A method and apparatus for suppressing losses in a transformer

Abstract

This application provides a method and apparatus for suppressing transformer losses. It extracts the electromagnetic harmonic characteristic layer, vibration state characteristic layer, and heat loss characteristic layer of a large transformer from historical multi-source operating data. Based on real-time acquisition of the current multi-source operating data and electromagnetic harmonic characteristic layer, it determines the harmonic interference mode of the large transformer. It generates a harmonic compensation current for the harmonic interference mode using the vibration state characteristic layer and heat loss characteristic layer. Based on the compensated multi-source operating data of the large transformer, it generates the measured harmonic spectrum and measured vibration spectrum under the current operating conditions of the large transformer to correct the control parameters of the frequency conversion modulation circuit in the large transformer, thereby suppressing transformer losses. Using the scheme of this application, transformer losses can be synergistically suppressed based on the transformer's vibration state and heat loss characteristics.

Description

Technical Field

[0001] This application relates to the field of loss suppression technology, and more specifically, to a method and apparatus for suppressing losses in a transformer. Background Technology

[0002] Loss suppression refers to the use of a series of technical means and management measures in fields such as electronic engineering, power systems, and signal transmission to minimize the unnecessary consumption of energy, signals, or resources during transmission, conversion, and use. Systematic loss suppression can not only significantly improve energy conversion efficiency and signal integrity, but is also the key to ensuring stable equipment operation, extending service life, and achieving green and low-carbon goals.

[0003] Large transformers are core equipment in power transmission and distribution systems. Under the influence of grid harmonics and load fluctuations, they generate significant harmonic interference, leading to increased additional electrical losses, aggravated abnormal vibrations, and excessive temperature rise in the core and windings. This severely reduces equipment operating efficiency and service life. Most existing transformer loss suppression methods only provide simple compensation for electrical harmonics without considering the correlation and coordinated control of harmonic interference, equipment vibration, and heat loss. Furthermore, they lack real-time data feedback and parameter closed-loop correction mechanisms after compensation, resulting in low compensation accuracy, poor control targeting, and limited loss suppression effects. These methods cannot meet the low-loss and high-stability operation requirements of large transformers. Therefore, how to coordinate the suppression of transformer losses based on the transformer's vibration state and heat loss characteristics has become a problem facing the industry. Summary of the Invention

[0004] This application provides a method and apparatus for suppressing transformer losses, which can synergistically suppress transformer losses based on the transformer's vibration state and thermal loss characteristics.

[0005] In a first aspect, this application provides a method for suppressing transformer losses, comprising the following steps: Acquire historical multi-source operating data of large transformers, including historical current, voltage, vibration, and temperature data of large transformers; Feature analysis was performed on the historical multi-source operating data to obtain the electromagnetic harmonic characteristic layer, vibration state characteristic layer, and heat loss characteristic layer of the large transformer. Real-time acquisition of current multi-source operating data of large transformers; based on the electromagnetic harmonic feature layer and the current and voltage data in the current multi-source operating data, harmonic interference identification of the operating status of large transformers is performed to obtain the harmonic interference modes of large transformers. The harmonic compensation current of the harmonic interference mode is generated through the vibration state characteristic layer and the heat loss characteristic layer, and the harmonic compensation current is injected into the neutral point or winding of the large transformer to offset the additional losses caused by harmonics. After the harmonic compensation current is injected, the multi-source operating data of the large transformer after compensation is collected. Based on the multi-source operating data after compensation, the measured harmonic spectrum and measured vibration spectrum of the large transformer under the current operating condition are generated. According to the spectral deviation between the measured harmonic spectrum and the measured vibration spectrum and the corresponding reference harmonic spectrum and reference vibration spectrum under the current operating condition of the large transformer, the control parameters of the frequency conversion modulation circuit in the large transformer are corrected to suppress the loss of the large transformer.

[0006] In some embodiments, performing feature analysis on the historical multi-source operating data to obtain the electromagnetic harmonic characteristic layer, vibration state characteristic layer, and heat loss characteristic layer of the large transformer specifically includes: The historical multi-source operation data is preprocessed to obtain preprocessed historical multi-source operation data; Harmonic characteristic analysis is performed on the voltage and current data in the preprocessed historical multi-source operating data to obtain the electromagnetic harmonic characteristic layer of the large transformer. The vibration data in the preprocessed historical multi-source operation data is subjected to state feature analysis to obtain the vibration state feature layer of the large transformer. The temperature data in the preprocessed historical multi-source operating data is analyzed for heat loss characteristics to obtain the heat loss characteristic layer of the large transformer.

[0007] In some embodiments, dedicated current and voltage acquisition units, piezoelectric vibration sensors, and distributed temperature sensing modules are deployed at key monitoring points on the high-voltage side, low-voltage side, core, enclosure, and windings of the large transformer to collect real-time multi-source operating data of the large transformer.

[0008] In some embodiments, harmonic interference identification of the operating state of a large transformer is performed based on the electromagnetic harmonic feature layer and the current and voltage data in the current multi-source operating data to obtain the harmonic interference modes of the large transformer, specifically including: Real-time electrical features are extracted from the current and voltage data in the current multi-source operating data; A difference analysis was performed on the real-time electrical characteristics and the electromagnetic harmonic characteristic layer to identify the harmonic interference components of the large transformer's operating state. The harmonic interference modes of the large transformer are determined by the harmonic interference components.

[0009] In some embodiments, generating the harmonic compensation current for the harmonic interference mode through the vibration state characteristic layer and the heat loss characteristic layer specifically includes: The harmonic interference mode is associated and matched with the vibration state feature layer and the heat loss feature layer to obtain the association matching result of the vibration and heat loss of the harmonic interference mode. The target compensation harmonic component for suppressing abnormal vibration and offsetting additional heat loss from harmonics is determined from the correlation matching results. The harmonic compensation current of the harmonic interference mode is determined based on the target compensation harmonic component.

[0010] In some embodiments, generating the measured harmonic spectrum and measured vibration spectrum of a large transformer under its current operating conditions based on the compensated multi-source operating data specifically includes: Harmonic analysis was performed on the compensated multi-source operating data to obtain the measured harmonic spectrum of the large transformer under the current operating conditions. The vibration data in the compensated multi-source operating data is subjected to spectral analysis to obtain the measured vibration spectrum of the large transformer under the current operating conditions.

[0011] In some embodiments, the control parameters of the frequency conversion modulation circuit in the large transformer are corrected based on the spectral deviation between the measured harmonic spectrum and the measured vibration spectrum and the reference harmonic spectrum and reference vibration spectrum corresponding to the current operating condition of the large transformer, so as to achieve loss suppression in the large transformer. Specifically, this includes: The measured harmonic spectrum and the measured vibration spectrum are matched point-by-point across the entire frequency band with the reference harmonic spectrum and the reference vibration spectrum corresponding to the current operating condition of the large transformer. The harmonic spectrum deviation between the measured harmonic spectrum and the corresponding reference harmonic spectrum, and the vibration spectrum deviation between the measured vibration spectrum and the reference vibration spectrum are calculated. Using the harmonic spectrum deviation and the vibration spectrum deviation as inputs, the control parameters of the frequency conversion modulation circuit in the large transformer are adaptively corrected to gradually reduce the difference between the measured spectrum and the reference, so as to suppress the loss of the large transformer.

[0012] Secondly, this application provides a transformer loss suppression device, comprising: The acquisition module is used to acquire historical multi-source operating data of large transformers, including historical current, voltage, vibration and temperature data of large transformers. The processing module is used to perform feature analysis on the historical multi-source operating data to obtain the electromagnetic harmonic feature layer, vibration state feature layer and heat loss feature layer of the large transformer. The processing module is also used to collect the current multi-source operating data of the large transformer in real time, and to identify the harmonic interference of the large transformer based on the electromagnetic harmonic feature layer and the current and voltage data in the current multi-source operating data, so as to obtain the harmonic interference mode of the large transformer. The processing module is also used to generate a harmonic compensation current for the harmonic interference mode through the vibration state characteristic layer and the heat loss characteristic layer, and inject the harmonic compensation current into the neutral point or winding of the large transformer to offset the additional losses generated by the harmonics. The execution module is used to collect the compensated multi-source operating data of the large transformer after the harmonic compensation current is injected, generate the measured harmonic spectrum and measured vibration spectrum of the large transformer under the current operating condition based on the compensated multi-source operating data, and correct the control parameters of the frequency conversion modulation circuit in the large transformer according to the spectral deviation between the measured harmonic spectrum and the measured vibration spectrum and the corresponding reference harmonic spectrum and reference vibration spectrum under the current operating condition of the large transformer, so as to achieve loss suppression of the large transformer.

[0013] Thirdly, this application provides a computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described transformer loss suppression method.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for suppressing transformer losses.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: The transformer loss suppression method and apparatus provided in this application comprehensively collects operating information of the transformer under different operating conditions by acquiring historical multi-source operating data of current, voltage, vibration, and temperature of a large transformer. Feature analysis of the historical multi-source operating data yields an electromagnetic harmonic characteristic layer, a vibration state characteristic layer, and a heat loss characteristic layer. The vibration state characteristic layer captures core characteristics such as vibration amplitude and frequency under different operating conditions, while the heat loss characteristic layer presents the distribution law and trend of heat loss under different operating conditions. These two layers, in conjunction with the electromagnetic harmonic characteristic layer, break through the traditional limitation of focusing only on harmonic suppression while ignoring vibration and heat loss, clarifying the correspondence between harmonic interference and transformer vibration and heat loss. Based on the electromagnetic harmonic characteristic layer and current current and voltage data, harmonic interference modes are identified. Combined with historical vibration state and heat loss characteristics, the correlation between harmonic interference and vibration and heat loss can be located, ensuring that the identified harmonic interference modes comprehensively reflect the relationship between harmonics, vibration, and heat loss. This approach avoids the drawbacks of traditional harmonic identification, which focuses solely on current and voltage parameters while neglecting the correlation between vibration and heat loss. By generating and injecting a suitable harmonic compensation current through vibration state feature layers and heat loss feature layers, the compensation current can not only effectively offset the additional heat loss caused by harmonic interference but also specifically suppress abnormal vibrations induced by harmonics. This achieves synergistic suppression of harmonic interference, abnormal vibrations, and additional heat loss, solving the problem of traditional methods that only compensate for harmonics and cannot simultaneously control vibration and heat loss. After compensation, multi-source operating data is collected to generate a measured spectrum. By combining this spectrum with a reference spectrum to calculate the deviation and dynamically correct the control parameters of the frequency conversion modulation circuit, the compensation accuracy can be adjusted according to the real-time vibration state and heat loss changes of the transformer. This ensures that the compensation current can continuously adapt to the requirements of vibration suppression and heat loss control, ultimately achieving comprehensive and stable suppression of transformer losses based on vibration state and heat loss characteristics. At the same time, it ensures the safety and stability of transformer operation and fully adapts to the synergistic loss suppression requirements of large transformers under multiple operating conditions. Using the above scheme, transformer losses can be synergistically suppressed based on the transformer's vibration state and heat loss characteristics. Attached Figure Description

[0016] Figure 1 This is an exemplary flowchart of a transformer loss suppression method according to some embodiments of this application; Figure 2 This is an exemplary flowchart illustrating the determination of harmonic interference modes according to some embodiments of this application; Figure 3 These are structural diagrams of transformers according to some embodiments of this application; Figure 4 This is a schematic diagram of the structure of a transformer loss suppression device according to some embodiments of this application; Figure 5This is a schematic diagram of the structure of a computer device for implementing a transformer loss suppression method according to some embodiments of this application. Detailed Implementation

[0017] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0018] refer to Figure 1 The figure is an exemplary flowchart of a transformer loss suppression method according to some embodiments of this application. The transformer loss suppression method mainly includes the following steps: In step 101, historical multi-source operating data of the large transformer is obtained, including historical current, voltage, vibration and temperature data of the large transformer.

[0019] It should be noted that the historical multi-source operating data in this application refers to the multi-dimensional monitoring data collected during the past operation of large transformers. It represents the long-term actual operating condition record of the transformer, comprehensively reflecting the electromagnetic characteristics, mechanical vibration characteristics and heat loss variation patterns of the transformer under different operating conditions. Specifically, it includes historical current data, voltage data, vibration data and temperature data of large transformers, which can be used to conduct multi-dimensional characteristic analysis of transformer operating status, thereby supporting the implementation of transformer harmonic interference identification and loss suppression.

[0020] In step 102, feature analysis is performed on the historical multi-source operating data to obtain the electromagnetic harmonic characteristic layer, vibration state characteristic layer, and heat loss characteristic layer of the large transformer.

[0021] In some embodiments, the following steps can be used to perform feature analysis on the historical multi-source operating data to obtain the electromagnetic harmonic characteristic layer, vibration state characteristic layer, and heat loss characteristic layer of the large transformer: The historical multi-source operation data is preprocessed to obtain preprocessed historical multi-source operation data; Harmonic characteristic analysis is performed on the voltage and current data in the preprocessed historical multi-source operating data to obtain the electromagnetic harmonic characteristic layer of the large transformer. The vibration data in the preprocessed historical multi-source operation data is subjected to state feature analysis to obtain the vibration state feature layer of the large transformer. The temperature data in the preprocessed historical multi-source operating data is analyzed for heat loss characteristics to obtain the heat loss characteristic layer of the large transformer.

[0022] In specific implementation, the historical multi-source operating data is preprocessed to obtain the preprocessed historical multi-source operating data, which can be achieved in the following way: First, data cleaning is performed, and the collected historical current, voltage, vibration, and temperature data are checked one by one to remove abnormal sampling values ​​caused by sensor failure or interference in the acquisition line, such as current and voltage values ​​exceeding the normal operating range of the transformer, sudden vibration data, and abnormally fluctuating temperature values. Missing data segments are supplemented by interpolation of adjacent valid sampling points to ensure data integrity; Second, noise suppression is performed, using a combination of hardware filtering and digital filtering. On the hardware side, a low-pass filter is used. The first step involves filtering high-frequency electromagnetic interference data from the field using a filter. Digitally, a moving average filtering method is used to smooth the acquired raw data, eliminating the impact of environmental noise on the data. The third step is to perform time synchronization and normalization processing. Using the clock of the transformer high-voltage side current acquisition channel as a reference, the data timing of each acquisition channel, including voltage, vibration, and temperature, is adjusted to ensure that the time nodes of data in different dimensions are consistent. At the same time, all data are normalized to the same numerical range to eliminate the influence of differences in the dimensions of different physical quantities. Finally, regular, clean, and time-uniform preprocessed historical multi-source operating data is obtained. Other methods can be used in other embodiments, which are not limited here.

[0023] In addition, in specific implementation, harmonic characteristic analysis is performed on the voltage and current data in the preprocessed historical multi-source operating data to obtain the electromagnetic harmonic characteristic layer of the large transformer. This can be achieved in the following way: harmonic characteristic analysis is performed on the preprocessed current and voltage data. The fast Fourier transform method is used to convert the current and voltage data in the time domain into frequency domain data, decompose the fundamental wave and each harmonic component, and statistically analyze the amplitude and phase shift of each harmonic under different operating conditions (including light load, rated load, overload, and different ambient temperatures). The harmonic distortion rate under each operating condition is calculated, and the law of harmonic component change with load and the mutual influence relationship between different harmonics are recorded. These parameters and laws reflecting the electromagnetic harmonic characteristics of the transformer are sorted and summarized to form an electromagnetic harmonic characteristic layer that can comprehensively characterize the electromagnetic harmonic state of the transformer. Other methods can also be used in other embodiments, which are not limited here.

[0024] In addition, in specific implementation, the vibration state characteristic layer of the large transformer can be obtained by performing state characteristic analysis on the vibration data in the preprocessed historical multi-source operating data in the following way: Using a conventional method combining time-domain statistical analysis and frequency-domain analysis, the peak value, effective value, average value, variance, and other key parameters of the vibration data are statistically analyzed in the time domain to determine the stability of the vibration data; in the frequency domain, the frequency components of the vibration data are decomposed using fast Fourier transform to identify the inherent vibration frequency of the transformer during normal operation and the abnormal vibration frequency caused by harmonic interference, analyze the variation law of vibration amplitude and characteristic frequency under different operating conditions, and clarify the relationship between the vibration data and the loosening of the transformer's internal iron core. The correlation between vibration data and mechanical states such as core loosening and winding deformation is established. Specifically, a multivariate nonlinear fitting model is constructed, with vibration characteristic frequency, amplitude, effective value, and time-domain variance as input variables, and quantitative indicators of core loosening and winding deformation as output variables. The correlation weights between each vibration characteristic and mechanical state are calibrated using historical data training and Pearson correlation coefficients. The correlation calculation formula between vibration characteristic parameters and the mechanical states of the core and windings is then fitted. These vibration-related characteristic parameters and their variation patterns are organized into a vibration state characteristic layer. Other methods can be used in other embodiments, which are not limited here.

[0025] It should be noted that the different operating conditions of large transformers mainly include different load states such as no-load operation, light-load operation, rated load operation and short-term overload operation. They also cover various operating scenarios such as grid voltage fluctuations, three-phase load imbalance, different ambient temperatures, different continuous operating durations, as well as differences in grid-side harmonic content and equipment start-up and shutdown switching.

[0026] In addition, in specific implementation, the heat loss characteristic analysis of the temperature data in the preprocessed historical multi-source operating data can be performed to obtain the heat loss characteristic layer of the large transformer. This can be achieved in the following way: Based on the heat transfer theory in the field of power equipment heat transfer, the temperature values ​​of key monitoring points such as the transformer core, windings, and tank are statistically analyzed. The temperature rise rate (the temperature rise per unit time) of each point and the temperature difference distribution between different points are analyzed. The temperature change trend under different operating conditions is tracked. Combined with the correlation between transformer loss and temperature (i.e., the greater the loss, the more obvious the temperature rise), the correspondence between temperature change and harmonic additional loss, iron loss, and copper loss is analyzed. That is, the relationship between temperature change and harmonic additional loss, iron loss, and copper loss is analyzed. Based on the correspondence between heat loss and copper loss, and combining the Fourier heat conduction formula and the power equipment loss calculation criteria, a heat loss-temperature correlation model is established with the calculated values ​​of harmonic additional loss, iron loss, and copper loss as independent variables, and the temperature rise rate and temperature difference distribution of key points of transformer core, winding, and tank as dependent variables. Harmonic distortion rate is introduced as a correction factor. The least squares method is used to fit historical data to obtain the correspondence between temperature change and various types of losses. The cumulative situation of heat loss under different operating conditions and the distribution law of heat loss in various parts of the transformer are statistically analyzed. These parameters and laws reflecting the heat generation and heat loss characteristics of the transformer are compiled and summarized to finally form a heat loss characteristic layer. Other methods can be used in other embodiments, which are not limited here.

[0027] It should be noted that the electromagnetic harmonic characteristic layer in this application represents the harmonic distribution and distortion characteristics of the transformer's electrical operation, reflecting the amplitude, phase, and harmonic variation law of each harmonic under different operating conditions. It can be used to identify the harmonic interference of the transformer in real time, determine the harmonic interference mode, and provide an electrical reference for harmonic compensation. The vibration state characteristic layer represents the vibration variation characteristics of the transformer's main structure, reflecting the transformer's mechanical state, steady-state vibration threshold, and abnormal vibration law under harmonic excitation. It can be used to match the vibration response corresponding to the harmonics and provide vibration constraints for the generation of harmonic compensation current. The heat loss characteristic layer represents the correlation characteristics of heat generation and energy loss in various parts of the transformer, reflecting the transformer's temperature rise, temperature difference distribution, and the generation and accumulation law of harmonic additional losses. It can be used to quantify the loss suppression target and generate a reasonable harmonic compensation current in conjunction with the vibration state characteristics.

[0028] In step 103, the current multi-source operating data of the large transformer is collected in real time. Based on the electromagnetic harmonic feature layer and the current and voltage data in the current multi-source operating data, the operating status of the large transformer is identified by harmonic interference to obtain the harmonic interference mode of the large transformer.

[0029] In practice, dedicated current and voltage acquisition units, piezoelectric vibration sensors, and distributed temperature sensing modules are deployed at key monitoring points on the high-voltage side, low-voltage side, core, tank, and windings of large transformers. A multi-channel synchronous sampling mechanism is adopted to collect instantaneous current, real-time voltage, body vibration data, and temperature data of key parts of the transformer in real time and in parallel with a unified clock reference. The collected raw analog data is isolated and amplified, filtered to resist electromagnetic interference, converted from analog to digital, and preprocessed with time alignment to remove environmental noise and electromagnetic interference in the field conditions. Finally, the current multi-source operating data with time synchronization, complete dimensions, and matching accuracy is obtained.

[0030] In some embodiments, reference Figure 2 The figure is an exemplary flowchart of determining harmonic interference modes in some embodiments of this application. In this embodiment, the harmonic interference identification of the operating state of the large transformer is performed based on the electromagnetic harmonic feature layer and the current and voltage data in the current multi-source operating data. The harmonic interference modes of the large transformer can be obtained by the following steps: In step 1031, real-time electrical features are extracted from the current and voltage data in the current multi-source operating data; In step 1032, a difference analysis is performed on the real-time electrical characteristics and the electromagnetic harmonic characteristic layer to identify the harmonic interference components of the large transformer's operating state. In step 1033, the harmonic interference mode of the large transformer is determined by the harmonic interference components.

[0031] In specific implementation, real-time electrical features are extracted from the current and voltage data in the current multi-source operating data, and this is achieved in the following manner: Current and voltage time-series data are extracted from the current multi-source operating data. Time-domain and frequency-domain analyses are performed on the extracted current and voltage time-series data to extract real-time electrical features. During the time-domain analysis, the instantaneous values ​​of current and voltage are statistically analyzed at each time point. The average, maximum, minimum, and fluctuation amplitudes of current and voltage within a certain time interval are calculated. Simultaneously, sudden changes in current and voltage values ​​are recorded, such as when the instantaneous amplitude exceeds the normal operating range. This process is then used to determine the current and voltage... Operational stability; During frequency domain analysis, the Fast Fourier Transform method is used to decompose the continuous current and voltage time series data in the time domain into components of different frequencies. The amplitude and phase shift of each harmonic are extracted one by one, and the proportion of each harmonic in the total current and total voltage is calculated. At the same time, the phase difference between current and voltage is statistically analyzed to determine the degree of harmonic distortion of current and voltage. All parameters obtained from the above time domain and frequency domain analysis are summarized and organized to form real-time electrical characteristics that can comprehensively and realistically reflect the current electrical operating status of large transformers. Other methods can also be used in other embodiments, which are not limited here.

[0032] Furthermore, in specific implementation, the difference analysis between the real-time electrical characteristics and the electromagnetic harmonic characteristic layer to identify the harmonic interference components of the large transformer's operating state can be achieved in the following way: First, retrieve the standard electrical characteristics that perfectly match the current operating condition of the transformer from the electromagnetic harmonic characteristic layer. This standard electrical characteristic is extracted based on preprocessed historical multi-source operating data and covers all electrical characteristic parameter benchmark values ​​during normal transformer operation under this condition. Using the difference calculation method in this field, calculate the absolute difference between each parameter in the real-time electrical characteristics and the corresponding standard electrical characteristic parameter, and simultaneously calculate the relative deviation, i.e., the ratio of the absolute difference to the standard parameter value. Combining the rated operating parameters of the large transformer and conventional standards in the power industry, set a reasonable deviation threshold. For example, the deviation threshold for the current harmonic amplitude is set to ±5% of the corresponding standard value. The voltage harmonic distortion rate deviation threshold is set to ±3%. This threshold setting method is a routine operation for those skilled in the art and does not require additional derivation. Parameters in the real-time electrical characteristics whose deviation exceeds the set threshold are judged as abnormal electrical characteristics. Then, combined with the harmonic variation law under different operating conditions recorded in the electromagnetic harmonic characteristic layer and the characteristics of the inherent background harmonics on the power grid side, the harmonic sources corresponding to the abnormal electrical characteristics are further distinguished. Among them, the characteristic fluctuation range of the inherent background harmonics on the power grid side is small and has low correlation with the transformer's own operating conditions, while the characteristic fluctuation of the abnormal harmonics generated by the transformer body is large and shows a clear pattern with the change of transformer load. After removing the abnormal electrical characteristics corresponding to the inherent background harmonics on the power grid side, the remaining abnormal electrical characteristics are the harmonic interference components under the operating state of the large transformer. Other methods can also be used in other embodiments, which are not limited here.

[0033] In addition, in specific implementation, the determination of the harmonic interference mode of the large transformer through the harmonic interference components can be achieved in the following way: by classifying, statistically analyzing and quantifying the harmonic interference components, the frequency, amplitude, phase shift, duration of operation, and trend of change over time of each harmonic interference component are statistically analyzed. Then, the interference type is classified according to the specific situation of the harmonic interference components. For example, a single frequency with a stable amplitude is a single harmonic interference, while multiple frequencies superimposed and amplitude fluctuating are composite harmonic interference. At the same time, based on the amplitude, frequency, and degree of influence on the transformer operation of the harmonic interference components, the specific characteristic parameters of each type of harmonic interference are identified. These interference types, characteristic parameters, and change patterns are summarized and integrated to finally determine the harmonic interference mode that characterizes the current harmonic interference situation of the large transformer. Other methods can also be used in other embodiments, which are not limited here.

[0034] It should be noted that the real-time electrical characteristics in this application represent the actual electrical operating state of the transformer at present, reflecting the amplitude fluctuations, harmonic distribution, and degree of harmonic distortion of the current and voltage. They can be used to compare with the electromagnetic harmonic characteristic layer to identify anomalies. The harmonic interference components represent the abnormal harmonic components generated by the transformer itself, reflecting the degree of harmonic interference on normal operation and the abnormal amplitude and frequency deviation. They can be used to identify the harmonic targets that need to be suppressed. The harmonic interference modes represent the type and manifestation of the current harmonic interference, reflecting the composition, variation law, and degree of impact on equipment loss and vibration. They can be used to directly guide the generation of harmonic compensation current and the parameter correction of the frequency conversion modulation circuit.

[0035] In step 104, the harmonic compensation current of the harmonic interference mode is generated through the vibration state characteristic layer and the heat loss characteristic layer, and the harmonic compensation current is injected into the neutral point or winding of the large transformer to offset the additional losses caused by harmonics.

[0036] In some embodiments, generating the harmonic compensation current of the harmonic interference mode through the vibration state characteristic layer and the heat loss characteristic layer can be achieved by the following steps: The harmonic interference mode is associated and matched with the vibration state feature layer and the heat loss feature layer to obtain the association matching result of the vibration and heat loss of the harmonic interference mode. The target compensation harmonic component for suppressing abnormal vibration and offsetting additional heat loss from harmonics is determined from the correlation matching results. The harmonic compensation current of the harmonic interference mode is determined based on the target compensation harmonic component.

[0037] In specific implementation, the harmonic interference mode is associated and matched with the vibration state feature layer and the heat loss feature layer to obtain the association matching result of the vibration and heat loss of the harmonic interference mode. This can be achieved in the following way: the identified harmonic interference mode is matched one by one with the vibration state feature layer and the heat loss feature layer. Based on the frequency, amplitude, trend and interference type of the current harmonic interference mode, the operating condition consistent with the current transformer load and ambient temperature is selected in the vibration state feature layer, and the vibration characteristic frequency corresponding to the harmonic interference mode is found. The vibration rate, vibration amplitude range, vibration stability threshold, and abnormal vibration patterns caused by harmonic excitation are analyzed. Simultaneously, in the heat loss characteristic layer, the temperature rise rate, temperature distribution difference, stable operation loss threshold, and additional heat loss variation patterns of key components such as windings and cores corresponding to this type of harmonic interference are matched. The harmonic interference characteristics, vibration response characteristics, and heat loss variation characteristics are synchronously correlated to obtain a correlation matching result that can intuitively reflect the degree of vibration anomaly and the magnitude of heat loss increment under the harmonic interference mode. Other methods can be used in other embodiments, which are not limited here.

[0038] In addition, in specific implementation, the target compensation harmonic component for suppressing abnormal vibration and offsetting additional heat loss caused by harmonics, determined by the correlation matching result, can be implemented in the following way: First, extract the frequency and amplitude parameters corresponding to the current harmonic interference from the correlation matching result, and simultaneously obtain the vibration exceeding the limit value, characteristic frequency deviation, temperature rise exceeding the standard value of the winding and core, and the increase of additional heat loss caused by harmonics caused by the harmonic interference. Then, using the normal vibration threshold of the corresponding working condition in the vibration state characteristic layer and the standard loss and temperature rise range of the corresponding working condition in the heat loss characteristic layer as constraint targets, and based on the corresponding correlation between harmonic interference and abnormal vibration and additional heat loss, determine the basic compensation parameters that have the same frequency and opposite phase as the harmonic interference component. Combined with the required vibration suppression amplitude and heat loss offsetting degree, gradually adjust and determine the amplitude of the compensation component so that while eliminating harmonic interference, the component can restore both transformer vibration and temperature rise to the normal operating range. Finally, the target compensation harmonic component that can simultaneously suppress abnormal vibration and offset additional heat loss caused by harmonics is obtained. Other methods can also be used in other embodiments, which are not limited here.

[0039] In addition, in specific implementation, the harmonic compensation current for the harmonic interference mode determined according to the target harmonic compensation component can be achieved in the following way: First, the core parameters such as the frequency, amplitude, and phase of the target harmonic compensation component are determined. Then, combined with the actual operating conditions of the large transformer, key electrical parameters such as the transformer winding impedance, neutral point connection form, and current load size are retrieved. Using the electrical parameter conversion and output configuration method in the field of active harmonic compensation in power systems, the amplitude parameter of the target harmonic compensation component is adapted and adjusted in combination with the winding impedance to ensure that the current amplitude can meet the requirements for eliminating harmonic interference, suppressing vibration, and reducing losses. At the same time, the current phase is kept completely consistent with the phase of the target harmonic compensation component, and the frequency is accurately matched with the frequency of the target harmonic compensation component. Then, according to the current operating sequence of the transformer, the current output timing is adjusted to avoid interference with the normal operating current of the transformer. Finally, a harmonic compensation current that can be directly injected into the transformer neutral point or winding, can accurately eliminate harmonic interference, and can also suppress vibration and cancel heat loss is determined. Other methods can also be used in other embodiments, which are not limited here.

[0040] It should be noted that the correlation matching results in this application represent the corresponding correlation between harmonic interference and abnormal equipment vibration and harmonic-induced heat loss, reflecting the degree of vibration exceeding limits and the magnitude of temperature rise exceeding limits caused by the current harmonic interference. This can be used to clarify the control targets for vibration suppression and loss cancellation, providing data basis for subsequent determination of compensation parameters. The target compensation harmonic component represents the harmonic compensation standard required to eliminate harmonic interference, suppress abnormal vibration, and reduce additional heat loss, reflecting the frequency, phase, and amplitude requirements for compensation. This can be used as the basis for determining the harmonic compensation current. The harmonic compensation current represents the compensation electrical quantity that can be directly applied to the transformer, reflecting the amplitude, phase, and frequency characteristics of the actual output compensation current. This can be injected into the transformer neutral point or winding to achieve real-time cancellation of harmonic interference, thereby reducing abnormal vibration and harmonic-induced heat loss.

[0041] In practice, the harmonic compensation current is injected into the neutral point or winding of a large transformer to offset the additional losses caused by harmonics. Specifically, based on the determined amplitude, phase, and frequency parameters of the harmonic compensation current, the frequency conversion modulation circuit and the active power compensation module are controlled to inject the harmonic compensation current into the neutral point or corresponding target winding of the large transformer in a directional and synchronous manner through an independent compensation output branch. This ensures that the injected compensation current and the harmonic interference current inside the transformer form a canceling relationship with equal amplitude and opposite phase. This implementation method can directly offset harmonic components from the source of electrical excitation, reduce the additional copper losses, iron losses, and eddy current losses caused by harmonics in the windings and core, and reduce the abnormal heating and energy loss caused by harmonics, thereby achieving active suppression of additional harmonic losses in the transformer.

[0042] In step 105, the compensated multi-source operating data of the large transformer after the harmonic compensation current is injected is collected. Based on the compensated multi-source operating data, the measured harmonic spectrum and measured vibration spectrum of the large transformer under the current operating condition are generated. According to the spectral deviation between the measured harmonic spectrum and the measured vibration spectrum and the corresponding reference harmonic spectrum and reference vibration spectrum under the current operating condition of the large transformer, the control parameters of the frequency conversion modulation circuit in the large transformer are corrected to suppress the loss of the large transformer.

[0043] In practice, after the harmonic compensation current is continuously injected and reaches a stable operating state, the existing multi-channel synchronous acquisition device of the transformer is used to synchronously acquire multi-dimensional operating data of the transformer after the compensation current is injected, using the same sampling frequency, clock reference and acquisition point as the previous real-time acquisition. The acquired raw data is isolated and noise-reduced, anti-electromagnetic interference processed, analog-to-digital converted and time-aligned, and the electrical interference and environmental noise introduced by the compensation device itself are eliminated, so as to obtain multi-source operating data that can truly and accurately reflect the actual operating state of the transformer after harmonic compensation.

[0044] In some embodiments, generating the measured harmonic spectrum and measured vibration spectrum of a large transformer under its current operating conditions based on the compensated multi-source operating data can be achieved through the following steps: Harmonic analysis was performed on the compensated multi-source operating data to obtain the measured harmonic spectrum of the large transformer under the current operating conditions. The vibration data in the compensated multi-source operating data is subjected to spectral analysis to obtain the measured vibration spectrum of the large transformer under the current operating conditions.

[0045] In specific implementation, harmonic analysis is performed on the compensated multi-source operating data to obtain the measured harmonic spectrum of the large transformer under its current operating conditions. This can be achieved in the following way: The compensated multi-source operating data is preprocessed by checking the compensated current, voltage, and vibration time-series data one by one, eliminating abnormal values ​​caused by acquisition interference and equipment fluctuations, smoothing the data using a moving average method, and aligning the time series of various data according to a unified time reference to ensure data integrity, continuity, stability, and reliability. Subsequently, harmonic analysis is performed on the preprocessed current and voltage time-series data. Relying on frequency domain analysis methods in the field of power monitoring, the electrical values ​​collected in chronological order are decomposed into components of different frequencies, distinguishing the fundamental wave and each harmonic component. The amplitude and distribution of harmonics at different frequencies are recorded one by one. The frequencies and corresponding amplitudes are then matched to form a measured harmonic spectrum that can intuitively reflect the actual harmonic situation after compensation under the current operating conditions of the transformer. At the same time, spectral analysis is carried out on the preprocessed vibration time series data. Using frequency domain analysis methods for mechanical equipment condition monitoring, the continuous vibration time series values ​​are decomposed into vibration components of different frequencies. Invalid frequency components caused by environmental vibration and external interference are filtered out, and effective vibration characteristics that can reflect the operating state of the transformer body are extracted. The vibration amplitudes corresponding to each frequency are statistically analyzed and regularized to obtain a measured vibration spectrum that can truly characterize the current structural vibration state of the transformer. Other methods can also be used in other embodiments, which are not limited here.

[0046] It should be noted that the measured harmonic spectrum in this application represents the actual electrical harmonic distribution of the transformer under the current operating conditions after compensation, reflecting the amplitude, frequency distribution and degree of harmonic distortion of each harmonic; the measured vibration spectrum represents the vibration state of the transformer's main structure after compensation, reflecting the vibration amplitude, abnormal vibration components and vibration stability of each frequency band.

[0047] In some embodiments, the control parameters of the frequency conversion modulation circuit in the large transformer are corrected based on the spectral deviation between the measured harmonic spectrum and the measured vibration spectrum and the reference harmonic spectrum and reference vibration spectrum corresponding to the current operating condition of the large transformer, so as to achieve loss suppression in the large transformer. This can be achieved by the following steps: The measured harmonic spectrum and the measured vibration spectrum are matched point-by-point across the entire frequency band with the reference harmonic spectrum and the reference vibration spectrum corresponding to the current operating condition of the large transformer. The harmonic spectrum deviation between the measured harmonic spectrum and the corresponding reference harmonic spectrum, and the vibration spectrum deviation between the measured vibration spectrum and the reference vibration spectrum are calculated. Using the harmonic spectrum deviation and the vibration spectrum deviation as inputs, the control parameters of the frequency conversion modulation circuit in the large transformer are adaptively corrected to gradually reduce the difference between the measured spectrum and the reference, so as to suppress the loss of the large transformer.

[0048] In specific implementation, the measured harmonic spectrum and the measured vibration spectrum are matched point-by-point across the entire frequency band with the reference harmonic spectrum and reference vibration spectrum corresponding to the current operating condition of the large transformer. The harmonic spectrum deviation between the measured harmonic spectrum and the corresponding reference harmonic spectrum, and the vibration spectrum deviation between the measured vibration spectrum and the reference vibration spectrum can be calculated in the following way: The reference harmonic spectrum and the reference vibration spectrum that perfectly match the current actual operating condition of the large transformer are retrieved. This reference spectrum is generated based on historical multi-source data under normal transformer operating conditions after preprocessing and analysis, and can characterize the standard state of harmonics and vibration under abnormal, low-loss operating conditions of the transformer. Subsequently, the measured harmonic spectrum and the reference harmonic spectrum, and the measured vibration spectrum and the reference vibration spectrum, are matched point-by-point across the entire frequency range. From the lowest to the highest frequency, the differences between the measured data and the reference data are compared at each frequency point. For the measured harmonic spectrum and the reference harmonic spectrum, the harmonic amplitude and phase parameters corresponding to each frequency point are compared in detail. Using the numerical difference calculation method in the field of power equipment monitoring, the absolute difference and relative deviation of the harmonic amplitude at each frequency point are calculated one by one. At the same time, the frequency ranges with deviations exceeding the standard and the maximum deviation are statistically analyzed. The deviation data of all frequency points are summarized to form a complete harmonic spectrum deviation. Similarly, for the measured vibration spectrum and the reference vibration spectrum, the differences in vibration amplitude are compared at each frequency point. The absolute difference and relative deviation of the vibration amplitude at each frequency point are calculated. The frequency ranges with vibration amplitude deviations exceeding the normal range are screened out, and the frequency and degree of deviation corresponding to abnormal vibration are identified. The vibration spectrum deviation is obtained after summarizing. Other methods can be used in other embodiments, which are not limited here.

[0049] It should be noted that the harmonic spectrum deviation in this application represents the deviation between the actual electrical harmonics after transformer harmonic compensation and the standard harmonic state, reflecting the elimination effect of harmonic compensation and the amplitude and distribution characteristics of residual harmonics. It can be used to specifically correct the output frequency, amplitude, and phase parameters of the frequency conversion modulation circuit. The vibration spectrum deviation represents the deviation between the actual structural vibration after transformer compensation and the standard vibration state, reflecting the vibration suppression effect and the amplitude and frequency band characteristics of abnormal vibration. It can be used to adjust the output gain and operating timing of the frequency conversion modulation circuit. Together, they provide a basis for the closed-loop correction of the frequency conversion modulation circuit, thereby achieving transformer loss suppression.

[0050] In addition, in specific implementation, using the harmonic spectrum deviation and the vibration spectrum deviation as inputs, the control parameters of the frequency conversion modulation circuit in the large transformer are adaptively corrected to gradually reduce the difference between the measured spectrum and the reference, thereby achieving loss suppression in the large transformer. This can be achieved in the following way: using the harmonic spectrum deviation and the vibration spectrum deviation as the core adjustment inputs, the closed-loop parameter adjustment method of the frequency conversion modulation circuit in the power system is adopted. Combining the working principle and output characteristics of the frequency conversion modulation circuit, the key control parameters of the circuit are finely adjusted in real time and continuously. According to the harmonic spectrum deviation, the output amplitude, phase, and output frequency of the frequency conversion modulation circuit are adjusted. For frequency ranges with large harmonic deviations, the compensation amplitude of the corresponding frequency band is appropriately adjusted, and the output phase is corrected to ensure that the compensation harmonics output by the circuit can more accurately cancel the residual harmonic interference inside the transformer, gradually reducing the harmonic spectrum deviation. At the same time, according to the vibration spectrum deviation, the output gain, triggering sequence, and working response speed of the frequency conversion modulation circuit are adjusted. For frequency ranges where the vibration deviation exceeds the standard, the circuit output rhythm is optimized to reduce the abnormal vibration caused by harmonic excitation and gradually control the vibration amplitude within the normal range corresponding to the reference spectrum. The entire correction process adopts a cyclical approach of "comparison, correction, re-comparison, and re-correction" to gradually reduce the difference between the measured harmonic spectrum, the measured vibration spectrum, and the corresponding reference spectrum. This ensures that the frequency conversion modulation circuit always maintains the optimal compensation working state, continuously weakens the electrical additional losses and vibration mechanical losses caused by harmonics, reduces the heating and wear of the iron core and windings, and ultimately achieves stable, accurate, and dynamic loss suppression of large transformers, ensuring that the transformer is in a low-loss and safe operating state for a long time. Other methods can also be used in other embodiments, which are not limited here.

[0051] In some embodiments, reference Figure 3The figure is a structural diagram of a transformer in some embodiments of this application. As shown in the figure, 1 is the transformer housing, which serves to enclose, protect and support the internal components; 2 is the iron core, which constitutes the core of the transformer's magnetic circuit; 3 is the winding, which is a key component for realizing power conversion and current transmission; 4 is the high-voltage terminal, which is used to connect to external high-voltage transmission lines; 5 is the neutral point terminal, which can be used to inject harmonic compensation current to realize transformer loss and vibration suppression.

[0052] In another aspect, in some embodiments, this application provides a transformer loss suppression device, referring to... Figure 4 The figure is a schematic diagram of the structure of a transformer loss suppression device according to some embodiments of this application. The transformer loss suppression device 400 includes: an acquisition module 401, a processing module 402, and an execution module 403, which are described below: The acquisition module 401 in this application is mainly used to acquire historical multi-source operating data of large transformers. The historical multi-source operating data includes historical current, voltage, vibration and temperature data of large transformers. Processing module 402, in this application, is used to perform feature analysis on the historical multi-source operation data to obtain the electromagnetic harmonic feature layer, vibration state feature layer and heat loss feature layer of the large transformer. It should be noted that the processing module 402 in this application is also used to collect the current multi-source operating data of the large transformer in real time, and to identify the harmonic interference of the large transformer based on the electromagnetic harmonic feature layer and the current and voltage data in the current multi-source operating data, so as to obtain the harmonic interference mode of the large transformer. Additionally, it should be noted that the processing module 402 in this application is also used to generate the harmonic compensation current of the harmonic interference mode through the vibration state characteristic layer and the heat loss characteristic layer, and inject the harmonic compensation current into the neutral point or winding of the large transformer to offset the additional losses generated by the harmonics. The execution module 403 in this application is mainly used to collect the compensated multi-source operating data of the large transformer after the harmonic compensation current is injected, generate the measured harmonic spectrum and measured vibration spectrum of the large transformer under the current operating condition based on the compensated multi-source operating data, and correct the control parameters of the frequency conversion modulation circuit in the large transformer according to the spectral deviation between the measured harmonic spectrum and the measured vibration spectrum and the corresponding reference harmonic spectrum and reference vibration spectrum under the current operating condition of the large transformer, so as to achieve loss suppression of the large transformer.

[0053] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described transformer loss suppression method.

[0054] In some embodiments, reference Figure 5 The figure is a schematic diagram of the structure of a computer device for implementing a transformer loss suppression method according to some embodiments of this application. The transformer loss suppression method in the above embodiments can be implemented by... Figure 5 The computer device shown is used to implement this, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0055] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0056] The communication bus 502 can be used to transmit information between the aforementioned components.

[0057] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CD-ROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital universal optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0058] The memory 503 stores program code for executing the scheme of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. The method used in the above embodiments can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0059] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0060] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single-CPU) processor or a multi-core (multi-CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0061] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0062] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described method for suppressing transformer losses.

[0063] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0064] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for suppressing transformer losses, characterized in that, Includes the following steps: Acquire historical multi-source operating data of large transformers, including historical current, voltage, vibration, and temperature data of large transformers; Feature analysis was performed on the historical multi-source operating data to obtain the electromagnetic harmonic characteristic layer, vibration state characteristic layer, and heat loss characteristic layer of the large transformer. Real-time acquisition of current multi-source operating data of large transformers; based on the electromagnetic harmonic feature layer and the current and voltage data in the current multi-source operating data, harmonic interference identification of the operating status of large transformers is performed to obtain the harmonic interference modes of large transformers. The harmonic compensation current of the harmonic interference mode is generated through the vibration state characteristic layer and the heat loss characteristic layer, and the harmonic compensation current is injected into the neutral point or winding of the large transformer to offset the additional losses caused by harmonics. After the harmonic compensation current is injected, the multi-source operating data of the large transformer after compensation is collected. Based on the multi-source operating data after compensation, the measured harmonic spectrum and measured vibration spectrum of the large transformer under the current operating condition are generated. According to the spectral deviation between the measured harmonic spectrum and the measured vibration spectrum and the corresponding reference harmonic spectrum and reference vibration spectrum under the current operating condition of the large transformer, the control parameters of the frequency conversion modulation circuit in the large transformer are corrected to suppress the loss of the large transformer.

2. The method as described in claim 1, characterized in that, Feature analysis was performed on the historical multi-source operating data to obtain the electromagnetic harmonic characteristic layer, vibration state characteristic layer, and heat loss characteristic layer of the large transformer, specifically including: The historical multi-source operation data is preprocessed to obtain preprocessed historical multi-source operation data; Harmonic characteristic analysis is performed on the voltage and current data in the preprocessed historical multi-source operating data to obtain the electromagnetic harmonic characteristic layer of the large transformer. The vibration data in the preprocessed historical multi-source operation data is subjected to state feature analysis to obtain the vibration state feature layer of the large transformer. The temperature data in the preprocessed historical multi-source operating data is analyzed for heat loss characteristics to obtain the heat loss characteristic layer of the large transformer.

3. The method as described in claim 1, characterized in that, By deploying dedicated current and voltage acquisition units, piezoelectric vibration sensors, and distributed temperature sensing modules at key monitoring points on the high-voltage side, low-voltage side, core, enclosure, and windings of large transformers, real-time multi-source operating data of the large transformers can be collected.

4. The method as described in claim 1, characterized in that, Based on the electromagnetic harmonic characteristic layer and the current and voltage data in the current multi-source operating data, harmonic interference identification is performed on the operating status of the large transformer, and the harmonic interference modes of the large transformer are specifically obtained as follows: Real-time electrical features are extracted from the current and voltage data in the current multi-source operating data; A difference analysis was performed on the real-time electrical characteristics and the electromagnetic harmonic characteristic layer to identify the harmonic interference components of the large transformer's operating state. The harmonic interference modes of the large transformer are determined by the harmonic interference components.

5. The method as described in claim 1, characterized in that, The generation of harmonic compensation current for the harmonic interference mode through the vibration state characteristic layer and the heat loss characteristic layer specifically includes: The harmonic interference mode is associated and matched with the vibration state feature layer and the heat loss feature layer to obtain the association matching result of the vibration and heat loss of the harmonic interference mode. The target compensation harmonic component for suppressing abnormal vibrations and offsetting additional heat loss from harmonics is determined based on the correlation matching results. The harmonic compensation current of the harmonic interference mode is determined based on the target compensation harmonic component.

6. The method as described in claim 1, characterized in that, The specific methods for generating the measured harmonic spectrum and measured vibration spectrum of a large transformer under its current operating conditions based on the compensated multi-source operating data include: Harmonic analysis was performed on the compensated multi-source operating data to obtain the measured harmonic spectrum of the large transformer under the current operating conditions. The vibration data in the compensated multi-source operating data is subjected to spectral analysis to obtain the measured vibration spectrum of the large transformer under the current operating conditions.

7. The method as described in claim 1, characterized in that, Based on the spectral deviation between the measured harmonic spectrum and the measured vibration spectrum and the corresponding reference harmonic spectrum and reference vibration spectrum under the current operating conditions of the large transformer, the control parameters of the frequency conversion modulation circuit in the large transformer are corrected to achieve loss suppression in the large transformer. Specifically, this includes: The measured harmonic spectrum and the measured vibration spectrum are matched point-by-point across the entire frequency band with the reference harmonic spectrum and the reference vibration spectrum corresponding to the current operating condition of the large transformer. The harmonic spectrum deviation between the measured harmonic spectrum and the corresponding reference harmonic spectrum, and the vibration spectrum deviation between the measured vibration spectrum and the reference vibration spectrum are calculated. Using the harmonic spectrum deviation and the vibration spectrum deviation as inputs, the control parameters of the frequency conversion modulation circuit in the large transformer are adaptively corrected to gradually reduce the difference between the measured spectrum and the reference, so as to suppress the loss of the large transformer.

8. A transformer loss suppression device, characterized in that, include: The acquisition module is used to acquire historical multi-source operating data of large transformers, including historical current, voltage, vibration and temperature data of large transformers. The processing module is used to perform feature analysis on the historical multi-source operating data to obtain the electromagnetic harmonic feature layer, vibration state feature layer and heat loss feature layer of the large transformer. The processing module is also used to collect the current multi-source operating data of the large transformer in real time, and to identify the harmonic interference of the large transformer based on the electromagnetic harmonic feature layer and the current and voltage data in the current multi-source operating data, so as to obtain the harmonic interference mode of the large transformer. The processing module is also used to generate a harmonic compensation current for the harmonic interference mode through the vibration state characteristic layer and the heat loss characteristic layer, and inject the harmonic compensation current into the neutral point or winding of the large transformer to offset the additional losses generated by the harmonics. The execution module is used to collect the compensated multi-source operating data of the large transformer after the harmonic compensation current is injected, generate the measured harmonic spectrum and measured vibration spectrum of the large transformer under the current operating condition based on the compensated multi-source operating data, and correct the control parameters of the frequency conversion modulation circuit in the large transformer according to the spectral deviation between the measured harmonic spectrum and the measured vibration spectrum and the corresponding reference harmonic spectrum and reference vibration spectrum under the current operating condition of the large transformer, so as to achieve loss suppression of the large transformer.

9. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing code, and the processor being configured to retrieve the code and execute the transformer loss suppression method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the transformer loss suppression method as described in any one of claims 1 to 7.

Images