Intelligent power grid transformer remote control and state feedback method

Abstract

The application discloses a smart grid transformer remote control and state feedback method, and particularly relates to the technical field of intelligent control of power equipment, and is used for solving the insulation damage risk caused by the dynamic mismatch of the existing cooling system remote start-stop operation and the real internal thermal state of the transformer; through real-time acquisition of load current and core vibration signals, the cooler is turned off during the load valley period to achieve energy saving; during the closing period, the outer shell leakage magnetic flux high harmonic component is synchronously measured to calculate the eddy current loss characteristic quantity, and a thermal accumulation risk dynamic evaluation model is established in combination with the oil film dielectric response characteristics; the oil flow stagnation state is identified based on the nonlinear dynamics characteristics of the core vibration; when the thermal accumulation risk value exceeds the threshold value and the oil flow stagnation is continued, a forced start instruction is generated; the instruction is executed and the minimum operation duration matching the thermal recovery characteristics of the transformer is maintained, the precise mapping of the internal thermal state and the precise control of the cooling intervention time are realized, and the insulation damage risk caused by thermal hysteresis is eliminated while the energy saving benefit is ensured.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for power equipment, and more specifically, to a method for remote control and status feedback of transformers in smart grids. Background Technology

[0002] In the operation and management of transformers in smart grids, remote control of cooling systems to optimize energy efficiency is a common practice. Existing methods typically rely on real-time top-level oil temperature data to control the start and stop of the cooler via remote commands: actively shutting down the cooling equipment during off-peak periods to reduce no-load losses, and restarting it after the oil temperature rises back to the threshold. This method is widely used in energy-saving control of transformers in various smart substations, and its technological foundation depends on the deployment of standard temperature sensor networks and remote communication architectures.

[0003] However, existing methods have drawbacks: there is a dynamic mismatch risk between the remote start-stop operation of the cooling system and the actual thermal state inside the transformer. Due to the uneven distribution of the thermal field inside the transformer, the temperature change in critical areas (such as winding hot spots) lags significantly behind the top oil temperature monitoring value. When the cooler is shut down during the off-peak period, the internal oil flow stagnates, causing heat to accumulate continuously in local areas, forming a monitoring blind spot. When the load suddenly increases, the control system delays the start of cooling by relying solely on the delayed oil temperature feedback, causing the temperature of the hidden hot area to exceed the standard rapidly in a short period of time. This reverse thermal shock effect triggered by the energy-saving control strategy is caused by the inability of the existing single-point temperature monitoring mechanism to characterize the actual thermal state inside the equipment, which will eventually lead to irreversible damage to the insulation material and affect the safety of remote control. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, embodiments of the present invention provide a method for remote control and status feedback of smart grid transformers to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] Methods for remote control and status feedback of smart grid transformers include:

[0007] S1. Real-time acquisition of transformer load current data and core vibration signals;

[0008] S2. When the load current data is lower than the preset load threshold, a cooler shutdown command is generated and executed.

[0009] S3. During the cooler shutdown period, perform the following operations:

[0010] Measure the higher harmonic components of the leakage flux of the transformer casing, and calculate the characteristic quantity of eddy current loss based on the higher harmonic components;

[0011] Obtain the dielectric response characteristics of transformer oil film;

[0012] The risk value of thermal accumulation is calculated based on the correlation parameters between the characteristic quantities of eddy current loss and the dielectric response characteristics.

[0013] S4. Identify oil flow stagnation state based on the nonlinear dynamic characteristics of iron core vibration signal;

[0014] S5. When the heat accumulation risk value exceeds the heat accumulation risk threshold and the oil flow stagnation continues, a cooler forced start command is generated.

[0015] S6. Execute the cooler forced start command and maintain the preset minimum running time.

[0016] In a preferred embodiment, real-time acquisition of transformer load current data and core vibration signals includes:

[0017] Load current data is collected using a current transformer;

[0018] The vibration signal of the transformer core is collected by an acceleration sensor installed in the transformer core clamp.

[0019] The current transformer's measurement accuracy meets the preset current accuracy requirements, and the acceleration sensor's frequency response range covers the core vibration fundamental frequency and harmonic components.

[0020] In a preferred embodiment, when the load current data is lower than a preset load threshold, a cooler shutdown command is generated and executed, including:

[0021] Obtain the three-phase effective values ​​of real-time load current data;

[0022] Calculate the arithmetic mean of the three-phase effective values ​​as the current load level;

[0023] Compare the current load level with the preset load threshold;

[0024] If the current load level remains below the preset load threshold for a preset duration, a cooler shutdown command will be generated.

[0025] Send a cooler shutdown command to the cooler drive circuit via the remote control interface;

[0026] The cooler drive circuit disconnects the cooler power supply contactor coil circuit.

[0027] In a preferred embodiment, the higher harmonic components are obtained by measuring the transformer casing grounding wire current signal using a Rogowski coil array, and extracting the harmonic components that are significantly higher than the power frequency from the current signal as higher harmonic components.

[0028] In a preferred embodiment, the dielectric response characteristics of the transformer oil film are obtained by applying a step voltage excitation to the transformer oil film and recording the polarization current response, and then obtaining the dielectric relaxation time parameters by fitting a relaxation model.

[0029] In a preferred embodiment, identifying the oil flow stagnation state based on the nonlinear dynamic characteristics of the core vibration signal includes:

[0030] Phase space reconstruction of the core vibration signal generates a state trajectory;

[0031] Calculate the recursive quantitative parameters of the state trajectory, including deterministic and laminar indices;

[0032] When the deterministic index is lower than the preset deterministic threshold and the laminar flow index is higher than the preset laminar flow threshold, the oil flow stagnation state is determined to be established.

[0033] In a preferred embodiment, phase space reconstruction is achieved through time-delay embedding, and recursive quantitative parameters are obtained by quantifying the diagonal structural features of the recursive graph.

[0034] In a preferred embodiment, when the heat buildup risk value exceeds the heat buildup risk threshold and the oil flow stagnation continues, a forced start command for the cooler is generated, including:

[0035] Obtain the current thermal accumulation risk value and the preset thermal accumulation risk threshold;

[0036] Verify the persistence of the oil flow stagnation indicator;

[0037] When the heat accumulation risk value continuously exceeds the preset heat accumulation risk threshold and reaches the preset risk judgment time, and the oil flow stagnation status flag remains valid, a cooler forced start command containing a forced start command code is generated.

[0038] In a preferred embodiment, the oil flow stagnation state is continuously verified by the duration of a status flag, and the preset heat accumulation risk threshold is dynamically adjusted according to the ambient temperature.

[0039] In a preferred embodiment, executing the cooler forced start command and maintaining a preset minimum running time includes:

[0040] The cooler drive circuit analyzes the cooler forced start command and closes the cooler power supply contactor coil circuit.

[0041] Start the runtime timer and keep the contactor coil circuit closed until the runtime timer reaches the preset minimum runtime.

[0042] The preset minimum operating time is set based on the transformer thermal recovery time, and the contactor coil circuit closure status is verified through auxiliary contact position feedback.

[0043] Compared with the prior art, the present invention has the following beneficial effects:

[0044] 1. By using a multi-dimensional state perception and collaborative decision-making mechanism, the accuracy and safety of transformer thermal state control are significantly improved. A dynamic correlation model between eddy current loss characteristics and oil film dielectric response is established. The abnormal leakage flux distribution is reflected by high-order harmonic components. The thermal aging state of the oil film is captured by combining dielectric relaxation parameters, realizing early quantitative assessment of internal heat accumulation risk. This overcomes the defect of single-point temperature monitoring being insensitive to internal hot spots, accurately maps the real thermal state of key areas of the transformer, and provides a more reliable decision basis for remote control.

[0045] 2. By identifying the oil flow stagnation state through the nonlinear dynamic characteristics of core vibration, and combining it with dual-condition triggering logic (exceeding the thermal risk limit and the oil flow stagnation continuing) to generate a forced start command, the timing of cooling intervention is strictly matched with the internal thermal state. The execution process dynamically links the minimum running time with the transformer's thermal recovery characteristics. Through a load rate-driven duration adjustment algorithm, the cooling effect is ensured to be sufficient. While maintaining energy-saving benefits, the risk of insulation damage caused by thermal hysteresis effect is eliminated, achieving a balance between safety and energy efficiency. Attached Figure Description

[0046] Figure 1 This is a flowchart of the method for remote control and status feedback of smart grid transformers according to the present invention. Detailed Implementation

[0047] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0048] Example: Figure 1 The present invention provides a method for remote control and status feedback of smart grid transformers, comprising:

[0049] S1. Real-time acquisition of transformer load current data and core vibration signals;

[0050] S2. When the load current data is lower than the preset load threshold, a cooler shutdown command is generated and executed.

[0051] S3. During the cooler shutdown period, perform the following operations:

[0052] Measure the higher harmonic components of the leakage flux of the transformer casing, and calculate the characteristic quantity of eddy current loss based on the higher harmonic components;

[0053] Obtain the dielectric response characteristics of transformer oil film;

[0054] The risk value of thermal accumulation is calculated based on the correlation parameters between the characteristic quantities of eddy current loss and the dielectric response characteristics.

[0055] S4. Identify oil flow stagnation state based on the nonlinear dynamic characteristics of iron core vibration signal;

[0056] S5. When the heat accumulation risk value exceeds the heat accumulation risk threshold and the oil flow stagnation continues, a cooler forced start command is generated.

[0057] S6. Execute the cooler forced start command and maintain the preset minimum running time.

[0058] S1. Real-time acquisition of transformer load current data and core vibration signals, specifically implemented as follows:

[0059] Real-time acquisition of transformer load current data is achieved through current transformers. The current transformers employ an epoxy resin casting structure, and their rated primary current is determined based on the transformer's capacity range. For example, for medium-sized distribution transformers (capacity range 630kVA to 2500kVA), a 1500A / 5A ratio current transformer is selected. This type of current transformer meets a 0.5-level accuracy requirement. This accuracy is achieved through calibration tests at multiple typical operating conditions during manufacturing, including light load (20% rated current), rated load (100% rated current), and short-time overload (120% rated current), ensuring that the measurement error does not exceed ±0.5% across the entire range. After on-site installation, accuracy is verified by injecting a standard test current. The test current value is 50% of the rated current, lasting 180 seconds, and the deviation between the measurement result and the standard meter reading must consistently be less than 0.5%.

[0060] The core vibration signal is acquired using an accelerometer. The accelerometer is a triaxial piezoelectric type, and its installation location is strictly limited to the clamping area on the transformer core. The specific installation process is as follows: First, determine the center point of the clamping area on the core according to the transformer structural drawings; this point is 50mm from the edge of the yoke. Then, use a magnetic base to fix the sensor; the magnetic force must reach at least 50N. Finally, perform contact surface treatment, including removing oil stains with acetone and polishing the surface with 400-grit sandpaper to ensure complete contact between the sensor's bottom surface and the metal surface.

[0061] The frequency response range of the accelerometer needs to cover the fundamental frequency and harmonic components of the iron core vibration. The fundamental frequency range is determined by the rated operating frequency of the transformer; for example, a rated frequency of 50Hz corresponds to a fundamental frequency band of 45Hz to 55Hz. The harmonic coverage range is determined based on the characteristics of oil flow stagnation and must include the 10th harmonic component (500Hz). Therefore, the lower limit of the sensor's frequency response is set to 0.5Hz to capture ultra-low frequency vibrations, and the upper limit is set to 2000Hz to cover higher harmonics. Frequency response verification is completed through a frequency sweep test: applying sinusoidal vibration excitation in logarithmic steps within the range of 0.5Hz to 2000Hz, the sensor output amplitude fluctuation is considered acceptable if it is controlled within ±3dB.

[0062] The achievement of the preset current accuracy requirements includes two stages: the factory calibration stage, in which no less than 10 calibration points are selected in the range of 5% to 120% of the rated current for full-range calibration; and the field verification stage, in which online tests are performed regularly, with the test current being a typical value between 30% and 80% of the rated current, for a duration of no less than 60 seconds, and the deviation between the measurement result and the reference value is consistently less than 0.5% to be considered as meeting the requirements.

[0063] Signal synchronization employs a hardware clock synchronization mechanism. Specifically, the load current data and the core vibration signal share the same high-precision GPS / BeiDou dual-mode clock source, achieving a time synchronization accuracy of ±1μs. The data acquisition unit packages data at fixed time intervals (10ms), and the data packets contain complete time stamp information, instantaneous three-phase current values ​​(unit: A), and triaxial vibration acceleration values ​​(unit: m / s²).

[0064] The fault handling mechanism is designed as follows: When the open-circuit voltage on the secondary side of the current transformer exceeds 5V, the protection circuit immediately short-circuits the output terminal; when the vibration signal amplitude exceeds the upper limit of the sensor range for 3 consecutive seconds, it automatically switches to the next higher range and records the range switching event. Raw data storage uses a first-in-first-out (FIFO) cyclic buffer with a buffer depth of 6000 sampling points (corresponding to 60 seconds of data) to ensure complete traceability of fault events.

[0065] S2. When the load current data is lower than the preset load threshold, a cooler shutdown command is generated and executed. The specific implementation is as follows:

[0066] The acquisition of real-time load current data using three-phase RMS values ​​is achieved through a standard power parameter measurement procedure. The specific process is as follows: The raw three-phase current signal acquired from the current transformer first undergoes anti-aliasing filtering. The filter is a Butterworth second-order low-pass filter with a cutoff frequency set to 40% of the sampling frequency. The filtered signal is then synchronously sampled at a fixed sampling rate, set to 80 times the rated frequency. For each power frequency cycle, the true RMS value calculation method is applied to the sampling points. This method includes three consecutive steps: first, calculating the square of the current value at each sampling point; second, calculating the arithmetic mean of the squares over a complete power frequency cycle; and third, performing a square root operation on the average value. Finally, the RMS value of each phase current (unit: amperes) is output.

[0067] The arithmetic mean of the three-phase effective values ​​is calculated as the current load level. Specifically, the effective values ​​of phase A, phase B, and phase C currents are added together, and the sum is divided by the number of phases, 3. When an abnormality is detected in a phase, the abnormality is determined by the effective value of that phase being zero for five consecutive power frequency cycles or exceeding 200% of the rated current. In this case, the abnormal phase data is automatically excluded. If only a single phase has effective data remaining, that phase value is directly used as the current load level, and a data abnormality recording event is triggered. The abnormal event record includes the abnormal phase identifier, duration, and last effective value.

[0068] The method for setting the preset load threshold is based on the transformer's no-load characteristics. The specific implementation process is as follows: After the transformer's initial commissioning or major overhaul, a continuous 72-hour no-load operation test is conducted, recording the no-load current values ​​within an ambient temperature range of 20℃ to 30℃; the arithmetic mean of these no-load current values ​​is calculated; the average is multiplied by a safety factor to obtain the final threshold. The safety factor ranges from 1.2 to 1.8, with the specific value determined according to the transformer's insulation class: 1.2-1.4 for Class A insulation, 1.4-1.6 for Class B insulation, and 1.6-1.8 for Class F insulation. The threshold data is stored in non-volatile memory and supports modification via a password-verified remote configuration interface. The modification record includes the operator ID, timestamp, and original / new value.

[0069] The current load level is compared with a preset load threshold using comparison logic with hysteresis. The hysteresis width is set to ±2% of the threshold, forming two boundary values: an upper threshold and a lower threshold. The comparison process is continuous, entering a pre-trigger state only when the current load level is below the lower threshold, and immediately exiting this state when it is above the upper threshold. The comparison result is displayed in real time on the monitoring interface, and an event log is generated when the state changes.

[0070] The determination of a sustained load below the threshold is achieved through a time accumulation mechanism. The preset duration is determined based on the transformer's thermal capacity characteristics, ranging from 30 to 300 seconds. The specific implementation process is as follows: a timer is started when entering the pre-trigger state, incrementing in 100-millisecond increments; if the current load level exceeds the upper threshold during the timing process, the timer is immediately reset; the determination condition is met when the timer's accumulated value reaches the preset duration. The remaining time percentage is displayed in real-time during the timing process, updating the display value every second.

[0071] The process for generating the cooler shutdown command is as follows: After the condition is met, the control processor generates a command data packet containing three fields. The first field is the command type identifier, with the shutdown command fixed at 0x0001; the second field is the current timestamp accurate to milliseconds, in the format of year-month-day-hour-minute-second-millisecond; the third field is a 16-bit cyclic redundancy check code calculated based on the first two fields. The command data packet is written to the transmission queue buffer, and the local status indicator is simultaneously updated to a yellow flashing mode.

[0072] The cooler shutdown command is sent to the cooler drive circuit via a remote control interface using an industry-standard communication protocol. The physical layer uses RS-485 differential signal transmission with a communication rate of 19200bps; the data link layer uses the MODBUS-RTU protocol, and the data frame contains five parts: a 1-byte slave address, a 1-byte function code (06H write register), a 2-byte register address (0002H instruction register), a 2-byte instruction data (0x0001), and a 2-byte CRC checksum. Before transmission, the signal undergoes optocoupler isolation, with the isolation medium having a withstand voltage of no less than 2500Vrms, and a response timeout of 300ms.

[0073] The execution process of disconnecting the cooler drive circuit from the contactor coil circuit involves three steps: First, the received instruction data packet is parsed, verifying address matching, function code correctness, and CRC check pass. Second, the output relay is driven; the relay is an electromagnetic normally open contact type with a rated load current of 10A and a coil drive voltage of 24VDC. Third, the contactor status changes are monitored; the coil circuit current is detected by a Hall current sensor, and execution is confirmed successful when the current value remains below 5mA for 10ms. Upon completion, a status report is generated, including the execution result code (0 for success / 1 for failure), contactor position status (0 for open / 1 for closed), and execution timestamp.

[0074] S3. During the cooler shutdown period, the operation is performed as follows:

[0075] The high-order harmonic components of the transformer casing leakage flux are measured using a Rogowski coil array. The Rogowski coil array consists of three independent coils, installed at the grounding wire connections at the top, middle, and bottom of the transformer casing, respectively. The coils are made of flexible PCB substrate, with an inner diameter of, for example, 50 mm, a number of turns of, for example, 1000, and a sensitivity coefficient of, for example, 0.1 V / A. The measurement process is as follows: First, the raw grounding wire current signal is acquired at a sampling rate of 2 MS / s; then, signal preprocessing is performed, including 50 Hz power frequency notch filtering (Q value, for example, 30, attenuation, for example, 60 dB) and baseline calibration (to eliminate DC offset); finally, harmonic components with frequencies not lower than 1 kHz are separated using a digital filter bank. The filter bank contains, for example, 20 fourth-order Butterworth bandpass filters, with center frequencies uniformly distributed logarithmically from 1 kHz to 100 kHz, and a bandwidth of 10% of the center frequency.

[0076] The frequency domain energy integration method is used to calculate the characteristic quantity of eddy current loss based on higher harmonic components. The specific process is as follows: Full-wave rectification and low-pass filtering (cutoff frequency, e.g., 10Hz) are performed on the harmonic components output from each filter channel to obtain the envelope amplitude of each frequency band. The envelope amplitudes are then weighted and accumulated according to frequency, with the weighting coefficient proportional to the square of the frequency. This results in an eddy current loss weight matrix, where rows correspond to the measurement position (top / middle / bottom), columns correspond to the frequency band, and element values ​​are the weighted amplitude integral results. The weighting coefficient setting rule is: a weight of 1.0 corresponds to a reference frequency of 1kHz, and the weight increases by a factor of 100 for every tenfold increase in frequency. This rule is based on the physical principle of eddy current loss being proportional to the square of the frequency in the law of electromagnetic induction.

[0077] The dielectric response characteristics of the transformer oil film are obtained through step voltage excitation. The excitation source is a high-voltage DC generator with an output voltage range of, for example, ±500V and a rise time of less than 10μs. Before applying voltage, the oil film state is checked: excitation is only applied if the measured oil film resistance is greater than 1GΩ. The excitation mode is a bipolar step sequence: first, a voltage of, for example, +200V is applied for 100ms, then switched to, for example, -200V for 100ms, and finally returned to zero. The polarization current response is acquired using a high-impedance current probe (input impedance, for example, 1TΩ), with a sampling rate of, for example, 100kS / s and a resolution of, for example, 0.1pA. Gold-plated electrodes are used at the probe-oil film contact point, and the contact pressure is maintained between 0.5N and 1N.

[0078] The dielectric relaxation time parameters were obtained by fitting a relaxation model using the Cole-Cole model. The model includes four parameters: static dielectric constant εs, high-frequency dielectric constant ε∞, relaxation time τ, and distribution coefficient α. The fitting process is as follows: First, the polarization current response is denoised using wavelet thresholding (wavelet basis, e.g., db4, decomposition level, e.g., 5 levels); then, the dielectric loss factor spectrum is calculated, with a frequency range of 0.01Hz-10kHz; finally, the model parameters are solved using nonlinear least squares fitting, with the iteration termination condition being that the number of iterations exceeds 100. Each iteration includes two steps: parameter update and residual evaluation. The relaxation time τ is output as the core parameter, and its physical meaning characterizes the polarization response speed of oil molecules.

[0079] The calculation of the thermal accumulation risk value based on the correlation parameters between eddy current loss characteristics and dielectric response characteristics is performed in three steps. The first step calculates the Frobenius norm of the eddy current loss weight matrix, which is the square root of the sum of the squares of all elements in the matrix. This calculation is implemented on an embedded processor (e.g., TI C6748) using floating-point instructions for acceleration. The second step calculates the derivative of the dielectric relaxation time parameter using the central difference method: dτ / dt = [τ(t+Δt) - τ(t-Δt)] / (2Δt), with a time step Δt, for example, 60 seconds. The third step multiplies the norm and derivative to obtain the basic correlation parameters. This calculation is performed every 5 minutes, and the result is stored in a circular buffer.

[0080] The final calculation of the heat accumulation risk value is: the basic correlation parameter multiplied by the temperature compensation coefficient. The temperature compensation coefficient is dynamically adjusted according to the ambient temperature, with the following adjustment rules: the coefficient is 1.0 at a reference temperature of 25℃, increasing by 0.02 for every 1℃ increase in temperature, and decreasing by 0.01 for every 1℃ decrease. The ambient temperature is obtained using a temperature sensor independently installed on the shaded side of the transformer, with a sensor accuracy of, for example, ±0.5℃, and a sampling interval of 5 minutes. The adjustment slope of the temperature compensation coefficient is calibrated through laboratory thermal aging tests; for example, an increase of 0.02 / ℃ in the coefficient at 40℃ can accurately compensate for thermal errors.

[0081] S4. Identifying oil flow stagnation based on the nonlinear dynamic characteristics of core vibration signals, specifically implemented as follows:

[0082] Phase space reconstruction of the iron core vibration signal to generate the state trajectory is achieved through a time-delay embedding method. This method involves determining two core parameters: the embedding dimension uses a spurious neighborhood method, which involves gradually increasing the dimension value and calculating the proportion of spurious neighborhood points in each dimension. When the proportion drops to, for example, below 5%, the final dimension is determined. The time delay uses an autocorrelation function method, calculating the autocorrelation function of the vibration signal and taking the delay corresponding to the first zero-crossing point of the function as the optimal delay value. The reconstruction process involves shifting the original vibration signal in time according to the optimal delay value, generating a multidimensional state vector sequence. Each state vector contains several consecutive sampling points, and the set of state vectors constitutes the state trajectory. The state trajectory is visualized as a point set distribution in a multidimensional space, typically with three to seven dimensions.

[0083] The recursive quantitative parameters for calculating state trajectories, including deterministic and laminar flow indices, are achieved through recursive graph analysis. The recursive graph construction process involves: calculating the Euclidean distance between all state vectors to form a distance matrix; setting a recursion threshold, and marking locations with distances less than the threshold as recursive points; the recursive graph is essentially a two-dimensional distribution of these recursive points. The deterministic indices are calculated by statistically analyzing the line segments formed by consecutive recursive points along the diagonal of the recursive graph, and determining the proportion of the total length of line segments exceeding a minimum length threshold out of all recursive points. The laminar flow indices are calculated by statistically analyzing the line segments formed by consecutive recursive points along the vertical direction, and determining the proportion of the total length of line segments exceeding a minimum length threshold out of all recursive points. The minimum length threshold is set to a multiple of the sampling interval, for example, ten times the sampling interval time.

[0084] The methods for setting the preset deterministic threshold and the preset laminar flow threshold are based on the normal operation benchmark of the transformer. The specific process is as follows: During normal operation of the transformer cooling system, multiple sets of core vibration signals are continuously collected; phase space reconstruction and recursive quantitative parameter calculation are performed on each set of signals; the distribution ranges of deterministic and laminar flow indicators are statistically analyzed; the lower limit of the deterministic indicator distribution range is multiplied by an adjustment coefficient to obtain the preset deterministic threshold; the upper limit of the laminar flow indicator distribution range is multiplied by an adjustment coefficient to obtain the preset laminar flow threshold. The adjustment coefficient ranges, for example, from 0.90 to 0.95, and is adjusted according to the transformer's operating years, decreasing by, for example, 0.03 for every five years of operation. Threshold data is stored in non-volatile memory and supports remote calibration and updates.

[0085] When the deterministic index is below a preset deterministic threshold and the laminar flow index is above a preset laminar flow threshold, the oil flow stagnation state is determined. The determination logic includes three verification steps: the first step checks the data validity, requiring the input signal-to-noise ratio to be no less than a preset signal-to-noise ratio threshold, such as 30dB; the second step performs continuous verification, requiring the determination condition to be met continuously for more than a preset duration, such as 30 seconds; the third step performs cross-validation, checking whether the energy distribution of the vibration signal spectrum within the same time period conforms to the characteristics of oil flow stagnation, specifically manifested as an increase in the proportion of low-frequency band (one to three times the power frequency) energy exceeding the normal baseline, such as 20%. An oil flow stagnation state flag is generated after all three steps pass; otherwise, a data re-acquisition process is triggered.

[0086] The process of quantifying the diagonal structure features of a recursive graph includes feature extraction and normalization. The diagonal structure features include three sub-features: the longest diagonal length, the average diagonal length, and the density of recursive points along the diagonal direction. The quantization method is as follows: First, all diagonal segments in the recursive graph are identified, and the length of each segment is recorded. Then, the length of the longest segment, the arithmetic mean of the lengths, and the ratio of the total number of segments to the area of ​​the recursive graph are calculated. Finally, the three sub-feature values ​​are divided by their respective baseline values ​​under normal operating conditions to obtain normalized feature values. The normalized feature values ​​are input into a pre-defined classification model, which outputs deterministic and laminar flow indices. The classification model uses a linear discriminant analysis algorithm, and the weight coefficients are obtained through training with historical data.

[0087] S5. When the heat accumulation risk value exceeds the heat accumulation risk threshold and the oil flow stagnation continues, a forced start command for the cooler is generated, specifically implemented as follows:

[0088] The current thermal accumulation risk value and the preset thermal accumulation risk threshold are obtained through a data interface. The current thermal accumulation risk value is directly read from the real-time cache of the calculation results in step S3. This cache uses a double buffering mechanism to ensure data consistency and is updated every 5 minutes. The preset thermal accumulation risk threshold is stored in non-volatile memory, and its baseline value is determined through transformer thermal model simulation: under rated load conditions, different ambient temperature scenarios are simulated, and the risk value when the oil film hot spot temperature reaches the critical value of the insulation material is recorded. The value after a safety margin of 20% is taken as the baseline threshold. The threshold data table contains the temperature-threshold correspondence and supports modification through a password-authenticated remote configuration interface. For example, the threshold baseline value setting process is as follows: in the simulation environment, the ambient temperature is set to 25℃, and rated load is applied. When the oil film hot spot temperature reaches 105℃, the risk value is recorded as 0.8. After taking a safety margin of 20%, the baseline threshold is 0.64.

[0089] Verifying the persistence of the oil flow stagnation status flag is implemented through a status monitoring thread. The oil flow stagnation status flag is derived from the Boolean value output in step S4 and is stored in the status register. The persistence verification involves three steps: first, checking the flag bit validity; invalid states include uninitialized registers or incorrect checksums; second, verifying the flag bit continuity, requiring the flag bit to remain true for 300 consecutive sampling periods (e.g., 5 minutes); and finally, performing timestamp verification to ensure the flag update time is within a valid time window (e.g., 10 minutes). The verification process is executed once per second, and the results are recorded in the status log. When a flag bit transition is detected, a timestamp refresh operation is triggered synchronously.

[0090] When the heat accumulation risk value continuously exceeds the preset heat accumulation risk threshold for a preset risk judgment time and the oil flow stagnation status flag remains valid, a cooler forced start command containing a forced start command code is generated. The determination of continuously exceeding the threshold uses a sliding window mechanism: the window length is equal to the preset risk judgment time, and the heat accumulation risk value of all sampling points within the window must be greater than the dynamic threshold corresponding to the current ambient temperature. The preset risk judgment time is set based on the transformer thermal time constant, with a typical range of, for example, 3 to 10 minutes. When the conditions are met, the control processor generates a 32-bit instruction code. The code structure contains four fields: the first field is the instruction type code, with the forced start command fixed as hexadecimal F0; the second field is the urgency level indicator, graded according to the proportion of the risk value exceeding the threshold, for example, exceeding 10% is level one, and exceeding 20% ​​is level two; the third field is the current timestamp accurate to milliseconds; and the fourth field is the cyclic redundancy check code of the first three fields. Before the instruction data packet is written to the sending queue, it is formatted as a start character (hexadecimal 7E), length byte, instruction content, and end character (hexadecimal 0D).

[0091] The verification of the continuous existence of the oil flow stagnation state through the duration of the status flag is specifically implemented as follows: a minimum duration requirement is set, which is related to the formation time of the transformer oil flow stagnation. The verification method uses a timer accumulation: when the status flag first becomes true, the timer is started with a timing step of 100 milliseconds; the timer continuously accumulates while the status flag remains true; if the status flag becomes false, the timer is immediately reset; when the accumulated value of the timer reaches the minimum duration, the persistence condition is determined to be met. The minimum duration is set according to the transformer capacity classification. The capacity classification standards are: less than 10 MVA is Level 1, with a minimum duration of, for example, 3 minutes; 10 MVA to 50 MVA is Level 2, with a minimum duration of, for example, 5 minutes; and greater than 50 MVA is Level 3, with a minimum duration of, for example, 10 minutes. The time period setting can be modified through the maintenance interface.

[0092] The preset heat buildup risk threshold is dynamically adjusted based on ambient temperature using a temperature compensation algorithm. Ambient temperature is acquired using three independently installed platinum resistance temperature sensors located at different heights on the shaded side of the transformer body, with a sampling interval of 5 minutes. The dynamic adjustment rules are as follows: a reference temperature of 25℃ corresponds to a reference threshold; for every 1℃ increase in temperature, the threshold decreases by a specific percentage of the reference value; for every 1℃ decrease in temperature, the threshold increases by a specific percentage of the reference value. The adjustment ratio is calibrated through accelerated aging tests in the laboratory. The test method involves setting different temperature conditions in a constant temperature chamber, measuring the change in the rate of heat buildup in the oil film, and determining the adjustment ratio of the threshold for every 1℃ change in temperature; a typical value is, for example, 0.5%. The adjustment calculation is performed every 30 minutes, and the calculation process includes three steps: calculating the average temperature, querying the proportional coefficient, and updating the threshold. The result is written to the threshold register.

[0093] S6. Execute the cooler forced start command and maintain the preset minimum running time. The specific implementation is as follows:

[0094] The forced start instruction for the cooler is decoded by parsing the cooler drive circuit. The parsing process consists of four steps: First, verifying that the start and end characters of the instruction data packet conform to the hexadecimal format of 7E and 0D; second, checking whether the length field matches the number of bytes in the instruction content; third, calculating the cyclic redundancy check (CRC) code and comparing it with the check field in the data packet; and finally, extracting the instruction type code to confirm whether it is a forced start command identified by hexadecimal F0. An error handling mechanism is triggered when parsing fails: the first failure is logged and a retransmission is requested; three consecutive failures trigger a level one alarm and lock the instruction receiving port. Successfully parsed instruction data is stored in the instruction register, and the instruction queue buffer is cleared. Instruction parsing is implemented using a dedicated decoding chip (e.g., STM32F407), with a clock frequency of, for example, 72MHz.

[0095] The drive circuit closes the cooler-powered contactor coil circuit to perform a physical connection operation. Upon receiving a valid command, the drive circuit first disconnects the protective interlock device of the current output relay; then, it applies a rated DC voltage (e.g., 24V) and a drive current (e.g., 100mA) to the relay coil; finally, it detects the relay coil circuit current, and determines successful drive operation when the current reaches 90% of the rated value. After the contactor coil circuit closes, the main contacts connect to a three-phase AC power supply, with a voltage range of (e.g., 380V to 480V) and a frequency of 50Hz or 60Hz. The drive operation execution time is limited to 500 milliseconds; failure to complete within this timeout is considered a failure. The contactor is an electromagnetic normally open contact type with a rated load current (e.g., 10A) and a mechanical life of (e.g., 100,000 cycles).

[0096] The operation duration timer and maintenance of the contactor coil circuit closure are implemented using a hardware timer. The timer employs a watchdog-type integrated circuit (e.g., MAX6818), with a temperature-compensated crystal oscillator as the clock source, achieving an accuracy of, for example, ±10ppm. When the timer is started, a preset minimum operating duration value is written, and the timer begins counting down, simultaneously triggering state-locking logic: before the timer reaches zero, all external commands to shut down the cooler, including manual operation signals and automatic control signals, are blocked. During the maintenance period, the contactor status is checked every second, with the closure status verified by real-time monitoring of the coil circuit current and auxiliary contact position. The timer has a power-off memory function, and a backup battery can maintain data for, for example, 72 hours.

[0097] The preset minimum operating time is set according to the transformer's thermal recovery time. The specific rule is as follows: the base duration is a specific multiple of the time required for the transformer to reach its rated temperature rise under no-load conditions. The setting process includes: during the transformer commissioning test, recording the time required for the oil temperature to drop to ambient temperature after the cooler is shut down; multiplying this time by a safety factor to obtain the base duration; adjusting the base duration according to the current load factor, with the duration increasing by a specific percentage for every 10% increase in load factor. The safety factor range is, for example, 1.2 to 1.5, and the load factor adjustment percentage is, for example, increasing the duration by 5% for every 10% increase in load. The duration setting value is stored in non-volatile memory and supports remote modification. Setting example: transformer thermal recovery time 20 minutes, safety factor 1.3, load factor 60%, then the final duration = 20 × 1.3 × (1 + 0.3) = 33.8 minutes.

[0098] The contactor coil circuit closure status is verified via auxiliary contact position feedback using a dual-channel detection mechanism. The main detection channel connects to the normally open contact of the auxiliary contact; when the contactor closes, the contact conducts and outputs a high-level signal. The backup detection channel connects to a current transformer to monitor the coil circuit current value in real time. The verification logic is as follows: the closure status is determined to be valid when both the high-level signal from the main channel and the current from the backup channel are greater than a threshold; if the main channel fails, the status is determined to be valid when the current from the backup channel is greater than the threshold for 100 milliseconds; verification fails when both channels fail. The threshold current is set to, for example, 80% of the rated coil current. The auxiliary contact and the main contact are mechanically linked, with a positional deviation of less than 0.5 mm.

[0099] In addressing the risk of transformer thermal runaway, this embodiment establishes a dynamic correlation model between eddy current loss characteristics and oil film dielectric response. It quantifies the risk of thermal accumulation by multiplying higher harmonic components (reflecting abnormal leakage flux distribution) with dielectric relaxation parameters (characterizing the degree of oil film thermal aging). This cross-scale electromagnetic-thermal-dielectric correlation overcomes the limitations of traditional single-parameter threshold judgments. In the oil flow stagnation identification stage, recursive quantitative analysis of the nonlinear dynamic characteristics of core vibration is employed. The synergistic changes of deterministic and laminar indices capture microscopic oil flow stagnation phenomena, making it more sensitive to early oil flow obstruction compared to conventional spectral analysis.

[0100] This embodiment employs a strict conditional triggering chain: a forced start command is generated only when both the heat accumulation risk value exceeds a threshold and the oil flow stagnation persists. This overcomes the concerns of those skilled in the art regarding misjudgment of a single parameter and dynamically correlates the minimum operating time with the transformer's thermal recovery characteristics, ensuring sufficient cooling effect through a load-rate-driven duration adjustment algorithm. The entire control process forms a closed-loop logic of "electromagnetic monitoring - vibration analysis - risk calculation - state verification - command execution - duration maintenance," with the output of each step serving as input constraints for the next stage.

[0101] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0102] It should be noted that this invention can be deployed on the device itself to realize embedded applications, or it can run on a PC or other terminal with a user interface, thereby meeting various hardware environments and usage requirements.

[0103] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, as a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, all or part of the processes or functions described in the embodiments of this application are generated. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that a computer can access or a data storage device such as a server or data center that includes one or more sets of available media. The available medium can be a magnetic medium (e.g., floppy disk, hard disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0104] Those skilled in the art will understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and modules described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0105] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0106] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0107] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0108] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0109] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0110] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for remote control and status feedback of smart grid transformers, characterized in that, include: S1. Real-time acquisition of transformer load current data and core vibration signals; S2. When the load current data is lower than the preset load threshold, a cooler shutdown command is generated and executed. S3. During the cooler shutdown period, perform the following operations: Measure the higher harmonic components of the leakage flux of the transformer casing, and calculate the characteristic quantity of eddy current loss based on the higher harmonic components; Obtain the dielectric response characteristics of transformer oil film; The risk value of thermal accumulation is calculated based on the correlation parameters between the characteristic quantities of eddy current loss and the dielectric response characteristics. S4. Identify oil flow stagnation state based on the nonlinear dynamic characteristics of iron core vibration signal; S5. When the heat accumulation risk value exceeds the heat accumulation risk threshold and the oil flow stagnation continues, a cooler forced start command is generated. S6. Execute the cooler forced start command and maintain the preset minimum running time.

2. The method for remote control and status feedback of smart grid transformers according to claim 1, characterized in that, Real-time acquisition of transformer load current data and core vibration signals, including: Load current data is collected using a current transformer; The vibration signal of the transformer core is collected by an acceleration sensor installed in the transformer core clamp. The current transformer's measurement accuracy meets the preset current accuracy requirements, and the acceleration sensor's frequency response range covers the core vibration fundamental frequency and harmonic components.

3. The method for remote control and status feedback of smart grid transformers according to claim 1, characterized in that, When the load current data is lower than the preset load threshold, a cooler shutdown command is generated and executed, including: Obtain the three-phase effective values ​​of real-time load current data; Calculate the arithmetic mean of the three-phase effective values ​​as the current load level; Compare the current load level with the preset load threshold; If the current load level remains below the preset load threshold for a preset duration, a cooler shutdown command will be generated. Send a cooler shutdown command to the cooler drive circuit via the remote control interface; The cooler drive circuit disconnects the cooler power supply contactor coil circuit.

4. The method for remote control and status feedback of smart grid transformers according to claim 1, characterized in that, The method for obtaining higher harmonic components is as follows: the current signal of the transformer casing grounding wire is measured by a Rogowski coil array, and the harmonic components that are significantly higher than the power frequency are extracted from the current signal as higher harmonic components.

5. The method for remote control and status feedback of smart grid transformers according to claim 1, characterized in that, The dielectric response characteristics of transformer oil film are obtained by applying a step voltage excitation to the transformer oil film and recording the polarization current response, and then obtaining the dielectric relaxation time parameters by fitting a relaxation model.

6. The method for remote control and status feedback of a smart grid transformer according to claim 1, characterized in that, Identifying oil flow stagnation states based on the nonlinear dynamic characteristics of core vibration signals, including: Phase space reconstruction of the core vibration signal generates a state trajectory; Calculate the recursive quantitative parameters of the state trajectory, including deterministic and laminar indices; When the deterministic index is lower than the preset deterministic threshold and the laminar flow index is higher than the preset laminar flow threshold, the oil flow stagnation state is determined to be established.

7. The method for remote control and status feedback of a smart grid transformer according to claim 6, characterized in that, Phase space reconstruction is achieved through time-delay embedding, and recursive quantitative parameters are obtained by quantifying the diagonal structure features of the recursive graph.

8. The method for remote control and status feedback of a smart grid transformer according to claim 1, characterized in that, When the heat accumulation risk value exceeds the heat accumulation risk threshold and the oil flow stagnation continues, a forced start command for the cooler is generated, including: Obtain the current thermal accumulation risk value and the preset thermal accumulation risk threshold; Verify the persistence of the oil flow stagnation indicator; When the heat accumulation risk value continuously exceeds the preset heat accumulation risk threshold and reaches the preset risk judgment time, and the oil flow stagnation status flag remains valid, a cooler forced start command containing a forced start command code is generated.

9. The method for remote control and status feedback of a smart grid transformer according to claim 8, characterized in that, The oil flow stagnation state is continuously verified by the duration of the state flag, and the preset heat accumulation risk threshold is dynamically adjusted according to the ambient temperature.

10. The method for remote control and status feedback of a smart grid transformer according to claim 1, characterized in that, Execute the forced start command for the cooler and maintain it for a preset minimum running time, including: The cooler drive circuit analyzes the cooler forced start command and closes the cooler power supply contactor coil circuit. Start the runtime timer and keep the contactor coil circuit closed until the runtime timer reaches the preset minimum runtime. The preset minimum operating time is set based on the transformer thermal recovery time, and the contactor coil circuit closure status is verified through auxiliary contact position feedback.

Images