Oil-immersed transformer dispatching and response system and method for power grid stability

Abstract

The application relates to the technical field of transformer scheduling, in particular to an oil-immersed transformer scheduling and response system and method for power grid stability, which comprises the following steps: synchronously collecting real-time operation state data and environmental meteorological data of a transformer; dynamically correcting a heat dissipation coefficient of a transformer thermal model based on the real-time air pressure; generating a winding hot spot temperature dynamic change track through the transformer thermal model; based on the dynamic change track, solving multi-dimensional dynamic thermal redundancy in parallel; taking the multi-dimensional dynamic thermal redundancy as a core constraint, performing rolling optimization scheduling through model predictive control, the rolling optimization scheduling taking maximization of total new energy consumption and minimization of transformer load fluctuation in a scheduling period as optimization objectives, generating a coordinated control instruction of the transformer and a new energy station; and executing the coordinated control instruction to dynamically adjust a power grid operation state. The application can realize collaborative improvement of transformer dynamic load safety and efficient new energy consumption.

Description

Technical Field

[0001] This application relates to the technical field of transformer dispatching, and in particular to an oil-immersed transformer dispatching and response system and method for grid stability. Background Technology

[0002] With the continuous increase in the proportion of renewable energy generation, the operating characteristics of power grids are becoming increasingly complex. This challenge is particularly prominent in high-altitude areas rich in natural energy resources but with harsh environments. For example, in the Qinghai-Tibet Plateau region, at an altitude of over 3,000 meters, power grid construction and stable operation face dual pressures: on the one hand, there is an urgent need for the development of local photovoltaic and wind power resources; on the other hand, extreme geographical and climatic conditions pose a serious threat to the long-term safe operation of critical power equipment. As a core device in the power grid that realizes power conversion and transmission, the operating status of oil-immersed transformers directly affects the reliability of the regional power grid.

[0003] Currently, power grid dispatching systems generally employ a control method based on fixed load thresholds to manage transformer operation. This method pre-sets the maximum allowable load current or overload curve of the transformer based on its design parameters under standard environmental conditions. In actual operation, when the monitored transformer load approaches or exceeds these static thresholds, the dispatching system issues alarms or executes control commands such as load shedding to prevent equipment overheating and damage. However, this traditional static threshold control method has inherent technical limitations when applied to special power grid scenarios involving high altitudes and high proportions of renewable energy integration, making it difficult to achieve a balance between ensuring equipment safety and improving operational efficiency. The specific reasons are as follows:

[0004] First, the heat dissipation efficiency of transformers is highly dependent on ambient atmospheric pressure; the low atmospheric pressure caused by increased altitude significantly weakens their convective heat dissipation capacity. However, existing models generally use fixed heat dissipation parameters under standard atmospheric pressure, which leads to a systematic overestimation of the actual heat dissipation capacity of transformers in high-altitude applications. This inherent bias at the model level renders the "fixed load threshold" set based on the model fundamentally inaccurate.

[0005] Then, in fluctuating scenarios with a high proportion of renewable energy integration, the aforementioned model bias further reveals a serious mismatch between the assessment dimensions and operational requirements. Photovoltaic and wind power output exhibits significant intermittency and randomness, requiring the dispatch system to accurately quantify the dynamic thermal load-bearing capacity of transformers at different time scales in the future: this includes the cumulative additional heat generation that can be withstood within the dispatch cycle, the instantaneous overload margin to cope with minute-level power surges, and the cumulative consumption of equipment insulation life due to operations. However, the static threshold method can only provide a single, isolated instantaneous load limit; its rigid assessment dimensions cannot characterize the aforementioned dynamic, multi-dimensional thermal safety boundaries, resulting in a lack of comprehensive and accurate basis for dispatch decisions.

[0006] Finally, the combined effect of static models and the limitations of single-dimensional approaches leads to rigidity and short-sightedness in control strategies. When the transformer's heat dissipation capacity is actually sufficient due to favorable environmental conditions, the fixed threshold prevents proactive efforts to increase the absorption of new energy sources, resulting in resource waste. Conversely, when its heat margin is quietly reduced due to low air pressure or load shocks, it is impossible to strengthen protection in advance, plunging the system into a dilemma where safety and efficiency are difficult to balance.

[0007] In summary, existing technologies exhibit a chain of technical defects, from model inaccuracy to evaluation failure, and then to control imbalance. The root cause lies in the inability of static models to respond to dynamic environments, which manifests in the inability of a single dimension to match diverse needs. The consequence is that rigid control cannot achieve dynamic optimization. Summary of the Invention

[0008] To achieve a synergistic improvement in both the dynamic load safety of transformers and the efficient absorption of new energy sources, this application provides an oil-immersed transformer dispatching and response system and method for grid stability.

[0009] Firstly, the oil-immersed transformer dispatching and response method for power grid stability provided in this application adopts the following technical solution: An oil-immersed transformer dispatching and response method for power grid stability includes the following steps:

[0010] The transformer's real-time operating status data and environmental meteorological data are collected synchronously. The real-time operating status data includes load current and top oil temperature, and the environmental meteorological data includes real-time air pressure and ambient temperature.

[0011] Based on the real-time air pressure dynamic correction transformer thermal model convective heat dissipation coefficient; using the top oil temperature as the initial state of the transformer thermal model, and driving the corrected transformer thermal model with the predicted load current sequence and ambient temperature sequence, the dynamic change trajectory of winding hot spot temperature in future scheduling cycles is generated.

[0012] Based on the dynamic change trajectory, the multidimensional dynamic thermal redundancy is calculated in parallel; the multidimensional dynamic thermal redundancy includes the cumulative energy redundancy used to evaluate the additional cumulative thermal load energy that the transformer can safely withstand in the future time window, the instantaneous peak redundancy used to evaluate the short-term overload capacity of the transformer, and the life credit redundancy used to quantify the equivalent normal aging life duration that can be consumed within the scheduling cycle.

[0013] Using the multidimensional dynamic thermal redundancy as the core constraint, rolling optimization scheduling is performed through model predictive control. The rolling optimization scheduling aims to maximize the total amount of new energy consumption and minimize the transformer load fluctuation within the scheduling period, and generates coordinated control commands for transformers and new energy power plants.

[0014] The coordinated control commands are executed to dynamically adjust the power grid operating status.

[0015] Optionally, the transformer thermal model is composed of the top oil temperature thermal balance equation and the winding hot spot temperature calculation equation.

[0016] The specific process for generating the dynamic change trajectory of winding hot spot temperature within the future scheduling cycle is as follows:

[0017] Calculate the corresponding total transformer loss sequence based on the predicted load current sequence;

[0018] Using the total transformer loss sequence and the predicted ambient temperature sequence as input, the top oil temperature thermal balance equation is solved by numerical integration to obtain the dynamic trajectory of the top oil temperature in the future scheduling cycle.

[0019] The dynamic trajectory of the top oil temperature and the load current sequence are input into the winding hot spot temperature calculation equation, and the corresponding dynamic change trajectory of the winding hot spot temperature is calculated and output point by point.

[0020] Optionally, the expression for the top oil temperature heat balance equation is:

[0021]

[0022] in, The heat capacity of transformer oil; Top oil temperature; For time; This represents the total transformer loss. The overall heat dissipation coefficient; The convective heat dissipation coefficient is dynamically corrected by real-time air pressure. Ambient temperature; This refers to the oil temperature index;

[0023] Total transformer losses The calculation formula is:

[0024]

[0025] in, This is the no-load loss; This is the load current; This is the winding resistance loss coefficient determined by the transformer design parameters.

[0026] Optionally, the expression for the winding hot spot temperature calculation equation is:

[0027]

[0028] in, This refers to the hot spot temperature of the winding. This refers to the steady-state temperature difference between the hot spot temperature of the winding and the top oil temperature when the transformer is running continuously at rated current. Rated current; This represents the winding index.

[0029] Optionally, the air pressure correction formula for the convective heat dissipation coefficient is:

[0030]

[0031] in, This represents the convective heat dissipation coefficient under standard atmospheric pressure; Indicates real-time air pressure; Indicates standard atmospheric pressure; This represents the experience index.

[0032] Optionally, the experience index The value range is from 0.5 to 0.8.

[0033] Optionally, the dynamic change trajectory is calculated using the following methods to determine the multidimensional dynamic thermal redundancy:

[0034] The cumulative energy redundancy is obtained by calculating the integral of the additional heat load energy that the transformer winding can withstand over time within the range where the dynamic change trajectory is lower than the preset maximum allowable hot spot temperature limit.

[0035] The instantaneous peak redundancy is obtained by extracting and analyzing the instantaneous temperature extreme values ​​of the dynamic change trajectory on a minute-level time scale and evaluating their margin with the transformer short-term overload safety threshold.

[0036] By inputting the dynamic change trajectory into the insulation thermal aging model, the equivalent normal aging lifetime duration that it will consume in future scheduling cycles is calculated, and the lifetime credit redundancy is obtained by comparing it with the preset cycle lifetime budget.

[0037] Optionally, during the rolling optimization scheduling process, constraint priorities are set for the multidimensional dynamic thermal redundancy: the safety margin requirement of the instantaneous peak redundancy is set as the highest priority constraint; the budget guarantee of the lifetime credit redundancy is set as the second highest priority constraint; and the non-negativity of the cumulative energy redundancy is set as the standard priority constraint.

[0038] Optionally, during the rolling optimization scheduling process, the system monitors the consumption status of the lifetime credit redundancy in real time; when it is determined that the real-time remaining amount of the lifetime credit redundancy is lower than a preset safety threshold, the weight coefficient of the lifetime credit redundancy in the optimization objective function is automatically increased, and the adjustment range of the weight coefficient is negatively correlated with the proportion of the remaining lifetime credit redundancy.

[0039] Secondly, the oil-immersed transformer dispatching and response system for power grid stability provided in this application adopts the following technical solution: The oil-immersed transformer dispatching and response system for power grid stability includes:

[0040] The data acquisition module is used to synchronously collect real-time operating status data of the transformer and environmental meteorological data;

[0041] The thermal model dynamic correction module is used to dynamically correct the convective heat dissipation coefficient in the transformer thermal model based on the real-time air pressure.

[0042] The temperature trajectory prediction module is used to drive the modified transformer thermal model with the current top oil temperature as the initial state to generate the dynamic change trajectory of the winding hot spot temperature in future scheduling cycles.

[0043] A parallel redundancy calculation module is used to calculate multi-dimensional dynamic thermal redundancy in parallel based on the dynamic change trajectory.

[0044] The optimization scheduling module is used to perform rolling optimization scheduling with the multi-dimensional dynamic hot redundancy as the core constraint, and generate coordinated control commands for transformers and new energy power plants.

[0045] The execution control module is used to execute the coordinated control commands to dynamically adjust the power grid operating status.

[0046] In summary, this application includes the following beneficial technical effects:

[0047] 1. This application solves the fundamental problem that traditional static thermal models systematically overestimate the actual heat dissipation capacity of transformers in high-altitude, low-pressure environments by introducing real-time air pressure data to dynamically correct the convective heat dissipation coefficient in the transformer thermal model. This significantly improves the calculation accuracy and adaptability of the thermal model under complex geographical and climatic conditions, and eliminates the safety assessment deviation caused by model inaccuracy from the source.

[0048] 2. This application constructs a multi-dimensional dynamic thermal safety assessment system by parallel calculation of cumulative energy redundancy, instantaneous peak redundancy and lifetime credit redundancy based on the dynamic change trajectory of winding hot spot temperature. This system overcomes the technical limitations of the traditional single fixed load threshold method, which cannot respond to the multi-timescale thermal load requirements under the background of new energy access. It realizes a comprehensive quantitative assessment of the transformer's short-term overload capacity, medium-term energy absorption and long-term lifetime consumption.

[0049] 3. This application embeds multidimensional dynamic thermal redundancy as a core constraint into the model predictive control framework for rolling optimization scheduling, realizing the coordinated optimization of the transformer dynamic thermal safety boundary and the power grid economic operation target. This not only ensures the operational safety of the equipment on the instantaneous, medium-term and long-term time scales, but also significantly improves the renewable energy consumption level and effectively smooths transformer load fluctuations, thus solving the contradiction between safety and efficiency that traditional rigid control strategies cannot balance.

[0050] 4. This application establishes a complete technology chain from data acquisition, model correction, temperature prediction, redundancy calculation to optimized scheduling and command execution, realizing real-time perception, dynamic prediction and precise control of transformer thermal state, effectively solving the chain of technical defects in the existing technology, from model inaccuracy to evaluation failure to control imbalance.

[0051] 5. This application innovatively introduces an adaptive weight adjustment mechanism for lifetime credit redundancy during the rolling optimization scheduling process. When lifetime resources are detected to be scarce, the priority of lifetime protection is automatically increased, realizing intelligent management and control of transformer insulation lifetime. This avoids overdrawing equipment life in pursuit of short-term benefits and significantly improves the long-term operational reliability of transformers. Attached Figure Description

[0052] Figure 1 This is an overall flowchart of the scheduling and response method according to an embodiment of this application;

[0053] Figure 2 This is a flowchart of step S1 of the scheduling and response method in an embodiment of this application;

[0054] Figure 3 This is a flowchart of step S2 of the scheduling and response method in an embodiment of this application;

[0055] Figure 4 This is a flowchart of step S3 of the scheduling and response method in an embodiment of this application;

[0056] Figure 5 This is a flowchart of step S4 of the scheduling and response method in an embodiment of this application;

[0057] Figure 6 This is a flowchart of step S5 of the scheduling and response method in an embodiment of this application;

[0058] Figure 7 This is an overall flowchart of the scheduling and response system in the embodiments of this application. Detailed Implementation

[0059] The following is in conjunction with the appendix Figure 1-7 This application will be described in further detail.

[0060] This application discloses a method for dispatching and responding to oil-immersed transformers for power grid stability. For example... Figure 1 As shown, a method for dispatching and responding to oil-immersed transformers for power grid stability includes the following steps:

[0061] S1, Multi-source data acquisition.

[0062] like Figure 2 As shown, this step simultaneously collects real-time operating status data and environmental meteorological data of the oil-immersed transformer. The real-time operating status data mainly includes load current and top oil temperature, while the environmental meteorological data includes real-time air pressure and ambient temperature. Specifically, it includes the following sub-steps:

[0063] S11. The system continuously measures the load current value through a current transformer installed in the transformer circuit.

[0064] S12. The system directly measures the top oil temperature by using a resistance temperature detector or fiber optic temperature sensor embedded in the transformer oil tank.

[0065] S13. The system obtains real-time air pressure values ​​through digital barometric pressure sensors deployed in meteorological monitoring stations near the transformer.

[0066] S14. The system obtains the ambient temperature value through a platinum resistance temperature sensor in the same meteorological monitoring station.

[0067] S15. All sensor output signals are synchronously sampled by the data acquisition unit. The data acquisition unit uses a high-precision analog-to-digital converter to convert analog signals into digital quantities and uses the global positioning system clock to add a uniform timestamp to all data points to achieve strict time alignment.

[0068] S16. The data acquisition unit transmits time-aligned data packets to the central controller in real time via industrial Ethernet or wireless communication module.

[0069] Step S1 integrates multiple sensors and achieves high-speed data synchronization, enabling the system to comprehensively capture instantaneous changes in the transformer's operating environment, especially the impact of real-time air pressure fluctuations on heat dissipation characteristics. This effectively overcomes the model bias caused by data asynchrony or missing data in traditional methods, significantly improves the reliability and environmental adaptability of transformer thermal state assessment, and ensures the temporal consistency between operating status and meteorological conditions, providing accurate input for subsequent dynamic correction of the thermal model.

[0070] S2, Dynamic correction of the thermal model.

[0071] like Figure 3As shown, this step involves dynamically correcting the convective heat dissipation coefficient in the transformer thermal model based on real-time air pressure to accurately reflect the impact of low-pressure environments on the transformer's heat dissipation capacity, ensuring that the transformer thermal model can adapt to the actual operating conditions in high-altitude areas. Specifically, it includes the following sub-steps:

[0072] S21. Read standard heat dissipation parameters. The system retrieves the reference convective heat dissipation coefficient under standard atmospheric pressure from the stored transformer technical files. This reference convective heat dissipation coefficient is a constant determined through type testing during the transformer design phase, and its value corresponds to the heat dissipation capacity under standard atmospheric pressure conditions.

[0073] S22. Obtain real-time air pressure data. The system receives and reads the real-time air pressure value transmitted in step S13. This real-time air pressure value represents the actual atmospheric pressure under the current environment.

[0074] S23. Calculate the corrected convective heat dissipation coefficient using the air pressure correction formula. The system takes the standard convective heat dissipation coefficient, real-time air pressure, and standard atmospheric pressure as inputs, and calculates the coefficient using an empirical air pressure correction formula based on the fundamental principles of fluid mechanics and heat transfer. The convective heat dissipation intensity of the transformer is closely related to the air density, which is approximately proportional to atmospheric pressure. In scenarios where forced convection is the primary heat dissipation method, the convective heat dissipation coefficient is related to the medium density. The power is directly proportional to the sum of its parts. Therefore, the corrected convective heat dissipation coefficient is equal to the standard convective heat dissipation coefficient multiplied by the nth power of the ratio of real-time air pressure to standard atmospheric pressure. The air pressure correction formula is:

[0075]

[0076] in, The convective heat dissipation coefficient is dynamically corrected by real-time air pressure. This represents the convective heat dissipation coefficient under standard atmospheric pressure; Indicates real-time air pressure; Indicates standard atmospheric pressure; This represents the experience index.

[0077] Among them, the experience index The following is the specific implementation process determined through a multi-dimensional verification method:

[0078] First, a climate simulation chamber was constructed in a laboratory environment, capable of precisely controlling key environmental parameters such as internal air pressure and temperature. Researchers placed transformer samples inside the chamber and conducted a series of temperature rise tests under different air pressure conditions. These tests covered typical operating conditions ranging from low to standard air pressure. In each test, the system recorded the dynamic response data of the transformer's load current and top oil temperature. In this way, a laboratory dataset encompassing the thermal behavior of transformers under various air pressure conditions was constructed.

[0079] Meanwhile, the system continuously collects historical operating data of transformers in actual operation in high-altitude areas. This field data includes long-term historical air pressure data, historical load current data, and historical top-layer oil temperature data. The data acquisition process follows unified standards and sampling frequencies to ensure data integrity and comparability. This constructs a field operating dataset reflecting the operating status of transformers in real, complex environments.

[0080] The system then merges the laboratory dataset and the field operation dataset. The data fusion process begins with time alignment and data cleaning, followed by integrating the controlled laboratory data with the measured field data to form a comprehensive and representative composite dataset.

[0081] Finally, with the optimization objective of minimizing the root mean square error between the predicted temperature from the transformer thermal model and the actual measured temperature, a nonlinear regression analysis algorithm was used for parameter fitting. This fitting process used a composite dataset as input and automatically adjusted the empirical index value through an iterative optimization algorithm until the optimal solution that minimizes the prediction error was found. Through this data-driven process, the optimal empirical index value suitable for high-altitude, low-pressure environments was finally obtained. Based on extensive experimental and field verification results, the stable value range of this empirical index is 0.5 to 0.8.

[0082] Step S2, by introducing real-time air pressure data and dynamically correcting the convective heat dissipation coefficient, solves the systematic deviation problem caused by the use of a fixed heat dissipation coefficient in traditional thermal models. This significantly improves the calculation accuracy of the transformer thermal model in high-altitude, low-pressure environments, enabling the model to more realistically simulate the actual heat dissipation process of the transformer. This lays a reliable model foundation for accurately predicting the winding hot spot temperature in subsequent steps.

[0083] S3, Temperature Trajectory Prediction.

[0084] like Figure 4 As shown, this step uses the current top-layer oil temperature obtained in step S1 as the initial state. It then uses the predicted load current sequence and ambient temperature sequence within the future scheduling cycle to drive the transformer thermal model corrected in step S2, ultimately generating a dynamic trajectory of the winding hot spot temperature. This trajectory accurately characterizes the thermal state evolution of the transformer winding over a future period, and specifically includes the following sub-steps:

[0085] S31. Calculate the total transformer loss sequence. Based on the predicted load current sequence, the system calculates the corresponding total transformer loss point by point. The total transformer loss includes two parts: no-load loss and load loss. No-load loss is the constant loss generated by the transformer core under energized conditions, and its value is determined by the transformer design parameters. Load loss is proportional to the square of the load current, specifically expressed as the product of the winding resistance loss coefficient and the square of the load current. The winding resistance loss coefficient is a constant derived from the transformer winding resistance and rated parameters. Based on the basic principles of electrical engineering (i.e., total loss = no-load loss + load loss), the system calculates each current value in the predicted load current sequence to generate the corresponding total transformer loss sequence. The calculation formula is as follows:

[0086]

[0087] in, This represents the total transformer loss. This is the no-load loss; This is the load current; This is the winding resistance loss coefficient determined by the transformer design parameters.

[0088] S32. Solve the dynamic trajectory of the top oil temperature. The system inputs the total transformer loss sequence and the predicted ambient temperature sequence into the corrected top oil temperature heat balance equation, and solves the equation using numerical integration to obtain the dynamic trajectory of the top oil temperature within the future scheduling cycle. The top oil temperature heat balance equation describes the dynamic balance relationship between the heat capacity, heat generation, and heat dissipation of the transformer oil. The equation is expressed as: the heat capacity of the transformer oil multiplied by the derivative of the oil temperature with respect to time equals the total transformer loss minus the product of the comprehensive heat dissipation coefficient and the convective heat dissipation coefficient corrected by the air pressure, multiplied by the oil temperature exponent raised to the power of the difference between the oil temperature and the ambient temperature. Here, the heat capacity of the transformer oil is a known physical parameter; the comprehensive heat dissipation coefficient and the oil temperature exponent are determined by the transformer design; the convective heat dissipation coefficient has been corrected by the air pressure in step S23. The numerical integration process uses the measured top oil temperature at the current moment as the initial value, and progressively calculates the top oil temperature value at each future time point, ultimately forming a continuous dynamic trajectory of the top oil temperature. Therefore, the expression of the top oil temperature heat balance equation is:

[0089]

[0090] in, The heat capacity of transformer oil; Top oil temperature; For time; This represents the total transformer loss. The overall heat dissipation coefficient; Ambient temperature; This refers to the oil temperature index.

[0091] S33. Calculate the dynamic trajectory of the winding hot spot temperature. The system inputs the top oil temperature dynamic trajectory obtained in step S32 and the predicted load current sequence into the winding hot spot temperature calculation equation, and calculates the corresponding winding hot spot temperature point by point. This calculation equation is established based on the heat conduction theory, expressing that the temperature of the hottest spot in the winding is composed of the top oil temperature and an additional temperature difference, which is equal to the product of the rated temperature difference and twice the winding exponent power of the ratio of the load current to the rated current. Wherein, the rated temperature difference is the difference between the steady-state hot spot temperature and the top oil temperature when the transformer is operating at rated current, determined by the transformer model; the rated current is the nominal value of the transformer; and the winding exponent is an empirical constant reflecting the characteristics of the winding thermal field distribution. By applying this equation point by point, the system converts the top oil temperature and load current at each time point into the winding hot spot temperature, ultimately generating a complete dynamic trajectory of the winding hot spot temperature. Therefore, the expression of the winding hot spot temperature calculation equation is:

[0092]

[0093] in, This refers to the hot spot temperature of the winding. This refers to the steady-state temperature difference between the hot spot temperature of the winding and the top oil temperature when the transformer is running continuously at rated current. Rated current; This represents the winding index.

[0094] Step S3, by sequentially calculating the total transformer loss, solving the dynamic trajectory of the top oil temperature, and calculating the dynamic change trajectory of the winding hot spot temperature, achieves accurate and dynamic prediction of the future thermal state of the transformer. This significantly improves the adaptability and overall accuracy of the winding hot spot temperature prediction results to high-altitude and low-pressure environments, providing a reliable and crucial temperature input for subsequent evaluation of the transformer's dynamic thermal load-bearing capacity. It effectively overcomes the prediction deviation problem caused by model inaccuracies in traditional static thermal evaluation methods.

[0095] S4. Parallel calculation of multidimensional dynamic thermal redundancy.

[0096] like Figure 5 As shown, this step calculates the multidimensional dynamic thermal redundancy in parallel based on the dynamic change trajectory of the winding hot spot temperature. The multidimensional dynamic thermal redundancy includes cumulative energy redundancy, instantaneous peak redundancy, and lifetime credit redundancy, and specifically includes the following sub-steps:

[0097] S41. Calculate the cumulative energy redundancy, which characterizes the total amount of additional heat load that the transformer can safely absorb over a longer time window. The specific steps are as follows:

[0098] S411. The system reads the preset maximum allowable hot spot temperature limit, which is a fixed safety threshold determined based on the heat resistance level of the transformer insulation material.

[0099] S412. The system identifies the time intervals during which the dynamic change trajectory of all winding hot spot temperatures is lower than the maximum allowable limit within future scheduling cycles.

[0100] S413. Within the aforementioned safe range, the system calculates the additional heat load energy that the transformer winding can withstand. This energy calculation is achieved by time integration of the additional heat load power, which is determined by the resistance characteristics of the winding and is proportional to the square of the load current.

[0101] S414. The system accumulates the extra energy that all safe zones can withstand during the entire scheduling cycle through integral calculation, and finally obtains the cumulative energy redundancy.

[0102] S42. Evaluate the instantaneous peak redundancy, which characterizes the transformer's ability to withstand minute-level power surges. The specific steps are as follows:

[0103] S421. The system performs a detailed analysis of the dynamic change trajectory of the winding hot spot temperature, paying particular attention to temperature fluctuations on short time scales of minutes.

[0104] S422. The system extracts all instantaneous temperature extreme points, including maximum points, of the dynamic change trajectory of the winding hot spot temperature within the entire scheduling cycle;

[0105] S423. The system compares each instantaneous temperature extreme value with the transformer insulation short-time overload safety threshold, which is usually higher than the allowable temperature limit for continuous operation and reflects the transformer's ability to withstand short-term thermal shock.

[0106] S424. The system calculates the difference between the temperature at each extreme point and the short-term overload safety threshold, i.e., the temperature margin.

[0107] S425. The system selects the minimum value among all temperature margins as the evaluation benchmark to obtain the instantaneous peak redundancy.

[0108] S43. Calculate lifetime credit redundancy, which is used to characterize the remaining available lifetime resources in the current cycle. The specific steps are as follows:

[0109] S431. The system inputs the complete dynamic change trajectory of the winding hot spot temperature into the insulation thermal aging model. This model is based on the thermal aging theory of insulation materials. Its core is the Arenius reaction rate equation, which expresses the exponential relationship between the aging rate of insulation materials and absolute temperature.

[0110] S432. The system calculates the cumulative effect of insulation aging caused by the predicted temperature trajectory in the future scheduling cycle based on the insulation thermal aging model. This cumulative effect is quantified as an equivalent lifetime consumed by normal operation at the reference temperature.

[0111] S433. The system compares the calculated equivalent life consumption with the life budget preset for this scheduling cycle. The life budget is usually allocated based on the total design life of the transformer and the operation and maintenance strategy.

[0112] S434. The system obtains the lifetime credit redundancy by calculating the difference between the lifetime budget and the predicted lifetime consumption.

[0113] Step S4 achieves a comprehensive and refined assessment of the transformer's thermal load capacity by parallel solving of three different dimensions of thermal redundancy indicators. This breaks through the limitations of traditional single-threshold assessment and can simultaneously characterize the transformer's safety boundaries in three key dimensions: energy accumulation, instantaneous overload, and long-term life consumption. This provides rich, accurate, and multi-dimensional decision-making basis for subsequent optimized scheduling and significantly improves the scheduling system's ability to tap the transformer's potential while ensuring equipment safety.

[0114] S5, Rolling Optimized Scheduling.

[0115] like Figure 6 As shown, this step uses cumulative energy redundancy, instantaneous peak redundancy, and lifetime credit redundancy as core constraints, and performs rolling optimization scheduling through model predictive control methods. This optimization process aims to maximize the total amount of new energy consumption and minimize transformer load fluctuations within the scheduling cycle, ultimately generating coordinated control commands for transformers and new energy power plants. Specifically, it includes the following sub-steps:

[0116] S51. Constructing the optimization objective function. This involves constructing a multi-objective function that integrates maximizing the total amount of new energy consumption with minimizing transformer load fluctuations, thus establishing a clear mathematical direction for the entire optimization scheduling process. The specific steps are as follows:

[0117] S511. The system establishes a mathematical function containing two key optimization objectives. The first optimization objective is to maximize the total amount of renewable energy consumed within the scheduling cycle. This objective is achieved by accumulating the total power injected into the grid by all renewable energy plants during the forecast period. Its mathematical expression is the maximization of the integral of the total renewable energy power over time. The second optimization objective is to minimize the transformer load fluctuation. This objective is achieved by calculating the integral of the square of the deviation between the transformer load current and its average value. Its mathematical expression is the minimization of the load current variance.

[0118] S512. The system combines the two objectives into a single comprehensive objective function by weighted summation, where the weight coefficients reflect the relative importance of different objectives in a specific operating scenario.

[0119] S52. Set multi-dimensional constraints. This involves specifying three different dimensions of thermal redundancy indices as rigid constraints with different priorities in the optimization model. This effectively embeds the dynamic thermal safety boundary of the transformer into the scheduling decision, ensuring that the primary premise of any scheduling scheme generated by subsequent optimization is to guarantee the safe operation of the transformer on instantaneous, medium-term, and long-term time scales. The specific steps are as follows:

[0120] S521. The system sets the temperature margin requirement corresponding to the instantaneous peak redundancy as the highest priority constraint to ensure that the winding hot spot temperature does not exceed the short-time overload safety threshold at any time.

[0121] S522. The system sets the equivalent lifetime consumption corresponding to lifetime credit redundancy as the second highest priority constraint, requiring that the total lifetime consumption within the scheduling cycle does not exceed the preset cycle lifetime budget.

[0122] S523. The system sets the energy integral value corresponding to the cumulative energy redundancy as a standard priority constraint, requiring that the value remain non-negative throughout the entire scheduling cycle.

[0123] S53. Perform model predictive control optimization. This method dynamically solves for the real-time control command that satisfies all safety constraints and is closest to the optimization objective through a rolling optimization mechanism of model predictive control. Based on the safety boundary set in step S52 and the optimization objective established in step S51, it recalculates the optimal solution at the beginning of each scheduling cycle, thus continuously adapting to the changing power grid state and environment. The specific steps are as follows:

[0124] S531. The system collects the actual measured values ​​of the power grid operation status at the current moment, including key parameters such as transformer load current and power output of new energy power plants, and uses these real-time data as the initial state for optimization calculation.

[0125] S532. Based on the acquired initial state, the system uses the power grid dynamic model to predict the evolution of the system state in the future within a finite time domain. This prediction process comprehensively considers uncertain factors such as the prediction of new energy output and the trend of load change, and generates a predicted trajectory of the system state.

[0126] S533. Based on the completion of state prediction, the system constructs a finite-time domain optimal control problem with the objective function of step S51 and the constraints of step S52 as the core. The solution of this optimization problem adopts a mature mathematical programming algorithm, and the optimal control sequence in the future time domain is obtained through numerical calculation.

[0127] S534. After the solution is completed, the system extracts the control command of the first time period from the optimal control sequence as the actual output at the current time and sends it to the actuator. At the next sampling time, the system re-executes the complete process from step S531 to step S534 to realize the continuous operation of rolling optimization.

[0128] S54. Implement adaptive adjustment of weighting coefficients. This ensures that when transformer insulation life resources are scarce, the system automatically strengthens control over long-term lifespan consumption. The specific steps are as follows:

[0129] S541. During the rolling optimization process, the system continuously monitors the real-time consumption status of lifetime credit redundancy, and the system presets a lifetime credit safety threshold as a trigger condition.

[0130] S542. When the system determines that the real-time remaining lifetime credit redundancy is lower than the safety threshold, the weight adjustment mechanism is automatically triggered. This mechanism increases the weight coefficient of the lifetime credit redundancy related constraints in the objective function. The adjustment range of the weight coefficient is negatively correlated with the ratio of the remaining lifetime credit redundancy to the initial value. That is, the lower the remaining ratio, the greater the increase in weight.

[0131] Step S5 transforms multidimensional dynamic thermal redundancy into optimization constraints and adopts rolling optimization and adaptive weight adjustment mechanisms to achieve coordinated optimization of transformer dynamic thermal safety and grid operation economy. This not only ensures the safe operation of transformers at different time scales, but also significantly improves the level of new energy consumption and smooths load fluctuations, overcoming the shortcomings of traditional static dispatching methods with fixed safety margins and single control objectives.

[0132] S6, Coordinate the execution of control commands.

[0133] This step translates commands into actual control actions through standard communication protocols and actuators, dynamically adjusting the power grid's operating status. It specifically includes the following sub-steps:

[0134] S61. The system receives and parses the coordination control commands from the optimization scheduling module. The command parsing process includes verifying data integrity, identifying the target device identifier, and extracting control parameters. The system verifies that the command format conforms to the preset communication protocol, ensuring that the command source is reliable and the content is valid.

[0135] S62. The system converts high-level control commands into executable signals for low-level equipment. For transformer load regulation, the system achieves the target load current by controlling the position of the on-load tap changer; for renewable energy power plants, the system converts power commands into specific modulation signals for the converter; the conversion process is based on the equipment control characteristic curves and the power grid dispatching procedures.

[0136] S63. The system sends control signals to the execution equipment via a remote terminal unit. The control process adopts a closed-loop regulation method. The system continuously monitors the actual load of the transformer and the actual output of the new energy power station, and compares them with the command values ​​in real time. When a deviation is detected, the system automatically performs dynamic correction to ensure that the operating status remains stable within the target range.

[0137] S64. The system implements safety interlock protection. The system monitors equipment protection signals in real time, and immediately suspends the current control command when an abnormal state is detected, prioritizing equipment safety. The system presets multiple protection thresholds and takes corresponding protective measures according to the severity of the abnormality.

[0138] S65. The system collects the results of instruction execution and feeds them back to the monitoring system. The system records key indicators such as actual control effect, response time, and steady-state error, and generates a complete execution report. This feedback data provides important status references for subsequent optimization of the scheduling cycle.

[0139] Step S6 ensures that the optimized scheduling results are accurately transformed into the actual operating state of the power grid through standardized instruction conversion and closed-loop execution mechanism, and establishes a reliable path from decision-making to execution, realizing precise control of power grid operation. At the same time, it ensures equipment safety through safety interlocking, and finally transforms the calculation results of the previous steps into actual technical benefits.

[0140] The implementation principle of the oil-immersed transformer scheduling and response method for power grid stability in this application is as follows: This application constructs a complete technology chain from accurate perception and dynamic prediction to optimized decision-making and closed-loop execution. First, by synchronously collecting transformer operating status and environmental meteorological data, and dynamically correcting thermal model parameters based on real-time air pressure, the fundamental problem of inaccurate heat dissipation capacity calculation caused by neglecting altitude changes in traditional static models is solved. Then, the corrected thermal model is used to generate dynamic trajectories of winding hot spot temperatures, and on this basis, the cumulative energy redundancy, instantaneous peak redundancy, and lifetime credit redundancy are calculated in parallel, thereby breaking through the evaluation limitation that a single fixed threshold cannot match multi-dimensional dynamic safety boundaries. Finally, with multi-dimensional dynamic thermal redundancy as the core constraint, rolling optimization scheduling is performed through model predictive control to generate and execute coordinated control commands. This achieves dynamic improvement of new energy absorption capacity and smoothing of load fluctuations while ensuring the thermal safety of the transformer across the entire time scale. It effectively solves the chain of technical defects in the prior art, from model inaccuracy to evaluation failure and then to control imbalance, and achieves a synergistic improvement in transformer dynamic load safety and efficient new energy absorption.

[0141] This application also discloses an oil-immersed transformer dispatching and response system for power grid stability. For example... Figure 7As shown, the system achieves complete intelligent transformer scheduling function through modular design, which includes a data acquisition module, a thermal model dynamic correction module, a temperature trajectory prediction module, a redundancy parallel solution module, an optimization scheduling module, and an execution control module.

[0142] The data acquisition module is responsible for synchronously acquiring real-time operating status data of the transformer and environmental meteorological data. This module integrates various sensing devices, including current transformers, temperature sensors, digital barometric pressure sensors, and more. The current transformer directly measures the load current in the transformer circuit; resistance temperature detectors or fiber optic temperature sensors embedded in the transformer oil tank directly measure the top oil temperature; digital barometric pressure sensors deployed at meteorological monitoring stations near the transformer acquire real-time barometric pressure values; and platinum resistance temperature sensors, also located at meteorological monitoring stations, acquire ambient temperature values. All sensor output signals are synchronously sampled by the data acquisition unit, which uses a high-precision analog-to-digital converter to convert analog signals into digital quantities and uses a GPS clock to add a unified timestamp to all data points to ensure strict time alignment. The data acquisition unit transmits the time-aligned data packets to the central processing system in real time via industrial Ethernet or a wireless communication module.

[0143] The dynamic correction module for the thermal model dynamically corrects the convective heat dissipation coefficient in the transformer thermal model based on real-time air pressure data. This module retrieves the baseline convective heat dissipation coefficient at standard atmospheric pressure from the stored transformer technical files and receives real-time air pressure measurements transmitted by the data acquisition module. Then, based on fundamental principles of fluid mechanics and heat transfer, the module calculates the corrected convective heat dissipation coefficient using an empirical air pressure correction formula. This calculation process considers the relationship between the transformer's convective heat dissipation intensity and air density, as well as the proportionality between air density and atmospheric pressure. In scenarios where forced convection cooling is dominant, the module uses an empirical index determined through multi-dimensional verification methods for calculation, ensuring that the corrected heat dissipation coefficient accurately reflects the impact of low-pressure environments.

[0144] The temperature trajectory prediction module uses the current top-layer oil temperature as the initial state to drive the modified transformer thermal model to generate the dynamic change trajectory of winding hot spot temperature within future scheduling cycles. This module first calculates the corresponding total transformer loss sequence based on the predicted load current sequence; then, it inputs the total transformer loss sequence and the predicted ambient temperature sequence into the modified top-layer oil temperature thermal balance equation, and solves it using numerical integration to obtain the dynamic trajectory of the top-layer oil temperature within future scheduling cycles; subsequently, it inputs the dynamic trajectory of the top-layer oil temperature and the load current sequence into the winding hot spot temperature calculation equation, calculating and outputting the corresponding dynamic change trajectory of winding hot spot temperature point by point.

[0145] The parallel redundancy calculation module calculates multi-dimensional dynamic thermal redundancy based on the dynamic temperature change trajectory of the winding hotspots. This module calculates the cumulative energy redundancy by integrating the additional thermal load energy the transformer winding can withstand over time within the range where the temperature trajectory is below the preset maximum allowable hotspot temperature limit; it extracts and analyzes the instantaneous temperature extremes of the temperature trajectory on a minute-level timescale to assess their margin with the transformer's short-term overload safety threshold, thus obtaining the instantaneous peak redundancy; and it inputs the temperature trajectory into the insulation thermal aging model to calculate the equivalent normal aging lifespan it will consume in future scheduling cycles, comparing this with the preset cycle life budget to obtain the lifespan credit redundancy.

[0146] The optimized scheduling module performs rolling optimization scheduling with multi-dimensional dynamic thermal redundancy as the core constraint, generating coordinated control commands for transformers and renewable energy power plants. This module constructs an optimization function aimed at maximizing the total renewable energy consumption and minimizing transformer load fluctuations within the scheduling cycle. It sets the safety margin requirement for instantaneous peak redundancy as the highest priority constraint, the budget guarantee for lifetime credit redundancy as the second highest priority constraint, and the non-negativity of cumulative energy redundancy as the standard priority constraint. The module employs model predictive control (MMC) for rolling optimization, recalculating the optimal control command based on the latest grid state in each scheduling cycle. The module also implements an adaptive adjustment mechanism for weight coefficients, automatically increasing their weight coefficients when lifetime credit redundancy is detected to be below the safety threshold.

[0147] The execution control module executes coordinated control commands to dynamically adjust the power grid's operating status. This module receives and parses the coordinated control commands generated by the optimization scheduling module, verifying that the command format conforms to the preset communication protocol. It then converts the higher-level control commands into executable signals for the lower-level equipment, including the position of the on-load tap changer on the control transformer and the modulation signals of the converter in the renewable energy plant. Subsequently, it sends control signals to the execution equipment through a remote terminal unit, continuously monitoring the actual operating status using a closed-loop regulation method and comparing it with the command values ​​in real time. Finally, it executes safety interlock protection, immediately suspending the control commands when an abnormal state is detected. The module also collects the command execution results and feeds them back to the monitoring system, forming a complete control closed loop.

[0148] Through the collaborative work of the above modules, this system has established a complete technology chain from data acquisition, model correction, state prediction, safety assessment to optimized scheduling and command execution. It has achieved accurate perception and dynamic prediction of transformer thermal state, constructed a multi-dimensional thermal safety assessment system, and achieved synergistic optimization of transformer dynamic thermal safety and power grid economic operation goals. Ultimately, it has significantly improved the level of new energy consumption and ensured the safe operation of the power grid.

[0149] The above are all preferred embodiments of this application, and are not intended to limit the scope of protection of this application. Therefore, all equivalent changes made in accordance with the structure, shape and principle of this application should be covered within the scope of protection of this application.

Claims

1. A method for dispatching and responding of oil-immersed transformer towards grid stability, characterized in that, Includes the following steps: The transformer's real-time operating status data and environmental meteorological data are collected synchronously. The real-time operating status data includes load current and top oil temperature, and the environmental meteorological data includes real-time air pressure and ambient temperature. Based on the real-time air pressure dynamic correction transformer thermal model convective heat dissipation coefficient; using the top oil temperature as the initial state of the transformer thermal model, and driving the corrected transformer thermal model with the predicted load current sequence and ambient temperature sequence, generating the dynamic change trajectory of winding hot spot temperature in future scheduling cycles. Based on the dynamic change trajectory, the multidimensional dynamic thermal redundancy is calculated in parallel. The multidimensional dynamic thermal redundancy includes cumulative energy redundancy for assessing the additional cumulative thermal load energy that the transformer can safely withstand within a future time window, instantaneous peak redundancy for assessing the transformer's short-term overload capacity, and lifetime credit redundancy for quantifying the equivalent normal aging life duration that can be consumed within a scheduling cycle. Using the multidimensional dynamic thermal redundancy as the core constraint, rolling optimization scheduling is performed through model predictive control. The rolling optimization scheduling aims to maximize the total amount of new energy consumption and minimize the transformer load fluctuation within the scheduling period. The optimization objectives are combined into an optimization objective function through a weighted summation method to generate coordinated control commands for transformers and new energy power plants. During the rolling optimization scheduling process, the system monitors the consumption status of the lifetime credit redundancy in real time; when it is determined that the real-time remaining amount of the lifetime credit redundancy is lower than the preset safety threshold, the weight coefficient of the lifetime credit redundancy in the optimization objective function is automatically increased, and the adjustment range of the weight coefficient is negatively correlated with the proportion of the remaining lifetime credit redundancy. The coordinated control commands are executed to dynamically adjust the power grid's operating status.

2. The method of claim 1, wherein, The transformer thermal model is composed of the top oil temperature thermal balance equation and the winding hot spot temperature calculation equation. The specific process for generating the dynamic change trajectory of winding hot spot temperature within the future scheduling cycle is as follows: The corresponding total transformer loss sequence is calculated based on the predicted load current sequence. Using the total transformer loss sequence and the predicted ambient temperature sequence as input, the top oil temperature thermal balance equation is solved by numerical integration to obtain the dynamic trajectory of the top oil temperature in the future scheduling cycle. The dynamic trajectory of the top oil temperature and the load current sequence are input into the winding hot spot temperature calculation equation, and the corresponding dynamic change trajectory of the winding hot spot temperature is calculated and output point by point.

3. The method according to claim 2, characterized in that, The expression for the top oil temperature heat balance equation is as follows: in, The heat capacity of transformer oil; Top oil temperature; For time; This represents the total transformer loss. The overall heat dissipation coefficient; The convective heat dissipation coefficient is dynamically corrected by real-time air pressure. The ambient temperature; This refers to the oil temperature index; Total transformer losses The calculation formula is: in, This is the no-load loss; This is the load current; This is the winding resistance loss coefficient determined by the transformer design parameters.

4. The method of claim 3, wherein, The expression for the equation for calculating the hot spot temperature of the winding is as follows: in, This refers to the hot spot temperature of the winding. This refers to the steady-state temperature difference between the hot spot temperature of the winding and the top oil temperature when the transformer is running continuously at rated current. Rated current; This represents the winding index.

5. The method of claim 4, wherein, The air pressure correction formula for the convective heat dissipation coefficient is: in, This represents the convective heat dissipation coefficient under standard atmospheric pressure; Indicates real-time air pressure; Indicates standard atmospheric pressure; This represents the experience index.

6. The method according to claim 5, characterized in that, The experience index The value range is from 0.5 to 0.

8.

7. The method according to claim 1, characterized in that, The dynamic change trajectory is calculated using the following methods to determine the multidimensional dynamic thermal redundancy: The cumulative energy redundancy is obtained by calculating the integral of the additional heat load energy that the transformer winding can withstand over time within the range where the dynamic change trajectory is lower than the preset maximum allowable hot spot temperature limit. The instantaneous peak redundancy is obtained by extracting and analyzing the instantaneous temperature extreme values ​​of the dynamic change trajectory on a minute-level time scale and evaluating their margin with the transformer short-term overload safety threshold. By inputting the dynamic change trajectory into the insulation thermal aging model, the equivalent normal aging lifetime duration that it will consume in future scheduling cycles is calculated, and the lifetime credit redundancy is obtained by comparing it with the preset cycle lifetime budget.

8. The method according to claim 7, characterized in that, During the rolling optimization scheduling process, constraint priorities are set for the multidimensional dynamic thermal redundancy: the safety margin requirement of the instantaneous peak redundancy is set as the highest priority constraint; the budget guarantee of the lifetime credit redundancy is set as the second highest priority constraint; and the non-negativity of the cumulative energy redundancy is set as the standard priority constraint.

9. A dispatching and response system for oil-immersed transformers oriented towards power grid stability, used to implement the method of any one of claims 1 to 8, characterized in that, include: The data acquisition module is used to synchronously collect real-time operating status data of the transformer and environmental meteorological data; The thermal model dynamic correction module is used to dynamically correct the convective heat dissipation coefficient in the transformer thermal model based on the real-time air pressure. The temperature trajectory prediction module is used to drive the modified transformer thermal model with the current top oil temperature as the initial state to generate the dynamic change trajectory of the winding hot spot temperature in future scheduling cycles. A parallel redundancy calculation module is used to calculate multi-dimensional dynamic thermal redundancy in parallel based on the dynamic change trajectory. The optimization scheduling module is used to perform rolling optimization scheduling with the multi-dimensional dynamic hot redundancy as the core constraint, and generate coordinated control commands for transformers and new energy power plants. The execution control module is used to execute the coordinated control commands to dynamically adjust the power grid operating status.

Images