Method and apparatus for predicting electrical risk of oil-filled paper-insulated equipment

Abstract

This application discloses a method and apparatus for predicting the electrical risk of oil-paper insulated equipment. The method includes: acquiring multiple oil and gas sample data with labels and confidence levels from the oil-paper insulated equipment at different time windows; generating a molecular model formulation of the mixture; constructing a molecular simulation mixture model of the mixture formulation at the current temperature; and performing geometric optimization, model annealing, and ensemble balancing to obtain a steady-state model. The method then calculates the total dipole moment of the molecular dynamics system of the steady-state model and the dielectric loss factor, using the planned quantitative value of the dielectric loss factor to assess the degree of electrical risk of the oil-paper insulated equipment. Therefore, this method provides a mechanistic and quantitative mapping from monitoring data to electrical risk, accurately calculates real-time dielectric parameters, considers individual equipment differences, quantifies risk values, and classifies them, improving the accuracy of early degradation identification and thus enhancing the accuracy of electrical risk monitoring for oil-paper insulated equipment.

Description

Technical Field

[0001] This application relates to the field of electrical risk monitoring technology, and more specifically, to a method and apparatus for predicting electrical risks in oil-paper-insulated equipment. Background Technology

[0002] Oil-immersed transformers are core equipment in power systems. The coexistence of new energy grid connection fluctuations and old equipment causes them to be subjected to multi-field coupling stress for a long time. Insulation deterioration can easily lead to faults such as partial discharge and breakdown. Moreover, the faults evolve quickly and the symptoms are weak. The traditional post-event judgment mode of periodic inspection can no longer meet the development needs of power systems for proactive protection and precise operation and maintenance.

[0003] The mainstream electrical risk diagnosis technologies for oil-insulated paper equipment similar to transformers are DGA (Dissolved Gas Analysis) and PD (Partial Discharge) monitoring technologies. Both of them have problems such as strong reliance on experience and broken mechanism chains. Furthermore, the fault characteristics are not unique and the operating conditions are highly sensitive. They are mainly based on static limit interpretation, making it difficult to update risks in real time. This can easily lead to missed or false alarms in early deterioration, and there is also a lack of comparability between equipment.

[0004] How to provide a more interpretable and quantifiable electrical risk prediction scheme for oil-paper insulated equipment, so as to improve the accuracy of electrical risk monitoring for oil-paper insulated equipment, is an issue that needs attention. Summary of the Invention

[0005] In view of the above problems, this application provides a method and apparatus for predicting electrical risks of oil-paper insulated equipment, so as to improve the accuracy of electrical risk monitoring of oil-paper insulated equipment.

[0006] To achieve the above objectives, the following specific solutions are proposed:

[0007] A method for predicting electrical risks in oil-impregnated paper insulation equipment includes:

[0008] Obtain multiple oil and gas sample data with labels and confidence levels from oil-impregnated paper insulation equipment at different time windows;

[0009] Based on the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment and the data of each oil and gas sample, a mixture molecular model formula is generated.

[0010] A molecular simulation model of the mixture formulation at the current temperature is constructed, and the molecular simulation model is subjected to geometric optimization, model annealing and ensemble equilibrium processing to obtain a steady-state model.

[0011] The dielectric loss factor is calculated by calculating the total dipole moment of the molecular dynamics system in the steady-state model.

[0012] Divide the dielectric loss factor by the specified limit to obtain the risk quantification value, and determine the electrical risk level of the oil-paper insulation equipment through the risk quantification value.

[0013] Optionally, acquiring multiple oil and gas sample data with labels and confidence levels at different time windows for oil-containing paper insulation equipment includes:

[0014] Multiple oil and gas sample data of the oil-paper insulation equipment under different time windows are obtained. Each oil and gas sample data includes multiple data indicators, including the trace water content, oil temperature and various gas contents in the insulating oil of the oil-paper insulation equipment.

[0015] Based on the numerical changes of each data indicator in each of the oil and gas sample data, a label and confidence level are added to the data indicator to obtain oil and gas sample data with labels and confidence levels.

[0016] Optionally, the method further includes:

[0017] Based on the data indicators with added tags and confidence levels, and the molecular model formulation of the mixture, a structured object is generated;

[0018] Construct a molecular simulation mixture model of the structured object at the current temperature.

[0019] Optionally, based on the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment and the oil and gas sample data of each item, a mixture molecular model formulation is generated, including:

[0020] Based on the oil and gas sample data described in each section, calculate the composition ratio of each dissolved gas fraction and trace water.

[0021] Using the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment as a framework, the aforementioned component proportions are added to generate a mixture molecular model formula.

[0022] Optionally, the step of calculating the dielectric loss factor by calculating the total dipole moment of the molecular dynamics system in the steady-state model includes:

[0023] Calculate the total dipole moment of the molecular dynamics system of the steady-state model under isothermal and isochoric ensemble;

[0024] Based on the total dipole moment of the molecular dynamics system, a normalized dipole autocorrelation function is constructed, and a Fourier transform is performed on the normalized dipole autocorrelation function to obtain the losses of the polarizable energy storage and polarization hysteresis conduction mechanisms.

[0025] The dielectric loss factor is calculated based on the polarizable energy storage and the polarization hysteresis conduction mechanism loss.

[0026] Optionally, the normalized dipole autocorrelation function is:

[0027]

[0028] in, Let be the trajectory of the total dipole moment of the molecular dynamics system over time. This represents the average value of the dipole moment at all moments during the simulation of the steady-state model. The total dipole moment of the system at the initial moment of the steady-state model simulation is given by... For average statistical calculation, for and The Euclidean inner product.

[0029] Optionally, a Fourier transform is performed on the normalized dipole autocorrelation function to obtain the losses of polarizable energy storage and polarization hysteresis conduction mechanisms, including:

[0030] The losses of polarizable energy storage and polarization-hysteresis conduction mechanisms can be calculated using the following formula:

[0031]

[0032] in, For polarizable energy storage, The loss is due to the polarization hysteresis conduction mechanism. The dielectric constant of optical frequency, For relaxation strength, Angular frequency, The equivalent DC conductivity of the material in the oil-impregnated paper insulation equipment is given. For the volume of the molecular dynamics system, The vacuum permittivity, Boltzmann's constant, This is the current thermodynamic temperature.

[0033] Optionally, based on the polarizable energy storage and the polarization hysteresis conduction mechanism loss, the dielectric loss factor is calculated, including:

[0034] Divide the polarization hysteresis conduction mechanism loss by the polarizable energy storage to obtain the dielectric loss factor.

[0035] Optionally, determining the electrical risk level of the oil-paper insulation equipment using the risk quantification value includes:

[0036] When the risk quantification value is less than the first quantification threshold, the electrical risk level of the oil-paper insulation equipment is determined to be normal.

[0037] When the risk quantification value is not less than the first quantification threshold and is less than the second quantification threshold, the electrical risk level of the oil-paper insulation equipment is determined to be of concern.

[0038] When the risk quantification value is not less than the second quantification threshold and less than the third quantification threshold, the electrical risk level of the oil-paper insulation equipment is determined to be an alarm.

[0039] When the risk quantification value is not less than the third quantification threshold, the electrical risk level of the oil-paper insulation equipment is determined to be critical.

[0040] An electrical risk prediction device for oil-impregnated paper insulation equipment, comprising:

[0041] The oil and gas sample data acquisition unit is used to acquire multiple oil and gas sample data with labels and confidence levels at different time windows for oil-containing paper insulation equipment.

[0042] The mixture molecular model formulation generation unit is used to generate a mixture molecular model formulation based on the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment and the oil and gas sample data of each of the oil and gas samples.

[0043] The model building unit is used to build a molecular simulation mixture model of the mixture molecular model formulation at the current temperature, and to perform geometric optimization, model annealing and ensemble equilibrium processing on the molecular simulation mixture model to obtain a steady-state model;

[0044] The dielectric loss factor calculation unit is used to calculate the dielectric loss factor by calculating the total dipole moment of the molecular dynamics system of the steady-state model.

[0045] The electrical risk level quantification unit is used to divide the dielectric loss factor by the specification limit to obtain the risk quantification value, and to determine the electrical risk level of the oil-paper insulation equipment through the risk quantification value.

[0046] Optionally, the oil and gas sample data acquisition unit includes:

[0047] The sample acquisition unit is used to acquire multiple oil and gas sample data of the oil-paper insulation equipment under different time windows. Each oil and gas sample data includes multiple data indicators, including the trace water content, oil temperature and various gas contents in the insulating oil of the oil-paper insulation equipment.

[0048] The label confidence level addition unit is used to add labels and confidence levels to each data indicator based on the numerical changes of each data indicator in each of the oil and gas sample data, so as to obtain oil and gas sample data with labels and confidence levels.

[0049] Optionally, the device may also include:

[0050] The structured object generation unit is used to generate structured objects based on various data indicators with added tags and confidence levels, as well as the molecular model formula of the mixture.

[0051] A molecular simulation mixture model building unit is used to build a molecular simulation mixture model of the structured object at the current temperature.

[0052] Optionally, the mixture molecular model formulation generation unit includes:

[0053] The composition ratio calculation unit is used to calculate the composition ratio of each dissolved gas fraction and trace water based on the oil and gas sample data mentioned above.

[0054] The formulation generation unit is used to generate a mixture molecular model formulation by adding the composition proportions based on the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment.

[0055] Optionally, the dielectric loss factor calculation unit includes:

[0056] The total dipole moment calculation unit for molecular dynamics systems is used to calculate the total dipole moment of the molecular dynamics system under the steady-state model in an isothermal and isochoric ensemble.

[0057] The Fourier transform unit is used to construct a normalized dipole autocorrelation function based on the total dipole moment of the molecular dynamics system, and to perform a Fourier transform on the normalized dipole autocorrelation function to obtain the losses of the polarizable energy storage and polarization hysteresis conduction mechanisms.

[0058] The dielectric loss factor determination unit is used to calculate the dielectric loss factor based on the polarizable energy storage and the polarization hysteresis conduction mechanism loss.

[0059] Optionally, the normalized dipole autocorrelation function is:

[0060]

[0061] in, Let be the trajectory of the total dipole moment of the molecular dynamics system over time. This represents the average value of the dipole moment at all moments during the simulation of the steady-state model. The total dipole moment of the system at the initial moment of the steady-state model simulation is given by... For average statistical calculation, for and The Euclidean inner product.

[0062] Optionally, the Fourier transform unit includes:

[0063] The first formula calculation unit is used to calculate the losses of polarizable energy storage and polarization-hysteresis conduction mechanisms using the following formula:

[0064]

[0065] in, For polarizable energy storage, The dielectric constant of optical frequency, For relaxation strength, Angular frequency, The equivalent DC conductivity of the material in the oil-impregnated paper insulation equipment is given. For the volume of the molecular dynamics system, The vacuum permittivity, Boltzmann's constant, This is the current thermodynamic temperature.

[0066] Optionally, the dielectric loss factor determination unit includes:

[0067] Divide the polarization hysteresis conduction mechanism loss by the polarizable energy storage to obtain the dielectric loss factor.

[0068] Optionally, the electrical risk quantification unit includes:

[0069] The normality quantification unit is used to determine that the electrical risk level of the oil-paper insulation equipment is normal when the risk quantification value is less than the first quantification threshold.

[0070] The attention level quantification unit is used to determine the electrical risk level of the oil-paper insulation equipment as requiring attention when the risk quantification value is not less than the first quantification threshold and less than the second quantification threshold.

[0071] An alarm level quantification unit is used to determine the electrical risk level of the oil-paper insulation equipment as an alarm when the risk quantification value is not less than the second quantification threshold and less than the third quantification threshold.

[0072] The urgency quantification unit is used to determine the electrical risk level of the oil-paper insulation equipment as critical when the risk quantification value is not less than the third quantification threshold.

[0073] By employing the above technical solution, this application acquires multiple oil and gas sample data with tags and confidence levels from oil-paper insulated equipment at different time windows. Based on the cellulose molecular model of the insulating paper of the oil-paper insulated equipment and each oil and gas sample data, a mixture molecular model formula is generated. A molecular simulation mixture model of the mixture molecular model formula at the current temperature is constructed. The molecular simulation mixture model is then subjected to geometric optimization, model annealing, and ensemble balancing to obtain a steady-state model. By calculating the total dipole moment of the molecular dynamics system of the steady-state model, the dielectric loss factor is calculated. The dielectric loss factor is divided by the specification limit to obtain a risk quantification value. The electrical risk level of the oil-paper insulated equipment is determined by the risk quantification value. Therefore, this application demonstrates a mechanistic and quantitative mapping from monitoring data to electrical risk, accurate calculation of real-time dielectric parameters, consideration of individual equipment differences, quantification of risk values, and graded judgment, thereby improving the accuracy of early degradation identification and enhancing the accuracy of electrical risk monitoring for oil-paper insulated equipment. Attached Figure Description

[0074] Various other advantages and benefits will become apparent to those skilled in the art upon reading the following detailed description of preferred embodiments. The accompanying drawings are for illustrative purposes only and are not intended to limit the scope of this application. Furthermore, the same reference numerals denote the same parts throughout the drawings. In the drawings:

[0075] Figure 1 This application provides a schematic flowchart for predicting electrical risks in oil-paper-insulated equipment.

[0076] Figure 2 A schematic diagram illustrating a process for calculating the dielectric loss factor using a steady-state model, provided as an embodiment of this application;

[0077] Figure 3 This is a schematic diagram of a device for predicting electrical risks in oil-paper-insulated equipment, provided as an embodiment of this application. Detailed Implementation

[0078] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0079] The proposed solution can be implemented based on a terminal with data processing capabilities, such as a computer, cloud, or server.

[0080] Next, combined Figure 1The electrical risk prediction method for oil-impregnated paper insulation equipment of this application may include the following steps:

[0081] Step S110: Obtain multiple oil and gas sample data with labels and confidence levels from oil-paper insulation equipment at different time windows.

[0082] The oil and gas sample data can include multiple indicators, which can be measured online by sensors on the oil-impregnated paper insulation equipment. These indicators include the trace water content, oil temperature, and the content of various gases in the insulating oil. The various gases can include... The contents of various gases can be determined by dissolved gas analysis (DGA), and the trace water content can be expressed as the trace water content in mineral oil (mg / kg). The time window can be set to ±30 minutes. Labels can include "normal", "sudden increase", "crossing the window", "inconsistent source", etc.

[0083] Specifically, when acquiring oil and gas sample data, quality control can be performed on the data. When a certain data indicator (such as the content of a certain gas or trace water) is obtained from oil and gas sample data from different sources within the same time window (for example, laboratory test data obtained from manual sampling and testing within the same time window measured by the sensor online), the laboratory value with higher accuracy can be given priority as the central reference to determine the numerical benchmark. At the same time, online monitoring data can be used to supplement the changing trend and continuity of the time series.

[0084] More specifically, step S110, the process of acquiring multiple oil and gas sample data with labels and confidence levels at different time windows for oil-paper insulation equipment, may include:

[0085] S111. Obtain multiple oil and gas sample data of oil-paper insulated equipment under different time windows.

[0086] Each of the oil and gas sample data includes multiple data indicators, including the trace water content, oil temperature, and various gas contents in the insulating oil of the oil-containing paper insulation equipment.

[0087] Specifically, gas, trace water, and temperature data within the same time window can be considered as a single sample. If a channel deviates too far (e.g., more than 1 hour), it can be marked as "crossing the window," and this sample can be excluded from the calculation initially. If the same time window contains both online values ​​and pre-set experimental values, all should be retained for quality control.

[0088] S112. Based on the numerical changes of each data indicator in each of the oil and gas sample data, add labels and confidence levels to the data indicators to obtain oil and gas sample data with labels and confidence levels.

[0089] Specifically, the growth rate of each gas and trace amount of water can be checked. For example, if hydrogen levels jump sharply within a day, acetylene shows continuous growth or a sudden appearance, or trace amounts of water rise significantly in a short period, it should be labeled as a "sudden increase." The data should not be deleted, but the confidence weight of this sample should be reduced when calculations are needed. When both exist within the same window, the laboratory value can be used as the central reference, and the online value can be used to supplement trends and continuity. If the two differ significantly, first check if maintenance and oil changes have been performed recently, if the instrument has been recently calibrated, or if the oil sample has been exposed a second time, etc. If so, it can be labeled as "suspected deviation," and this sample should not be included in the model for the time being. Assign a confidence level to each key quantity (e.g., normal, sudden increase, cross-window, inconsistent source, etc.). Pack the cleaned data into a clean sample package, including: the gas fractions for each window (with units standardized), the original value of the trace water fraction or mg / kg for the window (with temperature and caliber noted), the current temperature; labels (whether there is a sudden increase, whether it crosses the window, whether there is a dual-source conflict, etc.); any subsequent conclusions can be traced back to the original readings, conversion calibers, and cleaning traces through the sample package.

[0090] Step S120: Based on the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment and the oil and gas sample data of each sample, generate a mixture molecular model formula.

[0091] Specifically, using cellulose, the core insulating component of oil-impregnated paper insulation equipment, as the basic framework, and combining labeled and confidence-level oil and gas sample data from various time windows, the mass ratio of each dissolved gas in the oil is extracted as the core parameter. Simultaneously, based on the moisture balance curve of mineral oil and insulating paper, the water content of the mineral oil in the samples is converted to the actual water content in the insulating paper. By fully integrating the characteristics of the equipment's own insulation system with the gas and moisture content information in the samples, a molecular model formula for the paper-water-gas mixture, adapted to the current operating conditions and temperature range of the equipment, is finally reconstructed.

[0092] More specifically, step S120, the process of generating a molecular model formulation for the mixture based on the oil-containing paper insulation equipment and the oil and gas sample data, may include:

[0093] S121. Based on the data of each oil and gas sample, calculate the composition ratio of each dissolved gas fraction to trace water.

[0094] Specifically, for oil and gas sample data with various labels and confidence levels, the dissolved gas and trace water monitoring data in the oil can first be standardized and processed using standardized units. Then, corresponding calculation weights can be assigned based on the sample confidence level, eliminating invalid data interference. Subsequently, the calculations are performed. The proportion of various dissolved gases in the insulating oil is calculated to obtain the fraction of each dissolved gas. At the same time, the specific content of trace water in the oil phase is calculated to finally determine the precise composition ratio of each dissolved gas and trace water.

[0095] S122. Using the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment as the framework, add the composition ratio to generate a mixture molecular model formula.

[0096] Among them, the gas content can be expressed as the mass ratio in the mineral oil, and the water molecule content in the model can be expressed as the water content in the mineral oil converted into the water content in the insulating paper by using the moisture balance curve of the mineral oil and the insulating paper.

[0097] Specifically, the core component of the insulating paper in the oil-containing paper insulation equipment can be used as the basic framework. This component is mainly composed of cellulose. It is precisely matched with the actual insulating paper material characteristics of the equipment. That is, it is based on the cellulose molecular model. Then, the calculated proportions of each dissolved gas and the composition of trace water are precisely integrated into the framework system according to the current temperature range. It fully combines the actual proportion of the medium components under the actual operating conditions of the equipment to finally generate a paper-water-gas mixture molecular model formula that fits the actual operating conditions of the equipment.

[0098] Furthermore, structured objects can be generated based on various data indicators that have been tagged and assigned confidence levels, as well as the molecular model formulation of the mixture.

[0099] The structured object can contain information such as recipe, thermal properties, labels, and time windows. The structured object can be used as input for electrical parameter calculation, thus enabling the construction of a molecular simulation mixture model of the structured object at the current temperature.

[0100] Step S130: Construct a molecular simulation mixture model of the mixture formula at the current temperature, and perform geometric optimization, model annealing and ensemble equilibrium processing on the molecular simulation mixture model to obtain a steady-state model.

[0101] Specifically, firstly, based on the current operating temperature of the oil-paper insulation equipment and the formula of the generated paper-water-gas mixture molecular model, the specific molecular proportions and spatial distributions of cellulose, dissolved gases, and water molecules in the model are determined. Then, a molecular simulation mixture box is built according to the medium density parameters at that temperature. The third generation of condensed-phase optimized molecular potentials for atomistic simulation studies III (COMPASSIII) force field is selected to define the interatomic interaction rules. The Particle Mesh Ewald (PME) algorithm is used to handle long-range electrostatic interactions, restore the molecular composition and interaction characteristics of the equipment insulation system at the current temperature, and construct a molecular simulation mixture model that fits the actual operating conditions of the equipment. Then, for the completed molecular simulation mixture model, geometric optimization is first carried out. By adjusting the spatial arrangement of molecules and atoms, unreasonable steric hindrance and interatomic forces within the model are eliminated, allowing the molecular structure to tend towards the stable state with the lowest energy. Next, model annealing is performed to simulate the gradual temperature change process to overcome local energy barriers, further optimize the molecular configuration, and improve the model's adaptability to actual working conditions. Subsequently, isothermal and isobaric ensemble (NPT ensemble) equilibration of 5 ns and isothermal and isochoric ensemble (NVT ensemble) equilibration of 10 ns are performed sequentially. The pressure, temperature, and volume of the system are controlled to remain constant, and the process is continuously iterated until the key parameters of the model, such as density, energy, and pressure, tend to be stable without significant fluctuations. Finally, a steady-state molecular simulation model with physical properties and structural state that closely match reality is obtained.

[0102] Step S140: Calculate the dielectric loss factor by calculating the total dipole moment of the molecular dynamics system of the steady-state model.

[0103] Specifically, based on the obtained steady-state model, the trajectory of the total dipole moment of the molecular dynamics system over time can be recorded under the NVT ensemble. The normalized dipole autocorrelation function can be calculated through this trajectory, and the real and imaginary parts of the complex dielectric constant can be obtained through Fourier transform. The solution can be completed by combining key parameters such as relaxation strength. Finally, based on the ratio of the real to the imaginary parts of the dielectric constant, the dielectric loss factor of the insulation system of the oil-paper insulation equipment under the current operating conditions can be calculated.

[0104] More specifically, step S140, calculating the dielectric loss factor by calculating the total dipole moment of the molecular dynamics system in the steady-state model, is as follows: Figure 2 As shown, it includes:

[0105] Step S141: Calculate the total dipole moment of the molecular dynamics system of the steady-state model under isothermal and isochoric ensemble.

[0106] Here, the total dipole moment of the molecular dynamics system represents the trajectory of the total dipole moment of the system obtained by molecular dynamics under given temperature and volume conditions, with dimensions of coulomb-meter (C·m). The volume here represents the model volume established by the molecular dynamics simulation system.

[0107] Step S142: Construct a normalized dipole autocorrelation function based on the total dipole moment of the molecular dynamics system, and perform a Fourier transform on the normalized dipole autocorrelation function to obtain the losses of the polarizable energy storage and polarization hysteresis conduction mechanisms.

[0108] The normalized dipole autocorrelation function can be:

[0109]

[0110] in, Let be the trajectory of the total dipole moment of the molecular dynamics system over time. This represents the average value of the dipole moment at all moments during the simulation of the steady-state model. The total dipole moment of the system at the initial moment of the steady-state model simulation is given by... For statistical calculations, it can represent the statistical average over time or an ensemble. for and The Euclidean inner product is used to calculate the correlation between dipole vectors at different times.

[0111] Understandably, the normalized dipole autocorrelation function is used to represent the correlation between the initial dipole state and the dipole state at the lag time. Normalization makes... When it is necessary to When parameterizing the decay shape, the stretching exponential function is commonly used. Perform fitting, where A dimensionless prefactor, The characteristic duration (in seconds). The attenuation coefficient is 0 < ≤1 indicates the degree of broadening of the relaxed distribution.

[0112] Furthermore, the process of performing a Fourier transform on the normalized dipole autocorrelation function to obtain the polarizable energy storage and polarization hysteresis conduction mechanism losses may include:

[0113] The losses of polarizable energy storage and polarization-hysteresis conduction mechanisms can be calculated using the following formula:

[0114]

[0115] in, For polarizable energy storage, The loss is due to the polarization hysteresis conduction mechanism. The dielectric constant of optical frequency, For relaxation strength, Angular frequency, The equivalent DC conductivity of the material in the oil-impregnated paper insulation equipment is given. For the volume of the molecular dynamics system, The vacuum permittivity, Boltzmann's constant, This is the current thermodynamic temperature.

[0116] Understandably, the complex permittivity can be written as... ,in The real part indicates polarizable energy storage. The imaginary part represents the loss due to polarization hysteresis and the conduction mechanism, and the angular frequency. with line frequency The relationship is .

[0117] Step S143: Calculate the dielectric loss factor based on the polarizable energy storage and the polarization hysteresis conduction mechanism loss.

[0118] Specifically, the loss of the polarization hysteresis conductivity mechanism can be... Divided by the aforementioned polarizable energy storage The dielectric loss factor is obtained.

[0119] Among them, the dielectric loss factor is a dimensionless quantity that reflects the relative ratio of work loss to energy storage under a unit electric field strength; the frequency is taken as the power frequency and is compared under the same temperature and frequency range as the standard limit.

[0120] It should be noted that if the low-frequency leakage conductive path is considered separately, it will be treated as... Incorporating the form The conductive portion. In the case of discussing only the molecular-scale polarization mechanism, The polarization loss component refers to the dielectric loss factor used for benchmarking and classification when conducting power frequency assessments at engineering standards. The conductivity and polarization terms are then properly aligned and spliced ​​together at the same temperature and frequency to obtain the dielectric loss factor.

[0121] Step S150: Divide the dielectric loss factor by the specified limit to obtain the risk quantification value, and determine the electrical risk level of the oil-paper insulation equipment through the risk quantification value.

[0122] The specified limit can be the same as the specified value obtained under the same temperature and frequency when calculating the dielectric loss factor. Therefore, if the limit is given at other temperatures or frequencies, a simple mapping to the current temperature and power frequency must be performed first.

[0123] More specifically, step S150, dividing the dielectric loss factor by the specified limit to obtain a risk quantification value, and determining the electrical risk level of the oil-paper insulation equipment using the risk quantification value, may include the following situations:

[0124] The first method is to determine that the electrical risk level of the oil-paper insulation equipment is normal when the risk quantification value is less than the first quantification threshold.

[0125] For example, assuming the first quantification threshold is 0.6, when the risk quantification value U < 0.6, the electrical risk level of the oil-paper insulation equipment can be determined to be normal, and a green indicator light will be displayed.

[0126] The second approach is to determine the electrical risk level of the oil-paper insulation equipment as requiring attention when the risk quantification value is not less than the first quantification threshold and less than the second quantification threshold.

[0127] For example, assuming the first quantification threshold is 0.6 and the second quantification threshold is 0.9, when the risk quantification value 0.6 ≤ U < 0.9, the electrical risk level of the oil-paper insulation equipment can be determined to be of concern, and a yellow indicator light will be displayed.

[0128] The third method is to determine the electrical risk level of the oil-paper insulation equipment as an alarm when the risk quantification value is not less than the second quantification threshold and less than the third quantification threshold.

[0129] For example, assuming the second quantification threshold is 0.9 and the third quantification threshold is 1.0, when the risk quantification value 0.9 ≤ U < 1.0, the electrical risk level of the oil-paper insulation equipment can be determined to be an alarm, and an orange indicator light will be displayed.

[0130] Fourth, when the risk quantification value is not less than the third quantification threshold, the electrical risk level of the oil-paper insulation equipment is determined to be critical.

[0131] For example, assuming the third quantification threshold is 1.0, when the risk quantification value U≥1.0, the electrical risk level of the oil-paper insulation equipment can be determined to be critical, and a red indicator light will be displayed.

[0132] Furthermore, the terminal can ultimately output: the current level and limit U, the remaining margin 1-U (a negative value indicates that the limit has been exceeded), key trigger statements (such as the media loss factor limit xx%, or the media loss factor exceeding the limit xx%) and corresponding brief suggestions (green: routine monitoring; yellow: encrypted retesting and verification of the caliber; orange: load limiting or oil treatment and shortening the retesting interval; red: shutdown or emergency response).

[0133] The electrical risk prediction method for oil-paper insulated equipment provided in this embodiment acquires multiple oil and gas sample data with labels and confidence levels at different time windows. Based on the cellulose molecular model of the insulating paper of the oil-paper insulated equipment and each oil and gas sample data, a mixture molecular model formula is generated. A molecular simulation mixture model of the mixture molecular model formula at the current temperature is constructed. The molecular simulation mixture model is then subjected to geometric optimization, model annealing, and ensemble balancing to obtain a steady-state model. By calculating the total dipole moment of the molecular dynamics system of the steady-state model, the dielectric loss factor is calculated. The dielectric loss factor is divided by the specification limit to obtain the risk quantification value. The degree of electrical risk of the oil-paper insulated equipment is determined by the risk quantification value. Therefore, this method provides a mechanistic and quantitative mapping from monitoring data to electrical risk, accurately calculates real-time dielectric parameters, takes into account individual equipment differences, quantifies risk values, and classifies them, improving the accuracy of early degradation identification and thus enhancing the accuracy of electrical risk monitoring for oil-paper insulated equipment.

[0134] The apparatus for predicting electrical risks of oil-paper-insulated equipment provided in the embodiments of this application will be described below. The apparatus for predicting electrical risks of oil-paper-insulated equipment described below can be referred to in correspondence with the method for predicting electrical risks of oil-paper-insulated equipment described above.

[0135] See Figure 3 , Figure 3 This is a schematic diagram of a device for predicting electrical risks in oil-paper-insulated equipment, as disclosed in an embodiment of this application.

[0136] like Figure 3 As shown, the device may include:

[0137] The oil and gas sample data acquisition unit 10 is used to acquire multiple oil and gas sample data with labels and confidence levels under different time windows for oil-containing paper insulation equipment.

[0138] The mixture molecular model formulation generation unit 20 is used to generate a mixture molecular model formulation based on the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment and the oil and gas sample data of each of the oil and gas samples.

[0139] The model building unit 30 is used to build a molecular simulation mixture model of the mixture molecular model formulation at the current temperature, and to perform geometric optimization, model annealing and ensemble equilibrium processing on the molecular simulation mixture model to obtain a steady-state model.

[0140] The dielectric loss factor calculation unit 40 is used to calculate the dielectric loss factor by calculating the total dipole moment of the molecular dynamics system of the steady-state model.

[0141] The electrical risk level quantification unit 50 is used to divide the dielectric loss factor by the specification limit to obtain the risk quantification value, and to determine the electrical risk level of the oil-paper insulation equipment through the risk quantification value.

[0142] Optionally, the oil and gas sample data acquisition unit includes:

[0143] The sample acquisition unit is used to acquire multiple oil and gas sample data of the oil-paper insulation equipment under different time windows. Each oil and gas sample data includes multiple data indicators, including the trace water content, oil temperature and various gas contents in the insulating oil of the oil-paper insulation equipment.

[0144] The label confidence level addition unit is used to add labels and confidence levels to each data indicator based on the numerical changes of each data indicator in each of the oil and gas sample data, so as to obtain oil and gas sample data with labels and confidence levels.

[0145] Optionally, the device may also include:

[0146] The structured object generation unit is used to generate structured objects based on various data indicators with added tags and confidence levels, as well as the molecular model formula of the mixture.

[0147] A molecular simulation mixture model building unit is used to build a molecular simulation mixture model of the structured object at the current temperature.

[0148] Optionally, the mixture molecular model formulation generation unit includes:

[0149] The composition ratio calculation unit is used to calculate the composition ratio of each dissolved gas fraction and trace water based on the oil and gas sample data mentioned above.

[0150] The formulation generation unit is used to generate a mixture molecular model formulation by adding the composition proportions based on the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment.

[0151] Optionally, the dielectric loss factor calculation unit includes:

[0152] The total dipole moment calculation unit for molecular dynamics systems is used to calculate the total dipole moment of the molecular dynamics system under the steady-state model in an isothermal and isochoric ensemble.

[0153] The Fourier transform unit is used to construct a normalized dipole autocorrelation function based on the total dipole moment of the molecular dynamics system, and to perform a Fourier transform on the normalized dipole autocorrelation function to obtain the losses of the polarizable energy storage and polarization hysteresis conduction mechanisms.

[0154] The dielectric loss factor determination unit is used to calculate the dielectric loss factor based on the polarizable energy storage and the polarization hysteresis conduction mechanism loss.

[0155] Optionally, the normalized dipole autocorrelation function is:

[0156]

[0157] in, Let be the trajectory of the total dipole moment of the molecular dynamics system over time. This represents the average value of the dipole moment at all moments during the simulation of the steady-state model. The total dipole moment of the system at the initial moment of the steady-state model simulation is given by... For average statistical calculation, for and The Euclidean inner product.

[0158] Optionally, the Fourier transform unit includes:

[0159] The first formula calculation unit is used to calculate the losses of polarizable energy storage and polarization-hysteresis conduction mechanisms using the following formula:

[0160]

[0161] in, For polarizable energy storage, The loss is due to the polarization hysteresis conduction mechanism. The dielectric constant of optical frequency, For relaxation strength, Angular frequency, The equivalent DC conductivity of the material in the oil-impregnated paper insulation equipment is given. For the volume of the molecular dynamics system, The vacuum permittivity, Boltzmann's constant, This is the current thermodynamic temperature.

[0162] Optionally, the dielectric loss factor determination unit includes:

[0163] Divide the polarization hysteresis conduction mechanism loss by the polarizable energy storage to obtain the dielectric loss factor.

[0164] Optionally, the electrical risk quantification unit includes:

[0165] The normality quantification unit is used to determine that the electrical risk level of the oil-paper insulation equipment is normal when the risk quantification value is less than the first quantification threshold.

[0166] The attention level quantification unit is used to determine the electrical risk level of the oil-paper insulation equipment as requiring attention when the risk quantification value is not less than the first quantification threshold and less than the second quantification threshold.

[0167] An alarm level quantification unit is used to determine the electrical risk level of the oil-paper insulation equipment as an alarm when the risk quantification value is not less than the second quantification threshold and less than the third quantification threshold.

[0168] The urgency quantification unit is used to determine the electrical risk level of the oil-paper insulation equipment as critical when the risk quantification value is not less than the third quantification threshold.

[0169] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0170] The various embodiments in this specification are described in a progressive manner. Each embodiment focuses on the differences from other embodiments. The various embodiments can be combined as needed, and the same or similar parts can be referred to each other.

[0171] The above description of the disclosed embodiments enables those skilled in the art to make or use this application. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A method for predicting electrical risks in oil-paper-insulated equipment, characterized in that, include: Obtain multiple oil and gas sample data with labels and confidence levels from oil-impregnated paper insulation equipment at different time windows; Based on the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment and the data of each oil and gas sample, a mixture molecular model formula is generated. A molecular simulation model of the mixture formulation at the current temperature is constructed, and the molecular simulation model is subjected to geometric optimization, model annealing and ensemble equilibrium processing to obtain a steady-state model. The dielectric loss factor is calculated by calculating the total dipole moment of the molecular dynamics system in the steady-state model. Divide the dielectric loss factor by the specified limit to obtain the risk quantification value, and determine the electrical risk level of the oil-paper insulation equipment through the risk quantification value.

2. The method according to claim 1, characterized in that, The acquisition of multiple oil and gas sample data with labels and confidence levels at different time windows for oil-containing paper insulation equipment includes: Multiple oil and gas sample data of the oil-paper insulation equipment under different time windows are obtained. Each oil and gas sample data includes multiple data indicators, including the trace water content, oil temperature and various gas contents in the insulating oil of the oil-paper insulation equipment. Based on the numerical changes of each data indicator in each of the oil and gas sample data, a label and confidence level are added to the data indicator to obtain oil and gas sample data with labels and confidence levels.

3. The method according to claim 2, characterized in that, The method further includes: Based on the data indicators with added tags and confidence levels, and the molecular model formulation of the mixture, a structured object is generated; Construct a molecular simulation mixture model of the structured object at the current temperature.

4. The method according to claim 1, characterized in that, Based on the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment and the oil and gas sample data of each item, a mixture molecular model formulation is generated, including: Based on the oil and gas sample data described in each section, calculate the composition ratio of each dissolved gas fraction and trace water. Using the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment as a framework, the aforementioned component proportions are added to generate a mixture molecular model formula.

5. The method according to claim 1, characterized in that, The calculation of the total dipole moment of the molecular dynamics system in the steady-state model and the calculation of the dielectric loss factor include: Calculate the total dipole moment of the molecular dynamics system of the steady-state model under isothermal and isochoric ensemble; Based on the total dipole moment of the molecular dynamics system, a normalized dipole autocorrelation function is constructed, and a Fourier transform is performed on the normalized dipole autocorrelation function to obtain the losses of the polarizable energy storage and polarization hysteresis conduction mechanisms. The dielectric loss factor is calculated based on the polarizable energy storage and the polarization hysteresis conduction mechanism loss.

6. The method according to claim 5, characterized in that, The normalized dipole autocorrelation function is: in, Let be the trajectory of the total dipole moment of the molecular dynamics system over time. This represents the average value of the dipole moment at all moments during the simulation of the steady-state model. The total dipole moment of the system at the initial moment of the steady-state model simulation is given by... For average statistical calculation, for and The Euclidean inner product.

7. The method according to claim 6, characterized in that, Performing a Fourier transform on the normalized dipole autocorrelation function yields the losses from polarizable energy storage and polarization-hysteresis conduction mechanisms, including: The losses of polarizable energy storage and polarization-hysteresis conduction mechanisms can be calculated using the following formula: in, For polarizable energy storage, The loss is due to the polarization hysteresis conduction mechanism. The dielectric constant of optical frequency, For relaxation strength, Angular frequency, The equivalent DC conductivity of the material in the oil-impregnated paper insulation equipment is given. For the volume of the molecular dynamics system, The vacuum permittivity, Boltzmann's constant, This is the current thermodynamic temperature.

8. The method according to claim 5, characterized in that, Based on the polarizable energy storage and the polarization hysteresis conductivity mechanism losses, the dielectric loss factor is calculated, including: Divide the polarization hysteresis conduction mechanism loss by the polarizable energy storage to obtain the dielectric loss factor.

9. The method according to any one of claims 1-8, characterized in that, Determining the electrical risk level of the oil-paper-insulated equipment using the aforementioned risk quantification value includes: When the risk quantification value is less than the first quantification threshold, the electrical risk level of the oil-paper insulation equipment is determined to be normal. When the risk quantification value is not less than the first quantification threshold and is less than the second quantification threshold, the electrical risk level of the oil-paper insulation equipment is determined to be of concern. When the risk quantification value is not less than the second quantification threshold and less than the third quantification threshold, the electrical risk level of the oil-paper insulation equipment is determined to be an alarm. When the risk quantification value is not less than the third quantification threshold, the electrical risk level of the oil-paper insulation equipment is determined to be critical.

10. An electrical risk prediction device for oil-paper-insulated equipment, characterized in that, include: The oil and gas sample data acquisition unit is used to acquire multiple oil and gas sample data with labels and confidence levels at different time windows for oil-containing paper insulation equipment. The mixture molecular model formulation generation unit is used to generate a mixture molecular model formulation based on the cellulose molecular model of the insulating paper of the oil-containing paper insulation equipment and the oil and gas sample data of each of the oil and gas samples. The model building unit is used to build a molecular simulation mixture model of the mixture molecular model formulation at the current temperature, and to perform geometric optimization, model annealing and ensemble equilibrium processing on the molecular simulation mixture model to obtain a steady-state model; The dielectric loss factor calculation unit is used to calculate the dielectric loss factor by calculating the total dipole moment of the molecular dynamics system of the steady-state model. The electrical risk level quantification unit is used to divide the dielectric loss factor by the specification limit to obtain the risk quantification value, and to determine the electrical risk level of the oil-paper insulation equipment through the risk quantification value.

Images