Abnormal state early warning system and method for secondary circuit wiring terminal

Abstract

The invention relates to the technical field of power system automation, and discloses an abnormal state early warning system and method for a secondary circuit wiring terminal. The system comprises a parameter synchronous acquisition module, an electrical characteristic analysis module, a temperature rise correlation analysis module, a heat dissipation state switching and environment sensing module, a heat source property distinguishing module and a comprehensive positioning and defense module. According to the method, the temperature rise correlation is judged by constructing a thermoelectric coupling model by synchronously collecting the electrical quantity and the temperature quantity; a thermal inertia time constant is calculated by utilizing active aerodynamic disturbance of a cooling fan, so that functional heating and structural-level contact heating of the chip are distinguished; and meanwhile, the cabinet environment state is sensed based on fan impedance characteristics. The system can integrate the full-loop waveform distortion index and the heat source property, accurately judge the physical position of the fault and execute hierarchical defense, and effectively solves the problems that the fault type is difficult to distinguish and the positioning is fuzzy in the prior art.

Description

Technical Field

[0001] This invention relates to the field of power system automation technology, specifically to an abnormal state early warning system and method for secondary circuit terminals. Background Technology

[0002] The secondary circuit terminals in substations bear the critical task of transmitting voltage and current signals, and their contact status directly affects the stable operation of relay protection and control devices. Currently, abnormal condition monitoring of such connection components typically employs contact thermistors or non-contact infrared imaging temperature measurement technology. The basic working principle is to acquire the temperature values ​​of the terminal blocks or equipment backplane area in real time and compare the collected temperature data with the system's preset static safety threshold. When the monitored temperature amplitude or rate of temperature rise exceeds the set limit, the monitoring device will issue an overheat alarm signal. This temperature amplitude-based monitoring method, under conditions of stable equipment load and constant environment, can provide a certain degree of indication for obvious overheating faults.

[0003] However, under the complex operating conditions of substations, the aforementioned monitoring methods have limitations in their detection dimensions. The functional temperature rise of chips and the structural temperature rise caused by poor terminal contact during heavy load operation show a high degree of similarity in their temperature increase trends, making it difficult to effectively identify the physical properties of the heat source based solely on temperature amplitude. Furthermore, when nonlinear electrical characteristics caused by poor contact appear in the secondary circuit, existing technologies lack a mechanism to jointly analyze electrical waveform distortion and thermodynamic behavior, making it difficult to determine whether the root cause of the fault lies in a contact defect located on the equipment's local backplane or in electrical interference from a remote external circuit. This insufficient ability to physically locate faults makes it difficult for the system to quickly pinpoint the specific fault location when faced with complex fault characteristics.

[0004] Furthermore, the judgment logic based on temperature thresholds is highly dependent on the stability of the heat dissipation environment. During long-term operation, issues such as cabinet doors not being properly closed or dust accumulation clogging the air intake ducts may occur, altering the aerodynamic impedance and heat dissipation efficiency within the cabinet. Existing monitoring systems typically lack integrated sensing capabilities for the heat dissipation environment itself, and without dedicated sensors, they cannot identify changes in the physical state of airflow channels. When changes in heat dissipation conditions cause a passive increase in equipment temperature, the system may easily misinterpret this as an electrical contact fault. This lack of environmental condition awareness affects the objectivity and accuracy of monitoring results to some extent. Summary of the Invention

[0005] To address the shortcomings of existing technologies, this invention provides an abnormal state early warning system and method for secondary circuit terminals. This solves the problem that existing monitoring technologies rely solely on a single temperature or current threshold for alarms, failing to deeply correlate electrical and thermodynamic characteristics, thus making it difficult to effectively distinguish between functional heating of the chip and structural contact heating.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution: an abnormal state early warning system for secondary circuit terminals, comprising a parameter synchronous acquisition module, an electrical characteristic analysis module, a temperature rise correlation analysis module, a heat dissipation state switching and environmental perception module, a heat source property identification module, and a comprehensive positioning and defense module.

[0007] The parameter synchronous acquisition module is configured to connect with the underlying hardware of the intelligent measurement and control device, read the real-time sampling sequence of the secondary circuit load current and the onboard temperature data, and synchronously acquire the speed and drive current of the cooling fan. After aligning the acquired real-time sampling sequence, onboard temperature data, speed and drive current in time, the module is output.

[0008] The electrical characteristic analysis module is connected to the parameter synchronous acquisition module to receive current waveform data. By extracting frequency domain features, it calculates the waveform distortion index of the entire circuit and identifies whether there are nonlinear electrical fault characteristics in the circuit.

[0009] The temperature rise correlation analysis module is connected to the parameter synchronous acquisition module. It is used to construct a thermoelectric dynamic coupling model using the aligned temperature sequence and current sequence to determine whether the on-board temperature rise has a current correlation.

[0010] The heat dissipation state switching and environmental perception module are connected to the parameter synchronization acquisition module and the temperature rise correlation analysis module. When the temperature rise has a current correlation, the cooling fan is controlled to switch to the open-loop step control mode to apply active aerodynamic disturbance. It is also used to calculate the aerodynamic impedance deviation based on the fan drive current.

[0011] The heat source nature identification module is connected to the heat dissipation state switching and environmental sensing module. It is used to collect the transient thermal response curve of the on-board temperature and calculate the thermal inertia time constant during the application of active aerodynamic disturbance. Based on the magnitude of the thermal inertia time constant, it distinguishes whether the heat source is chip-level functional heat generation or structural-level contact heat generation.

[0012] The integrated positioning and defense module is connected to the electrical characteristic analysis module, temperature rise correlation analysis module, heat dissipation state switching and environmental perception module, and heat source nature identification module, respectively. It is used to comprehensively analyze the whole circuit waveform distortion index, temperature rise current correlation, aerodynamic impedance state and heat source nature classification results to determine the physical location and nature of the fault and execute a graded defense strategy.

[0013] In a preferred embodiment, the parameter synchronization acquisition module establishes a baseline correspondence between fan speed and load torque. This baseline correspondence is based on the principle of motor torque balance, specifically by constructing a fan load dynamics model using the motor torque constant, aerodynamic drag coefficient, fan rotational angular velocity, and motor bearing viscous friction coefficient. Under the baseline state of the cabinet door being closed and the dust filter being clean, the abnormal state early warning system controls the fan to operate at multiple steady-state speeds, acquires the drive current values ​​under each condition, and performs numerical fitting based on the fan load dynamics model to generate a standard impedance characteristic curve.

[0014] Furthermore, the electrical feature analysis module performs frequency domain feature extraction. This extraction process includes performing a Fast Fourier Transform on the acquired raw current waveform sequence, decomposing the current signal in the frequency domain into the fundamental frequency component, integer harmonic components, and high-frequency residual noise components in the non-harmonic frequency band. The calculation logic for the full-loop waveform distortion index is as follows: the ratio of the sum of the squares of the effective values ​​of each integer harmonic current to the effective value of the fundamental frequency current is calculated to characterize the degree of conventional harmonic distortion. The spectral energy proportion of the high-frequency residual noise component is introduced as a nonlinear correction factor, and the full-loop waveform distortion index is generated by weighting and correcting the degree of conventional harmonic distortion.

[0015] In one specific implementation, the temperature rise correlation analysis module constructs a thermoelectric dynamic coupling model. Based on Joule's law, this model establishes a physical rule that temperature change is proportional to the square of the effective current value, and introduces a thermal hysteresis time parameter representing the rate of heat conduction. The temperature rise correlation analysis module calculates the theoretically estimated temperature for the current moment based on historical current data, compares the theoretically estimated temperature with the actually collected onboard temperature, and calculates the numerical deviation between the two. When the numerical deviation remains within the allowable error range, the current onboard temperature rise is determined to have a current correlation.

[0016] The heat dissipation state switching and environmental sensing module controls the cooling fan to switch to open-loop step control mode. Specifically, it temporarily stops the intelligent speed regulation logic that automatically adjusts the fan speed according to the temperature, and directly sends a pre-set high speed command to the fan, forcing the fan to leave the current state and run continuously with a constant high airflow. This creates a stable and strong airflow cooling environment inside the rack, allowing the heat-generating components to enter a rapid cooling state.

[0017] Furthermore, the heat source identification module distinguishes heat sources. The process of distinguishing heat sources includes analyzing the collected temperature-time curve and calculating the thermal inertia time constant, which represents the rate of cooling at the heat source. The thermal inertia time constant is compared with a preset benchmark value: when the thermal inertia time constant is less than the benchmark value, it indicates a fast cooling rate, and is determined to be self-heating generated by the chip's operation; when the thermal inertia time constant is greater than or equal to the benchmark value, it indicates a slow cooling rate, and is determined to be heat generated by poor contact at structural joints.

[0018] In the implementation of environmental sensing, the heat dissipation state switching and environmental sensing module identify the rack door status and the degree of air intake channel blockage. The method for identifying the rack door status and air intake channel blockage includes finding the theoretically expected reference current value at the current actual fan speed on a pre-generated standard impedance characteristic curve; comparing the actual fan drive current with the reference current value, and calculating the difference between the two as the aerodynamic impedance deviation. When the difference is lower than a set negative threshold, it indicates an abnormal decrease in fan operating resistance, and the rack door is determined to be open; when the difference is higher than a set positive threshold, it indicates an abnormal increase in fan operating resistance, and the air intake channel is determined to be blocked.

[0019] The logic for determining the physical location of a fault using the integrated positioning and defense module is as follows: When the full-circuit waveform distortion index exceeds the preset normal range, but the onboard temperature rise does not have a current correlation, the fault point is determined to be outside the equipment, belonging to a remote circuit contact hazard; when the full-circuit waveform distortion index exceeds the preset normal range, and the onboard temperature rise has a current correlation, and the heat source is confirmed to be structural-level contact heating, the fault point is determined to be located at the equipment backplane, belonging to a local backplane contact fault.

[0020] For different fault locations, the integrated location and defense module implements differentiated defense strategies: if the fault point is determined to be outside the equipment, only an alarm signal is issued to prompt the staff to check the external lines, and the current normal operation of the equipment is maintained; if the fault point is determined to be at the back panel of the equipment, two-level protection measures are immediately activated: the first level measure is to directly adjust the cooling fan to the maximum speed to forcibly reduce the temperature; if the onboard temperature continues to rise and exceeds the set safety limit value when the cooling fan is running at the maximum speed, the second level measure is executed, and a trip command is sent to cut off the current circuit to prevent the wiring terminals from burning out.

[0021] A second aspect of the present invention provides a method for early warning of abnormal conditions of secondary circuit terminals.

[0022] The abnormal state early warning method includes the following steps:

[0023] By utilizing high-precision synchronous sampling technology, electrical quantity data and onboard temperature data with strictly aligned timestamps of the secondary circuit are acquired in real time to construct a multi-dimensional synchronous dataset.

[0024] Frequency domain feature extraction is performed on the acquired current waveform to separate the power frequency fundamental wave from the high-frequency noise component, and the waveform distortion index of the entire circuit is calculated to identify whether there are nonlinear electrical fault characteristics caused by poor contact in the circuit.

[0025] A thermoelectric dynamic coupling model based on Joule's law is constructed, and the temperature residual value is calculated using the temperature sequence and current sequence within the sliding time window to determine whether the current on-board temperature rise has a current correlation.

[0026] Real-time monitoring of the drive current change of the heat dissipation actuator at a constant speed, and calculation of the aerodynamic impedance deviation, are used to identify the opening and closing status of the cabinet door and the degree of blockage of the air intake channel.

[0027] When an abnormal temperature rise with current correlation is detected, the control heat dissipation execution unit switches to open-loop step control mode to apply active aerodynamic disturbance, collects the transient thermal response curve of the heat source under strong convection heat dissipation field, and calculates the thermal inertia time constant to distinguish whether the heat source is chip-level functional heat generation or structural-level contact heat generation.

[0028] By combining electrical waveform distortion index, temperature rise current correlation, aerodynamic impedance state and thermal inertia time constant, a multidimensional fault feature vector is constructed. Based on spatiotemporal correlation logic, the physical location and nature of the fault are determined, and a graded defense strategy including alarm, blocking or tripping is executed accordingly.

[0029] This invention provides an abnormal state early warning system and method for secondary circuit terminals. It has the following beneficial effects:

[0030] 1. This invention applies an open-loop step control signal to the cooling fan through a heat dissipation state switching and environmental sensing module, constructing a strong convective heat dissipation field within the cabinet and calculating the thermal inertia time constant. This mechanism utilizes the significant differences in physical materials and volume between semiconductor chips and metal terminals. By quantifying the transient cooling rate of the heat source under forced cooling conditions, it can accurately distinguish between chip-level functional heat generation and structural-level contact heat generation, effectively solving the technical problem that traditional monitoring methods cannot identify the type of fault source based solely on temperature values.

[0031] 2. This invention, through the collaborative operation of the electrical characteristic analysis module and the temperature rise correlation analysis module, constructs a joint judgment logic for the correlation between the full-loop waveform distortion index and the onboard temperature rise current. When nonlinear distortion of the current waveform is detected, the system can accurately determine whether the fault point is located in a remote circuit outside the equipment or on the backplane of the equipment itself, based on whether the local temperature change conforms to the thermoelectric coupling model of Joule's law. This avoids erroneous disconnection operations caused by external line interference and achieves precise isolation of the physical location of the fault.

[0032] 3. This invention establishes a baseline relationship between fan speed and drive current, and calculates the aerodynamic impedance deviation in real time, achieving sensorless sensing of the cabinet ventilation environment. This method utilizes the dynamic balance principle between fan load torque and fluid resistance, enabling accurate identification of abnormal states such as accidental opening of the cabinet door or blockage of the air intake channel without increasing additional hardware costs. This ensures the reliable operation of the equipment's heat dissipation system and guarantees the accuracy of the thermal environment data required for heat source identification. Attached Figure Description

[0033] Figure 1 This is a system framework diagram of the present invention;

[0034] Figure 2 This is a detailed flowchart of the parameter synchronization acquisition module of the present invention;

[0035] Figure 3 This is a flowchart illustrating the collaborative workflow between the electrical characteristic analysis module and the temperature rise correlation analysis module of the present invention.

[0036] Figure 4 This is a flowchart of the heat source property identification based on active aerodynamic disturbance of the present invention;

[0037] Figure 5 This is a flowchart of the cabinet environment perception and physical state monitoring based on aerodynamic impedance changes according to the present invention.

[0038] Figure 6 This is a flowchart illustrating the fault determination and defense strategy execution of the integrated positioning and defense module of the present invention.

[0039] Figure 7 This is a flowchart of the method of the present invention.

[0040] Among them, 10 is the parameter synchronous acquisition module; 20 is the electrical characteristic analysis module; 30 is the temperature rise correlation analysis module; 40 is the heat dissipation state switching and environmental perception module; 50 is the heat source property identification module; and 60 is the comprehensive positioning and defense module. Detailed Implementation

[0041] The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0042] See attached document Figure 1 , Figure 1 This is a structural block diagram of an abnormal state early warning system for secondary circuit terminals according to an embodiment of the present invention. The present invention provides an abnormal state early warning system for secondary circuit terminals, comprising:

[0043] The parameter synchronous acquisition module 10 is configured to read the operating data inside the intelligent measurement and control device, ensure that the data such as secondary circuit current and PCB board temperature are aligned in time sequence, and synchronously acquire the drive current or real-time power data of the cooling fan motor to establish a benchmark correspondence between fan speed and load torque.

[0044] The electrical characteristic analysis module 20 is configured to analyze the current waveform using the device's waveform recording function, and to monitor potential contact hazards at any position in the circuit by identifying minute distortions in the current waveform, utilizing the full-circuit transmission characteristics of electrical signals, and generating abnormal signals characterizing the electrical connection status of the entire circuit.

[0045] Temperature rise correlation analysis module 30 is configured to compare the changing trends of board temperature and load current. Based on Joule's law, when a significant positive correlation is found between the board temperature change trend and current fluctuation, it is preliminarily determined that the heat source is resistive heating from the current loop.

[0046] The heat dissipation state switching and environmental perception module 40 is configured to establish a stable strong convection heat dissipation environment when a suspected abnormality is detected. At the same time, it uses the principle of fluid mechanics and the physical characteristic that opening the cabinet door will cause the aerodynamic impedance of the air intake channel to decrease, and judges the cabinet door status by identifying the change in fan load torque.

[0047] The heat source identification module 50 is configured to monitor the rate of temperature drop of the board after the fan runs at full speed. It utilizes the physical difference between the large thermal inertia of external cables and the small thermal inertia of chips to distinguish the faulty heat source attributes based on the rate of temperature drop.

[0048] The integrated positioning and defense module 60 is configured to determine the fault location by combining the results of global electrical feature analysis and local thermal feature analysis, and to correlate electrical abnormality signals with cabinet door opening signals in a time sequence to identify security abnormalities. At the same time, it assesses the degree of dust screen blockage based on the long-term trend of fan load current.

[0049] The specific implementation methods of each module of the present invention will be described in detail below with reference to the accompanying drawings.

[0050] See attached document Figure 2 The parameter synchronization acquisition module 10 is configured to interact with the underlying hardware of the intelligent measurement and control device to obtain electrical and physical environmental parameters during device operation. The intelligent measurement and control device described in this embodiment refers to an industrial field electronic device equipped with a microprocessor computing unit, a high-precision analog quantity acquisition circuit, and peripheral control interfaces, specifically including but not limited to microprocessor protection devices, integrated measurement and control units, or intelligent control devices for switchgear. At the hardware level, the intelligent measurement and control device integrates a current transformer interface for acquiring current and voltage signals, an onboard sensor for monitoring internal status, and an active fan assembly for heat dissipation. The parameter synchronization acquisition module 10 performs the following steps to achieve data acquisition, timing alignment, and benchmark construction:

[0051] Step S101: Configure the underlying hardware data reading interface. The parameter synchronous acquisition module 10 establishes a communication connection with the analog-to-digital converter, onboard sensors, and fan control unit by calling the driver program inside the intelligent measurement and control device. For current data, the parameter synchronous acquisition module 10 directly reads the sampling buffer of the analog-to-digital converter to obtain the real-time sampling sequence of the secondary circuit load current. The real-time sampling sequence must contain the raw waveform data without digital filtering, and the sampling frequency is set to no less than 80 points per cycle to retain high-frequency signal components for subsequent contact fault characteristic analysis. For temperature data, the parameter synchronous acquisition module 10 reads the thermistor value located in the power device area of ​​the printed circuit board as the onboard temperature and reads the temperature sensor value located at the air inlet of the device as the ambient temperature inside the cabinet.

[0052] Step S102: Acquire fan operating parameters and establish an aerodynamic impedance model. The parameter synchronous acquisition module 10 reads the cooling fan speed feedback signal in real time and synchronously acquires the real-time drive current or power data of the fan motor. In order to use the fan as a sensor to sense air fluid impedance, this step establishes a mapping relationship between the drive current and the air resistance torque based on the motor torque balance principle. The specific calculation model is as follows:

[0053] ;

[0054] in, The collected fan motor drive current, Let be the torque constant of the motor. The aerodynamic drag coefficient (which is directly related to the unobstructedness of the air intake channel and the back pressure of the cabinet). Let be the angular velocity of the fan rotation. Let be the coefficient of viscous friction of the motor bearing. This is the inherent Coulomb friction torque. The parameter synchronous acquisition module 10 uses this formula to convert the acquired current and speed data into physical indicators characterizing the external air resistance, given the known inherent parameters of the motor.

[0055] Step S103: Time-series alignment processing of multi-source data. Given that the current sampling frequency is much higher than the update frequency of temperature and fan data, the parameter synchronization acquisition module 10 internally sets up a sliding buffer queue to process the data stream. The parameter synchronization acquisition module 10 uses the update time of the temperature data as the reference time and extracts current data and fan parameters within a preset time window before and after the reference time. For current data, the parameter synchronization acquisition module 10 calculates the effective value within the preset time window as a steady-state parameter, while simultaneously storing the original waveform segment; for fan speed and drive current, the parameter synchronization acquisition module 10 uses the arithmetic mean method to eliminate fluctuations during the commutation process. Through step S103, the parameter synchronization acquisition module 10 strictly matches electrical, thermal, and mechanical quantities with different sampling rates on the time axis, forming a synchronized dataset containing the effective value of current, current waveform, onboard temperature, ambient temperature, fan speed, and drive current.

[0056] Step S104: Construct a baseline characteristic curve for fan speed versus load torque. During the initial system configuration phase, after confirming that the cabinet door is closed and the dust filter is clean, the parameter synchronization acquisition module 10 controls the cooling fan to operate in a stepped manner within a preset speed range. At each stable speed point, the parameter synchronization acquisition module 10 records the corresponding drive current value and speed value, and generates a standard impedance characteristic curve through data fitting. The standard impedance characteristic curve serves as a reference for subsequent system operation. When the real-time monitored drive current value deviates significantly from the standard impedance characteristic curve, it indicates that the aerodynamic drag coefficient... Changes have occurred, which leads to the inference of changes in the physical environment of the server rack.

[0057] The analog-to-digital conversion sampling principle, thermistor temperature measurement technology, and motor drive control logic involved in the above steps are mature applications in the field of electronic technology. This embodiment directly adopts existing technical solutions and will not be elaborated further here. The parameter synchronous acquisition module 10, through the above processing, provides standardized data support for subsequent fault diagnosis and environmental perception.

[0058] See attached document Figure 3 The electrical feature analysis module 20 and the temperature rise correlation analysis module 30 work together, using the synchronous dataset output by the parameter synchronous acquisition module 10 to perform non-invasive monitoring from two dimensions: electrical waveform characteristics and thermo-electrical coupling relationship. The electrical feature analysis module 20 and the temperature rise correlation analysis module 30 identify contact hazards and classify heat source attributes by performing the following steps:

[0059] The electrical characteristic analysis module 20 is configured to identify nonlinear contact faults in the circuit based on the full-loop transmission characteristics of the electrical signal. Since the secondary circuit is a series circuit in electrical connection, according to Kirchhoff's current law, any micro-arc effect or nonlinear resistance change caused by poor contact at any location in the circuit (including the remote terminal box located outside the switchgear) will act as an interference source superimposed on the current signal and transmitted to the measurement and control device. The electrical characteristic analysis module 20 utilizes this physical characteristic to perform the following processing:

[0060] Step S201: High-frequency feature extraction of the current waveform. The electrical feature analysis module 20 performs a fast Fourier transform on the input raw current waveform data, converting the time-domain signal into a frequency-domain signal. The electrical feature analysis module 20 is equipped with digital filter logic to separate the 50Hz or 60Hz power frequency fundamental component and integer harmonic components in the frequency domain, thereby extracting the broadband high-frequency non-stationary signal caused by the breakdown of the contact surface oxide layer and micro-arc discharge.

[0061] Step S202: Calculate the waveform distortion index of the entire circuit. To quantify the severity of the contact hazard, the electrical characteristic analysis module 20 calculates the waveform distortion index based on the separated spectral components. The specific calculation model is as follows:

[0062] ;

[0063] in, This is the waveform distortion index. This is the effective value of the fundamental frequency current. The effective value of the nth harmonic current. The maximum harmonic order is set (40 in this embodiment). The total energy of the current signal. The residual noise energy in the non-harmonic frequency band is equal to the total energy minus the fundamental frequency energy and all preceding frequencies. The residual value after the subharmonic energy This is the noise weighting coefficient, used to adjust the proportion of non-periodic random contact noise in the distortion index. The electrical characteristic analysis module 20 uses this formula to comprehensively characterize both periodic waveform distortion and non-periodic contact noise.

[0064] Step S203: Generate an electrical anomaly early warning signal. The electrical feature analysis module 20 will calculate in real time... The comparison is made with a preset benchmark threshold. To accommodate the differences in electromagnetic environments at various industrial sites, the benchmark threshold in this embodiment is not a single fixed constant, but rather constructed based on a dynamic background noise baseline and superimposed with a preset safety margin. Specifically, the average level of the full-loop waveform distortion index during the initial operation of the system is used as the background noise baseline. And set the baseline threshold to ( This is the sensitivity coefficient, typically taken as 1.5 to 2.0.

[0065] like If the value continuously exceeds the benchmark threshold for a preset time window (e.g., 5 consecutive power frequency cycles) to eliminate interference from transient switching surges, it indicates the presence of nonlinear impedance components in the circuit. At this time, the electrical characteristic analysis module 20 generates a signal characterizing the abnormal electrical connection state of the entire circuit. The abnormal signal clearly indicates the presence of potential contact defects in the circuit, but it has not yet distinguished whether the fault point is located on the local backplane or at a remote terminal.

[0066] The temperature rise correlation analysis module 30 is configured to distinguish the physical properties of heat sources. By analyzing the dynamic correlation between heat and current, it eliminates non-fault-related temperature rises caused by increased chip computing load. The temperature rise correlation analysis module 30 performs the following processing:

[0067] Step S301: Construct a thermo-electric dynamic coupling model. Based on Joule's law, the heating power of resistive contact is proportional to the square of the current. The temperature rise correlation analysis module 30 utilizes the onboard temperature sequence within a sliding time window. and current RMS value sequence A linear regression residual analysis model was constructed to determine the correlation between the two. The calculation formula is as follows:

[0068] ;

[0069] in, This is the temperature residual value. The current temperature of the board. For ambient temperature, The contact resistance thermal sensitivity coefficient of the circuit characterizes the temperature rise caused by the square of a unit current. For at any time The square of the effective value of the secondary circuit load current. The heat conduction delay time characterizes the hysteresis characteristic of current-induced heat conduction to the temperature sensor. This is a constant term. In this embodiment of the invention, and These are the inherent parameters of the system obtained by least-squares fitting and calibration of historical operating data when the device is in a healthy state.

[0070] Step S302: Logical determination of heat source attributes. Temperature rise correlation analysis module 30 monitors temperature residual values ​​in real time. The range of change. If the temperature residual value... Maintaining a near-zero preset error range indicates that the temperature change trajectory and the current square change trajectory conform to the pre-calibrated physical model, and the temperature rise is mainly caused by the resistive heating effect of the current loop. In this case, the temperature rise correlation analysis module 30 determines that the current temperature rise has a current correlation. If the absolute temperature value exceeds the limit, it is determined that there is an overheating fault in the current loop. If the temperature residual value... A significant positive deviation occurred, namely the onboard temperature sequence. The increase far exceeded the current. The range that the change can explain indicates the existence of an additional heat source independent of the current loop. At this point, the temperature rise correlation analysis module 30 determines that the temperature rise is caused by the high-load operation of the onboard chip, which is a non-faulty functional heating.

[0071] Step S303: Output preliminary screening results. The temperature rise correlation analysis module 30 combines the output of the electrical feature analysis module 20 to make a comprehensive judgment. When the electrical feature analysis module 20 detects waveform distortion and the temperature rise correlation analysis module 30 determines that the temperature rise has a current correlation, the system confirms the existence of a real electrical contact fault and triggers subsequent modules to accurately identify the fault location.

[0072] The Fast Fourier Transform algorithm, digital filtering technique, and linear regression analysis method involved in the above steps are well-known technologies in the field of signal processing and data analysis. Those skilled in the art can implement them using corresponding software algorithms according to actual needs, and will not be elaborated further here.

[0073] See attached document Figure 4 Based on active control theory, the heat dissipation state switching and environmental perception module 40 and the heat source property identification module 50 construct a closed-loop identification mechanism for the physical properties of heat sources. The heat dissipation state switching and environmental perception module 40 is responsible for switching heat dissipation conditions, while the heat source property identification module 50 is responsible for analyzing thermal response data. Together, they achieve accurate classification of heat source properties by executing the following steps:

[0074] Step S401: Apply active aerodynamic disturbance. When the system receives a temperature anomaly signal with current correlation from the temperature rise correlation analysis module 30, the heat dissipation state switching and environmental sensing module 40 executes control logic switching, interrupting the original PID closed-loop regulation process of the cooling fan based on temperature feedback, and converting the fan control mode to an open-loop step control mode. The heat dissipation state switching and environmental sensing module 40 sends a full-scale (100% duty cycle) or preset high-speed control command to the fan drive unit, forcing the cooling fan into a constant high airflow operating state. This operation creates a strong convective heat dissipation field with constant flow velocity inside the cabinet, causing a step increase in the convective heat transfer coefficient flowing over the surface of the heat-generating element, forcing the heat-generating element into a transient cooling process.

[0075] Step S402: Acquire transient thermal response curve. At the initial moment when the heat dissipation state switching and environmental sensing module 40 applies aerodynamic disturbance, the heat source property identification module 50 simultaneously starts high-frequency sampling and recording, reading the values ​​of the onboard temperature sensor at a frequency of not less than 10Hz, and generating a temperature decrease trajectory sequence over time. To eliminate the influence of the ambient temperature baseline, the heat source property identification module 50 converts the acquired absolute temperature sequence into a temperature rise decay sequence relative to the ambient temperature inside the cabinet. The acquisition process continues until the absolute value of the temperature change rate is less than a preset convergence threshold, or reaches a preset diagnostic time window (e.g., 60 seconds), thereby obtaining a complete step response curve.

[0076] Step S403: Calculate the thermal inertia time constant. The heat source nature identification module 50 identifies parameters from the acquired transient thermal response curves based on Newton's law of cooling and the lumped parameter method. Given the significant differences in the physical structure of heat sources with different properties: chip-type functional heat sources consist of microgram-level silicon wafers and packaging materials, resulting in extremely low heat capacity; while terminal contact-type fault heat sources involve large-mass metal components such as copper busbars, bolts, and cables, resulting in higher heat capacity. This difference in physical properties manifests as an order-of-magnitude difference in the thermal inertia time constant under strong air cooling conditions. The heat source nature identification module 50 calculates the real-time time constant of the current heat source using the following exponential decay model:

[0077] ;

[0078] Taking the logarithm of both sides of the above equation and transforming it, the time constant can be calculated.

[0079] ;

[0080] in, for The difference between the onboard temperature and the ambient temperature at any given time;

[0081] The initial temperature rise at the instant the disturbance is applied;

[0082] is the base of the natural logarithm (a constant approximately equal to 2.71828);

[0083] This is a time variable calculated from the start of the perturbation.

[0084] The thermal inertia time constant to be identified;

[0085] This refers to a later sampling time selected during the diagnostic process;

[0086] This refers to the earliest sampling time selected during the diagnostic process;

[0087] The natural logarithm operator;

[0088] for Temperature rise at any given moment;

[0089] for The temperature rise at any given moment. Furthermore, the thermal inertia time constant is defined in a physical sense. :

[0090] ;

[0091] in, The convective heat transfer thermal resistance at the current wind speed. This is the equivalent heat capacity of the heat source (i.e., the product of the mass of the heating element and its specific heat capacity). Since the fan is forced to operate at a constant high speed in step 5401, the convective heat transfer conditions on the surface of the heat source are constant, i.e. They have been forced to be fixed at the same level, therefore the calculated The value directly reflects the heat source Size.

[0092] Step S404: Physical classification of heat source properties. The heat source property identification module 50 will calculate the real-time constant. With preset feature threshold A comparison is made. The feature threshold in this embodiment... It is the upper limit of the theoretical time constant calculated based on the standard package thermal resistance and specific heat capacity of the device's main processor chip, and is usually set to 1.5 to 2 times the chip's thermal response time constant.

[0093] If calculated This indicates that the heat source has a small heat capacity, and the temperature drops rapidly under strong winds. The heat source identification module 50 determines that the heat source is a chip-level functional heat source. This situation corresponds to heat accumulation caused by a sudden increase in the chip's computing load. The system marks it as non-fault overheating and maintains high-speed cooling until the temperature returns to normal.

[0094] If calculated This indicates that the heat source has a large heat capacity, and the temperature drops slowly, exhibiting a significant thermal hysteresis effect. The heat source nature identification module 50 determines that the heat source is a structural-level contact heat source. This situation corresponds to Joule heating caused by excessive contact resistance at the backplane wiring terminals or copper busbar connections, which is a substantial physical connection fault.

[0095] Step S405: Generate a confirmatory diagnostic report. When the determination result is structural-level contact heating, the heat source nature identification module 50 generates a high-level fault alarm signal, which indicates that the fault nature is a local physical connection fault. The heat dissipation state switching and environmental perception module 40 and the heat source nature identification module 50 utilize the principle of thermal inertia in physics to effectively distinguish between chip overheating and contact faults by actively changing heat dissipation conditions and observing response feedback, thus solving the technical problem that the physical properties of the heat source cannot be identified solely by temperature values.

[0096] The PID control logic, lumped parameter thermal model, and exponential function calculation involved in the above steps are mature technologies in the fields of automatic control and heat transfer. This embodiment focuses on using these technologies to construct identification logic for specific fault scenarios, and the basic theories will not be described in detail here.

[0097] See attached document Figure 5 The heat dissipation state switching and environmental sensing module 40 utilizes the load torque characteristics of the cooling fan motor in constant speed control mode to construct a cabinet physical status monitoring logic based on aerodynamic impedance changes. Without relying on additional physical limit switches or photoelectric sensors, the cabinet physical status monitoring logic achieves real-time sensing of the cabinet door opening / closing status and air intake passage unobstructedness by analyzing the stator current response of the fan motor under different airflow load conditions. The heat dissipation state switching and environmental sensing module 40 achieves the above functions by executing the following steps:

[0098] Step S411: Real-time aerodynamic load monitoring. When the intelligent monitoring and control device is in normal operation, the heat dissipation status switching and environmental sensing module 40 continuously reads the real-time speed of the cooling fan. With drive current To eliminate the impact of DC bus voltage fluctuations and motor coil temperature drift on current measurement, the heat dissipation state switching and environmental sensing module 40 calls the standard impedance characteristic curve stored in the parameter synchronization acquisition module 10. The heat dissipation state switching and environmental sensing module 40 uses the current real-time speed... Use the index key to retrieve the corresponding reference theoretical current value from the standard impedance characteristic curve. This benchmark theoretical current value characterizes the rated torque current of the motor required to maintain the speed under standard factory operating conditions where the cabinet door is closed and the air intake passage is clean and unobstructed.

[0099] Step S412: Calculate the impedance deviation index. The heat dissipation state switching and environmental sensing module 40 quantifies the current aerodynamic impedance change based on the difference between real-time acquired data and baseline data. The heat dissipation state switching and environmental sensing module 40 uses a normalized current deviation algorithm to calculate the impedance deviation, in order to eliminate interference from nonlinear current changes caused by speed regulation. The calculation model is as follows:

[0100] ;

[0101] in, The deviation from aerodynamic impedance is expressed as a percentage.

[0102] This is the effective value of the fan motor drive current measured at the current moment;

[0103] This is the actual fan speed measured at the current moment;

[0104] The value was obtained by querying the standard impedance characteristic curve at a rotational speed of [missing information]. The standard reference current value at that time. This formula, in its physical essence, reflects the surplus or deficit of the electromagnetic torque currently output by the fan motor relative to the standard pneumatic load torque.

[0105] Step S413: Logical classification and identification of physical environment state. The heat dissipation state switching and environmental perception module 40 will calculate the impedance deviation. Each is compared with the preset positive blockage threshold. and negative no-load threshold A comparison was made. The above thresholds are empirical values ​​calibrated based on fluid dynamics experiments. It is usually set at +20% to +30%. The setting is typically -15% to -25%. The heat dissipation status switching and environmental sensing module 40 classifies and determines the physical environment status of the cabinet based on fluid dynamics principles as follows:

[0106] Category 1 status: Detection of failure of physical protection of the cabinet (i.e., illegal opening of the cabinet door).

[0107] when This indicates that the actual load current of the fan motor is significantly lower than the reference value. From an aerodynamic perspective, assuming the fan maintains a constant speed, this corresponds to the system's intake drag coefficient. A step drop occurs. This physical phenomenon is caused by the cabinet door being opened, allowing outside air to bypass the high-resistance dust filter and air intake grille, directly entering the cabinet through the low-resistance door opening, resulting in a reduction in the air reaction torque on the fan blades. At this time, the heat dissipation state switching and environmental sensing module 40 determines that the cabinet door is in the "cabinet door open" state, and this logic enables the monitoring of the physical integrity of the cabinet.

[0108] The second category is monitoring the cleanliness of the heat dissipation channel environment (i.e., the dustproof screen is dirty and clogged).

[0109] when This indicates that the actual load current of the fan motor is significantly higher than the reference value. From an aerodynamic perspective, this corresponds to the system's intake drag coefficient. A significant increase occurs. This physical phenomenon is caused by the dust filter at the air inlet accumulating a large amount of dust or lint, resulting in a reduction in the effective cross-sectional area for airflow. This forces the fan motor to output greater electromagnetic torque to overcome the increased static pressure resistance and maintain the target speed. At this time, the heat dissipation state switching and environmental sensing module 40 determines that the device is in an "air intake channel blockage" state, indicating that the environment where the cabinet is located is severely dusty or the maintenance cycle has expired.

[0110] Step S414: Generate environmental perception and security alarm signals. Based on the identification results of step S413, the heat dissipation state switching and environmental perception module 40 generates corresponding non-electrical protection signals. For events determined to be "rack door open," the heat dissipation state switching and environmental perception module 40 generates a security intrusion alarm and triggers the event logging function of the background system to replace the traditional access control sensor signal. For events determined to be "air intake channel blocked," the heat dissipation state switching and environmental perception module 40 generates a maintenance alarm signal, prompting staff to clean the dust filter and outputting the current blockage percentage on the human-machine interface. Through the above processing, the heat dissipation state switching and environmental perception module 40 achieves the dual functions of heat dissipation control and environmental perception using a single heat dissipation actuator without adding dedicated sensor hardware.

[0111] See attached document Figure 6 The integrated positioning and defense module 60, as the system's integrated decision-making and processing unit, is configured to receive and fuse output status signals from the electrical characteristic analysis module 20, the temperature rise correlation analysis module 30, the heat dissipation state switching and environmental perception module 40, and the heat source nature identification module 50. Based on the logical combination of multi-dimensional status characteristics, the integrated positioning and defense module 60 constructs spatiotemporal correlation judgment logic to distinguish the spatial location and physical nature of the fault, and executes a graded defense strategy accordingly. The integrated positioning and defense module 60 achieves fault location and differentiated handling by executing the following steps:

[0112] Step S601: Construct a multi-dimensional fault feature vector. In each processing cycle, the integrated location and defense module 60 synchronously collects the status flags of the full-loop waveform distortion index, the current correlation flag of the temperature rise signal, the classification results of the thermal inertia time constant, and the deviation status of the aerodynamic impedance. These Boolean or enumerated status signals are combined to form a real-time multi-dimensional fault feature vector. In the specific software implementation, the multi-dimensional fault feature vector is solidified into a multi-bit status register or a data structure containing multiple member variables. Each bit or member corresponds to the judgment result of a preceding analysis module, thus providing standardized input data for subsequent logic gate circuits or lookup table-based decision-making.

[0113] Step S602: Execute the spatial fault location logic. The integrated location and defense module 60 identifies the physical location of the fault based on the inconsistency between electrical signal characteristics and thermal signal characteristics in spatial propagation. The specific judgment logic is as follows: When the electrical feature analysis module 20 detects that the full-loop waveform distortion index exceeds the standard, while at the same time, the temperature rise correlation analysis module 30 does not detect an abnormal increase in the device's onboard temperature related to current, the integrated location and defense module 60 determines that the current fault is a remote loop contact hazard. The physical basis of this logic is that the conductive electrical distortion signal of the current loop can be transmitted to the measuring device without attenuation along the wire, but the Joule heat generated at the remote fault point cannot affect the temperature sensor inside the device through thermal conduction or thermal radiation. Conversely, when the electrical feature analysis module 20 detects that the full-loop waveform distortion index exceeds the standard, and the temperature rise correlation analysis module 30 simultaneously detects an abnormal increase in onboard temperature with strong current correlation, and the heat source nature identification module 50 confirms it as structural-level heating with large heat capacity, the integrated location and defense module 60 determines that the current fault is a local backplane contact fault. At this point, the fault is located inside the device or on the backplane terminals directly connected to it.

[0114] Step S603: Execute a physical fault defense strategy. Based on the location results from step S602 and the classification results from the heat source identification module 50, the integrated location and defense module 60 matches and executes differentiated active defense actions. For cases determined to be local backplane contact faults, due to the risk of directly burning out the device or causing a fire, the integrated location and defense module 60 generates a highest-priority tripping lockout signal, drives the device's output relay to disconnect the relevant protection circuit, and sends a serious device fault alarm to the monitoring master station via the substation communication protocol, requesting immediate intervention from maintenance personnel. For cases determined to be remote circuit contact hazards, since the fault point is not on the device itself and does not temporarily endanger the device's safety, the integrated location and defense module 60 generates a warning-level signal, only sending a secondary circuit connection anomaly alarm to the monitoring master station and suggesting that maintenance personnel inspect external wiring terminals, without performing a tripping operation, thereby avoiding unnecessary power outages. In cases where the problem is determined to be functional overheating at the chip level, the integrated positioning and defense module 60 generates a heat dissipation enhancement command, locks the fan to maintain high-speed operation, and temporarily suspends the temperature-based overheat protection trip logic to prevent false tripping of the protection due to normal high-load operation of the chip.

[0115] Step S604: Execute interlocking defense based on environmental awareness. The integrated positioning and defense module 60 executes interlocking protection related to physical environment safety based on the heat dissipation status switching and the aerodynamic impedance status output by the environmental awareness module 40. When an open cabinet door is detected, the integrated positioning and defense module 60, in addition to issuing a security alarm, automatically locks the local control function or remote parameter setting modification permissions of the device by modifying the internal access control table, preventing unauthorized personnel from misoperating or maliciously modifying the device. When an air intake blockage is detected, the integrated positioning and defense module 60, while issuing an operation and maintenance alarm, calls the operating system's task scheduling interface to appropriately reduce the execution priority of non-critical background tasks, thereby reducing the overall power consumption of the chip and preventing the chip temperature from exceeding limits under deteriorating heat dissipation conditions.

[0116] The integrated positioning and defense module 60, through logical verification of the aforementioned multi-dimensional information, upgrades single threshold judgment to comprehensive scenario analysis, effectively solving the problem of protection maloperation or failure to operate caused by the inability to distinguish fault location and physical properties in traditional monitoring schemes. Regarding the implementation of logic gate circuits, state machine programming, and communication message framing involved in the above steps, those skilled in the art can implement them using field-programmable gate array logic or embedded software code according to the actual hardware platform. These are conventional technical means in the field of industrial control and will not be elaborated upon here.

[0117] See attached document Figure 7 , Figure 7 This is a flowchart of an abnormal state early warning method for secondary circuit wiring terminals according to an embodiment of the present invention. The present invention provides an abnormal state early warning method for secondary circuit wiring terminals, which is mainly executed by a processor inside an intelligent measurement and control device, and includes the following steps:

[0118] Step S100: Using high-precision synchronous sampling technology, electrical quantity data and onboard temperature data with strictly aligned timestamps of the secondary circuit are acquired in real time to construct a multi-dimensional synchronous dataset.

[0119] Step S200: Perform frequency domain feature extraction on the acquired current waveform, separate the power frequency fundamental wave and high-frequency noise components, calculate the full circuit waveform distortion index, and compare the full circuit waveform distortion index with the dynamic background noise baseline to identify whether there are nonlinear electrical fault characteristics caused by poor contact in the circuit.

[0120] Step S300: Construct a thermo-electric dynamic coupling model based on Joule's law, and use the temperature sequence and current sequence within the sliding time window to calculate the temperature residual value in order to determine whether the current onboard temperature rise has a current correlation, thereby initially screening out the heat source.

[0121] Step S400: Monitor the change of drive current of the heat dissipation execution unit at a constant speed in real time, calculate the aerodynamic impedance deviation, and compare the aerodynamic impedance deviation with the preset fluid impedance threshold to identify the opening and closing status of the cabinet door and the degree of blockage of the air intake channel.

[0122] Step S500: When a temperature rise anomaly with current correlation is detected, the heat dissipation execution unit is controlled to switch to open-loop step control mode to apply active aerodynamic disturbance, the transient thermal response curve of the heat source under strong convection heat dissipation field is collected, and the thermal inertia time constant is calculated to distinguish whether the heat source is chip-level functional heat dissipation or structural-level contact heat dissipation.

[0123] In step S600, a multi-dimensional fault feature vector is constructed by integrating the electrical waveform distortion index, temperature rise current correlation, aerodynamic impedance state, and thermal inertia time constant. Based on the spatiotemporal correlation logic, the physical location and nature of the fault are determined, and a graded defense strategy including alarm, blocking, or tripping is executed accordingly.

Claims

1. An abnormal state early warning system for secondary circuit terminals, characterized in that, include: The parameter synchronous acquisition module is configured to connect with the underlying hardware of the intelligent measurement and control device, read the real-time sampling sequence of the secondary circuit load current and the onboard temperature data, and synchronously acquire the speed and drive current of the cooling fan. The acquired data is then output after being aligned in time sequence. The electrical characteristic analysis module is connected to the parameter synchronous acquisition module and is configured to receive current waveform data. It calculates the full-circuit waveform distortion index by extracting frequency domain features and identifies whether there are nonlinear electrical fault characteristics in the circuit. The temperature rise correlation analysis module is connected to the parameter synchronous acquisition module. It is configured to use the aligned temperature sequence and current sequence to build a thermoelectric dynamic coupling model and determine whether the on-board temperature rise has current correlation. The heat dissipation state switching and environmental perception module is connected to the parameter synchronous acquisition module and the temperature rise correlation analysis module. It is configured to control the cooling fan to switch to the open-loop step control mode to apply active aerodynamic disturbance when the temperature rise has a current correlation. It is also used to calculate the aerodynamic impedance deviation based on the fan drive current. The heat source nature identification module is connected to the heat dissipation state switching and environmental sensing module. It is configured to collect the transient thermal response curve of the on-board temperature and calculate the thermal inertia time constant during the application of active aerodynamic disturbance. Based on the magnitude of the thermal inertia time constant, it distinguishes whether the heat source is chip-level functional heat generation or structural-level contact heat generation. The integrated positioning and defense module is connected to the electrical characteristic analysis module, temperature rise correlation analysis module, heat dissipation state switching and environmental perception module, and heat source nature identification module, respectively. It is configured to comprehensively analyze the whole circuit waveform distortion index, temperature rise current correlation, aerodynamic impedance state, and heat source nature classification results to determine the physical location and nature of the fault and execute a graded defense strategy.

2. The abnormal state early warning system for secondary circuit terminals according to claim 1, characterized in that, The parameter synchronous acquisition module establishes the baseline correspondence between fan speed and load torque in the following way: based on the principle of motor torque balance, a fan load dynamic model is constructed using the motor torque constant, aerodynamic drag coefficient, fan rotational angular velocity, and motor bearing viscous friction coefficient; under the baseline state of cabinet door closed and dust filter clean, the fan is controlled to run at steady-state speed conditions of multiple gradients, the drive current value under each condition is collected, and numerical fitting is performed based on the fan load dynamic model to generate a standard impedance characteristic curve.

3. The abnormal state early warning system for secondary circuit terminals according to claim 1, characterized in that, The specific method by which the electrical feature analysis module performs frequency domain feature extraction is as follows: A fast Fourier transform is performed on the acquired raw current waveform sequence to decompose the current signal in the frequency domain into the power frequency fundamental component, integer harmonic components, and high-frequency residual noise components in the non-harmonic frequency band. The calculation logic of the full-loop waveform distortion index is as follows: the ratio of the sum of squares of the effective values ​​of each integer harmonic current to the effective value of the power frequency fundamental current is calculated to characterize the degree of conventional harmonic distortion, and the spectral energy ratio of the high-frequency residual noise component is introduced as a nonlinear correction factor. The full-loop waveform distortion index is generated by weighting and correcting the degree of conventional harmonic distortion.

4. The abnormal state early warning system for secondary circuit terminals according to claim 1, characterized in that, The temperature rise correlation analysis module constructs the thermoelectric dynamic coupling model as follows: based on Joule's law, a physical rule is set that the temperature change is proportional to the square of the effective value of the current, and a thermal hysteresis time parameter representing the heat conduction speed is introduced; the theoretical estimated temperature at the current moment is calculated based on the current data at historical moments, and the theoretical estimated temperature is compared with the actual collected onboard temperature to calculate the numerical deviation between the two; when the numerical deviation is within the allowable error range, it is determined that the current onboard temperature rise has current correlation.

5. An abnormal state early warning system for secondary circuit terminals according to claim 1, characterized in that, The specific method by which the heat dissipation state switching and environmental perception module controls the cooling fan to switch to the open-loop step control mode is as follows: temporarily stop the intelligent speed regulation logic of the fan automatically adjusting its speed according to the temperature, directly send a set high speed command to the fan, force the fan to leave the current state and run continuously with a constant high air volume, create a stable strong air cooling environment inside the cabinet, and enable the heat-generating parts to enter a rapid cooling state.

6. The abnormal state early warning system for secondary circuit terminals according to claim 1, characterized in that, The specific method by which the heat source property identification module distinguishes heat sources is as follows: Analyzing the collected temperature decrease curve over time, calculating the thermal inertia time constant representing the rate of cooling of the heat source; and comparing the thermal inertia time constant with a preset benchmark value. When the thermal inertia time constant is less than the reference value, it means that the cooling rate is fast, and it is determined to be the self-heating generated by the chip during operation; When the thermal inertia time constant is greater than or equal to the reference value, it means that the cooling rate is slow, which is determined to be heat generation caused by poor contact at the structural connection.

7. An abnormal state early warning system for secondary circuit terminals according to claim 1, characterized in that, The way the heat dissipation state switching and environmental perception module identifies the cabinet door status and the degree of blockage of the air intake channel is: based on the current actual speed of the fan, find the theoretically required reference current value at that speed on the pre-generated standard impedance characteristic curve. The actual driving current of the fan during operation is compared with the reference current value, and the degree of difference between the two is calculated as the aerodynamic impedance deviation. When the difference is below the set negative limit, it indicates that the fan operating resistance has decreased abnormally, and it is determined that the cabinet door is in the open state. When the difference exceeds the set positive limit, it indicates that the fan's operating resistance has increased abnormally, and the intake passage is determined to be blocked.

8. An abnormal state early warning system for secondary circuit terminals according to claim 1, characterized in that, The logic for determining the physical location of the fault by the integrated positioning and defense module is as follows: when the full-circuit waveform distortion index exceeds the preset normal range, but the onboard temperature rise does not have a current correlation, the fault point is determined to be outside the equipment, belonging to a remote circuit contact hazard; when the full-circuit waveform distortion index exceeds the preset normal range, and the onboard temperature rise has a current correlation, and the heat source nature is confirmed to be structural-level contact heating, the fault point is determined to be located at the equipment backplane, belonging to a local backplane contact fault.

9. An abnormal state early warning system for secondary circuit terminals according to claim 1, characterized in that, The integrated positioning and defense module executes the differentiated defense strategy as follows: if the fault point is determined to be outside the equipment, only an alarm signal is issued to prompt the staff to check the external lines, and the current normal operation of the equipment is maintained; if the fault point is determined to be located on the back panel of the equipment, two-level protection measures are immediately activated: the first level measure is to directly adjust the cooling fan to the maximum speed to forcibly reduce the temperature; if the onboard temperature continues to rise and exceeds the set safety limit value when the cooling fan is running at the maximum speed, the second level measure is executed, and a trip command is sent to cut off the current circuit to prevent the wiring terminals from burning out.

10. A method for early warning of abnormal conditions of secondary circuit terminals, and an early warning system for abnormal conditions of secondary circuit terminals according to any one of claims 1-9, characterized in that, Includes the following steps: Step S100: Using high-precision synchronous sampling technology, electrical quantity data and onboard temperature data with strictly aligned timestamps of the secondary circuit are acquired in real time to construct a multi-dimensional synchronous dataset. Step S200: Perform frequency domain feature extraction on the acquired current waveform, separate the power frequency fundamental wave from the high-frequency noise component, and calculate the waveform distortion index of the entire circuit to identify whether there are nonlinear electrical fault characteristics caused by poor contact in the circuit. Step S300: Construct a thermoelectric dynamic coupling model based on Joule's law, and calculate the temperature residual value using the temperature sequence and current sequence within the sliding time window to determine whether the current onboard temperature rise has current correlation. Step S400: Monitor the change in drive current of the heat dissipation actuator at a constant speed in real time, calculate the aerodynamic impedance deviation, and identify the opening and closing status of the cabinet door and the degree of blockage of the air intake channel. Step S500: When a temperature rise anomaly with current correlation is detected, the heat dissipation execution unit is controlled to switch to open-loop step control mode to apply active aerodynamic disturbance, the transient thermal response curve of the heat source under strong convection heat dissipation field is collected, and the thermal inertia time constant is calculated to distinguish whether the heat source is chip-level functional heat dissipation or structural-level contact heat dissipation. In step S600, a multi-dimensional fault feature vector is constructed by integrating the electrical waveform distortion index, temperature rise current correlation, aerodynamic impedance state, and thermal inertia time constant. Based on the spatiotemporal correlation logic, the physical location and nature of the fault are determined, and a graded defense strategy including alarm, blocking, or tripping is executed accordingly.

Images