A performance evaluation method and system of an industrial internet connector

Abstract

The application relates to the technical field of industrial automation control and system state monitoring, and discloses a performance evaluation method and system of an industrial internet connector, which comprises the following steps: the method uses the common-mode noise generated in the power supply loop of the device itself as a diagnostic probe in reverse, establishes the intrinsic harmonic spectrum series relationship representing the stability of the noise source itself when establishing the initial frequency spectrum fingerprint baseline, and in the subsequent periodic evaluation, firstly, the spectrum series relationship is pre-checked, and then, based on the drift index of the current frequency spectrum fingerprint relative to the baseline and the change slope in the continuous time window, the performance evolution of the connector is evaluated. The application shifts the focus of evaluation from the apparent correctness of data transmission to the gradual change trend of the physical structure of the bearing connection, and can identify that the irreversible physical state degradation process of the connector is occurring.

Description

Technical Field

[0001] This invention relates to a performance evaluation method and system for industrial internet connectors, belonging to the field of industrial automation control and system status monitoring technology. Background Technology

[0002] In modern intelligent manufacturing and large-scale automation systems, modern communication protocols based on the Ethernet physical layer, such as EtherNet / IP or Profinet, have become the key data arteries connecting controllers, sensors, actuators, and information systems. Their stable and reliable operation is the foundation for ensuring the efficiency and safety of the entire production process. To ensure the integrity of data transmission, these modern communication protocols have built-in error correction and automatic retransmission mechanisms. When there is transient interference or slight signal distortion in the link, the redundancy operation at the protocol level can maintain the final success of data interaction. This constitutes the cornerstone of the high reliability of current industrial networks.

[0003] However, this high robustness at the protocol level creates a deep-seated problem when faced with the gradual degradation of the connector's own physical performance: connectors in industrial environments, subjected to long-term micro-vibration, temperature cycling, or environmental corrosion, experience gradual changes in their internal contact interfaces, such as stress relaxation of the elastic pins or slight oxidation of the contact surface. In the early stages of this degradation, although the physical connection is no longer ideal, it has not yet reached the point of complete failure, and its impact on data frames is often sporadic and weak. The protocol's retransmission mechanism at this time will sacrifice the minor... Compensation for these damaged data frames is achieved at the cost of small transmission delays. As a result, from the perspective of upper-layer applications or network monitoring software, the data flow still appears normal, and key indicators such as packet loss rate and bit error rate show no obvious changes. The system presents a false impression of health. This sub-healthy state of the physical layer, masked by the protocol's self-healing capabilities, constitutes a typical silent risk in system operation and maintenance. It prevents maintenance personnel from intervening during the optimal preventive maintenance window until the degradation of physical performance accumulates to the point that it exceeds the protocol's error correction limit, causing frequent device disconnections or communication interruptions, at which point the problem erupts.

[0004] Faced with this challenge, some conventional approaches in this field also have inherent limitations. First, relying on statistical indicators at the data level for judgment, as mentioned above, is a lagging method and cannot achieve early warning. Second, during the design or offline maintenance phase, precision instruments such as vector network analyzers are used to test the RF parameters of connectors. Although this method is accurate, its high equipment cost, dependence on professional personnel, and requirement to interrupt system operation determine that it cannot serve as an online and system-wide daily monitoring method. Specifically, existing technologies mainly have the following shortcomings: 1. Existing online monitoring methods generally focus on the transmission results at the protocol layer or data layer, lacking a way to directly penetrate to the physical layer health status of the connector itself for assessment; 2. For the gradual and irreversible performance degradation process of connectors caused by physical reasons, existing methods cannot provide trend-based early warnings and usually only respond passively after the fault has occurred. Therefore, the technical problem to be solved by this invention is how to build an online evaluation method that does not interfere with normal data communication on the basis of using general-purpose low-cost hardware, so that it can bypass the health spoofing of communication protocols and directly perceive and quantify the early progressive performance degradation trend of industrial connectors at the physical level. Summary of the Invention

[0005] This invention provides a performance evaluation method and system for industrial internet connectors. Its main purpose is to solve the problem that existing technologies cannot identify and track the progressive degradation trend of the connector's physical performance in advance when the data communication appears to be normal.

[0006] To achieve the above objectives, this invention provides a performance evaluation method for industrial internet connectors. This method is applied to evaluate the evolution of the connector's physical performance during a stage where the data communication carried by the connector appears normal due to protocol layer error correction mechanisms. The method includes:

[0007] Step a: On the downstream side of the connector, the endogenous power common-mode noise signal generated by its own power supply circuit on the device ground line is collected, and based on the spectral characteristics of the endogenous power common-mode noise signal, a baseline noise fingerprint characterizing the initial health state of the connector and a baseline harmonic spectrum relationship characterizing the self-stability of the power supply circuit corresponding to the endogenous power common-mode noise signal are determined.

[0008] Step b: During equipment operation, periodically collect the current common-mode noise signal on the ground wire, and determine a current noise fingerprint and a current harmonic spectrum relationship;

[0009] Step c: Before performing connector performance evaluation, a mandatory source credibility determination is performed, that is, it is determined whether the current harmonic spectrum relationship is consistent with the baseline harmonic spectrum relationship. If they are consistent, it is confirmed that the currently acquired common-mode noise signal is valid as an evaluation probe and subsequent steps are triggered.

[0010] Step d: Based on the comparison between the current noisy fingerprint and the baseline noisy fingerprint, calculate a current fingerprint drift index that quantifies the degree of deviation between the two.

[0011] Step e: Based on a series of current fingerprint drift indices obtained within a preset time window, determine an evaluation slope that characterizes the trend of the series of indices, and evaluate the evolution of the physical performance of the connector based on the evaluation slope.

[0012] Preferably, step a, determining the baseline noise fingerprint characterizing the initial health state of the connector, includes: performing a fast Fourier transform on the endogenous power supply common-mode noise signal to obtain the baseline power spectrum; and automatically identifying multiple stable harmonic frequency points whose power spectral density values ​​are among the top of a preset peak order in the baseline power spectrum, and constructing the baseline noise fingerprint by combining all frequency points and their corresponding power spectral density values. Step d, calculating the current fingerprint drift index, which quantifies the degree of deviation between the two, is determined by the following formula: ,in, This represents the current fingerprint drift index. The number of frequency points contained in the baseline noise fingerprint. The first in the baseline noisy fingerprint The power spectral density values ​​corresponding to each frequency point The first in the current noisy fingerprint The power spectral density value corresponding to each frequency point.

[0013] Preferably, step a, determining the baseline harmonic spectrum relationship characterizing the inherent stability of the power supply circuit corresponding to the endogenous power supply common-mode noise signal, includes: selecting at least two stable harmonic peaks of different frequencies in the baseline power spectrum, and calculating the ratio between the power spectral density values ​​of these two stable harmonic peaks, using this ratio as the baseline harmonic spectrum relationship between the corresponding stable harmonic peaks; step c, determining whether the current harmonic spectrum relationship matches the baseline harmonic spectrum relationship, includes: when the deviation between the current harmonic spectrum relationship and the baseline harmonic spectrum relationship is less than a preset source stability threshold, it is determined that the current harmonic spectrum relationship matches the baseline harmonic spectrum relationship.

[0014] Preferably, step e, which involves determining an evaluation slope characterizing the trend of a series of current fingerprint drift indices obtained within a preset time window, and evaluating the evolution of the connector's physical performance based on the evaluation slope, includes: performing linear regression analysis on a series of current fingerprint drift indices obtained within the preset time window to obtain an evaluation slope characterizing the trend of a series of indices; determining that the connector's physical performance is in a deterioration trend when the evaluation slope is consistently positive and exceeds a preset degradation trend slope threshold; and determining that the connector has suffered a transient event impact rather than performance degradation when the evaluation slope is close to zero but there are single or a few numerical abrupt changes in the series of current fingerprint drift indices.

[0015] Preferably, the method further includes the step of determining the degradation mode of physical performance, the steps of which include: in step a, calculating and recording a baseline higher-order statistical moment based on the time-domain waveform of the endogenous power supply common-mode noise signal; in step b, periodically calculating the current higher-order statistical moment of the current common-mode noise signal time-domain waveform; and, when the evaluation result of step e indicates that the physical performance of the connector is in a degradation trend, distinguishing the degradation mode of physical performance based on the comparison between the current higher-order statistical moment and the baseline higher-order statistical moment; wherein, the baseline higher-order statistical moment is the baseline kurtosis value, and the current higher-order statistical moment is the current kurtosis value.

[0016] Preferably, the step of distinguishing the degradation mode of physical performance based on the comparison between the current higher-order statistical moments and the baseline higher-order statistical moments includes: when the current kurtosis value does not increase significantly compared with the baseline kurtosis value, the degradation mode is determined to be progressive physical decay; when the current kurtosis value increases significantly compared with the baseline kurtosis value, the degradation mode is determined to be discontinuous contact deterioration caused by micro-arcsing or mechanical vibration of the contact surface.

[0017] Preferably, the method further includes a step of compensating for temperature changes, the step of which includes: in step a, determining a baseline noise floor power based on the noise floor band without harmonic peaks in the spectrum of the endogenous power supply common-mode noise signal; and, before performing step d, further including: determining a current noise floor power based on the current common-mode noise signal, and performing temperature compensation on the fingerprint drift index calculated in step d according to the difference between the current noise floor power and the baseline noise floor power, so as to eliminate the influence of ambient temperature changes on the evaluation results.

[0018] Preferably, the method further includes a step of identifying intermittent contact defects, the step of which includes: during the power-on startup of the device connected to the connector, continuously acquiring a series of instantaneous noise fingerprints at a frequency higher than that of periodically acquiring the current common-mode noise signal on the ground wire and calculating the corresponding instantaneous fingerprint drift index to form a fingerprint drift index sequence; and calculating the standard deviation of the fingerprint drift index sequence itself, and when the standard deviation exceeds a preset connection stability threshold, determining that the connector has a potential intermittent contact defect caused by a loose connection.

[0019] Preferably, the method further includes a step of defining the noise source, which includes: simultaneously acquiring the differential mode noise signal between the data signal pairs carried by the connector while acquiring the current common mode noise signal in step b; when it is determined in step e that the connector has been subjected to a transient event, further comparing the energy changes of the current common mode noise signal and the differential mode noise signal; if the energy of the current common mode noise signal and the energy of the differential mode noise signal increase significantly in the same proportion, then the root cause of the transient event is defined as external electromagnetic interference, thereby distinguishing between external environmental problems and connector-related problems.

[0020] A performance evaluation system for industrial internet connectors, the system comprising:

[0021] A baseline determination module is configured to: acquire the intrinsic power common-mode noise signal generated by the power supply circuit on the device ground line at the downstream side of the connector; and determine a baseline noise fingerprint characterizing the initial reference state of the connector and a baseline harmonic spectrum relationship characterizing the characteristics of the power supply circuit corresponding to the intrinsic power common-mode noise signal based on the spectrum of the intrinsic power common-mode noise signal.

[0022] A periodic acquisition module is configured to periodically acquire the current common-mode noise signal on the ground wire during device operation to obtain a current noise fingerprint and a current harmonic spectrum relationship.

[0023] A source verification module receives the baseline harmonic spectrum relationship and the current harmonic spectrum relationship, and is configured to: compare and verify the two before performing connector performance evaluation, and output an enable signal only when the deviation between the current harmonic spectrum relationship and the baseline harmonic spectrum relationship is within a preset source stability range;

[0024] A drift index calculation module receives a baseline noise fingerprint and a current noise fingerprint, and in response to an enable signal, is configured to compare the current noise fingerprint with the baseline noise fingerprint to calculate a current fingerprint drift index that quantifies the degree of deviation between the two.

[0025] A performance evaluation module receives a series of current fingerprint drift indices output by the drift index calculation module within a preset time window, and is configured to: calculate an evaluation slope characterizing the rate of change of the series of indices, and evaluate the evolution state of the physical performance of the connector based on the evaluation slope.

[0026] Compared with the prior art, the beneficial effects of the present invention are:

[0027] 1. This invention utilizes the inherent, persistent common-mode noise in the device's own power supply circuit as a conduction medium. By acquiring the spectral characteristics of this noise signal downstream of the connector, an initial baseline noise fingerprint characterizing the connector's physical state is established. In subsequent operation, this method does not isolately judge whether the spectral characteristics of a single measurement deviate, but continuously records its current fingerprint drift index relative to the initial baseline noise fingerprint. Based on the change in the evaluation slope of this index within a continuous preset time window, the evolution of the connector's physical performance is evaluated. This approach shifts the focus of evaluation from the apparent correctness of data transmission to the gradual change trend of the physical structure carrying the connection itself. Therefore, the error correction and retransmission capabilities of modern communication protocols mask early signs of physical degradation, allowing this method to identify irreversible physical degradation processes occurring in the connector even when the data is still normal.

[0028] 2. In the process of determining the initial baseline noise fingerprint, this invention also determines the intrinsic power ratio between the main harmonic components in the noise spectrum. During each subsequent periodic evaluation, the method first checks whether the currently acquired noise spectrum maintains this intrinsic ratio. Only when this ratio is stable, proving that the characteristics of the noise source itself as the probe signal have not changed, will the method continue to perform subsequent drift index calculation and trend evaluation. This preliminary verification step based on the signal's own characteristics prevents the entire evaluation system from making incorrect judgments about the connector status due to changes in the noise source itself caused by external factors such as drastic changes in equipment load. This strictly limits the validity of the evaluation conclusion to the situation caused by changes in the physical path of the connector itself.

[0029] 3. The method of this invention, while performing frequency domain analysis on noise signals to track long-term performance evolution trends, also simultaneously characterizes the probability distribution of the time-domain waveform of the same acquired endogenous power supply common-mode noise signal, paying particular attention to the changes in its higher-order statistical moments. Once the performance degradation trend revealed by the frequency domain analysis is established, the system immediately compares the current higher-order statistical moments of the current waveform with the baseline higher-order statistical moments in the initial healthy state. A smooth waveform distribution evolution corresponds to a gradual physical decay process such as material corrosion or fatigue; while a waveform distribution superimposed with a large number of instantaneous spikes and exhibiting a significant non-Gaussian morphology points to discontinuous mechanical contact deterioration problems such as micro-arcing caused by vibration at the contact surface. By combining the trend judgment in the frequency domain with the morphology judgment in the time domain, this method provides more detailed information about its physical causes, providing a basis for judging the degradation mode for subsequent maintenance decisions.

[0030] 4. This invention also provides a mechanism for identifying intermittent contact defects. This mechanism is activated the instantaneously upon each power-on of the device, utilizing the instantaneous electrical and thermodynamic pressure applied to the connector contact points by the power-on surge current. Within this brief, unsteady startup window, it continuously captures the instantaneous changes in noise fingerprints at a frequency far higher than conventional monitoring. The method does not focus on the average offset of this series of instantaneous noise fingerprints, but rather calculates the dispersion of the sequence itself. For a physically stable connector, its fingerprint drift index sequence remains smooth under this impact; while for a connector with potential for loose connections, its sequence will exhibit violent, irregular jumps, leading to an abnormally increased dispersion. By capturing the transient response characteristics during this power-on process, occasional millisecond-level contact defects that are difficult to detect in steady-state operation but may lead to the loss of critical instructions can be identified. Attached Figure Description

[0031] Figure 1 This is a schematic diagram of the baseline establishment process for a performance evaluation method for an industrial internet connector according to the present invention.

[0032] Figure 2 This is a functional module architecture diagram of a performance evaluation system for industrial internet connectors according to the present invention;

[0033] Figure 3 This is a schematic diagram of the data processing flow for a performance evaluation method for an industrial internet connector according to the present invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions of the present invention will be described in detail below. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0035] The performance evaluation method and system for industrial internet connectors provided by this invention can be applied to a specific industrial scenario, such as an automated guided vehicle (AGV) scheduling system in an automated warehouse, to evaluate the performance status of industrial Ethernet connectors connected to the backbone network. In such scenarios, minor degradation of the physical contact state of the connector, although initially compensated for by the error correction mechanism of the communication protocol, may have a cumulative effect that could lead to the instantaneous loss of scheduling instructions, causing process interruption or safety risks. In downstream devices of the connector, such as the controller of the AGV, the operation of its power supply unit generates switching power supply noise with stable spectral characteristics. This noise is transmitted as a common-mode signal in the system's ground network. This signal is used as a diagnostic probe signal in this method. When the workload of the device itself changes, such as when the AGV's drive motor switches from no-load to load, this noise is detected. When the device is fully loaded, the noise characteristics of its power supply unit may also change. To distinguish the signal changes caused by the degradation of the connector's physical path from the drift of the diagnostic probe signal source itself, this method performs a complexation procedure that includes calibration of the signal source's own characteristics when establishing the initial health baseline of the connector. In the initial stage of the device's first power-on or confirmation of health, the baseline determination module, located downstream of the connector, acquires the time-domain waveform of the common-mode noise signal generated by its own power supply circuit on the device's ground wire. The baseline determination module performs a fast Fourier transform on the acquired time-domain waveform to obtain the baseline power spectrum. In this baseline power spectrum, the module automatically identifies multiple stable harmonic frequency points with power spectral density values ​​at the top of the preset peak ranking, such as the top 5 harmonic peaks in power, using a peak search algorithm. These frequency points and their corresponding power spectral density values ​​are then associated with this algorithm. (in ,and Together, these constitute a baseline noise fingerprint characterizing the initial health state of the connector. Simultaneously, to establish a stability benchmark for the noise source itself, the baseline determination module selects at least two stable harmonic peaks of different frequencies from these identified stable harmonics, for example, selecting the fundamental frequency peak (…). ) and second harmonic peak value ( ), and calculate the ratio between the two peak power spectral densities, i.e. The ratio is recorded as a baseline harmonic spectrum relationship, which characterizes the inherent proportional features of the specific power source noise in the spectral structure.

[0036] During normal operation, the periodic acquisition module is woken up at a preset frequency, for example, once per hour, to repeat the acquisition and analysis process of the baseline determination phase. That is, it acquires the current ground common-mode noise signal and uses FFT to determine a spectrum containing the current harmonic frequency points and their power spectral density values. The current noise fingerprint, and a current harmonic spectrum relationship calculated based on the same rules. Before performing connector performance calculations, the source verification module performs a source credibility determination, which determines the current harmonic spectrum relationship. Relationship with stored baseline harmonic spectrum The system compares the two signals and calculates the deviation. Only when the deviation is less than a preset source stability threshold, for example, less than 5%, does the source verification module determine that the currently acquired common-mode noise signal is valid as an evaluation probe and output an enable signal to the drift index calculation module. If the deviation exceeds the threshold, the system determines that the source characteristics of the current measurement cycle are unstable, and the system will abandon the current calculation and wait for the next evaluation cycle. This procedure aims to suppress the judgment of the connector status caused by the change of the noise source itself due to factors such as drastic changes in equipment load.

[0037] Upon receiving the enable signal from the source verification module, the drift index calculation module is activated. This module compares the valid current noise fingerprint with the baseline noise fingerprint and calculates a current fingerprint drift index that quantifies the degree of deviation between the two. The index is determined by the following formula: ,in, This represents the current fingerprint drift index. The number of frequency points contained in the baseline noise fingerprint. The first in the baseline noisy fingerprint The power spectral density value corresponding to each frequency point, and The first in the current noisy fingerprint The power spectral density values ​​corresponding to each frequency point are then received by the performance evaluation module within a preset time window, for example, the past 24 hours, from the drift index calculation module. Values, used to evaluate the evolution of connector performance, this module considers this series of values. Linear regression analysis is performed on the values ​​to obtain an evaluation slope that characterizes the trend of the series of indices. When the evaluation slope is consistently positive and exceeds a preset degradation trend slope threshold, the system determines that the physical performance of the connector is in a degradation trend. When the evaluation slope is close to zero, but there are one or a few numerical abrupt changes in the current fingerprint drift index of the series, it is determined that the connector has suffered a transient event impact, rather than performance degradation.

[0038] Once the performance evaluation module determines that the connector is in a progressive degradation trend, this method also includes parallel analysis of the time-domain waveform distribution of the noise signal to further distinguish the physical modes of degradation. While the baseline determination module establishes the baseline noise fingerprint, a higher-order statistical moment of the baseline is calculated and recorded for the time-domain waveform of the common-mode noise signal acquired in the same segment, specifically the baseline kurtosis value. Correspondingly, the periodic acquisition module also calculates the current kurtosis value of the current time-domain waveform in parallel each time it acquires the current noise fingerprint. When the performance evaluation module determines that the physical performance of the connector is deteriorating, the system immediately compares the current kurtosis value. Compared with baseline kurtosis value If the current kurtosis value Compared with baseline kurtosis value If no significant increase has been observed, the degradation mode is determined to be a gradual physical degradation caused by factors such as material corrosion or elastic fatigue; if the current kurtosis value... Compared to baseline kurtosis If the increase is significant, the degradation mode is determined to be discontinuous contact deterioration caused by factors such as micro-arcsing at the contact surface or mechanical vibration.

[0039] For intermittent contact problems caused by loose connections, which are short-lived, this method also includes a screening mechanism activated at specific times. This mechanism utilizes the instantaneous electrical and thermodynamic stress applied to the connector contact points by the surge current during device power-on as a test of connection stability. During the power-on of the device connected to the connector, the system continuously acquires a series of instantaneous noise fingerprints at a frequency higher than the frequency of periodically acquiring the current common-mode noise signal on the ground wire, and calculates the corresponding instantaneous fingerprint drift index, thus forming a fingerprint drift index sequence. Subsequently, the system calculates the dispersion of the fingerprint drift index sequence itself, such as the standard deviation. A connector with a physically stable connection will have a low dispersion of its fingerprint drift index sequence under this impact; while a connector with a potential loose connection will exhibit irregular fluctuations in its sequence, leading to increased dispersion. When the calculated dispersion exceeds a preset connection stability threshold, the system determines that the connector has a potential intermittent contact problem caused by a loose connection. To ensure that the evaluation results are within a wide range... To ensure reliability in temperature-controlled operating scenarios, the system utilizes the noise floor in the noise spectrum as a measure of temperature change to achieve temperature compensation. When determining the baseline noise fingerprint, a baseline noise floor power is also determined based on the noise floor frequency band without harmonic peaks in the spectrum. Before calculating the current fingerprint drift index, the system first determines a current noise floor power and compensates for the calculated current fingerprint drift index based on the difference between the current noise floor power and the baseline noise floor power to reduce the impact of ambient temperature changes on the evaluation results. In addition, to distinguish between external electromagnetic interference and connector-related issues, the system simultaneously acquires the differential-mode noise signal between the data signal pairs carried by the connector while acquiring the current common-mode noise signal. When the connector is determined to be subjected to a transient event, the system further compares the energy changes of the common-mode noise signal and the differential-mode noise signal. If both energy increases significantly and proportionally, the root cause of the transient event is identified as external electromagnetic interference, which helps to distinguish between external environmental issues and connector-related issues.

[0040] Example 1: In a continuously operating automated storage and retrieval system (AS / RS), an AGV (Automated Guided Vehicle) responsible for transferring high-value goods experiences a slow, gradual decrease in the contact pressure of its internal pins due to prolonged exposure to minor mechanical vibrations at the track joints. Simultaneously, the AGV's operational commands require frequent switching between empty and full-load states, causing drastic changes in the load and power consumption of its motor drive system. During system operation, when the AGV performs a picking operation from empty to full load, the load on its power supply unit increases, altering the spectral characteristics of its common-mode noise signal. At this point, the periodic acquisition module acquires the current noise signal, and the source verification module calculates the current harmonic spectrum. ,Should Relationship between the value and the baseline harmonic spectrum initially stored in the system After comparison, the deviation exceeded the preset source stability threshold. Based on this, the source verification module determined that the signal acquired in this instance did not meet the prerequisites for evaluation and suppressed the subsequent current fingerprint drift index. The system only performs subsequent drift index calculations when the harmonic spectrum of the signal source is consistent with the baseline, thereby distinguishing signal fluctuations caused by changes in equipment operating conditions from changes in the physical path of the connector itself.

[0041] When the AGV is in a relatively stable load phase, such as cruising or returning empty, the harmonic spectrum of its common-mode noise source stabilizes. The source verification module confirms the signal validity for subsequent acquisition cycles, allowing the drift index calculation module and performance evaluation module to operate continuously. Over several days of operation, the performance evaluation module records a series of current fingerprint drift indices. After linear regression analysis, the value showed a consistently positive evaluation slope, indicating that the physical conduction characteristics of the connector were undergoing irreversible degradation. This trend-based analysis shifted the focus of the evaluation from offline RF parameter measurements to online monitoring of the changing trends of physical path conduction characteristics reflected by the device's own circuit signals. Simultaneously with the performance evaluation module determining the degradation trend, the system performed high-order statistical moment analysis on the time-domain noise waveforms collected during the same period. The analysis results showed that the current kurtosis value... Compared to the initially stored baseline kurtosis value The increase in the current fingerprint drift index trend indicates progressive degradation of the connector, while the increase in kurtosis further indicates that the physical mode of degradation is related to discontinuous mechanical contact problems. Combining the analysis results of these two dimensions, the maintenance instructions output by the system can simultaneously include the state evolution trend of the connector and the physical mode of degradation information. Based on the maintenance instructions, the maintenance personnel tightened the designated connector of the AGV during a planned maintenance window, avoiding the potential contact failure caused by minor vibrations, and thus avoiding an unplanned production downtime event that could have occurred in subsequent operation due to a momentary connection failure.

[0042] Example 2: To verify the ability of the method of the present invention to track and recognize patterns in the progressive physical degradation process of connectors, an accelerated aging test platform was built. The purpose was to objectively quantify the correlation between the key indicators output by the connector performance evaluation method and the evolution of the connector's physical state under controlled mechanical vibration and transient shock conditions. The test platform consisted of an industrial PC and a PLC, connected by a pair of industrial Ethernet connectors and cables for continuous data communication. The connector assembly was fixed on a six-axis vibration table, which applied continuous random vibration conforming to industrial equipment environmental standards to simulate long-term mechanical fatigue effects. A data acquisition device was used to collect the common-mode noise signal generated by the power supply circuit of the PLC device downstream of the connector. The acquisition interval was set to 1 hour. This parameter was determined after weighing the timeliness of data updates against the data processing load of the system during long-term operation. This interval provides sufficient time resolution for subsequent trend analysis. The test was divided into two phases: initial baseline establishment and accelerated aging test, lasting a total of 500 hours. In the initial phase (T=0h), the vibration table was not working, and the system operated in a stable state. According to the procedures in the specific implementation method, the baseline noise fingerprint, baseline harmonic spectrum relationship, and baseline kurtosis value were determined. The value is 3.12. During the accelerated aging test phase (T=1h to T=500h), the vibration table is started, and the system records the current fingerprint drift index once per hour. Compared with the current kurtosis value At the 300th hour, the test system applied a transient mechanical impact to the connector for 100ms to simulate the condition of an accidental collision. Throughout the test, the Ethernet communication packet loss rate of the system remained at a normal level.

[0043] During the experiment, the current fingerprint drift index The changes exhibit phased characteristics, especially in the initial 100 hours (T=100h). The value was only 0.04; subsequently, the value began to show an almost linear increase, reaching 0.11 at the 200th hour (T=200h), and 0.39 at the 500th hour (T=500h) after the test ended. The corresponding evaluation slope also steadily increased from the initial 0.0004 FDI / h to 0.0007 FDI / h. This trend corresponds to the physical process of stress relaxation and micro-wear on the internal metal contact surface of the connector under continuous vibration. Meanwhile, the current kurtosis value... The value remained near the baseline for most of the testing period, for example, at 200 hours it was 3.21. However, when a transient mechanical shock was applied at 300 hours, the current kurtosis value... The value instantaneously increased to 15.8, and then dropped back to 3.25 in the next sampling period (T=301h) after the impact ended. This phenomenon indicates that the kurtosis index can respond to discontinuous mechanical events and correlate with... The results show that by tracking the current fingerprint drift index within a continuous time window, the gradual degradation trend can be distinguished. The slope of the change can quantify the gradual degradation of the connector's physical performance caused by mechanical vibration. Simultaneously, the current kurtosis value of the time-domain waveform can be monitored in parallel. This method can distinguish this gradual degradation trend from discontinuous events caused by external transient shocks. Under the condition that the data communication layer appears normal, this method provides a way to conduct early trend assessment and pattern recognition of the physical layer health status of the connector.

[0044] Example 3: This example combines Figures 1 to 3 This paper describes the implementation of a performance evaluation method and system for an industrial internet connector, such as... Figure 1 As shown, the process is initiated by the operator issuing a baseline establishment command. After receiving the command, the evaluation system activates the baseline determination module. This module first requests the data acquisition unit to collect common-mode noise from the device ground of the connector equipment. After the data acquisition unit obtains the time-domain noise signal and sends it back as raw noise data, the baseline determination module sequentially performs Fast Fourier Transform (FFT), identifies the main harmonic frequency points, calculates the baseline noise fingerprint, determines the baseline harmonic spectrum relationship, and calculates the baseline kurtosis value. Finally, the baseline parameters, including the noise fingerprint, harmonic relationship, and kurtosis value, are stored, and the evaluation system returns a baseline establishment completion signal to the operator.

[0045] like Figure 2 As shown, Figure 2 Its core is built around three major functional blocks. At the data input end, the inherent power supply common-mode noise signal generated by the device itself first enters the data acquisition and preprocessing block. The periodic acquisition module, source verification module, and drift index calculation module in this block work together and process the raw signal into an FDI data stream based on the baseline data provided by the benchmark establishment and management block. The FDI data stream is then sent to the core evaluation and diagnosis block. The performance evaluation module, degradation physical mode analysis module, and intermittent contact failure identification module in this block perform comprehensive analysis based on the input data stream and finally output the performance evolution evaluation result. The benchmark for the entire evaluation process is established and maintained by the baseline determination module in the benchmark establishment and management block.

[0046] like Figure 3As shown, the process begins with the acquisition and transformation of the common-mode noise signal at the device ground wire (1.0). The resulting current power spectrum is sent to the evaluation parameter calculation stage (2.0) and compared with the baseline data (fingerprint, spectrum) in the composite benchmark database to generate the fingerprint drift index and source verification result. At the same time, the resulting current time-domain waveform is sent to the time-domain feature analysis stage (3.0) and compared with the baseline kurtosis value in the database to analyze the kurtosis change trend. Finally, the fingerprint drift index and source verification result output by the evaluation parameter calculation stage (2.0), and the kurtosis change trend output by the time-domain feature analysis stage (3.0) are integrated into the comprehensive evaluation and decision-making stage (4.0) to form an in-depth evaluation report, which is then submitted to the operation and maintenance personnel, constituting a complete closed-loop monitoring and decision-making process.

[0047] Example 4: In an onboard control system operating in an environment where the temperature fluctuates between -20°C and +85°C, the industrial Ethernet connector used must withstand measurement interference caused by benign drift of the entire electrical system parameters due to temperature changes. Furthermore, it must distinguish between transient contact failures with potential degradation due to vibration and normal electrical fluctuations during vehicle startup and operation. Therefore, before deploying the evaluation method of this invention to this specific system, a standardized engineering calibration procedure is used to determine the system's unique temperature compensation coefficient and the threshold for judging intermittent contact failures. This calibration procedure is performed in a controlled experimental environment, placing the connector and its downstream equipment system in a programmable temperature and humidity test chamber, ensuring the connector itself is in an initial healthy state with good physical contact. To determine the temperature compensation coefficient, while the equipment is running stably at a preset baseline workload, the temperature and humidity test chamber is programmed to perform multiple slow reciprocating cycles within the temperature range of -20°C to +85°C. During this process, the system continuously collects and records two data pairs at high frequency: the noise floor power, which serves as a temperature characterization quantity. And the current fingerprint drift index caused solely by temperature changes in the absence of physical degradation. After collecting a dataset covering the entire temperature range, the system performs linear regression analysis on the data to establish... Follow A linear relationship model for the changing relationship was established, and the slope of this model was determined to be the temperature compensation coefficient specific to this system. And in subsequent actual operation, it was used to analyze the current fingerprint drift index. Compensation is performed to reduce the impact of temperature changes on the evaluation results.

[0048] To determine the connection stability threshold for identifying intermittent contact defects, the system was subjected to 100 consecutive power-off cycles at room temperature while the connector was still in a healthy state. At the initial stage of each power-on startup, the system continuously acquired a series of instantaneous fingerprint drift indices at a high sampling frequency and calculated the dispersion of the sequence. This yielded a statistical sample distribution of 100 discrete values ​​characterizing the response of a healthy connector under power-on shock. After calculating the mean and standard deviation of this distribution, the connection stability threshold was set to the mean of the distribution plus six times the standard deviation. This threshold was stored and used as a benchmark for judging whether there are abnormal fluctuations in the connector's power-on response during subsequent device startups.

[0049] Example 5: On a production line where the evaluation method of this invention has been deployed, when a connector or its downstream device unit that has been monitored for a long time is replaced due to routine maintenance or failure, the system provides a baseline update procedure. After the hardware replacement is completed, a baseline update command can be sent to the evaluation system through a service interface. Upon receiving the command, the system clears the previously stored baseline noise fingerprint and baseline harmonic spectrum relationship. and baseline kurtosis value An initial baseline establishment process is then executed, and under the current new hardware configuration, a completely new set of baseline data is re-acquired and determined as a new reference benchmark for subsequent periodic evaluations.

[0050] During the long-term operation of the system, if the source verification module determines the current harmonic spectrum relationship in multiple consecutive monitoring cycles of a preset number, Relationship with baseline harmonic spectrum If the deviation continues to exceed the source stability threshold, the system will identify this phenomenon as a non-transient change in the characteristics of the noise source. At this time, the system will mark the current baseline as questionable and, in the next scheduled low-load operation window of the equipment, initiate the baseline update procedure to archive the old baseline data and establish a new baseline that matches the current noise source characteristics.

[0051] Example 6: Before applying the evaluation method of this invention to a specific model of industrial control equipment, in order to establish an evaluation benchmark and decision threshold applicable to that model of equipment, an offline standardized model construction and parameter calibration procedure needs to be executed. The input of this procedure is a set of samples of the same model of equipment that have been confirmed as new and in good connection status through independent physical testing, and its output is a standardized evaluation model applicable to all equipment of that model. The execution of this procedure first involves constructing a reference baseline noise fingerprint that can characterize the health status of the equipment of that model, and a set of stable harmonic frequency points for constructing the fingerprint. In a controlled environment with standard electromagnetic shielding and temperature and humidity, the aforementioned equipment samples are used... For each sample device, common-mode noise signal was collected and its power spectrum was calculated under a uniform baseline workload. By superimposing and analyzing the power spectra of all samples, several harmonic frequencies with high frequency consistency and high average power spectral density were selected in the entire sample space. This set of statistically verified frequency points was used as the standard fingerprint frequency set applicable to this model of equipment. Subsequently, the average power spectral density value of this set of frequency points in all healthy samples was calculated and recorded to construct a statistically significant prototype baseline noise fingerprint. This prototype baseline was fixed in the evaluation model and used to perform initial state benchmark verification for subsequent production of the same model of equipment.

[0052] The procedure also includes the calibration of a degradation trend slope threshold, i.e., determining a current fingerprint drift index that can effectively trigger early warning. The slope of change is determined by selecting a portion of the aforementioned equipment samples for accelerated aging testing. While applying continuous mechanical vibration or thermal cycling stress conforming to relevant industry standards to this portion of the samples, in addition to calculating the evaluation slope using the method of this invention, a parameter that directly reflects the physical state of the connector, such as the contact resistance or high-frequency insertion loss of the connector, is simultaneously monitored using an independent high-precision measuring instrument. When the degree of degradation of this independent physical parameter reaches a pre-set maintenance intervention point requiring preventive maintenance, the evaluation slope value calculated by the method of this invention at this moment is recorded. By statistically averaging the evaluation slope values ​​corresponding to this maintenance intervention point for multiple samples and combining them with a preset safety margin, a degradation trend slope threshold applicable to this model of equipment is finally determined. This threshold is also embedded in the evaluation model.

[0053] 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.

[0054] 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 performance evaluation method for an industrial internet connector, applied to evaluate the evolution of the connector's physical performance during a stage where the data communication carried by the connector appears normal due to protocol layer error correction mechanisms, characterized in that... The method includes: Step a: On the downstream side of the connector, the endogenous power common-mode noise signal generated by its own power supply circuit on the device ground line is collected, and based on the spectral characteristics of the endogenous power common-mode noise signal, a baseline noise fingerprint characterizing the initial health state of the connector and a baseline harmonic spectrum relationship characterizing the self-stability of the power supply circuit corresponding to the endogenous power common-mode noise signal are determined. Step b: During equipment operation, periodically collect the current common-mode noise signal on the ground wire, and determine a current noise fingerprint and a current harmonic spectrum relationship; Step c: Before performing connector performance evaluation, a mandatory source credibility determination is performed, that is, it is determined whether the current harmonic spectrum relationship is consistent with the baseline harmonic spectrum relationship. If they are consistent, it is confirmed that the currently acquired common-mode noise signal is valid as an evaluation probe and subsequent steps are triggered. Step d: Based on the comparison between the current noisy fingerprint and the baseline noisy fingerprint, calculate a current fingerprint drift index that quantifies the degree of deviation between the two. Step e: Based on a series of current fingerprint drift indices obtained within a preset time window, determine an evaluation slope that characterizes the trend of the series of indices, and evaluate the evolution of the physical performance of the connector based on the evaluation slope. The step a, determining the baseline noise fingerprint characterizing the initial health state of the connector, includes: performing a fast Fourier transform on the endogenous power supply common-mode noise signal to obtain the baseline power spectrum; and automatically identifying multiple stable harmonic frequency points whose power spectral density values ​​are among the top of a preset peak order in the baseline power spectrum, and using all frequency points and their corresponding power spectral density values ​​to constitute the baseline noise fingerprint; the current fingerprint drift index, which quantifies the degree of deviation between the two, is calculated in step d using the following formula: ,in, This represents the current fingerprint drift index. The number of frequency points contained in the baseline noise fingerprint. The first in the baseline noisy fingerprint The power spectral density values ​​corresponding to each frequency point The first in the current noisy fingerprint The power spectral density value corresponding to each frequency point.

2. The performance evaluation method for an industrial internet connector according to claim 1, characterized in that, The step a of determining the baseline harmonic spectrum relationship characterizing the inherent stability of the power supply circuit corresponding to the common-mode noise signal of the endogenous power source includes: selecting at least two stable harmonic peaks of different frequencies in the baseline power spectrum, and calculating the ratio between the power spectral density values ​​of the two stable harmonic peaks, and using this ratio as the baseline harmonic spectrum relationship between the corresponding stable harmonic peaks; the step c of determining whether the current harmonic spectrum relationship matches the baseline harmonic spectrum relationship includes: when the deviation between the current harmonic spectrum relationship and the baseline harmonic spectrum relationship is less than a preset source stability threshold, it is determined that the current harmonic spectrum relationship matches the baseline harmonic spectrum relationship.

3. The performance evaluation method for an industrial internet connector according to claim 1, characterized in that, Step e, which involves determining the evaluation slope characterizing the trend of a series of current fingerprint drift indices obtained within a preset time window, and evaluating the evolution of the connector's physical performance based on the evaluation slope, includes: performing linear regression analysis on a series of current fingerprint drift indices obtained within a preset time window to obtain the evaluation slope characterizing the trend of a series of indices; determining that the physical performance of the connector is in a deteriorating trend when the evaluation slope is consistently positive and exceeds a preset degradation trend slope threshold; and determining that the connector has suffered a transient event impact rather than performance degradation when the evaluation slope is close to zero but there are single or a few numerical abrupt changes in the series of current fingerprint drift indices.

4. The performance evaluation method for an industrial internet connector according to claim 3, characterized in that, The method also includes a step of determining the degradation mode of physical performance, the steps of which include: in step a, calculating and recording a baseline higher-order statistical moment based on the time-domain waveform of the endogenous power supply common-mode noise signal; in step b, periodically calculating the current higher-order statistical moment of the current common-mode noise signal time-domain waveform; and, when the evaluation result of step e indicates that the physical performance of the connector is in a degradation trend, distinguishing the degradation mode of physical performance based on the comparison between the current higher-order statistical moment and the baseline higher-order statistical moment; wherein, the baseline higher-order statistical moment is the baseline kurtosis value, and the current higher-order statistical moment is the current kurtosis value.

5. The performance evaluation method for an industrial internet connector according to claim 4, characterized in that, Based on the comparison between the current higher-order statistical moments and the baseline higher-order statistical moments, the steps for distinguishing the degradation mode of physical performance include: when the current kurtosis value does not increase significantly compared with the baseline kurtosis value, the degradation mode is determined to be progressive physical decay; when the current kurtosis value increases significantly compared with the baseline kurtosis value, the degradation mode is determined to be discontinuous contact deterioration caused by micro-arcsing or mechanical vibration of the contact surface.

6. The performance evaluation method for an industrial internet connector according to claim 1, characterized in that, The method also includes a step of compensating for temperature changes, which includes: in step a, determining a baseline noise floor power based on the noise floor band without harmonic peaks in the spectrum of the endogenous power common-mode noise signal; and before performing step d, further including: determining a current noise floor power based on the current common-mode noise signal, and performing temperature compensation on the current fingerprint drift index calculated in step d based on the difference between the current noise floor power and the baseline noise floor power, so as to eliminate the influence of ambient temperature changes on the evaluation results.

7. The performance evaluation method for an industrial internet connector according to claim 1, characterized in that, The method also includes a step of identifying intermittent contact defects, which includes: during the power-on startup of the device connected to the connector, continuously acquiring a series of instantaneous noise fingerprints at a frequency higher than that of periodically acquiring the current common-mode noise signal on the ground wire and calculating the corresponding instantaneous fingerprint drift index to form a fingerprint drift index sequence; and calculating the standard deviation of the fingerprint drift index sequence itself. When the standard deviation exceeds a preset connection stability threshold, it is determined that the connector has a potential intermittent contact defect caused by a loose connection.

8. The performance evaluation method for an industrial internet connector according to claim 3, characterized in that, The method also includes a step of defining the noise source, which includes: simultaneously acquiring the differential mode noise signal between the data signal pairs carried by the connector while acquiring the current common mode noise signal in step b; when it is determined in step e that the connector has been subjected to a transient event, further comparing the energy changes of the current common mode noise signal and the differential mode noise signal; if the energy of the current common mode noise signal and the energy of the differential mode noise signal increase significantly in the same proportion, the root cause of the transient event is defined as external electromagnetic interference, thereby distinguishing between external environmental problems and connector-related problems.

9. A performance evaluation system for an industrial internet connector, employing the performance evaluation method for an industrial internet connector as described in claim 1, characterized in that, The system includes: A baseline determination module is configured to: acquire the intrinsic power common-mode noise signal generated by the power supply circuit on the device ground line at the downstream side of the connector; and determine a baseline noise fingerprint characterizing the initial reference state of the connector and a baseline harmonic spectrum relationship characterizing the characteristics of the power supply circuit corresponding to the intrinsic power common-mode noise signal based on the spectrum of the intrinsic power common-mode noise signal. A periodic acquisition module is configured to periodically acquire the current common-mode noise signal on the ground wire during device operation to obtain a current noise fingerprint and a current harmonic spectrum relationship. A source verification module receives the baseline harmonic spectrum relationship and the current harmonic spectrum relationship, and is configured to: compare and verify the two before performing connector performance evaluation, and output an enable signal only when the deviation between the current harmonic spectrum relationship and the baseline harmonic spectrum relationship is within a preset source stability range; A drift index calculation module receives a baseline noise fingerprint and a current noise fingerprint, and in response to an enable signal, is configured to compare the current noise fingerprint with the baseline noise fingerprint to calculate a current fingerprint drift index that quantifies the degree of deviation between the two. A performance evaluation module receives a series of current fingerprint drift indices output by the drift index calculation module within a preset time window, and is configured to: calculate an evaluation slope characterizing the rate of change of the series of indices, and evaluate the evolution state of the physical performance of the connector based on the evaluation slope.

Images