Integrated terminal connection state detection method and system based on sensor

Abstract

The invention relates to the technical field of integrated terminal connection state detection, in particular to an integrated terminal connection state detection method and system based on a sensor. Comprising the following steps: acquiring instantaneous voltage information and instantaneous current information; when a current impact event exists in the instantaneous current information, calculating a transient impedance parameter; transient characteristics are extracted from the transient impedance parameters, and a normal operation reference range is established in a normal connection state; obtaining transient characteristics and comparing the transient characteristics with a normal operation reference range to obtain a deviation condition or a change trend; and judging whether an early mechanical loosening risk criterion is met or not according to the deviation condition or the change trend, and if yes, judging that an early mechanical loosening risk exists and outputting early warning information. The problems that in the prior art, an integrated terminal connection state detection method is difficult to recognize early mechanical loosening risks sensitively in real time, and early and transient thermal signals are difficult to effectively capture under the background of strong electromagnetic interference and mechanical vibration, so that fault early warning is not timely are solved.

Description

Technical Field

[0001] This application relates to the field of integrated terminal connection status detection technology, and more specifically, to a sensor-based integrated terminal connection status detection method and system. Background Technology

[0002] In the fields of power systems and industrial automation, integrated terminals are crucial components for equipment connections, and their contact stability directly affects equipment safety. Currently, checking the terminal connection status typically relies on manual inspection or analysis after equipment shutdown. This makes it difficult to monitor electrical parameters (such as contact resistance and current distribution) in real time, resulting in potential risks such as poor contact, overheating, or loosening going undetected in a timely manner.

[0003] In large industrial equipment environments, integrated terminals must withstand not only instantaneous current surges caused by loads but also mechanical vibrations. The combined effect of current surges and vibrations can accelerate the loosening of terminal connections. However, the early signs of loosening are characterized by brief, weak temperature pulses, which traditional temperature monitoring methods struggle to capture, leading to missed or delayed alarms until the loosening becomes severe.

[0004] Furthermore, the distribution cabinet operates in a complex environment where the combined effects of strong mechanical vibration and current surges make the loosening process of terminal connections more difficult to predict. Two impact forces of different frequencies and amplitudes act on the fastening bolts, accelerating the loosening process and posing a challenge to detection.

[0005] To address the aforementioned issues, existing technologies urgently need improvement. Summary of the Invention

[0006] This application discloses a sensor-based integrated terminal connection status detection method and system, which aims to solve the problems in the prior art where integrated terminal connection status detection methods are difficult to identify early mechanical loosening risks in real time and with high sensitivity, and are difficult to effectively capture early and transient thermal signals under strong electromagnetic interference and mechanical vibration backgrounds, thus leading to untimely fault warnings.

[0007] The technical solution of this application is as follows: In a first aspect, this application discloses a sensor-based integrated terminal connection status detection method, the method comprising: Simultaneously acquire instantaneous voltage information at both ends of the connection point of the integrated terminal, as well as instantaneous current information flowing through the integrated terminal; When the instantaneous current information satisfies the preset impact criterion to indicate the existence of a current impact event, the instantaneous voltage information and the instantaneous current information are kept in time correspondence within the time window corresponding to the current impact event, and the transient impedance parameters of the connection point under the impact are calculated based on the instantaneous voltage information and the instantaneous current information. Transient features are extracted from transient impedance parameters to characterize the microscopic contact state of the connection point. When the integrated terminal is in a normal connection state, establish a normal operating reference range for transient characteristics; During the operation of the integrated terminal, transient characteristics are continuously acquired and compared with the normal operation reference range to obtain the deviation or trend of the transient characteristics; Based on the deviation or trend of change, determine whether the deviation or trend of change meets the preset early mechanical loosening risk criteria. The early mechanical loosening risk criteria include: the criterion for deviation is that the transient characteristics exceed the normal operation reference range, and / or the criterion for trend of change is that the deviation continuously meets the preset deviation conditions within the preset observation window. When the deviation or trend meets the early mechanical loosening risk criterion, it is determined that there is an early mechanical loosening risk at the connection point of the integrated terminal, and a warning message is output.

[0008] Secondly, this application also discloses a sensor-based integrated terminal connection status detection system, the system comprising: The information acquisition module is used to synchronously acquire the instantaneous voltage information at both ends of the connection point of the integrated terminal, as well as the instantaneous current information flowing through the integrated terminal; The impedance calculation module is used to ensure that the instantaneous voltage information and the instantaneous current information are time-correlated within the time window corresponding to the current impact event when the instantaneous current information meets the preset impact criterion and indicates the existence of a current impact event. Based on the instantaneous voltage information and the instantaneous current information, the transient impedance parameters of the connection point under the impact are calculated. The feature extraction module is used to extract transient features from transient impedance parameters to characterize the microscopic contact state of the connection point; The reference range establishment module is used to establish a normal operating reference range for transient characteristics when the integrated terminal is in a normal connection state. The change tracking module is used to continuously acquire transient characteristics during the operation of the integrated terminal and compare the transient characteristics with the normal operation reference range to obtain the deviation or change trend of the transient characteristics; The judgment module is used to determine whether the deviation or trend meets the preset early mechanical loosening risk criteria based on the deviation or trend. The early mechanical loosening risk criteria include: the criterion for deviation is that the transient characteristics exceed the normal operation reference range, and / or the criterion for trend is that the deviation continuously meets the preset deviation conditions within the preset observation window. The early warning module is used to determine that there is an early mechanical loosening risk at the connection point of the integrated terminal when the deviation or trend meets the early mechanical loosening risk criterion, and outputs early warning information.

[0009] Beneficial Effects: The sensor-based integrated terminal connection status detection method disclosed in this application focuses on the transient impedance parameters under current surges. This allows for real-time and sensitive identification of minute gaps or subtle changes in contact status caused by early mechanical loosening at the connection point. These changes do not cause a significant and sustained increase in the overall temperature of the integrated terminal in the early stages. Therefore, this application enables early fault warning, avoiding the limitations of traditional temperature-based monitoring methods, significantly improving the timeliness and accuracy of fault detection, and effectively ensuring the safety of power systems and industrial automation equipment operation. Attached Figure Description

[0010] Figure 1 This is a schematic flowchart of a sensor-based integrated terminal connection status detection method provided in this application.

[0011] Figure 2 A flowchart of a sensor-based integrated terminal connection status detection system provided in this application.

[0012] In the diagram: 1. Information acquisition module; 2. Impedance calculation module; 3. Feature extraction module; 4. Reference range establishment module; 5. Change tracking module; 6. Judgment module; 7. Early warning module. Detailed Implementation

[0013] The technical solutions of this application will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are merely some embodiments of this application, and not all embodiments. The components of this application described and shown in the accompanying drawings can generally be arranged and designed in various different configurations. Therefore, the following detailed description of the embodiments of this application provided in the accompanying drawings is not intended to limit the scope of the claimed application, but merely to illustrate selected embodiments of this application. All other embodiments obtained by those skilled in the art based on the embodiments of this application without inventive effort are within the scope of protection of this application.

[0014] It should be noted that similar reference numerals and letters in the following figures indicate similar items; therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures. Furthermore, in the description of this application, terms such as "first," "second," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance.

[0015] In the fields of power systems and industrial automation, integrated terminals are crucial components connecting critical equipment, and the stability of their contact directly affects the safety of the entire equipment operation. However, traditional detection methods mostly rely on manual on-site inspections or equipment shutdown analysis, making it difficult to monitor changes in electrical parameters such as contact resistance and current distribution in real time. This results in the failure to promptly warn of potential risks such as poor contact, overheating, or loosening. Especially in large industrial equipment environments, integrated terminals not only withstand instantaneous high-current surges but also cope with continuous mechanical vibrations. Both factors together accelerate the mechanical loosening of the connection. This early loosening does not cause a significant and sustained increase in the overall temperature of the integrated terminal. Existing temperature threshold-based monitoring methods are unable to effectively capture early transient thermal signals, often leading to missed or delayed alarms.

[0016] Reference Figure 1 In response, this application proposes a sensor-based integrated terminal connection status detection method, comprising: S1000: Synchronously acquire instantaneous voltage information at both ends of the connection point of the integrated terminal, as well as instantaneous current information flowing through the integrated terminal; S2000: When the instantaneous current information meets the preset impact criterion and indicates the existence of a current impact event, the instantaneous voltage information and the instantaneous current information are kept in time correspondence within the time window corresponding to the current impact event, and the transient impedance parameters of the connection point under the impact are calculated based on the instantaneous voltage information and the instantaneous current information. S3000: Extract transient features from transient impedance parameters to characterize the microscopic contact state of the connection point; S4000: Establishes a normal operating reference range for transient characteristics when the integrated terminal is in a normal connection state; S5000: During the operation of the integrated terminal, transient characteristics are continuously acquired and compared with the normal operating reference range to obtain the deviation or trend of the transient characteristics; S6000: Based on the deviation or trend of change, determine whether the deviation or trend of change meets the preset early mechanical loosening risk criteria. The early mechanical loosening risk criteria include: the criterion for deviation is that the transient characteristics exceed the normal operation reference range, and / or the criterion for trend of change is that the deviation continuously meets the preset deviation conditions within the preset observation window. S7000: When the deviation or trend meets the early mechanical loosening risk criterion, it determines that there is an early mechanical loosening risk at the connection point of the integrated terminal and outputs a warning message.

[0017] An integrated terminal block is a component that connects wires to equipment. It contains one or more connection points, which are critical for current transmission. Instantaneous voltage and current information are voltage and current values ​​collected over extremely short timescales (e.g., microseconds or nanoseconds) to reflect the dynamic electrical characteristics during current surge events. A current surge event is a phenomenon where current rapidly increases or decreases within a very short time due to equipment startup, switching operation, or malfunction. Transient impedance parameters are the dynamic impedance characteristics exhibited by the connection point under current surge conditions. Transient characteristics are indicators extracted from transient impedance parameters that sensitively reflect changes in the microscopic contact state of the connection point, such as peak impedance, rise time, and oscillation frequency. The normal operating reference range is the range or set of values ​​or patterns within which the transient characteristics should be when the integrated terminal block is in a healthy connection state. Early mechanical loosening risk refers to a risk state where, although there are no obvious macroscopic signs of failure at the connection point, initial signs of loosening such as decreased contact pressure and reduced contact area are observed at the microscopic level, potentially leading to serious failure.

[0018] Synchronous acquisition of instantaneous voltage and current information can be achieved by connecting a high-bandwidth voltage sensor in parallel across the connection point and a high-bandwidth current sensor in series in the loop flowing through the integrated terminal. For example, a Hall effect current sensor can be used to measure instantaneous current, and a differential probe connected to a high-speed data acquisition card can be used to measure instantaneous voltage; the sensors need to have sufficiently high sampling rates and accuracy to capture rapid changes in current surge events. To ensure "time correspondence," a unified clock or hardware synchronization trigger can be used for synchronous acquisition, and the fixed delay of the sampling link can be calibrated and compensated to ensure that the instantaneous voltage and instantaneous current information within the time window correspond one-to-one at the same sampling moment.

[0019] When the instantaneous current information meets the preset impact criterion, indicating the presence of a current impact event, the transient impedance parameter is calculated within a time window. The preset impact criterion can be set as the instantaneous current rise rate exceeding a certain threshold, or the instantaneous current peak value exceeding a certain threshold; for example, when the current is detected to rise from 10 amperes to 1000 amperes within 10 microseconds, a current impact event can be identified. The length of the time window can be set according to the duration of the current impact event; for example, it can be set from 100 microseconds before the start of the current impact to 500 microseconds after the end of the impact. The transient impedance parameter can be calculated by dividing the instantaneous voltage information by the instantaneous current information; to avoid numerical distortion caused by excessively small instantaneous current information, calculations can be performed only on sampling points that meet the preset effective current conditions within the time window, or a lower limit can be set for the instantaneous current information and outliers can be removed or interpolated. In addition to time-domain ratio calculation, more complex signal processing methods, such as wavelet transform or Fourier transform, can be used to analyze the spectral relationship between voltage and current signals and assist in characterizing the transient impedance parameter.

[0020] Transient features can be extracted from transient impedance parameters using a combined time-domain and frequency-domain analysis. In the time domain, the peak value, rise time, fall time, full width at half maximum (FWHM), oscillation frequency, and damping coefficient of the transient impedance parameters can be calculated. In the frequency domain, spectral analysis can be performed on the transient impedance parameters to extract the energy distribution of specific frequency bands. These transient features are used to reflect the dynamic response of the connection point under impact, such as minute deformations of the contact interface and arc discharge, thereby improving the sensitivity to changes in the microscopic contact state.

[0021] When the integrated terminal is in a normal connection state, a normal operation reference range can be established. After installation, commissioning, or repair and repositioning, transient characteristics can be continuously collected and statistically analyzed or modeled over a period of time during normal operation. For example, the mean, standard deviation, maximum, and minimum values ​​of transient characteristics can be calculated, and a confidence interval can be set as the normal operation reference range. Alternatively, machine learning methods can be used to learn a set of transient characteristic patterns under normal operation through cluster analysis or anomaly detection algorithms, and this set of patterns can be determined as the normal operation reference range.

[0022] During the operation of the integrated terminal, transient characteristics are continuously acquired and compared with the normal operating reference range to obtain deviations or trends. If the transient characteristic is numerical, the difference between it and the average value of the normal operating reference range can be calculated to determine whether it exceeds the boundary and constitutes a deviation. If the transient characteristic is a vector or pattern, its similarity to the pattern of the normal operating reference range can be calculated to quantify the deviation. The trend can be obtained by tracking the directionality and persistence of the deviation over a period of time, which is manifested as the transient characteristic continuously moving in a certain direction over a period of time.

[0023] Early mechanical loosening risk criteria are used to transform deviations or trends into actionable risk assessment conditions. These criteria may include: a deviation criterion is that transient characteristics exceed the normal operating reference range, and / or a trend criterion is that the deviation continuously meets preset deviation conditions within a preset observation window. For example, if the peak value of the transient impedance exceeds the upper limit of the normal operating reference range three times consecutively, or if the rise time of the transient impedance continues to increase over the past 24 hours, then the early mechanical loosening risk criterion can be considered met.

[0024] When a deviation or trend meets the early mechanical loosening risk criteria, the connection point of the integrated terminal is determined to have an early mechanical loosening risk, and a warning message is output. The warning message may include the risk level, the time of occurrence, the possible cause, and the recommended handling measures; for example, the warning message may be output through audible and visual alarms, SMS notifications, emails, or integration into a SCADA system.

[0025] The core technical concept of this application lies in using current surge events as "probes" to reveal changes in the microscopic contact state through the transient electrical response of the connection point under the impact, thereby avoiding the lag effect of the thermal inertia of the integrated terminal body on temperature monitoring. Compared with temperature monitoring-based methods, this method is more sensitive to changes in small gaps or contact areas and can maintain detection effectiveness under strong electromagnetic interference and mechanical vibration. For example, in heavy industrial manufacturing workshops, the periodic, instantaneous high current surges generated when new servo-driven large stamping equipment is working will produce small, impactful repulsive forces at the connection of integrated terminals, accelerating the loosening of fastening bolts; due to the large heat capacity of integrated terminals, the brief pulsed heat will be quickly absorbed and dispersed, resulting in a very gradual and small increase in surface temperature, often submerged in normal temperature fluctuations and unable to trigger the warning threshold; this method, by comparing transient impedance parameters with transient characteristics, can provide early warning before significant heat generation due to connection deterioration and gain time for intervention.

[0026] In another embodiment of this application, step S3000 is further proposed to include: S3100: Performs high-pass filtering on transient impedance parameters to isolate low-frequency components in the transient impedance parameters; S3200: Performs spectral analysis on the transient impedance parameters after high-pass filtering to obtain the energy characteristics of a preset high-frequency band; S3300: Within the time window corresponding to the current surge event, select sampling points where the instantaneous current information meets the preset effective current conditions, construct a voltage-current dynamic relationship trajectory based on the instantaneous voltage information and instantaneous current information corresponding to the sampling point time, and construct an impedance-current dynamic relationship trajectory based on the transient impedance parameter and instantaneous current information, and calculate the geometric parameters of the voltage-current dynamic relationship trajectory and / or the impedance-current dynamic relationship trajectory. S3400: Performs differentiation on the rising edge of transient impedance parameters to detect impedance spikes and / or oscillations with a duration of a preset very short time. S3500: Dynamically learns and updates the transient feature extraction reference benchmark of the detection results of energy characteristics, geometric parameters, and impedance spikes and / or oscillations under normal operating conditions; S3600: Based on the current energy characteristics, geometric parameters, and the detection results of impedance spikes and / or oscillations, the transient characteristics are extracted relative to the reference reference to form transient characteristics.

[0027] Specifically, high-pass filtering of transient impedance parameters refers to removing the DC component and low-frequency noise from the transient impedance parameter signal using digital or analog filters to highlight the high-frequency transient response. This high-frequency transient response is usually more directly related to changes in the micro-contact state inside the connection point (such as micro-arcs and rapid fluctuations in contact resistance), thereby increasing the proportion of effective information in subsequent feature extraction.

[0028] Among them, performing spectral analysis on the transient impedance parameters after high-pass filtering can be understood as converting the time-domain signal to the frequency domain through methods such as fast Fourier transform, and calculating the energy in the preset high-frequency band as energy characteristics; for example, the energy in the range of 1kHz to 100kHz can be focused on to quantify the high-frequency energy changes accompanied by micro-discharge or contact instability generated at the connection point under impact.

[0029] In practical applications, selecting sampling points within the time window corresponding to a current surge event that meet preset effective current conditions for instantaneous current information means processing data only within a time period where the current is sufficiently large to ensure measurement validity. This avoids invalid calculations or introducing ratio distortion when the current is too small or noise dominates. A voltage-current dynamic relationship trajectory is constructed based on the instantaneous voltage and current information corresponding to the sampling point time, and an impedance-current dynamic relationship trajectory is constructed based on the transient impedance parameters and instantaneous current information. These are used to characterize the dynamic nonlinear relationship between voltage, current, and impedance during the surge. Furthermore, geometric parameters such as the area, slope, curvature, or coordinates of specific points on the trajectory can be calculated to quantify the nonlinear behavior and energy dissipation characteristics of the contact interface.

[0030] Furthermore, differential operations are performed on the rising edge of the transient impedance parameter to detect drastic changes in impedance that occur in a very short time, such as impedance spikes or high-frequency oscillations caused by instantaneous separation or re-contact of the contact points; such impedance spikes and / or oscillations can often serve as sensitive indicators of early mechanical loosening.

[0031] In addition, a transient feature extraction reference benchmark is dynamically learned and updated based on the detection results of energy characteristics, geometric parameters, and impedance spikes and / or oscillations under normal operating conditions. This benchmark is used to adapt the feature extraction process to the natural aging, environmental changes, or load fluctuations of the connection point under normal operating conditions, so that the transient features formed subsequently can better reflect abnormal conditions rather than normal operating condition fluctuations.

[0032] Therefore, based on the current energy characteristics, geometric parameters, and the detection results of impedance spikes and / or oscillations, the deviation and / or change pattern of the reference benchmark are extracted relative to the transient characteristics to form transient characteristics. These transient characteristics are used to integrate multi-dimensional information and consider their changes relative to the dynamic benchmark, thereby constructing a transient characteristic representation that is more sensitive and discriminative to the early mechanical loosening risk, in order to serve subsequent risk assessment.

[0033] In some preferred embodiments, it is assumed that the integrated terminal experiences a current surge event during operation. First, the sensor synchronously acquires the instantaneous voltage information at both ends of the connection point and the instantaneous current information flowing through the terminal. When the instantaneous current information meets the preset surge criterion, the transient impedance parameter of the connection point is calculated based on the above data within the corresponding time window. To extract transient features characterizing the microscopic contact state of the connection point from the transient impedance parameter, the transient impedance parameter is first subjected to high-pass filtering to remove low-frequency interference such as the power grid frequency and highlight high-frequency components. Subsequently, the high-pass filtered transient impedance parameter is subjected to spectral analysis, for example, by calculating its energy in the 5kHz to 50kHz frequency band using fast Fourier transform as an energy feature. Simultaneously, sampling points with instantaneous currents exceeding a certain threshold are selected within the time window of the current surge event. Voltage-current dynamic relationship trajectories are plotted using the instantaneous voltage and current information corresponding to these sampling points, and impedance-current dynamic relationship trajectories are plotted using transient impedance parameters and instantaneous current information. The area and maximum slope of the trajectories are calculated as geometric parameters. Furthermore, differential operations are performed on the rising edge of the transient impedance parameters to detect impedance spikes or high-frequency oscillations lasting less than 10 microseconds. The detection results of the aforementioned energy characteristics, geometric parameters, and impedance spikes and / or oscillations are compared with a transient feature extraction reference benchmark dynamically learned and updated by the system. For example, when the current energy characteristic is significantly higher than the reference benchmark, the trajectory shape corresponding to the geometric parameters is significantly distorted, or a previously unseen impedance spike is detected, the aforementioned deviations and / or change patterns are combined to form a transient feature vector. This transient feature vector is then compared with the normal operating reference range to determine if there is an early risk of mechanical loosening.

[0034] In another embodiment of this application, it is further proposed to continuously acquire transient features and compare the transient features with the normal operating reference range to obtain the deviation or trend of the transient features, specifically including: S5100: While continuously acquiring transient characteristics, it simultaneously acquires environmental vibration information and power grid fluctuation information related to the integrated terminal connection. S5200: Performs noise separation processing on transient characteristics, separating the electromagnetic interference components indicated by power grid fluctuation information from the transient characteristics to obtain the separated transient characteristics; S5300: Based on the separation of transient characteristics and environmental vibration information, feature correlation analysis is performed to obtain correlation indicators that characterize the correlation between the separation of transient characteristics and environmental vibration information; S5400: Establish an environmental interference mode reference benchmark based on historical correlation indicators when the integrated terminal is in a normal connection state. S5500: Compare the correlation index with the environmental interference pattern reference benchmark to identify environmental interference components that match the environmental interference pattern, and suppress, remove and / or compensate the separated transient features to obtain the interference-free transient features. S5600: Compares the transient characteristics after interference removal with the normal operating reference range to obtain the deviation or trend of the transient characteristics.

[0035] Specifically, synchronously acquiring environmental vibration information and power grid fluctuation information related to the integrated terminal connection point refers to collecting mechanical vibration data in the environment and power grid voltage and current fluctuation data in real time through vibration sensors and power grid monitoring equipment deployed near the integrated terminal connection point. Among them, environmental vibration information can include parameters such as vibration frequency and vibration amplitude, which are used to characterize the impact of external mechanical stress on the connection point. Power grid fluctuation information can include voltage sags, swells, harmonic distortions, etc., which are used to characterize the interference of the electromagnetic environment on transient characteristic measurements. Its purpose is to provide multi-source data support for subsequent interference separation, identification and compensation.

[0036] Among these steps, noise separation processing of transient features involves separating the electromagnetic interference components indicated by power grid fluctuation information from the transient features to obtain the separated transient features. This can be understood as using signal processing techniques such as wavelet denoising, Fourier transform combined with band-stop filtering to identify and remove electromagnetic noise components caused by power grid fluctuations based on synchronously acquired power grid fluctuation information. Specifically, the spectral characteristics of power grid fluctuation information can be analyzed first to determine the main interference frequencies, and then filters or denoising strategies can be designed to process the transient features, thereby isolating electromagnetic interference and improving the accuracy and comparability of the separated transient features.

[0037] In practical applications, feature correlation analysis based on the separated transient features and environmental vibration information to obtain correlation indicators that characterize the correlation refers to using methods such as cross-correlation analysis, Granger causality test, or neural network-based correlation models to quantify the influence intensity and action mode between the separated transient features and environmental vibration information. For example, the correlation coefficient between the two can be calculated, or a prediction model can be constructed to assess the degree of influence of environmental vibration on the separated transient features. The purpose is to provide quantifiable evidence for environmental interference pattern recognition and subsequent processing.

[0038] Furthermore, establishing an environmental interference mode reference benchmark based on historical correlation indicators when the integrated terminal is in a normal connection state refers to continuously collecting correlation indicators during the normal operation of the integrated terminal and performing statistical analysis on historical data (such as mean, variance, distribution range, etc.) to form a reference model or threshold set, thereby obtaining an environmental interference mode reference benchmark that can be used to determine the "normal environmental interference level / mode". Its purpose is to provide a reference benchmark for subsequent identification and decision-making.

[0039] Furthermore, by comparing the correlation indicators with the environmental disturbance model reference benchmark, the environmental disturbance components introduced by environmental vibration are identified. The separated transient features are then suppressed, eliminated, and / or compensated to obtain the disturbance-free transient features. This can be understood as adopting a graded processing strategy based on the comparison results: when the correlation indicators fall within the normal range allowed by the environmental disturbance model reference benchmark, the current environmental disturbance is determined to be an expected disturbance, and the separated transient features can be slightly compensated or weighted and suppressed; when the correlation indicators deviate significantly from the environmental disturbance model reference benchmark (e.g., persistently high correlation or pattern abrupt change), abnormal environmental disturbance is determined to exist, and the affected separated transient features can be subject to stronger filtering suppression, elimination of disturbed data points, and / or model compensation to reduce the impact of the abnormal disturbance on the features. The purpose is to ensure that the features ultimately involved in risk assessment reflect the true microscopic contact state of the connection point as much as possible rather than external disturbances, and to suppress disturbance components that are consistent with historical environmental disturbance patterns and do not exhibit continuous deviations.

[0040] Finally, the transient characteristics after interference removal are compared with the normal operating reference range to obtain the deviation or trend of the transient characteristics. Since the aforementioned noise separation, suppression, elimination and / or compensation processes have reduced the contribution of electromagnetic and mechanical environmental interference, the comparison results can more accurately characterize the degree of deviation of the transient characteristics from the normal operating reference range and its trend over time.

[0041] The proposed solution addresses the issue of transient characteristics being susceptible to environmental interference through a closed-loop mechanism: synchronous acquisition of environmental vibration and power grid fluctuation information, separation of electromagnetic noise, quantification of correlation indicators, establishment of environmental interference mode reference benchmarks, and graded suppression / removal / compensation. Power grid fluctuation information guides the separation of electromagnetic interference components, environmental vibration information quantifies the correlation between mechanical interference and the separated transient characteristics, and environmental interference mode reference benchmarks distinguish between normal and abnormal disturbances, thereby selecting suppression, removal, and / or compensation processes of varying intensities. This reduces the risk of misjudgment and underreporting and improves stability in complex industrial environments.

[0042] In another embodiment of this application, it is further proposed that, when the deviation or trend of change meets the early mechanical loosening risk criterion, after determining that there is an early mechanical loosening risk at the connection point of the integrated terminal, the method further includes: S7100: Acquires criticality level information, load level information and potential fault impact level information for connection points of multiple integrated terminals. Criticality level information is used to characterize the importance of the connection point in the power supply circuit or production line. Load level information is used to characterize the load intensity of the connection point within a preset statistical time window. Potential fault impact level information is used to characterize the degree of impact of connection point failure on power supply continuity or production. S7200: Calculates the risk priority value of each integrated terminal connection point based on criticality level information, load level information, and potential fault impact level information; S7300: Generates early warning ranking results based on risk priority values, arranging multiple early warning messages according to a preset ranking rule based on risk priority values; S7400: Determine the risk level corresponding to each early warning message based on the correspondence between the risk priority value and the preset classification rules; Output warning information, including: S7500: Outputs multiple warning messages according to the warning sorting results, and executes the corresponding preset notification strategy according to the risk level to issue graded warning messages.

[0043] Specifically, after determining that there is an early risk of mechanical loosening at the connection point of the integrated terminal, in order to achieve more refined risk management and early warning, it is necessary to obtain multi-dimensional information related to the connection point. Among them, the criticality level information is used to measure the importance of the connection point in the entire power supply circuit or production line. For example, if a connection point is located on the main power supply line of core production equipment, its criticality level is usually set to high. The load level information is used to characterize the load intensity borne by the connection point within a preset statistical time window, such as the current intensity or power consumption flowing through the connection point, which reflects the actual operating pressure borne by the connection point. The potential fault impact level information is used to characterize the impact of the connection point failure on the continuity of power supply or the production process. For example, a connection point that causes the entire production line to stop is usually set to a high potential fault impact level. The above information can be obtained through preset databases, sensor data, or manual input to ensure that subsequent assessments have a verifiable data source.

[0044] Furthermore, after obtaining criticality level information, load level information, and potential fault impact level information, these are used as inputs to calculate the risk priority value of each integrated terminal connection point through a preset algorithm or model. This risk priority value is a comprehensive indicator used to quantify the urgency and importance of the early mechanical loosening risk of the connection point. For example, it can be calculated using weighted average, fuzzy logic reasoning, or machine learning models. Different information can be assigned different weights to reflect their relative importance in risk assessment, thereby making the risk priority value comparable and interpretable.

[0045] Therefore, the warning ranking result is generated based on the risk priority value, so that multiple warning information are arranged according to the preset sorting rules, such as sorting them in descending order of risk priority value from high to low. This allows the warning information corresponding to the connection point with the highest risk to be displayed or processed first, avoiding congestion in the handling when multiple points alarm at the same time.

[0046] Meanwhile, based on the correspondence between risk priority values ​​and preset grading rules, the risk level corresponding to each warning message is determined. The preset grading rules are used to map continuous risk priority values ​​to discrete risk levels, such as "low risk", "medium risk", "high risk", "emergency risk", etc., and can be set according to actual application scenarios and risk management strategies, thereby achieving a unified definition and executable stratification of risk levels.

[0047] Finally, when outputting early warning information, multiple early warning messages are output according to the early warning ranking results. This ensures that maintenance personnel prioritize the most urgent or important risks and execute the corresponding preset notification strategies based on the risk level. For example, for early warnings at the "urgent risk" level, relevant personnel can be notified immediately via SMS, telephone, or audible and visual alarms; for early warnings at the "medium risk" level, notifications can be sent via email or system messages; and for early warnings at the "low risk" level, they can be recorded in the system log or displayed on the interface. This hierarchical early warning mechanism enables precise response and resource allocation.

[0048] In some preferred embodiments, assuming a factory has 100 integrated terminal connection points, and 5 connection points are identified as having an early risk of mechanical loosening during a certain inspection, traditional methods may only output 5 identical warning messages. However, the solution in this application further processes the information: the system first obtains the criticality level information of these 5 connection points (e.g., connection point A is located on the core production line, level high; connection point B is located on auxiliary equipment, level medium), load level information (e.g., connection point A has high cumulative energy dissipation, high load level; connection point B has low load level), and potential fault impact level information (e.g., failure of connection point A will cause production line shutdown, impact level high; failure of connection point B only affects local functions, impact level medium). Then, it calculates the risk priority value (e.g., connection point A is 0.95, connection point B is 0.60, connection point C is 0.80, connection point D is 0.55, and connection point E is 0.70), and generates a warning ranking result based on this (e.g., A > C > E > B > ...). The system determines the risk level (connection point A is "emergency risk", connection points C and E are "high risk", and connection points B and D are "medium risk") based on preset classification rules (e.g., above 0.85 is emergency, 0.70-0.85 is high risk, and 0.50-0.70 is medium risk) according to the sorting results. Finally, the system outputs warning information and executes notification strategies according to the sorting results. For example, for the "emergency risk" of connection point A, the system immediately sends SMS and telephone notifications to the workshop director and maintenance team; for the "high risk" of connection points C and E, the system sends email notifications and generates work orders; and for the "medium risk" of connection points B and D, the system only displays the warning information on the system interface and records logs. This allows maintenance personnel to quickly identify and prioritize the handling of critical risks to avoid major accidents.

[0049] In another embodiment of this application, obtaining load level information is further proposed, including: S7110: Within the time window corresponding to the current surge event, calculate the energy dissipation value of the current surge event based on the instantaneous voltage information and instantaneous current information. S7120: Accumulate the energy dissipation values ​​of multiple current impact events within a preset statistical time window to obtain the cumulative energy dissipation. S7130: Uses accumulated power dissipation as load level information.

[0050] Specifically, the energy dissipation value refers to the energy consumption represented by the instantaneous voltage and current information at both ends of the connection point during a current surge event. It can be calculated by integrating the product of the instantaneous voltage and current information within the time window corresponding to the current surge event. For example, a digital integration algorithm can be used to multiply the instantaneous voltage and current at each sampling point and accumulate these products over the entire time window to obtain the total energy dissipation of the surge event. The purpose is to quantify the instantaneous thermal or mechanical stress caused to the connection point by each current surge.

[0051] The preset statistical time window is a specific time period used to accumulate power dissipation values, such as several hours, a day, a week, or a longer period. Its setting is adjusted according to the periodicity of load changes and the real-time requirements of risk assessment. Its purpose is to capture the overall load situation of the connection point within a sufficiently long time span, so as to avoid the randomness of a single impact event from having too much impact on the load level assessment.

[0052] In practical applications, cumulative energy dissipation refers to the sum of energy dissipation values ​​of all current surge events occurring within a preset statistical time window. For example, the system can maintain a sliding time window. Whenever a new current surge event occurs and its energy dissipation value is calculated, it is added to the cumulative amount within the current window, and the energy dissipation values ​​of older events exceeding the window's time range are removed. The purpose is to provide a dynamically updated indicator reflecting the recent total energy load of the connection point. Using this cumulative energy dissipation as load level information can more accurately reflect the actual workload and energy loss borne by the connection point over a period of time, thus providing input for subsequent risk priority calculations.

[0053] In some preferred embodiments, the following specific example illustrates the situation: Assuming an integrated terminal block is operating in an industrial device, its connection status needs to be continuously monitored. The system first synchronously acquires instantaneous voltage information and instantaneous current information flowing through both ends of the connection point using sensors. When the instantaneous current information meets a preset impact criterion, such as the current exceeding a certain threshold in a very short time, the system identifies a current impact event.

[0054] For this current surge event, within the corresponding time window, the system calculates the energy dissipation value of the surge based on the collected instantaneous voltage and current information. Specifically, if the instantaneous voltage is U(t) and the instantaneous current is I(t), the energy dissipation value E can be obtained by integrating U(t) * I(t) within the surge time window [t1, t2], i.e., E = ∫[t1, t2] U(t) * I(t) dt.

[0055] The system continuously monitors and calculates the energy dissipation value of each current surge event. Simultaneously, the system sets a preset statistical time window, such as 24 hours. Within each 24-hour period, the system accumulates the energy dissipation values ​​of all detected current surge events to obtain a cumulative energy dissipation. For example, if 50 current surge events occur within a certain 24-hour period, with energy dissipation values ​​of E1, E2, ..., E50 for each surge, then the cumulative energy dissipation is ΣEi.

[0056] This accumulated energy dissipation is then used as the load level information for the connection point over the past 24 hours. For example, a high accumulated energy dissipation indicates that the connection point has been subjected to a large energy load over the past 24 hours, and the risk of mechanical loosening may be exacerbated by the high load. This load level information is then input into the calculation of the risk priority value, and together with other criticality level information and potential fault impact level information, a comprehensive assessment of the early mechanical loosening risk of the connection point is conducted, generating corresponding early warning information. In this way, the load level assessment is more refined and physicalized, and can more accurately reflect the actual operating status of the connection point.

[0057] In another embodiment of this application, a method for obtaining criticality level information and potential fault impact level information is further proposed, specifically including: S7140: For each integrated terminal connection point, obtain the current production task priority of the equipment associated with the integrated terminal connection point in real time; S7150: Determine the key process node identifiers of the associated equipment based on the current production task priority, and determine the criticality level information based on the current production task priority and the key process node identifiers according to the preset criticality assessment rules. S7160: A security event level library is pre-built. The security event level library includes the correspondence between connection point attributes and security event levels. The connection point attributes include the equipment category of the device associated with the connection point and / or the power supply circuit identifier to which the connection point belongs. Based on the connection point attributes, the security event level corresponding to the connection point is determined in the security event level library. S7170: Based on the safety incident level and according to the preset loss assessment rules, determine the estimated data of production stoppage loss, and according to the preset impact assessment rules, determine the potential failure impact level information based on the safety incident level and the estimated production stoppage loss data.

[0058] Specifically, real-time acquisition of the current production task priority of the equipment associated with the connection point of the integrated terminal means that the system continuously reads the importance level of the current task of the associated equipment from the production plan or scheduling system and dynamically updates it as the production batch or order changes; its purpose is to provide real-time input for subsequent determination of criticality level information.

[0059] Specifically, determining the key process node identifiers for associated equipment based on the current production task priority involves combining the process route / process node table to identify whether the current process of the task is a critical stage, and generating key process node identifiers accordingly. Then, based on preset criticality assessment rules, the criticality level information is determined according to the current production task priority and the key process node identifiers. For example, when the production task has a high priority and the connection point is located at a critical process node, the criticality level information is rated as the highest according to the preset criticality assessment rules.

[0060] In practical applications, pre-building a safety event level database refers to establishing a database of correspondences between "connection point attributes and safety event levels." The connection point attributes include the equipment category of the associated equipment (e.g., motor, sensor, control cabinet, etc.) and / or the power supply circuit identifier to which the connection point belongs. Based on the connection point attributes, the system determines the safety event level corresponding to the connection point in the safety event level database, which is used to characterize the severity of the event that may be caused by the failure of the connection point (e.g., minor shutdown, partial production stoppage, or major safety accident).

[0061] Furthermore, determining the estimated production stoppage loss data based on the safety incident level and according to preset loss assessment rules refers to inputting the determined safety incident level into the economic loss assessment model to obtain the estimated production stoppage loss data (which may include direct and indirect economic losses, such as order delay penalties); then, based on preset impact assessment rules, determining the potential failure impact level information based on the safety incident level and the estimated production stoppage loss data. For example, connection points with a high safety incident level and large estimated production stoppage loss data are assessed as having the highest potential failure impact level.

[0062] In some preferred embodiments, the following specific example illustrates the situation: Imagine an automated production line where an integrated terminal connection point A is connected to a core processing device used to produce high-value products.

[0063] First, the system obtains the current production task priority of the equipment associated with connection point A in real time. For example, if the current production plan shows that the equipment is executing a "urgent order" production task, its priority is identified as "high" by the system.

[0064] Next, based on the "high" priority, the system determines that the equipment associated with connection point A is in a "critical processing stage" in the current process flow, and determines the criticality level information of connection point A as "extremely high" according to the preset criticality assessment rules (for example, the rules stipulate that "high priority task and critical processing stage" corresponds to the highest criticality level).

[0065] Simultaneously, the system queries a pre-built safety event level database. This database records the attributes of connection point A: the equipment category is "core processing equipment," and the power supply circuit identifier is "main power supply circuit 1." Based on these attributes, the safety event level database indicates that if connection point A fails, the potential safety event level is "major production stoppage accident."

[0066] Based on the safety incident level of this "major production stoppage accident" and according to the preset loss assessment rules (for example, the rules are calculated based on factors such as production stoppage time, product value, and order penalties), the system determines the estimated production stoppage loss to be "500,000 yuan / hour".

[0067] Finally, based on the preset impact assessment rules (for example, the rules take into account the safety incident level of "major production stoppage accident" and the estimated production stoppage loss data of "500,000 yuan / hour"), the system determines the potential failure impact level of connection point A to be "extremely high".

[0068] The criticality level information and potential failure impact level information of connection point A were accurately and dynamically obtained, providing a solid data foundation for subsequent risk priority calculation.

[0069] In another embodiment of this application, S7400 is further proposed to include: S7410: Obtain the historical distribution characteristics of the risk priority values ​​corresponding to the connection points of each integrated terminal; S7420: Based on historical distribution characteristics and in conjunction with preset risk management strategies, set initial risk level boundary points for risk levels to form preset classification rules; S7430: Compare the new risk priority value generated during the operation with the initial risk level boundary point to determine the risk level corresponding to the new risk priority value, and record the number of risk level switching times within the preset observation window; S7440: When the new risk priority value reaches a preset number of times within the preset boundary tolerance range, and the number of switching reaches the preset switching threshold, it is determined that there is an ambiguous area at the initial risk level boundary point. S7450: For fuzzy areas, collect risk priority numerical samples falling into the fuzzy areas and corresponding operation and maintenance handling result information. The operation and maintenance handling result information includes actual fault consequence information and / or operation and maintenance personnel feedback information. S7460: Based on the risk priority numerical samples and operation and maintenance handling results, the initial risk level boundary points are adjusted to obtain the updated risk level boundary points; S7470: Calculate the boundary clarity index based on the updated risk level boundary points, and continue to iterate and update the risk level boundary points until the preset boundary clarity requirements are met when the boundary clarity index does not meet the preset boundary clarity requirements. S7480: Based on the risk level boundary points that meet the preset boundary clarity requirements, the risk priority values ​​are range-determined to determine the risk level corresponding to each warning message.

[0070] Specifically, obtaining the historical distribution characteristics of the risk priority values ​​corresponding to the connection points of each integrated terminal means that the system continuously collects and analyzes the risk priority value data of each connection point under different time periods and different working conditions, and extracts features that reflect statistical regularities and trends, such as statistical quantities such as mean, variance, and skewness, or their probability density function and cumulative distribution function; these historical distribution characteristics provide a data basis for setting the initial risk level boundary points.

[0071] Among them, setting the initial risk level boundary point based on historical distribution characteristics and in combination with the preset risk management strategy means that after obtaining the historical distribution pattern of risk priority values, and in combination with the enterprise or industry's preset risk management goals and strategies (e.g., tolerance for high-risk events, attention to low-risk events, etc.), the boundaries between different risk levels are initially defined; the initial risk level boundary point forms the initial preset classification rule, which is used to initially determine the risk level.

[0072] In practical applications, the new risk priority value generated during the operation is compared with the initial risk level boundary point to determine the risk level corresponding to the new risk priority value, and the number of risk level switching times within the preset observation window is recorded. This means that the risk level is classified after comparing the new risk priority value calculated in real time with the current initial risk level boundary point, and the number of risk level switching times between different levels of the connection point is continuously counted within the preset observation window to quantify the classification stability near the boundary.

[0073] Furthermore, when a new risk priority value continuously reaches a preset number of times within a preset boundary tolerance range, and the number of switching reaches a preset switching threshold, it is determined that the initial risk level boundary point has a fuzzy region. This means that when the risk priority value frequently falls within a small range (preset boundary tolerance range) near a certain boundary point, and the risk level frequently switches within a preset observation window and reaches a preset switching threshold, it is determined that the boundary point has a fuzzy region, making it difficult for the system to classify stably.

[0074] Specifically, for fuzzy regions, risk priority numerical samples falling into the fuzzy region and corresponding operation and maintenance (O&M) handling results information are collected. The O&M handling results information includes actual fault consequences information and / or O&M personnel feedback information. This means that after identifying the fuzzy region, the system collects risk priority numerical samples falling into the fuzzy region and associates them with the corresponding actual O&M data (such as whether a fault occurred, the severity of the fault, O&M personnel's opinions on the warning handling results, etc.) to form real-world evidence that can be used to calibrate boundary points.

[0075] Therefore, adjusting the initial risk level boundary points based on the risk priority numerical samples and operation and maintenance handling results information to obtain updated risk level boundary points means using the risk priority numerical samples and their corresponding operation and maintenance handling results information, and using methods such as machine learning algorithms, expert systems or optimization models to correct the initial risk level boundary points, so that the updated risk level boundary points are closer to the actual risk consequences and handling experience.

[0076] Specifically, the process of calculating boundary clarity indices based on updated risk level boundary points, and iteratively updating risk level boundary points until they meet preset boundary clarity requirements, refers to calculating boundary clarity indices (such as the proportion of samples falling within the preset boundary tolerance range, the frequency of risk level switching, etc.) after obtaining updated risk level boundary points. If the boundary clarity indices do not meet preset boundary clarity requirements, the "collection-adjustment-evaluation" update process is repeated. To avoid the mixing of statistical calibers due to old and new boundaries, the statistical starting point for the number of risk level switching times within the preset observation window is reset after each updated risk level boundary point is obtained.

[0077] Finally, based on the risk level boundary points that meet the preset boundary clarity requirements, the risk priority values ​​are determined by interval judgment to determine the risk level corresponding to each early warning information. This means that after the risk level boundary points are iteratively optimized to meet the preset boundary clarity requirements, the system uses the risk level boundary points to determine the interval of real-time risk priority values, thereby assigning a stable and accurate risk level to each early warning information.

[0078] In some preferred embodiments, assuming that during the initial operation of a certain integrated terminal connection point, the initial risk level boundary points of its risk priority value are set to 0.5 and 0.8, corresponding to low risk, medium risk, and high risk levels, respectively. After a period of operation, the system finds that the risk priority value of this connection point frequently fluctuates between 0.48 and 0.52, and within a preset observation window, its risk level frequently switches between low risk and medium risk, with the number of switches reaching a preset switching threshold. At this point, the system determines that there is a fuzzy area at the initial risk level boundary point of 0.5. For this fuzzy area, the system automatically collects risk priority value samples falling within the range of 0.48 to 0.52. Simultaneously, the system queries the corresponding maintenance and handling results information for these samples. For example, for a sample with a risk priority value of 0.51, if the actual maintenance record shows that the connection point did not malfunction, and the maintenance personnel report that the warning is non-emergency; and for a sample with a risk priority value of 0.49, if the actual maintenance record shows that the connection point subsequently experienced slight mechanical loosening, but without serious consequences. Based on these risk priority numerical samples and operational handling results, the system uses machine learning algorithms (such as support vector machines or decision trees) to adjust the boundary points. After calculation, the system may adjust the boundary point between low and medium risk from 0.5 to 0.53. After adjustment, the system calculates new boundary clarity indicators, such as assessing whether the fluctuation of risk priority values ​​near the new boundary points has decreased, and whether the frequency of risk level switching has decreased. If the boundary clarity indicators still do not meet the preset requirements, the system will continue to iterate and adjust until the boundary points can clearly and stably divide the risk levels. Finally, when the boundary points are optimized to meet the requirements, all new risk priority values ​​will be determined based on these optimized boundary points to determine their accurate risk levels and issue corresponding graded early warning information.

[0079] In another embodiment of this application, the historical distribution characteristics of the risk priority values ​​corresponding to the connection points of each integrated terminal are obtained, including the following steps: S7411: Real-time acquisition of production batch information, ambient temperature information, equipment maintenance records, and equipment version information related to the equipment associated with the connection point of the integrated terminal; S7412: Segment the historical data of risk priority values ​​based on production batch information; S7413: Divide the segmented historical data into temperature ranges based on ambient temperature information; S7414: Based on equipment maintenance records, mark the maintenance cycle for historical data after the temperature range is divided; S7415: Based on the equipment version information, classify the historical data marked with the maintenance cycle according to the equipment version to obtain multiple subsets of the historical data after classification; S7416: For each subset of historical data, calculate the mean, variance, and skewness, and construct a dynamic distribution feature set of risk priority values ​​based on the mean, variance, and skewness corresponding to each subset of historical data. S7417: When production batch information, ambient temperature information, equipment maintenance records, or equipment version information changes, an update to the dynamic distribution feature set is triggered to obtain the updated historical distribution features.

[0080] Specifically, real-time acquisition of production batch information, ambient temperature information, equipment maintenance records, and equipment version information related to the equipment associated with the connection point of the integrated terminal refers to the system continuously updating and recording key contextual data affecting the operating status and risk priority values ​​of the connection point through sensor acquisition, database query, or manual input. Among them, production batch information is used to reflect the inherent characteristics of manufacturing process and material batch, ambient temperature information is used to characterize the thermal environment of the connection point and affect material performance and aging process, equipment maintenance records are used to record historical operations such as repair, maintenance, and replacement of parts to reflect possible changes in the status of the connection point due to maintenance, and equipment version information is used to reflect hardware or software design iterations and indicate possible new features or performance changes.

[0081] Furthermore, based on production batch information, historical data on risk priority values ​​are segmented to differentiate initial conditions or aging rates caused by different production batches and reduce statistical bias. Subsequently, based on ambient temperature information, the segmented historical data is divided into temperature ranges to categorize and analyze the impact of temperature on risk priority values; for example, high or low temperatures may accelerate or slow down the mechanical loosening process. On this basis, based on equipment maintenance records, the historical data after temperature range segmentation is marked with maintenance cycles to differentiate data distribution before and after maintenance or within different maintenance cycles and to assess the impact of maintenance activities on the connection point status; for example, maintenance-restored connection points may exhibit different risk priority value distributions. Finally, based on equipment version information, the historical data after maintenance cycle marking is categorized by equipment version to obtain multiple categorized historical data subsets. These subsets are grouped by equipment version and consider differences in design or performance between different versions, ensuring that each historical data subset corresponds to a specific operating condition constrained by production batch, ambient temperature, maintenance cycle, and equipment version.

[0082] For each subset of historical data, the mean, variance, and skewness are calculated. Based on the mean, variance, and skewness corresponding to each subset of historical data, a dynamic distribution feature set of risk priority values ​​is constructed. The mean represents the central tendency, the variance represents the dispersion, and the skewness represents the asymmetry of the distribution. The dynamic distribution feature set is used to summarize and characterize the distribution characteristics of risk priority values ​​under different operating conditions, providing a quantifiable basis for setting subsequent risk level boundary points. When production batch information, ambient temperature information, equipment maintenance records, or equipment version information changes, the dynamic distribution feature set is updated. During the update process, newly added historical data of risk priority values ​​within the update cycle are included to ensure that the historical distribution features can continuously reflect the latest operating conditions and equipment status and maintain timeliness and accuracy.

[0083] In another embodiment of this application, the step of setting the initial risk level boundary point of the risk level based on historical distribution characteristics and in conjunction with a preset risk management strategy includes: S7421: Real-time acquisition of production batch information, ambient temperature information, equipment maintenance records, and equipment version information related to the equipment associated with the connection point of the integrated terminal; S7422: Based on production batch information, ambient temperature information, equipment maintenance records, and equipment version information, historical data of risk priority values ​​are grouped in multiple dimensions to obtain multiple subsets of historical data after grouping. S7423: For each subset of historical data, calculate the mean, variance, and skewness, and construct a dynamic distribution feature set of risk priority values ​​under different working conditions based on the mean, variance, and skewness corresponding to each subset of historical data. S7424: Based on a dynamic distribution feature set, perform distribution change point detection and / or cluster analysis on risk priority values ​​to obtain the change points and / or natural cluster centers of the risk priority value distribution. S7425: In conjunction with a pre-defined risk management strategy, adjust the points of change and / or natural cluster centers to determine the initial risk level boundary points for the risk level; S7426: When production batch information, ambient temperature information, equipment maintenance records, or equipment version information change, the dynamic distribution feature set is reconstructed, and the initial risk level boundary point is reset.

[0084] Specifically, real-time acquisition of production batch information, ambient temperature information, equipment maintenance records, and equipment version information aims to provide multi-dimensional contextual information for subsequent refined analysis of historical data on risk priority values, thereby characterizing the specific operating environment and status of the integrated terminal connection point. Based on this multi-dimensional information, the historical data on risk priority values ​​are grouped to isolate data under different operating conditions and ensure the relevance of the analysis. For example, data from the same production batch, similar ambient temperature range, same maintenance cycle, and same equipment version are grouped together, making each subset of historical data more homogeneous.

[0085] For each subset of historical data after grouping, statistical parameters such as mean, variance, and skewness are calculated to quantify the central tendency, dispersion, and distribution shape of risk priority values ​​under that specific working condition. The mean reflects the average level, the variance characterizes the volatility, and the skewness reveals the symmetry of the distribution. Based on this, a dynamic distribution feature set under different working conditions is constructed to more comprehensively characterize the distribution characteristics of risk priority values ​​and provide a data basis for setting subsequent boundary points.

[0086] Furthermore, based on the dynamic distribution feature set, distribution change point detection and / or cluster analysis are performed to automatically identify significant change points and / or natural cluster centers in the risk priority numerical distribution. The change points may correspond to abrupt changes in statistical characteristics, and the natural cluster centers may correspond to concentrated areas of similar values. Together, they are used to characterize the natural boundary line from the normal state to the slight anomaly and then to the severe anomaly, thereby providing candidate boundary basis for the preliminary classification of risk levels.

[0087] Based on this, the identified change points and / or natural cluster centers are adjusted in conjunction with the preset risk management strategy to determine the initial risk level boundary points. The adjustment is used to align the data-driven candidate boundaries with the enterprise's risk tolerance, safety production standards, cost-benefit analysis and other constraints, to ensure that the initial risk level boundary points meet both statistical characteristics and business feasibility, and form a feasible initial classification basis.

[0088] In some preferred embodiments, the following specific example illustrates the situation: Assuming that the historical distribution of risk priority values ​​for a certain integrated terminal connection point varies significantly across different production batches, ambient temperatures (e.g., high temperatures in summer and low temperatures in winter), maintenance cycles (e.g., immediately after maintenance and near the maintenance period), and equipment versions, the system collects this operating condition information in real time. When it detects that the current production batch is "Batch A," the ambient temperature is in the "high temperature range," the equipment maintenance record shows "80% maintenance cycle completed," and the equipment version is "V2.0," the system filters out a subset of data that meets these conditions from historical data, calculates the mean, variance, and skewness of this subset, and constructs a dynamic distribution feature set for this specific operating condition. Subsequently, the system uses a Gaussian mixture model for cluster analysis to identify categories such as "normal," "slightly abnormal," and "severely abnormal." The system uses natural cluster centers for three risk levels: "normal," "high," and "low." It then fine-tunes the clustering results using preset risk management strategies (e.g., company regulations requiring immediate shutdown and inspection when risk priority values ​​exceed a certain threshold), thus setting precise initial risk level boundary points for the current operating conditions. When operating conditions change, such as an ambient temperature shifting from "high temperature range" to "low temperature range," the system triggers a dynamic reconstruction of the distribution feature set and filters out historical data subsets under the conditions of "Batch A," "Low Temperature Range," "80% Maintenance Cycle Completed," and "V2.0." It recalculates statistical parameters and performs cluster analysis to reset the initial risk level boundary points adapted to the "low temperature range" operating conditions. This ensures that risk level determination is based on a reference benchmark closest to actual operating conditions and provides more accurate early warnings.

[0089] Reference Figure 2 The specific embodiments of this application also disclose an integrated terminal connection status detection system based on sensors, the system comprising: Information acquisition module 1 is used to synchronously acquire instantaneous voltage information at both ends of the connection point of the integrated terminal, as well as instantaneous current information flowing through the integrated terminal; Impedance calculation module 2 is used to ensure that the instantaneous voltage information and instantaneous current information are time-correlated within the time window corresponding to the current impact event when the instantaneous current information meets the preset impact criterion to indicate the existence of a current impact event, and to calculate the transient impedance parameters of the connection point under impact based on the instantaneous voltage information and instantaneous current information. Feature extraction module 3 is used to extract transient features from transient impedance parameters to characterize the micro-contact state of the connection point; Reference range establishment module 4 is used to establish a normal operating reference range for transient characteristics when the integrated terminal is in a normal connection state. The change tracking module 5 is used to continuously acquire transient characteristics during the operation of the integrated terminal and compare the transient characteristics with the normal operation reference range to obtain the deviation or change trend of the transient characteristics; The judgment module 6 is used to determine whether the deviation or trend meets the preset early mechanical loosening risk criteria based on the deviation or trend. The early mechanical loosening risk criteria include: the deviation criterion is that the transient characteristics exceed the normal operation reference range, and / or the trend criterion is that the deviation continuously meets the preset deviation conditions within the preset observation window. The early warning module 7 is used to determine that there is an early mechanical loosening risk at the connection point of the integrated terminal when the deviation or trend meets the early mechanical loosening risk criterion, and outputs early warning information.

[0090] Through its modular design, the system enables real-time, highly sensitive detection of the connection status of integrated terminals.

[0091] The above description is merely an embodiment of this application and is not intended to limit the scope of protection of this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of this application should be included within the scope of protection of this application.

Claims

1. A sensor-based integrated terminal connection status detection method, characterized in that, The method includes: Simultaneously acquire instantaneous voltage information at both ends of the connection point of the integrated terminal, as well as instantaneous current information flowing through the integrated terminal; When the instantaneous current information satisfies the preset impact criterion to indicate the existence of a current impact event, within the time window corresponding to the current impact event, the instantaneous voltage information and the instantaneous current information are kept in time correspondence, and the transient impedance parameters of the connection point under impact are calculated based on the instantaneous voltage information and the instantaneous current information. Transient features characterizing the microscopic contact state of the connection point are extracted from the transient impedance parameters. When the integrated terminal is in a normal connection state, establish a normal operating reference range for the transient characteristics; During the operation of the integrated terminal, the transient characteristics are continuously acquired, and the transient characteristics are compared with the normal operation reference range to obtain the deviation or trend of the transient characteristics; Based on the deviation or trend of change, determine whether the deviation or trend of change meets the preset early mechanical loosening risk criteria. The early mechanical loosening risk criteria include: the criterion for the deviation is that the transient characteristics exceed the normal operation reference range, and / or the criterion for the trend of change is that the deviation continuously meets the preset deviation conditions within the preset observation window. When the deviation or trend of change meets the early mechanical loosening risk criterion, it is determined that there is an early mechanical loosening risk at the connection point of the integrated terminal, and a warning message is output.

2. The sensor-based integrated terminal connection status detection method according to claim 1, characterized in that, Extracting transient features from the transient impedance parameters to characterize the microscopic contact state of the connection point includes: The transient impedance parameters are subjected to high-pass filtering to isolate the low-frequency components in the transient impedance parameters; Spectral analysis is performed on the transient impedance parameters after high-pass filtering to obtain the energy characteristics of a preset high-frequency band. Within the time window corresponding to the current surge event, sampling points where the instantaneous current information meets the preset effective current conditions are selected. A voltage-current dynamic relationship trajectory is constructed based on the instantaneous voltage information and the instantaneous current information corresponding to the sampling point time. An impedance-current dynamic relationship trajectory is constructed based on the transient impedance parameter and the instantaneous current information. The geometric parameters of the voltage-current dynamic relationship trajectory and / or the impedance-current dynamic relationship trajectory are calculated. Differential operation is performed on the rising edge of the transient impedance parameter to detect whether there are impedance spikes and / or oscillations with a duration of a preset very short time. The transient feature extraction reference benchmark is dynamically learned and updated based on the energy characteristics, geometric parameters, and impedance spikes and / or oscillations detected under normal operating conditions. The transient characteristics are formed by extracting the degree of deviation and / or change pattern of the reference benchmark relative to the current energy characteristics, the geometric parameters, and the detection results of the impedance spikes and / or oscillations.

3. The sensor-based integrated terminal connection status detection method according to claim 1, characterized in that, Continuously acquire the transient characteristics and compare them with the normal operating reference range to obtain the deviation or trend of the transient characteristics, including: While continuously acquiring the transient characteristics, environmental vibration information and power grid fluctuation information related to the integrated terminal connection are acquired simultaneously; The transient characteristics are subjected to noise separation processing to separate the electromagnetic interference components indicated by the power grid fluctuation information from the transient characteristics, so as to obtain the separated transient characteristics; Feature correlation analysis is performed based on the separated transient features and the environmental vibration information to obtain correlation indicators that characterize the correlation between the separated transient features and the environmental vibration information; An environmental interference mode reference benchmark is established based on historical correlation indicators when the integrated terminal is in a normal connection state. The correlation index is compared with the environmental interference mode reference benchmark to identify environmental interference components that match the environmental interference mode, and the separated transient features are suppressed, eliminated and / or compensated to obtain the interference-free transient features. The transient characteristics after interference removal are compared with the normal operating reference range to obtain the deviation or trend of the transient characteristics.

4. The sensor-based integrated terminal connection status detection method according to claim 1, characterized in that, When the deviation or trend of change meets the early mechanical loosening risk criterion, after determining that the connection point of the integrated terminal has an early mechanical loosening risk, the method further includes: Acquire the criticality level information, load level information, and potential fault impact level information corresponding to the connection points of multiple integrated terminals. The criticality level information is used to characterize the importance of the connection point in the power supply circuit or production line. The load level information is used to characterize the load intensity of the connection point within a preset statistical time window. The potential fault impact level information is used to characterize the degree of impact of the failure of the connection point on power supply continuity or production. Based on the criticality level information, the load level information, and the potential fault impact level information, calculate the risk priority value of each connection point of the integrated terminal; Based on the risk priority value, an early warning ranking result is generated, so that multiple early warning messages are arranged according to a preset ranking rule of the risk priority value; Based on the correspondence between the risk priority values ​​and the preset grading rules, the risk level corresponding to each of the warning messages is determined; Outputting early warning information includes: outputting multiple early warning messages according to the early warning sorting result, and executing a corresponding preset notification strategy based on the risk level to issue graded early warning information.

5. The sensor-based integrated terminal connection status detection method according to claim 4, characterized in that, Obtaining the load level information includes: Within the time window corresponding to the current surge event, the energy dissipation value of the current surge event is calculated based on the instantaneous voltage information and the instantaneous current information. The cumulative energy dissipation is obtained by accumulating the energy dissipation values ​​of multiple current impact events within a preset statistical time window. The accumulated power dissipation is used as the load level information.

6. The sensor-based integrated terminal connection status detection method according to claim 4, characterized in that, Obtaining the criticality level information and the potential fault impact level information includes: For each connection point of the integrated terminal, the current production task priority of the device associated with the connection point of the integrated terminal is obtained in real time. Based on the current production task priority, the key process node identifiers of the associated equipment are determined, and the criticality level information is determined based on the current production task priority and the key process node identifiers according to the preset criticality assessment rules. A security event level library is pre-built. The security event level library includes the correspondence between connection point attributes and security event levels. The connection point attributes include the equipment category of the device associated with the connection point and / or the power supply circuit identifier to which the connection point belongs. Based on the connection point attributes, the security event level corresponding to the connection point is determined in the security event level library. Based on the safety event level and according to the preset loss assessment rules, the estimated production stoppage loss data is determined, and according to the preset impact assessment rules, the potential failure impact level information is determined based on the safety event level and the estimated production stoppage loss data.

7. The sensor-based integrated terminal connection status detection method according to claim 4, characterized in that, Based on the correspondence between the risk priority values ​​and preset grading rules, the risk level corresponding to each of the warning messages is determined, including: Obtain the historical distribution characteristics of the risk priority values ​​corresponding to the connection points of each integrated terminal; Based on the historical distribution characteristics and in conjunction with the preset risk management strategy, the initial risk level boundary point of the risk level is set to form the preset classification rule; The new risk priority value generated during the operation is compared with the initial risk level boundary point to determine the risk level corresponding to the new risk priority value, and the number of times the risk level is switched within the preset observation window is recorded. When the new risk priority value reaches a preset number of times within the preset boundary tolerance range, and the number of switching reaches a preset switching threshold, it is determined that there is an ambiguous region at the initial risk level boundary point. For the fuzzy region, risk priority numerical samples falling into the fuzzy region and operation and maintenance handling result information corresponding to the risk priority numerical samples are collected. The operation and maintenance handling result information includes actual fault consequence information and / or operation and maintenance personnel feedback information. Based on the risk priority numerical sample and the operation and maintenance handling result information, the initial risk level boundary point is adjusted to obtain the updated risk level boundary point. The boundary clarity index is calculated based on the updated risk level boundary points, and if the boundary clarity index does not meet the preset boundary clarity requirements, the risk level boundary points are iteratively updated until the preset boundary clarity requirements are met. Based on the risk level boundary points that meet the preset boundary clarity requirements, the risk priority values ​​are range-determined to determine the risk level corresponding to each of the warning messages.

8. The sensor-based integrated terminal connection status detection method according to claim 7, characterized in that, Obtain the historical distribution characteristics of the risk priority values ​​corresponding to the connection points of each integrated terminal, including: Real-time acquisition of production batch information, ambient temperature information, equipment maintenance records, and equipment version information related to the equipment associated with the connection point of the integrated terminal; Based on the production batch information, the historical data of the risk priority value is segmented; Based on the ambient temperature information, the segmented historical data is divided into temperature ranges; Based on the equipment maintenance records, the historical data after the temperature range is divided is marked with maintenance cycles; Based on the device version information, the historical data marked with maintenance cycle is categorized by device version to obtain multiple categorized subsets of historical data; For each subset of historical data, the mean, variance, and skewness are calculated, and a dynamic distribution feature set of the risk priority values ​​is constructed based on the mean, variance, and skewness corresponding to each subset of historical data. When the production batch information, the ambient temperature information, the equipment maintenance record, or the equipment version information changes, an update to the dynamic distribution feature set is triggered to obtain the updated historical distribution features.

9. The sensor-based integrated terminal connection status detection method according to claim 7, characterized in that, Based on the historical distribution characteristics and in conjunction with the preset risk management strategy, the initial risk level boundary point of the risk level is set, including: Real-time acquisition of production batch information, ambient temperature information, equipment maintenance records, and equipment version information related to the equipment associated with the connection point of the integrated terminal; Based on production batch information, ambient temperature information, equipment maintenance records, and the equipment version information, the historical data of the risk priority value is grouped in multiple dimensions to obtain multiple subsets of historical data after grouping. For each subset of historical data, the mean, variance, and skewness are calculated, and a dynamic distribution feature set of the risk priority value under different working conditions is constructed based on the mean, variance, and skewness corresponding to each subset of historical data. Based on the dynamic distribution feature set, perform distribution change point detection and / or cluster analysis on the risk priority values ​​to obtain the change points and / or natural cluster centers of the risk priority value distribution. In conjunction with the preset risk management strategy, the change points and / or natural cluster centers are adjusted to determine the initial risk level boundary points of the risk level; When the production batch information, the ambient temperature information, the equipment maintenance record, or the equipment version information changes, the reconstruction of the dynamic distribution feature set is triggered, and the initial risk level boundary point is reset.

10. A sensor-based integrated terminal connection status detection system, characterized in that, The system includes: The information acquisition module is used to synchronously acquire the instantaneous voltage information at both ends of the connection point of the integrated terminal, as well as the instantaneous current information flowing through the integrated terminal; The impedance calculation module is used to ensure that the instantaneous voltage information and the instantaneous current information are time-correlated within a time window corresponding to the current impact event when the instantaneous current information meets the preset impact criterion to indicate the existence of a current impact event, and to calculate the transient impedance parameters of the connection point under impact based on the instantaneous voltage information and the instantaneous current information. The feature extraction module is used to extract transient features from the transient impedance parameters to characterize the microscopic contact state of the connection point; A reference range establishment module is used to establish a normal operating reference range for the transient characteristics when the integrated terminal is in a normal connection state. The change tracking module is used to continuously acquire the transient characteristics during the operation of the integrated terminal, and compare the transient characteristics with the normal operation reference range to obtain the deviation or change trend of the transient characteristics; The judgment module is used to determine whether the deviation or change trend meets the preset early mechanical loosening risk criteria based on the deviation or change trend. The early mechanical loosening risk criteria include: the criterion for the deviation is that the transient characteristics exceed the normal operation reference range, and / or the criterion for the change trend is that the deviation continuously meets the preset deviation conditions within the preset observation window. The early warning module is used to determine that there is an early mechanical loosening risk at the connection point of the integrated terminal when the deviation or trend meets the early mechanical loosening risk criterion, and outputs early warning information.

Images