Transformer oil circulation intelligent cooling method and system

Abstract

The invention provides a transformer oil circulation intelligent cooling method and system. Wherein the dynamic correlation model is constructed by detecting temperature and flow velocity data of different depths in the transformer, and periodic direction deflection control is applied in an oil flow path to form spiral rotation disturbance so as to trigger local energy exchange. And based on the predicted oil temperature evolution trend, the power intensity and disturbance amplitude parameters of the oil pump are synchronously adjusted, and dynamic balance between the flowing state and temperature distribution is established. And in combination with disturbance trajectory offset characteristics, dynamically matching frequency and phase parameters of direction deflection, so that the heat conduction direction and the main stream direction form directional cooperation. And model parameters are updated by utilizing heat conduction path characteristics and dynamic balance constraints, and a control instruction for inhibiting temperature sudden change is generated through reverse constraints, so that intelligent cooling regulation and control are realized. According to the technical scheme, oil flow is regulated and controlled through cooperation of spiral disturbance and the dynamic model, sudden temperature change is restrained, heat conduction is optimized, and the cooling efficiency and stability of the transformer are improved.

Description

Technical Field

[0001] The present application relates to the technical field of transformer cooling control, and in particular to a transformer oil circulation intelligent cooling method and system. Background Art

[0002] As power system load density increases, the heat losses generated by high-capacity transformers during operation increase significantly. Traditional oil circulation cooling systems face the risk of uneven oil temperature distribution, accelerated insulation aging caused by local overheating, and uncontrolled temperature rise at hot spots. The heat dissipation lag caused by a single oil flow path is particularly prominent in scenarios with new energy grid integration and impact loads. An intelligent cooling method is urgently needed that can dynamically adjust the oil flow disturbance pattern and synchronously match the evolution of oil temperature distribution to improve the synergy between heat conduction direction and oil flow path.

[0003] The current mainstream solution is a closed-loop control system based on a fixed-frequency oil pump drive and temperature threshold feedback. This solution uses temperature sensors placed in the transformer winding area to collect local temperature rise data. Combined with a preset oil pump speed-temperature mapping table, this solution uses a PID algorithm to adjust the pump's power output in real time. Static guide vanes are placed in the oil circuit to generate basic eddy currents. This system attempts to balance the oil temperature distribution through periodic speed increases and decreases. Its core control logic relies on a combination of linear flow rate regulation and a fixed guide structure.

[0004] The existing solution has three key defects. The direction and intensity of the eddy currents generated by the static guide vanes cannot be dynamically adjusted, resulting in insufficient spatial matching between the heat conduction direction and the mainstream oil path, and energy retention in local areas; the oil pump power regulation is only triggered based on the temperature threshold, lacks the feedforward prediction capability of the oil temperature distribution evolution trend, and cannot actively generate inhibitory disturbances before the temperature suddenly changes; the PID parameter solidification causes the system response to lag behind the dynamic coupling process of the oil flow and temperature field, especially in the nonlinear scenario where the oil viscosity changes with temperature, which is easy to cause control oscillation, but aggravates the local temperature rise fluctuations. Summary of the Invention

[0005] The present application provides a transformer oil circulation intelligent cooling method and system to solve the problems of solidification of oil disturbance direction and insufficient coordination of heat conduction paths in the prior art.

[0006] In a first aspect, the present application provides a transformer oil circulation intelligent cooling method, comprising:

[0007] By detecting the temperature distribution and flow velocity at different depths inside the transformer, a distribution model reflecting the dynamic relationship between temperature and flow velocity is generated;

[0008] Applying periodic directional deflection control to the oil flow path to form an oil rotation disturbance with spiral motion characteristics, and triggering local energy exchange during the flow process through the oil rotation disturbance;

[0009] Based on the mapping relationship between the temperature change trend and the flow velocity in the distribution model, the evolution direction of the oil temperature distribution is predicted, and the oil delivery dynamic intensity parameter and the disturbance amplitude parameter of the directional deflection control are simultaneously regulated to form a dynamic equilibrium relationship between the oil flow state and the temperature distribution;

[0010] Dynamically matching the frequency change mode and phase offset mode of the directional deflection control according to the trajectory deviation characteristics generated during the oil rotation disturbance process, so that the heat conduction direction of the local energy exchange and the oil flow direction form a directional cooperative relationship;

[0011] Based on the heat conduction path distribution characteristics of the directional cooperative relationship and the parameter constraints in the dynamic balance relationship, the dynamic correlation parameters in the distribution model are updated, and through the inverse constraint relationship between the disturbance amplitude parameter and the dynamic intensity parameter, a dynamic balance control instruction set that suppresses sudden changes in temperature distribution is generated to realize an intelligent cooling method for transformer oil circulation.

[0012] Optionally, based on the mapping relationship between the temperature change trend and the flow rate in the distribution model, the oil temperature distribution evolution direction is predicted, and the oil delivery power intensity parameter and the disturbance amplitude parameter of the direction deflection control are synchronously regulated to form a dynamic equilibrium relationship between the oil flow state and the temperature distribution, including:

[0013] Extracting data on the corresponding mapping relationship between the temperature change rate and the oil flow rate from the distribution model to determine the dominant trend of the oil temperature distribution evolution direction;

[0014] According to the expansion range of the dominant trend, an adjustment range of the oil delivery power intensity parameter and a variation range of the disturbance amplitude parameter of the direction deflection control are established;

[0015] By detecting the dynamic difference between the oil flow rate and the temperature distribution in real time, the compensation ratio of the power intensity parameter and the disturbance amplitude parameter is calculated, and the power intensity parameter value and the disturbance amplitude parameter value of the direction deflection control are synchronously adjusted according to the compensation ratio, so that the dynamic difference between the oil flow rate and the temperature distribution is maintained within the preset threshold, thereby forming a dynamic equilibrium relationship.

[0016] Optionally, applying periodic directional deflection control to the oil flow path to form an oil rotation disturbance with spiral motion characteristics, and triggering local energy exchange during the flow process through the oil rotation disturbance, including:

[0017] A plurality of flow guide components are arranged in the oil flow path, and the control unit drives the flow guide components to deflect in a preset periodic direction, so that the oil flow direction changes periodically and alternately;

[0018] According to the deflection angle and period parameters of the guide component, a spiral trajectory deviation feature of the oil flow path is generated, so that the oil generates a rotational disturbance during the spiral motion;

[0019] Through the centrifugal effect of the rotational disturbance, the oil is caused to interpenetrate between regions of different radii on the spiral track, triggering local energy exchange between the high-temperature region and the low-temperature region inside the oil.

[0020] Optionally, based on the trajectory deviation characteristics generated during the oil rotation disturbance, the frequency change mode and the phase deviation mode of the directional deflection control are dynamically matched so that the heat conduction direction of the local energy exchange and the oil flow direction form a directional cooperative relationship, including:

[0021] Extracting the maximum offset and offset direction from the spiral trajectory offset characteristics generated during the oil rotation disturbance process, and generating an adjustment coefficient for the frequency change mode of the directional deflection control;

[0022] Based on the difference between the maximum offset and a preset offset threshold, determining an adjustment range of the phase offset mode so as to form a preset phase difference between the deflection actions of different flow guide components;

[0023] By superimposing the phase difference and the frequency adjustment coefficient, the rotation period and centrifugal strength of the spiral motion of the oil are adjusted, so that the angle between the heat conduction direction inside the oil and the oil flow direction approaches the preset angle range, forming a directional cooperative relationship.

[0024] Optionally, updating the dynamic correlation parameters in the distribution model based on the heat conduction path distribution characteristics of the directional cooperative relationship and the parameter constraints in the dynamic equilibrium relationship includes:

[0025] According to the energy transfer direction and transfer rate in the heat conduction path distribution characteristics of the directional cooperative relationship, modifying the associated weight parameters of temperature and flow rate in the distribution model;

[0026] In combination with the constraint range of the dynamic intensity parameter and the disturbance amplitude parameter in the dynamic balance relationship, the correction amplitude of the associated weight parameter is limited;

[0027] The corrected association weight parameters are substituted into the distribution model, and the mapping relationship between the temperature change trend and the flow velocity is recalculated to complete the update of the dynamic association parameters.

[0028] Optionally, a dynamic balance control instruction set for suppressing sudden changes in temperature distribution is generated through an inverse constraint relationship between the disturbance amplitude parameter and the power intensity parameter, thereby realizing an intelligent cooling method for transformer oil circulation, including:

[0029] Establish an inverse correlation between the product of the disturbance amplitude parameter and the dynamic intensity parameter and the temperature distribution mutation threshold;

[0030] When the product exceeds the mutation threshold, a first type of instruction is generated to reduce the power intensity parameter or the disturbance amplitude parameter; when the product is lower than the mutation threshold, a second type of instruction is generated to increase the power intensity parameter or the disturbance amplitude parameter;

[0031] By combining the first and second types of instructions, a dynamic balance control instruction set is generated to control the dynamic difference between the oil flow rate and the temperature distribution to always be within the safe range of the mutation threshold.

[0032] Optionally, generating a spiral trajectory deviation feature of the oil flow path according to the deflection angle and period parameters of the flow guide component so as to generate a rotational disturbance in the oil during the spiral motion includes:

[0033] Setting an initial deflection angle of the guide component to a first angle value, and increasing or decreasing the deflection angles of adjacent guide components according to a preset incremental gradient based on a period parameter of the guide component to form a deflection angle sequence of multiple guide components;

[0034] According to the time interval of the cycle parameter, each guide component is controlled to switch to a target angle value in the deflection angle sequence at a corresponding time node, so that the oil flow direction forms a continuous directional offset between adjacent guide components;

[0035] Through the superposition of the continuous directional offsets, the oil flow path forms a continuous spiral trajectory offset feature in three-dimensional space;

[0036] The rotation radius change of the spiral trajectory offset feature is detected in real time. If the rotation radius change exceeds a preset range, the incremental gradient of the deflection angle sequence or the time interval of the period parameter is dynamically adjusted until the rotation radius of the spiral trajectory offset feature meets the preset condition.

[0037] In a second aspect, the present application provides a transformer oil circulation intelligent cooling system, comprising:

[0038] A detection module is used to detect the temperature distribution state and flow velocity state at different depths inside the transformer and generate a distribution model reflecting the dynamic relationship between temperature and flow velocity;

[0039] a control module for applying periodic directional deflection control to the oil flow path to form an oil rotation disturbance with spiral motion characteristics, and triggering local energy exchange during the flow process through the oil rotation disturbance;

[0040] a control module for predicting the direction of oil temperature distribution evolution based on the mapping relationship between temperature change trend and flow velocity in the distribution model, and synchronously controlling the oil delivery dynamic intensity parameter and the disturbance amplitude parameter of the directional deflection control, so as to form a dynamic equilibrium relationship between the oil flow state and the temperature distribution;

[0041] a matching module for dynamically matching the frequency change pattern and the phase offset pattern of the directional deflection control according to the trajectory offset characteristics generated during the oil rotation disturbance, so that the heat conduction direction of the local energy exchange forms a directional cooperative relationship with the oil flow direction;

[0042] A generation module is used to update the dynamic correlation parameters in the distribution model based on the heat conduction path distribution characteristics of the directional cooperative relationship and the parameter constraints in the dynamic balance relationship, and generate a dynamic balance control instruction set that suppresses sudden changes in temperature distribution through the inverse constraint relationship between the disturbance amplitude parameter and the dynamic intensity parameter, thereby realizing an intelligent cooling method for transformer oil circulation.

[0043] In a third aspect, an embodiment of the present application provides a computing device comprising a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a transformer oil circulation intelligent cooling method as described in the first aspect above.

[0044] In a fourth aspect, an embodiment of the present application provides a computer storage medium storing a computer program. When the computer program is executed by a computer, the transformer oil circulation intelligent cooling method as described in the first aspect is implemented.

[0045] In the embodiment of the present application, by real-time detection of the internal temperature and flow velocity of the transformer and construction of a dynamic correlation distribution model, the spatial coupling relationship between the oil temperature field and the flow velocity field can be accurately captured, providing multi-dimensional data support for predicting the temperature evolution direction; by applying periodic directional deflection control to generate spiral rotation disturbances, it is possible to break through the limitations of the static guide structure, trigger turbulence effects in the oil flow, and significantly enhance the local energy exchange efficiency; by synchronously regulating the dynamic intensity and disturbance amplitude parameters based on the temperature-velocity mapping relationship, it is possible to achieve a dynamic balance between the oil flow state and the temperature distribution, and suppress local temperature rise imbalance in advance; further combining the disturbance trajectory offset characteristics to dynamically match the frequency and phase parameters of the directional deflection, it is possible to optimize the directional coordination between the heat conduction direction and the main oil path, and reduce invalid energy retention; finally, by inversely constraining the dynamic parameters of the disturbance amplitude and dynamic intensity and iteratively updating the model, it is possible to generate a control instruction set that adaptively suppresses temperature mutations, improve the system's response accuracy and robustness to nonlinear working conditions, and thus comprehensively solve the temperature rise fluctuation problems caused by uneven oil temperature distribution, insufficient coordination of heat conduction paths, and control lag.

[0046] Furthermore, by extracting mapping data of temperature change rate and flow rate from the distribution model, the dominant trend of oil temperature evolution is determined; based on the trend expansion range, an adjustment range for dynamic intensity and disturbance amplitude is established; by real-time detection of the dynamic difference between flow rate and temperature distribution, the compensation ratio is calculated and the parameters are adjusted synchronously to maintain the difference within the threshold, forming a dynamic balance. This solution achieves coordinated decoupling control of dynamic intensity and disturbance amplitude through a dynamic difference feedback mechanism and compensation ratio calculation, overcoming the poor adaptability to nonlinear scenarios caused by the fixed parameters of traditional PID. At the same time, based on the interval prediction of the dominant trend, active compensation adjustment is made before the temperature distribution deviates from the threshold, avoiding temperature rise oscillations caused by hysteresis control.

[0047] These and other aspects of the present application will become more readily apparent from the description of the following embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without any creative work.

[0049] Figure 1 A flow chart of a transformer oil circulation intelligent cooling method provided by the present application is shown;

[0050] Figure 2 A scenario diagram showing a transformer oil circulation intelligent cooling method provided by the present application is shown;

[0051] Figure 3 A schematic diagram of the structure of a transformer oil circulation intelligent cooling system provided by the present application is shown;

[0052] Figure 4 A schematic structural diagram of a computing device provided by the present application is shown. DETAILED DESCRIPTION

[0053] In order to enable people skilled in the art to better understand the solution of the present application, the technical solution in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application.

[0054] In some of the processes described in the specification and claims of this application and the above-mentioned figures, multiple operations that appear in a specific order are included, but it should be clearly understood that these operations may not be executed in the order in which they appear in this document or may be executed in parallel. The serial numbers of the operations, such as 101, 102, etc., are only used to distinguish between different operations, and the serial numbers themselves do not represent any order of execution. In addition, these processes may include more or fewer operations, and these operations may be executed in sequence or in parallel. It should be noted that the descriptions of "first", "second", etc. in this document are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to being different types.

[0055] Research has found that in the field of transformer oil circulation cooling, the current mainstream solution is based on a closed-loop control strategy using a fixed-frequency oil pump drive and temperature threshold feedback. Its core relies on static guide vanes to generate basic eddy currents and linear PID parameter adjustment. However, this solution has inherent flaws: the direction and intensity of the eddy currents generated by the static guide vanes cannot be dynamically adjusted, resulting in low efficiency in the coordination between the heat conduction direction and the main oil flow path, and the problem of energy retention in local areas is difficult to eliminate; the temperature threshold relies on a feedback trigger mechanism, lacking the ability to predict the dynamic evolution trend of the oil temperature distribution and unable to actively intervene before a temperature change occurs; and the rigidity of PID parameters makes the system poorly adaptable in nonlinear scenarios where oil viscosity varies with temperature, prone to regulation lag and oscillation, and in turn exacerbating local temperature fluctuations. These issues collectively restrict the response speed and thermal balance performance of the transformer cooling system.

[0056] In response to the above-mentioned defects, the present invention proposes an intelligent cooling method for transformer oil circulation based on spiral disturbance control and temperature-flow rate dynamic coupling. The method constructs a real-time temperature-flow rate dynamic correlation model to accurately predict the evolution direction of the oil temperature distribution, and combines a programmable direction deflection device to impose periodic spiral disturbances on the oil flow path to generate a multi-dimensional rotating energy exchange path, replacing the fixed eddy current pattern of the traditional static guide vane; further, the control instructions are dynamically generated through the inverse constraint relationship between the disturbance amplitude parameter and the dynamic intensity parameter to achieve active coordinated balance between the oil flow state and the temperature field. This scheme solves the problem of insufficient coordination of the heat conduction path through adaptive control of the direction of the spiral disturbance, eliminates the hysteresis of the temperature threshold feedback through the feedforward prediction mechanism based on the dynamic model, and improves the control robustness under nonlinear working conditions through the adaptive iteration of the inverse constraint parameters, ultimately achieving effective suppression of sudden changes in the oil temperature distribution and significant optimization of the cooling efficiency.

[0057] The following will be combined with the drawings in the embodiments of this application to clearly and completely describe the technical solutions in the embodiments of this application. Obviously, the embodiments described are only part of the embodiments of this application, not all of the embodiments. Based on the embodiments in this application, all other embodiments obtained by those skilled in the art without making creative efforts are within the scope of protection of this application.

[0058] Figure 1 A flowchart of a transformer oil circulation intelligent cooling method is provided for an embodiment of the present application. Figure 1 As shown, the method includes:

[0059] 101. By detecting the temperature distribution state and flow velocity state at different depths inside the transformer, a distribution model reflecting the dynamic relationship between temperature and flow velocity is generated;

[0060] In the above scheme, the temperature distribution state refers to the set of temperature values ​​at different spatial locations within the transformer, collected in real time by distributed fiber optic sensors or thermocouple arrays, and is used to characterize the spatial heterogeneity of the oil thermal field. The flow velocity state is a set of oil velocity vectors in the oil channel, obtained by an electromagnetic flowmeter or a pressure gradient inversion algorithm, reflecting the dynamic characteristics of the oil flow. The distribution model refers to a coupled relationship model between temperature and flow velocity in the spatiotemporal dimensions, established using a long-short-term memory network. It predicts the evolution trend of the oil temperature distribution by analyzing the nonlinear mapping relationship between temperature gradient and flow velocity gradient.

[0061] In an embodiment of the present application, fiber Bragg grating temperature sensors and electromagnetic flow rate sensors are deployed at key positions of transformer windings, cores and oil channels to form a three-dimensional monitoring network, and the original data of temperature and flow rate are periodically and synchronously collected; secondly, the Kalman filter algorithm is used to denoise the data, and a continuous temperature field and flow rate field distribution map is generated through the Kriging spatial interpolation algorithm to eliminate the data blind spots caused by the sparsity of the sensor layout; then, the long short-term memory network (LSTM) is trained using historical data to construct a distribution model of nonlinear dynamic association between temperature gradient and flow rate gradient, the temperature-flow rate field data of the time series is input, and the predicted value of the oil temperature distribution evolution direction in the future time step is output; finally, the real-time collected data is input into the distribution model, and the network weight parameters are dynamically updated through the back propagation algorithm to make the model adaptive to nonlinear working conditions such as changes in oil viscosity.

[0062] In practical applications, for example, a 220 kV oil-immersed transformer has 24 fiber-optic temperature sensors and six electromagnetic flow meters installed in its oil channels. During one operation, the sensors detected an abnormal temperature gradient at the oil channel entrance, reaching 6°C / m, while the corresponding flow velocity was only 0.8 m / s. The LSTM model predicted that the temperature in this area would rise to 85°C within 20 seconds. The system immediately triggered the subsequent spiral perturbation control step, adjusting the guide vane deflection angle to +12° and increasing the oil pump power to 110% of its rated power. Ultimately, within 15 seconds, the temperature gradient was reduced to 2°C / m, and the flow velocity was restored to 1.5 m / s.

[0063] 102. Applying periodic directional deflection control to the oil flow path to form an oil rotation disturbance with spiral motion characteristics, and triggering local energy exchange during the flow process through the oil rotation disturbance;

[0064] Optionally, step 102 may specifically include the following steps:

[0065] 1021. Arrange a plurality of flow-guiding components in the oil flow path, and drive the flow-guiding components to deflect in a preset periodic direction by a control unit, so that the oil flow direction changes periodically and alternately;

[0066] 1022. Generate a spiral trajectory offset feature of the oil flow path according to the deflection angle and period parameters of the guide component, so that the oil generates a rotational disturbance during the spiral motion;

[0067] Among them, step 1022 may specifically include the following processes: setting the initial deflection angle of the guide component to a first angle value, and increasing or decreasing the deflection angles of adjacent guide components according to a preset incremental gradient based on the periodic parameter of the guide component to form a deflection angle sequence of multiple guide components; according to the time interval of the periodic parameter, controlling each guide component to switch to the target angle value in the deflection angle sequence at the corresponding time node, so that the oil flow direction forms a continuous directional offset between adjacent guide components; through the superposition of the continuous directional offsets, the oil flow path forms a continuous spiral trajectory offset feature in the three-dimensional space; real-time detection of the rotation radius change of the spiral trajectory offset feature, if the rotation radius change exceeds the preset range, dynamically adjusting the incremental gradient of the deflection angle sequence or the time interval of the periodic parameter until the rotation radius of the spiral trajectory offset feature meets the preset condition.

[0068] 1023. Through the centrifugal effect of the rotational disturbance, the oil is caused to interpenetrate between different radius areas of the spiral trajectory, triggering local energy exchange between the high-temperature area and the low-temperature area inside the oil.

[0069] In the above scheme, the guide component refers to an adjustable guide vane installed in the transformer oil channel, which is driven by a servo motor and is used to change the direction of oil flow; the periodic parameters include the deflection frequency of the guide vane, such as 0.5 to 2 Hz, and the phase difference, that is, the time interval between the deflection actions of adjacent guide vanes; the spiral trajectory offset feature refers to the three-dimensional spiral flow path formed by the periodic deflection of the guide vane, and its rotation radius is related to the incremental gradient of the deflection angle; the rotation radius variation is the radius fluctuation value of the spiral motion trajectory of the oil monitored in real time by particle image velocimetry technology, which is used for feedback adjustment of the guide vane parameters.

[0070] In this embodiment of the present application, multiple sets of adjustable guide vanes are first installed in the transformer oil channel at step 1021, with each set spaced 0.5 meters apart. These vanes are driven by a servo motor or hydraulic actuator for periodic deflection. A control unit sends a command to the actuator based on a preset initial deflection angle, i.e., a first angle value, such as +10°, and a periodic parameter, such as a time interval Δt = 2 seconds. At t = 0, the guide vanes deflect +10°, shifting the oil flow to the left. At t = 2 seconds, the guide vanes switch to -10°, shifting the oil flow to the right, thereby creating a periodic alternating flow. Then, step 1022 is used to generate a spiral trajectory offset feature of the oil flow path: the initial deflection angle of the guide component is set to a first angle value, such as +8°, ​​and the deflection angles of adjacent guide components are adjusted according to a preset incremental gradient, such as Δθ = 2° / piece, based on the periodic parameter of the guide component to generate a deflection angle sequence of adjacent guide blades. For example, the deflection angles of guide blades 1 to n are +8°, ​​+10°, and +12°, respectively, forming an increasing deflection angle sequence; and each guide component is controlled in stages to switch to a target angle value in the deflection angle sequence at a corresponding time node according to a time interval Δt = 1 second of the periodic parameter. Guide blade 1 switches to +8° at t = 0. , guide vane 2 switches to +10° at t=1 second, and guide vane 3 switches to +12° at t=2 seconds. The deflection directions of adjacent guide vanes are continuously offset by time-sharing switching, and the oil flow path is superimposed as a continuous spiral trajectory offset feature in three-dimensional space; the rotation radius R of the spiral trajectory offset feature is detected in real time using particle image velocimetry PIV or an electromagnetic flowmeter. If it is detected that the rotation radius R exceeds a preset range, such as R>0.3 meters, the incremental gradient Δθ is dynamically adjusted through the fuzzy control algorithm, for example, from 2° to 1.5°, or the time interval Δt is shortened, for example, from 1 second to 0.8 seconds, until the rotation radius converges to the target value, such as R=0.25 meters. Finally, in step 1023, the centrifugal effect of the rotational disturbance is utilized to force the oil to interpenetrate between different radius areas of the spiral trajectory. For example, the high-temperature oil in the 85°C area near the winding diffuses toward the periphery, while the low-temperature oil such as the 70°C area at the edge of the oil channel penetrates toward the center, thereby achieving forced mixing of high and low temperature oils and local energy exchange. For example, the oil temperature in the high-temperature area drops from 85°C to 78°C, and the oil temperature in the low-temperature area rises from 70°C to 75°C.

[0071] In practical applications, consider the thermal management of a 220kV oil-immersed transformer under sudden load conditions. When a temperature gradient in the middle of the transformer's oil channel rises abnormally to 5°C / m due to localized eddy currents, the system initiates a spiral perturbation control process. First, six sets of adjustable guide vanes are installed axially within the oil channel, spaced 0.5 meters apart. These vanes are linked to a PID controller via a servo motor. The PID controller's proportional coefficient, Kp, is set to 1.2, the integral time, Ti, is set to 0.8 seconds, and the differential time, Td, is set to 0.1 seconds to ensure rapid response when switching the guide vane angle. The system sets an initial deflection angle, θ0, of +10°, with a switching period, Δt, of 2 seconds. At t = 0, the guide vanes rotate +10°, forcing the oil flow to the left. Two seconds later, the deflection switches to -10° in the opposite direction, creating a periodic alternation of directions. Subsequently, based on the guide vane angle sequence generation rule, adjacent guide vanes were incremented by a gradient of Δθ = 2°, generating an angle sequence of +10°, +12°, and +14°. The angles were switched at intervals of Δt = 1 second: the first vane switched to +10° at t = 0, the second to +12° at t = 1 second, and so on, superimposing each other to form a spiral trajectory. Real-time monitoring using particle image velocimetry (PIV) revealed that the spiral radius R reached 0.4 meters, exceeding the safety threshold of 0.3 meters. At this point, the fuzzy controller, based on the spiral radius deviation ΔR = R measured - R target and the rate of change of the deviation dΔR / dt, triggered a dynamic adjustment strategy: Δθ was reduced from 2° to 1.5°, while Δt was shortened to 1.5 seconds, converging the spiral radius to 0.28 meters. During this process, the high-temperature oil was centrifugally thrown toward the edge of the oil channel, mixing with the cooler oil at the periphery. Simultaneously, the cooler oil permeated into the hotter central region. Distributed fiber optic temperature sensors verified that the high-temperature zone dropped from 82°C to 78°C, while the low-temperature zone rose from 75°C to 77°C, narrowing the temperature difference to 1°C / m. Ultimately, the system reduced the local temperature gradient from 5°C / m to 2°C / m within 15 seconds, evenly distributing the oil temperature and effectively preventing the risk of overheating of the insulation material.

[0072] This solution generates controllable spiral disturbances by periodically deflecting the flow guide components. This forcibly disrupts the oil's thermal boundary layer, enhancing dynamic mixing between high and low temperature zones, significantly improving heat dissipation efficiency and mitigating the risk of localized overheating. Furthermore, a dynamic parameter adjustment mechanism based on real-time feedback ensures the stability and adaptability of the spiral trajectory, making it suitable for thermal management requirements under complex transformer operating conditions.

[0073] 103. Based on the mapping relationship between the temperature change trend and the flow velocity in the distribution model, the oil temperature distribution evolution direction is predicted, and the oil delivery dynamic intensity parameter and the disturbance amplitude parameter of the directional deflection control are simultaneously regulated to form a dynamic equilibrium relationship between the oil flow state and the temperature distribution;

[0074] Optionally, step 103 may specifically include the following steps:

[0075] 1031. Extract data on the corresponding mapping relationship between the temperature change rate and the oil flow rate from the distribution model to determine the dominant trend of the oil temperature distribution evolution direction;

[0076] 1032. Establishing an adjustment range for an oil delivery power intensity parameter and a variation range for a disturbance amplitude parameter for directional deflection control based on the expansion range of the dominant trend;

[0077] 1033. By detecting the dynamic difference between the oil flow rate and the temperature distribution in real time, the compensation ratio of the power intensity parameter and the disturbance amplitude parameter is calculated, and the power intensity parameter value and the disturbance amplitude parameter value of the direction deflection control are synchronously adjusted according to the compensation ratio, so that the dynamic difference between the oil flow rate and the temperature distribution is maintained within a preset threshold, thereby forming a dynamic equilibrium relationship.

[0078] In the above scheme, the distribution model refers to a spatiotemporal coupling model of temperature and flow velocity established based on an LSTM network. It predicts the evolution trend of the oil temperature distribution by analyzing the nonlinear mapping relationship between temperature gradients and flow velocity gradients. The dominant trend refers to the core direction of the oil temperature distribution over time, such as the diffusion of high-temperature areas toward specific regions or the contraction of low-temperature areas. The oil delivery dynamic intensity parameter refers to the oil pump power or flow regulation parameters such as pump speed or valve opening, which directly affect the flow velocity. The disturbance amplitude parameter guides the deflection angle increment gradient or switching period of the flow component, determining the intensity and frequency of the spiral disturbance. The dynamic difference refers to the deviation between the real-time measured oil flow velocity and the predicted temperature distribution value.

[0079] In this embodiment, step 1031 first extracts data mapping the temperature change rate and oil flow rate from a distribution model for the spatiotemporal coupling of temperature and flow rate established based on an LSTM network. For example, if the distribution model output indicates a temperature change rate of 0.5 degrees Celsius per second at the oil channel inlet, the corresponding flow rate must be greater than or equal to 1.2 meters per second to suppress temperature rise. Key features are extracted from the LSTM weight matrix using a gradient descent method, and the Sobel operator is used to identify the direction of temperature distribution evolution. The dominant trend in the oil temperature distribution evolution is determined to be the diffusion of the high-temperature zone toward the center of the oil channel. Next, step 1032 establishes, based on the extension range of this dominant trend, such as a high-temperature zone diffusion rate of 0.3 meters per second, an adjustment range for oil pump power or flow rate control parameters, such as pump speed or valve opening, and oil delivery dynamic intensity parameters that directly affect flow rate, such as an oil pump power adjustment range of 80% to 120%, as well as a variation range for disturbance amplitude parameters, such as a guide vane deflection angle increment gradient of 1 to 3 degrees and a switching period of 0.5 to 2 seconds. Finally, step 1033 is used to detect the dynamic difference between the oil flow rate and the temperature distribution in real time. For example, the measured flow rate is 1.0 meters per second and the predicted demand value is 1.2 meters per second. The fuzzy PID controller is used to calculate the compensation ratio of the power intensity parameter and the disturbance amplitude parameter, and the oil pump power increment and deflection angle adjustment are output to stabilize the dynamic difference between the oil flow rate and the temperature distribution within the preset threshold. For example, the absolute value of the flow rate deviation is less than 0.1 meters per second and the temperature deviation is less than 1 degree Celsius, thereby achieving a dynamic balance between the oil flow state and the temperature distribution.

[0080] In practical applications, a sudden load increase in the middle of a 220 kV oil-immersed transformer's oil channel caused a temperature gradient of 4 degrees Celsius per meter. The LSTM distribution model predicted that the high-temperature zone would spread toward the top, with a temperature change rate of 0.4 degrees Celsius per second. This corresponded to a required flow rate of 1.3 meters per second, but the measured flow rate was only 1.0 meters per second. Based on the dominant trend of the oil temperature distribution's expansion rate of 0.25 meters per second, the oil pump power adjustment range was set between 90% and 130%, the guide vane deflection angle increment gradient range between 1.5 degrees and 3.5 degrees, and the switching cycle range between 0.6 seconds and 1.8 seconds. Upon real-time detection of a negative oil flow rate deviation of 0.3 meters per second, the fuzzy-PID controller increased the pump power output to 125%, increased the guide vane deflection angle increment gradient from 2 degrees to 2.8 degrees, and shortened the switching cycle from 1.5 seconds to 1.0 seconds. After regulation, the flow rate recovered to 1.25 meters per second, the temperature gradient dropped to 2.5 degrees Celsius per meter, and the flow rate deviation converged to negative 0.05 meters per second, reaching the dynamic equilibrium threshold.

[0081] This solution dynamically analyzes the coupled relationship between oil temperature distribution and flow velocity, accurately predicts the direction of temperature evolution, and simultaneously adjusts the oil delivery power and disturbance amplitude to achieve a dynamic balance between flow state and temperature distribution. Based on the model's prediction of dominant trends, it adaptively adjusts the oil pump power intensity and the disturbance parameters of the flow guide components, compensating for deviations in flow velocity and temperature in real time, effectively suppressing local overheating and improving heat dissipation uniformity. Through a closed-loop feedback control mechanism, the system can quickly respond to complex operating conditions such as changes in oil viscosity or load fluctuations, ensuring the coordinated optimization of the transformer's internal thermal and flow fields, significantly enhancing operational stability and thermal management reliability.

[0082] 104. Based on the trajectory deviation characteristics generated during the oil rotation disturbance, dynamically match the frequency change mode and phase deviation mode of the directional deflection control so that the heat conduction direction of the local energy exchange forms a directional cooperative relationship with the oil flow direction;

[0083] Optionally, step 104 may specifically include the following steps:

[0084] 1041. Extract the maximum offset and offset direction from the spiral trajectory offset characteristics generated during the oil rotation disturbance process, and generate an adjustment coefficient for the frequency change mode of the directional deflection control;

[0085] 1042. Determine an adjustment range of the phase shift mode based on a difference between the maximum shift amount and a preset shift threshold value, so that a preset phase difference is formed between the deflection actions of different flow guide components;

[0086] 1043. By combining the phase difference and the frequency adjustment coefficient, the rotation period and centrifugal strength of the oil spiral motion are adjusted so that the angle between the heat conduction direction inside the oil and the oil flow direction approaches the preset angle range, thus forming a directional cooperative relationship.

[0087] In the above scheme, the trajectory offset characteristics refer to the geometric parameters of the three-dimensional flow path formed by the oil during the spiral disturbance process, including the spiral radius, pitch and rotation direction. The frequency change mode refers to the frequency adjustment rule for switching the deflection direction of the guide component, for example, gradually increasing from a low frequency of 0.5Hz to a high frequency of 1Hz to enhance the disturbance intensity. The phase offset mode refers to the time difference control strategy for the deflection actions of adjacent guide components. For example, the deflection interval between guide vane 1 and guide vane 2 is Δt seconds, forming a phase difference. The maximum offset refers to the maximum spatial deviation distance of the central axis of the spiral trajectory relative to the axis of the oil channel. The directional synergy relationship refers to adjusting the spiral disturbance parameters to make the oil flow direction and the heat conduction direction, such as the heat transfer path from high temperature to low temperature, tend to be consistent, thereby improving the heat dissipation efficiency.

[0088] In an embodiment of the present application, first, the maximum offset and offset direction of the spiral trajectory offset feature generated during the oil rotation disturbance are extracted by using particle image velocimetry PIV or electromagnetic flowmeter through step 1041. For example, it is detected that the center axis of the spiral trajectory offset feature deviates 0.2 meters to the right from the axis of the oil channel, and the offset direction is clockwise. Based on the maximum offset and the preset threshold, such as the difference Δd=0.05 meters when the safety threshold is 0.15 meters, the frequency adjustment coefficient k_f=Δd / threshold of the frequency change mode of the directional deflection control is generated. If Δd>0, the deflection frequency of the guide component is increased according to the ratio of k_f; if Δd<0, the frequency is reduced. For example, when Δd=0.05 meters, k_f=0.05 / 0.15≈0.33, and the frequency is increased from 1Hz to 1.33Hz.

[0089] Next, in step 1042, the adjustment amplitude of the phase offset mode is determined based on the difference Δd between the maximum offset and the preset offset threshold. For example, when Δd = 0.05 meters exceeds the threshold of 0.02 meters, the phase difference between adjacent deflectors is set to Δt = 0.5 seconds, causing deflector blade 1 to deflect +10° at t = 0 and deflector blade 2 to deflect +12° at t = 0.5 seconds, forming a continuous perturbation controlled by phase difference. A Kalman filter algorithm is used to predict the offset trend, and fuzzy control rules are combined to dynamically adjust the phase difference, ensuring that the preset phase difference is achieved between the deflection actions of different deflectors.

[0090] Finally, in step 1043, the frequency adjustment coefficient k_f is superimposed on the phase difference Δt to control the rotation period T = 1 / f (f is the frequency) and the centrifugal strength F_c ∝ Rω of the oil spiral motion. 2 , R is the spiral radius, and ω is the angular velocity. For example, when the frequency is increased to 1.33 Hz and the phase difference Δt = 0.5 seconds, the spiral period is shortened to 0.75 seconds. The increased centrifugal force forces the angle between the heat conduction direction and the flow direction to decrease from 30° to 10°, bringing the angle between the heat conduction direction within the oil and the flow direction closer to the preset angle range, forming a directional synergistic relationship.

[0091] In practical applications, a 220 kV oil-immersed transformer's oil channel, partially blocked on the right side, caused its spiral trajectory to shift 0.25 meters to the right. Particle image velocimetry (PIV) detected that the maximum offset in the spiral trajectory offset signature generated during the oil rotation disturbance exceeded a preset threshold, generating a frequency adjustment coefficient and increasing the deflection frequency of the guide components from 1 Hz to 1.67 Hz. Based on the offset difference, the phase difference between adjacent guide components was set to 0.3 seconds: guide vane 1 deflected +12 degrees at t = 0 seconds, and guide vane 2 deflected +14 degrees at t = 0.3 seconds. After superimposing the frequency and phase adjustments, the rotation period of the oil's spiral motion was shortened to 0.6 seconds. The increased centrifugal strength reduced the angle between the heat conduction direction and the flow direction from 40 degrees to 15 degrees, allowing the high-temperature oil to diffuse rapidly along the flow direction. The temperature gradient on the right side decreased significantly within 20 seconds, bringing the angle between the heat conduction direction within the oil and the flow direction closer to the preset angle range, forming a directional synergistic relationship.

[0092] This solution dynamically matches frequency changes and phase offset patterns to precisely control the rotational period and centrifugal intensity of the oil's spiral motion, creating a directional synergy between the oil's heat conduction direction and its flow direction, significantly enhancing the efficiency of heat transfer from high-temperature areas to low-temperature areas. This method adaptively corrects flow path offsets, optimizes the consistency of the heat dissipation path, effectively suppresses localized temperature accumulation, and improves the transformer's overall thermal balance and operational reliability.

[0093] 105. Based on the heat conduction path distribution characteristics of the directional cooperative relationship and the parameter constraints in the dynamic balance relationship, the dynamic correlation parameters in the distribution model are updated, and through the inverse constraint relationship between the disturbance amplitude parameter and the dynamic intensity parameter, a dynamic balance control instruction set that suppresses sudden changes in temperature distribution is generated to realize an intelligent cooling method for transformer oil circulation.

[0094] Optionally, step 105 may specifically include the following steps:

[0095] 1051. Modify the associated weight parameters of temperature and flow velocity in the distribution model according to the energy transfer direction and transfer rate in the heat conduction path distribution characteristics of the directional cooperative relationship;

[0096] 1052. Based on the constraint ranges of the dynamic intensity parameter and the disturbance amplitude parameter in the dynamic balance relationship, the correction amplitude of the associated weight parameter is limited;

[0097] 1053. Substitute the corrected association weight parameter into the distribution model, recalculate the mapping relationship between the temperature change trend and the flow velocity, and complete the update of the dynamic association parameter.

[0098] 1054. Establish an inverse correlation between the product of the disturbance amplitude parameter and the dynamic intensity parameter and the temperature distribution mutation threshold;

[0099] 1055. When the product exceeds the mutation threshold, a first type of instruction is generated to reduce the power intensity parameter or the disturbance amplitude parameter; when the product is lower than the mutation threshold, a second type of instruction is generated to increase the power intensity parameter or the disturbance amplitude parameter;

[0100] 1056. By combining the first type of instructions with the second type of instructions, a dynamic balance control instruction set is generated to control the dynamic difference between the oil flow rate and the temperature distribution to always be within the safe range of the mutation threshold.

[0101] In the above scheme, the heat conduction path distribution characteristic refers to the directional heat transfer path from the high-temperature area of ​​the oil to the low-temperature area formed by spiral perturbations, which is determined by the synergistic effect of centrifugal force and flow direction. Dynamic correlation parameters refer to the weight matrix parameters in the LSTM model that characterize the nonlinear relationship between temperature and flow velocity. Inverse constraint relationships refer to dynamic strength parameters, such as the product of oil pump power and perturbation amplitude parameters such as the guide vane angle increment, which is negatively correlated with the temperature mutation threshold. That is, the larger the product, the lower the mutation risk. The dynamic balance control instruction set refers to a sequence of instructions generated based on real-time operating conditions to adjust oil pump power and guide vane parameters, used to maintain a dynamic balance between temperature and flow velocity. The mutation threshold refers to the critical condition for a sudden change in the oil temperature distribution, such as when the local temperature rise rate exceeds the safety limit.

[0102] In this embodiment, step 1051 first modifies the temperature-flow velocity association weight parameters in the LSTM distribution model based on the heat conduction path distribution characteristics of the directional synergy relationship, for example, the energy transfer rate from the high-temperature zone to the oil channel outlet is 0.2 degrees Celsius per second. A gradient descent method is used to calculate the gradient adjustment of the weight matrix of the associated weight parameters. If the energy transfer direction deviates from the model-predicted direction by more than a preset angle, the associated weight parameters are modified proportionally to the deviation. Next, step 1052 limits the modified associated weight parameters based on the constraints of the dynamic equilibrium relationship's power intensity parameter and the disturbance amplitude parameter, for example, a maximum pump power limit of 130% and a guide vane angle increment of no more than 3 degrees. A saturation function is used to force weight parameter values ​​that exceed the bounds to a safe range. Next, step 1053 substitutes the limited associated weight parameters into the LSTM distribution model, and the mapping between temperature change trends and flow velocity is recalculated. For example, after the update, the model predicts that the flow velocity requirement in the high-temperature zone will increase from 1.2 meters per second to 1.5 meters per second to match the actual heat conduction path.

[0103] Then, in step 1054, an inverse correlation is established between the product of the disturbance amplitude parameter and the dynamic intensity parameter and the temperature distribution mutation threshold. For example, the product is defined as the product of the guide vane angle increment gradient Δθ and the oil pump power P. When Δθ × P exceeds the preset mutation threshold, the system is considered stable; when it is below the threshold, a mutation risk is determined. In step 1055, an instruction is generated based on the comparison of this product with the mutation threshold: if the product exceeds the mutation threshold, a first-category instruction to reduce the oil pump power or reduce the guide vane angle increment is generated; if the product is below the mutation threshold, a second-category instruction to increase the oil pump power or angle increment is generated. For example, when Δθ = 2 degrees and P = 120%, the product is 240, exceeding the threshold of 200, triggering the first-category instruction to reduce P to 110%. Finally, step 1056 dynamically combines the combination of the first and second types of instructions into a dynamic balance control instruction set. For example, the oil pump power is adjusted to 110% at the same time while keeping the guide vane angle increment unchanged, or the power is maintained but the angle increment is increased to 2.5 degrees. The dynamic difference between the oil flow rate and the temperature distribution is controlled to always be within the safe range of the mutation threshold, thereby suppressing the risk of temperature mutation.

[0104] In practical applications, a thermal management scenario involving a 220 kV oil-immersed transformer operating during high summer temperatures is exemplified. One day, the ambient temperature rose to 38 degrees Celsius, and the transformer load factor suddenly increased to 95%. The temperature gradient in the middle of the oil channel rose from 2 degrees Celsius per meter to 5 degrees Celsius per meter within 10 minutes, with the local temperature rise rate reaching 0.6 degrees Celsius per second, approaching the insulation material's temperature limit. The system initiated a dynamic balancing control process. Distributed fiber optic temperature sensors detected that the heat conduction path in the high-temperature zone was shifting toward the top of the oil channel. The energy transfer direction deviated by 15 degrees from the direction predicted by the LSTM model, with a transfer rate of 0.5 degrees Celsius per second. Based on the energy transfer direction and rate of the heat conduction path distribution characteristics derived from the directional synergy relationship, the associated weight parameter of the distribution model was corrected from 0.85 to 0.78 using gradient descent. Subsequently, because the correction amplitude fell below the preset lower limit of 0.8, the dynamic limiting rule was triggered, forcing the value to a safe value of 0.8. The updated LSTM distribution model predicted an increase in the flow velocity requirement in the high-temperature zone from 1.3 meters per second to 1.6 meters per second, identifying the top oil channel as a critical heat dissipation path. The system recalculated the mapping between temperature trends and flow velocity, setting a threshold of 300 for the product of the disturbance amplitude parameter and the power intensity parameter. The real-time detection of the product of a 2.5-degree guide vane angle increment and 120% oil pump power was exactly 300, indicating that the system was in a critical state. Due to the continued rise in ambient temperature, the product was predicted to drop to 290 in 10 seconds, triggering a second type of instruction to increase the guide vane angle increment to 2.8 degrees while maintaining the oil pump power at 120%, bringing the product value to 336. After executing the command, the flow rate increased from 1.5 meters per second to 1.65 meters per second. The oil in the high-temperature area flowed rapidly along the top oil channel. Within 30 seconds, the temperature gradient dropped to 2.8 degrees Celsius per meter, and the temperature rise rate dropped to 0.3 degrees Celsius per second. The angle between the heat conduction direction and the flow direction was reduced from 25 degrees to 8 degrees. The dynamic difference between the oil flow rate and the temperature distribution was always controlled within the safe range of the mutation threshold, and the directional synergy effect was significant.

[0105] This solution dynamically updates model parameters and establishes inverse constraints to generate an adaptive control instruction set, precisely regulating the dynamic balance between oil flow rate and temperature distribution. The system mitigates the risk of sudden temperature changes in real time, maintains the consistency of heat conduction paths and flow directions, and enhances the stability and intelligence of transformer oil circulation cooling, adapting to the demands for efficient thermal management under complex operating conditions.

[0106] The following is a complete embodiment of steps 101 to 105:

[0107] Taking a 220 kV oil-immersed transformer as an example, when its load suddenly increases to near the rated maximum capacity during the high temperature period in summer, the ambient temperature rises significantly, and the temperature gradient in the middle of the oil channel rises from a low level to an abnormal value in a short period of time. The local temperature rise rate reaches the dangerous threshold, triggering the intelligent cooling system to start full-process regulation.

[0108] Multiple fiber Bragg grating temperature sensors and electromagnetic flowmeters are deployed at key locations within the windings, core, and oil channels, forming a high-density monitoring network. This network monitors the temperature distribution and flow velocity at various depths within the transformer. By integrating filtering and spatial interpolation algorithms, an LSTM distribution model dynamically correlates the continuous temperature and flow velocity fields. This model detects a significant temperature gradient anomaly at the oil channel entrance, corresponding to a flow velocity far below expectations. Based on historical data, the LSTM distribution model predicts that the oil temperature in the corresponding area will rapidly rise to a critical value, triggering subsequent control measures. Based on the mapping relationship between temperature trends and flow velocity in the LSTM distribution model, guide vanes within the oil channel apply periodic directional deflection control to the oil flow path at a preset initial deflection angle and period, switching the direction. This generates a spiraling oil perturbation. This perturbation triggers local energy exchange within the flow. Real-time detection indicates that the spiral radius of the perturbation exceeds a safe range. The guide vane angle increment gradient and switching period are dynamically adjusted to converge the spiral radius to the target range. The centrifugal effect drives the high-temperature oil to diffuse toward the periphery and the low-temperature oil to penetrate toward the center, rapidly reducing the temperature gradient to a safe level.

[0109] Based on the mapping relationship between temperature change trends and flow velocity in the distribution model, the LSTM distribution model identifies the diffusion trend of high-temperature zones. If the measured flow velocity falls below the predicted requirement, a dynamic adjustment range for the oil pump power and diversion parameters is set. Through real-time compensation control, the oil pump power is increased and the guide vane angle increment is increased to restore the flow velocity to a level that matches the heat dissipation requirements, thereby achieving a dynamic equilibrium between the oil flow state and the temperature distribution. After detecting the trajectory deviation characteristics generated during the oil rotational disturbance, the guide vane switching frequency is simultaneously increased and phase difference control is introduced to reduce the angle between the heat conduction direction and the flow direction, enhancing the diffusion efficiency of the high-temperature zone along the preset path and establishing a directional synergistic relationship between the heat conduction direction of the local energy exchange and the oil flow direction. Based on the heat conduction path distribution characteristics of this directional synergistic relationship and the parameter constraints in the dynamic equilibrium relationship, the dynamic correlation parameters in the distribution model are updated. Combined with the inverse constraints, a control instruction set is generated to maintain the synergistic effect between the oil pump power and diversion parameters, ensure dynamic equilibrium between the flow velocity and temperature distribution, suppress the temperature rise rate, and implement an intelligent cooling method for transformer oil circulation.

[0110] This solution uses a multi-physics field three-dimensional perception network to capture the spatiotemporal distribution characteristics of oil temperature and flow rate in real time, and combines the LSTM model to accurately predict the evolution trend of high-temperature areas, providing a data basis for dynamic control; generates spiral disturbances to break the oil thermal stratification structure, forcing high and low temperature oils to mix to reduce abnormal temperature gradients, and optimizes the heat conduction path through frequency-phase coordinated control to make the heat dissipation direction and flow direction close to each other; dynamically adjusts the oil pump power and diversion parameters based on the inverse constraint relationship to maintain the dynamic balance of flow rate and temperature distribution and suppress the risk of local temperature rise mutations; finally, through closed-loop intelligent control, parameter self-update and instruction set dynamic optimization are achieved to ensure the heat dissipation stability and operational reliability of the transformer under complex working conditions, significantly improve heat dissipation efficiency and extend equipment life, and provide full-process technical guarantees for safe operation in high-load scenarios.

[0111] Figure 3 The present invention provides a schematic diagram of a transformer oil circulation intelligent cooling system. Figure 3 As shown, the system includes:

[0112] The detection module 31 is used to generate a distribution model reflecting the dynamic relationship between temperature and flow velocity by detecting the temperature distribution state and flow velocity state at different depths inside the transformer;

[0113] a control module 32 for applying periodic directional deflection control to the oil flow path to form an oil rotation disturbance with spiral motion characteristics, and triggering local energy exchange during the flow process through the oil rotation disturbance;

[0114] a control module 33 for predicting the direction of oil temperature distribution evolution based on the mapping relationship between temperature change trend and flow velocity in the distribution model, and synchronously controlling the oil delivery power intensity parameter and the disturbance amplitude parameter of the directional deflection control to achieve a dynamic equilibrium relationship between the oil flow state and the temperature distribution;

[0115] a matching module 34 for dynamically matching the frequency variation pattern and the phase offset pattern of the directional deflection control according to the trajectory offset characteristics generated during the oil rotation disturbance, so that the heat conduction direction of the local energy exchange forms a directional cooperative relationship with the oil flow direction;

[0116] A generation module 35 is used to update the dynamic correlation parameters in the distribution model based on the heat conduction path distribution characteristics of the directional cooperative relationship and the parameter constraints in the dynamic balance relationship, and generate a dynamic balance control instruction set that suppresses sudden changes in temperature distribution through the inverse constraint relationship between the disturbance amplitude parameter and the dynamic intensity parameter, thereby realizing an intelligent cooling method for transformer oil circulation.

[0117] Figure 3 The transformer oil circulation intelligent cooling system can perform Figure 1 The implementation principles and technical effects of the transformer oil circulation intelligent cooling method described in the illustrated embodiment are not further elaborated. The specific manner in which the various modules and units of the transformer oil circulation intelligent cooling system in the above embodiment perform their operations has been described in detail in the relevant embodiments of the method and will not be further elaborated here.

[0118] In one possible design, Figure 3 The transformer oil circulation intelligent cooling system of the embodiment shown can be implemented as a computing device, such as Figure 4 As shown, the computing device may include a storage component 41 and a processing component 42;

[0119] The storage component 41 stores one or more computer instructions, wherein the one or more computer instructions are called and executed by the processing component 42 .

[0120] The processing component 42 is used for the above Figure 1 The embodiment provides a transformer oil circulation intelligent cooling method.

[0121] The processing component 42 may include one or more processors to execute computer instructions to perform all or part of the steps in the above method. Of course, the processing component may also be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above method.

[0122] The storage component 41 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile memory device, or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic memory, flash memory, magnetic disk, or optical disk.

[0123] Of course, a computing device may also include other components, such as input / output interfaces, display components, communication components, etc.

[0124] The input / output interface provides an interface between the processing component and the peripheral interface module, which can be an output device, an input device, etc.

[0125] The communication component is configured to facilitate, among other things, wired or wireless communications between the computing device and other devices.

[0126] Among them, the computing device can be a physical device or an elastic computing host provided by a cloud computing platform, etc. In this case, the computing device can refer to a cloud server, and the above-mentioned processing components, storage components, etc. can be basic server resources rented or purchased from the cloud computing platform.

[0127] The present application also provides a computer storage medium storing a computer program, wherein the computer program can achieve the above-mentioned Figure 1 The embodiment shown is a transformer oil circulation intelligent cooling method.

[0128] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0129] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate, and the components shown as units may or may not be physical units, i.e., they may be located in one location or distributed across multiple network units. Some or all of the modules may be selected based on actual needs to achieve the objectives of the present embodiment. Persons of ordinary skill in the art will be able to understand and implement the present invention without inventive effort.

[0130] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, or of course, by hardware. Based on this understanding, the essence of the above technical solution or the part that contributes to the existing technology can be embodied in the form of a software product. The computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, a magnetic disk, an optical disk, etc., and includes a number of instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or certain parts of the embodiments.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them. Although the present application has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A transformer oil circulation intelligent cooling method, characterized in that: include: By detecting the temperature distribution and flow velocity at different depths inside the transformer, a distribution model reflecting the dynamic relationship between temperature and flow velocity is generated; Applying periodic directional deflection control to the oil flow path to form an oil rotation disturbance with spiral motion characteristics, and triggering local energy exchange during the flow process through the oil rotation disturbance; Based on the mapping relationship between the temperature change trend and the flow velocity in the distribution model, the evolution direction of the oil temperature distribution is predicted, and the oil delivery dynamic intensity parameter and the disturbance amplitude parameter of the directional deflection control are simultaneously regulated to form a dynamic equilibrium relationship between the oil flow state and the temperature distribution; Dynamically matching the frequency change mode and phase offset mode of the directional deflection control according to the trajectory deviation characteristics generated during the oil rotation disturbance process, so that the heat conduction direction of the local energy exchange and the oil flow direction form a directional cooperative relationship; Based on the heat conduction path distribution characteristics of the directional cooperative relationship and the parameter constraints in the dynamic balance relationship, the dynamic correlation parameters in the distribution model are updated, and through the inverse constraint relationship between the disturbance amplitude parameter and the dynamic intensity parameter, a dynamic balance control instruction set that suppresses sudden changes in temperature distribution is generated to realize an intelligent cooling method for transformer oil circulation.

2. The method according to claim 1, characterized in that Based on the mapping relationship between the temperature change trend and the flow rate in the distribution model, the evolution direction of the oil temperature distribution is predicted, and the oil delivery dynamic intensity parameter and the disturbance amplitude parameter of the directional deflection control are simultaneously regulated to form a dynamic equilibrium relationship between the oil flow state and the temperature distribution, including: Extracting data on the corresponding mapping relationship between the temperature change rate and the oil flow rate from the distribution model to determine the dominant trend of the oil temperature distribution evolution direction; According to the expansion range of the dominant trend, an adjustment range of the oil delivery power intensity parameter and a variation range of the disturbance amplitude parameter of the direction deflection control are established; By detecting the dynamic difference between the oil flow rate and the temperature distribution in real time, the compensation ratio of the power intensity parameter and the disturbance amplitude parameter is calculated, and the power intensity parameter value and the disturbance amplitude parameter value of the direction deflection control are synchronously adjusted according to the compensation ratio, so that the dynamic difference between the oil flow rate and the temperature distribution is maintained within the preset threshold, thereby forming a dynamic equilibrium relationship.

3. The method according to claim 1, characterized in that Applying periodic directional deflection control to the oil flow path to form an oil rotation disturbance with spiral motion characteristics, and triggering local energy exchange during the flow process through the oil rotation disturbance, including: A plurality of flow guide components are arranged in the oil flow path, and the control unit drives the flow guide components to deflect in a preset periodic direction, so that the oil flow direction changes periodically and alternately; According to the deflection angle and period parameters of the guide component, a spiral trajectory deviation feature of the oil flow path is generated, so that the oil generates a rotational disturbance during the spiral motion; Through the centrifugal effect of the rotational disturbance, the oil is caused to interpenetrate between regions of different radii on the spiral track, triggering local energy exchange between the high-temperature region and the low-temperature region inside the oil.

4. The method according to claim 1, wherein Dynamically matching the frequency change mode and phase offset mode of the directional deflection control based on the trajectory deviation characteristics generated during the oil rotation disturbance process so that the heat conduction direction of the local energy exchange forms a directional cooperative relationship with the oil flow direction, including: Extracting the maximum offset and offset direction from the spiral trajectory offset characteristics generated during the oil rotation disturbance process, and generating an adjustment coefficient for the frequency change mode of the directional deflection control; Based on the difference between the maximum offset and a preset offset threshold, determining an adjustment range of the phase offset mode so as to form a preset phase difference between the deflection actions of different flow guide components; By superimposing the phase difference and the frequency adjustment coefficient, the rotation period and centrifugal strength of the spiral motion of the oil are adjusted, so that the angle between the heat conduction direction inside the oil and the oil flow direction approaches the preset angle range, forming a directional cooperative relationship.

5. The method according to claim 1, wherein Based on the heat conduction path distribution characteristics of the directional cooperative relationship and the parameter constraints in the dynamic equilibrium relationship, updating the dynamic correlation parameters in the distribution model includes: According to the energy transfer direction and transfer rate in the heat conduction path distribution characteristics of the directional cooperative relationship, modifying the associated weight parameters of temperature and flow rate in the distribution model; In combination with the constraint range of the dynamic intensity parameter and the disturbance amplitude parameter in the dynamic balance relationship, the correction amplitude of the associated weight parameter is limited; The corrected association weight parameters are substituted into the distribution model, and the mapping relationship between the temperature change trend and the flow velocity is recalculated to complete the update of the dynamic association parameters.

6. The method according to claim 1, wherein By means of the inverse constraint relationship between the disturbance amplitude parameter and the dynamic intensity parameter, a dynamic balance control instruction set for suppressing sudden changes in temperature distribution is generated, thereby realizing an intelligent cooling method for transformer oil circulation, including: Establish an inverse correlation between the product of the disturbance amplitude parameter and the dynamic intensity parameter and the temperature distribution mutation threshold; When the product exceeds the mutation threshold, a first type of instruction is generated to reduce the power intensity parameter or the disturbance amplitude parameter; when the product is lower than the mutation threshold, a second type of instruction is generated to increase the power intensity parameter or the disturbance amplitude parameter; By combining the first and second types of instructions, a dynamic balance control instruction set is generated to control the dynamic difference between the oil flow rate and the temperature distribution to always be within the safe range of the mutation threshold.

7. The method according to claim 3, characterized in that Generating a spiral trajectory deviation feature of the oil flow path according to the deflection angle and period parameters of the guide component so as to generate a rotational disturbance in the oil during the spiral motion, including: Setting an initial deflection angle of the guide component to a first angle value, and increasing or decreasing the deflection angles of adjacent guide components according to a preset incremental gradient based on a period parameter of the guide component to form a deflection angle sequence of multiple guide components; According to the time interval of the cycle parameter, each guide component is controlled to switch to a target angle value in the deflection angle sequence at a corresponding time node, so that the oil flow direction forms a continuous directional offset between adjacent guide components; Through the superposition of the continuous directional offsets, the oil flow path forms a continuous spiral trajectory offset feature in three-dimensional space; The rotation radius change of the spiral trajectory offset feature is detected in real time. If the rotation radius change exceeds a preset range, the incremental gradient of the deflection angle sequence or the time interval of the period parameter is dynamically adjusted until the rotation radius of the spiral trajectory offset feature meets the preset condition.

8. A transformer oil circulation intelligent cooling system, characterized in that: include: By detecting the temperature distribution and flow velocity at different depths inside the transformer, a distribution model reflecting the dynamic relationship between temperature and flow velocity is generated; Applying periodic directional deflection control to the oil flow path to form an oil rotation disturbance with spiral motion characteristics, and triggering local energy exchange during the flow process through the oil rotation disturbance; Based on the mapping relationship between the temperature change trend and the flow velocity in the distribution model, the evolution direction of the oil temperature distribution is predicted, and the oil delivery dynamic intensity parameter and the disturbance amplitude parameter of the directional deflection control are simultaneously regulated to form a dynamic equilibrium relationship between the oil flow state and the temperature distribution; Dynamically matching the frequency change mode and phase offset mode of the directional deflection control according to the trajectory deviation characteristics generated during the oil rotation disturbance process, so that the heat conduction direction of the local energy exchange and the oil flow direction form a directional cooperative relationship; Based on the heat conduction path distribution characteristics of the directional cooperative relationship and the parameter constraints in the dynamic balance relationship, the dynamic correlation parameters in the distribution model are updated, and through the inverse constraint relationship between the disturbance amplitude parameter and the dynamic intensity parameter, a dynamic balance control instruction set that suppresses sudden changes in temperature distribution is generated to realize an intelligent cooling method for transformer oil circulation.

9. A computing device, characterized in that It comprises a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are used to be called and executed by the processing component to implement a transformer oil circulation intelligent cooling method as described in any one of claims 1 to 7.

10. A computer storage medium, characterized in that A computer program is stored, and when the computer program is executed by a computer, the transformer oil circulation intelligent cooling method according to any one of claims 1 to 7 is implemented.

Images