A method for monitoring electrical parameters of a steering arm machining device drive system

Abstract

The present application relates to the technical field of electrical parameter measurement of driving system, and discloses a kind of electrical parameter monitoring method of steering arm processing equipment driving system, comprising: calibrating the harmonic amplitude ratio under different load index, and establishing load ratio compensation atlas;When online monitoring, real-time acquisition instantaneous ratio and current load index;Based on atlas and current load index, determine the expected health ratio, and compare it with instantaneous ratio to determine cable state, the present application uses load index dynamic calibration harmonic ratio measurement reference, realizes the separation of motor load change and cable characteristic change on electrical measurement, avoids the measurement misjudgment caused by processing load fluctuation, and improves the monitoring reliability under dynamic working condition.

Description

Technical Field

[0001] This invention relates to a method for monitoring the electrical parameters of a steering arm processing equipment drive system, belonging to the field of drive system electrical parameter measurement technology. Background Technology

[0002] Currently, in precision machining equipment and servo drive systems, electrical parameter measurement is fundamental for condition monitoring and fault diagnosis. These systems typically rely on measurement points installed at the driver end, connected to the motor's point of action via power cables. The assumption is that the electrical parameters measured at the driver end reflect the electrical state of the motor. However, power cables are not ideal transmission channels; they are complex networks of distributed resistance, inductance, and capacitance parameters, which distort the high-frequency switching signals generated by the driver. During actual operation, cables experience dynamic changes in distributed parameters and contact resistance due to heating, repeated bending, connector vibration, or oxidation. Existing electrical parameter measurement methods generally focus on extracting signals characterizing the motor load from strong noise backgrounds, such as filtering or using algorithms to eliminate switching harmonic interference. However, these methods treat the dynamic changes in the cable as unobservable disturbance sources. This leads to measurement confusion; when electrical parameters fluctuate, the measurement system cannot fundamentally distinguish whether the fluctuation originates from a genuine change in the motor load or from a drift in the RLC characteristics of the measurement channel, i.e., the cable itself.

[0003] Besides the lack of measurement dimensions, existing control strategies suffer from logical blind spots in handling the relationship between load fluctuations and transmission distortion. For example, Chinese invention patent CN114633115B discloses a composite grinding machine for automotive steering arm brackets. Although the technical solution uses a pressure sensor to provide real-time feedback signal strength and automatically adjust the position of the grinding head to ensure machining coaxiality, the judgment logic assumes that the electrical transmission circuit between the sensor and the motor is a constant and ideal channel. However, under harsh conditions such as high-frequency vibration and coolant splashing during steering arm machining, if the cable connectors oxidize, leading to increased impedance and signal attenuation, the existing technology cannot distinguish whether the signal change originates from workpiece mechanical alignment deviation or cable electrical characteristic degradation. The lack of health status perception of the measurement channel itself makes the system prone to misjudging structural faults in the cable layer as abnormal loads in the machining layer, generating a large number of false alarms or missed alarms, limiting the practical application of electrical monitoring in high-reliability scenarios. This unreliability of the measurement benchmark exposes all upper-level diagnostic algorithms to the risk of failure, which is a long-standing and unresolved problem in the field of electrical parameter measurement.

[0004] Therefore, how to provide an electrical parameter monitoring method that can first calibrate and verify the health status of the measurement channel itself online before load measurement, and solve the measurement uncertainty caused by channel status drift, is the technical problem to be solved by this invention. Summary of the Invention

[0005] To address the problems mentioned in the background art, the technical solution of the present invention is as follows: A method for monitoring the electrical parameters of a steering arm processing equipment drive system, comprising the following steps:

[0006] When the drive system is in a healthy baseline state, the driver output current signal is acquired;

[0007] Determine the first amplitude of the first carrier harmonic frequency and the second amplitude of the second carrier harmonic frequency in the output current signal;

[0008] The first and second amplitude values ​​of the drive system under different load indices are collected. The load index is an electrical parameter that characterizes the real-time load of the motor.

[0009] Calculate the corresponding healthy baseline ratio based on different load indices and their corresponding first and second amplitudes, and establish a load ratio compensation map between the load index and the healthy baseline ratio.

[0010] When the drive system is in online monitoring mode, the driver output current signal is collected in real time, and the current load index is obtained in real time;

[0011] The real-time acquired output current signal is processed to extract the instantaneous first amplitude and instantaneous second amplitude in real time, and to calculate the instantaneous ratio in real time.

[0012] Based on the load ratio compensation map and the current load index, determine the expected health ratio;

[0013] The cable condition of the drive system is determined based on the comparison between the instantaneous ratio and the expected health ratio.

[0014] Preferably, the first carrier harmonic frequency is the base frequency of the driver's PWM switching frequency, and the second carrier harmonic frequency is an odd multiple of the PWM switching frequency.

[0015] Preferably, the step of processing the real-time acquired output current signal to extract the instantaneous first amplitude and the instantaneous second amplitude in real time, and to calculate the instantaneous ratio in real time, includes: inputting the real-time acquired output current signal in parallel to a first bandpass filter with a center frequency of the first carrier harmonic frequency and a second bandpass filter with a center frequency of the second carrier harmonic frequency; and obtaining the output amplitudes of the first bandpass filter and the second bandpass filter respectively as the instantaneous first amplitude and the instantaneous second amplitude.

[0016] Preferably, the step of determining the cable status of the drive system based on the comparison result of the instantaneous ratio and the expected health ratio includes: when the deviation between the instantaneous ratio and the expected health ratio exceeds a preset deviation threshold, the cable status of the drive system is determined to be abnormal.

[0017] Preferably, in the healthy baseline state, the method further includes: determining a first phase of the first carrier harmonic frequency and a second phase of the second carrier harmonic frequency; calculating and storing the healthy baseline phase difference based on the first phase and the second phase; in the online monitoring state, the method further includes: extracting the instantaneous first phase and the instantaneous second phase in real time; calculating the instantaneous phase difference in real time; and the step of determining the cable state of the drive system further includes: determining the fault mode of the cable state of the drive system based on the comparison result of the instantaneous phase difference and the healthy baseline phase difference.

[0018] Preferably, the step of determining the fault mode of the cable condition includes: calculating the amplitude ratio deviation, which is the difference between the instantaneous ratio and the expected healthy ratio; calculating the phase difference deviation, which is the difference between the instantaneous phase difference and the healthy baseline phase difference; and classifying the fault mode into resistive fault, reactive fault, and combined fault based on the amplitude ratio deviation and the phase difference deviation.

[0019] Preferably, the method further includes: detecting whether the drive system is in a stopped state; when the drive system is in a stopped state, controlling the inverter of the drive system to inject a non-rotating common-mode detection voltage signal into the cable; measuring the common-mode current generated by the common-mode detection voltage signal; calculating the common-mode impedance characteristics based on the common-mode detection voltage signal and the common-mode current, and comparing the common-mode impedance characteristics with a pre-stored stopped health baseline to determine the insulation status of the cable.

[0020] Preferably, in the healthy baseline state, the method further includes: simultaneously acquiring the electrical temperature index while acquiring the first and second amplitudes of the drive system under different load indices; the load ratio compensation graph further includes the compensation relationship between the electrical temperature index and the healthy baseline ratio; in the online monitoring state, the method further includes: acquiring the current electrical temperature index in real time; and, the step of determining the expected health ratio specifically involves determining the expected health ratio based on the load ratio compensation graph and simultaneously based on the current load index and the current electrical temperature index.

[0021] Preferably, the electrical temperature index is a parameter determined by measuring the resistance value of the motor windings.

[0022] Preferably, in the healthy baseline state, the method further includes: measuring and storing the healthy background noise baseline, where the healthy background noise baseline is the spectral amplitude of the driver output current signal in at least one preset quiet frequency band, the quiet frequency band being far from the first carrier harmonic frequency and the second carrier harmonic frequency; in the online monitoring state, the method further includes: measuring the instantaneous background noise in the quiet frequency band in real time; and the step of determining the cable status of the drive system further includes: determining whether there is an intermittent fault in the cable of the drive system based on the comparison result between the instantaneous background noise and the healthy background noise baseline.

[0023] Compared with the prior art, the beneficial effects of the present invention are:

[0024] 1. In the monitoring of electrical parameters of the drive system of steering arm processing equipment, a self-calibration mechanism for electrical measurement benchmarks is constructed by establishing a healthy baseline ratio and calculating the instantaneous ratio in real time during monitoring. Utilizing the ratio of two harmonic amplitudes, rather than a single amplitude, the measurement method has inherent insensitivity to common drifts affecting the entire system, such as fluctuations in the driver bus voltage or overall temperature changes, which tend to cause the two amplitudes to change in the same direction. This ratio is also highly sensitive to structural changes in the cable that alter the high-frequency impedance characteristics of the system, such as connector oxidation or cable strand breakage, which cause non-proportional distortion in the response at different frequencies. This transforms the judgment of cable structural faults from fuzzy monitoring of absolute amplitude changes to clear identification of distortions in the system's frequency response characteristics. In principle, it distinguishes between systemic drift interference and structural fault signals in electrical measurements, solving the technical problem of distinguishing between false alarms and true alarms in the field of electrical parameter measurement.

[0025] 2. By introducing the load index as a second electrical measurement dimension, and establishing a compensation spectrum of the harmonic ratio and load index under a healthy baseline state, the monitoring benchmark is transformed from a static value to an expected health ratio that dynamically changes with the operating conditions. During online monitoring, the expected health ratio under the current operating conditions is obtained by querying the real-time load index spectrum synchronously. The instantaneous ratio measured in real time is compared with this expected health ratio. The load index, as an electrical parameter, is used to dynamically calibrate the measurement benchmark of high-frequency harmonics, thereby achieving the separation of two different physical sources, namely motor load changes and cable characteristic changes, in electrical measurements. This avoids measurement misjudgments caused by drastic fluctuations in processing load and improves the reliability of electrical parameter monitoring under dynamic operating conditions.

[0026] 3. By utilizing the phase information generated during the harmonic amplitude extraction process, and by additionally storing the healthy baseline phase difference characterizing the system's phase frequency characteristics during the healthy baseline state, synchronous comparison is performed during online monitoring to construct a two-dimensional electrical measurement matrix of amplitude ratio deviation and phase difference deviation. Since cable faults with different physical properties have different effects on the system's amplitude frequency characteristics and phase frequency characteristics, the phase data obtained during amplitude extraction is used to expand electrical parameter monitoring from a simple fault presence or absence judgment to fault mode classification diagnosis, thereby enhancing the depth and practical value of electrical measurement. Attached Figure Description

[0027] Figure 1 This is a flowchart of the electrical parameter monitoring logic based on the load ratio compensation spectrum of the present invention.

[0028] Figure 2 This is a comparison chart of monitoring deviations of different test groups under sudden load changes and fault conditions in this invention;

[0029] Figure 3 This is a timing diagram showing the signal interaction between the monitoring system of the present invention and the parallel filtering and fault determination. Detailed Implementation

[0030] To make the technical solution, purpose, and beneficial effects of the present invention clearer, the technical solution of the present invention will be described in detail below. It should be noted that the specific embodiments described herein are only used to explain the present invention and are not intended to limit the scope of protection of the present invention.

[0031] This invention discloses a method for monitoring the electrical parameters of a steering arm processing equipment drive system, including a health baseline calibration stage and an online monitoring status diagnosis stage. The calibration stage collects the harmonic amplitude values ​​of the driver output current signal under different load indices when the drive system is healthy, calculates the corresponding health baseline ratio, and establishes a load ratio compensation spectrum between the load index and the health baseline ratio. The diagnosis stage collects instantaneous harmonic amplitude values ​​in real time to calculate the instantaneous ratio, synchronously acquires the current load index, determines the expected health ratio based on the spectrum and the current load index, and finally determines the cable status of the drive system by comparing the instantaneous ratio with the expected health ratio. In this technical solution, the high-frequency carrier harmonics generated by the PWM switch action in the driver output current signal are used as inherent signals to detect the characteristics of the cable RLC network. At least two frequency points are selected, namely the first carrier harmonic frequency and the second carrier harmonic frequency, wherein the first carrier harmonic frequency is preferably the driver PWM switch frequency. The fundamental frequency of the frequency is used, while the second carrier harmonic frequency is preferably an odd multiple of the PWM switching frequency, such as the 3rd or 5th harmonic. Two harmonic amplitudes, namely the first and second amplitudes, are used, and their ratio is calculated as a monitoring quantity. This is because individual amplitudes are easily affected by systematic drift such as fluctuations in the driver bus voltage, while the ratio of the two amplitudes can reduce the impact of such common interferences. It also remains sensitive to high-frequency impedance characteristic distortions caused by cable connector oxidation, broken strands, etc. The execution of the health baseline calibration procedure requires the drive system, especially the power cable, to be in a known good condition, such as after new equipment installation or maintenance confirmation. During calibration, the servo driver's control program enables the motor to operate stably at multiple preset load points. At each load point, the system simultaneously performs two measurements: one is to collect electrical parameters characterizing the real-time load of the motor as a load index. In the servo system, the torque command output by the driver or the q-axis current feedback value can be directly used. The process involves two main steps: first, acquiring the driver's output current signal and using high-resolution spectral analysis, such as Fast Fourier Transform, to determine the first amplitude at the first carrier harmonic frequency and the second amplitude at the second carrier harmonic frequency; then, calculating the corresponding healthy baseline ratio under this load index, i.e., the ratio of the second amplitude to the first amplitude; traversing all preset load points, storing the obtained multiple sets of load index and corresponding healthy baseline ratio data pairs as a data structure to establish a load ratio compensation spectrum. In engineering implementation, this can be a lookup table or a polynomial fitting function, defining the inherent electrical response characteristics of the harmonic ratio as the motor load changes under healthy cable conditions.

[0032] In online monitoring mode, this method is executed in real-time and in parallel by the processor inside the driver. The processor acquires the current load index in real time and simultaneously collects the driver's output current signal. This time-domain signal is then fed in parallel into two digital bandpass filters. The center frequency of the first bandpass filter is set to the first carrier harmonic frequency, and the center frequency of the second bandpass filter is set to the second carrier harmonic frequency. The output amplitudes of the two filters are acquired as the instantaneous first and second amplitudes, and their ratio is calculated in real time to obtain the instantaneous ratio. The core judgment logic of this monitoring method is as follows: First, based on the real-time acquired current load index, the established load ratio compensation spectrum is consulted to determine the expected health ratio matching the current operating condition. Then, the instantaneous ratio is compared with the expected health ratio. When the absolute value of the deviation exceeds a preset deviation threshold, the system determines that the cable condition of the drive system is abnormal. This deviation threshold can be set by statistically analyzing the ratios under different operating conditions in the healthy baseline state. The method is determined based on the normal fluctuation range. To further enhance the diagnostic depth of the measurement, this method can also utilize the phase information accompanying the extraction of harmonic amplitude. When calibrating the healthy baseline, the first phase of the first carrier harmonic frequency and the second phase of the second carrier harmonic frequency are determined simultaneously, and the phase difference of the healthy baseline is calculated and stored. During online monitoring, the instantaneous first phase and instantaneous second phase are extracted in real time, and the instantaneous phase difference is calculated. By constructing a two-dimensional diagnostic matrix of amplitude ratio deviation and phase difference deviation, the fault modes of the cable condition are distinguished. For example, if the amplitude ratio deviation is small but the phase difference deviation is small, it is judged as a resistive fault, while if the phase difference deviation is large, it is judged as a reactive fault or a composite fault. Since the electrical parameters of the cable and motor are affected by temperature, in order to eliminate the measurement deviation caused by temperature drift, the load ratio compensation spectrum can also introduce a temperature dimension. When calibrating the healthy baseline, the electrical temperature index is collected simultaneously. Preferably, the measured motor winding resistance value is strongly correlated with the system temperature by means of measurement, such as applying a DC bias voltage to the q-axis.

[0033] This establishes a multi-dimensional compensation spectrum including load index and electrical temperature index, further comprising the compensation relationship between the electrical temperature index and the healthy baseline ratio. During online monitoring, the system synchronously acquires the current load index and current electrical temperature index in real time, and determines the expected healthy ratio after simultaneous load and temperature compensation based on this multi-dimensional spectrum. This method may also include a self-check procedure for electrical parameters in a shutdown state. When the drive system is detected to be in a shutdown state, the system can utilize the inverter hardware of the driver to simultaneously inject a non-rotating common-mode detection voltage signal into the three phases of the cable. This common-mode signal will not generate torque in the motor, ensuring detection safety. The common-mode impedance characteristics are calculated by measuring the common-mode current generated by this common-mode voltage. Since the common-mode current mainly reflects the distributed capacitance and insulation resistance of the cable to ground, this common-mode impedance characteristic is compared with the pre-stored shutdown health baseline to determine the insulation status of the cable and achieve a safety pre-inspection before startup. To monitor intermittent faults that are difficult to detect through harmonic ratios, this method can also introduce background noise monitoring. In the healthy baseline state, one or more quiet frequency bands far away from all major harmonic frequencies are selected in the spectrum of the driver output current signal, and the spectral amplitude within the frequency band is measured and stored as the healthy background noise baseline. During online monitoring, the instantaneous background noise within the quiet frequency band is measured in real time. When the instantaneous background noise is higher than the healthy background noise baseline, it can be determined that the cable of the drive system has intermittent faults such as partial discharge and poor micro-contact.

[0034] Example 1: In a specific application scenario of steering arm machining, a large five-axis machining center is used for heavy-duty cutting of high-hardness alloy forgings. The servo drive system is connected to the spindle motor via a long-distance power cable, and during machining, it reciprocates at high speed with the cable chain and is subjected to vibration. This working condition presents a technical challenge to the monitoring of electrical parameters: the system must distinguish whether the fluctuation of electrical parameters is due to drastic changes in the cutting load or to oxidation or micro-fractures of the connectors caused by long-term bending and vibration of the cable. This example adopts the monitoring method described in the specific implementation. This method has been used in the equipment commissioning phase, i.e., when the cable is in a healthy state, by controlling the motor to run under different torque outputs, a load ratio compensation spectrum is calibrated and established. This spectrum stores the ratio of the q-axis current feedback value as the load index to the corresponding healthy baseline, and the third harmonic amplitude of the PWM switching frequency as the second amplitude. With as the first amplitude The functional relationship between the ratios is such that at a certain moment during the machining process, the machine tool executes a heavy-load cutting command, and the tool cuts into the workpiece at full width, causing a surge in the motor load. The monitoring system inside the driver simultaneously acquires electrical parameters in two measurement dimensions: one is the current load index, i.e., the q-axis current, which instantaneously and significantly increases to 80% of the rated value; the other is the instantaneous ratio calculated after the real-time current signal is collected and filtered by dual bandpass filters. The load also fluctuated due to changes in motor impedance. The system immediately performed dynamic benchmark calibration, and based on the increased load index, queried the load ratio compensation chart to determine the expected health ratio corresponding to the current 80% rated load. Then, it compared the instantaneous ratio with the expected health ratio obtained from the dynamic query and found that the deviation between the two was only 0.8%, which did not exceed the preset deviation threshold of 5%. The system determined that the cable condition was normal, and the fluctuation of electrical parameters was attributed to normal changes in processing conditions.

[0035] After the equipment has been running continuously for several months, the cable connectors have undergone chronic oxidation due to long-term vibration, resulting in increased contact resistance. At this time, the machine tool performs a similar heavy-load cutting condition again, and the current load index also climbs to 80% of the rated value. The system once again consults the spectrum based on this load index to obtain the expected health ratio corresponding to the previous condition. However, due to the change in the high-frequency impedance characteristics of the system caused by cable oxidation, the instantaneous ratio calculated this time has a structural deviation, resulting in a deviation of 8.2% from the expected health ratio, exceeding the 5% deviation threshold. The system determines that the cable condition of the drive system is abnormal and outputs a measurement channel fault alarm, rather than a motor overload alarm. This embodiment introduces the load index, an orthogonal electrical measurement parameter, to construct a load ratio compensation spectrum, so that the monitoring benchmark changes from a static value to an expected health ratio that is dynamically adjusted according to the operating conditions. This measurement method uses the known electrical parameter, the load index, to dynamically calibrate another monitoring quantity.

[0036] Example 2: To verify the ability of the method of the present invention to distinguish between abnormal cable conditions and motor load fluctuations during measurement, a test platform for the drive system of a steering arm processing equipment was built. This platform includes a servo driver connected to a servo motor via a 30-meter power cable. A programmable magnetic powder brake is connected to the motor shaft end to simulate different processing loads. The PWM switching frequency of the test driver was set to 10kHz, the sampling frequency of the data acquisition system was 200kS / s, and the first carrier harmonic frequency for online monitoring was selected as 10kHz, and the second carrier harmonic frequency as 30kHz. Two control groups were set up: control group A, control group B, and the test group of the present invention, i.e., test group C. Control group A used conventional monitoring methods, monitoring only the 10kHz fundamental amplitude. The absolute change was used to set a fixed alarm threshold; control group B used the uncompensated harmonic ratio method, i.e., monitoring... The ratio of this to the fixed healthy baseline calibrated under no-load conditions. Comparison; Test group C adopted the method of the present invention, using the load ratio compensation spectrum established by calibrating under healthy cable conditions through different load points, i.e., 0% to 100% of rated torque. Based on this, dynamic compensation is performed. The test process simulates two working conditions: Working condition 1, load change, in which a step load of 10% to 90% of rated torque is applied through a magnetic powder brake while the cable is in good condition; Working condition 2, cable fault, in which a 1.5 ohm contact resistor is connected in series at the terminal of the power cable to simulate connector oxidation fault while the motor maintains a stable load of 30% of rated torque. The test data records the alarm status of the three methods under the two working conditions, as shown in Table 1.

[0037] Table 1: Results of the Comparison Test of Monitoring Methods

[0038]

[0039] The test results show that the method in control group A cannot distinguish between load changes and faults, generating false alarms in operating condition one; although the method in control group B uses a ratio, its reference is static and cannot adapt to impedance changes caused by motor load variations, also generating false alarms in operating condition one; test group C, using the method of this invention, utilizes the load index... Query load ratio compensation graph To obtain a dynamic and expected health ratio. Correctly follow load changes to make the instantaneous ratio and The deviation remained within the set 5% threshold during the load surge in Condition 1, without generating false alarms. Meanwhile, during the cable fault in Condition 2, due to... The deviation cannot be explained by load changes, and The deviation exceeds the threshold.

[0040] Example 3: This example combines Figures 1 to 3 A method for monitoring electrical parameters of a steering arm processing equipment drive system is described, such as... Figure 1As shown, the process logic first executes two data acquisition branches in parallel. On the one hand, it acquires the current load index, such as the q-axis current, which characterizes the real-time load of the motor, and optionally acquires the electrical temperature index based on the motor winding resistance value. On the other hand, it collects the driver output current signal to obtain the high-frequency PWM switching harmonic signal. Subsequently, on the one hand, it uses the calibration data under the healthy baseline state, i.e., the load ratio compensation spectrum, to determine the expected healthy ratio by querying the spectrum and performing dynamic interpolation. On the other hand, it performs parallel bandpass filtering on the current signal to separate the first carrier frequency and the second carrier frequency, and calculates the instantaneous ratio based on the real-time extracted harmonic amplitude ratio. After this, the process enters the comparison and analysis stage, calculates the deviation between input A, i.e., the instantaneous ratio, and input B, i.e., the baseline expected value, and determines whether the deviation exceeds the preset deviation threshold. If the judgment result is no, the system state is normal and the fluctuation is attributed to the processing load fluctuation. If the judgment result is yes, the cable state is determined to be abnormal and the fault nature is determined to be a structural fault.

[0041] like Figure 2 As shown, this bar chart, with the monitoring deviation percentage (%) as the vertical axis and the test group as the horizontal axis, displays the test results of control group A, control group B, and test group C under different operating conditions. The legend clearly distinguishes between the deviation under load change conditions (represented by horizontal bars) and the deviation under cable fault conditions (represented by diagonal bars). The data shows that control group A exhibits high monitoring deviation under both load change and cable fault conditions; control group B shows a decrease in deviation under load change conditions but still maintains a high deviation under cable fault conditions; while test group C has extremely low monitoring deviation under load change conditions and only exhibits high deviation under cable fault conditions. Figure 3 As shown, the timing interaction logic demonstrates the signal flow between the monitoring system, the first bandpass filter, the second bandpass filter, the data storage, and the alarm system. The monitoring system first acquires the driver output current signal and obtains the current load index, i.e., the q-axis current. In the parallel signal processing stage, the monitoring system sends the input signal to the first bandpass filter with a center frequency of 10kHz to output the instantaneous first amplitude A1, and to the second bandpass filter with a center frequency of 30kHz to output the instantaneous second amplitude A3. Subsequently, the monitoring system calculates the instantaneous ratio A3 / A1, queries the load ratio compensation spectrum from the data storage unit to obtain the returned expected health ratio, and compares the instantaneous ratio with the expected health ratio. If the deviation is within the threshold, the cable condition is determined to be normal. If the deviation exceeds the threshold, i.e., >5%, an abnormal cable condition alarm is sent to the alarm system.

[0042] Example 4: This example illustrates the load ratio compensation graph. The standardized procedures aim to eliminate measurement uncertainties introduced by high-frequency impedance nonlinearity changes caused by motor load variations. This is a prerequisite for achieving online dynamic compensation. The initial state of the procedures is defined as follows: the drive system is in a healthy baseline state; the cable has been confirmed to be intact through measurements such as insulation resistance and continuity resistance tests; the enabling environment includes the servo drive, the motor, and an external load device capable of applying and stabilizing the feedback torque value, such as a dynamometer or magnetic powder brake, whose torque control accuracy is not lower than the rated torque. The execution steps of the procedure are as follows: First, define the load calibration point set; within the rated torque range of the motor, select... A discrete load index point, As a calibration point, it covers the nonlinear response region of the load. The values ​​should generally not be less than 10 points; the preferred distribution is to select 11 points evenly at 10% intervals between 0% and 100% of the rated torque, i.e., 0%, 10%, 20% up to 100%; the second step is to perform gradient data acquisition; the system starts from the first calibration point, i.e., 0% load, and controls the external load device to apply the corresponding torque; the driver enters torque control mode and monitors the current load index, i.e., the q-axis current feedback value and the motor speed; when the load index fluctuates less than within a 1.0 second time window... Once the rotational speed stabilizes, the system determines that a stable measurement state has been reached; the system immediately acquires a data frame, which includes: the load index under the current stable state. And high-speed data acquisition, such as 200 kS / s, with a length of ,like The driver outputs current signal at each point. The system then controls the load device to switch to the next calibration point and repeats the above steady-state judgment and data acquisition process until all [devices are in use]. All data collection at each calibration point has been completed.

[0043] The third step involves performing offline feature extraction and map construction; processing the collected data... Each data frame is processed sequentially by the system: for each current signal... Apply a high-resolution FFT, such as a 16384-point Hanning window FFT; accurately measure the first carrier harmonic frequency in the obtained spectrum. amplitude at Second carrier harmonic frequency amplitude at Calculate the calibration point. Corresponding healthy baseline ratio Step 4: Define the map storage and interpolation methods; the system will... Group of data pairs, As a lookup table, it is stored in the driver's non-volatile memory to complete the calibration of the load ratio compensation spectrum; the interpolation method used during online monitoring is defined, and the method that balances computational load and accuracy is linear interpolation, that is, when the current load index is acquired in real time... Landing at the calibration point and Between, the expected health ratio pass The corresponding point between these two points and The standardization procedure, which establishes the standardization process, transforms the construction of the load ratio compensation map from a description of measurement principles into a reproducible engineering process consisting of deterministic equipment specifications, quantitative calibration points, steady-state judgment criteria, and interpolation algorithms.

[0044] Example 5: This example further elaborates on the quantization calibration procedure for the two-dimensional diagnostic matrix used for fault mode classification; it is executed when the drive system is in a healthy baseline state, with the aim of calculating the amplitude ratio deviation. Phase difference deviation Provides reproducible judgment boundaries to distinguish cable faults of different physical properties; measures and stores the healthy baseline phase difference on healthy cables. Perform known fault injection calibration: Step 1, insert a cable terminal block into the cable terminal block. The ohm calibration resistor is used to simulate a resistive fault. Under this condition, the system measurement parameters are run, and the deviation data points that stabilize at this point are recorded. , If obtained Step two: Remove the resistor and connect it between the cable core and the shield ground. The calibration capacitor is used to simulate a capacitive fault, such as insulation dampness; the deviation data points are then measured and recorded again. , If obtained Based on these two measurement calibration points, the system can establish a quantized classification boundary and set the phase difference deviation threshold as follows: The amplitude ratio deviation threshold is During online monitoring, when and When, it is determined to be a resistive fault, when and When this occurs, it is determined to be a reactive fault.

[0045] This embodiment also illustrates the health background noise baseline used for intermittent fault monitoring. The calibration procedure for its alarm thresholds; also performed under healthy baseline conditions: First step, select the quiet frequency band. On the FFT spectrum of the driver output current signal, select the one that is far away Such as 10kHz, and For example, 30kHz and its main sidebands, feasible choices are... to The second step is to calibrate the baseline, control the motor to run under no-load and healthy cable conditions, and calculate and store the data in the quiet frequency band. The average spectral amplitude within, such as Stored as a healthy baseline noise level The third step is to calibrate the thresholds, and the system continuously monitors the system when it is in a healthy state. The fluctuation is calculated, and its statistical standard deviation σ is determined. N ,like Set alarm threshold Set to N base +6σ N That is, about During online status monitoring, the system performs parallel computation. Instantaneous background noise within the frequency band ,once Exceed This means that an intermittent fault is identified, such as poor cable micro-contact. This embodiment further elaborates on the shutdown health baseline used for offline self-testing. The calibration procedure; during the initial equipment commissioning, confirming that the cable is dry and its insulation resistance is greater than [value missing]. The system executes while in a shutdown state; it initiates a self-test program, reuses the driver inverter, and injects power into the three phases U, V, and W of the cable. , common-mode detection voltage signal The system at this time measures the total modal current flowing through the grounding loop. To obtain stable measurement values, such as System calculation of common-mode impedance characteristics ,get Store this value as the shutdown health baseline. Simultaneously set alarm thresholds, such as ; In subsequent equipment shutdown self-tests, if the system measures Below If the insulation condition of the cable deteriorates or becomes damp, it is determined that the cable insulation condition has deteriorated or that the cable has become damp.

[0046] Example 6: This example illustrates the preset deviation threshold used for measurement. The standardized calibration procedure and the low-computational-load implementation method for extracting the instantaneous first and second amplitudes from the digital bandpass filter in real time; to determine the statistically significant deviation threshold, a baseline fluctuation assessment procedure is first executed; this procedure is executed under the premise that the drive system is in a healthy baseline state and the cable is intact, and the control motor traverses all calibrated load index points in the load ratio compensation spectrum. At each load index point The system obtains the data from the query map. Simultaneously, continuous data collection Instantaneous ratio of each measurement cycle , Take 1000 to calculate this The instantaneous ratio and Deviation sequence between ; Calculate the deviation sequence Standard deviation σ i After traversing all load points, select all σ values. i The maximum value σ in max , the σ max Characterizes the inherent fluctuation level of the health system under full operating conditions; preset deviation threshold. That is, set it as a multiple of the maximum standard deviation, ΔR Th =k•σ max ,in Here is the confidence coefficient. Usually, it is taken as 3 to 6, when When σ is measured max for The threshold can then be set to To achieve real-time amplitude extraction with low computational cost, the digital bandpass filter is preferably a second- or fourth-order infinite impulse response (IIR) filter; to extract the instantaneous first amplitude. For example, the implementation steps include: First, real-time acquisition of the driver output current signal. The input center frequency is the first carrier harmonic frequency. The first bandpass filter yields the filtered time-domain signal. The second step is to... The signal undergoes root mean square (RMS) calculation within a sliding time window. Performed within, window length Preferred The signal within the window is calculated in real time as an integer multiple of the period, such as 5 periods. The third step is to multiply the RMS value by the calibration coefficient. The result obtained is taken as the first instantaneous amplitude. Instantaneous second amplitude The calculation process is the same, with the center frequency of the corresponding second bandpass filter set as the second carrier harmonic frequency. .

[0047] This embodiment describes the standardized procedure for establishing the multidimensional compensation map mentioned in the specific implementation, which also includes the load index. and electrical temperature index The initial state definition of the procedure is the same as in Example 4, namely, a healthy cable and a platform with a dynamometer; electrical temperature index. By measuring the resistance of the motor windings The calibration procedure is as follows: First, define the temperature calibration point set, such as selecting three temperature points. for i.e., chiller for Instant temperature controller for Instantaneous heating; second step, at the first temperature point ( Under the following conditions, the complete process of gradient data acquisition and offline feature extraction in Example 4 is executed to measure the motor winding resistance at the corresponding temperature. As , obtain a set One data pair, The third step is to control the motor to run, raise its temperature, and stabilize it at the second temperature point. ( ), measure the corresponding Repeat step two again to obtain the second set. One data pair, Fourth step: Repeat the operation until all temperature points are calibrated; the system stores a three-dimensional lookup table. During online monitoring, the system obtains the current load index in real time. and the current electrical temperature index By performing bilinear interpolation in this three-dimensional lookup table, the expected health ratio, after both load and temperature compensation, is determined. The load ratio compensation map is constructed in the controller's non-volatile memory unit as follows: Each discrete data pair is structured as a two-dimensional lookup table, and each data pair is mapped to a calibration load index. Ratio to corresponding healthy baseline During the online monitoring operation cycle, the processor obtains the real-time load index. And retrieve the value range to which the lookup table belongs. A linear interpolation algorithm is used based on the formula. Calculate the expected health ratio under current operating conditions to ensure that the output baseline value continuously and monotonically responds to small changes in the load exponent, eliminating step measurement errors caused by data gaps between discrete calibration points; preset deviation thresholds. , The instantaneous ratio sample sequence of each sampling period is used to calculate the statistical standard deviation of the sequence, which characterizes the inherent volatility level at the load point. The system traverses all... The standard deviation data of each calibration node is used to filter the global maximum value σ. max , set the deviation threshold Set as the global maximum value σ max With confidence coefficient product, The value is set to an integer between 3 and 6 based on the principle of normal distribution coverage probability.

[0048] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0049] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A method of monitoring electrical parameters of a drive system of a knuckle processing apparatus, characterized in that, The method comprises the following steps: collecting the drive output current signal when the drive system is in a healthy baseline state; determining a first amplitude value of a first carrier harmonic frequency and a second amplitude value of a second carrier harmonic frequency in the output current signal; collecting the first amplitude value and the second amplitude value under different load indices of the drive system, the load indices being electrical parameters representing real-time load of the motor; calculating corresponding health baseline ratios based on the different load indices and the corresponding first amplitude value and second amplitude value, and establishing a load ratio compensation atlas between the load indices and the health baseline ratios; collecting the drive output current signal in real time when the drive system is in an online monitoring state, and obtaining a current load index in real time; processing the collected output current signal in real time to extract an instantaneous first amplitude value and an instantaneous second amplitude value in real time, and calculate an instantaneous ratio in real time; determining an expected health ratio based on the load ratio compensation atlas and the current load index; determining the cable state of the drive system based on a comparison result of the instantaneous ratio and the expected health ratio; and, when in the healthy baseline state, further comprising: determining a first phase of the first carrier harmonic frequency and a second phase of the second carrier harmonic frequency; calculating and storing a health baseline phase difference based on the first phase and the second phase; when in the online monitoring state, further comprising: extracting an instantaneous first phase and an instantaneous second phase in real time; calculating an instantaneous phase difference in real time; and the step of determining the cable state of the drive system further comprises: determining a failure mode of the cable state of the drive system based on a comparison result of the instantaneous phase difference and the health baseline phase difference; the step of determining the failure mode of the cable state comprises: calculating an amplitude ratio deviation, the amplitude ratio deviation being a difference between the instantaneous ratio and the expected health ratio; calculating a phase difference deviation, the phase difference deviation being a difference between the instantaneous phase difference and the health baseline phase difference; and distinguishing the failure mode into a resistive failure, a reactive failure and a composite failure based on the amplitude ratio deviation and the phase difference deviation; the method further comprises: detecting whether the drive system is in a shutdown state; when the drive system is in the shutdown state, controlling an inverter of the drive system to inject a non-rotating common-mode detection voltage signal into the cable; measuring a common-mode current generated by the common-mode detection voltage signal; calculating a common-mode impedance feature based on the common-mode detection voltage signal and the common-mode current, and comparing the common-mode impedance feature with a pre-stored shutdown health baseline to determine an insulation state of the cable; when in the healthy baseline state, further comprising: measuring and storing a health background noise baseline, the health background noise baseline being a spectral amplitude of the drive output current signal in at least one pre-set quiet frequency band, the quiet frequency band being far away from the first carrier harmonic frequency and the second carrier harmonic frequency; when in the online monitoring state, further comprising: measuring an instantaneous background noise in the quiet frequency band in real time; and the step of determining the cable state of the drive system further comprises: determining whether the cable of the drive system has an intermittent failure based on a comparison result of the instantaneous background noise and the health background noise baseline.

2. The method of claim 1, wherein the method further comprises: The first carrier harmonic frequency is a fundamental frequency of a PWM switching frequency of the drive, and the second carrier harmonic frequency is an odd multiple frequency of the PWM switching frequency.

3. The method of claim 1, wherein the method further comprises: The step of processing the real-time collected output current signal to real-time extract the instantaneous first amplitude and the instantaneous second amplitude, and real-time calculate the instantaneous ratio, comprises: inputting the real-time collected output current signal into a first band-pass filter with a center frequency of the first carrier harmonic frequency and a second band-pass filter with a center frequency of the second carrier harmonic frequency in parallel; obtaining the output amplitudes of the first band-pass filter and the second band-pass filter respectively as the instantaneous first amplitude and the instantaneous second amplitude.

4. The method of claim 1, wherein the method further comprises: The step of determining the cable state of the drive system based on the comparison result of the instantaneous ratio and the expected healthy ratio, comprises: when the deviation of the instantaneous ratio and the expected healthy ratio exceeds a preset deviation threshold, determining that the cable state of the drive system is abnormal.

5. The method of claim 1, wherein the method further comprises: In the healthy baseline state, further comprising: synchronously collecting an electrical temperature index when collecting the first amplitude and the second amplitude of the drive system under different load indices; the load ratio compensation map further comprises a compensation relationship between the electrical temperature index and the healthy baseline ratio; in the online monitoring state, further comprising: real-time obtaining a current electrical temperature index; and the step of determining the expected healthy ratio is specifically based on the load ratio compensation map, and simultaneously according to the current load index and the current electrical temperature index, determining the expected healthy ratio.

6. The method of claim 5, wherein the method further comprises: The electrical temperature index is a parameter determined by measuring the motor winding resistance value.

Images