized through training. The training data includes a large number of spindle sample data with known defects and normal states. By learning from the sample data, the model can automatically identify different types of defects and fault modes.

[0021] According to the description of this application, the collected current signals, voltage signals and eddy current signals usually have time characteristics, that is, the spindle has current signals, voltage signals and eddy current signals at every moment during its operation. For this reason, the intelligent diagnosis model here can usually be a composite network that combines a convolutional neural network with a long short-term memory network, so as to achieve the purpose of rapid and accurate prediction of defects.

[0022] In some optional embodiments, in step S1, a voltage transformer is coupled to the electrical connection part of the spindle to collect the voltage signal of the spindle during operation; a current transformer is connected in series to the power supply circuit of the spindle to measure the current signal of the spindle during operation; an array structure composed of multiple eddy current sensors arranged around the spindle is moved along the surface of the spindle to collect the eddy current signal of the spindle during operation.

[0023] refer to Figure 2In this embodiment, a high-precision voltage transformer is used to be tightly coupled with the electrical connection part of the spindle, and the voltage signal of the spindle during operation is collected in real time. It can accurately measure the voltage changes of different phase sequences and amplitudes, and effectively capture abnormal conditions such as voltage fluctuations, phase loss, and overvoltage. At the same time, the current transformer is connected in series to the power supply circuit of the spindle to accurately measure the working current of the spindle. The effective value, peak value, harmonic components and other parameters of the current can be accurately detected, and hidden faults such as current overload, short circuit, and leakage can be discovered in time. On the other hand, multiple high-sensitivity eddy current sensors are arranged around the key parts of the spindle to form a specific array structure to ensure that the surface area of ​​the spindle can be fully covered. The selection and layout of the sensors are optimized according to the size, shape and expected detection accuracy of the spindle to maximize the sensitivity and reliability of defect detection.

[0024] When collecting voltage and current signals, while the spindle is rotating, the voltage transformer and current transformer convert high voltage and high current signals into low voltage signals that can be collected, and then amplify and filter them through the signal conditioning circuit before sending them to the signal collection and processing system. The signal collection and processing system monitors and analyzes the collected voltage and current signals in real time, and calculates electrical parameters such as voltage RMS, current RMS, and power factor. Among them, for current signals, a current transformer combined with a sampling resistor is used for collection. The current transformer converts large current into a small current signal, and then converts it into a voltage signal through a sampling resistor for collection. At the same time, an overcurrent protection circuit is added to prevent excessive current from damaging the collection unit.

[0025] When collecting eddy current signals, the control unit (PLC) automatically adjusts the spacing and initial position of each sensor of the array eddy current detection according to the input parameters, so that it is in the best starting state for detection. At the same time, the rotation drive mechanism of the detection bench is started to make the spindle rotate slowly and evenly at a predetermined speed (such as 500-1500rpm, set according to the spindle type and detection requirements) to prepare for subsequent detection. Each sensor of the array eddy current detection is driven by a linear motor to move along the axial and circumferential directions of the spindle according to the preset trajectory, and conduct a comprehensive scanning and detection of the spindle surface and near surface. The sensor collects eddy current signals in real time and transmits them to the signal acquisition and processing system. During the detection process, the movement speed and sampling frequency of the sensor are dynamically adjusted according to the detection accuracy requirements. For example, for high-precision detection requirements, the sensor movement speed can be reduced to 1-5mm / s, and the sampling frequency can be increased to more than 10kHz to ensure that tiny defect signals can be accurately captured.

[0026] In some optional implementations, step S1 further includes: using an adaptive filtering algorithm to perform noise reduction processing on the voltage signal and the current signal; and using a wavelet transform algorithm to perform noise reduction processing on the collected eddy current signal.

[0027] It is understandable that the collected voltage signal, current signal and eddy current signal all need to be processed for noise reduction to remove noise interference in the signal, improve the signal-to-noise ratio of the signal, and ensure the accuracy of subsequent analysis. Among them, for the eddy current signal, by selecting the appropriate wavelet basis function and the number of decomposition layers, the signal is decomposed into sub-bands of different frequencies, and after removing the high-frequency noise component, the signal is reconstructed to obtain the denoised eddy current signal.

[0028] With the above-mentioned voltage signal, current signal and eddy current signal, deep data mining can be performed in step S2, and characteristic parameters reflecting electrical performance, such as voltage imbalance, current harmonic distortion rate, power factor, etc., can be extracted from the voltage and current signals. Characteristic parameters related to the surface defects of the spindle can be extracted from the eddy current signal, such as calculating the rate of change of signal amplitude, phase offset and other characteristic parameters, and the frequency spectrum characteristics of the signal can be calculated using fast Fourier transform (FFT), which can also be used as characteristic parameters.

[0029] Finally, in step S3, the above characteristic parameters are combined to form a characteristic vector. In an alternative implementation, each characteristic parameter needs to be normalized within a specified range. According to the output results of the intelligent diagnosis model, it is accurately determined whether the spindle has surface defects, electrical faults, and the type and severity of the faults. The fault type can be, for example, surface cracks, holes, insulation aging, or winding short circuits, and the severity of the fault can be, for example, mild, moderate, or severe.

[0030] In some optional implementations, step S3 further includes: Step S31, obtaining attribute parameters and / or operation parameters of the spindle; Step S32: Calling a corresponding intelligent diagnosis model to predict the fault type according to the attribute parameters and / or operation parameters of the spindle.

[0031] This embodiment decides to use different intelligent diagnostic models according to the attribute parameters and / or operating parameters of the spindle. Of course, different intelligent diagnostic models use different training data and thus have different weight parameters and decision boundaries. The attribute parameters of the spindle may be, for example, the material and size of the spindle, and the operating parameters of the spindle may be, for example, the rated voltage, rated current, speed range, etc. of the spindle.

[0032] In some optional implementations, after step S3, the method further includes: Step S4: storing the confirmed fault prediction results and the corresponding model input information in a database as new sample data for the intelligent diagnosis model to learn.

[0033] In this embodiment, the intelligent diagnosis model is optimized through continuous iteration to adapt to the changes in detection requirements of different spindles and working conditions, thereby improving the overall accuracy and adaptability of the detection method.

[0034] In an alternative implementation, the specific position of the defect or fault on the spindle can be located based on the diagnostic results and combined with the position information of the sensor, and the inspection results of the spindle can be intuitively displayed in a graphical manner on the display screen of the host computer. At the same time, a detailed inspection report can be generated, including various inspection data, diagnostic conclusions, performance evaluation indicators, and maintenance suggestions.

[0035] The second aspect of the present application provides a spindle detection device based on array eddy current, voltage and current corresponding to the above method, mainly comprising: A signal acquisition module is used to obtain the voltage signal, current signal and eddy current signal of the spindle during operation; A data preprocessing module is used to determine the voltage imbalance, current harmonic distortion rate and power factor according to the voltage signal and current signal of the main shaft, and calculate the signal amplitude change rate and phase offset according to the eddy current signal; The fault type prediction module is used to form a feature vector consisting of voltage imbalance, current harmonic distortion rate, power factor, signal amplitude change rate and phase offset, and determine the fault type based on a pre-trained intelligent diagnosis model. The fault types include surface cracks of varying severity, holes, wear, electrical short circuit, electrical open circuit and electrical overload.

[0036] In some optional embodiments, in the signal acquisition module, a voltage transformer is coupled to the electrical connection part of the spindle to collect the voltage signal of the spindle during operation; a current transformer is connected in series to the power supply circuit of the spindle to measure the current signal of the spindle during operation; an array structure composed of multiple eddy current sensors arranged around the spindle is moved along the surface of the spindle to collect the eddy current signal of the spindle during operation.

[0037] In some optional implementations, the signal acquisition module includes: A voltage and current signal noise reduction unit, used to perform noise reduction processing on voltage and current signals using an adaptive filtering algorithm; The eddy current signal noise reduction unit is used to perform noise reduction processing on the collected eddy current signal by adopting a wavelet transform algorithm.

[0038] In some optional implementations, the fault type prediction module includes: A spindle parameter acquisition unit, used to acquire attribute parameters and / or operation parameters of the spindle; The intelligent diagnosis model selection unit is used to call the corresponding intelligent diagnosis model to predict the fault type according to the attribute parameters and / or operation parameters of the spindle.

[0039] In some optional implementations, the spindle detection device further includes: The model optimization module is used to store the confirmed fault prediction results and the corresponding model input information in the database as new sample data for the intelligent diagnosis model to learn.

[0040] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application shall be based on the protection scope of the claims.

Claims

1. A spindle detection method based on array eddy current, voltage and current, characterized in that: include: Step S1, obtaining a voltage signal, a current signal and an eddy current signal of the spindle during operation; Step S2, determining the voltage imbalance, current harmonic distortion rate and power factor according to the voltage signal and current signal of the main shaft, and calculating the signal amplitude change rate and phase offset according to the eddy current signal; Step S3, the voltage imbalance, current harmonic distortion rate, power factor, signal amplitude change rate and phase offset are combined into a feature vector, and the fault type is determined based on a pre-trained intelligent diagnosis model. The fault type includes surface cracks, holes, wear, electrical short circuit, electrical open circuit and electrical overload of different severity.

2. The spindle detection method based on array eddy current, voltage and current as claimed in claim 1, characterized in that: In step S1, a voltage transformer is coupled to the electrical connection part of the spindle to collect the voltage signal of the spindle during operation; a current transformer is connected in series to the power supply circuit of the spindle to measure the current signal of the spindle during operation; An array structure consisting of a plurality of eddy current sensors arranged around the main shaft moves along the main shaft surface to collect eddy current signals of the main shaft during operation.

3. The spindle detection method based on array eddy current, voltage and current as claimed in claim 1, characterized in that: Step S1 further comprises: Adopt adaptive filtering algorithm to reduce noise of voltage and current signals; The wavelet transform algorithm is used to reduce the noise of the collected eddy current signal.

4. The spindle detection method based on array eddy current, voltage and current as claimed in claim 1, characterized in that: Step S3 further comprises: Step S31, obtaining attribute parameters and / or operation parameters of the spindle; Step S32: Calling a corresponding intelligent diagnosis model to predict the fault type according to the attribute parameters and / or operation parameters of the spindle.

5. The spindle detection method based on array eddy current, voltage and current as claimed in claim 1, characterized in that: After step S3, the method further comprises: Step S4: storing the confirmed fault prediction results and the corresponding model input information in a database as new sample data for the intelligent diagnosis model to learn.

6. A spindle detection device based on array eddy current, voltage and current, characterized in that: include: A signal acquisition module is used to obtain the voltage signal, current signal and eddy current signal of the spindle during operation; A data preprocessing module is used to determine the voltage imbalance, current harmonic distortion rate and power factor according to the voltage signal and current signal of the main shaft, and calculate the signal amplitude change rate and phase offset according to the eddy current signal; The fault type prediction module is used to form a feature vector consisting of voltage imbalance, current harmonic distortion rate, power factor, signal amplitude change rate and phase offset, and determine the fault type based on a pre-trained intelligent diagnosis model. The fault types include surface cracks of varying severity, holes, wear, electrical short circuit, electrical open circuit and electrical overload.

7. The spindle detection device based on array eddy current, voltage and current as claimed in claim 6, characterized in that: In the signal acquisition module, a voltage transformer is coupled with the electrical connection part of the spindle to collect the voltage signal of the spindle during operation; a current transformer is connected in series to the power supply circuit of the spindle to measure the current signal of the spindle during operation; An array structure consisting of a plurality of eddy current sensors arranged around the main shaft moves along the main shaft surface to collect eddy current signals of the main shaft during operation.

8. The spindle detection device based on array eddy current, voltage and current as claimed in claim 6, characterized in that: The signal acquisition module comprises: A voltage and current signal noise reduction unit, used to perform noise reduction processing on voltage and current signals using an adaptive filtering algorithm; The eddy current signal noise reduction unit is used to perform noise reduction processing on the collected eddy current signal by adopting a wavelet transform algorithm.

9. The spindle detection device based on array eddy current, voltage and current as claimed in claim 6, characterized in that: The fault type prediction module includes: A spindle parameter acquisition unit, used to acquire attribute parameters and / or operation parameters of the spindle; The intelligent diagnosis model selection unit is used to call the corresponding intelligent diagnosis model to predict the fault type according to the attribute parameters and / or operation parameters of the spindle.

10. The spindle detection device based on array eddy current, voltage and current as claimed in claim 6, characterized in that: The spindle detection device also includes: The model optimization module is used to store the confirmed fault prediction results and the corresponding model input information in the database as new sample data for the intelligent diagnosis model to learn.

Images