Cooling optimization method and system of CLCC converter transformer, medium and equipment

Abstract

The invention relates to a heat dissipation optimization method, system, medium and equipment for a CLCC converter transformer, and the method comprises the steps: carrying out the continuous infrared thermal imaging of the surface of a radiator of the CLCC converter transformer, obtaining the temperature distribution data of spatial resolution, and synchronously collecting the real-time load data, the top oil temperature and the environment temperature of the CLCC converter transformer; normalizing load data to a standard load reference, and constructing a load-temperature normalized characteristic curve in combination with the temperature distribution data; based on the load-temperature normalized characteristic curve, establishing a heat exchange balance model representing a mapping relation between the load change and the heat dissipation demand; constructing a load prediction model based on a historical load and an environment temperature, and predicting an oil temperature change trend in a future preset time window in combination with the heat exchange balance model; and dynamically adjusting the switching number and the start-stop time sequence of the radiator groups in advance according to the predicted thermal load change and the current running state of the radiator so as to maintain the top oil temperature within a set target range.

Description

Optimization methods, systems, media, and equipment for heat dissipation of CLCC converter transformers Technical Field

[0001] This invention relates to the field of power system converter transformer technology, and in particular to a heat dissipation optimization method, system, medium and equipment for a CLCC converter transformer. Background Technology

[0002] The Controllable Line Commutated Converter (CLCC) introduces a fully controlled IGBT auxiliary branch into the traditional six-pulse bridge structure, effectively solving the commutation failure problem through a forced commutation mechanism. However, when the high-frequency voltage pulses generated during the CLCC commutation process propagate in the converter transformer windings, they can induce multi-conductor transmission line effects due to distributed parameter coupling, leading to problems such as local overheating of the windings and increased core vibration, posing a threat to the long-term stable operation of the transformer.

[0003] As a core component of converter stations, the heat dissipation performance of converter transformers directly impacts equipment lifespan and system reliability. Traditional heat dissipation management often relies on fixed-group radiator switching or simple temperature control strategies, which are ill-suited to the rapid load fluctuations and uneven heat distribution characteristic of CLCC (Converter Capacitor) operation. Existing technologies, control methods based on oil temperature feedback suffer from response lag, while load forecasting-based heat dissipation strategies lack high-precision temperature distribution data. Furthermore, harmonic and eddy current losses generated during the mixed commutation process of CLCC converter transformers further complicate heat dissipation design. In recent years, researchers have attempted to optimize temperature control performance by improving radiator structure or introducing intelligent algorithms, but these efforts have not fundamentally solved the problems of oil temperature fluctuations and localized overheating under dynamic loads. Especially with the widespread application of CLCC converter valves, developing an intelligent heat dissipation management method capable of real-time temperature monitoring combined with load forecasting has become an urgent need to improve the operational safety and economy of converter transformers. This invention presents an innovative solution within this technological context.

[0004] The information disclosed in the background section is only for enhancing the understanding of the background of this invention, and therefore may contain information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0005] This invention provides a method, system, medium, and equipment for optimizing the heat dissipation of a CLCC converter transformer, which can achieve constant oil temperature control and dynamic optimization of heat dissipation resources.

[0006] A heat dissipation optimization method for a CLCC converter transformer includes: performing continuous infrared thermal imaging on the surface of the CLCC converter transformer radiator to obtain spatially resolved temperature distribution data; simultaneously collecting real-time load data, top oil temperature, and ambient temperature of the CLCC converter transformer; normalizing the load data to a standard load reference and constructing a load-temperature normalized characteristic curve based on the temperature distribution data; establishing a heat transfer balance model characterizing the mapping relationship between load changes and heat dissipation requirements based on the load-temperature normalized characteristic curve; constructing a load prediction model based on historical load and ambient temperature, and predicting the oil temperature change trend within a preset time window based on the heat transfer balance model; and dynamically adjusting the number of radiator groups and their start-up / shutdown sequence in advance according to the predicted heat load changes and the current radiator operating status to maintain the top oil temperature within a set target range.

[0007] In the heat dissipation optimization method for a CLCC converter transformer, the normalization process includes: normalizing the original load data to the [0,1] interval according to the rated capacity, and normalizing the temperature distribution data according to the upper limit of safe operation, so that the load and temperature establish a monotonically increasing correlation curve under the same dimensionless scale.

[0008] In the aforementioned heat dissipation optimization method for a CLCC converter transformer, the heat transfer balance model establishes a load-heat dissipation relationship function based on the principle of constant transformer oil temperature, and its expression is: in, This represents the total heat loss of the transformer. The heat transfer coefficient, Due to the temperature difference between the oil and the environment, To maximize heat dissipation area, the oil temperature change trend is predicted based on real-time load data, and the number of radiator groups switched on and off is dynamically adjusted.

[0009] In the heat dissipation optimization method for a CLCC converter transformer, the load prediction model uses time series analysis or machine learning algorithms to predict the load value for the next 5–15 minutes based on historical load and ambient temperature data from the past 30 minutes to 2 hours, and adjusts the radiator group in advance by 0.5–2 control cycles accordingly.

[0010] In the heat dissipation optimization method for a CLCC converter transformer, the switching of the radiator groups follows the following: when the predicted load increases and the oil temperature change trend exceeds the set threshold, the number of radiator groups is increased in advance; when the predicted load decreases and the oil temperature change trend is lower than the target value, the shutdown of some radiators is delayed to avoid frequent start-stop; the switching action is controlled by zone based on the temperature distribution uniformity evaluation results, and priority is given to strengthening the operation of radiators corresponding to hot spots.

[0011] In the heat dissipation optimization method for a CLCC converter transformer, the load-temperature normalization curve and temperature prediction model are periodically corrected online based on actual operating data to adapt to model deviations caused by transformer aging, oil quality changes, or environmental condition drift.

[0012] In the aforementioned heat dissipation optimization method for CLCC converter transformers, the prediction formula of the load forecasting model is as follows: ,in, To predict load, For historical load sequences, For ambient temperature, It is a time variable.

[0013] The system for optimizing the heat dissipation of CLCC converter transformers includes a multispectral temperature monitoring platform. This platform performs continuous infrared thermal imaging on the surface of the CLCC converter transformer's heatsink to acquire spatially resolved temperature distribution data. It simultaneously collects real-time load data, top oil temperature, and ambient temperature of the CLCC converter transformer. The multispectral temperature monitoring platform includes: an infrared imaging system comprising an infrared multispectral analyzer, an optical lens, and a data acquisition card to output a heatsink surface temperature matrix; an operating parameter acquisition system comprising a load monitoring device, an oil temperature probe, and an ambient temperature sensor; and a control and data processing system comprising an embedded controller, a communication module, and a host computer for data fusion and modeling. The system includes: a calculation and control command issuance unit; a normalization unit that normalizes the load data to a standard load reference and constructs a load-temperature normalized characteristic curve based on the temperature distribution data; a modeling unit that establishes a heat exchange balance model representing the mapping relationship between load changes and heat dissipation requirements based on the load-temperature normalized characteristic curve, constructs a load prediction model based on historical load and ambient temperature, and predicts the oil temperature change trend within a preset time window based on the heat exchange balance model; and an optimization module that dynamically adjusts the number of radiator groups and their start-stop sequence in advance according to the predicted heat load changes and the current radiator operating status to maintain the top oil temperature within the set target range.

[0014] A computer storage medium including computer instructions that, when run on a computer, cause the computer to perform the method.

[0015] An electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method when executing the program.

[0016] Compared with the prior art, the present invention has the following advantages: The present invention normalizes the load data to a standard load curve and dynamically adjusts the radiator group switching strategy through a heat exchange balance model to achieve constant oil temperature control. Attached Figure Description

[0017] Various other advantages and benefits of the present invention will become apparent to those skilled in the art upon reading the detailed description of the preferred embodiments below. The accompanying drawings are for illustrative purposes only and are not intended to limit the invention. It is obvious that the drawings described below are merely some embodiments of the invention, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort. Furthermore, the same reference numerals denote the same parts throughout the drawings.

[0018] In the accompanying drawings: Figure 1 is a flowchart of the present invention; Figure 2 is a structural diagram of the heat dissipation management system of the present invention; Figure 3 is a schematic diagram of the infrared temperature monitoring principle of the present invention; Figure 4 is a sequence diagram of load and temperature under the operating conditions of the CLCC converter transformer of the present invention; Figure 5 is a load-temperature normalized curve diagram of the present invention; Figure 6 is a logic diagram of radiator group switching of the present invention; Figure 7 is a typical load curve diagram of the CLCC converter transformer; Figure 8 is a deep neural network architecture of the present invention; Figure 9 is a temperature prediction model diagram of an embodiment of the present invention.

[0019] The present invention will be further explained below with reference to the accompanying drawings and embodiments. Detailed Implementation

[0020] Specific embodiments of the present invention will now be described in more detail with reference to Figures 1 through 9. While specific embodiments of the invention are shown in the figures, it should be understood that the invention may be implemented in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this invention will be thorough and complete, and will fully convey the scope of the invention to those skilled in the art.

[0021] It should be noted that certain terms are used in the specification and claims to refer to specific components. Those skilled in the art will understand that different terms may be used to refer to the same component. This specification and claims do not distinguish components based on differences in terminology, but rather on differences in function. The terms "comprising" or "including" used throughout the specification and claims are open-ended and should be interpreted as "comprising but not limited to." The following descriptions are preferred embodiments for carrying out the invention; however, these descriptions are for the purpose of understanding the general principles of the specification and are not intended to limit the scope of the invention. The scope of protection of this invention is determined by the appended claims.

[0022] To facilitate understanding of the embodiments of the present invention, further explanations and descriptions will be provided below with reference to the accompanying drawings and specific embodiments. The accompanying drawings do not constitute a limitation on the embodiments of the present invention.

[0023] As shown in Figures 1 to 8, the heat dissipation optimization method for CLCC converter transformers includes the following steps: Continuous infrared thermal imaging of the CLCC converter transformer radiator surface is performed to obtain spatially resolved temperature distribution data; real-time load data, top oil temperature, and ambient temperature of the CLCC converter transformer are simultaneously collected; the load data is normalized, and combined with the temperature distribution data, a load-temperature normalized characteristic curve is constructed; a heat transfer balance model characterizing the mapping relationship between load changes and heat dissipation requirements is established based on the load-temperature normalized characteristic curve; a load prediction model is constructed based on historical load and ambient temperature, and the oil temperature change trend within a future preset time window is predicted in conjunction with the heat transfer balance model; based on the real-time load prediction results and ambient temperature data, the oil temperature change trajectory within the future time window is numerically calculated using the heat balance differential equation; a multi-scenario prediction and analysis mechanism is established, considering the impact of different load change patterns and environmental conditions; based on the prediction results, the period of risk of oil temperature exceeding the limit is identified, and a preventive heat dissipation control strategy is generated.

[0024] Based on the predicted changes in heat load and the current operating status of the radiators, the number of radiator groups to be switched on and off and the start-up and shutdown sequence are dynamically adjusted in advance to maintain the top oil temperature within the set target range.

[0025] In a preferred embodiment of the heat dissipation optimization method for a CLCC converter transformer, the normalization process includes: normalizing the original load data to the [0,1] interval according to the rated capacity, and normalizing the temperature distribution data according to the upper limit of safe operation, so that the load and temperature establish a monotonically increasing correlation curve under the same dimensionless scale.

[0026] In a preferred embodiment of the heat dissipation optimization method for a CLCC converter transformer, the heat transfer balance model establishes a load-heat dissipation relationship function based on the principle of constant transformer oil temperature, and its expression is: in, This represents the total heat loss of the transformer. The heat transfer coefficient, Due to the temperature difference between the oil and the environment, To maximize heat dissipation area, the oil temperature change trend is predicted based on real-time load data, and the number of radiator groups switched on and off is dynamically adjusted.

[0027] In a preferred embodiment of the heat dissipation optimization method for a CLCC converter transformer, the load prediction model employs a time series analysis method or a machine learning algorithm, as shown in Figure 8. Based on historical load and ambient temperature data from the past 30 minutes to 2 hours, the load value for the next 5–15 minutes is predicted, and the radiator group is adjusted accordingly 0.5–2 control cycles in advance.

[0028] In a preferred embodiment of the heat dissipation optimization method for a CLCC converter transformer, the radiator group switching follows the following: when the predicted load increases and the oil temperature change trend exceeds a set threshold, the number of radiator groups is increased in advance; when the predicted load decreases and the oil temperature change trend is lower than the target value, the shutdown of some radiators is delayed to avoid frequent start-ups and shutdowns; the switching action is controlled by zone based on the temperature distribution uniformity assessment results, and priority is given to strengthening the operation of radiators corresponding to hot spots.

[0029] In a preferred embodiment of the heat dissipation optimization method for a CLCC converter transformer, the load-temperature normalization curve and temperature prediction model are periodically corrected online based on actual operating data to adapt to model deviations caused by transformer aging, oil quality changes, or environmental condition drift.

[0030] In a preferred embodiment of the heat dissipation optimization method for a CLCC converter transformer, the prediction formula of the load forecasting model is as follows: ,in, To predict load, For historical load sequences, For ambient temperature, It is a time variable.

[0031] A system for optimizing the heat dissipation of a CLCC converter transformer includes a multispectral temperature monitoring platform. This platform performs continuous infrared thermal imaging on the surface of the CLCC converter transformer's heatsink to acquire spatially resolved temperature distribution data. It simultaneously collects real-time load data, top oil temperature, and ambient temperature of the CLCC converter transformer. The multispectral temperature monitoring platform includes: an infrared imaging system comprising an infrared multispectral analyzer, an optical lens, and a data acquisition card to output a heatsink surface temperature matrix; an operating parameter acquisition system comprising a load monitoring device, an oil temperature probe, and an ambient temperature sensor; and a control and data processing system comprising an embedded controller, a communication module, and a host computer for data fusion. The system includes model calculation and control command issuance; a normalization unit that normalizes load data to a standard load baseline and constructs a load-temperature normalized characteristic curve based on the temperature distribution data; a modeling unit that establishes a heat exchange balance model representing the mapping relationship between load changes and heat dissipation requirements based on the load-temperature normalized characteristic curve, constructs a load prediction model based on historical load and ambient temperature, and predicts the oil temperature change trend within a preset time window based on the heat exchange balance model; and an optimization module that dynamically adjusts the number of radiator groups and their start-stop sequence in advance based on the predicted heat load changes and the current radiator operating status to maintain the top oil temperature within the set target range.

[0032] A computer storage medium including computer instructions that, when run on a computer, cause the computer to perform the method.

[0033] An electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor implements the method when executing the program.

[0034] In one embodiment, the method includes: building a multispectral temperature monitoring platform based on the infrared thermal imaging detection principle; synchronously acquiring surface temperature distribution of converter transformer radiators and transformer load data based on the multispectral temperature monitoring platform; normalizing the load data to a standard load curve based on the temperature and load data and theoretical calculations; and dynamically adjusting the radiator group switching strategy through a heat exchange balance model to achieve constant oil temperature control.

[0035] The multispectral temperature monitoring platform consists of three parts: an infrared imaging system composed of an infrared multispectral analyzer, optical lens, and data acquisition card; a condition parameter acquisition system composed of temperature sensors, oil temperature probes, and load monitoring devices; and a control and data processing system composed of an embedded controller, communication module, and host computer software. The infrared multispectral analyzer continuously images the radiator surface to obtain temperature data in a 512×640 matrix, enabling real-time monitoring of the heat load distribution of the converter transformer. The heat exchange balance model, based on the principle of constant transformer oil temperature, establishes a load-heat dissipation relationship function, predicts heat load changes using real-time load data, and dynamically adjusts the number of radiator groups switched on and off.

[0036] The radiator group switching strategy is based on load forecast results and adjusts the radiator operating status in advance. Specifically, it includes: automatically increasing the number of radiator groups when the load is higher than the set threshold; reducing the number of radiator groups when the load is lower than the set threshold; and starting or stopping the radiators in advance according to the load change trend to maintain the oil temperature within the set range.

[0037] Temperature data collected by an infrared multispectral analyzer and transformer load data are normalized and plotted as a load-temperature characteristic curve for real-time evaluation of the cooling system efficiency and transformer operating status. The method also includes establishing a load prediction model based on historical load data and ambient temperature data to adjust cooling strategies in advance.

[0038] Referring to Figures 1 and 2, a method for optimizing the heat dissipation management of CLCC converter transformers is described. The method includes the following steps in sequence: (1) Based on the principle of infrared thermal imaging detection, a multispectral temperature monitoring platform is built. The platform mainly consists of three parts: an infrared imaging system composed of an infrared multispectral analyzer, an optical lens, and a data acquisition card; a working condition parameter acquisition system composed of a temperature sensor, an oil temperature probe, and a load monitoring device; and a control and data processing system composed of an embedded controller, a communication module, and host computer software.

[0039] (2) Calibrate the infrared multispectral analyzer and temperature sensor to ensure data acquisition accuracy.

[0040] (3) Collect the surface temperature distribution of the radiator and the transformer load data simultaneously and perform normalization processing.

[0041] (4) Based on the load-temperature characteristic curve, establish a heat exchange balance model.

[0042] (5) Adjust the switching strategy of radiator groups dynamically based on real-time load data and forecast results.

[0043] (6) Monitor oil temperature changes, verify control effectiveness, and optimize model parameters.

[0044] (7) Analyze the operating data and output a heat dissipation management report.

[0045] Referring to Figure 3, the surface of the radiator is continuously photographed using an infrared multispectral analyzer to obtain temperature data in a 512×640 matrix, enabling real-time monitoring of the heat load distribution of the converter transformer.

[0046] Refer to Figure 4, which shows the load and temperature sequence of the CLCC converter transformer under operating conditions according to the present invention. This figure illustrates the synchronous change trend of transformer load and radiator surface temperature over the same time period. Analysis of this sequence clearly shows that when the load increases, the temperature increases accordingly; when the load decreases, the temperature decreases as well. This synchronous relationship confirms that load change is one of the main factors affecting transformer temperature, providing an empirical basis for establishing a load-temperature prediction model. Simultaneously, the figure also reflects a certain lag in temperature response under certain operating conditions, which provides an important reference for optimizing heat dissipation control strategies.

[0047] Refer to Figure 5 for the load-temperature normalized curve of this invention. This figure normalizes the collected load and temperature data separately, eliminating the influence of dimensions and allowing for comparison and analysis on the same scale. The normalized curve clearly shows the quantitative relationship between load and temperature: the curve shows a monotonically increasing trend, indicating a positive correlation between load and temperature; the change in the curve slope reflects the dynamic characteristics of the heat dissipation system efficiency. When the curve slope increases, it indicates a larger temperature rise caused by a unit increase in load, and relatively insufficient heat dissipation efficiency; when the curve slope decreases, it indicates that the heat dissipation system is working well and can effectively control the temperature rise. This normalized curve provides an intuitive basis for real-time evaluation of the heat dissipation system performance.

[0048] Refer to Figure 6 for the radiator group switching logic diagram of the present invention. This logic is based on a load prediction model and temperature monitoring data. When the predicted load exceeds a set threshold or the monitored temperature exceeds a safe range, the radiator group is automatically activated; when the load is below the set threshold and the temperature is stable within a safe range, the radiator group operation is reduced. By adjusting the cooling strategy in advance, the oil temperature is maintained within the normal operating range.

[0049] Refer to Figure 7 for a typical load curve of the CLCC converter transformer. This curve can be used to determine the load change trend at different time periods. Combined with ambient temperature data, the future load value can be calculated through a load prediction model to adjust the switching of radiator groups in advance.

[0050] Figure 8 illustrates a load prediction model based on a deep learning-based intelligent prediction method. This model employs a deep neural network architecture, constructing a multi-branch fusion deep learning network, including: a convolutional neural network branch for receiving and processing infrared temperature image data to extract spatial distribution features; a recurrent neural network branch for receiving and processing load time-series data to capture time-dependent features; a fully connected network branch for receiving and processing scalar features such as environmental parameters; and the recurrent neural network also receives the outputs of the convolutional neural network branch and the fully connected network branch. An attention fusion module operates on the fully connected network branch, employing an attention mechanism to adaptively weight the importance of spatial distribution features, time-dependent features, and scalar features to improve prediction accuracy. The load prediction model is trained via the fully connected network branch to obtain the prediction output, and further: the prediction error is evaluated and output using a combined loss function of mean squared error and cross-entropy; the gradient of each layer's parameters is calculated using the chain rule; and the network weights are adjusted using the Adam optimization algorithm, with the learning rate dynamically decaying, to obtain a trained model for application in real-world scenarios.

[0051] Refer to Figure 9, which illustrates the temperature prediction model of this embodiment of the invention. This figure demonstrates the performance of the temperature prediction model established based on historical load data, ambient temperature data, and real-time monitoring data. The figure compares the curves of the actual monitored temperature and the model-predicted temperature, showing a high degree of agreement between the predicted and actual temperature curves, indicating that the temperature prediction model established in this invention has high accuracy. This prediction model comprehensively considers multiple factors such as load change trends, the influence of ambient temperature, and equipment operating characteristics, enabling it to predict transformer temperature change trends in advance, providing a decision-making basis for the intelligent switching of radiator groups. Through the temperature prediction model, the system can predict temperature changes 5-10 minutes in advance, enabling proactive adjustments to the heat dissipation strategy. This effectively avoids the oil temperature fluctuation problem caused by response lag in traditional control methods, significantly improving the accuracy and stability of oil temperature control.

[0052] In the above analysis results, the heat dissipation management strategy based on infrared thermal imaging detection and load prediction includes functions such as temperature distribution monitoring, load normalization processing, heat exchange balance model establishment and dynamic adjustment and control of radiators. Through real-time data acquisition and model calculation, constant oil temperature control is achieved, that is, the heat dissipation management of CLCC converter transformers is optimized.

[0053] Furthermore, the multispectral infrared thermal imaging system of this invention breaks through the limitations of traditional point-based temperature measurement, achieving high spatiotemporal resolution monitoring of the entire temperature field of the radiator surface. It can accurately identify local hot spots and uneven heat distribution caused by CLCC high-frequency voltage pulses. The normalization processing of load and temperature data eliminates dimensional differences and constructs a universal load-temperature characteristic curve, enabling quantitative analysis and efficiency evaluation of thermal response behavior under different operating conditions and environmental conditions on a unified scale. The dynamic model based on the heat exchange balance principle explicitly correlates electrical losses with heat dissipation capacity, providing a physically interpretable decision basis for constant oil temperature control. The predictive model that integrates historical load and ambient temperature endows the system with feedforward control capability, which can predict the trend of heat load changes 5-10 minutes in advance and drive the radiator group to switch in a forward-looking manner, effectively overcoming the inherent thermal inertia lag problem of traditional oil temperature feedback control. The aforementioned technical means together form a closed loop of "sensing-modeling-prediction-control", which not only suppresses the fluctuation of oil temperature, but also significantly reduces auxiliary energy consumption by starting and stopping the radiator group on demand, while extending the insulation life of the transformer. This fundamentally solves the heat dissipation bottleneck problem of CLCC converter transformers under the coupling effect of rapid load change and high frequency harmonics.

[0054] Although embodiments of the present invention have been described above in conjunction with the accompanying drawings, the present invention is not limited to the specific embodiments and application fields described above. The specific embodiments described above are merely illustrative and instructive, and not restrictive. Those skilled in the art can make many other forms based on the guidance of this specification and without departing from the scope of protection of the claims of the present invention, and all of these are within the scope of protection of the present invention.

Claims

1. A method for optimizing heat dissipation of a CLCC converter transformer, characterized in that, The process includes the following steps: performing continuous infrared thermal imaging on the surface of the CLCC converter transformer radiator to obtain spatially resolved temperature distribution data; simultaneously acquiring real-time load data, top oil temperature, and ambient temperature of the CLCC converter transformer; normalizing the load data to a standard load benchmark and constructing a load-temperature normalized characteristic curve based on the temperature distribution data; establishing a heat transfer balance model based on the load-temperature normalized characteristic curve to characterize the mapping relationship between load changes and heat dissipation requirements; constructing a load prediction model based on historical load and ambient temperature, and predicting the oil temperature change trend within a preset time window based on the heat transfer balance model; and dynamically adjusting the number of radiator groups and their start-up / shutdown sequence in advance according to the predicted heat load changes and the current radiator operating status to maintain the top oil temperature within a set target range.

2. The heat dissipation optimization method for a CLCC converter transformer according to claim 1, characterized in that, Preferably, the normalization process includes: normalizing the original load data to the [0,1] interval according to the rated capacity, and normalizing the temperature distribution data according to the upper limit of safe operation, so that the load and temperature establish a monotonically increasing correlation curve under the same dimensionless scale.

3. The heat dissipation optimization method for a CLCC converter transformer according to claim 1, characterized in that, The heat exchange balance model establishes a load-heat dissipation relationship function based on the principle of constant transformer oil temperature, and its expression is as follows: in, This represents the total heat loss of the transformer. The heat transfer coefficient, Due to the temperature difference between the oil and the environment, To maximize heat dissipation area, the oil temperature change trend is predicted based on real-time load data, and the number of radiator groups switched on and off is dynamically adjusted.

4. The heat dissipation optimization method for a CLCC converter transformer according to claim 1, characterized in that, The load prediction model uses time series analysis or machine learning algorithms to predict the load value for the next 5–15 minutes based on historical load and ambient temperature data from the past 30 minutes to 2 hours, and adjusts the radiator group accordingly 0.5–2 control cycles in advance.

5. The heat dissipation optimization method for a CLCC converter transformer according to claim 1, characterized in that, The switching of radiator groups follows the following principles: when the predicted load increases and the oil temperature change trend exceeds the set threshold, the number of radiator groups is increased in advance; when the predicted load decreases and the oil temperature change trend is lower than the target value, the shutdown of some radiators is delayed to avoid frequent start-stop operations; the switching action is controlled by zones based on the temperature distribution uniformity assessment results, and priority is given to strengthening the operation of radiators corresponding to hot spots.

6. The heat dissipation optimization method for a CLCC converter transformer according to claim 1, characterized in that, The load-temperature normalization curve and temperature prediction model are periodically calibrated online based on actual operating data to adapt to model deviations caused by transformer aging, oil quality changes, or environmental condition drift.

7. The heat dissipation optimization method for a CLCC converter transformer according to claim 1, characterized in that, The prediction formula for the load forecasting model is: ,in, To predict load, For historical load sequences, For ambient temperature, It is a time variable.

8. A system for optimizing the heat dissipation of a CLCC converter transformer according to any one of claims 1-7, characterized in that, It includes a multispectral temperature monitoring platform, which performs continuous infrared thermal imaging on the surface of the CLCC converter transformer radiator to acquire spatially resolved temperature distribution data, and simultaneously collects real-time load data, top oil temperature, and ambient temperature of the CLCC converter transformer. The multispectral temperature monitoring platform includes: an infrared imaging system, comprising an infrared multispectral analyzer, an optical lens, and a data acquisition card to output a radiator surface temperature matrix; an operating parameter acquisition system, comprising a load monitoring device, an oil temperature probe, and an ambient temperature sensor; and a control and data processing system, comprising an embedded controller, a communication module, and a host computer, used for data fusion, model calculation, and control command delivery. The system comprises: a load normalization unit, which normalizes the load data to a standard load baseline and, in conjunction with the temperature distribution data, constructs a load-temperature normalized characteristic curve; a modeling unit, which establishes a heat exchange balance model representing the mapping relationship between load changes and heat dissipation requirements based on the load-temperature normalized characteristic curve, constructs a load prediction model based on historical loads and ambient temperatures, and, in conjunction with the heat exchange balance model, predicts the oil temperature change trend within a preset time window; and an optimization module, which dynamically adjusts the number of radiator groups and their start-stop sequence in advance based on the predicted heat load changes and the current radiator operating status, in order to maintain the top oil temperature within a set target range.

9. A computer storage medium, characterized in that, The storage medium includes computer instructions that, when executed on a computer, cause the computer to perform the method as described in any one of claims 1-7.

10. An electronic device, characterized in that, The electronic device includes: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the method as described in any one of claims 1-7.

Images