Intelligent monitoring method for winding type transformer iron core annealing process

Abstract

The invention discloses an intelligent monitoring method for an annealing process of a wound transformer iron core, and particularly relates to the technical field of intelligent monitoring. The method comprises the following steps: acquiring temperature data and magnetic field intensity data generated in an annealing process of a wound transformer iron core, and carrying out time domain synchronization processing and abnormal data labeling; identifying an iron core micro-deformation area and a corresponding temperature anomaly area according to the synchronous data, and extracting temperature anomaly fluctuation characteristics; analyzing a dynamic coupling relationship between the temperature abnormal fluctuation characteristic and the iron core stress change, and outputting characteristic data of an iron core magnetic field and stress interaction; synchronously analyzing the interference intensity of abnormal temperature fluctuation on the annealing heat distribution uniformity, and outputting temperature disturbance intensity data; and constructing a coupling anomaly evolution model, generating iron core annealing anomaly evaluation data, quantifying a quality fluctuation risk in an annealing process, and generating intelligent early warning information. The monitoring precision and reliability of the iron core annealing process are effectively improved, and the annealing quality of the transformer iron core is guaranteed.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology, and more specifically, to an intelligent monitoring method for the annealing process of wound transformer cores. Background Technology

[0002] The core of a wound transformer is usually made of continuous silicon steel sheets or strips using specialized winding equipment, and is widely used in power transformers as a key magnetic component.

[0003] Existing quality assessment methods for the annealing process of wound transformer cores only focus on single indicators such as local temperature or static magnetic properties. They cannot effectively monitor and accurately identify the multi-field coupling effects between the magnetic field, temperature field, and stress field during the annealing process and their impact on the core performance. This makes it difficult to detect and warn of coupled quality defects in the annealing process in a timely manner, increasing the risk of overall quality fluctuations and uncertainties after the annealing of wound transformer cores. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide an intelligent monitoring method for the annealing process of wound transformer cores to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution: A method for intelligent monitoring of the annealing process of wound transformer cores includes the following steps: S1: Collect temperature and magnetic field strength data generated by the core of the wound transformer during the annealing process, perform time-domain synchronization processing and anomaly data labeling, and generate time-series synchronized data of temperature and magnetic field. S2: Based on the time-series synchronization data of temperature and magnetic field, identify the micro-deformation region of the iron core and the corresponding temperature anomaly region, and generate the temperature anomaly fluctuation characteristics of the micro-deformation region of the iron core. S3: Based on the abnormal temperature fluctuation characteristics of the micro-deformation region of the iron core, analyze the dynamic coupling effect of stress change on the magnetic field distribution, and output the characteristic data of the interaction between the iron core magnetic field and stress. S4: Based on the abnormal temperature fluctuation characteristics of the micro-deformation region of the iron core, analyze the interference intensity of abnormal temperature changes on the uniformity of heat distribution during iron core annealing, and output the temperature disturbance intensity data. S5: Based on the characteristic data of the interaction between the magnetic field and stress in the iron core and the data of the temperature disturbance intensity of the iron core, a coupled anomaly evolution model of magnetic field-thermal field-stress field is established to generate iron core annealing anomaly assessment data. S6: Assess the quality fluctuation risk of the iron core annealing process based on the iron core annealing anomaly assessment data, and generate intelligent early warning information for iron core annealing anomalies.

[0006] In a preferred embodiment, S1 specifically refers to: Real-time temperature data of the core of a wound transformer during the annealing process was collected. Real-time magnetic field strength data of the core of a wound transformer during the annealing process were collected. Real-time temperature data and real-time magnetic field strength data are aligned according to a unified time axis and integrated synchronously into a unified time-series data sequence. Abnormal data points that exceed the preset normal range in a unified time-series data sequence are marked as abnormal data to form time-series synchronized data of temperature and magnetic field.

[0007] In a preferred embodiment, S2 specifically refers to: The magnetic field strength difference and corresponding temperature difference between adjacent sampling points are calculated based on time-series synchronous data to generate magnetic field gradient sequences and temperature gradient sequences. The initial candidate region for micro-deformation of the iron core is determined in the magnetic field gradient sequence based on a preset magnetic field gradient threshold. Based on the preset temperature gradient threshold, candidate areas for temperature anomalies are determined in the temperature gradient sequence, and their spatial positions are matched with the initial candidate areas for micro-deformation of the iron core to determine the micro-deformation areas and temperature anomaly areas of the iron core that completely overlap in spatial position. The time-domain mean, time-domain variance, and frequency-domain amplitude of temperature data within the temperature anomaly region are calculated to generate the temperature anomaly fluctuation characteristics of the micro-deformation region of the iron core.

[0008] In a preferred embodiment, S3 specifically refers to: Based on the abnormal temperature fluctuation characteristics of the micro-deformation region of the iron core, and combined with the thermal expansion coefficient and Young's modulus of the iron core material, a sequence of iron core stress variation is generated. The core stress variation sequence is registered with the time-series synchronous data to construct a joint matrix of stress and magnetic field. Correlation analysis was performed on the joint matrix of stress and magnetic field to obtain the magnetic stress coupling coefficient sequence. Screen the coupling enhancement intervals in the magnetic stress coupling coefficient sequence and mark the corresponding iron core spatial locations as magnetic stress interaction enhancement regions; The mean, variance, and frequency domain amplitude of the coupling coefficient are extracted in the magnetic stress interaction enhancement region to output the characteristic data of the interaction between the core magnetic field and stress.

[0009] In a preferred embodiment, S4 specifically refers to: Based on the temperature anomaly fluctuation characteristics of the micro-deformation region of the iron core, the spatial coordinate position corresponding to the micro-deformation region of the iron core is determined, and the real-time temperature data sequence of the spatial coordinate position in the time-series synchronization data is extracted. A spatial neighborhood with a preset spatial radius is constructed, centered on the spatial coordinate position corresponding to the micro-deformation region of the iron core. Extract the temporal temperature difference of real-time temperature data sequences in the spatial neighborhood; The interference intensity of abnormal temperature changes in the micro-deformation region of the iron core on the overall annealing heat distribution uniformity of the iron core is determined based on the time-domain temperature difference, and the temperature disturbance intensity data of the micro-deformation region of the iron core is output.

[0010] In a preferred embodiment, S5 specifically refers to: The characteristic data of the interaction between the magnetic field and stress in the iron core are synchronized and aligned with the temperature disturbance intensity data of the micro-deformation region of the iron core according to the time axis to generate a coupled input matrix; Normalize the coupled input matrix and unify the data dimensions; A coupled anomaly evolution model of magnetic field-thermal field-stress field is established based on the coupled input matrix; The magnetic field-thermal field-stress field coupling index sequence is calculated based on the coupling anomaly evolution model, and the coupling anomaly interval is screened based on the preset coupling threshold to output the iron core annealing anomaly assessment data.

[0011] In a preferred embodiment, S6 specifically refers to: The core annealing anomaly assessment data was divided into multiple time windows, and the mean and variance of the core annealing anomaly assessment data were extracted within each time window. The deviation of the average and variance of the iron core annealing anomaly assessment data within each time window from the preset normal threshold is calculated to obtain the iron core annealing quality fluctuation risk index. The risk level of each time window in the iron core annealing process is determined based on the risk index of iron core annealing quality fluctuation. For time windows where the risk level exceeds the preset risk level threshold, mark them as periods of abnormal annealing quality, and generate intelligent early warning information for abnormal core annealing that includes the risk level, start and end time, and corresponding core spatial location.

[0012] The technical effects and advantages of the intelligent monitoring method for the annealing process of wound transformer cores of this invention are as follows: By synchronously acquiring and annotating the temperature and magnetic field strength fields during the annealing process of wound transformer cores, comprehensive online monitoring of the core annealing status is achieved. Based on time-series synchronous data, the micro-deformation regions and corresponding temperature anomaly regions of the core are accurately located, directly linking temperature anomaly fluctuation characteristics with micro-deformation behavior, thus improving the ability to identify latent defects. By analyzing the dynamic coupling effect of temperature anomaly fluctuations and stress changes on the magnetic field distribution, characteristic data reflecting the stress-magnetic field interaction are obtained, achieving a quantitative description of multi-field coupling effects. A quantitative assessment of the interference intensity of abnormal core temperature changes on the uniformity of the annealed core heat distribution provides accurate interference indicators for the uniformity of the annealed temperature field distribution. A three-dimensional coupled evolution model including magnetic field, thermal field, and stress field is established to predict and evaluate the coupling anomaly evolution process, generating core annealing anomaly assessment data. By assessing the quality fluctuation risk during the core annealing process, real-time early warning of annealing quality fluctuation risks can be provided, significantly improving the stability and reliability of the annealing process, ensuring the final magnetic performance consistency and product qualification rate of the transformer core. Attached Figure Description

[0013] Figure 1 This is a schematic diagram of an intelligent monitoring method for the annealing process of a wound transformer core according to the present invention. Detailed Implementation

[0014] The technical solutions of 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0015] Example Figure 1 This invention provides an intelligent monitoring method for the annealing process of wound transformer cores, which includes the following steps: S1: Collect temperature and magnetic field strength data generated by the core of the wound transformer during the annealing process, perform time-domain synchronization processing and anomaly data labeling, and generate time-series synchronized data of temperature and magnetic field. S2: Based on the time-series synchronization data of temperature and magnetic field, identify the micro-deformation region of the iron core and the corresponding temperature anomaly region, and generate the temperature anomaly fluctuation characteristics of the micro-deformation region of the iron core. S3: Based on the abnormal temperature fluctuation characteristics of the micro-deformation region of the iron core, analyze the dynamic coupling effect of stress change on the magnetic field distribution, and output the characteristic data of the interaction between the iron core magnetic field and stress. S4: Based on the abnormal temperature fluctuation characteristics of the micro-deformation region of the iron core, analyze the interference intensity of abnormal temperature changes on the uniformity of heat distribution during iron core annealing, and output the temperature disturbance intensity data. S5: Based on the characteristic data of the interaction between the magnetic field and stress in the iron core and the data of the temperature disturbance intensity of the iron core, a coupled anomaly evolution model of magnetic field-thermal field-stress field is established to generate iron core annealing anomaly assessment data. S6: Assess the quality fluctuation risk of the iron core annealing process based on the iron core annealing anomaly assessment data, and generate intelligent early warning information for iron core annealing anomalies.

[0016] S1: Collect temperature and magnetic field strength data generated during the annealing process of the wound transformer core, perform time-domain synchronization processing and anomaly data annotation, and generate time-series synchronized data of temperature and magnetic field, including: Real-time temperature data of the core of a wound transformer during the annealing process was collected. A wound transformer core is a core structure with a circular or rectangular cross-section, manufactured from continuous silicon steel sheets or strips using specialized winding equipment. It is primarily used for the magnetic circuit composition of transformers. Due to the mechanical stress incurred during the winding process, the wound transformer core requires high-temperature heat treatment during annealing to relieve stress and improve the material's magnetic properties. The annealing process involves the orderly heating, holding, and cooling of the wound transformer core in a high-temperature furnace. The accuracy and stability of temperature control directly determine the final magnetic properties of the core. Therefore, to monitor the dynamic temperature changes during annealing, temperature sensors are used to continuously collect real-time temperature data at various local spatial locations within the core in the annealing furnace. Multiple K-type or N-type thermocouple temperature sensors are arranged at different spatial locations around the wound transformer core, continuously collecting and recording the real-time temperature at different spatial locations at a frequency of 1 to 10 data points per second, forming a real-time temperature data set including timestamps and spatial location markers. For example, the real-time temperature data collected at a certain point in time might be: the real-time temperature at location coordinates (x1, y1, z1) is 550 degrees Celsius, the real-time temperature at location coordinates (x2, y2, z2) is 548 degrees Celsius, and the real-time temperature at location coordinates (x3, y3, z3) is 549 degrees Celsius.

[0017] Real-time magnetic field strength data of the core of a wound transformer during the annealing process were collected. To monitor the magnetic field changes during the core annealing process, multiple magnetic field strength sensors need to be deployed around the core of the wound transformer. Taking fluxgate magnetometers or Hall effect sensors as examples, these sensors measure and collect data on the magnetic field strength changes during the annealing process in real time. The magnetic field strength sensors are installed at multiple specific locations inside the annealing furnace, continuously collecting and recording the magnetic field strength at different locations, forming a real-time magnetic field strength data set with timestamps and spatial coordinate markers. For example, at a certain time point, the simultaneously obtained magnetic field strength data might be: 200 Gauss at coordinates (x1, y1, z1), 198 Gauss at coordinates (x2, y2, z2), and 199 Gauss at coordinates (x3, y3, z3).

[0018] Real-time temperature data and real-time magnetic field strength data are aligned according to a unified time axis and integrated synchronously into a unified time-series data sequence. Since temperature and magnetic field data originate from different sensors and may have different sampling start times and frequencies, they need to be aligned along a unified time axis. This can be achieved by using interpolation or linear interpolation to fill in missing data points, ensuring that both temperature and magnetic field strength data have corresponding values ​​at each time point. For example, temperature data might have a data point at 1 second, 2 seconds, and 3 seconds, while magnetic field strength data might be collected at 1.5 seconds, 2.5 seconds, and 3.5 seconds. Therefore, linear interpolation is needed to adjust the magnetic field data to the same 1-second, 2-second, and 3-second time points as the temperature data, thus obtaining a unified time-series data sequence. Using this method, a time-series data sequence can be obtained. For instance, the data combination obtained at 2 seconds might be: temperature 550 degrees Celsius and magnetic field strength 200 Gauss at coordinates (x1, y1, z1); temperature 548 degrees Celsius and magnetic field strength 198 Gauss at coordinates (x2, y2, z2), and so on. Through the above synchronous integration process, the time-series data sequence accurately reflects the real-time temperature and real-time magnetic field changes during the annealing process of the wound transformer core.

[0019] Abnormal data points that exceed the preset normal range in a unified time-series data sequence are marked as abnormal data to form time-series synchronized data of temperature and magnetic field; By using preset temperature and magnetic field strength threshold ranges, data exceeding reasonable physical or process settings within a unified time-series data sequence are automatically identified and marked as abnormal data. For example, the preset normal temperature range is 520 to 560 degrees Celsius during annealing, and the normal magnetic field strength range is 190 to 210 Gauss. If, at a certain moment and location, the temperature data is 570 degrees Celsius, it exceeds the preset normal temperature range, and therefore the corresponding data point is marked as abnormal data in the data sequence. Similarly, if, at a certain moment and location, the magnetic field strength data is 215 Gauss, it also exceeds the preset normal magnetic field strength range, and the corresponding data point is marked as abnormal data in the data sequence. For example, the time-series synchronization data sequence of temperature and magnetic field after anomaly data annotation processing may be as follows: at time point t=2 seconds, at location (x1, y1, z1), the temperature is 550 degrees Celsius and the magnetic field strength is 200 gauss, both of which are normal data; at time point t=3 seconds, at location (x2, y2, z2), the temperature is 570 degrees Celsius (anomaly data marker) and the magnetic field strength is 202 gauss.

[0020] S2: Based on the time-series synchronized data of temperature and magnetic field, identify the micro-deformation region of the iron core and the corresponding temperature anomaly region, and generate the temperature anomaly fluctuation characteristics of the micro-deformation region of the iron core, including: The magnetic field strength difference and corresponding temperature difference between adjacent sampling points are calculated based on time-series synchronous data to generate magnetic field gradient sequences and temperature gradient sequences. Time-series synchronized data refers to the temperature and magnetic field data generated through synchronized processing. It reflects the real-time temperature and magnetic field strength data of the wound transformer core at each time point and spatial location during the annealing process. To reflect the gradient characteristics of the dynamic changes in temperature and magnetic field during annealing, it is necessary to calculate the changes in magnetic field strength and temperature between adjacent sampling time points. For example, taking the position coordinates (x1, y1, z1) as an example, the magnetic field strength at two adjacent sampling points at the 1st and 2nd seconds are 200 Gauss and 202 Gauss, respectively. The difference in magnetic field strength is calculated to be 2 Gauss. Simultaneously, the real-time temperatures at the same two adjacent sampling points are 550 degrees Celsius and 552 degrees Celsius, respectively, resulting in a temperature difference of 2 degrees Celsius. By continuously executing the above calculation process, the magnetic field gradient sequence and temperature gradient sequence, composed of the magnetic field strength difference and temperature difference, are obtained, reflecting the dynamic changes of magnetic field and temperature over time.

[0021] The initial candidate region for micro-deformation of the iron core is determined in the magnetic field gradient sequence based on a preset magnetic field gradient threshold. The magnetic field gradient threshold is a pre-set threshold based on the changes in magnetic field strength and the magnetostrictive effect of the core during the annealing process of a wound transformer. The magnetostrictive effect indicates that when the core undergoes minute mechanical deformation, it is accompanied by a change in magnetic field strength. Therefore, by setting an appropriate magnetic field gradient threshold, such as 1.5 Gauss / second, the intervals in the magnetic field gradient sequence where the magnetic field gradient is greater than 1.5 Gauss / second are considered to have potentially undergone magnetic field changes, implying the possibility of micro-deformation in the core. For example, at position coordinates (x2, y2, z2), if the magnetic field gradient reaches 2 Gauss / second between the 2nd and 3rd second, exceeding the preset magnetic field gradient threshold, then the spatial location of the core at position coordinates (x2, y2, z2) can be marked as the initial candidate region for micro-deformation of the core, providing the initial spatial location of the micro-deformation.

[0022] Based on the preset temperature gradient threshold, candidate areas for temperature anomalies are determined in the temperature gradient sequence, and their spatial positions are matched with the initial candidate areas for micro-deformation of the iron core to determine the micro-deformation areas and temperature anomaly areas of the iron core that completely overlap in spatial position. The temperature gradient threshold is also pre-set based on empirical rules of the annealing process and the characteristics of the core material, for example, a temperature gradient threshold of 1.2 degrees Celsius per second. When the temperature gradient in the temperature gradient sequence exceeds 1.2 degrees Celsius per second, it indicates an abnormal temperature fluctuation at the corresponding spatial location. For example, at the location coordinates (x2, y2, z2), if the temperature gradient reaches 1.5 degrees Celsius per second from the 2nd to the 3rd second, exceeding the set temperature gradient threshold, then this location is marked as a candidate area for temperature anomalies. The initial candidate areas for core micro-deformation determined by the magnetic field gradient sequence and the candidate areas for temperature anomalies determined by the temperature gradient sequence are matched one by one in spatial location. If a location coordinate (e.g., (x2, y2, z2)) is found to exceed the corresponding threshold in both the magnetic field gradient and temperature gradient, and the spatial location is completely consistent, then the spatial location corresponding to the location coordinates (x2, y2, z2) is confirmed as both the core micro-deformation area and the temperature anomaly area. By using the criterion of complete spatial overlap, the core area where actual micro-deformation and temperature anomalies occur is accurately identified.

[0023] Calculate the time-domain mean, time-domain variance, and frequency-domain amplitude of temperature data within the temperature anomaly region to generate the temperature anomaly fluctuation characteristics of the core micro-deformation region. Within the defined micro-deformation region and temperature anomaly region of the iron core, statistical and frequency domain features of the real-time temperature data within these regions are extracted to quantify the abnormal fluctuation characteristics. Specifically, the time domain mean refers to the average value of all real-time temperature data within a time window; the time domain variance refers to the degree of fluctuation of the real-time temperature data relative to the average value; and the frequency domain amplitude is calculated by transforming the real-time temperature data sequence into frequency space using Fourier transform, and determining the amplitude corresponding to the characteristic frequency. For example, for the defined coordinates (x2, y2, z2) of the micro-deformation region of the iron core, the real-time temperature data collected within the time window from the 2nd to the 6th second are 552, 554, 553, 556, and 555 degrees Celsius, respectively. The calculated time domain mean of the temperature data within the time window is 554 degrees Celsius, the time domain variance is 2.0 degrees Celsius, and the frequency domain analysis reveals an amplitude of 1.5 degrees Celsius at the characteristic frequency of temperature fluctuation at 0.2 Hz. Through the above statistical and frequency domain analyses, the amplitude, stability, and variation pattern of local abnormal temperature fluctuations in the micro-deformation region of the iron core during annealing are reflected.

[0024] S3: Based on the abnormal temperature fluctuation characteristics of the micro-deformation region of the iron core, analyze the dynamic coupling effect of stress changes on the magnetic field distribution, and output the characteristic data of the interaction between the iron core magnetic field and stress, including: Based on the abnormal temperature fluctuation characteristics of the micro-deformation region of the iron core, and combined with the thermal expansion coefficient and Young's modulus of the iron core material, a sequence of iron core stress variation is generated. To determine the impact of abnormal temperature fluctuations on the actual mechanical properties of the transformer core, it is necessary to comprehensively consider the physical properties of the core material, specifically its coefficient of thermal expansion and Young's modulus. The coefficient of thermal expansion represents the degree of change in length of the core material under unit temperature variation; Young's modulus characterizes the rigidity of the core material within its elastic deformation range. By combining the abnormal temperature fluctuation characteristics of the core's micro-deformation region with the core material's coefficient of thermal expansion and Young's modulus, the dynamic changes in stress under abnormal temperature fluctuations can be calculated, generating a core stress change sequence. For example, if the core material is silicon steel sheet with a coefficient of thermal expansion of 11.5 × 10⁻⁶ degrees Celsius and a Young's modulus of 195 GPa, at the coordinate position of the micro-deformation region (e.g., position coordinates (x2, y2, z2)), when the abnormal temperature fluctuation characteristics show a rapid fluctuation from 550 degrees Celsius to 560 degrees Celsius, the magnitude of the thermal stress change at the corresponding position can be calculated using the coefficient of thermal expansion and Young's modulus, based on a temperature difference of 10 degrees Celsius. Using the formula Δσ=E·α·ΔT, where Δσ represents the stress change, E is Young's modulus, α is the coefficient of thermal expansion, and ΔT is the temperature change, substituting the above parameters, we obtain the stress change Δσ=195GPa×11.5×10^-6 degrees Celsius×10 degrees Celsius=22.425MPa. Following the same calculation method, temperature anomaly fluctuation data at multiple time points were continuously processed to obtain stress change values ​​sequentially, forming a continuous core stress change sequence. The core stress change sequence reflects the dynamic evolution trend and amplitude of stress in local areas of the core caused by abnormal temperature changes during annealing.

[0025] The core stress variation sequence is registered with the time-series synchronous data to construct a joint matrix of stress and magnetic field. To accurately study the dynamic correlation between core stress variation and magnetic field strength, it is necessary to match the core stress variation sequence and the magnetic field strength data sequence in the time dimension, a process known as time axis registration. For example, the sampling interval for the core stress variation sequence is 1 second, and the sampling interval for the magnetic field strength data is also 1 second. By aligning the core stress variation sequence and the magnetic field strength data sequence at each time point using seconds as the unit, for instance, at the location coordinates (x2, y2, z2), the core stress variation values ​​from the 1st to the 5th second are 22.425 MPa, 23.500 MPa, 21.300 MPa, 24.000 MPa, and 23.100 MPa, respectively; at the corresponding time points from the 1st to the 5th second, the magnetic field strength data are 200 Gauss, 201 Gauss, 202 Gauss, 200 Gauss, and 199 Gauss, respectively. By aligning the stress and magnetic field data using time points as coordinates, a joint stress-magnetic field matrix is ​​constructed. The stress-magnetic field joint matrix reflects the numerical information of stress and magnetic field intensity in local areas of the iron core at each time point.

[0026] Correlation analysis was performed on the joint matrix of stress and magnetic field to obtain the magnetic stress coupling coefficient sequence. Correlation analysis employs numerical statistical methods, such as calculating the covariance between magnetic field strength and stress data within each time window and determining the correlation coefficient to generate the magnetic-stress coupling coefficient. The magnetic-stress coupling coefficient typically ranges from 0 to 1, with values ​​closer to 1 indicating a stronger dynamic correlation between magnetic field strength and stress. For example, using the Pearson correlation coefficient method, a magnetic-stress coupling coefficient of 0.85 is obtained for the time window from 1 to 5 seconds, indicating a positive correlation between the magnetic field and stress. By continuously performing this calculation within multiple consecutive time windows, a sequence of magnetic-stress coupling coefficients can be obtained; each value in the sequence reflects the dynamic correlation between magnetic field strength and stress within the corresponding time window.

[0027] Screen the coupling enhancement intervals in the magnetic stress coupling coefficient sequence and mark the corresponding iron core spatial locations as magnetic stress interaction enhancement regions; The coupling enhancement interval refers to the continuous time period in the magnetic stress coupling coefficient sequence where the magnetic stress coupling coefficient exceeds a pre-set magnetic stress coupling coefficient threshold. For example, if the magnetic stress coupling coefficient threshold is set to 0.75, and the magnetic stress coupling coefficient sequence is 0.80, 0.82, and 0.85 respectively from the 1st to the 3rd second, all exceeding the magnetic stress coupling coefficient threshold, then the continuous time period is marked as the coupling enhancement interval, and the corresponding spatial location, such as the location coordinates (x2, y2, z2), is marked as the magnetic stress interaction enhancement region, thus identifying the core spatial region where the interaction between the magnetic field and stress is significantly enhanced.

[0028] In the magnetic stress interaction enhancement region, the mean value of the coupling coefficient, the variance of the coupling coefficient and the frequency domain amplitude are extracted to output the characteristic data of the interaction between the core magnetic field and stress. The characteristic data were obtained through statistical and frequency domain characteristic analysis. For example, in the magnetic stress interaction enhancement zone at position coordinates (x2, y2, z2), the coupling coefficients were 0.80, 0.82, and 0.85 in the first to third seconds, respectively. The calculated mean of the coupling coefficients was 0.823, and the variance was 0.00042. Simultaneously, through Fourier spectrum analysis, the frequency domain amplitude of the magnetic stress coupling coefficient at a frequency of 0.5 Hz was calculated to be 0.015. Together, these data constitute the characteristic data of the interaction between the iron core magnetic field and stress.

[0029] S4: Based on the abnormal temperature fluctuation characteristics of the micro-deformation region of the iron core, analyze the interference intensity of abnormal temperature changes on the uniformity of heat distribution during iron core annealing, and output temperature disturbance intensity data, including: Based on the temperature anomaly fluctuation characteristics of the micro-deformation region of the iron core, the spatial coordinate position corresponding to the micro-deformation region of the iron core is determined, and the real-time temperature data sequence of the spatial coordinate position in the time-series synchronization data is extracted. Temperature anomaly fluctuation characteristics include the time-domain mean, variance, and frequency-domain amplitude of temperature data within a specific time interval, reflecting the amplitude, frequency, and stability of local temperature changes. Analysis of these temperature anomaly fluctuation characteristics can confirm the location coordinates of the micro-deformation region. For example, in the annealing process of a wound transformer core, the location coordinates of the micro-deformation region are determined to be (x2, y2, z2). To analyze the real-time temperature changes in the micro-deformation region, a complete real-time temperature data sequence corresponding to the location coordinates is extracted based on time-series synchronized data of temperature and magnetic field. This real-time temperature data sequence records the real-time temperature changes at each moment during the entire annealing process of the core, demonstrating the entire process of temperature change in the micro-deformation region. For example, the real-time temperature data for position coordinates (x2, y2, z2) from the 1st second to the 10th second are 550 degrees Celsius, 551 degrees Celsius, 552 degrees Celsius, 560 degrees Celsius, 563 degrees Celsius, 559 degrees Celsius, 558 degrees Celsius, 556 degrees Celsius, 555 degrees Celsius and 553 degrees Celsius, respectively, reflecting the dynamic process of rapid local temperature rise and subsequent gradual drop during annealing in the micro-deformation region.

[0030] A spatial neighborhood with a preset spatial radius is constructed, centered on the spatial coordinate position corresponding to the micro-deformation region of the iron core. To quantify the impact of abnormal temperature changes in the micro-deformation region of the iron core on the uniformity of the overall annealing heat distribution, an appropriate spatial range is defined around the micro-deformation region to obtain sufficient spatial comparison temperature data; this defined spatial range is called the spatial neighborhood. The spatial neighborhood is a three-dimensional or two-dimensional region centered on the spatial coordinates of the micro-deformation region (e.g., (x2, y2, z2)) and extending outwards in all directions of the iron core with a predetermined spatial radius. The size of the predetermined spatial radius is determined based on the iron core dimensions and actual process conditions, for example, from 10 mm to 50 mm. For example, if the predetermined spatial radius is set to 20 mm, then the spatial neighborhood is defined with the coordinates (x2, y2, z2) as the center and a radius of 20 mm; the spatial neighborhood may contain several different spatial coordinate positions, such as (x2+5, y2, z2), (x2-10, y2, z2), (x2, y2+15, z2), etc., forming a set of spatial positions. By setting up spatial neighborhoods, real-time temperature data of the micro-deformation region and its neighboring regions can be compared and analyzed to assess the degree of interference of abnormal temperature changes in the micro-deformation region on the overall temperature uniformity of the core.

[0031] Extract the temporal temperature difference of real-time temperature data sequences in the spatial neighborhood; Within a spatial neighborhood, real-time temperature data sequences at all spatial locations are compared to calculate the temperature differences within the time domain. These temperature differences reflect the non-uniform distribution of the temperature field near the micro-deformation region. For example, if the real-time temperature data at the spatial neighborhood coordinates (x2+5, y2, z2) are 552°C, 553°C, 554°C, 555°C, and 556°C, and the real-time temperature data at (x2-10, y2, z2) are 551°C, 552°C, 553°C, 554°C, and 555°C, then the real-time temperature data in the micro-deformation region (x2, y2, z2) from the 3rd to the 7th second are 560°C, 563°C, 559°C, 558°C, and 556°C. By comparing real-time temperature data from different spatial locations, a significant difference can be observed between the temperature of the micro-deformation region and the surrounding spatial locations. For example, at the 4th second, the temperature of the micro-deformation region is 563 degrees Celsius, while the temperatures of the surrounding spatial locations are all below 556 degrees Celsius, a difference of 7 degrees Celsius. This reflects the temporal temperature difference characteristics within the spatial neighborhood. Calculating these temporal temperature difference characteristics reflects the degree of interference caused by temperature anomalies in the micro-deformation region on the thermal field uniformity of the spatial neighborhood.

[0032] Based on the time-domain temperature difference, determine the interference intensity of the abnormal temperature change in the micro-deformation region of the iron core on the overall annealing heat distribution uniformity of the iron core, and output the temperature disturbance intensity data of the micro-deformation region of the iron core. The disturbance intensity is calculated by statistically analyzing the difference between the real-time temperature data of the micro-deformation region and the real-time temperature data of other locations in the spatial neighborhood over time. The mean method or the root mean square error method is used to process the temperature difference to quantify the degree of temperature disturbance caused by the micro-deformation region to the spatial neighborhood. For example, during the temperature fluctuation period from the 3rd to the 7th second, the average temperature difference between the micro-deformation region and the neighboring location is calculated per second, resulting in temperature differences of 6°C, 7°C, 5°C, 4°C, and 3°C. By averaging these temperature differences, the average disturbance intensity is obtained as 5°C, reflecting the average degree of temperature disturbance caused by the micro-deformation region to the neighborhood. Simultaneously, the standard deviation of the temperature difference is calculated to reflect the stability of the disturbance; for example, the standard deviation of the disturbance intensity is 1.414°C. Furthermore, frequency domain analysis methods are used to calculate the spectral amplitude of the temperature difference to assess the periodicity and dynamic trend of the disturbance. For example, frequency domain analysis determined that the main disturbance frequency was 0.25 Hz and the frequency domain amplitude was 1.2 degrees Celsius. Together, they formed the temperature disturbance intensity data of the micro-deformation region of the iron core, reflecting the degree of interference and change characteristics of the temperature anomaly in the micro-deformation region on the overall annealing heat distribution uniformity of the iron core.

[0033] S5: Based on the characteristic data of the interaction between the core magnetic field and stress, and the core temperature disturbance intensity data, a coupled anomaly evolution model of the magnetic field-thermal field-stress field is established to generate core annealing anomaly assessment data, including: The characteristic data of the interaction between the magnetic field and stress in the iron core are synchronized and aligned with the temperature disturbance intensity data of the micro-deformation region of the iron core according to the time axis to generate a coupled input matrix; The characteristic data of the interaction between the core magnetic field and stress include the mean, variance, and frequency domain amplitude of the coupling coefficient within the enhanced magnetic-stress interaction region. These characteristic data represent the changing trends and magnitudes of the dynamic interaction between the magnetic field and mechanical stress in the micro-deformation region of the core. The temperature disturbance intensity data in the micro-deformation region of the core is represented by the mean, standard deviation, and frequency domain amplitude of the disturbance intensity, reflecting the degree and variation of the interference of local temperature anomalies in the micro-deformation region on the overall uniformity of the core temperature distribution during annealing. Since the characteristic data of the interaction between the core magnetic field and stress and the temperature disturbance intensity data originate from different analysis paths, they need to be synchronized along a unified time axis to demonstrate the synchronous changing relationship between the three physical fields: the core magnetic field, mechanical stress, and temperature disturbance. For example, the time sampling points of the characteristic data of the interaction between the magnetic field and stress in the iron core and the temperature disturbance intensity data of the micro-deformation region of the iron core can be uniformly adjusted to be sampled once every 1 second. If the sampling points of the characteristic data of the interaction between the magnetic field and stress were originally at integer second positions such as 1 second, 2 seconds, 3 seconds, etc., and the sampling points of the temperature disturbance intensity data were at 1.5 seconds, 2.5 seconds, the sampling points of the temperature disturbance intensity data can be adjusted to integer second positions using linear interpolation. For example, at the 2-second position, if the mean value of the magnetic stress coupling coefficient in the magnetic stress interaction characteristic data is 0.82, the variance is 0.0004, and the frequency domain amplitude is 0.015, and the temperature disturbance intensity data after interpolation adjustment has an average disturbance intensity of 5 degrees Celsius, a standard deviation of 1.4 degrees Celsius, and a frequency domain amplitude of 1.2 degrees Celsius, then together at the 2-second position on the time axis, they constitute a set of synchronous characteristic data of magnetic field-stress-temperature. Following the above method, the data at each sampling time point are uniformly aligned to form a multi-dimensional data sequence based on the time axis, with each time point containing multiple parameters such as magnetic field, stress, and temperature. This sequence is defined as the coupling input matrix. The coupling input matrix reflects the synchronous dynamic relationship between the three physical fields—magnetic field, stress, and temperature—at each moment during the iron core annealing process.

[0034] Normalize the coupled input matrix and unify the data dimensions; The coupled input matrix contains parameters from three different physical fields: magnetic field, mechanical stress, and temperature. These parameters differ in their numerical dimensions and ranges. For example, magnetic field strength is measured in Gaussians, typically ranging from tens to hundreds; mechanical stress is measured in megapascals (MPa), usually ranging from several to tens of MPa; and temperature disturbance intensity is measured in degrees Celsius, ranging from a few degrees Celsius. To avoid the influence of these differences in numerical magnitudes, all data in the coupled input matrix is ​​normalized, unifying the numerical ranges of each data dimension to a common standard range, such as between 0 and 1. A minimum-maximum normalization method is used, where the minimum and maximum values ​​of each parameter sequence are calculated using the formula: Normalized data value = (Original data value - Minimum value) / (Maximum value - Minimum value), thus transforming all data into the interval between 0 and 1. Using the above methods, each physical field parameter is uniformly normalized, and the data of different physical quantities in the final coupled input matrix are all within a uniform dimensionless numerical range, eliminating the influence of differences in dimensions and amplitudes between different data.

[0035] A coupled anomaly evolution model of magnetic field-thermal field-stress field is established based on the coupled input matrix; The coupled anomaly evolution model comprises a set of coupled equations and a time-series parameter matrix. The coupled equations are mathematical equations established based on physical laws and statistical analysis methods, reflecting the quantitative relationship of the dynamic interaction between magnetic field, stress, and temperature. For example, linear or nonlinear multiple regression analysis is used to establish the mathematical equations, with changes in magnetic field strength, stress, and temperature disturbances as independent variables, and the coupled anomaly state defined as the dependent variable. For instance: Coupled anomaly state index = Coefficient A × Normalized magnetic field data + Coefficient B × Normalized stress data + Coefficient C × Normalized temperature data, where coefficients A, B, and C are determined based on historical data regression analysis. The time-series parameter matrix is ​​a data matrix composed of synchronously normalized data at each time point. Each row represents a time point, and each column represents the values ​​of magnetic field, stress, and temperature, respectively. Together, the coupled equations and the time-series parameter matrix form the coupled anomaly evolution model of the magnetic field-thermal field-stress field, enabling the quantification and calculation of the complex coupling relationship between the magnetic, thermal, and stress fields.

[0036] The magnetic field-thermal field-stress field coupling index sequence is calculated based on the coupling anomaly evolution model, and the coupling anomaly interval is screened based on the preset coupling threshold to output the iron core annealing anomaly assessment data. Based on the coupling anomaly evolution model, data from each time point in the time series parameter matrix are substituted into the coupling equations for calculation, generating the magnetic field-thermal field-stress field coupling index for each moment. Coupling indices from multiple consecutive moments form a coupling index sequence. For example, the coupling index ranges from 0 to 1; values ​​closer to 1 indicate a stronger coupling anomaly between the three physical fields. A preset coupling threshold (e.g., 0.8) is then used to filter the coupling indices for consecutive moments, identifying consecutive time periods exceeding the preset threshold as coupling anomaly intervals. For example, if the coupling indices from the 5th to the 8th second are 0.81, 0.85, 0.83, and 0.80 respectively, then this time period is identified as a coupling anomaly interval. The core annealing anomaly assessment data consists of the magnetic field-thermal field-stress field coupling indices within these coupling anomaly intervals.

[0037] S6: Based on the core annealing anomaly assessment data, evaluate the quality fluctuation risk of the core annealing process and generate intelligent early warning information for core annealing anomalies, including: The core annealing anomaly assessment data was divided into multiple time windows, and the mean and variance of the core annealing anomaly assessment data were extracted within each time window. The time window length is preset according to the actual core annealing process requirements and can be 5 seconds, 10 seconds, or longer. For example, if a standard time window length of 10 seconds is determined, the annealing process time, for example, from second 0 to second 100, is divided into independent time windows of 10 seconds each, generating windows from second 0 to second 10, second 10 to second 20, second 20 to third 30, and so on, forming multiple consecutive time windows of equal length. The mean and variance of all coupled indicators within each time window are calculated. The mean is used to quantify the overall level of coupled indicators within each time window, and the variance is used to assess the fluctuation range of the data within each window. For example, within a time window from the 10th to the 20th second, if the core annealing anomaly assessment data within that time window are 0.81, 0.83, 0.85, 0.84, 0.82, 0.86, 0.87, 0.85, 0.84, and 0.83 respectively, then the calculated average is 0.84, and the calculated variance is 0.0004. This process is repeated for each time window to obtain the average and variance of the core annealing anomaly assessment data for multiple time windows.

[0038] The deviation of the average and variance of the iron core annealing anomaly assessment data within each time window from the preset normal threshold is calculated to obtain the iron core annealing quality fluctuation risk index. The preset normal threshold is a standard determined based on historical data of the coupling index sequence of the magnetic field-thermal field-stress field of the iron core under normal annealing process conditions. It reflects the coupling characteristics between the magnetic field, thermal field, and stress field under normal annealing process conditions. For example, after analyzing a large amount of historical process data, the average threshold for the magnetic field-thermal field-stress field coupling index under normal conditions in the iron core annealing process was determined to be 0.75, and the variance threshold was 0.0002. The iron core annealing quality fluctuation risk index is quantitatively described by comparing the average value and variance calculated within each time window with the normal threshold, thus describing the severity of the abnormal assessment data deviating from the normal state within each time window. For example, for the time window from the 10th to the 20th second, the calculated average value is 0.84, and the variance is 0.0004. The deviation is calculated as follows: the deviation of the average value is (0.84-0.75) / 0.75×100%=12%, and the deviation of the variance is (0.0004-0.0002) / 0.0002×100%=100%. Using a weighted method, for example, with a 60% weight for the deviation of the average value and a 40% weight for the deviation of the variance, the final core annealing quality fluctuation risk index for each time window is determined to be: 12% × 60% + 100% × 40% = 47.2%. The same method is used to obtain the core annealing quality fluctuation risk index for all time windows, reflecting a quantitative assessment of the quality fluctuation risk of the core annealing process within each time window.

[0039] The risk level of each time window in the iron core annealing process is determined based on the risk index of iron core annealing quality fluctuation. The risk level is divided into three levels: low risk, medium risk, and high risk. Each level corresponds to a range of risk indicators for core annealing quality fluctuations. For example, a risk indicator between 0% and 20% is defined as low risk, indicating that the core annealing process is within a basically normal range; between 20% and 40% is medium risk, indicating that the core annealing process has a certain degree of abnormal risk and requires attention; and above 40% is high risk, indicating that the core annealing process deviates significantly from the normal range and there is a significant risk of quality abnormalities. Based on the calculated risk indicator for core annealing quality fluctuations in each time window, for example, if the indicator is 47.2% in the 10th to 20th second window, then the risk level for this time window is determined to be high risk according to the risk level classification rules. If the indicator is 15% in the 20th to 30th second time window, then the risk level is determined to be low risk. Through this method, the risk level of the core annealing process within each time window is determined.

[0040] For time windows where the risk level exceeds the preset risk level threshold, mark them as periods of abnormal annealing quality, and generate intelligent early warning information for abnormal core annealing that includes the risk level, start and end time, and corresponding core spatial location. A preset risk level threshold is used to determine abnormal states requiring intervention or handling. For example, a risk level of medium or higher (including medium and high risk levels) is set as the preset risk level threshold. All time windows where the risk level exceeds the threshold are marked and defined as annealing quality abnormality periods. For example, a high-risk level within a time window from the 10th to the 20th second is marked, defining this time window as an annealing quality abnormality period. Simultaneously, intelligent early warning information for core annealing abnormalities during this period is generated, including the abnormal risk level (high risk level), the start and end times of the abnormality (from the 10th to the 20th second), and the corresponding spatial coordinates of the abnormal core location (e.g., (x2, y2, z2)). This intelligent early warning information is output to the monitoring terminal or equipment management platform in a visual graphical or list format, promptly prompting operators to manually intervene or inspect the abnormal core annealing area to ensure the stability and reliability of the core annealing process quality.

[0041] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters and thresholds in the formulas are set by those skilled in the art according to the actual situation.

[0042] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0043] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0044] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0045] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0046] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0047] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0048] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0049] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0050] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for intelligent monitoring of the annealing process of a wound transformer core, characterized in that, Includes the following steps: S1: Collect temperature and magnetic field strength data generated by the core of the wound transformer during the annealing process, perform time-domain synchronization processing and anomaly data labeling, and generate time-series synchronized data of temperature and magnetic field. S2: Based on the time-series synchronization data of temperature and magnetic field, identify the micro-deformation region of the iron core and the corresponding temperature anomaly region, and generate the temperature anomaly fluctuation characteristics of the micro-deformation region of the iron core. S3: Based on the abnormal temperature fluctuation characteristics of the micro-deformation region of the iron core, analyze the dynamic coupling effect of stress change on the magnetic field distribution, and output the characteristic data of the interaction between the iron core magnetic field and stress. S4: Based on the abnormal temperature fluctuation characteristics of the micro-deformation region of the iron core, analyze the interference intensity of abnormal temperature changes on the uniformity of heat distribution during iron core annealing, and output the temperature disturbance intensity data. S5: Based on the characteristic data of the interaction between the magnetic field and stress in the iron core and the data of the temperature disturbance intensity of the iron core, a coupled anomaly evolution model of magnetic field-thermal field-stress field is established to generate iron core annealing anomaly assessment data. S6: Assess the quality fluctuation risk of the iron core annealing process based on the iron core annealing anomaly assessment data, and generate intelligent early warning information for iron core annealing anomalies.

2. The intelligent monitoring method for the annealing process of a wound transformer core according to claim 1, characterized in that, S1, specifically: Real-time temperature data of the core of a wound transformer during the annealing process was collected. Real-time magnetic field strength data of the core of a wound transformer during the annealing process were collected. Real-time temperature data and real-time magnetic field strength data are aligned according to a unified time axis and integrated synchronously into a unified time-series data sequence. Abnormal data points that exceed the preset normal range in a unified time-series data sequence are marked as abnormal data to form time-series synchronized data of temperature and magnetic field.

3. The intelligent monitoring method for the annealing process of a wound transformer core according to claim 2, characterized in that, S2, specifically: The magnetic field strength difference and corresponding temperature difference between adjacent sampling points are calculated based on time-series synchronous data to generate magnetic field gradient sequences and temperature gradient sequences. The initial candidate region for micro-deformation of the iron core is determined in the magnetic field gradient sequence based on a preset magnetic field gradient threshold. Based on the preset temperature gradient threshold, candidate areas for temperature anomalies are determined in the temperature gradient sequence, and their spatial positions are matched with the initial candidate areas for micro-deformation of the iron core to determine the micro-deformation areas and temperature anomaly areas of the iron core that completely overlap in spatial position. The time-domain mean, time-domain variance, and frequency-domain amplitude of temperature data within the temperature anomaly region are calculated to generate the temperature anomaly fluctuation characteristics of the micro-deformation region of the iron core.

4. The intelligent monitoring method for the annealing process of a wound transformer core according to claim 3, characterized in that, S3, specifically: Based on the abnormal temperature fluctuation characteristics of the micro-deformation region of the iron core, and combined with the thermal expansion coefficient and Young's modulus of the iron core material, a sequence of iron core stress variation is generated. The core stress variation sequence is registered with the time-series synchronous data to construct a joint matrix of stress and magnetic field. Correlation analysis was performed on the joint matrix of stress and magnetic field to obtain the magnetic stress coupling coefficient sequence. Screen the coupling enhancement intervals in the magnetic stress coupling coefficient sequence and mark the corresponding iron core spatial locations as magnetic stress interaction enhancement regions; The mean, variance, and frequency domain amplitude of the coupling coefficient are extracted in the magnetic stress interaction enhancement region to output the characteristic data of the interaction between the core magnetic field and stress.

5. The intelligent monitoring method for the annealing process of a wound transformer core according to claim 4, characterized in that, S4, specifically: Based on the temperature anomaly fluctuation characteristics of the micro-deformation region of the iron core, the spatial coordinate position corresponding to the micro-deformation region of the iron core is determined, and the real-time temperature data sequence of the spatial coordinate position in the time-series synchronization data is extracted. A spatial neighborhood with a preset spatial radius is constructed, centered on the spatial coordinate position corresponding to the micro-deformation region of the iron core. Extract the temporal temperature difference of real-time temperature data sequences in the spatial neighborhood; The interference intensity of abnormal temperature changes in the micro-deformation region of the iron core on the overall annealing heat distribution uniformity of the iron core is determined based on the time-domain temperature difference, and the temperature disturbance intensity data of the micro-deformation region of the iron core is output.

6. The intelligent monitoring method for the annealing process of a wound transformer core according to claim 5, characterized in that, S5, specifically: The characteristic data of the interaction between the magnetic field and stress in the iron core are synchronized and aligned with the temperature disturbance intensity data of the micro-deformation region of the iron core according to the time axis to generate a coupled input matrix; Normalize the coupled input matrix and unify the data dimensions; A coupled anomaly evolution model of magnetic field-thermal field-stress field is established based on the coupled input matrix; The magnetic field-thermal field-stress field coupling index sequence is calculated based on the coupling anomaly evolution model, and the coupling anomaly interval is screened based on the preset coupling threshold to output the iron core annealing anomaly assessment data.

7. The intelligent monitoring method for the annealing process of a wound transformer core according to claim 6, characterized in that, S6, specifically: The core annealing anomaly assessment data was divided into multiple time windows, and the mean and variance of the core annealing anomaly assessment data were extracted within each time window. The deviation of the average and variance of the iron core annealing anomaly assessment data within each time window from the preset normal threshold is calculated to obtain the iron core annealing quality fluctuation risk index. The risk level of each time window in the iron core annealing process is determined based on the risk index of iron core annealing quality fluctuation. For time windows where the risk level exceeds the preset risk level threshold, mark them as periods of abnormal annealing quality, and generate intelligent early warning information for abnormal core annealing that includes the risk level, start and end time, and corresponding core spatial location.

Images