Automatic test method and system for power grid transformer overload caused by new energy

Abstract

The application provides a new energy caused power grid transformer overload automatic test method and system, and relates to the transformer life loss evaluation technical field. The application realizes the closed loop control of the whole process of loading-measuring-evaluating-correcting by the programmable inverter-energy storage device for cyclic loading of the transformer, real-time collection of the electrical quantity, temperature and gas content in the oil, dynamic calculation of the hot spot temperature and life loss by using the coupled thermal-life model, automatic adjustment of the loading sequence and termination of the test when the threshold is reached. The overload capacity and life loss report is automatically generated and uploaded after the test is completed. The application improves the test authenticity, life evaluation accuracy and field application value.

Description

Technical Field

[0001] This invention relates to the field of transformer life loss assessment technology, and in particular to an automatic testing method for power grid transformer overload caused by new energy sources. Background Technology

[0002] With the large-scale grid connection of intermittent new energy sources such as wind power and photovoltaics to the distribution network and main grid, the power of transformers in some areas has long exhibited a "bidirectional, pulsating" characteristic, resulting in a "heavy load-overload-reverse transmission" cycle.

[0003] To ensure equipment safety in scenarios with a high proportion of renewable energy, domestic and international research is accelerating on electromagnetic transient modeling of main transformers, assessment of multi-field coupling losses, and construction of test platforms. However, the focus is mainly on simulation and offline evaluation; on-site automated testing and evaluation for "frequent power flow switching-overload" conditions remains lacking. Provincial grid companies have included "reliability assessment and improvement of transformers with frequent power flow switching" in their science and technology project requirements, demonstrating the industry's urgent need for automated testing tools.

[0004] Current transformer life loss assessments mainly rely on empirical formulas or single physical models, which suffer from insufficient accuracy, poor adaptability, and weak real-time performance, making it difficult to cope with the complex overload-unloading cycles caused by new energy sources. Furthermore, the lack of an integrated testing-online calculation-intelligent judgment closed-loop solution makes it impossible to quickly quantify the impact of overload on key indicators such as hotspot temperature and life loss factor (LLF). Summary of the Invention

[0005] To overcome the shortcomings of existing technologies, the purpose of this invention is to provide an automatic testing method for overload of power grid transformers caused by new energy sources. This method realizes closed-loop control of the entire process of loading, measurement, evaluation, and correction, thereby improving the authenticity of the test, the accuracy of life assessment, and the value of field application.

[0006] To achieve the above objectives, the present invention provides the following solution:

[0007] An automatic testing method for power grid transformer overload caused by new energy sources includes:

[0008] Collect historical data on renewable energy output in the target area, and generate test sequences including normal load, heavy load, overload and reverse power flow sections based on the historical data on renewable energy output;

[0009] The programmable inverter-energy storage device is connected in parallel to the low-voltage side of the transformer under test, and the transformer under test is cyclically loaded according to the test sequence.

[0010] The transformer's measured values ​​are obtained by collecting primary and secondary voltage and current, and simultaneously collecting top oil temperature, winding hot spot temperature, and gas content in the oil.

[0011] Calculate the hot spot temperature and life loss factor based on the transformer measurements;

[0012] If the lifespan loss factor or the hot spot temperature reaches a preset threshold, the cyclic loading process is stopped.

[0013] The test sequence parameters are corrected based on the hot spot temperature and the life loss factor, and an overload capacity and life loss report is automatically generated after the test is completed.

[0014] Preferably, historical power output data of new energy sources in the target area is collected, and based on the historical power output data, a test sequence including normal load, heavy load, overload, and reverse power flow segments is generated using a random-probabilistic fusion algorithm, including:

[0015] Collection coverage for no less than 1 year, with a sampling interval of [missing information]. Wind power and solar power active power curves And perform linear interpolation of missing segments and Normalization;

[0016] Based on the active power curves of wind power and photovoltaic power We fitted Weibull and log-normal distributions to wind speed and light intensity respectively, and constructed a joint probability density using Gaussian Copula to maintain the correlation.

[0017] For each sampling point The combined output is generated according to the following formula:

[0018]

[0019] in Normalized load ratio;

[0020] when satisfy Classified as a normal load segment, For heavy load sections, For overload section, This is the reverse current flow section;

[0021] A three-point moving average is performed on the classification results of the normal load segment, the overload segment, and the reverse power flow segment to obtain a test sequence that includes the normal, heavy load, overload, and reverse power flow segments.

[0022] in, To contribute to the synthesis of new energy sources; The active power curves for wind power and photovoltaic power are the historical output curves. This is the random disturbance amplification factor; The joint probability density; The quantile function is a normal distribution and u t~U(0,1); The rated capacity of the transformer under test; It is an adjustable amplitude modulation factor; This represents a typical load cycle within 24 hours. Used to divide the test segments.

[0023] Preferably, the programmable inverter-energy storage device is connected in parallel to the low-voltage side of the transformer under test, and the transformer under test is cyclically loaded according to the test sequence, including:

[0024] The isolation transformer connects the AC output terminal of the programmable inverter-energy storage device to the low-voltage side bus of the transformer under test, completing the electrical connection and communication interface configuration;

[0025] The test sequence is sent to the inverter control module in the form of time-target power pairs, and the number of cycles, overload ratio and single segment duration are set.

[0026] The inverter device is based on the target power value at time t. Adjust the actual output power in real time To satisfy the error limit relationship:

[0027] ;

[0028] Prediction time window before the power stage switching point of the test sequence Internally, the control module calculates the power slope in advance:

[0029]

[0030] And based on this slope, a linear smooth adjustment is performed to achieve a flexible transition between power segments;

[0031] The measured power of the transformer primary side is collected synchronously during the loading process. When satisfied If any of the following faults are detected: current exceeds the set limit, voltage exceeds the limit, or DC bias is abnormal as determined by the inverter, the control system will automatically pause the loading process and record the abnormal segment data.

[0032] in, Allowable loading error; Total loading test duration; This is the power deviation alarm threshold.

[0033] Preferably, the power stage switching points include: normal-heavy load, heavy load-overload, and overload-reverse transmission.

[0034] Preferably, the primary and secondary voltage and current are acquired by a synchronous phasor measurement unit, and the top oil temperature, winding hot spot temperature, and oil gas content are acquired simultaneously to obtain transformer measurement values, including:

[0035] During transformer operation, corresponding electrical signals are obtained from the voltage and current sampling circuits on the primary and secondary sides, and the data is aligned based on a unified time reference.

[0036] Representative locations were selected at the top of the transformer tank, the middle of the winding, and the bottom to continuously acquire data on the changes in top oil temperature and winding hot spot temperature.

[0037] Online sampling and analysis of transformer operating oil samples were performed to extract the content of typical gases, including acetylene, hydrogen, and methane, to reflect thermal aging and discharge trends.

[0038] The electrical signals, the change data, and the gas content are organized into the transformer measurement values ​​according to a unified time axis.

[0039] Preferably, the transformer measurements are input into a coupled thermal-lifetime model to calculate the hot spot temperature and lifetime loss factor, including:

[0040] The transformer measurements are divided into equal-length discrete time periods in chronological order.

[0041] Within each discrete time period, the winding hot spot temperature is calculated using a coupled thermal model;

[0042] Based on the hotspot temperature, the corresponding lifetime loss factor is generated by calling the lifetime model.

[0043] The lifetime loss factor is accumulated over all discrete time periods to obtain the real-time lifetime loss curve of the test process, and an overload stop judgment is triggered when the accumulated value reaches a preset threshold.

[0044] Preferably, the formula for calculating the lifetime loss factor of the hot spot temperature is:

[0045]

[0046] in, This refers to the hot spot temperature of the winding. Real-time top oil temperature; Heating of hot spots under rated load conditions; This is the real-time load current; This is the transformer's rated current. Load index; This is the lifespan acceleration factor; The reference temperature is 100°C; LLF is the lifetime loss factor.

[0047] Preferably, the test sequence parameters are corrected based on the hotspot temperature and the lifetime loss factor to ensure that the target temperature rise error is within the allowable range, including:

[0048] At the end of each loading cycle, the winding hot spot temperature and cumulative lifetime loss factor are read.

[0049] Calculate the deviation between the hot spot temperature and the target temperature rise. If the absolute value of the deviation is greater than the allowable error or the cumulative life loss factor exceeds the threshold, proceed to the next step; otherwise, maintain the current test sequence.

[0050] Adjust the target power and duration for the next cycle;

[0051] Insert the corrected target power and duration into the corresponding positions in the test sequence, and execute in the next loop;

[0052] When the hot spot temperature deviation of two consecutive cycles does not exceed the allowable error and the cumulative lifetime loss factor no longer increases to the threshold, the test sequence is determined to have converged, and the correction process ends.

[0053] Preferably, the formula for correcting the target power and duration of the next cycle is:

[0054]

[0055] in, This refers to the hot spot temperature of the winding. The target hotspot temperature; This represents the current cumulative lifespan loss factor. This is the lifespan loss threshold; and The target power and duration are as follows, before correction. and The corrected target power and duration.

[0056] Preferably, after the test is completed, an overload capacity and life loss report is automatically generated and uploaded to the management platform, including:

[0057] After the entire testing process is completed, the load data, hot spot temperature records, life loss factors and corresponding time tags for all time periods are classified and summarized.

[0058] Based on the summarized data, the temperature rise trend curve, life loss curve and hot spot temperature distribution map of the transformer under different load levels were plotted, and abnormal points or sudden change segments were marked.

[0059] A standardized test report file is automatically generated according to a preset template, which includes the test number, test scenario description, key parameter changes, anomaly triggering conditions, and evaluation conclusions.

[0060] The generated test report file is named with a unique identifier and uploaded to the designated power equipment health management platform or assessment system through a secure channel for subsequent scheduling, operation and maintenance or risk warning.

[0061] After successful upload, record the report generation time, upload time, and platform return status, and archive a local copy to complete the entire closed loop.

[0062] This invention also provides an automatic testing system for power grid transformer overload caused by new energy sources, characterized in that it includes:

[0063] Test sequence construction module: Collects historical data of renewable energy output in the target area, and generates test sequences including normal load, heavy load, overload and reverse power flow sections based on the historical data of renewable energy output;

[0064] Test transformer cyclic loading module: The programmable inverter-energy storage device is connected in parallel to the low voltage side of the test transformer, and the test transformer is cyclically loaded according to the test sequence;

[0065] Test module: Calculates hot spot temperature and life loss factor based on the transformer measurement values; if the life loss factor or the hot spot temperature reaches a preset threshold, stops the cyclic loading process; corrects the test sequence parameters according to the hot spot temperature and the life loss factor, and automatically generates an overload capacity and life loss report after the test is completed.

[0066] Furthermore, historical power output data of new energy sources in the target area is collected, and based on this historical power output data, a random-probabilistic fusion algorithm is used to generate test sequences including normal load, heavy load, overload, and reverse power flow segments, including:

[0067] Collection coverage for no less than 1 year, with a sampling interval of [missing information]. Wind power and solar power active power curves And perform linear interpolation of missing segments and Normalization;

[0068] Based on the active power curves of wind power and photovoltaic power We fitted Weibull and log-normal distributions to wind speed and light intensity respectively, and constructed a joint probability density using Gaussian Copula to maintain the correlation.

[0069] For each sampling point The combined output is generated according to the following formula:

[0070]

[0071] in Normalized load ratio;

[0072] when satisfy Classified as a normal load segment, For heavy load sections, For overload section, This is the reverse current flow section;

[0073] A three-point moving average is performed on the classification results of the normal load segment, the overload segment, and the reverse power flow segment to obtain a test sequence that includes the normal, heavy load, overload, and reverse power flow segments.

[0074] in, To contribute to the synthesis of new energy sources; The active power curves for wind power and photovoltaic power are the historical output curves. This is the random disturbance amplification factor; The joint probability density; The quantile function is a normal distribution and u t ~U(0,1); The rated capacity of the transformer under test; It is an adjustable amplitude modulation factor; This represents a typical load cycle within 24 hours. Used to divide the test segments.

[0075] Furthermore, the programmable inverter-energy storage device is connected in parallel to the low-voltage side of the transformer under test, and the transformer under test is cyclically loaded according to the test sequence, including:

[0076] The isolation transformer connects the AC output terminal of the programmable inverter-energy storage device to the low-voltage side bus of the transformer under test, completing the electrical connection and communication interface configuration;

[0077] The test sequence is sent to the inverter control module in the form of time-target power pairs, and the number of cycles, overload ratio and single segment duration are set.

[0078] The inverter device is based on the target power value at time t. Adjust the actual output power in real time To satisfy the error limit relationship:

[0079] ;

[0080] Prediction time window before the power stage switching point of the test sequence Internally, the control module calculates the power slope in advance:

[0081]

[0082] And based on this slope, a linear smooth adjustment is performed to achieve a flexible transition between power segments;

[0083] The measured power of the transformer primary side is collected synchronously during the loading process. When satisfied If any of the following faults are detected: current exceeds the set limit, voltage exceeds the limit, or DC bias is abnormal as determined by the inverter, the control system will automatically pause the loading process and record the abnormal segment data.

[0084] in, Allowable loading error; Total loading test duration; This is the power deviation alarm threshold.

[0085] Furthermore, the power stage switching points include: normal-heavy load, heavy load-overload, and overload-reverse transmission.

[0086] Furthermore, the primary and secondary voltage and current are acquired through a synchronous phasor measurement unit, and the top oil temperature, winding hot spot temperature, and oil gas content are simultaneously acquired to obtain transformer measurement values, including:

[0087] During transformer operation, corresponding electrical signals are obtained from the voltage and current sampling circuits on the primary and secondary sides, and the data is aligned based on a unified time reference.

[0088] Representative locations were selected at the top of the transformer tank, the middle of the winding, and the bottom to continuously acquire data on the changes in top oil temperature and winding hot spot temperature.

[0089] Online sampling and analysis of transformer operating oil samples were performed to extract the content of typical gases, including acetylene, hydrogen, and methane, to reflect thermal aging and discharge trends.

[0090] The electrical signals, the change data, and the gas content are organized into the transformer measurement values ​​according to a unified time axis.

[0091] Furthermore, the transformer measurements are input into a coupled thermal-lifetime model to calculate the hot spot temperature and lifetime loss factor, including:

[0092] The transformer measurements are divided into equal-length discrete time periods in chronological order.

[0093] Within each discrete time period, the winding hot spot temperature is calculated using a coupled thermal model;

[0094] Based on the hotspot temperature, the corresponding lifetime loss factor is generated by calling the lifetime model.

[0095] The lifetime loss factor is accumulated over all discrete time periods to obtain the real-time lifetime loss curve of the test process, and an overload stop judgment is triggered when the accumulated value reaches a preset threshold.

[0096] Furthermore, the formula for calculating the lifetime loss factor of the hotspot temperature is as follows:

[0097]

[0098] in, This refers to the hot spot temperature of the winding. Real-time top oil temperature; Heating of hot spots under rated load conditions; This is the real-time load current; This is the transformer's rated current. Load index; This is the lifespan acceleration factor; The reference temperature is 100°C; LLF is the lifetime loss factor.

[0099] Furthermore, the test sequence parameters are corrected based on the hotspot temperature and the lifetime loss factor to ensure that the target temperature rise error is within the allowable range, including:

[0100] At the end of each loading cycle, the current winding hot spot temperature and cumulative lifetime loss factor are read;

[0101] Calculate the deviation between the hot spot temperature and the target temperature rise. If the absolute value of the deviation is greater than the allowable error or the cumulative life loss factor exceeds the threshold, proceed to the next step; otherwise, maintain the current test sequence.

[0102] Adjust the target power and duration for the next cycle;

[0103] Insert the corrected target power and duration into the corresponding positions in the test sequence, and execute in the next loop;

[0104] When the hot spot temperature deviation of two consecutive cycles does not exceed the allowable error and the cumulative lifetime loss factor no longer increases to the threshold, the test sequence is determined to have converged, and the correction process ends.

[0105] Furthermore, the formula for correcting the target power and duration for the next cycle is as follows:

[0106]

[0107] in, This refers to the hot spot temperature of the winding. The target hotspot temperature; This represents the current cumulative lifespan loss factor. This is the lifespan loss threshold; and The target power and duration are as follows, before correction. and The corrected target power and duration.

[0108] Furthermore, after the test is completed, an overload capacity and lifespan loss report is automatically generated and uploaded to the management platform, including:

[0109] After the entire testing process is completed, the load data, hot spot temperature records, life loss factors and corresponding time tags for all time periods are classified and summarized.

[0110] Based on the summarized data, the temperature rise trend curve, life loss curve and hot spot temperature distribution map of the transformer under different load levels were plotted, and abnormal points or sudden change segments were marked.

[0111] A standardized test report file is automatically generated according to a preset template, which includes the test number, test scenario description, key parameter changes, anomaly triggering conditions, and evaluation conclusions.

[0112] The generated test report file is named with a unique identifier and uploaded to the designated power equipment health management platform or assessment system through a secure channel for subsequent scheduling, operation and maintenance or risk warning.

[0113] After successful upload, record the report generation time, upload time, and platform return status, and archive a local copy to complete the entire closed loop.

[0114] The present invention discloses the following technical effects:

[0115] (1) Based on historical wind power and photovoltaic power output data, this invention integrates probability models and disturbance functions to construct a controllable “normal-heavy load-overload-reverse transmission” cyclic test sequence, which can realistically simulate the dynamic impact of frequent power flow switching on transformers and make up for the shortcomings of existing technologies that cannot reproduce typical new energy operation scenarios under test conditions.

[0116] (2) The present invention automatically loads the test sequence onto the low-voltage side of the transformer through the inverter-energy storage control system, and adjusts the loading power and duration in real time by combining hot spot temperature rise and life loss factor feedback, forming a closed-loop mechanism of "test-measurement-evaluation-correction", avoiding manual debugging and improving the accuracy and efficiency of the test.

[0117] (3) The present invention adopts a coupled hot spot temperature-life loss model, which dynamically calculates the life loss factor based on the online measured top oil temperature, current and other parameters. It can accurately quantify the aging degree of transformers under different overload levels and overcome the defects of traditional empirical models such as weak applicability and large delay.

[0118] (4) By judging whether the hot spot temperature and life factor exceed the limit in real time, the present invention can automatically trigger the test to stop and output the overload bearing boundary, making the transformer's limit operation level concrete and data-driven, filling the problem that the existing technology cannot quantify the transformer bearing limit under the impact of new energy.

[0119] (5) This invention realizes the unified collection, time alignment and archiving of electrical parameters, hot spot temperature and gas in oil throughout the entire process, and automatically generates standardized evaluation reports, providing structured and traceable data support for power grid regulation, equipment operation and maintenance and life cycle management.

[0120] (6) The method of the present invention does not rely on a fixed site structure and can be deployed on a test platform or in a real station environment. It has the versatility to conduct rapid load testing and life evaluation of any type of power transformer under high-penetration new energy conditions, which significantly improves the application scope and engineering practicality. Attached Figure Description

[0121] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the embodiments will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0122] Figure 1 This is a flowchart of a method provided in an embodiment of the present invention. Detailed Implementation

[0123] 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 skilled in the art without creative effort are within the scope of protection of the present invention.

[0124] The purpose of this invention is to provide an automatic testing method for overload of power grid transformers caused by new energy sources. It realizes closed-loop control of the entire process of loading, measurement, evaluation and correction, thereby improving the authenticity of the test, the accuracy of life assessment and the value of field application.

[0125] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0126] Figure 1 The method flowchart provided in the embodiments of the present invention is as follows: Figure 1 As shown, this invention provides an automatic testing method for power grid transformer overload caused by new energy sources, comprising:

[0127] Step 100: Collect historical power output data of new energy sources in the target area, and based on the historical power output data, use a random-probabilistic fusion algorithm to generate test sequences including normal load, heavy load, overload, and reverse power flow segments; including:

[0128] Collection coverage for no less than 1 year, with a sampling interval of [missing information]. Wind power and solar power active power curves And perform linear interpolation of missing segments and Normalization;

[0129] Based on the active power curves of wind power and photovoltaic power We fitted Weibull and log-normal distributions to wind speed and light intensity respectively, and constructed a joint probability density using Gaussian Copula to maintain the correlation.

[0130] For each sampling point The combined output is generated according to the following formula:

[0131]

[0132] in Normalized load ratio;

[0133] when satisfy Classified as a normal load segment, For heavy load sections, For overload section, This is the reverse current flow section;

[0134] A three-point moving average is performed on the classification results of the normal load segment, the overload segment, and the reverse power flow segment to obtain a test sequence that includes the normal, heavy load, overload, and reverse power flow segments.

[0135] in, To contribute to the synthesis of new energy sources; The active power curves for wind power and photovoltaic power are the historical output curves. This is the random disturbance amplification factor; The joint probability density; The quantile function is a normal distribution and u t ~U(0,1); The rated capacity of the transformer under test; It is an adjustable amplitude modulation factor; This represents a typical load cycle within 24 hours. Used to divide the test segments.

[0136] Step 200: Connect the programmable inverter-energy storage device in parallel to the low-voltage side of the transformer under test, and perform cyclic loading on the transformer under test according to the test sequence; including:

[0137] The isolation transformer connects the AC output terminal of the programmable inverter-energy storage device to the low-voltage side bus of the transformer under test, completing the electrical connection and communication interface configuration;

[0138] The test sequence is sent to the inverter control module in the form of time-target power pairs, and the number of cycles, overload ratio and single segment duration are set.

[0139] The inverter device is based on the target power value at time t. Adjust the actual output power in real time To satisfy the error limit relationship:

[0140]

[0141] Prediction time window before the power stage switching point of the test sequence Internally, the control module calculates the power slope in advance:

[0142]

[0143] And based on this slope, a linear smooth adjustment is performed to achieve a flexible transition between power segments;

[0144] The measured power of the transformer primary side is collected synchronously during the loading process. When satisfied If any of the following faults are detected: current exceeds the set limit, voltage exceeds the limit, or DC bias is abnormal as determined by the inverter, the control system will automatically pause the loading process and record the abnormal segment data.

[0145] in, Allowable loading error; Total loading test duration; This is the power deviation alarm threshold. Power level switching points include: Normal-Heavy Load, Heavy Load-Overload, and Overload-Reverse Feed.

[0146] Step 300: Acquire primary and secondary voltage and current using a synchronous phasor measurement unit, and simultaneously acquire top oil temperature, winding hot spot temperature, and oil gas content to obtain transformer measurement values; including:

[0147] During transformer operation, corresponding electrical signals are obtained from the voltage and current sampling circuits on the primary and secondary sides, and the data is aligned based on a unified time reference.

[0148] Representative locations were selected at the top of the transformer tank, the middle of the winding, and the bottom to continuously acquire data on the changes in top oil temperature and winding hot spot temperature.

[0149] Online sampling and analysis of transformer operating oil samples were performed to extract the content of typical gases, including acetylene, hydrogen, and methane, to reflect thermal aging and discharge trends.

[0150] The electrical signals, the change data, and the gas content are organized into the transformer measurement values ​​according to a unified time axis.

[0151] Step 400: Input the transformer measurements into the coupled thermal-lifetime model to calculate the hot spot temperature and lifetime loss factor; including:

[0152] The transformer measurements are divided into equal-length discrete time periods in chronological order.

[0153] Within each discrete time period, the winding hot spot temperature is calculated using a coupled thermal model;

[0154] Based on the hotspot temperature, the corresponding lifetime loss factor is generated by calling the lifetime model.

[0155] The lifetime loss factor is accumulated over all discrete time periods to obtain the real-time lifetime loss curve of the test process, and an overload stop judgment is triggered when the accumulated value reaches a preset threshold.

[0156] Preferably, the formula for calculating the lifetime loss factor of the hot spot temperature is:

[0157]

[0158] in, This refers to the hot spot temperature of the winding. Real-time top oil temperature; Heating of hot spots under rated load conditions; This is the real-time load current; This is the transformer's rated current. Load index; This is the lifespan acceleration factor; The reference temperature is 100°C; LLF is the lifetime loss factor.

[0159] The coupled thermal-life model employs a dynamic differential equation system, with real-time step size control ensuring local errors do not exceed preset values; the charging and discharging capacity of the programmable inverter-energy storage device is not less than twice the rated capacity of the transformer under test; and the time synchronization accuracy of the synchronous measurement is not greater than 1 μs.

[0160] Step 500: If the lifetime loss factor or the hot spot temperature reaches a preset threshold, stop the cyclic loading process; including:

[0161] At the end of each loading cycle, the current winding hot spot temperature and cumulative lifetime loss factor are read;

[0162] Calculate the deviation between the hot spot temperature and the target temperature rise. If the absolute value of the deviation is greater than the allowable error or the cumulative life loss factor exceeds the threshold, proceed to the next step; otherwise, maintain the current test sequence.

[0163] Adjust the target power and duration for the next cycle;

[0164] Insert the corrected target power and duration into the corresponding positions in the test sequence, and execute in the next loop;

[0165] When the hot spot temperature deviation of two consecutive cycles does not exceed the allowable error and the cumulative lifetime loss factor no longer increases to the threshold, the test sequence is determined to have converged, and the correction process ends.

[0166] Step 600: Correct the test sequence parameters based on the hotspot temperature and the lifetime loss factor to ensure that the target temperature rise error is within the allowable range; the formula for correcting the target power and duration of the next cycle is:

[0167]

[0168] in, This refers to the hot spot temperature of the winding. The target hotspot temperature; This represents the current cumulative lifespan loss factor. This is the lifespan loss threshold; and The target power and duration are as follows, before correction. and The corrected target power and duration.

[0169] Step 700: After the test is completed, an overload capacity and lifespan loss report will be automatically generated and uploaded to the management platform; including:

[0170] After the entire testing process is completed, the load data, hot spot temperature records, life loss factors and corresponding time tags for all time periods are classified and summarized.

[0171] Based on the summarized data, the temperature rise trend curve, life loss curve and hot spot temperature distribution map of the transformer under different load levels were plotted, and abnormal points or sudden change segments were marked.

[0172] A standardized test report file is automatically generated according to a preset template, which includes the test number, test scenario description, key parameter changes, anomaly triggering conditions, and evaluation conclusions.

[0173] The generated test report file is named with a unique identifier and uploaded to the designated power equipment health management platform or assessment system through a secure channel for subsequent scheduling, operation and maintenance or risk warning.

[0174] After successful upload, record the report generation time, upload time, and platform return status, and archive a local copy to complete the entire closed loop.

[0175] Specifically, the process of step 100 in this embodiment is as follows:

[0176] This embodiment extracts historical active power curves of wind power and photovoltaic power for a target area for at least one year from the scheduling database, with a fixed sampling time interval, such as five minutes. To ensure data integrity, missing data segments in the curves are filled using linear interpolation. Then, each wind power and photovoltaic curve is normalized based on its respective historical maximum value, bringing all data to between zero and one. After normalization, this embodiment fits a Weibull probability distribution to the wind speed data samples and a log-normal distribution to the solar irradiance samples. To reflect the correlation between wind power and photovoltaic fluctuations, this embodiment uses the Gaussian Copula method to construct a joint probability density distribution. The aforementioned "joint probability density" refers to a joint distribution that preserves the correlation between the two variables by mapping the independent distributions of wind speed and solar irradiance to a unified normal space and then inversely transforming them back to the original space, serving as the statistical basis for subsequent random disturbance generation.

[0177] This embodiment generates "synthetic renewable energy output" by superimposing random disturbances based on normalized historical output. The specific process includes: First, the amplitude of the random disturbance is referenced to the standard deviation in the joint probability density function, reflecting the historical correlation fluctuation range. Second, a random disturbance amplification factor is added, typically chosen between 0.05 and 0.20, empirically selected based on the historical data change rate to balance sequence repeatability and volatility. Furthermore, pseudo-randomness is introduced through the inverse operation of the normal distribution quantile function; the required disturbance sequence is generated from uniformly distributed pseudo-random numbers between zero and one. The synthetic renewable energy output is also superimposed with a typical 24-hour load cycle through an adjustable amplitude modulation factor, with the modulation factor ranging from zero to 0.30, to match the daily variation characteristics of the actual electricity load. Finally, the synthetic renewable energy output value is divided by the measured transformer rated capacity to obtain the normalized load ratio. This normalized load ratio serves as the basis for subsequent operating condition classification.

[0178] This embodiment classifies each sampling point into different load conditions based on the numerical range of the normalized load ratio: when the normalized load ratio meets the following conditions... Classified as a normal load segment, For heavy load sections, For overload section, This is the reverse power flow section. To reduce the impact of data jitter on the stability of operating condition discrimination, this embodiment also performs a three-point moving average processing on the operating condition discrimination results. That is, the operating condition label at each time point is determined by averaging the discrimination results of that point and the sampling points before and after it, which can effectively suppress misjudgments caused by single-point data fluctuations. The operating condition labels after the above moving average form a complete test sequence including multiple operating scenarios such as normal, heavy load, overload, and reverse power flow. The test sequence can be directly used as program instruction output to control the inverter or energy storage device to perform quantitative and timed automatic loading on the transformer under test, realizing high-fidelity simulation testing of overload conditions.

[0179] Specifically, the process of step 200 in this embodiment is as follows:

[0180] In this embodiment, the AC output terminal of the programmable inverter-energy storage device is electrically connected to the low-voltage bus of the transformer under test via an isolation transformer, and the parameter configuration and functional debugging of the communication interface are completed. Subsequently, according to the aforementioned generated test sequence, the target power value and its duration at each moment are sent to the control module of the inverter device in the form of a "time-target power pair". At the same time, during the sending process, parameters such as the number of times the test sequence needs to be executed repeatedly, the ratio of each overload segment to the rated capacity, and the operating time range of a single load segment are set synchronously. The "programmable inverter-energy storage device" is a power electronic device that can flexibly output power according to the command and realize bidirectional energy flow, and has both inverter function and energy storage regulation. The "test sequence" is a power command set formed by relying on the historical output of new energy sources and integrating probabilistic disturbance mechanisms. It includes load segments such as normal load, heavy load, overload, and reverse power flow, and each item is the target power demand at a specific time point.

[0181] Furthermore, "number of cycles" refers to the number of rounds in which the entire test sequence is repeated completely, "overload ratio" refers to the rate of increase in the rated capacity of the transformer by different sections, and "single section duration" is the duration of maintenance for each typical load section. During loading, the inverter needs to automatically adjust the actual output power according to the target power value corresponding to the current time point, and continuously monitor the deviation between the actual output and the target value. When the deviation exceeds the allowable loading error range, which is generally set as a percentage of the rated power, such as ±3 to ±5 percentage points, the system will automatically alarm and suspend output if the limit is exceeded. Before the power level switching of the test sequence, the control module will enter the prediction time window in advance, dynamically calculate the rate of power change based on the upcoming target power change and the prediction time window duration, and gradually adjust the output power linearly according to this rate to achieve a smooth transition between sections and prevent electrical shocks caused by sudden power changes. Throughout the entire process, physical quantities such as primary side current, voltage, and active power of the transformer under test are collected and monitored in real time. If the current exceeds the preset safety limit, the voltage is outside the high and low limits, or an abnormal DC bias is detected inside the inverter (i.e., the iron core may be magnetically saturated due to the DC component), the control system automatically pauses loading and immediately records all test data during the period of triggering the abnormality for subsequent analysis. The power deviation alarm threshold is a preset limit, such as 10% of the rated capacity. The total loading test duration is determined by multiplying the duration of a single sequence by the number of cycles.

[0182] In this embodiment, the power level switching points in the load test sequence are preferably defined as the transition times between the normal and heavy load segments, the heavy load and overload segments, and the overload and reverse power flow segments, serving as key nodes for flexible transition control. Specifically, this embodiment automatically sets the starting trigger point for the power slope smoothing adjustment algorithm at the time points when the power level of the test sequence changes—that is, when switching from the normal load segment to the heavy load segment, from the heavy load segment to the overload segment, and from the overload segment to the reverse power flow segment. Within a preset prediction time window before these power level switching points arrive, the control module obtains the power targets for the current and target segments in advance, calculates the power change amplitude and time window length, and thus dynamically determines the rate of change during each power switching process. In actual control, the inverter-energy storage device continuously and uniformly adjusts the power output according to this rate of change, achieving a linear and smooth transition of the power value from the current level to the target level, avoiding mechanical or electrical shocks to the transformer caused by large or sudden adjustments. All switching processes and related parameters can be recorded in the control system log, enabling traceable and controllable management of the power switching process. The "prediction time window" is used to schedule power output adjustments in advance. The duration is generally set to a few seconds to tens of seconds, and can be set in real time according to the transformer's operating characteristics and on-site requirements to ensure smooth and reliable switching.

[0183] Optionally, this embodiment employs a synchronous phasor measurement unit to collect all core operating parameters of the transformer under test, achieving high timeliness and accuracy in obtaining transformer measurement values. Specifically, during the operation of the transformer under test, voltage and current sampling circuits are connected to both the primary and secondary sides. A standard time reference is established using a synchronous sampling clock, and all electrical signal data are synchronized at the millisecond level to achieve time alignment of data on both sides, obtaining accurate active and reactive power and phasor information. Simultaneously, multiple temperature sensors are installed in thermally sensitive areas such as the top of the transformer tank, the middle and lower parts of the windings, to record and continuously transmit data on the changes in top oil temperature, winding hot spot temperature, etc., over time, for dynamically assessing the thermal stress distribution of the transformer. Based on this, an online oil gas monitoring device is used to periodically or in real-time extract operating oil samples to analyze changes in the content of typical dissolved gases such as acetylene, hydrogen, and methane. The concentration and trends of these gas components can reflect potential fault states such as transformer thermal aging and partial discharge. All the electrical parameters, temperature data, and gas content information mentioned above are collected using a unified sampling clock and data communication protocol, and are archived synchronously with a consistent timestamp. They are then compiled in a unified manner in the form of "transformer measurement values" for transformer status assessment and anomaly early warning, enabling high-precision diagnosis of operating conditions and health status in a comprehensive and dynamic way.

[0184] Specifically, in this embodiment, the collected transformer measurements are divided into several discrete time periods of equal length, for example, each time period is one minute. Within each discrete time period, key data parameters such as top oil temperature and load current are compiled. The aforementioned "equal-length discrete time periods" refer to dividing the continuously sampled measurement data during operation into segments based on fixed-length time intervals, allowing data within each interval to be processed independently, ensuring the consistency of timeliness and accuracy of analysis for temperature, load, and other curves.

[0185] Within each defined discrete time period, this embodiment utilizes a coupled thermal model to calculate the winding hot spot temperature. The coupled thermal model is a calculation model that combines the electrical, thermodynamic, and material aging characteristics of the transformer. It can extrapolate the winding hot spot temperature within that time period based on real-time top oil temperature, load current, rated current, and transformer design parameters. The top oil temperature is obtained in real-time through a temperature sensor; the real-time load current is also directly collected by the online monitoring unit; the rated current and hot spot temperature rise under rated load are respectively the transformer nameplate parameters and design factory parameters, typically defined in the manufacturer's technical documentation. The "load index" is an empirical constant, commonly valued at 0.8, reflecting the nonlinear effect of load changes on heating.

[0186] Based on the winding hot spot temperature calculated for each time period, this embodiment further calls the lifetime model to evaluate the insulation aging rate for that time period and obtain the required lifetime loss factor. The lifetime model calculates the ratio of the actual aging rate of the insulation material to the reference rate using empirical lifetime acceleration theories (such as Arrhenius reaction rate theory) based on the difference between the hot spot temperature and the reference temperature, and uses this as the lifetime loss factor. The "lifetime acceleration coefficient" can be selected based on the chemical properties of the material and is usually an empirically determined value; the "reference temperature" is a widely used reference point for insulation lifetime conversion in the industry, generally based on 98 degrees Celsius; the aforementioned lifetime loss factor is specifically used to measure the rate of insulation lifetime consumption at a specific temperature level. A value greater than one indicates accelerated loss, while a value less than one indicates slowed aging.

[0187] Finally, the lifetime loss factors obtained from all equal-length discrete time periods are cumulatively accumulated one by one, that is, the lifetime loss is accumulated chronologically to obtain the real-time lifetime loss curve for the entire test process. Each factor usually needs to be multiplied by the duration of the corresponding time period (e.g., hours) to obtain the absolute loss value. When the accumulated result reaches or exceeds the preset lifetime loss threshold, this embodiment will trigger an overload stop judgment mechanism to actively stop the load and prevent excessive aging or damage to the transformer insulation material. The rolling accumulation process can intuitively reflect the lifetime loss trend of the transformer under test throughout the entire cycle, providing an accurate decision-making basis for equipment operation and maintenance and protection.

[0188] Specifically, the process of step 500 in this embodiment is as follows:

[0189] In this embodiment, during the cyclic loading process, the transformer's life loss factor and winding hot spot temperature are continuously monitored in real time, and these two core parameters are compared with their respective preset thresholds. When the life loss factor reaches or exceeds the life loss threshold, or the winding hot spot temperature reaches or exceeds the hot spot temperature threshold, the cyclic loading is immediately stopped. The aforementioned "life loss factor" is an indicator reflecting the insulation aging process, calculated using a life model based on data such as the aging rate of the transformer insulation material and the hot spot temperature. Its threshold is generally set according to the maximum allowable cumulative life loss of the transformer insulation material, typically selected as approximately 1% of the transformer's design life as the protection trigger point. The "hot spot temperature threshold" is set according to the national standards for transformers or the manufacturer's limit requirements, with common reference values ​​such as 140 degrees Celsius. Upon reaching any of the above thresholds, this embodiment immediately issues a shutdown command to the programmable inverter-energy storage device, causing it to actively disconnect from the electrical connection with the low-voltage side of the transformer and simultaneously stop issuing subsequent power commands. At the same time, the current operating status, the reason for the stop, and the measured data are fully recorded in the system log, and an automatic alarm can be issued if desired, so that relevant personnel can follow up and handle the situation promptly and conduct subsequent analysis. By following the above steps, the safety of the loading process and the reliability of the equipment operation are ensured, and equipment damage or failure caused by overload or excessive insulation aging is prevented.

[0190] Specifically, the process of step 600 in this embodiment is as follows:

[0191] At the end of each loading cycle, this embodiment employs an online data acquisition and processing mechanism to first read the actual hotspot temperature and cumulative lifetime loss factor during the current test cycle in real time. The aforementioned "hotspot temperature" refers to the temperature value of the hottest location of the transformer winding, calculated in real time using a coupled thermal model. This temperature comprehensively considers the influence of multiple factors such as load, electrical parameters, and ambient temperature. The "lifetime loss factor" is an insulation aging acceleration index accumulated through the lifetime model, reflecting the total lifetime loss of the transformer under the current test load. Both are stored synchronously in the test sequence.

[0192] After each cycle, this embodiment performs an error assessment on the actual hot spot temperature and the preset target temperature rise. The target temperature rise is the ideal hot spot temperature increase value specified in the transformer operation design, and the allowable error is the maximum allowable fluctuation range in engineering (e.g., ±5 degrees Celsius). If the absolute deviation between the actual hot spot temperature and the target temperature rise exceeds the above-mentioned allowable error, or the cumulative life loss factor has exceeded the preset life loss threshold (this threshold is obtained from the insulation material and safe life assessment), then the test sequence parameter correction process is initiated; otherwise, it is determined that all parameters in the current cycle meet expectations, the test sequence parameters for this round remain unchanged, and the next routine loading is initiated.

[0193] When adjustments are needed, this embodiment adjusts the target power and duration of the next cycle according to a preset adaptive adjustment rule based on the winding hotspot temperature, target temperature rise, cumulative life loss factor, and corresponding life loss threshold. The adjustment strategy is as follows: when the hotspot temperature is greater than the target temperature rise, or the life loss factor is close to or exceeds the threshold, the target power for the next cycle is moderately reduced, and the duration is correspondingly shortened; conversely, if the hotspot temperature is lower than the target temperature rise and the life loss factor is lower than the threshold, the target power can be moderately increased, and the duration appropriately extended. The specific adjustment range of the target power and duration is determined by combining the set step parameters with historical correction experience to gradually converge the hotspot temperature to the target value while avoiding exceeding the life loss limit.

[0194] The corrected target power and duration will be inserted into the corresponding positions in the test sequence, becoming the new parameters for the next round of execution. In subsequent loading, the system continues to automatically load according to the corrected parameters, continuously collecting data, making judgments, and making corrections to form a closed-loop optimization. This process is iterated repeatedly until the deviation between the hot spot temperature and the target temperature rise in two consecutive rounds does not exceed the allowable error, and the cumulative life loss factor is within the life loss threshold range. At this point, this embodiment determines that the test sequence parameters have converged, terminates the correction process, and proceeds to the next stage of performance analysis and evaluation, ensuring that the transformer's operational safety and test effectiveness are fully guaranteed.

[0195] Specifically, step 700 in this embodiment includes:

[0196] In this embodiment, after the entire testing process is completed, the load data, hotspot temperature records, lifespan loss factors, and their corresponding time stamp information acquired during the testing period are first automatically classified and summarized. The classification process includes splitting and archiving all raw measurement data according to load level, time sequence, and operating condition stage, and calculating the mean, peak value, and abnormal fluctuation range for different load segments to facilitate subsequent trend analysis and report output. All data processing processes use a unified time base to achieve accurate alignment between various measurement data and events.

[0197] After completing the data classification and summarization, this embodiment automatically plots the temperature rise trend curves, life loss curves, and hot spot temperature distribution maps of the transformer under various load levels based on the processed data. The curve plotting steps include segmenting and fitting the temperature rise and loss of each load segment according to the time-series data, and automatically identifying abnormal points or abrupt changes using statistical detection methods, which are then clearly highlighted or marked in the graph. All the above graphs use a standardized coordinate system, which facilitates a visual display of the transformer's thermal stress, loss process, and potential anomalies under different operating conditions, improving the scientific nature and traceability of subsequent maintenance decisions.

[0198] Finally, in this embodiment, according to a pre-set standard report template, the test number, test scenario description, key parameter changes, abnormal triggering conditions, and evaluation conclusions are embedded into an automatically generated overload capacity and life loss report file, and saved with a unique identifier. The report is automatically uploaded to the designated power equipment health management platform or evaluation system via an encrypted secure channel (such as HTTPS or VPN). After the upload is completed, the system synchronously records the report generation time, upload time, and the status or feedback information returned by the platform, and retains a copy of the original report on a local dedicated storage medium, realizing a closed-loop management system for the entire process of test data and analysis conclusions, ensuring security and traceability. The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for automatically testing power grid transformer overload caused by new energy, characterized in that, The method comprises the following steps: Collecting new energy output historical data of a target area, and generating a test sequence containing normal load, heavy load, overload and reverse power flow section based on the new energy output historical data; Connecting a programmable inverter-energy storage device to the low-voltage side of the transformer to be tested, and performing cyclic loading on the transformer to be tested according to the test sequence; Obtaining transformer measurement values by collecting primary and secondary voltage and current, and synchronously collecting top oil temperature, winding hot spot temperature and gas content in the oil; Calculating hot spot temperature and life loss factor based on the transformer measurement values; If the life loss factor or the hot spot temperature reaches a preset threshold, stopping the process of cyclic loading; According to the hot spot temperature and the life loss factor, correcting the test sequence parameters, and automatically generating an overload carrying capacity and life loss report after the test is completed; Inputting the transformer measurement values into a coupled thermal-life model to calculate the hot spot temperature and the life loss factor, comprising: Dividing the transformer measurement values into equal-length discrete time periods in chronological order; In each discrete time period, the winding hot spot temperature is calculated using a coupled thermal model; Based on the hot spot temperature, the corresponding life loss factor is generated by calling a life model; Rolling up the life loss factors of all discrete time periods to obtain a real-time life loss curve of the test process, and triggering an overload stop determination when the accumulated value reaches a preset threshold; The life loss factor calculation formula of the hot spot temperature is: wherein, is the winding hot-spot temperature; is the real-time top-oil temperature; is the hot-spot temperature rise under rated load conditions; is the real-time load current; is the transformer rated current; is the load factor; is the life acceleration factor; is the reference temperature; and LLF is the life loss factor.

2. The method of claim 1, wherein, Collecting new energy output historical data of a target area, and generating a test sequence containing normal load, heavy load, overload and reverse power flow section based on the new energy output historical data using a random-probability fusion algorithm, comprising: The wind power and photovoltaic active power curves covering not less than 1 year, with a sampling interval of , are collected, linear interpolation of missing segments is performed, and normalization is performed.​ Based on wind power, photovoltaic active power curve , respectively, to wind speed and light amount fitting Weibull distribution and lognormal distribution, using Gaussian Copula to construct joint probability density to maintain correlation; For each sample point The synthetic output is generated by the following equation: wherein is the normalized load ratio; When satisfies is classified as a normal load section, is a heavy load section, is an overload section, is a reverse power flow section; Performing three-point sliding average on the classification results of the normal load section, the overload section and the reverse power flow section to obtain a test sequence containing normal, heavy load, overload and reverse power flow sections; wherein, is the synthetic new energy output; is the wind power and photovoltaic active power curve, i.e. historical output; is the random disturbance amplification coefficient; is the joint probability density; is the quantile function of normal distribution and u t ~ U(0, 1); is the rated capacity of the measured transformer; is the adjustable amplitude modulation factor; is the typical load cycle within 24 hours; is used to divide each test section.

3. The method of claim 1, wherein the method further comprises: Connecting a programmable inverter-energy storage device to the low-voltage side of the transformer to be tested, and performing cyclic loading on the transformer to be tested according to the test sequence, comprising: Isolating transformer connects the AC output end of the programmable inverter-energy storage device to the low-voltage side bus of the transformer to be tested, completing electrical connection and communication interface configuration; The test sequence is sent to the inverter device control module in the form of time-target power pairs, setting the number of cycles, overload ratio and single section duration; The inverter device is based on the target power value at time t. Adjust the actual output power in real time To satisfy the error limit relationship: ； a prediction time window before the power level switching point of the test sequence arrives Within the control module, the power ramp is calculated in advance: And performing linear smoothing adjustment according to the slope to realize flexible transition between power sections; Synchronous acquisition of measured power on the primary side of the transformer during the loading process When any of the following conditions are met or any of the following faults is detected: current exceeds the set limit, voltage is out of limit, or DC bias abnormality determined internally by the inverter, the control system automatically suspends the loading process and records the abnormal section data; wherein, is the allowed loading error; is the total loading test duration; is the power deviation alarm threshold.

4. The method of claim 3, wherein the new energy causes the transformer overload of the power grid to be automatically tested. The power level switching points include: normal-heavy load, heavy load-overload, overload-reverse sending.

5. The method of claim 1, wherein the method further comprises: Obtaining transformer measurement values by synchronously collecting primary and secondary voltage and current, and synchronously collecting top oil temperature, winding hot spot temperature and gas content in the oil, comprising: During the operation of the transformer, corresponding electrical signals are obtained from the voltage and current sampling circuits on the primary and secondary sides, and data alignment is performed based on a unified time reference; Representative positions are selected at the top of the transformer tank, the middle and lower parts of the winding, and the change data of the top oil temperature and the winding hot spot temperature are continuously obtained; On-line sampling and analysis of transformer operating oil samples, extraction of typical gas content including acetylene, hydrogen and methane, for reflecting thermal aging and discharge trend; The electrical signal, the change data and the gas content are arranged on a unified time axis as the transformer measurement value.

6. The method of claim 1, wherein, According to the hotspot temperature and the life loss factor, the test sequence parameters are corrected to ensure that the target temperature rise error is within the allowable range, including: At the end of each loading cycle, the winding hotspot temperature and the cumulative life loss factor are read; The deviation of the hotspot temperature from the target temperature rise is calculated, and if the absolute value of the deviation is greater than the allowable error or the cumulative life loss factor exceeds the threshold, the next step is entered, otherwise the current test sequence is maintained; The target power and duration of the next cycle are corrected; The corrected target power and duration are inserted into the corresponding position of the test sequence and executed in the next cycle; When the hotspot temperature deviations of two consecutive cycles are both within the allowable error and the cumulative life loss factor no longer grows to the threshold, it is determined that the test sequence has converged, and the correction process is ended.

7. The method of claim 6, wherein the method further comprises: The formula for correcting the target power and duration of the next cycle is: wherein, is the winding hot-spot temperature; is the target hot-spot temperature; is the current cumulative lifetime loss factor; is the lifetime loss threshold; is the target power and duration before correction; is the target power and duration before correction; is the target power and duration after correction; is the target power and duration after correction.

8. The method of claim 1, wherein the method further comprises: After completing the test, an overload carrying capacity and life loss report is automatically generated, including: After the entire test process is completed, the load data, hotspot temperature record, life loss factor and corresponding time label of all time periods are classified and summarized; Based on the summarized data, the temperature rise trend curve, life loss curve and hotspot temperature distribution graph of the transformer under different load levels are drawn, and abnormal points or sudden changes are marked; A standardized test report file containing test number, test scene description, key parameter changes, abnormal trigger conditions and evaluation conclusions is automatically generated according to a preset template; The generated test report file is named with a unique identifier and uploaded to a designated power equipment health management platform or evaluation system through a secure channel for subsequent dispatching, operation and maintenance or risk warning calls; After successful upload, the report generation time, upload time and platform return status are recorded, and a local copy is archived, completing the whole process closed loop.

9. A new energy caused grid transformer overload automatic test system, characterized in that, including: Test sequence construction module: collect new energy output historical data of the target area, and generate a test sequence containing normal load, heavy load, overload and reverse flow segment based on the new energy output historical data; Measured transformer cycle loading module: a programmable inverter-energy storage device is connected to the measured transformer low-voltage side, and the measured transformer is loaded according to the test sequence; Test module: obtain transformer measurement values by collecting primary and secondary voltage and current, and synchronously collecting top oil temperature, winding hotspot temperature and gas content in oil; calculate the hotspot temperature and life loss factor based on the transformer measurement values; if the life loss factor or the hotspot temperature reaches a preset threshold, stop the cycle loading process; According to the hotspot temperature and the life loss factor, the test sequence parameters are corrected, and an overload carrying capacity and life loss report is automatically generated after the test is completed; The transformer measurement values are input into the coupled thermal-life model to calculate the hotspot temperature and the life loss factor, including: The transformer measurement values are divided into equal length discrete time periods in chronological order; In each of the discrete time periods, the winding hot spot temperature is calculated by using a coupled thermal model; Based on the hot spot temperature, a corresponding life loss factor is generated by calling a life model; The life loss factors of all discrete time periods are rolled up to obtain a real-time life loss curve of the test process, and an overload stop judgment is triggered when the cumulative value reaches a preset threshold; The life loss factor calculation formula of the hot spot temperature is: wherein, is the winding hot-spot temperature; is the real-time top-oil temperature; is the hot-spot temperature rise under rated load conditions; is the real-time load current; is the transformer rated current; is the load index; is the life-acceleration factor; is the reference temperature; and LLF is the life-loss factor.

10. The new energy induced grid transformer overload automatic test system according to claim 9, characterized in that, Collecting historical data of new energy output in a target area, and based on the historical data of new energy output, a test sequence containing normal load, heavy load, overload and reverse power flow section is generated by using a random-probability fusion algorithm, including: The wind power and photovoltaic active power curves covering not less than 1 year, with a sampling interval of 1 minute are collected , and linear interpolation of missing segments and normalization are performed; Based on wind power, photovoltaic active power curve , respectively, to wind speed and light amount fitting Weibull distribution and lognormal distribution, using Gaussian Copula to construct joint probability density to maintain correlation; For each sample point The synthetic output is generated by the following equation: wherein is the normalized load ratio; When satisfies is classified as a normal load section, is a heavy load section, is an overload section, is a reverse power flow section; Performing three-point sliding average on the classification results of the normal load section, the overload section and the reverse power flow section to obtain a test sequence containing normal, heavy load, overload and reverse power flow section; wherein, is the synthetic new energy output; is the wind power and photovoltaic active power curve, i.e., historical output; is the random disturbance amplification coefficient; is the joint probability density; is the quantile function of normal distribution and u t ~U(0,1); is the rated capacity of the measured transformer; is the adjustable amplitude modulation factor; is the typical load cycle within 24 hours; is used to divide each test section.

11. The new energy induced grid transformer overload automatic test system according to claim 9, characterized in that, The programmable inverter-energy storage device is connected in parallel to the low-voltage side of the transformer under test, and the transformer under test is cyclically loaded according to the test sequence, including: The AC output end of the programmable inverter-energy storage device is connected to the low-voltage side bus of the transformer under test through an isolation transformer to complete the electrical connection and communication interface configuration; The test sequence is sent to the inverter device control module in the form of time-target power pairs, and the number of cycles, overload ratio and single section duration are set; The inverter device is based on the target power value at time t. Adjust the actual output power in real time To satisfy the error limit relationship: ； a prediction time window before the power level switching point of the test sequence arrives Within the control module, the power ramp is calculated in advance: And according to the slope, linear smoothing adjustment is performed to realize flexible transition between power sections; Synchronous acquisition of measured power on the primary side of the transformer during the loading process When the following conditions are met or any of the following faults is detected: current exceeds the set limit, voltage is out of limit, or DC bias abnormality determined internally by the inverter, the control system automatically suspends the loading process and records the abnormal section data; wherein, is the allowable loading error; is the total loading test duration; is the power deviation alarm threshold.

12. The new energy induced grid transformer overload automatic test system according to claim 11, characterized in that, The power level switching points include: normal-heavy load, heavy load-overload, overload-reverse sending.

13. The new energy induced grid transformer overload automatic test system according to claim 9, characterized in that, The primary and secondary voltage and current are collected by the synchronous phasor measurement unit, and the top oil temperature, winding hot spot temperature and gas content in the oil are synchronously collected to obtain transformer measurement values, including: During the operation of the transformer, the corresponding electrical signals are obtained from the voltage and current sampling circuits on the primary and secondary sides, and the data is aligned based on a unified time reference; Representative positions are selected at the top of the transformer oil tank, the middle and lower parts of the winding, and the change data of the top oil temperature and the winding hot spot temperature are continuously obtained; The transformer operating oil sample is analyzed online to extract typical gas content including acetylene, hydrogen and methane, which is used to reflect the thermal aging and discharge trend; The electrical signals, change data and gas content are arranged on a unified time axis to obtain the transformer measurement values.

14. The new energy induced grid transformer overload automatic test system according to claim 9, characterized in that, According to the hot spot temperature and the life loss factor, the test sequence parameters are corrected to ensure that the target temperature rise error is within the allowable range, including: At the end of each loading cycle, the winding hot spot temperature and the cumulative life loss factor are read; The deviation of the hot spot temperature and the target temperature rise is calculated, if the absolute value of the deviation is greater than the allowable error or the cumulative life loss factor exceeds the threshold, then the next step is entered, otherwise the current test sequence is maintained; The target power and duration of the next cycle are corrected; The corrected target power and duration are inserted into the corresponding positions of the test sequence, and the next cycle is executed; When the hot spot temperature deviations of two consecutive cycles are both within the allowable error and the cumulative life loss factor does not grow to the threshold, it is determined that the test sequence has converged, and the correction process is ended.

15. The new energy induced grid transformer overload automatic test system according to claim 14, characterized in that, The formula for correcting the target power and duration of the next cycle is: wherein, is the winding hot-spot temperature; is the target hot-spot temperature; is the current cumulative lifetime loss factor; is the lifetime loss threshold; and is the target power and duration before correction; and is the target power and duration after correction.

16. The new energy induced grid transformer overload automatic test system according to claim 9, characterized in that, After completing the test, automatically generate overload bearing capacity and life consumption report and upload to the management platform, including: After the end of the whole test process, classify and summarize the load data, hot spot temperature record, life consumption factor and corresponding time label of all periods; Based on the summary data, draw the temperature rise trend curve, life consumption curve and hot spot temperature distribution graph of the transformer under different load levels, and mark the abnormal points or mutation sections; According to the preset template, automatically generate the standardized test report file containing test number, test scene description, key parameter change, abnormal trigger condition and evaluation conclusion; Name the generated test report file with a unique identifier and upload it to the designated power equipment health management platform or evaluation system through a secure channel for subsequent scheduling, operation and maintenance or risk warning call; After successful uploading, record the report generation time, uploading time and platform return status, and archive the local copy to complete the whole process closed loop.

Images