Online rapid evaluation method and system for mechanical state of transformer

Abstract

The invention relates to the technical field of transformers, and discloses a transformer mechanical state on-line rapid evaluation method and system, and the method comprises the steps: installing N pairs of signal generators and signal receivers on the wall of an oil tank of a to-be-detected transformer, and determining the model discrimination result of a scattering signal transfer matrix, according to the method, whether the mechanical state of the transformer to be tested is changed or not is assessed, whether the mechanical state of the transformer is changed or not can be assessed visually and accurately, dependence on indirect parameters is not needed, the assessment intuition and accuracy are remarkably improved, meanwhile, the assessment capacity of the mode for the mechanical state of the iron core is enhanced, other special methods do not need to be combined, and the cost is reduced. The problems that in the prior art, the evaluation result is not visual, the accuracy is limited, and the iron core mechanical state evaluation capacity is weak are effectively solved.

Description

Technical Field

[0001] The present invention relates to the technical field of transformers, and in particular, to an online rapid evaluation method and system for the mechanical state of transformers. Background Art

[0002] As an indispensable device in the power system, the core function of a transformer is the transmission and distribution of electrical energy. However, during operation, a transformer may encounter various faults, especially winding mechanical faults and core mechanical faults, which pose a serious threat to the stable operation of the power system.

[0003] Currently, the industry has developed various methods for evaluating the mechanical state of transformers, such as the short-circuit impedance method, the vibration frequency response method, and the vibration detection method. However, these methods all rely on indirect parameters such as the electrical and mechanical parameters of the transformer for indirect evaluation from the side, resulting in problems such as non-intuitive evaluation results and limited accuracy, and they are weak in evaluating the mechanical state of the core and often require the combination of other specialized methods. Summary of the Invention

[0004] Based on this, it is necessary to address the above problems and propose an online rapid evaluation method and system for the mechanical state of transformers, which can intuitively and accurately evaluate whether the mechanical state of the transformer has changed, without relying on indirect parameters, significantly improving the intuitiveness and accuracy of the evaluation. At the same time, the evaluation ability of this method for the mechanical state of the core has also been enhanced, and there is no need to combine other specialized methods, effectively solving the problems of non-intuitive evaluation results, limited accuracy, and weak evaluation ability for the mechanical state of the core in the prior art.

[0005] To achieve the above object, in a first aspect, the present invention provides an online rapid evaluation method for the mechanical state of a transformer, the method comprising: When the transformer to be tested is operating, controlling N signal generators to emit N scattered signals with different frequencies, and acquiring N×N scattered signals received by N signal receivers, wherein N pairs of signal generators and signal receivers are installed on the tank wall of the transformer to be tested, and N is greater than or equal to 2; Determining a transmitted scattered signal matrix according to the N scattered signals with different frequencies, and determining a received scattered signal matrix according to the N×N scattered signals; Determining a scattered signal transfer matrix according to the transmitted scattered signal matrix and the received scattered signal matrix; Inputting the scattered signal transfer matrix into a preset machine learning model to obtain a discrimination result; Evaluating whether the mechanical state of the transformer to be tested has changed according to the discrimination result.

[0006] Optionally, the received scattered signal matrix includes a received scattered signal amplitude matrix and a received scattered signal phase matrix, the emitted scattered signal matrix includes an emitted scattered signal amplitude matrix and an emitted scattered signal phase matrix, and the scattered signal transfer matrix includes a scattered signal amplitude transfer matrix and a scattered signal phase transfer matrix. Determining the scattered signal transfer matrix according to the emitted scattered signal matrix and the received scattered signal matrix includes: Determining the scattered signal amplitude transfer matrix according to the received scattered signal amplitude matrix and the emitted scattered signal amplitude matrix, and determining the scattered signal phase transfer matrix according to the received scattered signal phase matrix and the emitted scattered signal phase matrix.

[0007] Optionally, determining the scattered signal amplitude transfer matrix according to the received scattered signal amplitude matrix and the emitted scattered signal amplitude matrix, and determining the scattered signal phase transfer matrix according to the received scattered signal phase matrix and the emitted scattered signal phase matrix includes: Using the formula to determine the scattered signal amplitude transfer matrix and the scattered signal phase transfer matrix; wherein, , ; In the above formula, is the element in the nth row and nth column of the received scattered signal amplitude matrix, is the element in the nth row and 1st column of the emitted scattered signal amplitude matrix, is the element in the nth row and nth column of the received scattered signal phase matrix, is the element in the nth row and nth column of the emitted scattered signal phase matrix, is the element in the nth row and nth column of the scattered signal amplitude transfer matrix, is the element in the nth row and nth column of the scattered signal phase transfer matrix.

[0008] Optionally, the method further includes: Constructing a transformer physical simulation model of the transformer under test; When the transformer physical simulation model is running, controlling N simulation signal generators to emit N simulation scattered signals with different frequencies, and obtaining N×N scattered signals received by N simulation signal receivers, where N pairs of simulation signal generators and simulation signal receivers are installed on the oil tank wall of the transformer physical simulation model; Determining a simulation emitted scattered signal matrix according to N simulation scattered signals with different frequencies, and determining a simulation scattered signal matrix according to N×N scattered signals; Determine the simulation scattering signal transfer matrix based on the simulated scattering signal matrix and the simulated received scattering signal matrix; According to the normal mechanical state change range of the transformer under test under normal electrodynamic action, adjust the simulation parameters of the transformer physical simulation model, and each time an adjustment is made, return to execute the step of controlling N simulation signal generators to emit N simulation scattering signals with different frequencies, and obtain the N×N scattering signals received by N simulation signal receivers, until the adjustment is completed, obtain multiple simulation scattering signal transfer matrices, and use all the multiple simulation scattering signal transfer matrices as the normal scattering signal transfer matrices; According to the abnormal mechanical state change range of the transformer under test under the breakthrough of normal electrodynamic action, adjust the simulation parameters of the transformer physical simulation model, and each time an adjustment is made, return to execute the step of controlling N simulation signal generators to emit N simulation scattering signals with different frequencies, and obtain the N×N scattering signals received by N simulation signal receivers, until the adjustment is completed, obtain multiple simulation scattering signal transfer matrices, and use all the multiple simulation scattering signal transfer matrices as the abnormal scattering signal transfer matrices; Input multiple normal adjusted scattering signal transfer matrices and multiple abnormal adjusted scattering signal transfer matrices into the initial machine learning model for training to obtain the preset machine learning model.

[0009] Optionally, before the step of controlling N simulation signal generators to emit N simulation scattering signals with different frequencies and obtaining the N×N scattering signals received by N simulation signal receivers, the method further includes: Calibrate the simulation parameters of the transformer physical simulation model according to the reference scattering signal transfer matrix.

[0010] Optionally, the method further includes: When the transformer under test is not operating, control N signal generators to emit N scattering signals with different frequencies, obtain the N×N scattering signals received by N signal receivers, and use all the N scattering signals with different frequencies as the reference emitted scattering signals, and use all the N×N scattering signals as the reference received scattering signals; Determine the reference emitted scattering signal matrix according to the N reference emitted scattering signals with different frequencies, and determine the reference received scattering signal matrix according to the N×N reference received scattering signals; Determine the reference scattering signal transfer matrix according to the reference emitted scattering signal matrix and the reference received scattering signal matrix.

[0011] Optionally, the determining the emitted scattering signal matrix according to N scattering signals with different frequencies and the received scattering signal matrix according to N×N scattering signals includes: Amplitude and phase extraction are performed on each of the N scattered signals with different frequencies to obtain N transmitted scattered signal amplitudes and N transmitted scattered signal phases; An N×1 transmitted scattered signal amplitude matrix is determined based on the N transmitted scattered signal amplitudes, and an N×N transmitted scattered signal phase matrix is determined based on the N transmitted scattered signal phases; The N×1 transmitted scattered signal amplitude matrix and the N×N transmitted scattered signal phase matrix are used as the transmitted scattered signal matrix; Amplitude and phase extraction are performed on each of the N×N scattered signals to obtain N×N received scattered signal amplitudes and N×N received scattered signal phases; An N×1 received scattered signal amplitude matrix is determined based on the N×N received scattered signal amplitudes, and an N×N received scattered signal phase matrix is determined based on the N×N received scattered signal phases; The N×1 received scattered signal amplitude matrix and the N×N received scattered signal phase matrix are used as the received scattered signal matrix.

[0012] Optionally, controlling the N signal generators to emit N scattered signals with different frequencies and acquiring the N×N scattered signals received by the N signal receivers includes: Evenly divide the preset frequency domain range into N frequency domain ranges; Extract the j-th frequency of each frequency domain range to obtain a scattered signal frequency group, where the initial value of j is 1; Based on the scattered signal frequency group, control the N signal generators to emit N scattered signals with different frequencies and acquire the N×N scattered signals received by the N signal receivers; Evaluating whether the mechanical state of the transformer under test has changed according to the discrimination result includes: Return to execute the steps of determining the transmitted scattered signal matrix according to the N scattered signals with different frequencies and determining the received scattered signal matrix according to the N×N scattered signals until the discrimination result is obtained; Let j = j + 1, and return to execute the step of extracting the j-th frequency of each frequency domain range to obtain a scattered signal frequency group until j is equal to the maximum frequency sweep times to obtain multiple discrimination results; Evaluate whether the mechanical state of the transformer under test has changed according to all the discrimination results.

[0013] Optionally, controlling the N signal generators to emit N scattered signals with different frequencies and acquiring the N×N scattered signals received by the N signal receivers includes: Control N signal generators to emit N scattered signals with different frequencies, and obtain the N×N intermediate scattered signals received by N signal receivers at the i-th time, where the initial value of i is 1; Let i = i + 1, and return to execute the step of controlling N signal generators to emit N scattered signals with different frequencies and obtaining the N×N intermediate scattered signals received by N signal receivers at the i-th time, until the attenuation value of the intermediate scattered signals corresponding one-to-one between the N×N intermediate scattered signals received at the i-th time and the N×N intermediate scattered signals received at the first time is less than the attenuation threshold, and obtain the N×N intermediate scattered signals received multiple times; Among the N×N intermediate scattered signals received multiple times, take the N×N intermediate scattered signals received at any one time as the N×N scattered signals.

[0014] To achieve the above object, the present invention provides a transformer mechanical state online rapid evaluation system in the second aspect. The system includes N signal generators, N signal receivers and a processor; N pairs of signal generators and signal receivers are installed on the oil tank wall of the transformer to be tested; The processor is used to execute the method described in any item of the first aspect.

[0015] To achieve the above object, the present invention provides a transformer mechanical state online rapid evaluation device in the third aspect. The device includes: A control module, configured to control N signal generators to emit N scattered signals with different frequencies and obtain the N×N scattered signals received by N signal receivers when the transformer to be tested is operating, where N pairs of signal generators and signal receivers have been installed on the oil tank wall of the transformer to be tested, and N is greater than or equal to 2; A first determination module, configured to determine a transmitted scattered signal matrix according to the N scattered signals with different frequencies, and determine a received scattered signal matrix according to the N×N scattered signals; A second determination module, configured to determine a scattered signal transfer matrix according to the transmitted scattered signal matrix and the received scattered signal matrix; A model discrimination module, configured to input the scattered signal transfer matrix into a preset machine learning model to obtain a discrimination result; An evaluation module, configured to evaluate whether the mechanical state of the transformer to be tested has changed according to the discrimination result.

[0016] To achieve the above object, the present invention provides a computer-readable storage medium in the fourth aspect, storing a computer program, and when the computer program is executed by a processor, the processor is caused to execute the method described in any item of the first aspect.

[0017] To achieve the above object, in a fifth aspect, the present invention provides a computer device, including a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor is caused to execute the method described in any item of the first aspect.

[0018] By adopting the embodiment of the present invention, the following beneficial effects are achieved: Under the condition that the transformer to be tested is operating, the method controls N signal generators to emit N scattering signals with different frequencies, and obtains N×N scattering signals received by N signal receivers. Among them, N pairs of signal generators and signal receivers have been installed on the oil tank wall of the transformer to be tested, and N is greater than or equal to 2. Then, a transmitted scattering signal matrix is determined according to the N scattering signals with different frequencies, and a received scattering signal matrix is determined according to the N×N scattering signals. Then, a scattering signal transfer matrix is determined according to the transmitted scattering signal matrix and the received scattering signal matrix, and the scattering signal transfer matrix is input into a preset machine learning model to obtain a discrimination result. Finally, according to the discrimination result, it is evaluated whether the mechanical state of the transformer to be tested has changed; that is, by installing N pairs of signal generators and signal receivers on the oil tank wall of the transformer to be tested, and evaluating whether the mechanical state of the transformer to be tested has changed through the discrimination result of the model for determining the scattering signal transfer matrix, it is possible to intuitively and accurately evaluate whether the mechanical state of the transformer has changed, without relying on indirect parameters, significantly improving the intuitiveness and accuracy of the evaluation. At the same time, the evaluation ability of the mechanical state of the iron core is also enhanced, and there is no need to combine other special methods, effectively solving the problems of non-intuitive evaluation results, limited accuracy, and weak evaluation ability of the mechanical state of the iron core in the prior art. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings required for use in the description of the embodiments or the prior art. Obviously, the following drawings are only some embodiments of the present invention. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0020] Among them: Figure 1 It is a schematic diagram of a method for online rapid evaluation of the mechanical state of a transformer in an embodiment of the present application.

[0021] Figure 2 It is a schematic diagram of the installation relationship between the transformer to be tested, the signal generator, and the signal receiver shown in an embodiment of the present application; Figure 3 It is a schematic diagram of a device for online rapid evaluation of the mechanical state of a transformer in an embodiment of the present application; Figure 4Internal structure diagram of a computer device in some embodiments. Detailed implementation manners

[0022] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0023] As an indispensable device in the power system, the core function of a transformer is the transmission and distribution of electrical energy. However, a transformer may encounter various faults during operation, especially winding mechanical faults and core mechanical faults, which pose a serious threat to the stable operation of the power system.

[0024] Currently, the industry has developed various methods for evaluating the mechanical state of transformers, such as the short-circuit impedance method, the vibration frequency response method, and the vibration detection method. However, these methods all rely on indirect parameters such as the electrical and mechanical parameters of the transformer for indirect evaluation from the side, resulting in problems such as non-intuitive evaluation results and limited accuracy, and are weak in evaluating the mechanical state of the core, often requiring the combination of other specialized methods.

[0025] In view of the above problems, the present application proposes a method and system for online rapid evaluation of the mechanical state of a transformer, which can intuitively and accurately evaluate whether the mechanical state of the transformer has changed, without relying on indirect parameters, significantly improving the intuitiveness and accuracy of the evaluation. At the same time, the evaluation ability of this method for the mechanical state of the core has also been enhanced, without the need to combine other specialized methods, effectively solving the problems of non-intuitive evaluation results, limited accuracy, and weak evaluation ability for the mechanical state of the core in the prior art. The specific implementation principle will be described in detail in the following embodiments.

[0026] The present application provides a method for online rapid evaluation of the mechanical state of a transformer in the first aspect.

[0027] Please refer to Figure 1 , which is a schematic diagram of a method for online rapid evaluation of the mechanical state of a transformer in an embodiment of the present application. The method includes: Step 110: When the transformer to be tested is operating, control N signal generators to emit N scattered signals with different frequencies, and obtain N×N scattered signals received by N signal receivers, where N pairs of signal generators and signal receivers are installed on the tank wall of the transformer to be tested, and N is greater than or equal to 2.

[0028] Among them, the transformer to be tested refers to the transformer whose mechanical state needs to be evaluated online.

[0029] It should be noted that after controlling N signal generators to emit N scattered signals with different frequencies, the N scattered signals with different frequencies will be reflected in the oil tank of the transformer under test until they are received by N signal receivers, so as to obtain N×N scattered signals.

[0030] Furthermore, it should be noted that the frequencies of the N signal generators need to be different so as to emit N scattered signals with different frequencies, and the frequency bands of each signal receiver need to cover the frequencies of the N signal generators so as to receive scattered signals of all frequencies.

[0031] Regarding the installation method of the N signal generators and the N signal receivers, in some embodiments, a preset number of windows can be cut out on the oil tank wall of the transformer under test, and these windows are filled with insulating materials to form insulating detection windows. Finally, the N signal generators and the N signal receivers are installed on the oil tank wall of the transformer under test; among them, the specific number of the preset windows can be obtained and preset by the operator based on a large amount of experience, experiments or statistics. Of course, it can also be set by the operator according to actual needs.

[0032] In some embodiments, the specific number of the preset windows can be 2N or N, depending on the situation.

[0033] In this application, preferably, the paired signal generators and signal receivers are arranged on the diagonal lines of the oil tank of the transformer under test.

[0034] For example, assuming that the number of pairs of signal generators and signal receivers is 2, reference can be made to Figure 2 , which is a schematic diagram of the installation relationship between the transformer under test and the signal generators and signal receivers shown in the embodiments of this application. In this schematic diagram, 210 is the oil tank of the transformer under test, 220 is the signal generator, 230 is the signal receiver, and 240 is the insulating detection window.

[0035] Step 120: Determine the transmitted scattered signal matrix according to the N scattered signals with different frequencies, and determine the received scattered signal matrix according to the N×N scattered signals.

[0036] In some embodiments, the phases of the scattered signals at N different frequencies can be used to form an emitted scattered signal phase matrix, the amplitudes of the scattered signals at N different frequencies can be used to form an emitted scattered signal amplitude matrix, the integral values of the scattered signals at N different frequencies can be used to form an emitted scattered signal integral value matrix, or the average values of the data points of the scattered signals at N different frequencies can be used to form an emitted scattered signal data point average matrix. Then, at least one of the emitted scattered signal phase matrix, the emitted scattered signal amplitude matrix, the emitted scattered signal integral value matrix, and the emitted scattered signal data point average matrix is used as the emitted scattered signal matrix; similarly, the received scattered signal matrix is determined in a similar manner.

[0037] Step 130: Determine the scattered signal transfer matrix based on the emitted scattered signal matrix and the received scattered signal matrix.

[0038] It should be noted that the scattered signal transfer matrix corresponds to the emitted scattered signal matrix and the received scattered signal matrix; for example, when the scattered signal transfer matrix includes the emitted scattered signal phase matrix and the received scattered signal matrix includes the received scattered signal phase matrix, the scattered signal transfer matrix includes the scattered signal phase transfer matrix. When the scattered signal transfer matrix includes the emitted scattered signal phase matrix and the emitted scattered signal amplitude matrix, and the received scattered signal matrix includes the received scattered signal phase matrix and the received scattered signal amplitude matrix, the scattered signal transfer matrix includes the scattered signal phase transfer matrix and the scattered signal amplitude transfer matrix.

[0039] For the determination method of the scattered signal transfer matrix, in some embodiments, it can be determined by solving one unknown with two known numbers, that is, it can be determined by formulating equations.

[0040] Step 140: Input the scattered signal transfer matrix into a preset machine learning model to obtain a discrimination result.

[0041] Here, the preset machine learning model refers to a machine learning model that has been trained and can be directly used to predict and output a discrimination result based on the input scattered signal transfer matrix.

[0042] In some embodiments, a large number of scattering signal transfer matrices and the discrimination results corresponding to the scattering signal transfer matrices can be used. Then, all the scattering signal transfer matrices and the discrimination results corresponding to them are sequentially input into the initial machine learning model for training. After training to a certain extent, a trained preset machine learning model can be obtained. Among them, the discrimination result corresponding to the scattering signal transfer matrix can be used as the true value of the initial machine learning model. That is, by comparing this true value with the discrimination result output during the training process one by one, it can be determined whether the initial machine learning model is well-trained and meets the expected requirements.

[0043] In some embodiments, the discrimination result can be 1 or 0 (or approaching 1 or approaching 0), can also be true or false, or can also be right or wrong, etc., so as to clearly evaluate whether the mechanical state of the transformer under test has changed.

[0044] Step 150: According to the discrimination result, evaluate whether the mechanical state of the transformer under test has changed.

[0045] It should be noted that since the scattering signal matrix is obtained by the reflection of N scattering signals with different frequencies in the oil tank of the transformer under test, the scattering signal matrix can reflect the mechanical conditions in the oil tank of the transformer under test. And the scattering signal transfer matrix is also obtained based on the scattering signal matrix. Therefore, it can be used to evaluate whether the mechanical state of the transformer under test has changed through the model determination result of the scattering signal transfer matrix.

[0046] In some embodiments, if the determination result is 1 or approaching 1, it indicates that the mechanical state of the transformer under test has changed. If the determination result is 0 or approaching 0, it indicates that the mechanical state of the transformer under test has not changed.

[0047] It should also be noted that in this method of the present application, in addition to being able to evaluate whether the mechanical state of the winding and the core or their positions have changed, it can also evaluate whether the mechanical state or position of other mechanical components in the transformer under test has changed; for example, the spacers, clamping parts, fasteners, etc. of the transformer under test.

[0048] In the embodiments of the present application, by installing N pairs of signal generators and signal receivers on the oil tank wall of the transformer under test, and determining the model discrimination result of the scattering signal transfer matrix to evaluate whether the mechanical state of the transformer under test has changed, it can intuitively and accurately evaluate whether the mechanical state of the transformer has changed, without relying on indirect parameters, significantly improving the intuitiveness and accuracy of the evaluation. At the same time, the evaluation ability of the mechanical state of the core is also enhanced, without the need to combine other special methods, effectively solving the problems of non-intuitive evaluation results, limited accuracy, and weak evaluation ability of the mechanical state of the core in the prior art.

[0049] In addition, the online rapid assessment method for the mechanical state of the transformer proposed in this application, in addition to the above-mentioned ability to intuitively and accurately evaluate the mechanical state of the transformer and enhance the evaluation ability of the mechanical state of the iron core, also has the following advantages: Real-time monitoring and early warning: Through online rapid assessment, the mechanical state of the transformer can be monitored in real time, potential fault hazards can be detected in time, the expansion of faults can be avoided, the power outage time and economic losses can be reduced, and based on the discrimination result, the system can set an early warning threshold. When the discrimination result approaches or exceeds the threshold, the early warning mechanism is automatically triggered to notify the operation and maintenance personnel to take measures in time; Improve operation and maintenance efficiency: The traditional assessment of the mechanical state of the transformer often requires manual inspection, which is time-consuming and laborious. The method of this application can realize automatic assessment, reduce the frequency and intensity of manual inspection, improve operation and maintenance efficiency, and through the analysis of the scattering signal transfer matrix, the fault location can be more accurately located, providing detailed fault information for the operation and maintenance personnel for quick repair; Enhance system reliability: Through real-time monitoring and early warning, potential faults of the transformer can be detected in advance for preventive maintenance, the service life of the transformer can be extended, and the reliability of the power system can be improved. And timely detection and handling of mechanical faults of the transformer can avoid power outage accidents caused by faults and ensure the continuity and stability of power supply; Reduce operation and maintenance costs: Through preventive maintenance and accurate fault location, the maintenance costs and power outage losses caused by faults can be reduced. The method of this application can realize automatic assessment of the mechanical state of the transformer, reduce the dependence on professional operation and maintenance personnel, and optimize the allocation of human resources; Strong adaptability: The method of this application does not depend on the specific type or model of the transformer, has strong versatility and adaptability, can be widely applied to different types of transformers, and in addition to the mechanical state of the winding and the mechanical state of the iron core, it can also evaluate whether the mechanical state or position of other mechanical components in the transformer under test has changed, such as spacers, clamping parts, fasteners, etc.; Technical advancement: This application uses a machine learning model for discrimination, trains with a large amount of data, improves the accuracy and reliability of the assessment, and at the same time realizes rapid assessment. And the mechanical state information of the transformer is obtained through scattering signal technology, which is a novel and effective technical means with high technical content and innovation.

[0050] In a feasible implementation manner, the received scattering signal matrix in the above embodiment includes a received scattering signal amplitude matrix and a received scattering signal phase matrix, the emitted scattering signal matrix includes an emitted scattering signal amplitude matrix and an emitted scattering signal phase matrix, and the scattering signal transfer matrix includes a scattering signal amplitude transfer matrix and a scattering signal phase transfer matrix.

[0051] In step 130 of the above embodiments, determining the scattering signal transfer matrix based on the transmitted scattering signal matrix and the received scattering signal matrix includes: determining the scattering signal amplitude transfer matrix based on the received scattering signal amplitude matrix and the transmitted scattering signal amplitude matrix, and determining the scattering signal phase transfer matrix based on the received scattering signal phase matrix and the transmitted scattering signal phase matrix.

[0052] In the embodiments of the present application, by separately processing the amplitude and phase information of the scattering signal, not only the comprehensiveness and accuracy of the mechanical state evaluation of the transformer are improved, but also the sensitivity and reliability of the evaluation are enhanced, providing a strong guarantee for the safe and stable operation of the transformer.

[0053] It can be understood that improving the comprehensiveness of the evaluation: by simultaneously considering the amplitude and phase information of the scattering signal, the mechanical state of the transformer can be more comprehensively reflected. The amplitude information can reflect the intensity of the signal, while the phase information can reflect the timing characteristics of the signal. The combination of the two can provide richer state information; enhancing the accuracy of the evaluation: separately processing the amplitude and phase information can avoid the evaluation deviation caused by a single parameter. For example, some mechanical faults may mainly affect the amplitude of the signal, while other faults may mainly affect the phase of the signal. By comprehensively considering both, the mechanical state of the transformer can be more accurately judged; improving the sensitivity of the evaluation: since the amplitude and phase information have different sensitivities to the changes in the mechanical state, separately processing can more easily detect small state changes. This high sensitivity helps to detect and take measures in a timely manner at the initial stage of the fault to prevent the expansion of the fault; enhancing the reliability of the evaluation: separately processing the amplitude and phase information can mutually verify the evaluation results. If the evaluation results of the two are consistent, the reliability of the evaluation can be increased. If there are differences between the two, the reasons can be further analyzed to improve the accuracy of the evaluation; adapting to different fault types: different mechanical faults may have different effects on the amplitude and phase of the scattering signal. By separately processing the two, it is easier to adapt to the evaluation requirements of different types of faults and improve the generality and adaptability of the evaluation.

[0054] In a feasible implementation manner, determining the scattering signal amplitude transfer matrix based on the received scattering signal amplitude matrix and the transmitted scattering signal amplitude matrix, and determining the scattering signal phase transfer matrix based on the received scattering signal phase matrix and the transmitted scattering signal phase matrix in the above embodiments includes: Using the formula to determine the scattering signal amplitude transfer matrix and the scattering signal phase transfer matrix; where , ; In the above formula, is the element in the nth row and nth column of the received scattering signal amplitude matrix, is the element in the first column of the n-th row of the scattered signal amplitude matrix is the element in the n-th column of the n-th row of the received scattered signal phase matrix is the element in the n-th column of the n-th row of the transmitted scattered signal phase matrix is the element in the n-th column of the n-th row of the scattered signal amplitude transfer matrix is the element in the n-th column of the n-th row of the scattered signal phase transfer matrix.

[0055] In some embodiments, It can also be expressed as the amplitude of the n×n-th scattered signal in the received scattered signal amplitude matrix, It can also be expressed as the amplitude of the n-th scattered signal in the transmitted scattered signal amplitude matrix, It can also be expressed as the phase of the n×n-th scattered signal in the received scattered signal phase matrix, It can also be expressed as the phase of the n-th scattered signal in the transmitted scattered signal phase matrix.

[0056] In the embodiments of the present application, by defining formulas to determine the scattered signal amplitude transfer matrix and the scattered signal phase transfer matrix, a specific and quantifiable calculation method is provided for the on-line rapid assessment of the mechanical state of the transformer, enhancing the accuracy and reliability of the assessment.

[0057] It can be understood that providing a quantitative calculation method: in the actual process of evaluating the mechanical state of the transformer, how to accurately calculate the scattered signal amplitude transfer matrix and the scattered signal phase transfer matrix is a key issue. This implementation provides specific calculation formulas, making the evaluation process have a clear quantitative basis and avoiding subjectivity and uncertainty in the evaluation process; enhancing evaluation accuracy: through precise formula calculations, the values of the scattered signal amplitude transfer matrix and the scattered signal phase transfer matrix can be obtained more accurately. The values of these matrices directly reflect the information of the mechanical state inside the transformer tank. Accurate matrix values help to more precisely evaluate whether the mechanical state of the transformer has changed, improving the accuracy of the evaluation; improving evaluation reliability: the definitions of the parameters in the formulas are clear, such as the meanings of the elements in the received scattered signal amplitude matrix, the transmitted scattered signal amplitude matrix, the received scattered signal phase matrix, the transmitted scattered signal phase matrix, etc. This clear definition and calculation method reduce the sources of errors in the evaluation process, making the evaluation results more reliable. At the same time, the subsequent discrimination results based on accurate matrix values are also more credible, providing a stronger guarantee for the safe and stable operation of the transformer.

[0058] In a feasible implementation manner, the method in the above embodiment further includes: constructing a transformer physical simulation model of the transformer to be tested; when the transformer physical simulation model is running, controlling N simulation signal generators to emit N simulation scattering signals with different frequencies, and obtaining N×N scattering signals received by N simulation signal receivers, where N pairs of simulation signal generators and simulation signal receivers are installed on the oil tank wall of the transformer physical simulation model; determining a simulation emitted scattering signal matrix according to the N simulation scattering signals with different frequencies, and determining a simulation scattering signal matrix according to the N×N scattering signals; determining a simulation scattering signal transfer matrix according to the simulation emitted scattering signal matrix and the simulation received scattering signal matrix; according to the normal mechanical state change range of the transformer to be tested under the action of normal electrodynamic force, adjusting the simulation parameters of the transformer physical simulation model, and each time an adjustment is made, returning to execute the step of controlling N simulation signal generators to emit N simulation scattering signals with different frequencies and obtaining N×N scattering signals received by N simulation signal receivers until the adjustment ends, obtaining a plurality of simulation scattering signal transfer matrices, and taking all the plurality of simulation scattering signal transfer matrices as normal scattering signal transfer matrices; according to the abnormal mechanical state change range of the transformer to be tested under the action of breaking through the normal electrodynamic force, adjusting the simulation parameters of the transformer physical simulation model, and each time an adjustment is made, returning to execute the step of controlling N simulation signal generators to emit N simulation scattering signals with different frequencies and obtaining N×N scattering signals received by N simulation signal receivers until the adjustment ends, obtaining a plurality of simulation scattering signal transfer matrices, and taking all the plurality of simulation scattering signal transfer matrices as abnormal scattering signal transfer matrices; inputting the plurality of normal adjusted scattering signal transfer matrices and the plurality of abnormal adjusted scattering signal transfer matrices into an initial machine learning model for training to obtain a preset machine learning model.

[0059] Among them, both the normal mechanical state change range and the abnormal mechanical state change range can be obtained and set by the operator based on a large amount of experience, experiments or statistics. Of course, they can also be set by the operator according to actual needs.

[0060] In some embodiments, preferably in this application, when the normal mechanical state change range is set to [-M, +M], the abnormal mechanical state change range is set to [-3M, -M) and (+M, +3M]; where M is a positive integer.

[0061] For the adjustment method of the simulation parameters, in some embodiments, within the mechanical state change range, the simulation parameters of the transformer physical simulation model can be adjusted incrementally in sequence according to a preset step size; where the preset step size can be obtained and set by the operator based on a large amount of experience, experiments or statistics. Of course, it can also be set by the operator according to actual needs. For example, it can be set to 1mm.

[0062] For the limiting conditions for the end of adjustment, in some embodiments, the mechanical state change range can be used as the criterion. For example, within the mechanical state change range, the simulation parameters of the transformer physical simulation model are incrementally adjusted in sequence according to a preset step size. When the adjustment has been made from the lower limit to the upper limit of the mechanical state change range, it indicates that the adjustment has ended.

[0063] In the embodiments of the present application, by constructing a transformer physical simulation model and simulating the scattering signal transfer matrix under different mechanical states, rich and accurate training data is provided for the preset machine learning model, significantly improving the prediction accuracy and generalization ability of the preset machine learning model, thereby enhancing the overall performance of the on-line rapid evaluation method for the mechanical state of the transformer and effectively solving the problems of insufficient training data and limited prediction accuracy of the preset machine learning model.

[0064] It can be understood that providing rich and accurate training data: During the training process of a preset machine learning model, how to obtain rich and accurate training data is a key issue. This implementation method constructs a transformer physical simulation model, simulates different mechanical states of the transformer under normal electrodynamic forces and breakthrough normal electrodynamic forces, and obtains the corresponding scattering signal transfer matrix as training data, enabling the training process of the preset machine learning model to have a clear quantitative basis and avoiding subjectivity and uncertainty during the training process; enhancing prediction accuracy: By accurately simulating the scattering signal transfer matrix under different mechanical states, it can more accurately reflect the information of the mechanical state inside the transformer tank. These accurate scattering signal transfer matrices are input into the preset machine learning model as training data, which helps the preset machine learning model to more precisely learn the mapping relationship between the mechanical state and the scattering signal transfer matrix, thereby improving the prediction accuracy of the preset machine learning model; improving generalization ability: The transformer physical simulation model can simulate a variety of different mechanical state change ranges, including normal and abnormal states. By obtaining the scattering signal transfer matrix as training data under these different states, the preset machine learning model can learn a wider range of mechanical state change patterns, thereby improving its generalization ability, enabling the preset machine learning model to more accurately predict and discriminate when facing the evaluation of the actual mechanical state of the transformer; reducing the dependence on actual data: In practical applications, obtaining a large amount of accurate transformer mechanical state data often faces many difficulties, such as high data acquisition costs and inaccurate data annotation. This implementation method simulates the scattering signal transfer matrix under different mechanical states through the transformer physical simulation model, reducing the dependence of the preset machine learning model on actual data, reducing the costs of data acquisition and annotation, and at the same time improving the accuracy and reliability of the training data; accelerating the model training process: Since the transformer physical simulation model can quickly simulate the scattering signal transfer matrix under different mechanical states, it can accelerate the training process of the preset machine learning model. Compared with the time required for actual data acquisition and annotation, the transformer physical simulation model can generate a large amount of accurate training data in a short time, thereby improving the training efficiency of the preset machine learning model.

[0065] In a feasible implementation method, before controlling N simulation signal generators to emit N simulation scattering signals with different frequencies and obtaining the N×N scattering signals received by N simulation signal receivers, the method in the above embodiment further includes: calibrating the simulation parameters of the transformer physical simulation model according to the reference scattering signal transfer matrix.

[0066] Among them, the reference scattering signal transfer matrix refers to the reference scattering signal transfer matrix without faults and abnormalities, which can be obtained and preset in advance by the operator based on a large amount of experience, experiments or statistics. Of course, it can also be set by the operator according to actual needs.

[0067] It should be noted that for the matrix categories included in the reference scattering signal transfer matrix (such as the amplitude matrix and the phase matrix), corresponding to the scattering signal transfer matrix, the relevant content of the scattering signal transfer matrix in the above embodiments can be referred to, and will not be elaborated here.

[0068] In the embodiments of the present application, through calibration, the accuracy of the physical simulation model of the transformer is ensured, providing a reliable basis for subsequent training and evaluation based on this model.

[0069] It can be understood that by calibrating the simulation parameters of the physical simulation model of the transformer with reference to the scattering signal transfer matrix, the high accuracy of the simulation model in simulating the mechanical state of the transformer can be ensured, which helps to make the subsequent training and evaluation based on this model more reliable; reducing errors: the calibration process can eliminate or reduce the errors between the simulation model and the actual transformer, making the simulation results closer to the actual situation, which helps to improve the training effect and prediction accuracy of the preset machine learning model; improving training efficiency: an accurate simulation model can generate more accurate training data, thus accelerating the training process of the preset machine learning model. Compared with using an inaccurate simulation model or actual data, the calibrated simulation model can provide the data required for training more efficiently; enhancing evaluation reliability: the evaluation results based on an accurate simulation model are more reliable and can more accurately reflect the actual mechanical state of the transformer, which helps the operation and maintenance personnel to detect and handle potential fault hazards in a timely manner, improving the reliability and stability of the power system.

[0070] In a feasible implementation manner, the method in the above embodiments further includes: when the transformer to be tested is not operating, controlling N signal generators to emit N scattering signals with different frequencies, obtaining the N×N scattering signals received by N signal receivers, and taking the N scattering signals with different frequencies as the reference emitted scattering signals, and taking the N×N scattering signals as the reference received scattering signals; determining the reference emitted scattering signal matrix according to the N reference emitted scattering signals with different frequencies, and determining the reference received scattering signal matrix according to the N×N reference received scattering signals; determining the reference scattering signal transfer matrix according to the reference emitted scattering signal matrix and the reference received scattering signal matrix.

[0071] It should be noted that for the matrix categories included in the reference received scattering signal and the reference emitted scattering signal matrix (such as the amplitude matrix and the phase matrix), corresponding to the received scattering signal and the emitted scattering signal matrix, the relevant content of the received scattering signal and the emitted scattering signal matrix in the above embodiments can be referred to, and will not be elaborated here.

[0072] In the embodiments of the present application, by obtaining the reference scattering signal transfer matrix when the transformer to be measured is not operating, an accurate benchmark is provided for subsequent evaluation, improving the accuracy and reliability of the evaluation.

[0073] It can be understood that providing an accurate benchmark: obtaining the reference scattering signal transfer matrix when the transformer to be measured is not operating, which reflects the scattering signal characteristics of the transformer in a normal and fault-free state, providing an accurate benchmark for subsequent evaluation of whether the mechanical state of the transformer has changed. During the operation of the transformer, by comparing and analyzing with this reference scattering signal transfer matrix, it is possible to more accurately determine whether the mechanical state is abnormal; improving the evaluation accuracy: since the reference scattering signal transfer matrix is obtained when the transformer is not operating and there is no fault interference, its data has high accuracy and stability. Using this as a benchmark for evaluation can effectively avoid evaluation errors caused by factors such as the transformer's own operating state or external interference, thereby improving the accuracy of the evaluation of the transformer's mechanical state; enhancing the evaluation reliability: the accurate reference scattering signal transfer matrix provides a reliable basis for the evaluation process, making the evaluation results more credible. The operation and maintenance personnel can rely on this evaluation result to more confidently judge the mechanical state of the transformer and take corresponding maintenance measures in a timely manner to ensure the safe and stable operation of the power system.

[0074] In a feasible implementation manner, step 120 in the above embodiment, determining the transmitted scattering signal matrix according to the scattering signals of N different frequencies and determining the received scattering signal matrix according to N×N scattering signals, includes: extracting the amplitude and phase of each of the scattering signals of N different frequencies to obtain N transmitted scattering signal amplitudes and N transmitted scattering signal phases; determining an N×1 transmitted scattering signal amplitude matrix according to the N transmitted scattering signal amplitudes and determining an N×N transmitted scattering signal phase matrix according to the N transmitted scattering signal phases; using the N×1 transmitted scattering signal amplitude matrix and the N×N transmitted scattering signal phase matrix as the transmitted scattering signal matrix; extracting the amplitude and phase of each of the N×N scattering signals to obtain N×N received scattering signal amplitudes and N×N received scattering signal phases; determining an N×1 received scattering signal amplitude matrix according to the N×N received scattering signal amplitudes and determining an N×N received scattering signal phase matrix according to the N×N received scattering signal phases; using the N×1 received scattering signal amplitude matrix and the N×N received scattering signal phase matrix as the received scattering signal matrix.

[0075] In the embodiments of the present application, by finely processing the amplitude and phase information of the scattering signals, a more comprehensive scattering signal matrix is constructed, improving the accuracy and reliability of the evaluation of the transformer's mechanical state.

[0076] It can be understood that a comprehensive scattering signal matrix is constructed: by simultaneously extracting the amplitude and phase of the scattering signals at N different frequencies, a comprehensive scattering signal matrix including the amplitude matrix of the emitted scattering signals, the phase matrix of the emitted scattering signals, the amplitude matrix of the received scattering signals, and the phase matrix of the received scattering signals is constructed. This comprehensive matrix construction method can more completely reflect the characteristics of the scattering signals, provide richer information for the subsequent calculation of the scattering signal transfer matrix, thereby improving the accuracy and reliability of the evaluation; improving the evaluation accuracy: amplitude and phase are two important parameters of the scattering signal, which respectively reflect the intensity and timing characteristics of the signal. By considering these two parameters simultaneously, the evaluation bias caused by a single parameter can be avoided. For example, some mechanical faults may mainly affect the amplitude of the signal, while others may mainly affect the phase of the signal. By comprehensively considering the amplitude and phase information, the mechanical state of the transformer can be judged more accurately, improving the evaluation accuracy; enhancing the evaluation reliability: there is a certain correlation and complementarity between the amplitude and phase information. By processing these two parameters simultaneously, the evaluation results can be mutually verified, increasing the evaluation reliability. If the evaluation results of the amplitude and phase are consistent, the judgment of the mechanical state of the transformer can be further confirmed. If there are differences, the reasons can be further analyzed to improve the evaluation accuracy; adapting to different fault types: different mechanical faults may have different effects on the amplitude and phase of the scattering signal. By processing these two parameters simultaneously, it is easier to adapt to the evaluation requirements of different types of faults. For example, winding looseness may cause a significant change in the signal amplitude, while core offset may mainly affect the signal phase. By comprehensively considering the amplitude and phase information, different types of mechanical faults can be evaluated more comprehensively; providing a solid foundation for subsequent steps: the constructed comprehensive scattering signal matrix is the basis for the subsequent calculation of the scattering signal transfer matrix. An accurate scattering signal matrix can ensure that the calculation result of the scattering signal transfer matrix is more reliable, thereby providing accurate input data for the subsequent discrimination of the machine learning model. This helps to improve the accuracy and reliability of the entire evaluation process and provide a strong guarantee for the safe and stable operation of the transformer.

[0077] In a feasible implementation manner, step 110 in the above embodiment, controlling N signal generators to emit N scattering signals with different frequencies and obtaining N×N scattering signals received by N signal receivers includes: equally dividing the preset frequency domain range into N frequency domain ranges; extracting the j-th frequency of each frequency domain range to obtain a scattering signal frequency group, where the initial value of j is 1; according to the scattering signal frequency group, controlling N signal generators to emit N scattering signals with different frequencies and obtaining N×N scattering signals received by N signal receivers.

[0078] In step 150 of the above embodiments, based on the discrimination result, it is evaluated whether the mechanical state of the transformer under test has changed, including: returning to execute the steps of determining the transmitted scattering signal matrix according to the scattering signals of N different frequencies and determining the received scattering signal matrix according to the N×N scattering signals until the discrimination result is obtained; setting j = j + 1, and returning to execute the step of extracting the j-th frequency of each frequency range to obtain the scattering signal frequency group until j is equal to the maximum number of frequency sweeps to obtain multiple discrimination results; and evaluating whether the mechanical state of the transformer under test has changed according to all the discrimination results.

[0079] Among them, the preset frequency range and the maximum number of frequency sweeps can both be set in advance by the operator based on a large amount of experience, experiments or statistics. Of course, they can also be set by the operator according to actual needs.

[0080] It should be noted that for this frequency sweep process, the reference scattering signal transfer matrix in the above embodiments can have no frequency sweep process, that is, only one frequency sweep is required and multiple frequency sweeps are not needed. While the normal adjustment scattering signal transfer matrix and the abnormal adjustment scattering signal transfer matrix in the above embodiments can have a frequency sweep process, that is, multiple frequency sweeps can be performed to obtain multiple groups of normal adjustment scattering signal transfer matrices and multiple groups of abnormal adjustment scattering signal transfer matrices. And each group of normal adjustment scattering signal transfer matrices has multiple normal adjustment scattering signal transfer matrices, and each group of abnormal adjustment scattering signal transfer matrices has multiple abnormal adjustment scattering signal transfer matrices. Then, multiple groups of normal adjustment scattering signal transfer matrices and multiple groups of abnormal adjustment scattering signal transfer matrices are used for model training. The frequency sweep process can be referred to and will not be elaborated here.

[0081] For the evaluation method of evaluating whether the mechanical state of the transformer under test has changed according to all the discrimination results, in some embodiments, among all the discrimination results, if there is any discrimination result that is 0 or approaches 0, it is evaluated that the mechanical state of the transformer under test has changed; in other embodiments, among all the discrimination results, if there is a preset proportion of discrimination results that are 0 or approach 0, it is evaluated that the mechanical state of the transformer under test has changed; among them, the preset proportion can be set in advance by the operator based on a large amount of experience, experiments or statistics. Of course, it can also be set by the operator according to actual needs.

[0082] In the embodiments of the present application, multiple groups of discrimination results are obtained through the frequency sweep process, and the mechanical state of the transformer under test is comprehensively evaluated, which significantly improves the comprehensiveness and accuracy of the evaluation.

[0083] It is understandable that improving the comprehensiveness of evaluation: Evaluating based on a single scattering signal frequency or a limited set of frequencies may lead to an insufficiently comprehensive evaluation result, overlooking possible mechanical state changes at other frequencies. In this implementation method, through the frequency sweep process, the preset frequency domain range is evenly divided into multiple frequency domain ranges, and the frequencies within each frequency domain range are sequentially extracted for transmitting and receiving scattering signals, thereby obtaining multiple sets of discrimination results. This method can more comprehensively cover various possible mechanical state changes of the transformer under test, improving the comprehensiveness of evaluation and also enhancing the evaluation efficiency; enhancing evaluation accuracy: Multiple sets of discrimination results provide richer information, enabling the evaluation process to more accurately reflect the mechanical state of the transformer under test. By comprehensively analyzing multiple sets of discrimination results, errors caused by evaluating with a single frequency or a limited frequency set can be eliminated or reduced, improving the evaluation accuracy. For example, in some cases, certain mechanical faults may only exhibit obvious characteristics at specific frequencies, and the frequency sweep process can capture these characteristics, thus more accurately judging the mechanical state of the transformer; adapting to different fault types: Different mechanical faults may have different effects on different frequency components of the scattering signal. By obtaining multiple sets of discrimination results through the frequency sweep process, it is easier to adapt to the evaluation requirements of different types of faults. For example, some faults may mainly affect low-frequency components, while others may mainly affect high-frequency components. By comprehensively analyzing multiple sets of discrimination results, different types of mechanical faults can be more comprehensively evaluated; improving evaluation reliability: Multiple sets of discrimination results can be mutually verified, increasing the reliability of the evaluation. If there is consistency among multiple sets of discrimination results, the judgment of the mechanical state of the transformer can be further confirmed. If there are differences, the reasons can be further analyzed to improve the evaluation accuracy. This method helps reduce the risks of misjudgment and missed judgment, improving the evaluation reliability; optimizing the maintenance strategy: Based on the evaluation results of multiple sets of discrimination results, maintenance personnel can formulate more optimized maintenance strategies. For example, for frequency ranges with frequent anomalies, monitoring and inspection can be strengthened, and for frequency ranges judged to be in a normal state, the monitoring frequency can be appropriately reduced, thereby optimizing the allocation of maintenance resources and improving the maintenance efficiency; enhancing system robustness: The frequency sweep process can meet the evaluation requirements under different environments and working conditions, enhancing the robustness of the system. Under different environments and working conditions, the characteristics of the scattering signal may vary. By obtaining multiple sets of discrimination results through the frequency sweep process, different environments and working conditions can be adapted, improving the adaptability and robustness of the system.

[0084] In a feasible implementation manner, controlling the N signal generators to emit N scattered signals with different frequencies and obtaining the N×N scattered signals received by the N signal receivers in the above embodiments includes: controlling the N signal generators to emit N scattered signals with different frequencies, obtaining the N×N intermediate scattered signals received by the N signal receivers at the i-th time, where the initial value of i is 1; making i = i + 1, and returning to execute the step of controlling the N signal generators to emit N scattered signals with different frequencies and obtaining the N×N intermediate scattered signals received by the N signal receivers at the i-th time until the attenuation value of the intermediate scattered signals corresponding one by one between the N×N intermediate scattered signals received at the i-th time and the N×N intermediate scattered signals received at the first time is less than the attenuation threshold, thereby obtaining the N×N intermediate scattered signals received multiple times; among the N×N intermediate scattered signals received multiple times, taking the N×N intermediate scattered signals received at any one time as the N×N scattered signals.

[0085] Among them, the attenuation threshold can be obtained and preset by the operator based on a large amount of experience, experiments or statistics. Of course, it can also be set by the operator according to actual requirements.

[0086] In some embodiments, the attenuation threshold can be set to half of the scattered signal emitted by the signal generator.

[0087] It should be noted that since the N scattered signals emitted by the N signal generators have different frequencies, each signal receiver can identify the number of times of reception when receiving the N scattered signals reflected in the oil tank of the transformer under test.

[0088] Furthermore, it should be noted that when the scattered signal emitted by the signal generator is reflected in the oil tank of the transformer under test, it will gradually attenuate; in this application, the scattered signal emitted by the signal generator is defaulted to attenuate to 0 within one single sweep frequency.

[0089] In the embodiments of this application, stable scattered signals are obtained through the multiple reflection attenuation process, ensuring the accuracy and reliability of the evaluation data and improving the accuracy and efficiency of the transformer mechanical state evaluation.

[0090] It can be understood that to ensure the accuracy of the evaluation data: Inside the transformer oil tank, the scattered signal will undergo multiple reflections and attenuations until it reaches a stable state. In this implementation, the signal generator is controlled to emit scattered signals multiple times, and the intermediate scattered signals after multiple reflections are obtained by the signal receiver until the attenuation value of the intermediate scattered signal is less than the preset attenuation threshold, thereby ensuring that the obtained scattered signal is stable, avoiding evaluation errors caused by unstable signals. The stable scattered signal can more accurately reflect the mechanical state inside the transformer oil tank, providing a reliable data basis for subsequent signal processing and evaluation, and significantly improving the accuracy of the evaluation; To improve the evaluation efficiency: By setting the attenuation threshold, it can be automatically determined when the scattered signal has reached a stable state, thereby reducing the unnecessary number of signal acquisitions and improving the evaluation efficiency. In practical applications, this automatic judgment mechanism can significantly shorten the evaluation time, reduce the evaluation cost, and improve the evaluation efficiency; To enhance the reliability of the evaluation: The stable scattered signal can provide a more reliable evaluation result, reducing the risk of misjudgment and missed judgment caused by unstable signals. By obtaining a stable scattered signal through the multiple reflection and attenuation process and using the N×N intermediate scattered signals received at the same time as the N×N scattered signals, the consistency and reliability of the evaluation result can be ensured, providing a strong guarantee for the safe and stable operation of the transformer; To adapt to different working conditions and environments: Under different working conditions and environments, the attenuation characteristics and stable time of the scattered signal may vary. In this implementation, a stable scattered signal is obtained through the multiple reflection and attenuation process, which can adapt to the evaluation requirements under different working conditions and environments and improve the versatility and adaptability of the evaluation; To optimize the signal processing flow: After obtaining a stable scattered signal, more precise signal processing and analysis can be performed on it, such as filtering, denoising, etc., thereby further improving the accuracy and reliability of the evaluation. The stable scattered signal provides better input data for the subsequent signal processing flow, helping to optimize the entire evaluation process and improve the evaluation efficiency and quality; To reduce the requirements for equipment: By obtaining a stable scattered signal through the multiple reflection and attenuation process, the performance requirements for the signal generator and signal receiver can be reduced. In practical applications, this can reduce the equipment cost and improve the economy of the evaluation. At the same time, the stable scattered signal also helps to reduce the dependence on equipment accuracy and improve the robustness of the evaluation.

[0091] In a second aspect of the present application, an on-line rapid evaluation system for the mechanical state of a transformer is provided. The system includes N signal generators, N signal receivers, and a processor (not shown in the figure, and reference can be made to Figure 2 the schematic diagram shown).

[0092] In a feasible implementation, N pairs of signal generators and signal receivers are installed on the oil tank wall of the transformer to be measured; the processor is used to execute the method according to any one of the first aspect.

[0093] In the embodiments of the present application, the on-line mechanical state evaluation system of the transformer proposed in the present application provides a strong guarantee for the stable operation of the power system through beneficial effects such as integrated design, real-time on-line monitoring, flexible scalability, high-efficiency data processing ability, easy deployment and maintenance, and improved evaluation accuracy and reliability. This system not only solves the problems of non-intuitive evaluation results, limited accuracy, and weak evaluation ability for the mechanical state of the iron core in the prior art, but also reduces the operation and maintenance costs and improves the overall efficiency of the power system.

[0094] It can be understood that system integration and automation: By integrating N signal generators, N signal receivers, and a processor into one system, the automation of on-line evaluation of the mechanical state of the transformer is realized. This integrated design makes the evaluation process more efficient and convenient, reduces the need for manual intervention, and improves the accuracy and reliability of the evaluation; Real-time and on-line monitoring: The system can collect and process scattered signal data in real time. Through the real-time analysis of the data by the processor, on-line monitoring of the mechanical state of the transformer is realized. This real-time nature enables the operation and maintenance personnel to timely understand the operation state of the transformer, discover and handle potential faults in a timely manner, and avoid greater losses caused by the deterioration of the faults; Flexibility and scalability: The number of signal generators and signal receivers in the system can be flexibly configured according to actual needs. This flexibility enables the system to adapt to different types of transformers and different evaluation requirements. At the same time, with the continuous development of technology, the system can also be easily expanded and upgraded to adapt to new evaluation methods and technologies that may appear in the future; High-efficiency data processing ability: As the core component of the system, the processor has strong data processing ability. It can quickly process a large amount of scattered signal data and extract useful information for evaluating the mechanical state of the transformer. This high-efficiency data processing ability improves the efficiency and accuracy of the evaluation, enabling the system to obtain reliable evaluation results in a short time; Easy deployment and maintenance: The deployment of the system is relatively simple. It only needs to install signal generators and signal receivers on the oil tank wall of the transformer to be measured and connect the processor. At the same time, the maintenance of the system is also very convenient. Due to the high degree of integration and automation of the system, the operation and maintenance personnel only need to conduct regular inspections and maintenance to ensure the normal operation of the system; Improve the accuracy and reliability of the evaluation: Through the precise analysis and processing of the scattered signal data by the processor, the system can more accurately evaluate the mechanical state of the transformer. Compared with traditional evaluation methods, this system does not need to rely on indirect parameters and can directly reflect the actual operation state of the transformer, thereby improving the accuracy and reliability of the evaluation.

[0095] The present application provides a device for on-line rapid evaluation of the mechanical state of a transformer in the third aspect.

[0096] Please refer to Figure 3, which is a schematic diagram of an on-line rapid evaluation device for the mechanical state of a transformer in an embodiment of the present application. The device 310 includes: A control module 311, configured to control N signal generators to emit N scattered signals with different frequencies when the transformer under test is operating, and obtain N×N scattered signals received by N signal receivers, where N pairs of signal generators and signal receivers are installed on the oil tank wall of the transformer under test, and N is greater than or equal to 2; A first determination module 312, configured to determine a transmitted scattered signal matrix according to the N scattered signals with different frequencies, and determine a received scattered signal matrix according to the N×N scattered signals; A second determination module 313, configured to determine a scattered signal transfer matrix according to the transmitted scattered signal matrix and the received scattered signal matrix; A model discrimination module 314, configured to input the scattered signal transfer matrix into a preset machine learning model to obtain a discrimination result; An evaluation module 315, configured to evaluate whether the mechanical state of the transformer under test has changed according to the discrimination result.

[0097] In the embodiment of the present application, the relevant content of the above control module 311, first determination module 312, second determination module 313, model discrimination module 314 and evaluation module 315 can refer to Figure 1 the content in the shown embodiment, which will not be elaborated here.

[0098] It should be noted that the device 310 of the present application further includes some other modules. It can be understood that the method of the present application has a one-to-one correspondence with the device 310. Therefore, some other modules of the device 310 of the present application are the corresponding content of the method of the present application in the above embodiment.

[0099] In the embodiment of the present application, by installing N pairs of signal generators and signal receivers on the oil tank wall of the transformer under test, and evaluating whether the mechanical state of the transformer under test has changed through the model discrimination result of the scattered signal transfer matrix, it is possible to intuitively and accurately evaluate whether the mechanical state of the transformer has changed, without relying on indirect parameters, significantly improving the intuitiveness and accuracy of the evaluation. At the same time, the evaluation ability of the mechanical state of the iron core is also enhanced, and there is no need to combine other special devices, effectively solving the problems of non-intuitive evaluation results, limited accuracy and weak evaluation ability of the mechanical state of the iron core in the prior art.

[0100] In addition, the on-line rapid assessment device for the mechanical state of the transformer proposed in this application, in addition to being able to intuitively and accurately evaluate the mechanical state of the transformer and enhance the ability to evaluate the mechanical state of the iron core as mentioned above, also has the following advantages: Real-time monitoring and early warning: Through on-line rapid assessment, the mechanical state of the transformer can be monitored in real time, potential fault hazards can be detected in time, the expansion of faults can be avoided, the power outage time and economic losses can be reduced, and based on the discrimination results, the system can set early warning thresholds. When the discrimination results approach or exceed the thresholds, the early warning mechanism will be automatically triggered to notify the operation and maintenance personnel to take measures in time; Improve operation and maintenance efficiency: The traditional assessment of the mechanical state of the transformer often requires manual inspections, which is time-consuming and laborious. The device of this application can realize automatic assessment, reduce the frequency and intensity of manual inspections, improve operation and maintenance efficiency, and through the analysis of the scattering signal transfer matrix, the fault location can be more accurately located, providing detailed fault information for the operation and maintenance personnel for quick repair; Enhance system reliability: Through real-time monitoring and early warning, potential faults of the transformer can be detected in advance, preventive maintenance can be carried out, the service life of the transformer can be extended, and the reliability of the power system can be improved. Moreover, the timely detection and handling of the mechanical faults of the transformer can avoid power outage accidents caused by faults and ensure the continuity and stability of power supply; Reduce operation and maintenance costs: Through preventive maintenance and accurate fault location, the repair costs and power outage losses caused by faults can be reduced. The device of this application can realize the automatic assessment of the mechanical state of the transformer, reduce the dependence on professional operation and maintenance personnel, and optimize the allocation of human resources; Strong adaptability: The device of this application does not depend on the specific type or model of the transformer, has strong versatility and adaptability, and can be widely applied to different types of transformers. In addition to the mechanical state of the winding and the iron core, it can also evaluate whether the mechanical state or position of other mechanical components in the transformer under test has changed, such as pads, clamping parts, fasteners, etc.; Technological advancement: This application uses a machine learning model for discrimination, trains with a large amount of data, improves the accuracy and reliability of the assessment, and at the same time realizes rapid assessment. Moreover, the mechanical state information of the transformer is obtained through scattering signal technology, which is a novel and effective technical means with high technical content and innovation.

[0101] In the fourth aspect, this application also provides a computer-readable storage medium storing a computer program, which when executed by a processor causes the processor to execute a method for on-line rapid assessment of the mechanical state of a transformer in the above method embodiment.

[0102] In the fifth aspect, this application also provides a computer device including a memory and a processor, the memory storing a computer program, which when executed by the processor causes the processor to execute a method for on-line rapid assessment of the mechanical state of a transformer in the above method embodiment.

[0103] Figure 4The internal structure diagram of a computer device in some embodiments is shown. The computer device may specifically be a terminal, a server, or a gateway. As Figure 4 shown, the computer device includes a processor, a memory, and a network interface connected via a system bus.

[0104] Among them, the memory includes a non-volatile storage medium and an internal memory. The non-volatile storage medium of the computer device stores an operating system and may also store a computer program. When the computer program is executed by the processor, the processor can implement each step in the above method embodiments. The internal memory may also store a computer program. When the computer program is executed by the processor, the processor can execute each step in the above method embodiments. Those skilled in the art can understand that Figure 4 the structure shown in is only a block diagram of some structures related to the solution of this application, and does not constitute a limitation on the computer device to which the solution of this application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine some components, or have different component arrangements.

[0105] Those of ordinary skill in the art can understand that all or part of the processes of implementing the above method embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a non-volatile computer-readable storage medium. When the program is executed, it may include the processes of the above method embodiments.

[0106] Among them, any reference to a memory, storage, database, or other medium used in the embodiments provided in this application may include non-volatile and / or volatile memories. Non-volatile memory may include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), or flash memory. Volatile memory may include random access memory (RAM) or an external cache memory. By way of illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0107] The technical features of the above embodiments can be combined arbitrarily. For the sake of concise description, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, it should be considered as the scope recorded in this specification.

[0108] The above-described embodiments merely represent several implementation manners of the present application. The description is relatively specific and detailed, but it should not be construed as a limitation on the scope of the patent of the present application. It should be noted that for those of ordinary skill in the art, without departing from the concept of the present application, several modifications and improvements can still be made, and these all belong to the protection scope of the present application. Therefore, the protection scope of the patent of the present application shall be subject to the appended claims.

Claims

1. An online rapid assessment method for the mechanical state of a transformer, characterized in that, The method includes: When the transformer under test is operating, controlling N signal generators to emit N scattering signals with different frequencies, and obtaining N×N scattering signals received by N signal receivers, where N pairs of signal generators and signal receivers are installed on the oil tank wall of the transformer under test, and N is greater than or equal to 2; Determining a transmitted scattering signal matrix according to the N scattering signals with different frequencies, and determining a received scattering signal matrix according to the N×N scattering signals; Determining a scattering signal transfer matrix according to the transmitted scattering signal matrix and the received scattering signal matrix; Inputting the scattering signal transfer matrix into a preset machine learning model to obtain a discrimination result; Evaluating whether the mechanical state of the transformer under test has changed according to the discrimination result.

2. The on-line rapid evaluation method for the mechanical state of a transformer according to claim 1, characterized in that The received scattering signal matrix includes a received scattering signal amplitude matrix and a received scattering signal phase matrix, the transmitted scattering signal matrix includes a transmitted scattering signal amplitude matrix and a transmitted scattering signal phase matrix, the scattering signal transfer matrix includes a scattering signal amplitude transfer matrix and a scattering signal phase transfer matrix, and determining the scattering signal transfer matrix according to the transmitted scattering signal matrix and the received scattering signal matrix includes: Determining the scattering signal amplitude transfer matrix according to the received scattering signal amplitude matrix and the transmitted scattering signal amplitude matrix, and determining the scattering signal phase transfer matrix according to the received scattering signal phase matrix and the transmitted scattering signal phase matrix.

3. The online rapid evaluation method for the mechanical state of a transformer according to claim 2, wherein Determining the scattering signal amplitude transfer matrix according to the received scattering signal amplitude matrix and the transmitted scattering signal amplitude matrix, and determining the scattering signal phase transfer matrix according to the received scattering signal phase matrix and the transmitted scattering signal phase matrix includes: Using the formula to determine the amplitude transfer matrix and the phase transfer matrix of the scattering signal; Among them, , ; In the above formula, is the element in the n-th row and n-th column of the received scattered signal amplitude matrix, is the element in the n-th row and 1st column of the transmitted scattered signal amplitude matrix, is the element in the n-th row and n-th column of the received scattered signal phase matrix, is the element in the n-th row and n-th column of the transmitted scattered signal phase matrix, is the element in the n-th row and n-th column of the scattered signal amplitude transfer matrix, is the element in the n-th row and n-th column of the scattered signal phase transfer matrix.

4. The on-line rapid evaluation method for the mechanical state of a transformer according to claim 1, characterized in that The method further includes: Constructing a transformer physical simulation model of the transformer under test; When the transformer physical simulation model is operating, controlling N simulation signal generators to emit N simulation scattering signals with different frequencies, and obtaining N×N scattering signals received by N simulation signal receivers, where N pairs of simulation signal generators and simulation signal receivers are installed on the oil tank wall of the transformer physical simulation model; Determining a simulated transmitted scattering signal matrix according to the N simulation scattering signals with different frequencies, and determining a simulated scattering signal matrix according to the N×N scattering signals; Determining a simulated scattering signal transfer matrix according to the simulated transmitted scattering signal matrix and the simulated received scattering signal matrix; Adjusting the simulation parameters of the transformer physical simulation model according to the normal mechanical state change range of the transformer under test under normal electrodynamic action, and each time adjustment is made, returning to execute the step of controlling N simulation signal generators to emit N simulation scattering signals with different frequencies and obtaining N×N scattering signals received by N simulation signal receivers until the adjustment is completed, obtaining a plurality of simulated scattering signal transfer matrices, and taking the plurality of simulated scattering signal transfer matrices as normal scattering signal transfer matrices; Adjust the simulation parameters of the physical simulation model of the transformer according to the abnormal mechanical state change range of the transformer to be measured under the action of breaking the normal electromagnetic force. And each time an adjustment is made, return to execute the step of controlling N simulation signal generators to emit N simulation scattering signals with different frequencies, and obtaining N×N scattering signals received by N simulation signal receivers, until the adjustment is completed, obtaining multiple simulation scattering signal transfer matrices, and taking all the multiple simulation scattering signal transfer matrices as abnormal scattering signal transfer matrices; Input multiple normal adjustment scattering signal transfer matrices and multiple abnormal adjustment scattering signal transfer matrices into the initial machine learning model for training to obtain the preset machine learning model.

5. The online rapid evaluation method for the mechanical state of a transformer according to claim 4, wherein Before the step of controlling N simulation signal generators to emit N simulation scattering signals with different frequencies and obtaining N×N scattering signals received by N simulation signal receivers, the method further includes: Calibrate the simulation parameters of the physical simulation model of the transformer according to the reference scattering signal transfer matrix.

6. The on-line rapid evaluation method for the mechanical state of a transformer according to claim 5, characterized in that The method further includes: When the transformer to be measured is not operating, control N signal generators to emit N scattering signals with different frequencies, obtain N×N scattering signals received by N signal receivers, and take all N scattering signals with different frequencies as reference emitted scattering signals, and take all N×N scattering signals as reference received scattering signals; Determine a reference emitted scattering signal matrix according to the N reference emitted scattering signals with different frequencies, and determine a reference received scattering signal matrix according to the N×N reference received scattering signals; Determine the reference scattering signal transfer matrix according to the reference emitted scattering signal matrix and the reference received scattering signal matrix.

7. The on-line rapid evaluation method for the mechanical state of a transformer according to claim 1, characterized in that The determining the emitted scattering signal matrix according to the N scattering signals with different frequencies and the determining the received scattering signal matrix according to the N×N scattering signals include: Extract the amplitude and phase of each of the N scattering signals with different frequencies to obtain N emitted scattering signal amplitudes and N emitted scattering signal phases; Determine an N×1 emitted scattering signal amplitude matrix according to the N emitted scattering signal amplitudes, and determine an N×N emitted scattering signal phase matrix according to the N emitted scattering signal phases; Take the N×1 emitted scattering signal amplitude matrix and the N×N emitted scattering signal phase matrix as the emitted scattering signal matrix; Extract the amplitude and phase of each of the N×N scattering signals to obtain N×N received scattering signal amplitudes and N×N received scattering signal phases; Determine an N×1 received scattering signal amplitude matrix according to the N×N received scattering signal amplitudes, and determine an N×N received scattering signal phase matrix according to the N×N received scattering signal phases; Take the N×1 received scattering signal amplitude matrix and the N×N received scattering signal phase matrix as the received scattering signal matrix.

8. The on-line rapid assessment method for the mechanical state of a transformer according to claim 1, characterized in that The controlling N signal generators to emit N scattering signals with different frequencies and obtaining N×N scattering signals received by N signal receivers includes: Evenly divide the preset frequency domain range into N frequency domain ranges; Extract the j-th frequency of each frequency domain range to obtain a scattering signal frequency group, where the initial value of j is 1; According to the scattering signal frequency group, control N signal generators to emit N scattering signals with different frequencies, and obtain N×N scattering signals received by N signal receivers; Evaluating whether the mechanical state of the transformer under test has changed according to the discrimination result includes: Return to execute the steps of determining the scattering signal matrix according to N scattering signals with different frequencies and determining the received scattering signal matrix according to N×N scattering signals until the discrimination result is obtained; Let j = j + 1, and return to execute the step of extracting the j-th frequency of each frequency domain range to obtain a scattering signal frequency group until j is equal to the maximum frequency sweep times, and obtain multiple discrimination results; Evaluate whether the mechanical state of the transformer under test has changed according to all the discrimination results.

9. The on-line rapid evaluation method for the mechanical state of a transformer according to claim 1 or 8, characterized in that The controlling N signal generators to emit N scattering signals with different frequencies and obtaining N×N scattering signals received by N signal receivers includes: Control N signal generators to emit N scattering signals with different frequencies, and obtain N×N intermediate scattering signals received by N signal receivers at the i-th time, where the initial value of i is 1; Let i = i + 1, and return to execute the step of controlling N signal generators to emit N scattering signals with different frequencies and obtaining N×N intermediate scattering signals received by N signal receivers at the i-th time until the attenuation value of the intermediate scattering signals corresponding one by one between the N×N intermediate scattering signals received at the i-th time and the N×N intermediate scattering signals received at the first time is less than the attenuation threshold, and obtain N×N intermediate scattering signals received multiple times; Among the N×N intermediate scattering signals received multiple times, take any one of the N×N intermediate scattering signals received as N×N scattering signals.

10. An on-line rapid evaluation system for the mechanical state of a transformer, characterized in that, The system includes N signal generators, N signal receivers and a processor; N pairs of signal generators and signal receivers are installed on the oil tank wall of the transformer under test; The processor is used to execute the method according to any one of claims 1 to 9.

Images