A transformer cooling control method, system, device and storage medium

Abstract

The present application relates to the technical field of transformer, and specifically relates to a transformer cooling control method, system, device and storage medium, comprising the following steps: obtaining electrical characteristic quantity signals and thermal state quantity signals of the transformer; based on the electrical characteristic quantity signals and the thermal state quantity signals, calculating a top layer oil temperature prediction value of the next time, and coupling a dynamically adjusted safety margin to determine an oil temperature threshold value, the safety margin being dynamically adjusted according to real-time electrical characteristic quantity signals; nonlinearly fusing the electrical characteristic quantity signals and the thermal state quantity signals to generate a composite control quantity; taking the oil temperature threshold value as a safety benchmark, and executing a hierarchical cooling strategy according to the composite control quantity. The present application solves the problems of lag, isolation and rigidity of the traditional transformer cooling control scheme by fusing electrical and thermal quantities, predicting oil temperature, dynamically adjusting the safety margin to determine the oil temperature threshold value, and nonlinearly generating the composite control quantity to realize hierarchical cooling, thereby improving the timeliness, accuracy and adaptability of the cooling control.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of transformers, in particular to a transformer cooling control method, system, device and storage medium. BACKGROUND

[0002] Power transformers are the core equipment for voltage transformation, power transmission and distribution in power systems, and their operation reliability is directly related to the safety and stability of the power grid. Among various types of transformers, oil-immersed transformers are widely used in power transmission and distribution networks due to their excellent insulation performance, high heat dissipation efficiency, and moderate maintenance cost, and they dominate in medium and high voltage power systems.

[0003] As the core equipment in power systems, the operating temperature of oil-immersed transformers directly affects the insulation performance and service life. Traditional cooling control methods generally rely on a single temperature threshold (such as a fixed 55℃ trigger for cooling), which is difficult to adapt to dynamic load changes in complex power grid environments, and has the following significant defects: (1) Response lag problem: the transformer oil temperature change has thermal inertia, and only triggering cooling by real-time top oil temperature signal cannot predict rapid temperature rise caused by sudden load changes (such as short circuit fault, large equipment start-stop); for example, when the high-voltage side current suddenly increases, the winding loss increases sharply, but the oil temperature rises with a delay, and traditional control may cause local overheating due to cooling start lag, accelerating insulation aging; (2) Signal isolation limitation: traditional strategies only rely on thermal state variables, ignoring the strong correlation between electrical characteristic signals and heating processes; for example, at the same oil temperature, the winding hot spot temperature under high load is much higher than that under low load, and relying only on oil temperature signals may cause the cooling strategy to deviate from the actual heating risk; (3) Threshold rigidity: fixed thresholds cannot adapt to environmental temperature fluctuations and load characteristic differences; in low temperature environments, fixed thresholds may cause excessive cooling, increasing energy consumption; while in high temperature and high load scenarios, fixed thresholds may be too high, leading to insufficient cooling and potential safety hazards; in addition, different types of transformers have different heat capacities and heat dissipation efficiencies, and a unified threshold cannot balance universality and precision; (4) Single control strategy: traditional cooling systems mostly use binary control mode of "all on / all off", which cannot dynamically adjust cooling intensity according to real-time operating conditions; for example, during slow load increase, all cooling devices may be fully engaged due to reaching the threshold, causing energy waste; while during sudden load increase, the temperature may be out of control due to insufficient cooling capacity.

[0004] In summary, traditional transformer cooling control methods have obvious shortcomings in safety, economy and adaptability, and an intelligent control scheme that integrates multi-dimensional signals and has predictive ability and dynamic adjustment characteristics is urgently needed. SUMMARY

[0005] The application aims to provide a transformer cooling control method, system, device and storage medium to solve the problems of response lag, signal isolation and threshold rigidity in traditional oil-immersed transformer cooling control.

[0006] The application is implemented by the following technical scheme: a transformer cooling control method, comprising the following steps:

[0007] Obtaining an electrical characteristic signal and a thermal state signal of an oil-immersed transformer;

[0008] Based on the electrical characteristic signal and the thermal state signal, a top oil temperature prediction value at the next time is calculated, and an oil temperature threshold is determined by coupling a dynamically adjusted safety margin, and the safety margin is dynamically adjusted according to the real-time electrical characteristic signal;

[0009] The electrical characteristic signal and the thermal state signal are nonlinearly fused to generate a composite control quantity;

[0010] Taking the oil temperature threshold as a safety reference, a hierarchical cooling strategy is executed according to the composite control quantity.

[0011] According to a preferred embodiment, the electrical characteristic signal at least includes a high-voltage side CT current signal and a current change rate, and the thermal state signal at least includes a top oil temperature signal and an ambient temperature signal of the oil-immersed transformer.

[0012] According to a preferred embodiment, the top oil temperature prediction value is obtained by using a pre-constructed LSTM prediction model based on the electrical characteristic signal and the thermal state signal.

[0013] According to a preferred embodiment, the method further comprises pre-processing the obtained electrical characteristic signal and thermal state signal, and the pre-processing includes normalization processing and sliding average filtering processing.

[0014] According to a preferred embodiment, the dynamic adjustment rule of the safety margin is:

[0015] When the pre-processed current change rate is greater than a preset load change threshold, the safety margin is reduced;

[0016] When the pre-processed high-voltage side CT current signal is less than a preset current threshold, the safety margin is increased.

[0017] According to a preferred embodiment, the calculation expression of the composite control quantity is , wherein, is the composite control quantity, and is a dynamic weight coefficient, increases with the high-voltage side CT current signal nonlinearly increases, The temperature difference between the top oil temperature signal and the ambient temperature signal. denoted as the rate of change of current.

[0018] According to a preferred embodiment, the staged cooling strategy is specifically as follows:

[0019] when When the value is greater than or equal to the first preset value, the cooling actuator operates in strong cooling mode;

[0020] when When the value is less than the first preset value and greater than the second preset value, the cooling actuator operates in balanced mode;

[0021] when When the value is less than or equal to the second preset value, the cooling actuator operates in energy-saving mode.

[0022] The present invention also provides a transformer cooling control system, applied to the transformer cooling control method described above, the system comprising:

[0023] The data acquisition module is used to acquire electrical characteristic signals and thermal state signals of the oil-immersed transformer.

[0024] The first processing module is used to calculate the predicted value of the top oil temperature at the next moment based on the electrical characteristic signal and the thermal state signal, and to determine the oil temperature threshold by coupling a dynamically adjusted safety margin. The safety margin is dynamically adjusted according to the real-time electrical characteristic signal.

[0025] The second processing module is used to nonlinearly fuse electrical characteristic signals and thermal state signals to generate composite control quantities.

[0026] The execution module is used to execute a graded cooling strategy based on the composite control quantity, with the oil temperature threshold as a safety benchmark.

[0027] The present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the transformer cooling control method described above.

[0028] The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the transformer cooling control method described above.

[0029] The technical scheme of the transformer cooling control method, system, device and storage medium provided by the application has at least the following advantages and beneficial effects: (1) The application obtains high-voltage side CT current signals, current change rate and other electrical characteristic quantity signals, combines top oil temperature, ambient temperature and other thermal state quantities, and uses an LSTM prediction model to calculate the top oil temperature prediction value at the next moment in advance, thereby breaking the hysteresis limitation of the traditional method of relying only on real-time temperature signals; at the same time, the load mutation trend can be perceived through the current change rate, and the cooling strategy can be adjusted in advance when the temperature has not yet significantly increased, thereby effectively avoiding the cooling lag caused by thermal inertia, improving the timeliness and predictability of the cooling control, and reducing the risk of insulation overheating; (2) The safety margin is dynamically adjusted according to real-time electrical characteristic quantity signals, and when the current change rate is greater than a preset load change threshold (such as load surge), the safety margin is reduced to tighten the oil temperature threshold and start strong cooling in advance; when the high-voltage side current is less than a preset current threshold (such as light load operation), the safety margin is increased to avoid excessive cooling, and this dynamic adjustment mechanism not only ensures the safety redundancy in high-load and high-risk scenarios, but also reduces unnecessary cooling energy consumption in low-load conditions, balances safety and economy, and realizes dynamic adaptation and precise protection of the safety margin; (3) The application generates a composite control quantity by nonlinearly fusing electrical characteristic quantity signals and thermal state quantities, overcoming the limitations of traditional single temperature signals; in the composite control quantity, the current square term can reflect the load loss heating intensity, the temperature difference term can reflect the actual heat dissipation pressure, and the current change rate term can capture the load dynamic trend, and the three work together to make the control strategy more suitable for the actual operation condition of the transformer; at the same time, the hierarchical cooling strategy can be flexibly switched according to the composite control quantity, adapt to different load intensity and environmental conditions, and avoid the energy waste or insufficient cooling problem of the traditional "all-on / all-off" mode, thereby enhancing the comprehensiveness and adaptability of the control strategy; (4) The signal noise interference is reduced through preprocessing, and the accuracy of the input data can be ensured. BRIEF DESCRIPTION OF DRAWINGS

[0030] Figure 1 A flowchart of a transformer cooling control method provided for the embodiment 1 of the application is shown in the figure.

[0031] Figure 2 A double signal coupling and hierarchical control principle diagram provided for the embodiment 1 of the application is shown in the figure. DETAILED DESCRIPTION

[0032] To make the purpose, technical scheme and advantages of the embodiments of the application clearer, the technical scheme in the embodiments of the application will be described clearly and completely below with reference to the drawings in the embodiments of the application. Obviously, the described embodiments are part of the embodiments of the application, rather than all the embodiments of the application. The components of the embodiments of the application described and shown in the drawings herein can be arranged and designed in various different configurations.

[0033] Embodiment 1

[0034] The embodiment provides a transformer cooling control method, Figure 1 For the flowchart of the transformer cooling control method, see Figure 1 The transformer cooling control method comprises the following steps:

[0035] Signal acquisition: obtaining electrical characteristic quantity signals and thermal state quantity signals of the oil-immersed transformer; in some preferred embodiments, the electrical characteristic quantity signals at least include high-voltage side CT current signals and current change rates, and the thermal state quantity signals at least include top layer oil temperature signals and ambient temperature signals of the oil-immersed transformer; in addition, other signals can be used instead of the high-voltage side CT current signals and the top layer oil temperature signals, such as vibration signals, noise analysis or infrared temperature monitoring, but whether these signals can accurately reflect the thermal state of the transformer and whether the real-time performance is sufficient need to be considered, which will not be described in detail here.

[0036] Signal preprocessing: in the embodiment, the obtained electrical characteristic quantity signals and thermal state quantity signals are preprocessed, and the preprocessing includes normalization processing and sliding average filtering processing. Specifically, the embodiment reduces signal noise interference through preprocessing, and can ensure the accuracy of input data.

[0037] Oil temperature prediction: in the embodiment, based on the electrical characteristic quantity signals and the thermal state quantity signals, a top layer oil temperature prediction value at the next moment is calculated, for example, a top layer oil temperature after 10 minutes is predicted; in some preferred embodiments, the top layer oil temperature prediction value is obtained based on the electrical characteristic quantity signals and the thermal state quantity signals by using a pre-constructed LSTM prediction model; in addition, other time series prediction models can be used instead, such as GRU, ARIMA or a simpler regression model; it is worth mentioning that a model-free control strategy can also be used, such as fuzzy control or PID control, but the parameters need to be adjusted dynamically, which will not be described in detail here.

[0038] Specifically, the embodiment obtains electrical characteristic quantity signals such as high-voltage side CT current signals and current change rates, combines thermal state quantity such as top layer oil temperature and ambient temperature, and uses an LSTM prediction model to calculate a top layer oil temperature prediction value at the next moment in advance, breaking the hysteresis limitation of traditional real-time temperature signals; at the same time, the current change rate can perceive the load mutation trend, so that the cooling strategy can be adjusted in advance when the temperature has not yet significantly increased, effectively avoiding the cooling lag caused by thermal inertia, improving the timeliness and predictability of the cooling control, and reducing the risk of insulation overheating.

[0039] Oil temperature threshold dynamic generation: In this embodiment, the oil temperature threshold is determined by coupling the top-level oil temperature prediction value with a dynamically adjusted safety margin, which is dynamically adjusted according to real-time electrical characteristic quantity signals; in addition, the oil temperature threshold can also be generated in other ways, such as based on the heat balance equation of the physical model, or using an expert system rule base, rather than a machine learning model, which will not be described in detail here.

[0040] In some preferred embodiments, the dynamic adjustment rule of the safety margin is: when the preprocessed current rate of change is greater than the preset load change threshold, the safety margin is reduced; when the preprocessed high-voltage side CT current signal is less than the preset current threshold, the safety margin is increased.

[0041] Specifically, the safety margin is dynamically adjusted according to real-time electrical characteristic quantity signals, and when the current rate of change is greater than the preset load change threshold (such as sudden load increase), the safety margin is reduced to tighten the oil temperature threshold and start the strong cooling in advance; when the high-voltage side current is less than the preset current threshold (such as light load operation), the safety margin is increased to avoid excessive cooling. This dynamic adjustment mechanism not only ensures safety redundancy in high-load and high-risk scenarios, but also reduces unnecessary cooling energy consumption in low-load conditions, balancing safety and economy, and achieving dynamic adaptation and precise protection of the safety margin.

[0042] Composite control quantity generation: Referring to FIG. 8, in this embodiment, the electrical characteristic quantity signals and the thermal state quantity signals are nonlinearly fused to generate a composite control quantity. Figure 2

[0043] In some preferred embodiments, the calculation expression of the composite control quantity is wherein, is the composite control quantity, and is a dynamic weight coefficient, increases with the high-voltage side CT current signal nonlinearly, is the temperature difference between the top-level oil temperature signal and the ambient temperature signal, is the current rate of change.

[0044] Specifically, this embodiment generates a composite control quantity by nonlinearly fusing electrical characteristic quantity signals and thermal state quantities, overcoming the limitations of traditional single temperature signals; in the composite control quantity, the current square term can reflect the load loss heating intensity, the temperature difference term reflects the actual heat dissipation pressure, and the current rate of change term captures the load dynamic trend, all of which work together to make the control strategy more suitable for the actual operation conditions of the transformer; at the same time, the staged cooling strategy can be flexibly switched according to the composite control quantity, adapting to different load intensities and environmental conditions, avoiding the energy waste or insufficient cooling problem of the traditional "all-on / all-off" mode, and enhancing the comprehensiveness and adaptability of the control strategy. ​

[0045] Performing cooling: in this embodiment, according to the composite control quantity, a hierarchical cooling strategy is performed based on the oil temperature threshold as a safety benchmark, and the control of the variable frequency oil pump, the multi-stage fan and the electric water valve is performed; in addition, other types of drivers or actuators can be used instead, such as solenoid valves, step motor controlled dampers, etc.

[0046] In some preferred embodiments, the hierarchical cooling strategy is specifically: When the oil temperature threshold is greater than or equal to the first preset value, the cooling actuator is operated in a strong cooling mode; when the oil temperature threshold is less than the first preset value and greater than the second preset value, the cooling actuator is operated in a balanced mode; when the oil temperature threshold is less than or equal to the second preset value, the cooling actuator is operated in an energy-saving mode. When the oil temperature threshold is greater than or equal to the first preset value, the cooling actuator is operated in a strong cooling mode; when the oil temperature threshold is less than the first preset value and greater than the second preset value, the cooling actuator is operated in a balanced mode; when the oil temperature threshold is less than or equal to the second preset value, the cooling actuator is operated in an energy-saving mode. When the oil temperature threshold is greater than or equal to the first preset value, the cooling actuator is operated in a strong cooling mode; when the oil temperature threshold is less than the first preset value and greater than the second preset value, the cooling actuator is operated in a balanced mode; when the oil temperature threshold is less than or equal to the second preset value, the cooling actuator is operated in an energy-saving mode.

[0047] Further, the transformer cooling control method provided in this embodiment further includes, when a sensor fault is detected or the prediction error of the LSTM prediction model is greater than a preset error threshold, switching to a backup control strategy and triggering a multi-stage alarm; wherein the backup control strategy is specifically to directly calculate the top oil temperature prediction value according to a preset temperature value and a high-voltage side CT current signal, wherein the preset temperature value is set to 65℃; the multi-stage alarm is specifically a local audible and visual alarm, a remote platform notification and an emergency shutdown; specifically, the combination of the multi-stage alarm and the backup strategy can effectively guarantee the robustness.

[0048] Embodiment 2

[0049] The embodiment provides a transformer cooling control system, which is applied to the transformer cooling control method of embodiment 1, and the system comprises a data acquisition module, a first processing module, a second processing module and an execution module.

[0050] The data acquisition module is configured to acquire the electrical characteristic quantity signal and the thermal state quantity signal of the oil-immersed transformer; the first processing module is configured to calculate the top oil temperature prediction value at the next time based on the electrical characteristic quantity signal and the thermal state quantity signal, and to determine the oil temperature threshold by coupling a dynamically adjusted safety margin, and the safety margin is dynamically adjusted according to the real-time electrical characteristic quantity signal; the second processing module is configured to perform nonlinear fusion on the electrical characteristic quantity signal and the thermal state quantity signal to generate a composite control quantity; and the execution module is configured to execute a hierarchical cooling strategy according to the composite control quantity based on the oil temperature threshold as a safety benchmark.

[0051] Embodiment 3

[0052] The embodiment provides an electronic device, which comprises a memory, a processor and a computer program stored in the memory and executable on the processor, and the processor implements the transformer cooling control method of embodiment 1 when executing the computer program.

[0053] Embodiment 4

[0054] The embodiment provides a computer readable storage medium, a computer program is stored on the computer readable storage medium, and the computer program is executed by a processor to implement the transformer cooling control method in the embodiment 1.

[0055] The above only is the preferred embodiment of the present application, and is not used to limit the present application, for the person skilled in the art, the present application can have various changes and changes. Any modification, equivalent replacement, improvement, etc. within the spirit and principle of the present application should be included in the protection scope of the present application.

Claims

1. A transformer cooling control method characterized by, The method comprises the following steps: obtaining electrical characteristic quantity signals and thermal state quantity signals of the oil-immersed transformer; using a pre-constructed LSTM prediction model to calculate a top-layer oil temperature prediction value at the next moment based on the electrical characteristic quantity signals and the thermal state quantity signals, and coupling a dynamically adjusted safety margin to determine an oil temperature threshold value, the safety margin being dynamically adjusted according to real-time electrical characteristic quantity signals; nonlinearly fusing the electrical characteristic quantity signals and the thermal state quantity signals to generate a composite control quantity; using the oil temperature threshold value as a safety benchmark, and executing a hierarchical cooling strategy according to the composite control quantity; The method further comprises: when a sensor fault is detected or an LSTM prediction model prediction error is greater than a preset error threshold, switching to a backup control strategy and triggering a multi-level alarm; wherein the backup control strategy specifically calculates a top-layer oil temperature prediction value according to a preset temperature value and a high-voltage side CT current signal.

2. The transformer cooling control method of claim 1, wherein The electrical characteristic quantity signals at least include a high-voltage side CT current signal and a current rate of change, and the thermal state quantity signals at least include a top-layer oil temperature signal and an ambient temperature signal of the oil-immersed transformer.

3. The transformer cooling control method of claim 2, wherein The method further comprises pre-processing the obtained electrical characteristic quantity signals and thermal state quantity signals, the pre-processing including normalization processing and sliding average filtering processing.

4. The transformer cooling control method of claim 3, wherein The dynamic adjustment rule of the safety margin is: when the pre-processed current rate of change is greater than a preset load change threshold, the safety margin is reduced; when the pre-processed high-voltage side CT current signal is less than a preset current threshold, the safety margin is increased.

5. The transformer cooling control method of claim 3, wherein The calculation expression of the compound control quantity is wherein, is the compound control quantity, and is a dynamic weight coefficient, with the high-voltage side CT current signal increasing the non-linear increment, is the temperature difference between the top layer oil temperature signal and the ambient temperature signal, is the current change rate.

6. The transformer cooling control method of claim 5, wherein The hierarchical cooling strategy specifically comprises: When greater than or equal to the first preset value, the cooling actuator is operated in a strong cooling mode. When when less than the first preset value and greater than the second preset value, the cooling actuator is operated in an equalization mode; When The cooling actuator is operated in the energy saving mode when the temperature is less than or equal to a second preset value.

7. A transformer cooling control system applied to the transformer cooling control method according to any one of claims 1 to 6, characterized by a system including: a data acquisition module for obtaining electrical characteristic quantity signals and thermal state quantity signals of the oil-immersed transformer; a first processing module for calculating a top-layer oil temperature prediction value at the next moment based on the electrical characteristic quantity signals and the thermal state quantity signals, and coupling a dynamically adjusted safety margin to determine an oil temperature threshold value, the safety margin being dynamically adjusted according to real-time electrical characteristic quantity signals; a second processing module for nonlinearly fusing the electrical characteristic quantity signals and the thermal state quantity signals to generate a composite control quantity; an execution module for using the oil temperature threshold value as a safety benchmark, and executing a hierarchical cooling strategy according to the composite control quantity.

8. An electronic device comprising a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, The processor executes the computer program to implement the transformer cooling control method according to any one of claims 1 to 6.

9. A computer-readable storage medium, characterized in that, The computer program is stored on the computer-readable storage medium and is executed by the processor to implement the transformer cooling control method according to any one of claims 1 to 6.

Images