Generator stator iron loss test method and system

Abstract

The invention discloses a generator stator iron loss test method and system, and the method comprises the following steps: 1, dynamically calculating a target voltage interval through a self-adaptive control algorithm at an excitation stage based on preset iron core structure parameters and target magnetic flux density, and executing the voltage adjustment in stages; step 2, in a test stage, synchronously acquiring multi-dimensional state information of the iron core; and 3, based on the multi-dimensional state information, calculating a loss value and a temperature rise value of the iron core through fusion of a finite element model and Kalman filtering, and generating a comprehensive evaluation result of the iron core stacking quality according to the loss value, the temperature rise value and a change trend thereof. Through the implementation of the invention, accurate voltage regulation in stages is realized, and overshoot is effectively suppressed; the detection precision of the iron core loss and the temperature rise is obviously improved; and the temperature rise trend can be accurately predicted and early warning can be triggered in the early stage of the test, so that the test safety and the evaluation reliability are greatly improved.

Description

Technical Field

[0001] This invention relates to the field of iron loss testing technology, and in particular to a method and system for testing iron loss of generator stator. Background Technology

[0002] As the core component of a generator, the quality of its stator core assembly directly affects the generator's operating efficiency, temperature rise characteristics, and service life. Iron loss testing is a crucial method for verifying the quality of stator core assembly. By measuring the core's losses and temperature rise under an alternating magnetic field, it assesses whether the core has defects such as inter-laminar short circuits or insulation damage. Traditional generator stator iron loss testing methods have the following main technical limitations:

[0003] Insufficient excitation control precision and the use of fixed voltage regulation rates in conventional test methods lack consideration for the adaptability of core structural parameters, resulting in overshoot during voltage regulation and affecting the accuracy of test data. In particular, when approaching the target voltage, excessively fast regulation rates can easily cause voltage fluctuations and make it difficult to stabilize within the target range.

[0004] The data acquisition dimension is limited, and existing technologies mostly use a limited number of measuring points for temperature monitoring, which cannot fully reflect the spatial temperature distribution characteristics of the iron core. At the same time, the acquisition of electrical parameters and temperature data often has a time synchronization problem, making it difficult to establish an accurate correlation analysis model.

[0005] The assessment method is significantly outdated. Traditional assessments mainly rely on the final data after the test, lacking the ability to provide early warning of abnormal conditions during the test. When the temperature rise exceeds the standard, the core may have already suffered irreversible thermal damage, increasing maintenance costs and safety risks.

[0006] The safety protection measures are inadequate. When the existing test system detects anomalies, it relies heavily on manual intervention, resulting in a slow response speed and a lack of automated safety linkage mechanisms, making it impossible to take protective measures quickly in emergency situations.

[0007] The system has low integration, the test equipment is scattered, the control and data acquisition systems operate independently, and there is a lack of a unified software platform for centralized management and automated control, which increases the complexity of operation and is prone to human error.

[0008] In summary, there is an urgent need for a generator stator iron loss testing method and system that can achieve precise excitation control, multi-dimensional synchronous data acquisition, intelligent early warning, and automated safety linkage, in order to solve the problems of insufficient control accuracy, delayed evaluation, and imperfect safety assurance in existing technologies. Summary of the Invention

[0009] To address the shortcomings of existing technologies, this invention provides a generator stator iron loss testing method and system, aiming to solve the aforementioned problems.

[0010] On the one hand, this application provides a method for testing the stator iron loss of a generator, including the following steps:

[0011] Step 1: During the excitation stage, based on the preset core structure parameters and target magnetic flux density, the target voltage range is dynamically calculated through an adaptive control algorithm, and staged voltage regulation is performed.

[0012] Step 2: During the testing phase, multi-dimensional status information of the iron core is collected simultaneously;

[0013] Step 3: Based on the multi-dimensional state information, calculate the loss value and temperature rise value of the iron core by fusing the finite element model and Kalman filter, and generate a comprehensive evaluation result of the iron core stacking quality based on the loss value, temperature rise value and their changing trends.

[0014] The comprehensive assessment includes early warning: acquiring the temperature rise rate at the initial stage of the test, calculating the predicted temperature rise of the core at the end of the test using a quadratic polynomial prediction model based on the core's thermal characteristics, and issuing an early warning signal before the test is completed if the predicted temperature rise exceeds a safety threshold. Adaptive control algorithms are used to achieve precise phased voltage adjustment, effectively suppressing overshoot; multi-dimensional state information is collected synchronously, combined with finite element model and Kalman filter fusion calculations, significantly improving the detection accuracy of core loss and temperature rise; and an innovative quadratic polynomial prediction model based on thermal characteristics is introduced, accurately predicting the temperature rise trend and triggering early warnings in the early stages of the test, realizing a shift from post-event judgment to pre-event warning, greatly improving test safety and assessment reliability.

[0015] In one implementation, step 1 includes core structure parameters such as core thickness, stacking factor, and silicon steel sheet grade, and the staged voltage regulation includes:

[0016] The voltage is rapidly increased at the first rate until the difference between the current output voltage and the target voltage value is less than a preset threshold.

[0017] Switch to the second rate for fine boosting until the current output voltage enters the target voltage range;

[0018] During the voltage stabilization process, an anti-integral saturation control algorithm based on Kalman filtering is adopted to suppress voltage overshoot.

[0019] As one implementation method, in the Kalman filter-based anti-integral saturation control algorithm, the state equation of the Kalman filter is:

[0020] X(k)=A·X(k-1)+B·U(k)+W(k),

[0021] Y(k) = C·X(k) + V(k),

[0022] Where X(k) is the state vector, A is the system matrix, B is the control input matrix, U(k) is the control input, W(k) is the process noise, Y(k) is the observation vector, C is the observation matrix, and V(k) is the observation noise. The eigenvalues ​​of the system matrix A satisfy |λ|<1 to ensure system stability.

[0023] In one implementation, step 2 includes multi-dimensional state information including spatially distributed temperature field data and time-series electrical parameter data.

[0024] As one implementation method, the synchronous acquisition of spatially distributed temperature field data is obtained by a temperature monitoring instrument that supports at least 64 channels, and the values ​​of multiple physical temperature measurement channels are virtually calculated to obtain derived parameters characterizing the overall or local thermal state of the iron core.

[0025] As one implementation method, in the quadratic polynomial prediction model based on the thermal properties of the iron core, the predicted relationship between temperature rise and time is as follows:

[0026] Predicted temperature rise = a × t² + b × t + c

[0027] Where t is the running time after the start of the test, and a, b, and c are coefficients determined based on the thermal characteristic parameters of the iron core.

[0028] Parameters a, b, and c are determined through the following steps:

[0029] Collect thermal characteristic parameters of the iron core at different temperatures;

[0030] Based on the properties of the iron core material, the heat distribution model of the iron core at different temperatures was calculated by finite element analysis.

[0031] Regression analysis was used to determine the relationship between parameters a, b, and c and the properties of the core material.

[0032] As one implementation method, in step 3, after the early warning signal is triggered, the following safety linkage measures are automatically executed:

[0033] Reduce the excitation current to 30-40% of the rated current;

[0034] Start the cooling system and increase the cooling water flow rate to 120% of the rated flow rate;

[0035] At the same time, the early warning information is sent to the remote monitoring platform via the GPRS network.

[0036] As one implementation method, the adaptive control algorithm is achieved through the following steps:

[0037] Obtain the core structure parameters;

[0038] Retrieve the excitation voltage adjustment parameters corresponding to the core structure parameters from the pre-stored database;

[0039] Based on the query results, the target voltage range and adjustment rate are dynamically calculated.

[0040] In one implementation method, the method is implemented through a unified software platform, which performs the following technical processes during execution:

[0041] In response to the user selecting manual mode in the graphical interface, it receives and executes discrete control commands for the voltage regulator and contactor, while continuously recording and updating the data display interface.

[0042] In response to the user's selection of automatic mode, the preset test logic sequence is invoked to automatically execute the entire process from closing, voltage boosting, voltage stabilization timing to voltage reduction and circuit breaking;

[0043] During the execution of the automatic mode, the platform monitors the electrical parameter data in real time. If the voltage value deviates from the preset target voltage range, a voltage regulation command is automatically generated and sent to the voltage regulator to maintain voltage stability.

[0044] Select the export mode to export data in a general format while maintaining the data collection process, according to the selected time period.

[0045] Furthermore,

[0046] On the other hand, this application provides a generator stator iron loss testing system, comprising:

[0047] The dynamic excitation control module is used to perform staged voltage regulation;

[0048] A multimodal data synchronous acquisition module is used to acquire the multi-dimensional state information;

[0049] The intelligent analysis and evaluation module is used to perform fusion calculations and generate comprehensive evaluation results;

[0050] The security linkage execution module is used to execute security linkage measures.

[0051] The substantial effects of this invention:

[0052] 1. In this invention, an adaptive control algorithm based on core structure parameters is used to dynamically calculate the target voltage range and perform staged voltage regulation, effectively suppressing voltage overshoot. The Kalman filter-based anti-integral saturation control algorithm ensures voltage stability during the voltage stabilization process, improving voltage control accuracy and significantly enhancing the reliability of test data. Furthermore, by synchronously acquiring spatially distributed temperature field data and time-series electrical parameters, combined with the fusion calculation method of finite element model and Kalman filter, accurate calculation of core loss and temperature rise values ​​is achieved. Based on the quadratic polynomial prediction model of core thermal characteristics, the final temperature rise can be predicted in the initial stage of the test, identifying potential overheating risks in advance and significantly improving the timeliness of quality assessment.

[0053] 2. In this invention, upon triggering an early warning signal, the system automatically executes coordinated measures such as reducing excitation current, increasing cooling flow, and remote alarm, forming a complete safety closed loop from detection and early warning to handling. Compared to traditional manual intervention methods, the response time is greatly shortened, significantly improving the safety of the testing process.

[0054] 3. This invention utilizes a unified software platform to enable flexible switching between manual, automatic, and export modes, supporting fully automated management from test preparation to data export. The system possesses self-learning and adaptive capabilities, allowing it to optimize control parameters based on different core characteristics, thus lowering the technical barrier for operators and improving testing efficiency.

[0055] 4. In this invention, by establishing a quantitative relationship between predicted temperature rise and core material properties, and combining finite element analysis and regression analysis, the accuracy of the temperature rise prediction model is improved, providing a scientific basis for early judgment of core stacking quality. Attached Figure Description

[0056] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0057] Figure 1 This is the logic block diagram of the iron loss test in Example 1.

[0058] Figure 2 This is a block diagram of the multi-dimensional data acquisition logic in Example 1.

[0059] Figure 3 This is a logic block diagram of the adaptive control algorithm in Example 1.

[0060] Figure 4This is a schematic diagram of the temperature recording window in Example 3.

[0061] Figure 5 This is a schematic diagram of the wiring terminals of the temperature monitoring instrument in Example 3. Detailed Implementation

[0062] To facilitate understanding of the present invention, a more detailed description is provided below with reference to the accompanying drawings and specific embodiments. It should be noted that when an element is described as being "fixed to" another element, it can be directly on the other element, or one or more intermediate elements may exist between them. When an element is described as being "connected to" another element, it can be directly connected to the other element, or one or more intermediate elements may exist between them. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this specification are for illustrative purposes only.

[0063] Unless otherwise defined, all technical and scientific terms used in this specification have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used in this specification is for the purpose of describing particular embodiments only and is not intended to limit the invention. The term "and / or" as used in this specification includes any and all combinations of one or more of the associated listed items.

[0064] Example 1:

[0065] Reference Figure 1 As shown, a method for testing the stator iron loss of a generator includes the following steps:

[0066] Step 1: During the excitation stage, based on the preset core structure parameters and target magnetic flux density, the target voltage range is dynamically calculated through an adaptive control algorithm, and staged voltage regulation is performed.

[0067] Step 2: During the testing phase, multi-dimensional status information of the iron core is collected simultaneously;

[0068] Step 3: Based on multi-dimensional state information, calculate the loss value and temperature rise value of the iron core by fusing the finite element model and Kalman filter, and generate a comprehensive evaluation result of the iron core stacking quality based on the loss value, temperature rise value and their changing trends.

[0069] The stacking quality assessment is based on the correlation between iron loss and temperature rise. When the relative deviation of the iron loss from the design standard value exceeds ±4.1% and the temperature rise rate exceeds 0.8℃ / minute, the stacking quality is deemed unqualified. The comprehensive assessment includes early warning: acquiring the temperature rise rate in the initial stage of the test (first 5 minutes), and calculating the predicted temperature rise of the iron core at the end of the test using a quadratic polynomial prediction model based on the core's thermal characteristics. If the predicted temperature rise exceeds the safety threshold (75℃), an early warning signal is issued before the test is completed. Adaptive control algorithms achieve precise phased voltage adjustment, effectively suppressing overshoot; synchronous acquisition of multi-dimensional state information, combined with finite element model and Kalman filter fusion calculation, significantly improves the detection accuracy of core loss and temperature rise; innovatively introducing a quadratic polynomial prediction model based on thermal characteristics, it can accurately predict the temperature rise trend and trigger early warnings in the early stages of the test, realizing a shift from post-event judgment to pre-event warning, greatly improving test safety and assessment reliability.

[0070] In one implementation, step 1 includes core structure parameters such as core thickness, stacking factor, and silicon steel sheet grade, and the phased voltage regulation includes:

[0071] Rapidly boost the voltage at a first rate (2-3% of rated voltage / second) until the difference between the current output voltage and the target voltage value is less than a preset threshold (0.5%).

[0072] Switch to the second rate (0.5% of rated voltage / second) for fine boosting until the current output voltage enters the target voltage range;

[0073] During the voltage stabilization process, an anti-integral saturation control algorithm based on Kalman filtering is adopted to suppress voltage overshoot.

[0074] As one implementation method, in the Kalman filter-based anti-integral saturation control algorithm, the state equation of the Kalman filter is:

[0075] X(k)=A·X(k-1)+B·U(k)+W(k),

[0076] Y(k) = C·X(k) + V(k),

[0077] Where X(k) is the state vector, A is the system matrix, B is the control input matrix, U(k) is the control input, W(k) is the process noise, Y(k) is the observation vector, C is the observation matrix, and V(k) is the observation noise. The eigenvalues ​​of the system matrix A satisfy |λ|<1 to ensure system stability.

[0078] As one implementation method, refer to Figure 2As shown, in step 2, the multi-dimensional state information includes spatially distributed temperature field data and time-series electrical parameter data. By synchronously collecting the multi-dimensional state information of the iron core, a completely new sensing system is constructed. This system enables subsequent intelligent analysis (predictive model), precise control (adaptive algorithm), and proactive safety (early warning), thereby comprehensively solving the fundamental problems of "one-sided diagnosis, delayed early warning, and coarse control" existing in the current technology.

[0079] As one implementation method, the spatially distributed temperature field data is acquired by a temperature monitoring instrument that supports at least 64 channels. The values ​​of multiple physical temperature measurement channels are then used for virtual calculations to obtain derived parameters characterizing the overall or local thermal state of the iron core. The 64-channel temperature monitoring instrument is arranged in a ring array on the surface of the iron core, with a spacing of 20-30 mm between each temperature sensor, a temperature acquisition frequency of 5-10 Hz, and a measurement accuracy of ±0.5℃.

[0080] As one implementation method, in the quadratic polynomial prediction model based on the thermal properties of the iron core, the predicted relationship between temperature rise and time is as follows:

[0081] Predicted temperature rise = a × t² + b × t + c

[0082] Where t is the running time (in minutes) after the start of the test, and the value of t ranges from 0 to 180 minutes; a, b, and c are coefficients determined based on the thermal characteristic parameters of the iron core (including specific heat capacity, thermal conductivity, and coefficient of thermal expansion).

[0083] Parameters a, b, and c are determined through the following steps:

[0084] Collect thermal characteristic parameters of the iron core at different temperatures (including specific heat capacity, thermal conductivity, and coefficient of thermal expansion).

[0085] Based on the properties of the iron core material, the heat distribution model of the iron core at different temperatures was calculated by finite element analysis.

[0086] Based on the specific heat capacity c_p, thermal conductivity λ, and coefficient of thermal expansion α of the core material, the coefficients a, b, and c are determined by the following relationships: a = f(c_p, λ, α) b = g(c_p, λ, α) c = h(c_p, λ, α).

[0087] As one implementation method, in step 3, after the early warning signal is triggered, the following safety linkage measures are automatically executed:

[0088] Reduce the excitation current to 30-40% of the rated current;

[0089] Start the cooling system and increase the cooling water flow rate to 120% of the rated flow rate;

[0090] At the same time, the early warning information is sent to the remote monitoring platform via the GPRS network.

[0091] As one implementation method, refer to Figure 3 As shown, the adaptive control algorithm is implemented through the following steps:

[0092] Obtain the core structure parameters (core thickness, stacking factor, silicon steel sheet grade);

[0093] Retrieve the excitation voltage adjustment parameters corresponding to the core structure parameters from the pre-stored database;

[0094] Based on the query results, the target voltage range and adjustment rate are dynamically calculated.

[0095] As one implementation method, the approach is implemented through a unified software platform that performs the following technical processes during execution:

[0096] In response to the user selecting manual mode in the graphical interface, it receives and executes discrete control commands for the voltage regulator and contactor, while continuously recording and updating the data display interface.

[0097] In response to the user's selection of automatic mode, the preset test logic sequence is invoked to automatically execute the entire process from closing, voltage boosting, voltage stabilization timing to voltage reduction and circuit breaking;

[0098] During the execution of automatic mode, the platform monitors electrical parameter data in real time. If the voltage value deviates from the preset target voltage range, it automatically generates a voltage regulation command and sends it to the voltage regulator to maintain voltage stability.

[0099] Select the export mode to export data in a general format while maintaining the data collection process, according to the selected time period.

[0100] As one implementation method, during execution, the unified software platform incorporates the following further detailed internal logic:

[0101] The pre-stored database contains various common silicon steel sheet grades (such as DW470, B35A300, etc.) and their corresponding optimal excitation parameter mapping tables. When the operator inputs a core thickness of 350mm, a stacking factor of 0.96, and a silicon steel sheet grade of B35A300, the algorithm automatically queries the mapping table to determine the initial rapid voltage boost rate as 2.5% of rated voltage / second, the fine voltage boost rate as 0.5% of rated voltage / second, and the target voltage range as [98.5% to 101.5% of rated voltage];

[0102] During the voltage stabilization process, the system uses the output voltage and load current as the state vector X(k) and estimates the optimal state of the system in real time through a Kalman filter, effectively filtering out interference from grid fluctuations and measurement noise. When the estimated value shows that the output voltage tends to exceed the target range, the anti-integral saturation PID controller will make fine adjustments in advance, thereby achieving control accuracy and stability far exceeding that of traditional PID controllers.

[0103] In addition to being displayed individually, the raw data collected by 64 temperature sensors are used to calculate three key derived parameters in real time through a virtual computing channel: 1) the average temperature on the back of the core (the average of the top 16 channels), 2) the maximum temperature difference in the tooth section (the difference between the maximum and minimum values ​​of the 32 channels in the middle section), and 3) the temperature growth rate in the hot spot area (a linear fit is performed on the data from the 5 channels with the highest temperatures to calculate the slope). These parameters provide a more direct basis for comprehensive evaluation.

[0104] Example 2:

[0105] This embodiment is basically the same as Embodiment 1, except that it provides a specific mechanism for automatically generating voltage regulation commands and sending them to the voltage regulator to maintain voltage stability:

[0106] For high-precision data sensing and state estimation, the system synchronously acquires the raw voltage U(t) and current I(t) signals from the output side of the voltage regulator at high frequency (50-100Hz). These raw data usually contain grid harmonics and measurement noise. To address this, the system introduces a Kalman filter, an optimal estimation algorithm, for preprocessing. By establishing the system state equation (X(k) = A·X(k-1) + B·U(k) + W(k)) and the observation equation (Y(k) = C·X(k) + V(k)), the filter can recursively calculate the optimal estimates of the true voltage and current values, U_hat and I_hat, and estimate their changing trends in real time, providing a clean and reliable input basis for control decisions.

[0107] Based on rule-based multi-mode intelligent decision-making, the control system compares the filtered voltage estimate U_hat with the preset target voltage range [U_target_min, U_target_max], and dynamically activates different control strategies according to the deviation, demonstrating its intelligent characteristics.

[0108] Large Deviation Fast Correction Mode: When the voltage deviates significantly from the target range (U_hat < U_target_min or U_hat > ​​U_target_max), the system uses high proportional gain (P control) or feedforward control to pull the voltage back to the vicinity of the target range at the maximum rate, ensuring a fast response.

[0109] Small Deviation Fine Voltage Regulation Mode: Once the voltage enters the target range, the system switches to a PID control algorithm with anti-windup. This algorithm is the core innovation; it dynamically calculates ΔV(k) = Kp * e(k) + Ki *∑e(j) + Kd * [e(k) - e(k-1)]. Its anti-saturation mechanism monitors whether the output command exceeds the physical limits of the actuator, automatically pausing the accumulation of the integral term when saturation occurs. This completely avoids voltage overshoot and continuous oscillation caused by integrator "saturation," achieving smooth and precise voltage regulation.

[0110] With adaptive optimization and instruction execution, the system further enhances its adaptability, dynamically fine-tuning the PID parameters (Kp, Ki, Kd) based on real-time calculated load characteristics (such as equivalent impedance) to adapt to different iron core test conditions and improve system robustness. Finally, the generated control commands are encapsulated into standard industrial protocol (such as Modbus TCP) data frames and sent to the embedded controller of the voltage regulator to drive servo motors or thyristors and other actuators for precise position adjustment.

[0111] With closed-loop feedback and continuous optimization, after the voltage regulator is activated, the system immediately starts a new cycle of "sensing-decision-execution". This high-frequency closed-loop process constitutes an autonomous system that can sense the system status in real time, make intelligent decisions on control strategies, and execute them accurately.

[0112] This automatic voltage regulation mechanism improves sensing accuracy through Kalman filtering, achieves intelligent overshoot-free control through multi-mode and anti-saturation PID algorithms, and enables precise execution through adaptive technology and industrial communication. Together, these measures ensure rapid, stable, and high-precision control of the test voltage, greatly improving the automation level and data reliability of iron loss testing.

[0113] Example 3:

[0114] This embodiment is basically the same as Embodiment 1, except that it provides a generator stator iron loss test system, including:

[0115] The dynamic excitation control module is used to perform staged voltage regulation, including:

[0116] Rapidly boost the voltage at the first rate (2% of rated voltage / second) until the difference between the current output voltage and the target voltage value is less than the preset threshold (0.5%).

[0117] Switch to the second rate (0.5% of rated voltage / second) for fine boosting until the current output voltage enters the target voltage range;

[0118] A Kalman filter-based anti-integral saturation control algorithm is used to suppress voltage overshoot.

[0119] The multimodal data synchronous acquisition module is used to acquire multi-dimensional state information, including:

[0120] Temperature field data was acquired and recorded using a 64-channel temperature monitoring instrument arranged in a ring array on the iron core surface (temperature sensor spacing 25mm, acquisition frequency 8Hz, accuracy ±0.5℃). Figure 4 , 5 As shown, the current temperature data of the temperature monitoring instrument is recorded, and a record is made every 15 minutes for 90 minutes. The temperature difference between the end time and the initial time, as well as the maximum and minimum temperature difference, are also recorded.

[0121] Synchronously collect electrical parameter data (voltage, current, power factor, harmonic content);

[0122] The intelligent analysis and evaluation module performs fusion calculations and generates comprehensive evaluation results, including:

[0123] The loss and temperature rise of the iron core are calculated by fusing the finite element model with Kalman filtering.

[0124] Early warning is provided based on a quadratic polynomial prediction model of the core thermal characteristics (predicted temperature rise = a × t² + b × t + c), where t is the running time after the start of the test (in minutes), and a, b, and c are coefficients determined according to the core thermal characteristic parameters.

[0125] The security linkage execution module is used to execute security linkage measures, including:

[0126] Reduce the excitation current to 35% of the rated current;

[0127] Start the cooling system and increase the cooling water flow rate to 120% of the rated flow rate;

[0128] The early warning information is sent to the remote monitoring platform via the GPRS network.

[0129] As one implementation method, the hardware architecture and data flow of the generator stator iron loss test system work in tandem, specifically as follows:

[0130] The system adopts a hardware-triggered synchronization method, where the temperature monitoring instrument and the data acquisition card (used for electrical parameters) share a synchronization clock signal issued by the industrial control computer, ensuring that each data frame has the same timestamp. This solves the problem of asynchronous "time shift" between electrical parameters and temperature data in traditional systems, laying a solid foundation for subsequent fusion calculations.

[0131] After receiving the early warning signal from the intelligent analysis and evaluation module, this module executes safety measures through a dual path: hard-wired connection (directly connected to the voltage regulator control circuit) and soft commands (sent via the software platform). For example, a command to reduce the excitation current is simultaneously sent to the voltage regulator via the analog output card, and a portion of the excitation circuit is cut off via the PLC, ensuring that the protection action can still be reliably executed in the event of a single component failure.

[0132] Example 4:

[0133] This embodiment is basically the same as Embodiment 1, except that it provides a customized test scheme applicable to large hydro-generators. This embodiment mainly demonstrates the adaptability of the present invention when facing large, special-structure generators.

[0134] Technical background: The stator core of large hydro-generators has a large diameter and high thermal inertia. Traditional testing methods are extremely time-consuming and cannot fully capture its complex heat distribution.

[0135] Implementation method adjustment:

[0136] The enhanced temperature monitoring network uses two cascaded 64-channel temperature monitoring instruments to form a distributed temperature monitoring network with 128 measuring points. The measuring points are arranged according to the principle of "radial layering and circumferential partitioning", focusing on monitoring key parts such as ventilation channels and toothed pressure plates.

[0137] The segmented adaptive excitation divides the entire iron core into four quadrants logically. The adaptive control algorithm dynamically fine-tunes the voltage of the excitation winding in each quadrant based on the imbalance of the initial temperature rise rate in each quadrant, so as to achieve uniform heating of the iron core and avoid local overheating.

[0138] To enhance the prediction model, which addresses the slow thermal response of large iron cores, the coefficients (a, b, c) of the quadratic polynomial prediction model are no longer fixed. Instead, they are dynamically corrected and rolled over at two time points, 10 minutes and 20 minutes after the start of the test, using the latest collected data. This makes the prediction results increasingly accurate as the test progresses.

[0139] This embodiment ensures precise and uniform experimental heating and reliable early warning even on large and complex structures by strengthening the monitoring network and control strategy, minimizing experimental risks.

[0140] Example 5:

[0141] This embodiment is basically the same as Embodiment 1, except that it provides a predictive maintenance and offline analysis scheme based on digital twins. This embodiment mainly demonstrates the extended application of the present invention in the in-depth mining and full lifecycle management of experimental data.

[0142] Technical background: Users are not only concerned with whether a single test is qualified, but also hope to establish a health record of the iron core to achieve predictive maintenance.

[0143] The implementation methods include:

[0144] Before the experiment begins, a high-fidelity finite element thermal-magnetic coupling simulation model is built on the cloud or local server based on the precise three-dimensional model and material parameters of the iron core, serving as the "digital twin" of the iron core.

[0145] Online data-driven and model calibration: During the experiment, multi-dimensional state information collected in real time (especially temperature rise and loss data in the first 30 minutes) is synchronously input into the digital twin; using this data, the boundary conditions and non-uniform material parameters of the simulation model are inverted and corrected through data assimilation technology, so that the behavior of the digital twin is highly consistent with the physical entity;

[0146] Predictive analysis and report generation: After the test, the system not only provides a report on the test, but also uses the corrected digital twin to perform "what if" analysis; for example, it can simulate the performance of the iron core under higher ambient temperatures or insulation aging after long-term operation, providing users with life prediction and risk assessment reports.

[0147] This embodiment upgrades a one-time quality inspection into a health management tool for the entire life cycle of the generator stator core, providing users with unprecedented in-depth insights and decision support, and greatly enhancing the added value of the product.

[0148] It should be noted that while the preferred embodiments of the present invention are provided in the specification and accompanying drawings, the present invention can be implemented in many different forms and is not limited to the embodiments described herein. These embodiments are not intended to impose additional limitations on the content of the present invention; their purpose is to provide a more thorough and comprehensive understanding of the disclosure of the present invention. Furthermore, the above-described technical features can be combined with each other to form various embodiments not listed above, all of which are considered to be within the scope of the present invention specification. Moreover, those skilled in the art can make improvements or modifications based on the above description, and all such improvements and modifications should fall within the protection scope of the appended claims.

Claims

1. A method for testing stator iron loss in a generator, characterized in that, Includes the following steps: Step 1: During the excitation stage, based on the preset core structure parameters and target magnetic flux density, the target voltage range is dynamically calculated through an adaptive control algorithm, and staged voltage regulation is performed. Step 2: During the testing phase, multi-dimensional status information of the iron core is collected simultaneously; Step 3: Based on the multi-dimensional state information, calculate the loss value and temperature rise value of the iron core by fusing the finite element model and Kalman filter, and generate a comprehensive evaluation result of the iron core stacking quality based on the loss value, temperature rise value and their changing trends. The comprehensive assessment includes early warning: obtaining the temperature rise rate at the initial stage of the test, calculating the predicted temperature rise of the iron core at the end of the test using a quadratic polynomial prediction model based on the thermal characteristics of the iron core, and issuing an early warning signal before the test is completed if the predicted temperature rise exceeds the safety threshold.

2. The generator stator iron loss test method according to claim 1, characterized in that, In step 1, the core structure parameters include core thickness, stacking factor, and silicon steel sheet grade, and the phased voltage regulation includes: The voltage is rapidly increased at the first rate until the difference between the current output voltage and the target voltage value is less than a preset threshold. Switch to the second rate for fine boosting until the current output voltage enters the target voltage range; During the voltage stabilization process, an anti-integral saturation control algorithm based on Kalman filtering is adopted to suppress voltage overshoot.

3. The generator stator iron loss test method according to claim 1, characterized in that, In the Kalman filter-based anti-integral saturation control algorithm, the state equation of the Kalman filter is: X(k)=A·X(k-1)+B·U(k)+W(k), Y(k) = C·X(k) + V(k), Where X(k) is the state vector, A is the system matrix, B is the control input matrix, U(k) is the control input, W(k) is the process noise, Y(k) is the observation vector, C is the observation matrix, and V(k) is the observation noise. The eigenvalues ​​of the system matrix A satisfy |λ|<1 to ensure system stability.

4. The generator stator iron loss test method according to claim 1, characterized in that, In step 2, the multi-dimensional state information includes spatially distributed temperature field data and time-series electrical parameter data.

5. The generator stator iron loss test method according to claim 4, characterized in that, The synchronously acquired spatially distributed temperature field data is obtained through a temperature monitoring instrument that supports at least 64 channels, and the values ​​of multiple physical temperature measurement channels are virtually calculated to obtain derived parameters characterizing the overall or local thermal state of the iron core.

6. The generator stator iron loss test method according to claim 1, characterized in that, In the quadratic polynomial prediction model based on the thermal properties of the iron core, the predicted relationship between temperature rise and time is as follows: Predicted temperature rise = a × t² + b × t + c Where t is the running time after the start of the test, and a, b, and c are coefficients determined based on the thermal characteristic parameters of the iron core. Parameters a, b, and c are determined through the following steps: Collect thermal characteristic parameters of the iron core at different temperatures; Based on the properties of the iron core material, the heat distribution model of the iron core at different temperatures was calculated by finite element analysis. Regression analysis was used to determine the relationship between parameters a, b, and c and the properties of the core material.

7. The generator stator iron loss test method according to claim 1, characterized in that, In step 3, after the early warning signal is triggered, the following safety linkage measures are automatically executed: Reduce the excitation current to 30-40% of the rated current; Start the cooling system and increase the cooling water flow rate to 120% of the rated flow rate; At the same time, the early warning information is sent to the remote monitoring platform via the GPRS network.

8. The generator stator iron loss test method according to claim 1, characterized in that, The adaptive control algorithm is implemented through the following steps: Obtain the core structure parameters; Retrieve the excitation voltage adjustment parameters corresponding to the core structure parameters from the pre-stored database; Based on the query results, the target voltage range and adjustment rate are dynamically calculated.

9. The generator stator iron loss test method according to claim 1, characterized in that, The method is implemented through a unified software platform, which performs the following technical processes during execution: In response to the user selecting manual mode in the graphical interface, it receives and executes discrete control commands for the voltage regulator and contactor, while continuously recording and updating the data display interface. In response to the user's selection of automatic mode, the preset test logic sequence is invoked to automatically execute the entire process from closing, voltage boosting, voltage stabilization timing to voltage reduction and circuit breaking; During the execution of the automatic mode, the platform monitors the electrical parameter data in real time. If the voltage value deviates from the preset target voltage range, a voltage regulation command is automatically generated and sent to the voltage regulator to maintain voltage stability. Select the export mode to export data in a general format while maintaining the data collection process, according to the selected time period.

10. A generator stator iron loss testing system, used to implement the test method as described in any one of claims 1-9, characterized in that, include: The dynamic excitation control module is used to perform staged voltage regulation; A multimodal data synchronous acquisition module is used to acquire the multi-dimensional state information; The intelligent analysis and evaluation module is used to perform fusion calculations and generate comprehensive evaluation results; The security linkage execution module is used to execute security linkage measures.

Images