Polishing robot millstone state intelligent monitoring system based on piezoelectric sensing

Abstract

The invention relates to the technical field of mechanical equipment state monitoring and fault diagnosis, in particular to a grinding robot millstone state intelligent monitoring system based on piezoelectric sensing, and the system comprises a signal collection module which is used for collecting a millstone vibration signal; the signal processing and feature extraction module is used for preprocessing the vibration signal and extracting the extraction time domain feature and the frequency domain feature of the processed vibration signal; and the intelligent diagnosis module is used for calculating a self-adaptive two-parameter fusion feature value P based on the time domain feature and the frequency domain feature, and judging the state of the grinding disc based on the two-parameter fusion feature value P. According to the method, the fusion feature value P is introduced, two parameters are intelligently fused into a single index, nonlinear weighting and an adaptive enhancement mechanism are adopted, instantaneous interference and real faults are effectively distinguished, and the false alarm rate is greatly reduced. The system complexity is reduced, and the decision reliability is improved.

Description

Technical Field

[0001] This invention relates to the field of mechanical equipment condition monitoring and fault diagnosis technology, specifically to a system for real-time online monitoring of the condition of a grinding disc in a rotary grinding machine, and particularly to an intelligent monitoring system for the condition of a grinding disc in a grinding robot based on piezoelectric sensing. Background Technology

[0002] The grinding disc is the direct actuator of the grinding equipment, and its condition directly affects the processing quality, efficiency, and equipment safety. Currently, monitoring the condition of the grinding disc mainly relies on manual experience, such as observing sparks, listening to sounds, or periodic shutdowns for inspection, which has the following prominent drawbacks: Highly subjective and delayed: It relies on operator experience, cannot be quantified, and by the time an anomaly is discovered, the workpiece has often already been damaged or the equipment has failed.

[0003] Inefficiency: Shutdowns for inspections interrupt production, affecting overall efficiency.

[0004] Unable to detect internal damage: Cracks, dynamic imbalances, and other hidden dangers inside the grinding disc are difficult to detect manually.

[0005] While some studies have attempted to indirectly determine the load by monitoring motor current, the current signal is susceptible to power grid fluctuations and interference from other moving parts of the equipment, and is insensitive to early wear and localized damage to the grinding disc, resulting in a high false alarm rate. Therefore, there is an urgent need for an intelligent monitoring system that can directly, accurately, and in real-time sense the health status of the grinding disc itself. Summary of the Invention

[0006] The present invention aims to solve one of the problems existing in the background art.

[0007] To address this, the present invention provides an intelligent monitoring system for the state of a grinding robot grinding disc based on piezoelectric sensing. By intelligently weighting and fusing the effective vibration value (RMS) and kurtosis value, a fused feature value P is generated, enabling a single indicator and highly reliable diagnosis of the grinding disc's health status, significantly simplifying the diagnostic logic and improving accuracy.

[0008] The technical solution adopted by this invention to solve its technical problem is: A smart monitoring system for the grinding wheel status of a grinding robot based on piezoelectric sensing includes, A signal acquisition module, which is used to acquire the vibration signal of the grinding disc; The signal processing and feature extraction module is used to preprocess the vibration signal and extract the time-domain and frequency-domain features of the processed vibration signal. The intelligent diagnostic module calculates an adaptive dual-parameter fusion feature value P based on time-domain and frequency-domain features, and judges the state of the grinding disc based on the dual-parameter fusion feature value P.

[0009] Furthermore, the signal processing and feature extraction module includes a hardware conditioning circuit, which includes a charge amplifier and a high-pass and low-pass filter, used to amplify and initially reduce the noise of the weak charge signal output by the sensor.

[0010] Furthermore, the time-domain features extracted by the signal processing and feature extraction module include the effective value (RMS) of vibration velocity and the kurtosis index. The effective value of vibration velocity is used to measure the total vibration energy, and the kurtosis index is used to evaluate the impact signal.

[0011] Furthermore, the signal processing and feature extraction module uses Fast Fourier Transform to convert the vibration signal from the time domain to the frequency domain to obtain a spectrum, which is used to identify specific fault frequencies.

[0012] Furthermore, the two-parameter fused feature value P = (RMS / RMS) base ) × (Kurtosis / Kurtosis base ) 0.6 × [1 + 0.2 × max(0, Kurtosis - Kurtosis base )], where RMS base As the reference vibration effective value, Kurtosis base This is the baseline kurtosis value.

[0013] Furthermore, the dual-parameter fusion feature value P is less than 1.0, indicating normal status; 1.0 ≤ P < 1.5, indicating slight wear; 1.5 ≤ P < 2.2, indicating severe wear; and P ≥ 2.2, indicating local damage or serious malfunction.

[0014] Furthermore, for the case where the dual-parameter fusion feature value P ≥ 1.5, it needs to be collected continuously for 3 sampling cycles before an alarm is officially triggered.

[0015] Furthermore, if the amplitude at the spindle rotation frequency in the spectrum exceeds 30% of the total energy, it is determined that there is a problem with the dynamic balance of the grinding wheel.

[0016] Furthermore, it also includes an output and execution module, wherein the output and execution module is used to display the diagnostic results of the intelligent diagnostic module.

[0017] Furthermore, the output and execution module includes a human-machine interface, a communication interface, and an alarm unit. The communication interface is connected to the intelligent diagnostic module, and the communication interface is also connected to the human-machine interface and the alarm unit.

[0018] The beneficial effects of this invention are: Simplified diagnostic logic and improved reliability: By introducing a fusion feature value P, two parameters are intelligently fused into a single indicator, reducing system complexity and improving decision reliability.

[0019] Precise wear quantification and early warning: The RMS term in the P-value formula directly quantifies the degree of wear, enabling precise early warning.

[0020] Adaptive anti-interference capability: The nonlinear weighting and adaptive enhancement mechanism of the P-value formula can effectively distinguish between instantaneous interference and real faults, and significantly reduce the false alarm rate.

[0021] Highly practical for engineering applications: The single indicator P is easier for on-site engineers to understand and operate, which is conducive to the promotion of technology.

[0022] Modular and versatile: The system is independent of the main equipment control and can be quickly installed on various rotary grinding equipment, making it widely applicable. Attached Figure Description

[0023] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0024] Figure 1 This is a schematic diagram of the intelligent monitoring system for the grinding wheel status of a grinding robot based on piezoelectric sensing in this invention.

[0025] Figure 2 This is a schematic diagram illustrating the working principle of the intelligent diagnostic module in this invention.

[0026] Figure 3 This is a schematic diagram of the kurtosis sensitivity analysis in this invention. Detailed Implementation

[0027] The present invention will now be described in further detail with reference to the accompanying drawings. These drawings are simplified schematic diagrams, illustrating only the basic structure of the invention, and therefore only show the components relevant to the invention.

[0028] In the description of this invention, it should be understood that the terms "center," "longitudinal," "lateral," "length," "width," "thickness," "upper," "lower," "front," "rear," "left," "right," "vertical," "horizontal," "top," "bottom," "inner," "outer," "clockwise," "counterclockwise," "axial," "radial," and "circumferential," etc., indicating orientation or positional relationships, are based on the orientation or positional relationships shown in the accompanying drawings and are only for the convenience of describing the invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation, and therefore should not be construed as a limitation of the invention. Furthermore, features defined with "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, unless otherwise stated, "a plurality of" means two or more.

[0029] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0030] A smart monitoring system for the state of a grinding robot grinding disc based on piezoelectric sensing includes a signal acquisition module, a signal processing and feature extraction module, an intelligent diagnostic module, and an output and execution module.

[0031] The signal acquisition unit includes multiple piezoelectric sensors mounted on the grinding disc. These piezoelectric sensors act as the "sensory nerves" of the system, with wideband piezoelectric accelerometers being preferred. Their key installation location is the spindle bearing housing or drive motor housing of the grinding equipment, connected rigidly via magnetic bases or threads to ensure low-loss transmission of high-frequency vibration signals.

[0032] The signal acquisition unit transmits the acquired vibration signal to the signal processing and feature extraction module, which processes the vibration signal.

[0033] The signal processing and feature extraction module includes a hardware conditioning circuit and a digital signal processor. The hardware conditioning circuit includes a charge amplifier and high-pass and low-pass filters, used to amplify and initially reduce the noise of the weak charge signal output by the sensor. The digital signal processor further processes the acquired digital signal and extracts the time-domain and frequency-domain features of the vibration signal. Time-domain features include the effective value (RMS) of vibration velocity, used to measure the total vibration energy; and kurtosis, which is extremely sensitive to impact signals. Then, the vibration signal is converted from the time domain to the frequency domain using a fast Fourier transform to obtain a spectrum, used to identify specific fault frequencies (such as rotational frequency, bearing fault frequency).

[0034] Based on the results obtained from the signal processing and feature extraction modules, the intelligent diagnostic module uses an adaptive two-parameter fusion algorithm to determine the real-time vibration state of the grinding disc in order to analyze its condition. The intelligent diagnostic module transmits the judgment results to the output and execution module, which includes a human-machine interface, a communication interface, and an alarm unit. The communication interface receives data from the intelligent diagnostic module and transmits the data to the human-machine interface, the alarm unit, the cloud, and other equipment control systems. The alarm unit triggers different levels of indication signals based on the diagnostic results.

[0035] In the intelligent diagnosis module, kurtosis and RMS are fused into a fused feature value P, where P = (RMS / RMS). base )× (Kurtosis / Kurtosis base ) k × [1 + 0.2 × max(0, Kurtosis - Kurtosis base )).

[0036] Among them, RMS base As the reference effective value of vibration (preferably 5.0 g), Kurtosis base The baseline kurtosis value is 4.0 (preferably). This formula achieves an intelligent balance between wear degree and impact characteristics through nonlinear weighting and adaptive enhancement mechanisms.

[0037] The exponent k controls the nonlinearity of the kurtosis effect. Sensitivity analysis shows that when the exponent k is 0.6, it effectively suppresses the influence of normal fluctuations while maintaining sensitivity to anomalous shocks. Sensitivity analysis of the exponent value (k) confirms that... Figure 3As shown, k=0.6 is key to achieving the optimal balance between diagnostic sensitivity and anti-interference capability. When k=0.4, the system is insensitive to shocks, posing a risk of missed detections; while when k≥0.8, the system is overly sensitive to transient interference, leading to an increased false alarm rate. This invention preferably uses k=0.6, ensuring that the fused feature value P can significantly respond to sustained high kurtosis values ​​characterizing local damage while effectively suppressing fluctuations caused by transient interference, thus minimizing the false alarm rate while maintaining a high detection rate. This value is the optimized solution for achieving high-reliability diagnosis in this scheme.

[0038] Figure 3 In the data, the kurtosis sensitivity analysis curve visually confirms the optimality of k=0.6. The curve shows that when k=0.6, the trajectory of P-value as the kurtosis increases is most ideal: it can quickly rise to the alarm threshold when the kurtosis value remains high (indicating local damage), and it can remain stable when the kurtosis value rises briefly (indicating instantaneous interference). This effectively avoids the false alarm tendency of the k=0.4 curve and the false alarm tendency of the k≥0.8 curve, thus graphically confirming the key role of this index value in diagnostic reliability.

[0039] To verify the diagnostic efficacy of the fused feature value P, Table 1 shows the statistical characteristics of RMS and kurtosis values ​​and the corresponding P-value calculation results under three typical working conditions.

[0040] Table 1 Diagnostic results under various operating conditions

[0041] Finally, a confusion matrix is ​​used to quantify the system performance improvement after introducing complementarity judgment. Table 2 compares the effects of two diagnostic strategies: "using only kurtosis values" and "integrating feature values ​​P".

[0042] Table 2 Comparison of Results from Different Diagnostic Strategies

[0043] After installing this system on an angle grinder used for grinding metal welds: Initial stage: The RMS value is stable at 3.5g, the kurtosis value is 3.0, and the calculated P=0.61. The system judges it as "normal".

[0044] After 2 hours of continuous operation: the RMS value slowly rises to 5.5g, the kurtosis value remains at 3.0, and the calculated P=0.94. The system issues a yellow warning for "slight wear".

[0045] After three hours of operation: the RMS value rose to 7.0g, and the kurtosis value soared to 8.5, resulting in a calculated P=4.20. The system immediately triggered a "local damage" red alarm. On-site inspection confirmed that there was a chipped edge on the grinding disc, and the system accurately identified the superimposed state of wear and damage.

[0046] Based on the above experimental data and comparative analysis, the following conclusions can be drawn: Verification of the diagnostic effectiveness of P-value: The fused feature value P can accurately distinguish different working conditions and has a natural resistance to instantaneous interference.

[0047] The false alarm rate is significantly reduced: the P-value fusion strategy reduces the false alarm rate of the system to "transient interference" to below 5%.

[0048] The diagnostic dimensions are comprehensive: a single indicator P can simultaneously identify two key failure modes: wear level and localized damage.

[0049] Example 1 In this embodiment, the intelligent monitoring system is installed as an independent module on a general-purpose industrial angle grinder.

[0050] Step 1: Sensor Installation. Securely attach the piezoelectric accelerometer to the surface of the angle grinder gearbox housing where vibration is most directly transmitted, using its dedicated magnetic mount. Ensure the mounting surface is clean and flat to guarantee efficient transmission of high-frequency vibration signals to the sensor.

[0051] Step 2: Connecting the Core Processor. Connect the signal line output from the sensor to the signal conditioning circuit board. This circuit board is responsible for powering the sensor and amplifying and filtering the raw signal. The conditioned signal is then sent to the high-precision analog-to-digital converter inside the microprocessor for sampling.

[0052] Step 3: Power Supply and Communication Setup. A rechargeable lithium battery powers the entire monitoring module. Simultaneously, the microprocessor is connected to the wireless communication module to send monitoring results to the host computer display interface.

[0053] Step 4: System Power-On Self-Test. After completing the connection, power on the system. The system automatically performs initialization operations such as sensor path detection and memory self-test. The indicator light flashes to indicate that the self-test has passed and the system has entered standby mode.

[0054] Step 5: Two-parameter adaptive fusion diagnostic rules The system calculates the fused feature value P and determines the state according to the following threshold rules: 1. Normal state determination Monitoring parameter: Fusion feature value P Judgment condition: P < 1.0 Output: The status flag is "Normal", and the green indicator light remains on. 2. Wear condition classification judgment Monitoring parameter: Fusion feature value P Judgment conditions: If 1.0 ≤ P < 1.5, it is judged as "slight wear"; If 1.5 ≤ P < 2.2, it is judged as "severe wear"; If P ≥ 2.2, it is judged as "partial damage or serious failure".

[0055] Output: Triggers the corresponding level of yellow, orange, or red alarm.

[0056] 3. Anti-interference mechanism To prevent false alarms, an abnormal state (P ≥ 1.5) must last for 3 sampling cycles before a formal alarm is triggered; instantaneous fluctuations are only recorded.

[0057] 4. Determination of dynamic imbalance Monitoring parameters: Vibration signal spectral characteristics Judgment criterion: The amplitude at the principal axis rotation frequency in the spectrum exceeds 30% of the total energy. Output result: Special alarm prompt.

[0058] In summary, this application introduces a fusion feature value P, intelligently fusing two parameters into a single index, reducing system complexity and improving decision reliability. The RMS term in the P-value formula directly quantifies the degree of wear, achieving accurate early warning. The nonlinear weighting and adaptive enhancement mechanism of the P-value formula effectively distinguishes between transient interference and real faults, significantly reducing the false alarm rate. The single index P is easier for field engineers to understand and operate, facilitating technology promotion. The system is independent of the equipment's main control unit and can be quickly installed on various rotary grinding equipment, making it widely applicable.

[0059] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined by the scope of the claims.

Claims

1. A smart monitoring system for the state of a grinding robot's grinding disc based on piezoelectric sensing, characterized in that, include, A signal acquisition module, which is used to acquire the vibration signal of the grinding disc; The signal processing and feature extraction module is used to preprocess the vibration signal and extract the time-domain and frequency-domain features of the processed vibration signal. The intelligent diagnostic module calculates an adaptive dual-parameter fusion feature value P based on time-domain and frequency-domain features, and judges the state of the grinding disc based on the dual-parameter fusion feature value P.

2. The intelligent monitoring system for the grinding wheel status of a grinding robot based on piezoelectric sensing according to claim 1, characterized in that, The signal processing and feature extraction module includes a hardware conditioning circuit, which includes a charge amplifier and a high-pass and low-pass filter, used to amplify and initially reduce the noise of the weak charge signal output by the sensor.

3. The intelligent monitoring system for the grinding wheel status of a grinding robot based on piezoelectric sensing according to claim 2, characterized in that, The time-domain features extracted by the signal processing and feature extraction module include the effective value (RMS) of vibration velocity and the kurtosis index. The effective value of vibration velocity is used to measure the total vibration energy, and the kurtosis index is used to evaluate the impact signal.

4. The intelligent monitoring system for the grinding wheel status of a grinding robot based on piezoelectric sensing according to claim 3, characterized in that, The signal processing and feature extraction module uses Fast Fourier Transform to convert the vibration signal from the time domain to the frequency domain to obtain a spectrum, which is used to identify specific fault frequencies.

5. The intelligent monitoring system for the grinding wheel status of a grinding robot based on piezoelectric sensing according to claim 1, characterized in that, The dual-parameter fusion feature value P = (RMS / RMS) base )×(Kurtosis / Kurtosis base ) 0.6 ×[1 + 0.2×max(0, Kurtosis-Kurtosis base )], where RMS base As the reference vibration effective value, Kurtosis base This is the baseline kurtosis value.

6. The intelligent monitoring system for the grinding wheel status of a grinding robot based on piezoelectric sensing according to claim 5, characterized in that, The dual-parameter fusion feature value P being less than 1.0 indicates a normal state; 1.0≤P < 1.5, judged as slight wear; 1.5≤P < 2.2, judged as severe wear; P≥2.2, judged as local damage or serious failure.

7. The intelligent monitoring system for the grinding wheel status of a grinding robot based on piezoelectric sensing according to claim 6, characterized in that, The alarm is only triggered when the dual-parameter fusion feature value P ≥ 1.5 is collected for three consecutive sampling cycles.

8. The intelligent monitoring system for the grinding wheel status of a grinding robot based on piezoelectric sensing according to claim 4, characterized in that, If the amplitude at the spindle rotation frequency in the spectrum exceeds 30% of the total energy, it is determined that there is a problem with the dynamic balance of the grinding wheel.

9. The intelligent monitoring system for the grinding wheel status of a grinding robot based on piezoelectric sensing according to claim 1, characterized in that, It also includes an output and execution module, wherein the output and execution module is used to display the diagnostic results of the intelligent diagnostic module.

10. The intelligent monitoring system for the grinding wheel status of a grinding robot based on piezoelectric sensing according to claim 9, characterized in that, The output and execution module includes a human-machine interface, a communication interface, and an alarm unit. The communication interface is connected to the intelligent diagnostic module, and the communication interface is also connected to the human-machine interface and the alarm unit.

Images