An integrated circuit fault diagnosis and early warning system

Abstract

The present invention discloses an integrated circuit fault diagnosis and early warning system, which relates to the technical field of integrated circuit fault diagnosis and early warning. It includes: a signal acquisition and processing module: used to collect and process the voltage, current signals and surrounding magnetic field signals of the integrated circuit; an electromagnetic coupling characteristic calculation module: used to calculate electromagnetic coupling characteristic parameters; a dynamic distortion index construction module: used to construct a dynamic distortion index; a sensitivity evaluation module: used to evaluate the sensitivity of the dynamic distortion index; a fault mode recognition module: used to perform fault mode recognition on the evaluation results to generate fault categories; a fault probability calculation module: used to calculate the fault probability according to the electromagnetic coupling characteristic parameters and the dynamic distortion index; a hierarchical fault response module: used to jointly execute a hierarchical fault response strategy according to the fault categories and the fault probabilities. Through accurate fault identification, multi-level response strategies and reliability improvement, the stable operation and long-term reliability of the integrated circuit are effectively guaranteed.

Description

Technical Field

[0001] The present invention relates to the technical field of integrated circuit fault diagnosis and early warning, and specifically provides an integrated circuit fault diagnosis and early warning system. Background Art

[0002] Integrated circuits (ICs) are indispensable core components in modern electronic devices, and the stability and reliability of their performance directly affect the operation quality of the system. During the operation of integrated circuits, due to complex current and voltage fluctuations, as well as external electromagnetic interference and other factors, faults may occur, leading to equipment failure or performance degradation. Therefore, real-time monitoring and diagnosis of integrated circuit faults, timely identification of potential faults and taking early warning measures have become important means to ensure the reliability of integrated circuits.

[0003] In the prior art, although certain degrees of fault detection can be carried out on integrated circuits by monitoring parameters such as current, voltage signals and temperature, there is a lack of a comprehensive diagnosis method that comprehensively considers various characteristics such as electromagnetic coupling and dynamic distortion. At the same time, traditional fault diagnosis technologies mostly rely on static analysis or simple statistical methods, lacking accurate identification and prediction of complex dynamic fault modes.

[0004] To solve the above problems, the present invention proposes an integrated circuit fault diagnosis and early warning system, which combines electromagnetic coupling characteristics, dynamic distortion indicators, sensitivity evaluation and chaos theory analysis, and provides a more accurate and efficient fault diagnosis method by collecting signals in real time and performing complex time-domain alignment and noise reduction processing. This system can identify various fault modes, calculate the fault probability, and adopt a hierarchical response strategy according to the fault category and probability, thereby improving the reliability and operation stability of integrated circuits. Summary of the Invention

[0005] Based on the above-mentioned disadvantages of the prior art, the purpose of the present invention is to provide an integrated circuit fault diagnosis and early warning system to solve the above technical problems.

[0006] To achieve the above purpose, the present invention provides the following technical solution: An integrated circuit fault diagnosis and early warning system, comprising:

[0007] A signal acquisition and processing module: used for real-time acquisition of voltage, current signals and surrounding magnetic field signals of the integrated circuit, and performing time-domain alignment and noise reduction processing;

[0008] An electromagnetic coupling characteristic calculation module: used for calculating electromagnetic coupling characteristic parameters based on the processed voltage signals and current signals, in combination with the magnetic field signals;

[0009] A dynamic distortion index construction module: used for constructing a dynamic distortion index according to the electromagnetic coupling characteristic parameters;

[0010] Sensitivity evaluation module: used to evaluate the sensitivity of the dynamic distortion index by using a quantization algorithm;

[0011] Fault mode recognition module: used to identify the fault mode of the evaluation result by using the chaos theory analysis method to generate fault categories;

[0012] Fault probability calculation module: used to calculate the fault probability according to the electromagnetic coupling characteristic parameters and the dynamic distortion index;

[0013] Hierarchical fault response module: used to jointly execute the hierarchical fault response strategy according to the fault category and the fault probability.

[0014] The present invention is further configured that the calculation logic of the electromagnetic coupling characteristic parameters is: , is the electromagnetic coupling coefficient, is the value of the current signal at time t, is the value of the magnetic field signal at time t, is the modulus of the magnetic field intensity collected at time t, is a constant.

[0015] The present invention is further configured that the calculation logic of the dynamic distortion index is: , is the dynamic impedance distortion rate, is the instantaneous impedance, is the average impedance, The calculation logic of is: , is the length of the sliding time window.

[0016] The present invention is further configured that the sensitivity evaluation is obtained by constructing a non-commutative operator, calculating the eigenvalue shift, and then calculating the sensitivity score according to the operator and the eigenvalue shift;

[0017] The construction logic of the non-commutative operator is: , is the Hilbert-Schmidt operator, is the dynamic distortion factor, is the voltage change, is the complex conjugate term of the voltage change;

[0018] The calculation logic of the eigenvalue shift is: , is the eigenvalue shift;

[0019] The calculation logic of the sensitivity score is: , is the sensitivity score, is the fluctuation degree of the change amount, is the length of the sliding time window, is the Planck constant, is the Boltzmann constant, is the electron temperature, and the calculation logic of the electron temperature is: , is the electron charge, is the transient thermal resistance, and the calculation logic of the transient thermal resistance is: , is the current change amount, and it is defined in the formula: .

[0020] The present invention is further configured such that the calculation logic of the dynamic distortion factor is: , is the dynamic distortion factor, is the vacuum permeability, is the electromagnetic phase mismatch degree, is the decay time constant, is the value of the current signal at time t, is the value of the magnetic field signal at time t, where , is the highest frequency of the signal, and the calculation logic of the electromagnetic phase mismatch degree is: , is the amplitude of the current signal, is the binary exclusive OR operation.

[0021] The present invention is further configured such that for the fault mode recognition, the fractal permeability is extracted by using the Poincaré section mapping, and the faults are classified to generate fault categories in combination with the maximum Lyapunov exponent;

[0022] The construction logic of the Poincaré section is: In the phase space , record the intersection points of the trajectory and the hyperplane , and the hyperplane is defined as: , where , and are coefficients;

[0023] The calculation logic of the fractal permeability is: , is the fractal permeability, is the measure of distance, is the number of points within the neighborhood with a radius of , The calculation logic of is: For the number of elements that meet the internal conditions, is the data point and the distance between;

[0024] The generation logic of the fault category is: , is the fault category, is the maximum Lyapunov exponent, is the weight vector, The calculation logic of , is the Jacobian matrix, and the Jacobian matrix The generation logic of is: ;

[0025] The fault category includes: when , it represents a short - circuit fault; when , it represents an open - circuit fault; when , it represents an aging fault; when , it represents an electromagnetic interference fault; when , it represents quantum tunneling leakage; when , it represents thermal runaway.

[0026] The present invention is further set such that the calculation logic of the fault probability is: , is the fault probability, is the weight coefficient;

[0027] Adjust the fault probability , and the adjustment logic of the fault probability is: , is the updated fault probability, is the fault probability calculated last time, is the fault probability calculated currently, is the smoothing factor.

[0028] The present invention is further set such that a dynamic threshold is calculated based on historical fault probability data, and the calculation logic of the dynamic threshold is: , is the dynamic threshold, is the mean difference of the fault probability, is the standard deviation of the fault probability, and are constant coefficients;

[0029] Set the maximum and minimum threshold ranges according to the historical dynamic threshold, and the range of the dynamic threshold is: , is the minimum threshold, is the maximum threshold.

[0030] The present invention is further configured such that, according to the fault category and the updated fault probability, a hierarchical fault response strategy is jointly executed. When and , an emergency shutdown and alarm are triggered; when and , a frequency reduction operation mode is started; when , the normal operation state is maintained without any operation.

[0031] The present invention is further configured such that the time domain alignment adopts a quantum tunneling synchronization protocol, the clock jitter is less than 1 ps, and the calculation logic for synchronizing error compensation of voltage signals and current signals is: , and the calculation logic for synchronizing error compensation of magnetic field signals and current signals is: , is the value of the voltage signal at time t, is the value of the current signal after adjusting the time offset , is the value of the current signal after adjusting the time offset , and are the offsets.

[0032] The present invention provides an integrated circuit fault diagnosis and early warning system, including a signal acquisition and processing module: for real-time acquisition of voltage, current signals and surrounding magnetic field signals of the integrated circuit, and performing time domain alignment and noise reduction processing; an electromagnetic coupling feature calculation module: for calculating electromagnetic coupling feature parameters based on the processed voltage and current signals in combination with the magnetic field signal; a dynamic distortion index construction module: for constructing a dynamic distortion index according to the electromagnetic coupling feature parameters; a sensitivity evaluation module: for evaluating the sensitivity of the dynamic distortion index using a quantization algorithm; a fault mode recognition module: for using the chaotic theory analysis method to perform fault mode recognition on the evaluation results to generate a fault category; a fault probability calculation module: for calculating the fault probability according to the electromagnetic coupling feature parameters and the dynamic distortion index; a hierarchical fault response module: for jointly executing a hierarchical fault response strategy according to the fault category and the fault probability, and the beneficial effects include:

[0033] 1. Precise fault identification: By real-time acquisition and time domain alignment of voltage, current and magnetic field signals of the integrated circuit, combined with electromagnetic coupling features and dynamic distortion indicators, potential fault signs in the integrated circuit can be accurately captured, realizing multi-dimensional fault diagnosis;

[0034] 2. Multi - level Fault Response Strategy: According to the fault category and probability, the system can automatically execute a hierarchical response strategy. For severe faults, it can trigger an emergency shutdown and alarm; for minor faults, it starts a frequency - reduction operation mode to ensure the continuous and stable operation of the system.

[0035] 3. Improve System Reliability: Through comprehensive electromagnetic coupling characteristic analysis and dynamic distortion monitoring, the system can timely detect problems such as electromagnetic interference and signal distortion, avoid the long - term impact of these problems on the functions of integrated circuits, and enhance the overall reliability and stability of the system.

[0036] The above description is only an overview of the technical solution of this application. In order to understand the technical means of this application more clearly, it can be implemented according to the content of the specification. And in order to make the above and other purposes, features and advantages of this application more obvious and understandable, the following specifically illustrates the specific embodiments of this application. Brief Description of the Drawings

[0037] In order to more clearly illustrate the technical solutions in the embodiments of the present invention, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings. In the drawings:

[0038] Figure 1 It is a structural diagram of an integrated circuit fault diagnosis and early warning system shown in an exemplary embodiment of the present invention. Detailed Embodiments

[0039] The following will illustrate the embodiments of the present invention with reference to the drawings and preferred embodiments. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. It should be understood that the preferred embodiments are only for explaining the present invention, rather than for limiting the protection scope of the present invention.

[0040] It should be noted that the diagrams provided in the following embodiments only illustrate the basic concept of the present invention in a schematic manner. Therefore, only the components related to the present invention are shown in the diagrams, rather than being drawn according to the number, shape and size of the components in actual implementation. The type, quantity and proportion of each component in its actual implementation can be arbitrarily changed, and the component layout type may also be more complex.

[0041] In the following description, numerous specific details are explored to provide a more thorough explanation of the embodiments of the present invention. However, it will be apparent to those skilled in the art that the embodiments of the present invention can be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring the embodiments of the present invention.

[0042] An integrated circuit fault diagnosis and early warning system, as Figure 1 shown, includes:

[0043] A signal acquisition and processing module: used to collect the voltage, current signals and the surrounding magnetic field signals of the integrated circuit in real time, and perform time-domain alignment and noise reduction processing;

[0044] An electromagnetic coupling characteristic calculation module: used to calculate the electromagnetic coupling characteristic parameters based on the processed voltage signals and current signals, in combination with the magnetic field signals;

[0045] A dynamic distortion index construction module: used to construct a dynamic distortion index according to the electromagnetic coupling characteristic parameters;

[0046] A sensitivity evaluation module: used to evaluate the sensitivity of the dynamic distortion index by using a quantization algorithm;

[0047] A fault mode recognition module: used to use the chaos theory analysis method to perform fault mode recognition on the evaluation results to generate fault categories;

[0048] A fault probability calculation module: used to calculate the fault probability according to the electromagnetic coupling characteristic parameters and the dynamic distortion index;

[0049] A hierarchical fault response module: used to jointly execute a hierarchical fault response strategy according to the fault categories and the fault probabilities;

[0050] Specifically, the signal acquisition and processing module is mainly responsible for real-time acquisition of voltage, current signals, and ambient magnetic field signals in the integrated circuit. The acquired signals will then undergo time-domain alignment and noise reduction processing. Time-domain alignment is a process to ensure signal synchronization, while noise reduction processing helps remove interference caused by environmental or device noise, ensuring the accuracy and reliability of the acquired data. The electromagnetic coupling feature calculation module calculates electromagnetic coupling feature parameters based on the processed voltage and current signals, combined with the synchronized magnetic field signals. The electromagnetic coupling feature parameters are key indicators for evaluating factors such as the degree of electromagnetic coupling and signal interference in the integrated circuit, and can reveal issues in aspects such as electromagnetic compatibility. The dynamic distortion index construction module further constructs a dynamic distortion index based on the calculated electromagnetic coupling feature parameters. The dynamic distortion index reflects the changes in electrical characteristics (such as impedance, signal quality, etc.) during the operation of the circuit and can reveal whether there are performance degradations or potential faults in the circuit. The sensitivity evaluation module uses a quantization algorithm to evaluate the sensitivity of the dynamic distortion index, that is, the response ability to circuit performance changes. Sensitivity evaluation can help identify which parameter changes may have a greater impact on the circuit under specific conditions, contributing to improving the accuracy of fault detection. The fault mode recognition module uses chaos theory analysis methods to identify fault modes from the evaluation results. By analyzing the results of dynamic distortion and sensitivity evaluation, different types of fault modes (such as short circuits, open circuits, aging, etc.) are identified, and fault categories are generated. The fault probability calculation module calculates the fault probability of the integrated circuit using electromagnetic coupling features and dynamic distortion indexes. The fault probability reflects the likelihood of the circuit malfunctioning and is the basis for subsequent decision-making and response strategies. The hierarchical fault response module jointly executes a hierarchical fault response strategy based on the fault category and the calculated fault probability. According to the severity of different faults, corresponding response operations can be triggered, including emergency shutdown and alarm, frequency reduction operation mode, and normal operation. Through collaborative work, the above modules form a comprehensive integrated circuit fault diagnosis and early warning system, which can detect potential faults in a timely manner under different working environments and take appropriate measures to ensure the safe and stable operation of the system.

[0051] The present invention is further configured such that the calculation logic of the electromagnetic coupling feature parameters is as follows: , is the electromagnetic coupling coefficient, is the value of the current signal at time t, is the value of the magnetic field signal at time t, is the magnitude of the magnetic field strength collected at time t, is a constant; specifically, the electromagnetic coupling feature parameter is used to describe the degree of interaction between current and magnetic field and is a key parameter for electromagnetic compatibility analysis; the current signal changes with time and represents the working state of the circuit; the magnetic field signal reflects the electromagnetic waves and interference generated by the current flow. Used to adjust the weight between current and magnetic field signals, the value range is By introducing the ratio of current and magnetic field signals and their time variation, the coupling strength between current and magnetic field can be more accurately characterized, helping to reveal problems such as electromagnetic interference and power loss.

[0052] The present invention is further configured such that the calculation logic of the dynamic distortion index is: , is the dynamic impedance distortion rate, is the instantaneous impedance, is the average impedance, The calculation logic is: , is the length of the sliding time window; specifically, the dynamic impedance distortion rate It is used to measure the degree of change in the circuit impedance during the operation of the integrated circuit, revealing possible abnormal fluctuations or fault signals in the circuit. The larger the value, the more drastic the circuit impedance change, which may mean that the circuit is undergoing abnormal changes; average impedance Reflects the impedance level of the circuit over a period of time; instantaneous impedance Reflects dynamic changes in circuit performance; by calculating the dynamic impedance distortion rate, it can accurately capture abnormal changes in the circuit, such as electromagnetic interference, instantaneous impedance changes, etc., which helps to identify potential faults in advance.

[0053] The present invention is further configured such that the sensitivity assessment is performed by constructing a non-commutative operator, calculating an eigenvalue shift, and then calculating a sensitivity score based on the operator and the eigenvalue shift; the construction logic of the non-commutative operator is: , is the Hilbert-Schmidt operator, is the dynamic distortion factor, For voltage changes, is the complex conjugate term of the voltage change; the calculation logic of the eigenvalue shift is: , is the eigenvalue offset; the calculation logic of the sensitivity score is: , Score sensitivity, is the degree of fluctuation of the change, is the length of the sliding time window, is Planck's constant, is the Boltzmann constant, is the electron temperature, the electron temperature The calculation logic is: , is the electron charge, is the transient thermal resistance, and the calculation logic of the transient thermal resistance is as follows: , is the change in current, and it is defined in the formula that: ; Specifically, the Hilbert - Schmidt operator is used to describe a physical system with non - commutation properties, representing the relationship between voltage change and dynamic distortion; the eigenvalue shift reflects the degree of change in voltage and current over a period of time; the sensitivity score is used to quantify the sensitivity of the system; the electron temperature is used to represent the average energy state of electrons in the circuit; the transient thermal resistance is used to represent the impedance in the thermal equilibrium state; by calculating the sensitivity score, the fault sensitivity of the circuit can be monitored and evaluated in real - time, potential fault points can be identified in advance, and through the real - time quantification of the system fault sensitivity, the fault response strategy can be optimized to improve the reliability and stability of the system in a complex environment.

[0054] The present invention is further set such that the calculation logic of the dynamic distortion factor is: , is the dynamic distortion factor, is the vacuum permeability, is the electromagnetic phase mismatch degree, is the decay time constant, is the value of the current signal at time t, is the value of the magnetic field signal at time t, where, , is the highest frequency of the signal, and the calculation logic of the electromagnetic phase mismatch degree is: , is the amplitude of the current signal, is the binary exclusive - OR operation; specifically, the dynamic distortion factor is used to describe the mutual influence and its change between current and magnetic field in the integrated circuit, evaluate the dynamic relationship and mismatch degree between the current signal and the magnetic field signal, and thus reflect the dynamic stability of the integrated circuit; the electromagnetic phase mismatch degree is used to measure the relative direction difference between the current signal and the magnetic field signal; is used to describe the rate at which the electromagnetic coupling effect decays with time, and its value depends on the circuit characteristics of the specific application and the frequency of the signal; the calculation of the dynamic distortion factor can monitor the dynamic relationship between current and magnetic field in real - time, thereby evaluating the electromagnetic interference and signal distortion in the circuit and timely detecting potential fault signs.

[0055] The present invention is further configured such that the fault mode recognition extracts the fractal permeability by using the Poincaré section mapping, and classifies the faults in combination with the maximum Lyapunov exponent to generate fault categories; the construction logic of the Poincaré section is as follows: in the phase space record the intersection points of the trajectory and the hyperplane ; the hyperplane is defined as: , where , , and are coefficients; the calculation logic of the fractal permeability is: , is the fractal permeability, is a measure of distance, is the number of points within the neighborhood with a radius of ; the calculation logic of is: , is the number of elements that meet the conditions within , is the data point and ; the generation logic of the fault categories is: , is the fault category, is the maximum Lyapunov exponent, is the weight vector, the calculation logic of is: , is the Jacobian matrix, and the generation logic of the Jacobian matrix is: ; the fault categories include: when , it represents a short - circuit fault; when , it represents an open - circuit fault; when , it represents an aging fault; when , it represents an electromagnetic interference fault; when , it represents a quantum tunneling leakage; when , it represents a thermal runaway; specifically, the Poincaré section mapping is a tool for analyzing periodic orbits in a dynamical system. Usually, a hyperplane is defined in the phase space, and the intersection points of the system are recorded. By studying these intersection points, the behavior of complex systems can be simplified and understood; the fractal permeability An index used to quantify the complexity of an integrated circuit, reflecting the self-similarity and non-linearity of the integrated circuit; The Lyapunov exponent is an index used to describe the stability of an integrated circuit. The maximum Lyapunov exponent reflects the sensitivity of the system to changes in initial conditions and determines the stability or chaotic nature of the system; The Jacobian matrix is used to describe how the system state changes over time; 、 and are the coefficients required when constructing the hyperplane of the Poincaré section. By adjusting 、 and , the section can be positioned at different locations in the phase space, thus affecting the fault mode recognition ability. The values are determined through experimental data; Using the method of combining the fractal permeability and the Lyapunov exponent with weights helps to accurately distinguish different types of faults and ensure more accurate fault prediction and response.

[0056] The present invention is further configured such that the calculation logic of the fault probability is: , is the fault probability, is the weight coefficient; Adjust the fault probability , and the adjustment logic of the fault probability is: , is the updated fault probability, is the fault probability calculated last time, is the fault probability calculated currently, is the smoothing factor; Specifically, the fault probability represents the likelihood of an integrated circuit failing at a certain point in time or over a period of time, reflecting the health state of the integrated circuit and helping to predict potential fault occurrences. Adjusting the fault probability is to smooth the system's response to changes in the fault probability; is used to adjust the relative importance of the electromagnetic coupling characteristics and the dynamic distortion index, and its value range is [0,1]; controls the influence degree of the historical fault probability and the current fault probability on the final result, and its value range is [0,1]; By smoothing the adjustment of the fault probability, it is possible to avoid unnecessary responses caused by drastic fluctuations in the fault probability. The introduction of the smoothing factor makes the fault response more stable and not overreact to short-term abnormal fluctuations. The accurate calculation and adjustment of the fault probability can help the system identify potential fault risks in advance, and then take appropriate preventive measures to reduce the likelihood of faults occurring and improve the stability and reliability of the system.

[0057] The present invention is further configured to calculate a dynamic threshold based on the historical data of the fault probability, and the calculation logic of the dynamic threshold is: , is the dynamic threshold, is the mean difference of the failure probability, is the standard deviation of the failure probability, and is a constant coefficient; the maximum and minimum threshold ranges are set according to the historical dynamic threshold, and the range of the dynamic threshold is: , is the minimum threshold, is the maximum threshold; specifically, the dynamic threshold is used to judge the change trend of the failure probability and provide an adaptive threshold value to trigger the response of the system or take measures; in order to prevent the threshold from being too large or too small, the minimum threshold and the maximum threshold are set to limit the change range of the dynamic threshold: ; the minimum threshold represents the boundary at which some preventive measures are still allowed when the system failure probability is the smallest. Further, obtain the historical dynamic threshold when some preventive measures are still allowed when all system failure probabilities are the smallest under historical data, and set the largest historical dynamic threshold as the minimum threshold ; the maximum threshold represents the maximum warning value when the system has an extremely high failure probability. When exceeding this value, the system must respond emergently. Further, obtain the historical dynamic threshold corresponding to the maximum warning value when all systems have an extremely high failure probability under historical data, and set the smallest historical dynamic threshold as the maximum threshold ; and are used for weighted mean, standard deviation and the change rate of the failure probability to adjust the influence of different factors on the dynamic threshold, and the value range is [0,1]; the dynamic threshold combines historical data, standard deviation and change rate, can accurately reflect the failure trend, and can react more sensitively when the failure risk increases by dynamically adjusting the threshold, and take preventive measures in time to reduce the occurrence probability of potential failures.

[0058] The present invention is further set to jointly execute a hierarchical failure response strategy according to the failure category and the updated failure probability. When and , trigger emergency shutdown and alarm; when and , start the frequency reduction operation mode; when , maintain the normal operation state without operation; specifically, the failure response strategy adopts three response modes according to different combinations of the failure probability and the failure category; when and When it occurs, an emergency shutdown is triggered and an alarm is given. At this time, the failure probability is higher than the maximum safety threshold, and the failure category is one that may cause significant damage or is irreparable, such as short circuit, quantum tunneling leakage, and thermal runaway. Immediate shutdown and alarm are required to ensure the safety of personnel and equipment; when and When it occurs, the frequency reduction operation mode is started. At this time, the failure probability has not reached the level of emergency shutdown, but it has exceeded the tolerance range of normal operation, and the failure category is a failure with moderate impact, such as open circuit failure, aging failure, and electromagnetic interference failure. Therefore, the burden will be reduced by reducing the operating frequency to maintain a safe working state; when When it occurs, the normal operation state is maintained without any operation. At this time, the failure probability is low, indicating a normal working state and no additional intervention is required; according to the dynamic changes of the failure probability and failure category, the response mode can be flexibly adjusted from emergency shutdown to frequency reduction operation to ensure appropriate handling according to the actual risk. By adjusting the operating state according to the failure probability, it can operate with high reliability and stability, and avoid irreparable damage caused by the excessive spread of failures.

[0059] The present invention is further configured such that the time domain alignment adopts a quantum tunneling synchronization protocol, the clock jitter is less than 1 ps, and the calculation logic for compensating the synchronization error between the voltage signal and the current signal is: , and the calculation logic for compensating the synchronization error between the magnetic field signal and the current signal is: , is the value of the voltage signal at time t, is the value of the current signal after adjusting the time offset , is the value of the current signal after adjusting the time offset , and are the offset amounts; specifically, the clock jitter target is to minimize the fluctuation of the clock signal as much as possible to ensure the accuracy during signal synchronization; calculate the synchronization error compensation between the voltage signal and the current signal to minimize the difference between the two and achieve synchronization alignment; calculate the synchronization error compensation between the magnetic field signal and the current signal to perform time domain alignment of the magnetic field signal and the current signal; and are the quantities for adjusting the time axis, and their values are adjusted through experiments or historical data without a fixed range; by accurately calculating the time offset amounts between the voltage, current, and magnetic field signals, it is ensured that these signals are optimally aligned in the time domain, improving the accuracy of the system.

[0060] As described above, it is only the specific implementation manner of this application, but the protection scope of this application is not limited thereto. Any person skilled in the art within the technical scope disclosed by this application can easily think of changes or substitutions, which should all be covered within the protection scope of this application. Therefore, the protection scope of this application shall be subject to the protection scope of the claims described above.

Claims

1. An integrated circuit fault diagnosis and early warning system, characterized in that Including: Signal acquisition and processing module: used to collect the voltage, current signals and surrounding magnetic field signals of the integrated circuit in real time, and perform time-domain alignment and noise reduction processing; Electromagnetic coupling characteristic calculation module: used to calculate the electromagnetic coupling characteristic parameters based on the processed voltage signal and current signal, combined with the magnetic field signal; Dynamic distortion index construction module: used to construct a dynamic distortion index according to the electromagnetic coupling characteristic parameters; Sensitivity evaluation module: used to evaluate the sensitivity of the dynamic distortion index by using a quantization algorithm; Fault mode recognition module: used to use the chaotic theory analysis method to perform fault mode recognition on the evaluation results to generate fault categories; Fault probability calculation module: used to calculate the fault probability according to the electromagnetic coupling characteristic parameters and the dynamic distortion index; Hierarchical fault response module: used to jointly execute the hierarchical fault response strategy according to the fault category and the fault probability.

2. The integrated circuit fault diagnosis and early warning system according to claim 1, characterized in that The calculation logic of the electromagnetic coupling characteristic parameters is as follows: κ(t) is the electromagnetic coupling coefficient, i(t) is the value of the current signal at time t, B(t) is the value of the magnetic field signal at time t, ||B(t)|| is the magnitude of the magnetic field strength collected at time t, and ∈ is a constant.

3. An integrated circuit fault diagnosis and early warning system according to claim 1, characterized in that, The calculation logic of the dynamic distortion index is as follows: ρ(t) is the dynamic impedance distortion rate, and Z inst (t) is the instantaneous impedance, and Z avg is the average impedance, and Z avg The calculation logic of is as follows: T is the length of the sliding time window.

4. An integrated circuit fault diagnosis and early warning system according to claim 1, characterized in that, The sensitivity evaluation is obtained by constructing a non-commutative operator, calculating the eigenvalue shift, and then calculating the sensitivity score according to the operator and the eigenvalue shift; The construction logic of the non-commutative operator is as follows: H(t) is a Hilbert-Schmidt operator, η(t) is a dynamic distortion factor, Δv(t) is a voltage change, and Δv(t) * is the complex conjugate term of the voltage change; The calculation logic of the eigenvalue shift is as follows; δλ(t) is the eigenvalue shift; The calculation logic of the sensitivity score is as follows: Q(t) is the sensitivity score, σ λ is the degree of fluctuation of the change amount, T is the length of the sliding time window, h is Planck's constant, k B is the Boltzmann constant, T e is the electron temperature, and the calculation logic of the electron temperature T e is as follows: q is the electron charge, R thermal is the transient thermal resistance, and the calculation logic of the transient thermal resistance R thermal is as follows: Δi is the change in current, and it is defined in the formula that Δi = 1 mA.

5. The integrated circuit fault diagnosis and early warning system according to claim 4, characterized in that The calculation logic of the dynamic distortion factor is as follows: η(t) is the dynamic distortion factor, μ0 is the vacuum permeability, φ(t) is the electromagnetic phase mismatch degree, τ is the decay time constant, i(t) is the value of the current signal at time t, and B(t) is the value of the magnetic field signal at time t, where f max is the highest frequency of the signal, and the calculation logic of the electromagnetic phase mismatch degree is as follows: ||i(t)|| is the amplitude of the current signal, is the binary exclusive OR operation.

6. The integrated circuit fault diagnosis and early warning system according to claim 1, characterized in that, For the fault mode recognition, the fractal permeability is extracted by using the Poincaré section mapping, and the fault is classified by combining the maximum Lyapunov exponent to generate the fault category; The construction logic of the Poincaré section is as follows: In the phase space (η(t), Q(t)), record the intersection points {(η k , Q k )} of the trajectory and the hyperplane ∑, where the hyperplane ∑ is defined as: aη(t) + bQ(t) = C, where a, b, and c are coefficients; The calculation logic of the fractal permeability is as follows: FP is the fractal permeability, r is a measure of distance, N(r) is the number of points within a neighborhood of radius r, and the calculation logic of N(r) is: N(r) = #{||(η k ,Q k ) - (η m ,Q m )|| < r}, #{·} is the number of elements that meet the conditions within {·}, and ||(η k ,Q k ) - (η m ,Q m )|| is the distance between the data points (η k ,Q k ) and (η m ,Q m ); The generation logic of the fault category is: y = argmax[FP·exp(Λ max )·||w c ||1], where y is the fault category, Λ max is the maximum Lyapunov exponent, w c is the weight vector, and the calculation logic of Λ max is: J(t) is the Jacobian matrix, and the generation logic of the Jacobian matrix J(t) is: The fault categories include: when y = 1, it represents a short circuit fault; when y = 2, it represents an open circuit fault; when y = 3, it represents an aging fault; when y = 4, it represents an electromagnetic interference fault; when y = 5, it represents quantum tunneling leakage; when y = 6, it represents thermal runaway.

7. An integrated circuit fault diagnosis and early warning system according to claim 1, characterized in that The calculation logic of the failure probability is as follows: P f is the failure probability, and β is the weight coefficient; Adjust the failure probability P f The adjustment logic for the failure probability is as follows: is the updated failure probability, is the failure probability calculated last time, is the failure probability calculated currently, and α is the smoothing factor.

8. An integrated circuit fault diagnosis and early warning system according to claim 7, characterized in that, Calculate the dynamic threshold based on the historical data of failure probability, and the calculation logic of the dynamic threshold is as follows: T new is the dynamic threshold, is the mean difference of the failure probability, is the standard deviation of the failure probability, and k1 and k2 are constant coefficients; Set the maximum and minimum threshold ranges according to the historical dynamic threshold, where the dynamic threshold T new ranges as follows: T min ≤T new ≤T max , where T min is the minimum threshold and T max is the maximum threshold.

9. An integrated circuit fault diagnosis and early warning system according to claim 8, characterized in that, Execute the hierarchical fault response strategy jointly according to the fault category and the updated fault probability. When and y ∈ {1, 5, 6}, trigger an emergency shutdown and alarm; when and y ∈ {2, 3, 4}, start the frequency reduction operation mode; when , maintain the normal operation state without any operation.

10. The integrated circuit fault diagnosis and early warning system according to claim 1, characterized in that, The time-domain alignment adopts a quantum tunneling synchronization protocol with a clock jitter less than 1 ps. The calculation logic for the synchronization error compensation of the voltage signal and the current signal is: δt = argmin||v(t) - i(t + δ)|| 2 , and the calculation logic for the synchronization error compensation of the magnetic field signal and the current signal is: δt = argmin||||B(t)|| - i(t + δ′0)||| 2 , where v(t) is the value of the voltage signal at time t, i(t + δ) is the value of the current signal after adjusting the time offset δ, i(t + δ′0) is the value of the current signal after adjusting the time offset δ′0, and δ and δ′0 are the offsets.

Images