An indirect method for measuring hot spot temperature in a prefabricated substation

Abstract

This invention belongs to the field of hotspot temperature measurement technology, specifically relating to an indirect method for measuring the hotspot temperature of a prefabricated substation. The steps include: establishing a coupled temperature and fluid field model of the prefabricated substation; obtaining the internal heat distribution and airflow characteristics of the prefabricated substation through simulation analysis; identifying the heat transfer path, heat accumulation area, and temperature-sensitive area of ​​the hotspot region; selecting candidate temperature measurement points along the heat transfer path and filtering to obtain a set of characteristic temperature measurement points; constructing an intermediate characteristic temperature inversion model to obtain intermediate characteristic temperatures; constructing a hotspot temperature inversion model, calculating the temperature rise rate and temperature rise increment based on the hotspot temperature, and constructing a graded early warning criterion for abnormal state identification; and performing online adaptive correction of the hotspot temperature inversion model to achieve non-intrusive online monitoring and status assessment of the hotspot temperature. This invention enables high-precision estimation and online monitoring of hotspot temperatures, improving the safety and reliability of equipment operation.

Description

Technical Field

[0001] This invention belongs to the field of hot spot temperature measurement technology, specifically relating to an indirect method for measuring hot spot temperature in a prefabricated substation. Background Technology

[0002] Prefabricated substations, as crucial power distribution equipment in power systems, are widely used in wind power, photovoltaic, and other new energy power generation systems, as well as urban power distribution networks. Prefabricated substations typically integrate various electrical equipment such as transformers, high- and low-voltage switchgear, and cable connection devices, offering advantages such as compact structure, small footprint, and convenient installation. However, due to the limited internal space and dense equipment arrangement, the heat generated during operation is difficult to dissipate promptly, easily leading to localized hot spots in transformer windings, conductor connections, and switch contacts. When the temperature of these hot spots continues to rise, it accelerates the aging of insulation materials, reduces equipment lifespan, and in severe cases, may even cause malfunctions or accidents, thus affecting the safe and stable operation of the power system. Therefore, accurate monitoring of the temperature of hot spots inside prefabricated substations has significant engineering importance and application value.

[0003] Currently, obtaining hotspot temperatures mainly relies on direct measurement or indirect estimation. Direct measurement typically employs fiber optic temperature sensors, placing sensing elements in transformer windings or key heat-generating components to achieve real-time hotspot temperature acquisition. This method offers high measurement accuracy, but the need to embed sensors inside the equipment increases manufacturing and installation costs and may impact the equipment structure. Furthermore, its implementation is challenging for already operational equipment, limiting its widespread application. On the other hand, infrared thermography, as a non-contact measurement method, is widely used in temperature detection of power equipment. This method obtains temperature information by detecting infrared signals radiated from the equipment surface, offering ease of operation. However, its measurement results are easily affected by ambient temperature, wind speed, dust, and measurement angle, and it struggles to penetrate the equipment casing to obtain the true internal hotspot temperature. Therefore, its application in enclosed structures like prefabricated substations is limited, and continuous online monitoring is often difficult to achieve.

[0004] In addition, temperature estimation methods based on empirical formulas or simplified thermal models are also used in engineering. These methods typically calculate hotspot temperatures based on the empirical relationship between load current and temperature rise, offering the advantage of simplicity. However, because they fail to fully consider the complex structural layout, airflow distribution, and changes in heat dissipation conditions within prefabricated substations, they often struggle to accurately reflect hotspot temperature changes in actual operation. This is especially true in renewable energy generation scenarios, where frequent load fluctuations and complex, variable operating conditions further reduce the applicability and accuracy of traditional empirical models.

[0005] Furthermore, because the interior of a prefabricated substation is a relatively enclosed space, the airflow organization and heat transfer process exhibit significant non-uniformity, resulting in substantial differences in temperature distribution across different locations. Hot spots are typically located inside windings or at connection interfaces—locations difficult to directly access—making it challenging for conventional temperature measurement methods to cover these critical areas. Simultaneously, most existing methods fail to effectively utilize the internal heat transfer paths and flow characteristics of the equipment, lacking in-depth analysis of the hot spot formation mechanism. This leads to a lack of targeted placement of temperature measurement points, thereby affecting the accuracy and stability of temperature estimation results.

[0006] In summary, existing technologies still have many shortcomings in monitoring hot spot temperatures in prefabricated substations. These shortcomings mainly include the difficulty in achieving non-intrusive measurements, the susceptibility of measurement results to environmental and operating conditions, insufficient adaptability to complex structures and airflow distributions, and the difficulty in achieving high-precision continuous online monitoring and early warning functions. Summary of the Invention

[0007] In view of the shortcomings of the prior art, the purpose of this invention is to provide an indirect method for measuring the temperature of hot spots in a prefabricated substation. Without the need to place sensors directly at the hot spot location, this method combines the internal temperature field and fluid field characteristics of the prefabricated substation, and by reasonably selecting characteristic temperature measurement points and constructing an effective temperature inversion model, can achieve high-precision estimation and online monitoring of hot spot temperatures, thereby improving the safety and reliability of equipment operation.

[0008] To achieve the above objectives, the present invention provides an indirect method for measuring the hot spot temperature of a prefabricated substation, comprising the following steps: S1. Based on simulation software and the actual structure of the prefabricated substation, establish a coupled temperature field and fluid field model of the prefabricated substation, and obtain the heat distribution and air flow characteristics inside the box through simulation analysis. S2. Based on the heat distribution and airflow characteristics inside the box, perform heat flow streamline analysis, and calculate the temperature gradient in combination with the spatial distribution of the temperature field to identify the heat transfer path, heat accumulation area and temperature change sensitive area in the hot spot area. S3. Select candidate temperature measurement points along the heat transfer path. Based on the joint screening strategy of heat flow streamline and temperature gradient, the candidate temperature measurement points are screened. After screening, the comprehensive evaluation function is used to obtain the comprehensive evaluation value. The candidate temperature measurement points with the comprehensive evaluation value greater than the preset evaluation threshold and the preset correlation requirements are selected as the feature temperature measurement point set. S4. Based on the characteristic temperature measurement points, temperature sensors are deployed to collect operating parameters and the temperature of the characteristic temperature measurement points during the operation of the box-type substation, and an intermediate characteristic temperature inversion model is constructed to obtain the intermediate characteristic temperature. S5. Based on intermediate characteristic temperatures and operating parameters, construct a hotspot temperature inversion model, calculate the temperature rise change rate and temperature rise rate according to the hotspot temperature, and construct a graded early warning criterion to identify abnormal states. S6. Utilize on-site inspection data, infrared temperature measurement data, or maintenance measurement data to perform online adaptive correction on the hotspot temperature inversion model, thereby achieving non-intrusive online monitoring and status assessment of hotspot temperatures.

[0009] As a preferred embodiment of the present invention, in S1, based on the actual structure of the prefabricated substation, and based on the actual geometric dimensions and assembly relationships of the enclosure, transformer, high and low voltage switchgear, cable joints, busbars, and heat dissipation devices, the three-dimensional modeling module of Comsol is used to complete the independent modeling of each component, and then the components are precisely assembled to form a complete three-dimensional physical model of the prefabricated substation. Define the parameters and material properties required for the simulation, including: Equipment structural parameters include the external dimensions of the enclosure, the thickness of the shell plates, the layout and thickness of the internal partitions, the installation spacing and orientation of each component, the cross-sectional dimensions and direction of the airflow channels, and the location, number, size and opening ratio of the ventilation openings. Material thermal conductivity characteristics: Set the thermal conductivity, specific heat capacity, surface emissivity, and contact thermal resistance of the corresponding materials for each component; Load loss, quantifying the specific values ​​of copper loss, iron loss of transformer, operating loss of circuit breakers and contactors in high and low voltage switchgear, and contact loss of cable joints. Ventilation conditions, including setting ambient temperature, ambient wind speed, inlet / outlet air boundary conditions of the ventilation openings, and the sealing / ventilation status of the enclosure; Add heat conduction and laminar flow physics fields in Comsol and complete the coupling settings to obtain a coupled model of temperature field and fluid field. Select the transient solver for calculation, and complete the simulation iteration by setting the preset time step and solution accuracy. Obtain the temperature distribution cloud map, heat flux density distribution, air velocity and direction, and heat flow streamline of each region inside the box. Obtain the heat distribution and air flow characteristics inside the box. Among them, the heat flow streamline is generated by solving the coupled model of temperature field and fluid field of box substation through computational fluid dynamics method. It is used to characterize the heat transfer path of hot spot area in box substation and quantify the spatial distribution of heat flux intensity.

[0010] As a preferred embodiment of the present invention, the specific process of identifying the heat transfer path, heat accumulation area, and temperature change sensitive area in S2 is as follows: Perform heat flow analysis to clarify the heat transfer path from transformer windings, busbars, cable joints, circuit breakers and contactors in high and low voltage switchgear to the external environment, distinguish conduction heat flow channels, convection heat flow channels and radiation heat flow channels, and determine the heat source and diffusion law of each hot spot area; By combining the spatial distribution data of the temperature field corresponding to the temperature distribution cloud map, spatial differentiation is performed on each location inside the box to calculate the magnitude and direction of the temperature gradient at each location, thus quantifying the degree of temperature change in each region. Identify heat transfer paths, heat convergence areas, and temperature-sensitive areas in hot spots: Identify the heat transfer paths in hot spots by continuously connected heat flow lines; identify heat convergence areas by areas where heat flow lines intersect, temperature peaks are concentrated, and heat flux density is higher than the average heat flux density of the chamber; and identify temperature-sensitive areas by areas where the absolute value of the temperature gradient is higher than the absolute value of the average temperature gradient of the chamber.

[0011] As a preferred embodiment of the present invention, in S3, the joint screening strategy based on heat flow streamlines and temperature gradient specifically involves screening a set of candidate temperature measurement points located on the heat transfer path, heat accumulation area, and temperature change sensitive area, based on heat flow streamline distribution, heat flux density, and temperature gradient. These candidate points have a heat flux density that is not lower than the average heat flux density of the chamber, and are located in the temperature change sensitive area with an absolute temperature gradient that is not lower than the absolute temperature gradient of the chamber. Based on four evaluation indicators—temperature gradient characteristics, heat flux intensity characteristics, path distance characteristics, and correlation with hotspot temperatures—a comprehensive evaluation function for candidate temperature measurement points is constructed. The comprehensive evaluation value is then calculated for each selected candidate temperature measurement point. The comprehensive evaluation function is as follows: ; In the formula, This is the comprehensive evaluation value of the i-th candidate temperature measurement point; Let be the temperature gradient feature of the i-th candidate temperature measurement point; The average temperature gradient inside the chamber; Let be the heat flux intensity characteristic quantity of the i-th candidate temperature measurement point; The average heat flux density inside the box; The distance from the i-th candidate temperature measurement point to the hot spot along the heat flow path is denoted as . This is the maximum heat flow path length inside the enclosure; Let be the Pearson correlation coefficient between the temperature sequence of the i-th candidate temperature measurement point and the temperature sequence of the hot spot. , , , They are respectively , , , Weighting coefficients; The preset relevance requirement is, It is greater than or equal to the preset correlation coefficient threshold.

[0012] As a preferred embodiment of the present invention, for the i-th candidate temperature measurement point, intermediate variables corresponding to temperature gradient characteristics, heat flux intensity characteristics, path distance characteristics, and the correlation between these characteristics and the hot spot temperature are constructed. , , , : ; The weight coefficients are determined according to the normalization method: ; In the formula, This represents the weight coefficient corresponding to the k-th evaluation indicator; This is the intermediate variable corresponding to the k-th evaluation indicator.

[0013] In a preferred embodiment of the present invention, the operating parameters in S4 include load current I and ambient temperature. Including external operating parameters, the number of temperature sensors deployed is n, and the temperature sequence of characteristic temperature measurement points is as follows: , This represents the temperature of the nth feature temperature measurement point; The intermediate characteristic temperature inversion model is as follows: ; In the formula, This is the intermediate characteristic temperature; This is an environmental compensation item; The inversion function of the intermediate characteristic temperature inversion model is constructed using a regression model.

[0014] As a preferred embodiment of the present invention, in S5, the hotspot temperature inversion model is expressed as follows: ; In the formula, For hotspot temperature; Let m be the intermediate feature temperature corresponding to the z-th intermediate feature position, and m be the number of intermediate feature temperatures. Let be the equivalent thermal resistance from the z-th intermediate feature location to the hot spot; The baseline intercept term; The coefficient for the square term of the load current; This is the coefficient for the ambient temperature term; For environmental compensation items, the coefficient is... The weighting coefficient for the z-th intermediate characteristic temperature; for The weighting coefficients.

[0015] As a preferred embodiment of the present invention, based on Calculate the temperature rise : ; In the formula, Ambient temperature; This includes one or more of the following parameters: solar radiation intensity, ventilation status, and radiative heat transfer parameters of the enclosure, used to reduce the influence of the external environment on the temperature measurement results; Calculate the rate of change of temperature rise : ; In the formula, For reference temperature rise; Calculate the rate of temperature rise : ; In the formula, t represents time.

[0016] As a preferred embodiment of the present invention, in S5, the graded early warning criterion specifically involves setting a first threshold and a second threshold, wherein the second threshold is greater than the first threshold. When either the rate of change of temperature rise or the rate of temperature rise exceeds the first threshold, it is determined to be an abnormal hotspot temperature state. When both the rate of change of temperature rise and the rate of temperature rise exceed the second threshold, it is determined to be a severe overheating state, and early warning or protection measures are implemented. The early warning or protection measures include issuing an alarm signal, starting ventilation or cooling devices, reducing load operation, or triggering protection tripping.

[0017] As a preferred embodiment of the present invention, in S6, the hotspot temperature inversion model is adaptively corrected online by recursive least squares method, Kalman filtering or incremental learning method.

[0018] The beneficial effects of this invention are: This invention, through comprehensive analysis of the internal temperature field and airflow characteristics of the equipment, can more clearly reflect the heat transfer process within the prefabricated substation and rationally select characteristic temperature measurement points accordingly. Compared to traditional methods that rely on experience or simple point placement, this invention is more targeted in its selection of measurement points, making subsequent temperature estimation more reliable.

[0019] This invention utilizes the relationship between temperature measurement point data and operating parameters to construct a hot spot temperature inversion model, enabling temperature estimation without direct contact with the hot spot location. It eliminates the need to place sensors in high-voltage areas such as windings, reducing the impact on equipment structure and demonstrating good engineering feasibility. It is particularly suitable for the retrofitting and application of already operational equipment.

[0020] This invention incorporates key operating factors such as load current and ambient temperature into the model, enabling the calculation results to be dynamically adjusted according to changes in operating conditions, thus more realistically reflecting the actual thermal state of the equipment. Modeling by combining heat flow paths and the degree of influence of measuring points helps improve adaptability under different structural and operating conditions, exhibiting better stability and accuracy compared to traditional methods.

[0021] This invention assesses the degree of hotspot temperature anomalies by setting a temperature rise change rate index, enabling tiered early warning systems. When abnormal temperature changes occur, it promptly reflects the changing trends in equipment operating status, providing maintenance personnel with a reference basis. This helps in the early detection of potential faults and the implementation of corresponding measures, thereby improving the safety and reliability of equipment operation. Overall, this invention not only achieves effective estimation of hotspot temperatures but also considers practicality and scalability, demonstrating promising application prospects. Attached Figure Description

[0022] Figure 1 This is a flowchart illustrating the principle of this invention; Figure 2 This is a schematic diagram of a temperature field slice obtained from simulation analysis in Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of the velocity field slice obtained from the simulation analysis in Embodiment 1 of the present invention; Figure 4 This is a schematic diagram of the main heat flow path and hot spot region extracted based on temperature field slicing results in Embodiment 1 of the present invention; Figure 5 This is a schematic diagram illustrating the selection of candidate temperature measurement points and characteristic temperature measurement points in Embodiment 1 of the present invention; Figure 6 This is a comparative schematic diagram of the inversion results for intermediate characteristic temperatures in Embodiment 1 of the present invention; Figure 7 This is a schematic diagram comparing the inversion results of hotspot temperature in Embodiment 1 of the present invention. Detailed Implementation

[0023] The embodiments of the present invention will be further described below with reference to the accompanying drawings: Example 1: As Figure 1 As shown, a method for indirect measurement of hot spot temperature in a prefabricated substation includes the following steps: S1. Based on simulation software and the actual structure of the prefabricated substation, establish a coupled temperature field and fluid field model of the prefabricated substation, and obtain the heat distribution and air flow characteristics inside the box through simulation analysis. S2. Based on the heat distribution and airflow characteristics inside the box, perform heat flow streamline analysis, and calculate the temperature gradient in combination with the spatial distribution of the temperature field to identify the heat transfer path, heat accumulation area and temperature change sensitive area in the hot spot area. S3. Select candidate temperature measurement points along the heat transfer path. Based on the joint screening strategy of heat flow streamline and temperature gradient, the candidate temperature measurement points are screened. After screening, the comprehensive evaluation function is used to obtain the comprehensive evaluation value. The candidate temperature measurement points with the comprehensive evaluation value greater than the preset evaluation threshold and the preset correlation requirements are selected as the feature temperature measurement point set. S4. Based on the characteristic temperature measurement points, temperature sensors are deployed to collect operating parameters and the temperature of the characteristic temperature measurement points during the operation of the box-type substation, and an intermediate characteristic temperature inversion model is constructed to obtain the intermediate characteristic temperature. S5. Based on intermediate characteristic temperatures and operating parameters, construct a hotspot temperature inversion model, calculate the temperature rise change rate and temperature rise rate according to the hotspot temperature, and construct a graded early warning criterion to identify abnormal states. S6. Utilize on-site inspection data, infrared temperature measurement data, or maintenance measurement data to perform online adaptive correction on the hotspot temperature inversion model, thereby achieving non-intrusive online monitoring and status assessment of hotspot temperatures.

[0024] In S1, based on the actual structure of the prefabricated substation, and based on the actual geometric dimensions and assembly relationships of the enclosure, transformer, high and low voltage switchgear, cable joints, busbars, and heat dissipation devices, the three-dimensional modeling module of Comsol is used to complete the independent modeling of each component, and then the components are precisely assembled to form a complete three-dimensional physical model of the prefabricated substation. Define the parameters and material properties required for the simulation, including: Equipment structural parameters include the external dimensions of the enclosure, the thickness of the shell plates, the layout and thickness of the internal partitions, the installation spacing and orientation of each component, the cross-sectional dimensions and direction of the airflow channels, and the location, number, size and opening ratio of the ventilation openings. Material thermal conductivity characteristics: Set the thermal conductivity, specific heat capacity, surface emissivity, and contact thermal resistance of the corresponding materials for each component; Load loss, quantifying the specific values ​​of copper loss, iron loss of transformer, operating loss of circuit breakers and contactors in high and low voltage switchgear, and contact loss of cable joints. Ventilation conditions, including setting ambient temperature, ambient wind speed, inlet / outlet air boundary conditions of the ventilation openings, and the sealing / ventilation status of the enclosure; Add heat conduction and laminar flow physics fields in Comsol and complete the coupling settings to obtain a coupled model of temperature field and fluid field. Select the transient solver for calculation, and complete the simulation iteration by setting the preset time step and solution accuracy. Obtain the temperature distribution cloud map, heat flux density distribution, air velocity and direction, and heat flow streamline of each region inside the box. Obtain the heat distribution and air flow characteristics inside the box. Among them, the heat flow streamline is generated by solving the coupled model of temperature field and fluid field of box substation through computational fluid dynamics method. It is used to characterize the heat transfer path of hot spot area in box substation and quantify the spatial distribution of heat flux intensity.

[0025] A temperature field slice obtained from simulation analysis of the coupled temperature field and fluid field model is shown below. Figure 2 As shown, from Figure 2 Three high-temperature hotspots can be clearly identified. The central region is the highest temperature point (approximately 150℃), with symmetrically distributed secondary high-temperature regions on both sides. The hotspots exhibit high temperature gradients and dense heat flow lines, allowing heat to diffuse along the heat flow paths to other areas of the chamber. This temperature field distribution provides a simulation data foundation for subsequent steps such as extracting the main heat flow path, selecting candidate temperature measurement points, and calculating equivalent thermal resistance.

[0026] and Figure 2 Velocity field slices at the same location, such as Figure 3 As shown, from Figure 3 As can be seen, the flow velocity is higher on the right side of the chamber (up to approximately 0.95 m / s), while it is lower on the left side, resulting in a significant difference in convective heat transfer. This velocity distribution is coupled with the temperature field distribution; the high-velocity region on the right exhibits stronger convective heat transfer, corresponding to better heat dissipation conditions in the second-highest temperature region on the right side of the temperature field. Conversely, the low-velocity region on the left shows weaker convective heat transfer, leading to heat accumulation. The velocity field data provides a fluid dynamic basis for subsequent steps in environmental compensation, heat dissipation condition analysis, and equivalent thermal resistance calculation.

[0027] In S2, the specific process of identifying the heat transfer path of hotspot areas, heat accumulation areas, and temperature-sensitive areas is as follows: Perform heat flow analysis to clarify the heat transfer path from transformer windings, busbars, cable joints, circuit breakers and contactors in high and low voltage switchgear to the external environment, distinguish conduction heat flow channels, convection heat flow channels and radiation heat flow channels, and determine the heat source and diffusion law of each hot spot area; By combining the spatial distribution data of the temperature field corresponding to the temperature distribution cloud map, spatial differentiation is performed on each location inside the box to calculate the magnitude and direction of the temperature gradient at each location, thus quantifying the degree of temperature change in each region. Identify heat transfer paths, heat convergence areas, and temperature-sensitive areas in hot spots: Identify the heat transfer paths in hot spots by continuously connected heat flow lines; identify heat convergence areas by areas where heat flow lines intersect, temperature peaks are concentrated, and heat flux density is higher than the average heat flux density of the chamber; and identify temperature-sensitive areas by areas where the absolute value of the temperature gradient is higher than the absolute value of the average temperature gradient of the chamber.

[0028] Figure 4 This is a schematic diagram of the main heat flow paths and hotspot regions extracted based on temperature field slicing results, with the temperature field distribution as the background. Figure 4 The hotspot locations (cyan dots) and multiple main heat transfer paths (white dashed lines) are marked. It can be seen that heat spreads from the central hotspot to the surrounding area along multiple main paths. The temperature gradient and heat flux density along these paths are high, making them key areas for selecting candidate temperature measurement points in subsequent steps.

[0029] In S3, based on the coupled temperature field and fluid field model, the main heat flow path (i.e., the region with dense heat flow streamlines and high temperature gradient) is extracted from the hot spot region. A series of potential measurement points are generated on these paths to form an initial candidate set. The combined heat flow line and temperature gradient screening strategy specifically involves using heat transfer paths, heat accumulation areas, and temperature change sensitive areas as the scope, and using heat flow line distribution, heat flux density, and temperature gradient as the basis to screen a set of candidate temperature measurement points that are located on the heat transfer path, have a heat flux density not lower than the average heat flux density of the chamber, and are located in the temperature change sensitive area, with an absolute temperature gradient not lower than the absolute temperature gradient of the chamber. Based on four evaluation indicators—temperature gradient characteristics, heat flux intensity characteristics, path distance characteristics, and correlation with hotspot temperatures—a comprehensive evaluation function for candidate temperature measurement points is constructed. The comprehensive evaluation value is then calculated for each selected candidate temperature measurement point. The comprehensive evaluation function is as follows: ; In the formula, This is the comprehensive evaluation value of the i-th candidate temperature measurement point; The temperature gradient feature of the i-th candidate temperature measurement point (the average temperature gradient at the location of the i-th candidate temperature measurement point). The average temperature gradient inside the chamber; Let be the heat flux intensity characteristic of the i-th candidate temperature measurement point (the average heat flux density at the location of the i-th candidate temperature measurement point). The average heat flux density inside the box; The path distance from the i-th candidate temperature measurement point to the hot spot along the heat flow path is given by the near-hot spot gain form, where the closer the distance, the greater the contribution. This is the maximum heat flow path length inside the enclosure; Let be the Pearson correlation coefficient between the temperature sequence of the i-th candidate temperature measurement point and the temperature sequence of the hot spot. , , , They are respectively , , , The weighting coefficients are given, and the sum of the four weighting coefficients is 1. The preset relevance requirement is, The correlation coefficient is greater than or equal to a preset threshold (set between 0.7 and 0.95). This is the preset evaluation threshold. The value range is 0.8 to 1.2.

[0030] Figure 5 This diagram illustrates the selection results of candidate and feature temperature measurement points. The background also shows the temperature field distribution, with yellow dots representing initial candidate temperature measurement points and green dots representing the final selected feature temperature measurement points. Figure 5 It can be seen that, after comprehensive evaluation of four indicators—temperature gradient, heat flux intensity, path distance, and correlation with hotspots—the characteristic temperature measurement points are mainly concentrated on the main heat flux path, and are close to the hotspot area, with high temperature gradient and heat flux density, which can effectively reflect the temperature changes of the hotspots.

[0031] For the i-th candidate temperature measurement point, construct dimensionless intermediate variables corresponding to the temperature gradient characteristics, heat flux intensity characteristics, path distance characteristics, and their correlation with the hotspot temperature. , , , : ; The weight coefficients are determined according to the normalization method: ; In the formula, This represents the weight coefficient corresponding to the k-th evaluation indicator; This is the intermediate variable corresponding to the k-th evaluation indicator.

[0032] The larger, the corresponding The larger; The larger, the corresponding The larger; The smaller, the corresponding The larger; The higher, the corresponding The larger.

[0033] In S4, the operating parameters include load current I and ambient temperature. Including external operating parameters, the number of temperature sensors deployed is n, and the temperature sequence of characteristic temperature measurement points is as follows: , This represents the temperature of the nth feature temperature measurement point; The intermediate characteristic temperature inversion model is as follows: ; In the formula, This is the intermediate characteristic temperature; This is an environmental compensation item; This is the inversion function of the intermediate characteristic temperature inversion model.

[0034] A regression model was constructed and trained based on simulation data and historical measured data. The model input consisted of the temperature sequence of characteristic temperature measurement points, load current, ambient temperature, and environmental compensation term. The output was the temperature corresponding to the intermediate characteristic position, resulting in a trained intermediate characteristic temperature inversion model. Real-time collected temperature sequences of characteristic temperature measurement points, load current, ambient temperature, and environmental compensation term were input into the trained intermediate characteristic temperature inversion model to obtain one or more intermediate characteristic temperatures, reflecting the key temperature levels along different thermal paths. The intermediate characteristic position refers to a representative location highly sensitive to changes in hotspot temperature along the heat flow path from the hotspot region to the characteristic temperature measurement point in the coupled temperature and fluid field model of the prefabricated substation.

[0035] The regression model can be any one of the following: support vector regression model, gradient boosting tree model, or BP neural network model.

[0036] External operating parameters include one or more of the following: wind speed, solar irradiance, and ventilation status parameters.

[0037] Figure 6 This section compares the inversion results for intermediate characteristic temperatures. Figure 6 The blue curve represents the actual intermediate characteristic temperature, while the orange curve represents the predicted intermediate characteristic temperature obtained by inverting data from characteristic temperature measurement points. Figure 6 As can be seen, the predicted curve closely matches the actual curve, accurately tracking the changing trend of intermediate characteristic temperature over time, with small inversion error, thus verifying the accuracy of the intermediate characteristic temperature inversion model.

[0038] In S5, the hotspot temperature inversion model is expressed as: ; In the formula, For hotspot temperature; Let m be the intermediate feature temperature corresponding to the z-th intermediate feature position, and m be the number of intermediate feature temperatures. Let be the equivalent thermal resistance from the z-th intermediate feature location to the hot spot; The baseline intercept term; The coefficient for the square term of the load current; This is the coefficient for the ambient temperature term; For environmental compensation items, the coefficient is... The weighting coefficient for the z-th intermediate characteristic temperature; for The weighting coefficients.

[0039] , , , , , The parameters are obtained by parameter identification from simulation data, experimental data and field operation data. The parameter identification takes the hot spot reference temperature as the target output and the intermediate characteristic temperature sequence, load current, ambient temperature, environmental compensation term and equivalent thermal resistance sequence from the intermediate characteristic position to the hot spot as input. The solution is obtained by using any one of the least squares method, ridge regression or recursive least squares method.

[0040] based on Calculate the temperature rise : ; In the formula, Ambient temperature; It can be seen that, It is determined by both the equipment's heat generation capacity and heat dissipation conditions; The impedance characteristics are determined by one or more of the following factors: the heat transfer path length, path cross-sectional area, thermal conductivity of the path material, thermal resistance of the contact interface, and structural connection method between the z-th intermediate feature location and the hot spot. This is based on Fourier's law of heat conduction and the series thermal resistance synthesis rule, and is used to describe the impedance characteristics during heat transfer. For example, the calculation method can be: ; In the formula, , , These represent the length of the heat transfer path from the z-th intermediate feature position to the hot spot, the thermal conductivity of the material in each segment, and the cross-sectional area of ​​the path, respectively. For the thermal resistance of the contact interface.

[0041] This includes one or more of the following parameters: solar radiation intensity, ventilation status, and radiative heat transfer parameters of the enclosure, used to reduce the influence of the external environment on the temperature measurement results; Calculate the rate of change of temperature rise : ; In the formula, For reference temperature rise; Calculate the rate of temperature rise : ; In the formula, t represents time.

[0042] Figure 7 This is a comparison of the inversion results of hotspot temperatures. Figure 7 The blue curve represents the actual temperature of the hotspot, the orange curve represents the predicted temperature calculated based on the hotspot temperature inversion model, and the green curve represents the corresponding temperature rise (the difference between the hotspot temperature and the ambient temperature). Figure 7 As can be seen, the predicted temperature curve is highly consistent with the actual temperature curve, accurately tracking the dynamic trend of hot spot temperature changes with load, with minimal inversion error. At the same time, the temperature rise curve clearly reflects the heat generation and heat dissipation characteristics of the equipment under different operating conditions, verifying the accuracy and effectiveness of the hot spot temperature inversion model.

[0043] In S5, the specific criteria for graded early warning are as follows: a first threshold and a second threshold are set, and the second threshold is greater than the first threshold. When either the rate of change of temperature or the rate of temperature rise exceeds the first threshold, it is determined to be an abnormal hot spot temperature state. When both the rate of change of temperature and the rate of temperature rise exceed the second threshold, it is determined to be a severe overheating state, and early warning or protection measures are implemented. The early warning or protection measures include issuing an alarm signal, starting ventilation or cooling devices, reducing load operation, or triggering protection trip.

[0044] Combining the indirect measurement method of hot spot temperature in prefabricated substations with the safety operation standards of power equipment, and referring to the temperature rise The recommended temperature is 90℃. The first threshold for the rate of change of temperature rise is set to 1.2, and the second threshold is set to 1.5. When the rate of temperature rise is in minutes, the first threshold is set to 1.0℃ / min and the second threshold is set to 1.8℃ / min. When either the rate of change of temperature rise or the rate of temperature rise exceeds the first threshold, it is determined that the hot spot temperature is abnormal. When both exceed the second threshold at the same time, it is determined that it is a serious overheating and triggers alarms, starts cooling or protection trips, etc. The first / second thresholds can be adaptively fine-tuned by ±5%~10% according to the season and load conditions.

[0045] The hot spot temperature inversion model must meet the constraints of the heat conduction mechanism, specifically: the hot spot temperature increases with the square of the load current and the intermediate characteristic temperature, decreases with the increase of the heat transfer path distance, and is not lower than the ambient temperature.

[0046] In S6, online adaptive correction of the hotspot temperature inversion model is achieved through recursive least squares, Kalman filtering, or incremental learning. Specifically, the measured hotspot temperature, characteristic temperature measurement point temperature, load current, ambient temperature, and external operating condition data are used as a sample set. The model parameters (including the baseline intercept term, load current squared term coefficient, ambient temperature term coefficient, environmental compensation term coefficient, intermediate characteristic temperature weight coefficient, and equivalent thermal resistance weight coefficient) are iteratively updated using recursive least squares, Kalman filtering, or incremental learning algorithms. Among these, recursive least squares corrects the parameters in real time through recursion, Kalman filtering dynamically adjusts the model state based on the prediction and measured residuals, and incremental learning continuously optimizes the model mapping relationship using new data. This enables the model to adapt to actual operating conditions, improving the accuracy and long-term stability of hotspot temperature inversion.

[0047] Example 2: The difference between this example and Example 1 is that... The determination is adaptively based on the statistical distribution of the comprehensive evaluation values ​​of the candidate temperature measurement points, that is: ; In the formula, The average value of the comprehensive evaluation of all candidate temperature measurement points; is the standard deviation of the comprehensive evaluation values ​​of all candidate temperature measurement points; K is the adjustment coefficient, with a value range of 0~0.5; when and If the correlation coefficient threshold is greater than or equal to the candidate temperature measurement point, the candidate temperature measurement point is determined as the characteristic temperature measurement point.

[0048] Furthermore, a path distance weighting coefficient and a thermal resistance correction term are introduced into the hotspot temperature inversion model. The path distance weighting coefficient is used to characterize the attenuation effect of the path distance between the intermediate feature location and the hotspot on the temperature contribution, and it decreases as the path distance increases. The thermal resistance correction term is used to characterize the influence of environmental factors on the heat transfer process and correct the equivalent thermal resistance.

[0049] Example 3: An indirect temperature measurement device for hot spots in a prefabricated substation, comprising: One or more processors; Memory, used to store one or more computer programs; When one or more programs are executed by one or more processors, the one or more processors perform the method in Embodiment 1 or Embodiment 2.

[0050] Example 4: A computer-readable storage medium having executable instructions stored thereon, which, when executed by a processor, cause the processor to perform the method in Example 1 or Example 2.

Claims

1. A method for indirect measurement of hot spot temperature in a prefabricated substation, characterized in that... Includes the following steps: S1. Based on simulation software and the actual structure of the prefabricated substation, establish a coupled temperature field and fluid field model of the prefabricated substation, and obtain the heat distribution and air flow characteristics inside the box through simulation analysis. S2. Based on the heat distribution and airflow characteristics inside the box, perform heat flow streamline analysis, and calculate the temperature gradient in combination with the spatial distribution of the temperature field to identify the heat transfer path, heat accumulation area and temperature change sensitive area in the hot spot area. S3. Select candidate temperature measurement points along the heat transfer path. Based on the joint screening strategy of heat flow streamline and temperature gradient, the candidate temperature measurement points are screened. After screening, the comprehensive evaluation function is used to obtain the comprehensive evaluation value. The candidate temperature measurement points with the comprehensive evaluation value greater than the preset evaluation threshold and the preset correlation requirements are selected as the feature temperature measurement point set. S4. Based on the characteristic temperature measurement points, temperature sensors are deployed to collect operating parameters and the temperature of the characteristic temperature measurement points during the operation of the box-type substation, and an intermediate characteristic temperature inversion model is constructed to obtain the intermediate characteristic temperature. S5. Based on intermediate characteristic temperatures and operating parameters, construct a hotspot temperature inversion model, calculate the temperature rise change rate and temperature rise rate according to the hotspot temperature, and construct a graded early warning criterion to identify abnormal states. S6. Utilize on-site inspection data, infrared temperature measurement data, or maintenance measurement data to perform online adaptive correction on the hotspot temperature inversion model, thereby achieving non-intrusive online monitoring and status assessment of hotspot temperatures.

2. The method for indirect measurement of hot spot temperature in a prefabricated substation according to claim 1, characterized in that: In S1, based on the actual structure of the prefabricated substation, and based on the actual geometric dimensions and assembly relationships of the enclosure, transformer, high and low voltage switchgear, cable joints, busbars, and heat dissipation devices, the three-dimensional modeling module of Comsol is used to complete the independent modeling of each component, and then the components are precisely assembled to form a complete three-dimensional physical model of the prefabricated substation. Define the parameters and material properties required for the simulation, including: Equipment structural parameters include the external dimensions of the enclosure, the thickness of the shell plates, the layout and thickness of the internal partitions, the installation spacing and orientation of each component, the cross-sectional dimensions and direction of the airflow channels, and the location, number, size and opening ratio of the ventilation openings. Material thermal conductivity characteristics: Set the thermal conductivity, specific heat capacity, surface emissivity, and contact thermal resistance of the corresponding materials for each component; Load loss, quantifying the specific values ​​of copper loss, iron loss of transformer, operating loss of circuit breakers and contactors in high and low voltage switchgear, and contact loss of cable joints. Ventilation conditions, including setting ambient temperature, ambient wind speed, inlet / outlet air boundary conditions of the ventilation openings, and the sealing / ventilation status of the enclosure; Add heat conduction and laminar flow physics fields in Comsol and complete the coupling settings to obtain a coupled model of temperature field and fluid field. Select the transient solver for calculation, and complete the simulation iteration by setting the preset time step and solution accuracy. Obtain the temperature distribution cloud map, heat flux density distribution, air velocity and direction, and heat flow streamline of each region inside the box. Obtain the heat distribution and air flow characteristics inside the box. Among them, the heat flow streamline is generated by solving the coupled model of temperature field and fluid field of box substation through computational fluid dynamics method. It is used to characterize the heat transfer path of hot spot area in box substation and quantify the spatial distribution of heat flux intensity.

3. The method for indirect measurement of hot spot temperature in a prefabricated substation according to claim 2, characterized in that: In S2, the specific process of identifying the heat transfer path, heat accumulation area, and temperature change sensitive area of ​​the hot spot is as follows: Perform heat flow analysis to clarify the heat transfer path from transformer windings, busbars, cable joints, circuit breakers and contactors in high and low voltage switchgear to the external environment, distinguish conduction heat flow channels, convection heat flow channels and radiation heat flow channels, and determine the heat source and diffusion law of each hot spot area; By combining the spatial distribution data of the temperature field corresponding to the temperature distribution cloud map, spatial differentiation is performed on each location inside the box to calculate the magnitude and direction of the temperature gradient at each location, thus quantifying the degree of temperature change in each region. Identify heat transfer paths, heat convergence areas, and temperature-sensitive areas in hot spots: Identify the heat transfer paths in hot spots by continuously connected heat flow lines; identify heat convergence areas by areas where heat flow lines intersect, temperature peaks are concentrated, and heat flux density is higher than the average heat flux density of the chamber; and identify temperature-sensitive areas by areas where the absolute value of the temperature gradient is higher than the absolute value of the average temperature gradient of the chamber.

4. The method for indirect measurement of hot spot temperature in a prefabricated substation according to claim 1, characterized in that: In S3, the joint screening strategy based on heat flow streamlines and temperature gradient specifically involves using heat transfer paths, heat accumulation areas, and temperature change sensitive areas as the scope, and using heat flow streamline distribution, heat flux density, and temperature gradient as the basis to screen a set of candidate temperature measurement points that are located on the heat transfer path, have a heat flux density not lower than the average heat flux density of the box, and are located in the temperature change sensitive area, with an absolute temperature gradient not lower than the absolute temperature gradient of the box. Based on four evaluation indicators—temperature gradient characteristics, heat flux intensity characteristics, path distance characteristics, and correlation with hotspot temperatures—a comprehensive evaluation function for candidate temperature measurement points is constructed. The comprehensive evaluation value is then calculated for each selected candidate temperature measurement point. The comprehensive evaluation function is as follows: ； In the formula, This is the comprehensive evaluation value of the i-th candidate temperature measurement point; Let be the temperature gradient feature of the i-th candidate temperature measurement point; The average temperature gradient inside the chamber; Let be the heat flux intensity characteristic quantity of the i-th candidate temperature measurement point; The average heat flux density inside the box; The distance from the i-th candidate temperature measurement point to the hot spot along the heat flow path is denoted as . This is the maximum heat flow path length inside the enclosure; Let be the Pearson correlation coefficient between the temperature sequence of the i-th candidate temperature measurement point and the temperature sequence of the hot spot. , , , They are respectively , , , Weighting coefficients; The preset relevance requirement is, It is greater than or equal to the preset correlation coefficient threshold.

5. The method for indirect measurement of hot spot temperature in a prefabricated substation according to claim 4, characterized in that: For the i-th candidate temperature measurement point, construct intermediate variables corresponding to the temperature gradient characteristics, heat flux intensity characteristics, path distance characteristics, and their correlation with the hot spot temperature. , , , : ； The weight coefficients are determined according to the normalization method: ； In the formula, This represents the weight coefficient corresponding to the k-th evaluation indicator; This is the intermediate variable corresponding to the k-th evaluation indicator.

6. The method for indirect measurement of hot spot temperature in a prefabricated substation according to claim 1, characterized in that, In S4, the operating parameters include load current I and ambient temperature. In addition to external operating parameters, the number of temperature sensors deployed is n, and the temperature sequence of characteristic temperature measurement points is as follows: , This represents the temperature of the nth feature temperature measurement point; The intermediate characteristic temperature inversion model is as follows: ； In the formula, This is the intermediate characteristic temperature; This is an environmental compensation item; The inversion function of the intermediate characteristic temperature inversion model is constructed using a regression model.

7. The method for indirect measurement of hot spot temperature in a prefabricated substation according to claim 6, characterized in that, In S5, the hotspot temperature inversion model is expressed as follows: ； In the formula, For hotspot temperature; Let m be the intermediate feature temperature corresponding to the z-th intermediate feature position, and m be the number of intermediate feature temperatures. Let be the equivalent thermal resistance from the z-th intermediate feature location to the hot spot; The baseline intercept term; The coefficient for the square term of the load current; This is the coefficient for the ambient temperature term; For environmental compensation items, the coefficient is... The weighting coefficient for the z-th intermediate characteristic temperature; for The weighting coefficients.

8. The method for indirect measurement of hot spot temperature in a prefabricated substation according to claim 7, characterized in that, based on Calculate the temperature rise : ； In the formula, Ambient temperature; This includes one or more of the following parameters: solar radiation intensity, ventilation status, and radiative heat transfer parameters of the enclosure, used to reduce the influence of the external environment on the temperature measurement results; Calculate the rate of change of temperature rise : ； In the formula, For reference temperature rise; Calculate the rate of temperature rise : ； In the formula, t represents time.

9. The method for indirect measurement of hot spot temperature in a prefabricated substation according to claim 1, characterized in that, In S5, the graded early warning criterion is specifically set as follows: a first threshold and a second threshold are set, and the second threshold is greater than the first threshold. When either the rate of change of temperature rise or the rate of temperature rise exceeds the first threshold, it is determined to be an abnormal hot spot temperature state. When both the rate of change of temperature rise and the rate of temperature rise exceed the second threshold, it is determined to be a severe overheating state, and early warning or protection measures are executed. The early warning or protection measures include issuing an alarm signal, starting ventilation or cooling devices, reducing load operation, or triggering protection trip.

10. The method for indirect measurement of hot spot temperature in a prefabricated substation according to claim 1, characterized in that, In S6, the hotspot temperature inversion model is adaptively corrected online using recursive least squares, Kalman filtering, or incremental learning methods.

Images