CN-122016055-A - Automatic overtemperature early warning method for substation equipment based on thermal imaging data

CN122016055ACN 122016055 ACN122016055 ACN 122016055ACN-122016055-A

Abstract

The invention relates to the technical field of monitoring and early warning of power equipment, in particular to a substation equipment overtemperature automatic early warning method based on thermal imaging data, which comprises the steps of collecting environmental flow field data and equipment thermal imaging data, and constructing a flow-thermal state vector comprising a fluid vector and an observation view angle; calculating effective hysteresis wind speed reflecting thermal hysteresis effect by utilizing time domain convolution, calculating fluid scouring anisotropy factor reflecting the relative relation between wind direction and observation visual angle, calculating fluid thermal impedance compensation gain by combining fluid state index, performing linear inversion on surface observation temperature by utilizing the compensation gain, reconstructing zero wind equivalent core temperature and performing phase space early warning. According to the invention, by introducing the intermediate calculation index fluid thermal impedance compensation gain, the accurate numerical compensation of the differential cooling effect of the windward side and the leeward side is realized, and the detection precision of the equipment overheat fault under complex weather is effectively improved.

Inventors

LU YONGFENG
SONG QINGLAN
LU YONGLI
YU FEIYAN

Assignees

山东山开电力有限公司

Dates

Publication Date: 20260512
Application Date: 20260416

Claims (10)

1. The automatic overtemperature early warning method for the substation equipment based on the thermal imaging data is characterized by comprising the following steps of: acquiring real-time ambient wind speed, wind direction azimuth, optical axis horizontal azimuth, ambient air temperature and infrared image sequences, and extracting surface observation temperature to construct a flow-heat state vector; Calculating an effective hysteresis wind speed reflecting thermal inertia by using a time domain convolution algorithm based on the flow-thermal state vector, and calculating a relative pneumatic attack angle reflecting the relative relation between an observation position and a flow field; Determining a fluid scouring anisotropy factor based on the relative aerodynamic angle of attack, calculating a fluid thermal impedance compensation gain in combination with the effective hysteretic wind speed; Correcting the difference value between the surface observed temperature and the ambient air temperature by using the fluid thermal impedance compensation gain, and superposing the inherent thermal resistance deviation correction amount of the system to obtain zero wind equivalent core temperature; And when the three-dimensional characteristic point formed by the effective lagged wind speed, the surface observation temperature and the zero wind equivalent core temperature falls into a hidden dangerous area, generating a first-stage overheat alarm signal.
2. The thermal imaging data-based substation equipment overtemperature automatic early warning method according to claim 1, wherein the constructing a flow-thermal state vector comprises: Processing the infrared image sequence by using a region-of-interest pooling algorithm to obtain a target surface observation temperature; And combining the real-time ambient wind speed, the wind direction azimuth angle, the optical axis horizontal azimuth angle, the surface observation temperature and the ambient air temperature at the same sampling moment to generate the flow-heat state vector at the moment.
3. The automatic pre-warning method for overtemperature of substation equipment based on thermal imaging data according to claim 1, wherein the relation formula for calculating the effective hysteresis wind speed reflecting thermal inertia is: In the formula, Represent the first The effective lag wind speed at the moment; representing a total number of time steps for historical backtracking; A time step index representing historical backtracking; representing a sampling time interval; Representing the thermal time constant of the device; representing the normalization coefficient; Represent the first The first time before Real-time ambient wind speed at each sampling instant.
4. The method for automatically warning of overtemperature of substation equipment based on thermal imaging data according to claim 1, wherein said calculating a relative aerodynamic attack angle comprises: calculating a difference between the azimuth angle of the wind direction and the horizontal azimuth angle of the optical axis; Acquiring an absolute value of the difference value, and determining the absolute value as the relative aerodynamic attack angle; The relative aerodynamic angle of attack is used to characterize the degree of deviation of the line of sight of the observation from the direction of incoming fluid flow.
5. The automatic pre-warning method for overtemperature of transformer substation equipment based on thermal imaging data according to claim 1, wherein the determined fluid scouring anisotropy factor satisfies the following relation: In the formula, Represent the first A fluid scouring anisotropy factor at time; representing the wake zone cooling attenuation coefficient, wherein the value of the cooling attenuation coefficient is equal to the ratio of the leeward Noval number to the windward Noval number; Represent the first Said relative aerodynamic angle of attack at a moment in time; representing a streamlined shape index.
6. The automatic pre-warning method for overtemperature of transformer substation equipment based on thermal imaging data according to claim 1, wherein the calculated fluid thermal impedance compensation gain satisfies the following relation: In the formula, Represent the first The fluid thermal impedance at the moment compensates the gain; Represent the first The fluid washout anisotropy factor at time; Represent the first The effective lag wind speed at time; Representing a sensitivity coefficient; Representing a flow index; representing a half-saturation constant for preventing zero denominator and adjusting a linear region range of the gain curve; representing the maximum physically allowable gain constant, Representing the baseline reference wind speed.
7. The automatic overtemperature early warning method for substation equipment based on thermal imaging data according to claim 1, wherein the obtaining of the zero wind equivalent core temperature comprises the following steps: Calculating the difference between the surface observed temperature and the ambient air temperature to obtain a surface temperature rise value; Calculating the product of the fluid thermal impedance compensation gain and the surface temperature rise value to obtain the compensated equivalent temperature rise; And calculating the accumulated sum of the compensated equivalent temperature rise, the ambient air temperature and the system inherent thermal resistance deviation correction quantity, and determining the accumulated sum as the zero wind equivalent core temperature.
8. The method for automatically warning about overtemperature of substation equipment based on thermal imaging data according to claim 1, wherein the generating a primary overtemperature alarm signal comprises: Constructing three-dimensional data points comprising the effective hysteresis wind speed, the difference between the surface observed temperature and the ambient air temperature, and the difference between the zero wind equivalent core temperature and the ambient air temperature; Mapping the three-dimensional data points into an early warning phase space; and if the three-dimensional data points fall into a preset dangerous area of the high internal heat low table Wen Yinbi in the early warning phase space, triggering the primary overheat alarm signal.
9. The automatic overtemperature early warning method for substation equipment based on thermal imaging data according to claim 1, wherein the method for determining the intrinsic thermal resistance deviation correction amount of the system comprises the following steps: Continuously collecting the surface observation temperature and the ambient air temperature in a preset period of no solar radiation and cold shutdown of the equipment; calculating an average value of the difference between the surface observed temperature and the ambient air temperature in the period; judging whether the average value is smaller than zero, if yes, determining the absolute value of the average value as the system inherent thermal resistance deviation correction quantity, and if not, setting the system inherent thermal resistance deviation correction quantity as zero.
10. The automatic overtemperature early warning method for substation equipment based on thermal imaging data according to claim 3, wherein the method for determining the thermal time constant of the equipment comprises the following steps: Under a windless constant temperature environment, step power excitation is applied to equipment of the same type, and a surface temperature rise curve is recorded; performing first-order inertia link fitting on the surface temperature rise curve; And extracting the corresponding time length when the surface temperature rise curve reaches the preset proportion of the steady state value, and determining the time length as the equipment thermal time constant.

Description

Automatic overtemperature early warning method for substation equipment based on thermal imaging data Technical Field The invention relates to the technical field of monitoring and early warning of power equipment. More particularly, the invention relates to an automatic overtemperature early warning method for substation equipment based on thermal imaging data. Background The infrared thermal imaging technology is widely applied to overheat fault inspection of high-voltage equipment of a transformer substation due to the characteristics of non-contact, long-distance and visualization. The existing infrared temperature measurement method is generally based on isotropic static radiation assumption, namely, the temperature distribution of the surface of the equipment is considered to be only dependent on the difference value between the internal heating power and the external environment air temperature, and the natural cooling condition around the default equipment is uniform. Based on this assumption, conventional thermometry models often build a simple linear mapping relationship by reading the surface temperature acquired by the infrared camera and attempting to introduce the ambient wind speed as a single correction variable, attempting to restore the true temperature by subtracting the temperature drop due to wind speed. However, in the microscopic physical environment in which the substation is actually operating, the above assumption has serious logic drawbacks. When strong crosswind exists in an outdoor environment, namely under the working condition of a high Reynolds number flow field, a complex asymmetric flow field is formed around cylindrical or prismatic power equipment such as a lightning arrester, a transformer porcelain sleeve and a transformer sleeve. Specifically, on the windward side of fluid impact, the fluid boundary layer is strongly compressed, the convective heat transfer coefficient is extremely high, so that the temperature of the side surface is forced to be reduced to be close to the ambient air temperature, and at the moment, if the temperature is observed only by an infrared image, the real overheating defect in the interior is easily covered, and the missing report is caused. In contrast, on the lee surface where the fluid deviates, the boundary layer separation of the fluid occurs and forms karman vortex streets, so that the local flow velocity suddenly drops or even flows back, the heat exchange coefficient is obviously lower than the average level, heat is easy to stay and accumulate in the area, and at the moment, if a correction formula based on the average wind speed is directly applied, the correction amount is excessively large, so that the core temperature with a virtual height is calculated, and false alarm is caused. In addition, large oil-filled devices have a large heat capacity, and the change in surface temperature thereof has a significant time-lag effect with respect to fluctuations in ambient wind speed. The current instantaneous wind speed can only reflect the cooling capacity at that moment, whereas the current surface temperature is actually the cumulative result of the continued action of wind speed over the past period of time. The prior art often directly uses the current instantaneous wind speed to correct the current surface temperature, ignoring the causal dislocation in the time dimension. The fundamental disadvantage of the prior art is therefore that the vector flow thermal coupling field with extremely strong directivity is calculated and described with isotropic scalar operators, i.e. single wind speed values that are not directional. The space-time mismatch of the calculation model and the physical entity is the root cause of the failure of the refined early warning under the complex meteorological conditions. How to construct a temperature inversion method capable of identifying the relative relation between an observation view angle and a wind direction and processing thermal inertia hysteresis becomes a technical problem to be solved in the field. Disclosure of Invention In order to solve the problem of temperature correction errors caused by neglecting flow field anisotropy and thermal inertia under strong crosswind working conditions in the prior art, the invention provides a substation equipment overtemperature automatic early warning method based on thermal imaging data, which comprises the following steps: acquiring real-time ambient wind speed, wind direction azimuth, optical axis horizontal azimuth, ambient air temperature and infrared image sequences, and extracting surface observation temperature to construct a flow-heat state vector; Calculating an effective hysteresis wind speed reflecting thermal inertia by using a time domain convolution algorithm based on the flow-thermal state vector, and calculating a relative pneumatic attack angle reflecting the relative relation between an observation position and a flow field; Determining