Method and system for evaluating temperature and stress states of closing resistor stack

Abstract

The invention discloses a temperature and stress state evaluation method and system for a closing resistor pile, and relates to the field of temperature and stress state evaluation, and the method comprises the steps: arranging optical fiber temperature measurement sensors at different parts of the resistor pile, constructing a two-dimensional temperature monitoring network, and transmitting data to a processing system through a wireless network; in combination with real-time current collection and power monitoring, heat generation and heat load distribution in the resistor pile are analyzed; stress-temperature coupling evaluation is realized by constructing a mathematical model, and the stress state of the resistor stack is analyzed; performing time sequence analysis on the temperature and the stress, establishing a thermal fatigue damage accumulation model based on a change rule, and evaluating the fatigue life; integrating the temperature, the stress and the fatigue damage degree, and adopting a machine learning algorithm to predict the fault risk. According to the invention, the working state of the resistor stack is monitored in multiple dimensions, the potential fault risk is accurately evaluated, and a basis is provided for optimized operation of equipment.

Description

Technical Field

[0001] The present invention belongs to the technical field of temperature and stress state assessment, and more specifically, relates to a method and system for assessing the temperature and stress state of a closing resistor stack. Background Art

[0002] As power systems become increasingly complex, the temperature, stress, and fatigue state of resistor stacks directly impact the safety, reliability, and long-term operational performance of equipment. Therefore, accurately assessing the temperature and stress state of resistor stacks is crucial for early warning of failures, optimizing maintenance cycles, and improving the safety of power equipment. However, existing methods and systems for assessing the temperature and stress state of closing resistor stacks still face several challenges and issues.

[0003] Currently, most resistor stack temperature and stress monitoring systems rely on a single sensor or measurement point, resulting in insufficient representativeness and accuracy of monitoring data. Especially in complex operating environments, where temperature and stress distribution may be uneven, traditional monitoring methods cannot fully reflect the overall operating status of the resistor stack. Different parts of the resistor stack may be subject to different loads and environmental influences, making it inaccurate to assess the overall temperature and stress status based on only partial data. In existing systems, sensor placement is often lacking in specificity, making it difficult to fully reflect the thermal load and stress distribution within the resistor stack. Many resistor stack monitoring systems fail to fully account for the complexity of the internal and external environments of the resistor stack, resulting in missing or significant errors in monitoring data. The accuracy and stability of temperature and stress sensors can also affect assessment results, especially under high temperature, high stress, or harsh environmental conditions. Current temperature and stress assessment methods rely on traditional physical modeling and empirical formulas, which often ignore the complex dynamic changes during resistor stack operation. Assessing the overall status of the resistor stack based on simple temperature and stress data is prone to errors. When faced with complex nonlinear relationships, traditional methods struggle to accurately predict faults and assess status. Summary of the Invention

[0004] In view of the problems of the above or existing methods and systems for evaluating the temperature and stress state of a closing resistor stack, the present invention is proposed.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] An embodiment of the present invention provides a method for evaluating the temperature and stress state of a closing resistor stack, comprising: arranging optical fiber temperature sensors at different locations of the resistor stack to form a two-dimensional temperature monitoring network, and transmitting temperature data to a data processing system via a wireless network;

[0007] Utilize real-time current acquisition and power monitoring, combined with the current input and output changes of the resistor stack, to estimate the heat generation and heat load distribution inside the resistor stack.

[0008] Construct a comprehensive thermal stress mathematical model and evaluate the stress state of the resistor stack through stress-temperature coupling;

[0009] Use Kalman filter to perform time series analysis on resistor stack temperature and stress;

[0010] Based on the changes in temperature and stress, a thermal fatigue damage accumulation model is constructed to evaluate the fatigue life of the resistor stack under set operating conditions;

[0011] Temperature, stress, and fatigue damage are combined to assess the failure risk of resistor stacks using machine learning algorithms.

[0012] As a preferred embodiment of the method for evaluating the temperature and stress state of a closing resistor stack according to the present invention, optical fiber temperature sensors are arranged at different locations of the resistor stack to form a two-dimensional temperature monitoring network, and temperature data is transmitted to a data processing system via a wireless network; the method comprises:

[0013] Fiber optic temperature sensors are placed at the input and output ends, resistor core, and power supply port of the resistor stack. Thermocouple arrays are evenly arranged in both horizontal and vertical dimensions according to the structure of the resistor stack.

[0014] The temperature data formula measured by the temperature sensor is:

[0015]

[0016] Where i is the sensor number, t is the time, f(Ti(t),R) is the heat conduction model, and R is the thermal conductivity of the resistor material.

[0017] As a preferred embodiment of the method for evaluating the temperature and stress state of a closing resistor stack according to the present invention, the method utilizes real-time current acquisition and power monitoring, combined with the current input and output changes of the resistor stack, to estimate the heat generation and heat load distribution within the resistor stack, including:

[0018] The current signal I(t) and voltage signal V(t) are obtained by the current sensor, and the power P(t) is calculated:

[0019]

[0020] Using the electric power and thermal characteristics model of the resistor stack, estimate the heat generation rate of the resistor stack:

[0021]

[0022] Where α is the thermal power conversion coefficient and Q(t) is the heat generated per unit time.

[0023] As a preferred embodiment of the method for evaluating the temperature and stress state of a closing resistor stack according to the present invention, the method of constructing a mathematical model of comprehensive thermal stress and evaluating the stress state of the resistor stack through stress-temperature coupling includes:

[0024] Use thermal expansion coefficient α T The thermal stress formula is constructed by combining the Young's modulus E and the temperature field distribution. Assuming that the closing resistor stack undergoes one-dimensional thermal expansion, the thermal stress generated during the temperature change process is calculated as follows:

[0025]

[0026] Where ΔT(t) is the temperature difference, α T is the thermal expansion coefficient of the material.

[0027] As a preferred solution of the temperature and stress state evaluation method of the closing resistor stack of the present invention, the method of using Kalman filtering to perform time series analysis on the temperature and stress of the resistor stack includes:

[0028] If the dynamic behavior of the system is linear, the state equation can be expressed as:

[0029]

[0030] Where x(t) is the state vector, representing temperature and stress, A is the state transfer matrix, describing the state change from the previous moment to the current moment, u(t) is the external control input, B is the control input matrix, describing the impact of the input on the state, and w(t) is the process noise, which is assumed to be zero-mean Gaussian white noise with a variance of Q.

[0031] If the measured values ​​of temperature and stress are affected by measurement noise, it can be expressed as:

[0032]

[0033] Where z(t) is the measured value, which is the actual measured value of temperature and stress respectively, H is the observation matrix, and v(t) is the measurement noise, which is assumed to be zero-mean Gaussian noise with variance R;

[0034] According to the time series temperature data Ti(t), the thermal stress prediction is optimized by Kalman filtering:

[0035]

[0036] Among them, K(t) is the Kalman gain, which reflects the weight of the current estimation error and prediction error.

[0037] As a preferred embodiment of the method for evaluating the temperature and stress state of a closing resistor stack according to the present invention, the method of constructing a thermal fatigue damage accumulation model based on temperature and stress changes to evaluate the fatigue life of the resistor stack under set operating conditions includes:

[0038] The damage degree D(t) of each thermal stress cycle is calculated by applying the Miner linear accumulation method based on the fluctuation frequency of temperature and stress:

[0039]

[0040] Among them, n i is the number of damages in the i-th stress cycle, N i is the fatigue life under this stress amplitude,

[0041] When the cumulative damage value D(t) is greater than a predetermined threshold, it indicates that the closing resistor stack has entered the critical fatigue stage and needs to be repaired.

[0042] As a preferred embodiment of the method for evaluating the temperature and stress state of a closing resistor stack according to the present invention, the method of combining temperature, stress, and fatigue damage degree and using a machine learning algorithm to evaluate the failure risk of the resistor stack includes:

[0043] Define the failure risk R as a classification label. The model can predict the failure risk by learning the relationship between temperature T, stress σ, damage degree D and failure risk. The specific model form is as follows:

[0044]

[0045] Where f is the decision function of the machine learning model, T is the temperature of the resistor stack, σ is the stress of the resistor stack, and D is the fatigue damage degree of the resistor stack;

[0046] If the failure probability P output by the model is low risk ≤0.7, the device is considered to be in the low-risk category;

[0047] If P high risk ≥0.7, the device is considered to be in the high-risk category.

[0048] A temperature and stress state assessment system for a closing resistor stack includes: a temperature monitoring module for arranging optical fiber temperature sensors at different locations of the resistor stack to form a two-dimensional temperature monitoring network, and transmitting temperature data to a data processing system via a wireless network;

[0049] The current and power analysis module is used to use real-time current acquisition and power monitoring, combined with the current input and output changes of the resistor stack, to infer the heat generation and heat load distribution inside the resistor stack;

[0050] Model building module, used to construct a comprehensive thermal stress mathematical model and evaluate the stress state of the resistor stack through stress-temperature coupling;

[0051] Timing analysis module, used to perform timing analysis on resistor stack temperature and stress using Kalman filtering;

[0052] The fatigue life assessment module is used to build a thermal fatigue damage accumulation model based on temperature and stress changes to assess the fatigue life of the resistor stack under set operating conditions;

[0053] The fault assessment module is used to combine temperature, stress, and fatigue damage to evaluate the failure risk of resistor stacks using machine learning algorithms.

[0054] The present invention has the following beneficial effects: By combining the temperature, stress, and fatigue damage of the resistor stack and utilizing a machine learning algorithm for fault risk assessment, the system can analyze the operating status and potential failure risks of the closing resistor stack from multiple dimensions, more accurately identifying potential equipment failures and providing early warnings, thereby reducing safety hazards caused by equipment failures. The system accurately collects relevant parameters of the closing resistor stack and accurately determines the operating status of the closing resistor. By comprehensively considering the complex relationship between the temperature, stress, and fatigue damage of the resistor stack, it can effectively avoid assessment bias caused by a single factor. BRIEF DESCRIPTION OF THE DRAWINGS

[0055] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for use in the description of the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0056] Figure 1 This is a flow chart of a method for evaluating the temperature and stress state of a closing resistor stack provided by an embodiment of the present invention.

[0057] Figure 2 A schematic structural diagram of a temperature and stress state assessment system for a closing resistor stack provided in an embodiment of the present invention.

[0058] In the drawings, the same or corresponding reference numerals denote the same or corresponding parts. DETAILED DESCRIPTION

[0059] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below with reference to the accompanying drawings.

[0060] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0061] Secondly, the term "one embodiment" or "embodiment" herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in various places throughout this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive of other embodiments.

[0062] Example

[0063] Reference below Figure 1 , Figure 1 This is a flow chart of a method for evaluating the temperature and stress state of a closing resistor stack according to an embodiment of the present invention. It should be noted that the embodiments of the present invention can be applied to any applicable scenario.

[0064] Figure 1 The process of the temperature and stress state evaluation method of a closing resistor stack provided by an embodiment of the present invention includes:

[0065] S1: Fiber optic temperature sensors are installed at different parts of the resistor stack to form a two-dimensional temperature monitoring network, and the temperature data is transmitted to the data processing system through a wireless network.

[0066] Preferably, optical fiber temperature sensors are arranged at the input end, output end, resistor core and power port of the resistor stack, and thermocouple arrays are evenly arranged in both horizontal and vertical dimensions according to the structure of the resistor stack;

[0067] The temperature data formula measured by the temperature sensor is:

[0068]

[0069] Where i is the sensor number, t is the time, f(Ti(t),R) is the heat conduction model, and R is the thermal conductivity of the resistor material.

[0070] Furthermore, multiple fiber optic temperature sensors are deployed at different locations along the resistor stack to ensure uniform distribution of temperature data both horizontally and vertically. Each sensor regularly collects temperature data, including the sensor number, acquisition time, and measured temperature. This collected temperature data is transmitted to a central control system via a wireless or wired network. The measured temperature data is processed and analyzed using the heat conduction model f(Ti(t),R), calculating the temperature distribution at each sensor point and analyzing temperature trends. Based on real-time monitoring data and the heat conduction model, the temperature status of the resistor stack and potential failure risks are assessed. Based on the assessment results, the intelligent system can trigger an alarm to warn of potential failures or automatically adjust equipment operating parameters to ensure safe operation.

[0071] Furthermore, suppose that the temperature sensors of a resistor stack are numbered from 1 to 10, distributed in different positions of the resistor stack, and the time interval is 5 seconds. The temperature data recorded by the system is shown in the table:

[0072]

[0073] Based on the above temperature data, the system will conduct further analysis according to the heat conduction model to determine the overall temperature distribution of the resistor stack and potential failure risks to ensure the safe and stable operation of the resistor stack.

[0074] S2: Utilize real-time current acquisition and power monitoring, combined with the current input and output changes of the resistor stack, to estimate the heat generation and heat load distribution inside the resistor stack.

[0075] Preferably, the current signal I(t) and the voltage signal V(t) are obtained by a current sensor, and the power P(t) is calculated:

[0076]

[0077] Using the electric power and thermal characteristics model of the resistor stack, estimate the heat generation rate of the resistor stack:

[0078]

[0079] Where α is the thermal power conversion coefficient and Q(t) is the heat generated per unit time.

[0080] Furthermore, assume that the current signal and voltage signal of a resistor stack at time t1t_1t1 are as follows:

[0081] The current signal I(t1) = 5 A; the voltage signal V(t1) = 10 V; then the instantaneous power P(t1) of the resistor stack is: P(t1) = 5 A × 10 V = 50 W. Assuming the thermal power conversion coefficient α = 0.85, the heat generated per unit time Q(t1) is: Q(t1) = 0.85 × 50 W = 42.5 W. In this way, the heat rate generated by the resistor stack at time t1 can be obtained as Q(t1) = 42.5 W.

[0082] S3: Construct a mathematical model of comprehensive thermal stress and evaluate the stress state of the resistor stack through stress-temperature coupling.

[0083] Preferably, the thermal expansion coefficient α is used T The thermal stress formula is constructed by combining the Young's modulus E and the temperature field distribution. Assuming that the closing resistor stack undergoes one-dimensional thermal expansion, the thermal stress generated during the temperature change process is calculated as follows:

[0084]

[0085] Where ΔT(t) is the temperature difference, α T is the thermal expansion coefficient of the material.

[0086] Furthermore, suppose we have a closing resistor stack whose resistance material has the following properties:

[0087] Young's modulus E=2.1×10 11 Pa; thermal expansion coefficient αT = 1.2 × 10 -5  1 / °C; temperature difference ΔT(t) = 50 °C (assuming the temperature difference increases from the initial temperature to 50°C).

[0088] Based on the above data, the thermal stress of the resistor stack material can be calculated.

[0089] Calculation: σ(t)=12.6MPa;

[0090] Therefore, when the temperature difference is 50°C, the thermal stress of the resistor stack material is 12.6 MPa.

[0091] Assuming that the temperature of the resistor stack has a gradient along the horizontal or vertical direction (for example, the temperature is higher in the center and lower at the edge), use the following steps to perform a point-by-point calculation:

[0092] Use an optical fiber temperature sensor or an infrared temperature sensor to measure the temperature of the resistor stack at different locations; for each measurement location, calculate the temperature difference ΔT between it and the initial temperature. i (t)=T i (t)-T ref , where T i(t) is the instantaneous temperature at position i, T ref is the reference temperature. Using the thermal expansion coefficient αT and Young's modulus E, for each position, the local thermal stress is calculated:

[0093]

[0094] If the overall thermal stress of the resistor stack needs to be solved, an overall thermal stress value can be obtained by taking the weighted average of all local stresses.

[0095] S4: Use Kalman filtering to perform timing analysis on resistor stack temperature and stress.

[0096] Preferably, if the dynamic behavior of the system is linear, the state equation is expressed as:

[0097]

[0098] Where x(t) is the state vector, representing temperature and stress, A is the state transfer matrix, describing the state change from the previous moment to the current moment, u(t) is the external control input, B is the control input matrix, describing the impact of the input on the state, and w(t) is the process noise, which is assumed to be zero-mean Gaussian white noise with a variance of Q.

[0099] If the measured values ​​of temperature and stress are affected by measurement noise, it can be expressed as:

[0100]

[0101] Where z(t) is the measured value, which is the actual measured value of temperature and stress respectively, H is the observation matrix, and v(t) is the measurement noise, which is assumed to be zero-mean Gaussian noise with variance R;

[0102] According to the time series temperature data Ti(t), the thermal stress prediction is optimized by Kalman filtering:

[0103]

[0104] Among them, K(t) is the Kalman gain, which reflects the weight of the current estimation error and prediction error.

[0105] Furthermore, suppose we have a resistor stack system whose temperature T(t) and stress σ(t) state vectors are expressed as:

[0106]

[0107] State transfer matrix A: Assuming that the changes of temperature and stress at each moment are linear, it is set as the unit matrix.

[0108]

[0109] Control input matrix B: Assuming that the influence of external control input on temperature and stress is known, it can be set as:

[0110]

[0111] Observation matrix H: Assume that the measurements depend only on the current state of temperature and stress.

[0112]

[0113] The process noise variance Q and the measurement noise variance R are assumed to be known constants and are usually obtained through system modeling or experiments.

[0114]

[0115] Assume initial conditions:

[0116] Initial state The initial temperature is 20°C and the stress is 5 MPa;

[0117] Initial error covariance P(0):

[0118]

[0119] Assume we have the following temperature and stress measurements:

[0120]

[0121] The state and error covariance are predicted based on the system model, and the state estimate is updated based on the measurement data. For each new time step, the prediction and update process is repeated until the optimal state estimate for all moments is obtained.

[0122] S5: Based on the changes in temperature and stress, a thermal fatigue damage accumulation model is constructed to evaluate the fatigue life of the resistor stack under set operating conditions.

[0123] Preferably, the damage degree D(t) of each thermal stress cycle is calculated by applying the Miner linear accumulation method based on the fluctuation frequency of temperature and stress:

[0124]

[0125] Among them, n i is the number of damages in the i-th stress cycle, N i is the fatigue life under this stress amplitude,

[0126] When the cumulative damage value D(t) is greater than a predetermined threshold, it indicates that the closing resistor stack has entered the critical fatigue stage and needs to be repaired.

[0127] Furthermore, assume that the temperature and stress data of the resistor stack are as follows:

[0128] Temperature fluctuation: T(t)=[20∘C, 22∘C, 25∘C, 23∘C, … ]

[0129] Stress fluctuation: σ(t)=[5MPa,7MPa,10MPa,8MPa,…]

[0130] Through cycle analysis, it is assumed that the maximum stress σmax(t) and minimum stress σmin(t) of each cycle are extracted from the stress data, and the stress amplitude Δσ is calculated:

[0131] Δσ1=10MPa-5MPa=5MPa, Δσ2=8MPa-4MPa=4MPa;

[0132] Assume that the relationship between the fatigue life of the material and the stress amplitude is:

[0133]

[0134] Assume that the constant obtained through experiment is C=10 6 and m=3, then for each cycle stress amplitude, estimate its fatigue life Ni:

[0135] For the first cycle, Δσ1=5MPa, so: N1=8000 times;

[0136] For the second cycle, Δσ2=4MPa, so: N2=15625 times;

[0137] According to the number of damages and fatigue life in each cycle, the Miner linear accumulation method is used to calculate the damage degree:

[0138] For the first cycle: D1(t)=0.000125;

[0139] For the second cycle: D2(t)=0.000064;

[0140] The cumulative damage degree D(t)D(t)D(t) is the sum of the damage degrees of all cycles:

[0141] D(t)=D1(t)+D2(t)=0.000125+0.000064=0.000189;

[0142] The threshold value of the cumulative damage degree is set to Dth=0.01. When D(t) exceeds this threshold, it means that the resistor stack has entered the critical fatigue stage and needs to be repaired.

[0143] At the current moment, D(t)=0.000189, which is much smaller than Dth=0.01, so the resistor stack does not need to be repaired immediately.

[0144] S6: Combine temperature, stress, and fatigue damage to assess the failure risk of resistor stacks using machine learning algorithms.

[0145] Preferably, the failure risk R is defined as a classification label. The model can predict the failure risk by learning the relationship between temperature T, stress σ, damage degree D and failure risk. The specific model form is as follows:

[0146]

[0147] Where f is the decision function of the machine learning model, T is the temperature of the resistor stack, σ is the stress of the resistor stack, and D is the fatigue damage degree of the resistor stack;

[0148] If the failure probability P output by the model is low risk ≤0.7, the device is considered to be in the low-risk category;

[0149] If P high risk ≥0.7, the device is considered to be in the high-risk category.

[0150] Furthermore, assume that the following data is obtained during real-time monitoring:

[0151] Temperature T = 75∘C, stress σ = 35 MPa, damage degree D = 0.05, the model prediction results are:

[0152] P lowrisk =0.45, P high risk =0.5;

[0153] According to the set threshold: high risk =0.55 is less than 0.7, so the device is judged to be low risk.

[0154] In order to achieve real-time monitoring and automatic alarm, an alarm system can be built. When the failure risk exceeds a predetermined threshold, the system will automatically issue an alarm to prompt maintenance personnel to conduct inspections or shut down the machine for maintenance. For example:

[0155] When P high risk ≥0.7, triggering a high-risk alert;

[0156] When P lowrisk ≤0.7 and P high risk <0.7, triggering a low-risk monitoring prompt.

[0157] After introducing the method of the exemplary embodiment of the present invention, next, reference is made to Figure 2A temperature and stress state evaluation system for a closing resistor stack according to an exemplary embodiment of the present invention is described. The system includes:

[0158] The temperature monitoring module is used to set optical fiber temperature sensors at different parts of the resistor stack to form a two-dimensional temperature monitoring network, and transmit temperature data to the data processing system through a wireless network;

[0159] The current and power analysis module is used to use real-time current acquisition and power monitoring, combined with the current input and output changes of the resistor stack, to infer the heat generation and heat load distribution inside the resistor stack;

[0160] Model building module, used to construct a comprehensive thermal stress mathematical model and evaluate the stress state of the resistor stack through stress-temperature coupling;

[0161] Timing analysis module, used to perform timing analysis on resistor stack temperature and stress using Kalman filtering;

[0162] The fatigue life assessment module is used to build a thermal fatigue damage accumulation model based on temperature and stress changes to assess the fatigue life of the resistor stack under set operating conditions;

[0163] The fault assessment module is used to combine temperature, stress, and fatigue damage to evaluate the failure risk of resistor stacks using machine learning algorithms.

[0164] Evaluate the overall operation process of the system;

[0165] Data acquisition stage: The temperature monitoring module collects the temperature data of the resistor stack in real time through the optical fiber temperature sensor, and the current and power analysis module obtains the current and voltage signals through the current sensor.

[0166] Data analysis phase: The model building module calculates thermal stress based on temperature and power data, and the time series analysis module uses Kalman filtering to optimize thermal stress prediction.

[0167] Condition assessment stage: The fatigue life assessment module calculates the fatigue life of the resistor stack based on the thermal fatigue damage accumulation model, and the fault assessment module uses a machine learning algorithm to assess the failure risk of the resistor stack.

[0168] Result output: The system displays the final evaluation results (temperature distribution, stress state, fatigue life and failure risk) to the user and issues a warning message when the predetermined threshold is exceeded.

[0169] The principle of fiber optic temperature measurement is that a continuous laser wave emitted by a laser diode enters the optical fiber. As the light wave travels down the fiber's glass core, it generates various types of radiation scattering, including Rayleigh scattering, Brillouin scattering, and Raman scattering. Raman scattering is the most temperature-sensitive type of scattered light. Raman scattering occurs at every point in the optical fiber, and the resulting Raman scattered light is uniformly distributed across the entire spatial angle. Raman scattering is caused by the energy exchange between the thermal vibrations of the optical fiber molecules and photons. Specifically, if a portion of the light energy is converted into thermal vibrations, it emits light with a longer wavelength than the source, known as Stokes light. If a portion of the thermal vibrations is converted into light energy, it emits light with a shorter wavelength than the source, known as anti-Stokes light. The intensity of Stokes light is negligibly affected by temperature, while the intensity of Anti-Stokes light varies with temperature. The ratio of the intensities of Anti-Stokes to Stokes light provides a functional relationship with temperature. When light is transmitted through an optical fiber, a portion of Raman scattered light (backward Raman scattered light) returns along the original path of the fiber and is received by the optical fiber detection unit. By measuring the changes in the intensity ratio of Anti-Stokes light to Stokes light in the back Raman scattered light, external temperature changes can be monitored. Triggered by a synchronous control unit, the optical transmitter generates a high-current pulse, which drives a semiconductor laser to generate a high-power optical pulse that is injected into the laser pigtail. The optical pulse output from the laser pigtail passes through an optical coupler into an optical fiber placed in a constant-temperature bath. This optical fiber is used for system calibration, and then enters a sensing fiber to sense the temperature field of the closing resistor stack. When the laser light is scattered in the fiber, the Raman backscattered light carrying temperature information returns to the optical coupler. The optical coupler not only couples the emitted light directly to the sensing fiber, but also couples the scattered Raman scattered light of a different wavelength from the emitted light to a spectrometer. The spectrometer consists of two optical filters with different central wavelengths, which respectively output Slokes light and Anti-Stokes light. These are transmitted to the data processing system via a wireless network, where the temperature is stored, displayed, and controlled.

[0170] Finally, it should be noted that the above-described embodiments are only specific implementation methods of the present invention, which are used to illustrate the technical solutions of the present invention, rather than to limit them. The scope of protection of the present invention is not limited thereto. Although the present invention has been described in detail with reference to the above-described embodiments, those skilled in the art should understand that any person skilled in the art can modify or easily conceive of changes to the technical solutions described in the above-described embodiments within the technical scope disclosed by the present invention, or replace some of the technical features therein with equivalents. Such modifications, changes, or replacements do not deviate from the spirit and scope of the technical solutions of the embodiments of the present invention, and should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention shall be subject to the scope of protection of the claims.

[0171] Furthermore, although the operations of the method of the present invention are described in a particular order in the accompanying drawings, this does not require or imply that these operations must be performed in this particular order, or that all illustrated operations must be performed to achieve the desired results. Additionally or alternatively, some steps may be omitted, multiple steps may be combined into one step, and / or one step may be decomposed into multiple steps.

Claims

1. A method for evaluating the temperature and stress state of a closing resistor stack, characterized in that: include: Step (1): Fiber optic temperature sensors are installed at different locations of the resistor stack to form a two-dimensional temperature monitoring network, and the temperature data is transmitted to the data processing system via a wireless network; Step (2): Utilize real-time current acquisition and power monitoring, combined with the current input and output changes of the resistor stack, to estimate the heat generation and heat load distribution inside the resistor stack; Step (3): Construct a mathematical model of comprehensive thermal stress and evaluate the stress state of the resistor stack through stress-temperature coupling; Step (4): Use Kalman filtering to perform time series analysis on the temperature and stress of the resistor stack; Step (5): Based on the changes in temperature and stress, a thermal fatigue damage accumulation model is constructed to evaluate the fatigue life of the resistor stack under the set operating conditions; Step (6): Combine temperature, stress, and fatigue damage to evaluate the failure risk of the resistor stack using a machine learning algorithm.

2. The method for evaluating the temperature and stress state of a closing resistor stack according to claim 1, wherein: The optical fiber temperature sensors are arranged at different parts of the resistor stack to form a two-dimensional temperature monitoring network, and the temperature data is transmitted to the data processing system via a wireless network, including: Optical fiber temperature sensors are placed at the input and output ends, resistor core, and power supply port of the resistor stack. Based on the structure of the resistor stack, the optical fiber temperature measurement array is evenly arranged in both the horizontal and vertical dimensions. The temperature data formula measured by the temperature sensor is: , Where i is the sensor number, t is the time, f(Ti(t),R) is the heat conduction model, and R is the thermal conductivity of the resistor material.

3. The method for evaluating the temperature and stress state of a closing resistor stack according to claim 1, wherein: The method utilizes real-time current acquisition and power monitoring, combined with the current input and output changes of the resistor stack, to estimate the heat generation and heat load distribution inside the resistor stack, including: The current signal I(t) and voltage signal V(t) are obtained by the current sensor, and the power P(t) is calculated: , Using the electric power and thermal characteristics model of the resistor stack, estimate the heat generation rate of the resistor stack: , Where α is the thermal power conversion coefficient and Q(t) is the heat generated per unit time.

4. The method for evaluating the temperature and stress state of a closing resistor stack according to claim 1, wherein: The mathematical model of comprehensive thermal stress is constructed to evaluate the stress state of the resistor stack through stress-temperature coupling, including: Use thermal expansion coefficient α T The thermal stress formula is constructed by combining the Young's modulus E and the temperature field distribution. Assuming that the closing resistor stack undergoes one-dimensional thermal expansion, the thermal stress generated during the temperature change process is calculated as follows: , Where ΔT(t) is the temperature difference, α T is the thermal expansion coefficient of the material.

5. The method for evaluating the temperature and stress state of a closing resistor stack according to claim 1, wherein: The use of Kalman filtering to perform time series analysis on the resistor stack temperature and stress includes: If the dynamic behavior of the system is linear, the state equation can be expressed as: , Where x(t) is the state vector, representing temperature and stress, A is the state transfer matrix, describing the state change from the previous moment to the current moment, u(t) is the external control input, B is the control input matrix, describing the impact of the input on the state, and w(t) is the process noise, which is assumed to be zero-mean Gaussian white noise with a variance of Q. If the measured values ​​of temperature and stress are affected by measurement noise, it can be expressed as: , Where z(t) is the measured value, which is the actual measured value of temperature and stress respectively, H is the observation matrix, and v(t) is the measurement noise, which is assumed to be zero-mean Gaussian noise with variance R; According to the time series temperature data Ti(t), the thermal stress prediction is optimized by Kalman filtering: , Among them, K(t) is the Kalman gain, which reflects the weight of the current estimation error and prediction error.

6. The method for evaluating the temperature and stress state of a closing resistor stack according to claim 1, wherein: The thermal fatigue damage accumulation model is constructed based on the changes in temperature and stress to evaluate the fatigue life of the resistor stack under the set operating conditions, including: The damage degree D(t) of each thermal stress cycle is calculated by applying the Miner linear accumulation method based on the fluctuation frequency of temperature and stress: , Among them, n i is the number of damages in the i-th stress cycle, N i is the fatigue life under this stress amplitude, When the cumulative damage value D(t) is greater than a predetermined threshold, it indicates that the closing resistor stack has entered the critical fatigue stage and needs to be repaired.

7. The method for evaluating the temperature and stress state of a closing resistor stack according to claim 1, wherein: The method combines temperature, stress, and fatigue damage to assess the failure risk of resistor stacks using a machine learning algorithm, including: Define the failure risk R as a classification label. The model can predict the failure risk by learning the relationship between temperature T, stress σ, damage degree D and failure risk. The specific model form is as follows: , Where f is the decision function of the machine learning model, T is the temperature of the resistor stack, σ is the stress of the resistor stack, and D is the fatigue damage degree of the resistor stack; If the failure probability P output by the model is low risk ≤0.7, the device is considered to be in the low-risk category; If P high risk ≥0.7, the device is considered to be in the high-risk category.

8. A temperature and stress state evaluation system for a closing resistor stack, characterized in that: include: The temperature monitoring module is used to set optical fiber temperature sensors at different parts of the resistor stack to form a two-dimensional temperature monitoring network, and transmit temperature data to the data processing system through a wireless network; The current and power analysis module is used to use real-time current acquisition and power monitoring, combined with the current input and output changes of the resistor stack, to infer the heat generation and heat load distribution inside the resistor stack; Model building module, used to construct a comprehensive thermal stress mathematical model and evaluate the stress state of the resistor stack through stress-temperature coupling; Timing analysis module, used to perform timing analysis on resistor stack temperature and stress using Kalman filtering; The fatigue life assessment module is used to build a thermal fatigue damage accumulation model based on temperature and stress changes to assess the fatigue life of the resistor stack under set operating conditions; The fault assessment module is used to combine temperature, stress, and fatigue damage to evaluate the failure risk of resistor stacks using machine learning algorithms.

Images