Dry-type transformer energy consumption comprehensive monitoring system based on multi-mode sensing

Abstract

The invention discloses a dry-type transformer energy consumption comprehensive monitoring system based on multi-modal sensing, and relates to the technical field of dry-type transformers. A multi-modal sensing subsystem deploys multi-modal sensing equipment in a dry-type transformer in advance and collects the operation state of the dry-type transformer to generate a characteristic value vector set; the local structure analysis subsystem is used for constructing a frequency-energy two-dimensional response map, judging triggering of a mechanical response mechanism and identifying a loose hot area; the disturbance analysis subsystem is used for identifying jump points where sudden jump occurs based on the loose hot area, screening out the jump points, deviated due to structural disturbance driving, of the saturation area according to the positions of the jump points, and obtaining a drift area; and the dissipation coupling subsystem combines the drift region and the loose hot region, identifies the coupling excitation region and combines the eigenvalue vector set to evaluate the influence degree of the energy consumption in the coupling excitation region on the total energy consumption in the dry-type transformer so as to realize the execution of a dynamic structure adjustment mechanism.

Description

Technical Field

[0001] The present invention relates to the technical field of dry-type transformers, and in particular to an integrated monitoring system for the energy consumption of dry-type transformers based on multi-modal sensing. Background Art

[0002] As a key device in the power distribution link, dry-type transformers are widely used in scenarios with high requirements for safety and stability, such as urban power distribution networks, data centers, and rail transit. During their operation, they are affected by the interweaving of multiple physical quantities such as structural vibration, electromagnetic disturbance, and thermal coupling. If there are internal structural looseness or magnetic flux path disturbances, it is extremely easy to cause local energy consumption anomalies, resulting in a decrease in overall efficiency or the accumulation of potential faults.

[0003] In the actual operation of dry-type transformers, the looseness of the iron core is often caused by mechanical vibration or aging of the fixed clamping, and is common in urban distribution substations, photovoltaic grid-connected equipment, or industrial automation control systems. This kind of structural deterioration does not immediately affect the overall power supply function in the early stage, but will cause a series of non-linear cross anomalies:

[0004] On the one hand, the instability of the iron core will lead to an increase in the release of local vibration energy, forming structural vibration coupling energy consumption;

[0005] On the other hand, such disturbances will cause the originally symmetrically closed magnetic flux channels to reorganize, forming new high-density magnetic aggregation regions, that is, the phenomenon of magnetic saturation region migration;

[0006] This kind of problem has long-term latency and multi-dimensional disturbance. It is manifested as the structural deviation of the iron core after being loaded and vibrated, the reorganization of the originally uniform and closed magnetic flux distribution, and the increase in the magnetic density in local areas. Existing monitoring methods mostly rely on current, voltage, and hot spot temperature parameters, which are difficult to accurately reflect the variation of the magnetic flux path and the change of local iron loss, and ignore the coupling identification of energy consumption anomalies caused by structural micro-changes. Summary of the Invention

[0007] In view of the above problems existing in the prior art, the present application provides an integrated monitoring system for the energy consumption of dry-type transformers based on multi-modal sensing.

[0008] The present disclosure provides an integrated monitoring system for the energy consumption of dry-type transformers based on multi-modal sensing, including,

[0009] A multi-modal perception subsystem, which pre-deploys multi-modal sensing devices inside the dry-type transformer to collect the operating state of the dry-type transformer to generate a set of eigenvalue vectors;

[0010] A local structure analysis subsystem, which constructs a two-dimensional frequency-energy response map, judges the triggering of the mechanical response mechanism, and identifies the loosening hot spot;

[0011] The perturbation analysis subsystem, based on the loose hot zone, identifies the jump points where mutations occur, and according to the positions of the jump points, filters out the jump points where the saturation zone is shifted by structural perturbations to obtain the drift region;

[0012] The dissipative coupling subsystem combines the drift region and the loose hot zone to identify the coupled excitation zone, and combines the eigenvalue vector set to evaluate the influence degree of the energy consumption in the coupled excitation zone on the total energy consumption in the dry-type transformer, so as to implement the execution of the dynamic structure adjustment mechanism.

[0013] Optionally, the multimodal perception subsystem includes a deployment unit and a collection unit;

[0014] The deployment unit is used to deploy multimodal sensing devices inside the dry-type transformer, including electromagnetic field density probes, three-dimensional vibration sensor arrays and thermal imaging probes, and integrate a unified clock reference module in each modal sensing device to achieve the timing alignment of data acquisition;

[0015] The collection unit is used to use the multimodal sensing devices to collect the vibration acceleration signals, magnetic flux density, structural vibration energy density and magnetic flux perturbation energy density at each position in real time, and generate an eigenvalue vector set after data preprocessing.

[0016] Optionally, the local structure analysis subsystem includes a spectrum generation unit, a mechanism trigger unit and an identification unit;

[0017] The spectrum generation unit collects the vibration acceleration signals in real time by the three-dimensional vibration sensor array to form a time-domain vibration data stream, performs signal preprocessing on the time-domain vibration data stream, performs time-segment window processing on the preprocessed time-domain vibration data stream, applies a window function to each time segment for weighting, and performs a fast Fourier transform on the time-domain vibration data stream within each windowed time segment to obtain the spectral results of multiple time periods. By splicing the spectral results of multiple time periods on the time axis, a frequency-energy two-dimensional response spectrum is generated;

[0018] The mechanism trigger unit extracts the modal resonance peaks under each frequency band condition from the frequency-energy two-dimensional response spectrum, compares them with the initial baseline modal frequencies, calculates and obtains the modal shift ratio, sets a continuous determination window. If the modal shift ratio exceeds the preset shift threshold within the continuous determination window, it indicates that the modal shift is abnormal. At this time, the mechanical response mechanism is triggered;

[0019] The identification unit receives the mechanical response mechanism, constructs a structural looseness index under different time segment conditions by weighted summation of the vibration energy concentration degree and the modal shift ratio, and performs a stabilized judgment through the time-domain averaging method to obtain a stabilized result. The stabilized result is compared with the preset looseness threshold. If the stabilized result exceeds the looseness threshold, the corresponding position is taken as the loose hot zone, and a spatial thermal map is generated according to the loose hot zone.

[0020] Optionally, the perturbation analysis subsystem includes a zone unit, a jump unit, a perturbation unit, and a trajectory unit;

[0021] The zone unit takes the loosening hot zone as the center and expands a grid range of one circle outward as the boundary zone, and combines the boundary zone with the loosening hot zone to generate a spatio-temporal field coverage zone;

[0022] The jump unit constructs a local magnetic flux density spatio-temporal field within the spatio-temporal field coverage zone and uses the slope mutation discrimination method to identify the instantaneous jump position. The specific content is as follows:

[0023] According to the local magnetic flux density spatio-temporal field, a time series is extracted at each position to calculate the first-order time derivative of the magnetic flux density;

[0024] Extract the maximum first-order time derivative of the magnetic flux density from the time series. If the maximum first-order time derivative of the magnetic flux density exceeds the mean value of the maximum first-order time derivative of the magnetic flux density at the corresponding position within the historical period, it is determined that a mutation jump occurs at the corresponding position and is marked as a jump point;

[0025] According to the jump points, record the jump point information, including time, position, and magnetic flux density.

[0026] Optionally, the perturbation unit presets a magnetic flux density saturation critical threshold. If the magnetic flux density at the jump point exceeds the magnetic flux density saturation critical threshold and is within the loosening hot zone, it is determined that the saturation zone is shifted due to structural perturbation drive; if the magnetic flux density at the jump point exceeds the magnetic flux density saturation critical threshold but is not within the loosening hot zone, it is determined that the generation of the saturation zone is a non-mechanical drive source;

[0027] The trajectory unit determines the migration path of the saturation zone in space over time according to the jump points where the saturation zone is shifted due to structural perturbation drive, and obtains the drift trajectory vector.

[0028] Optionally, the dissipation coupling subsystem includes a coupling analysis unit, a quantitative analysis unit, and a comparison and optimization unit;

[0029] The coupling analysis unit determines the drift region according to the drift trajectory vector, and superimposes the drift region on the spatial heat map to obtain the overlap degree. The overlap degree is used to reflect whether there is a high degree of consistency between the structural vibration excitation and the magnetic flux anomaly. The specific content is as follows: , where, is the overlap degree, is the drift region, is the part of the spatial heat map where there is a loosening hot zone, is the spatially overlapping region, is the spatially union region; if the overlap degree exceeds the preset overlap threshold, the spatially overlapping region is marked as the coupling excitation zone.

[0030] Optionally, the quantitative analysis unit performs feature extraction on the eigenvalue vector set to obtain the structural vibration energy density and magnetic flux disturbance energy density at different positions in the dry-type transformer. Combining with the coupling excitation region, it analyzes the energy consumption behavior caused by structural vibration excitation in the coupling excitation region. By using the weighted summation algorithm for the structural vibration energy density and magnetic flux disturbance energy density in the coupling excitation region, it evaluates the unit-time coupling energy consumption of the coupling excitation region in the current dry-type transformer, specifically as follows: , where is the unit-time coupling energy consumption of the coupling excitation region, is the position coordinate point in the coupling excitation region, is the coupling excitation region, is the unit-time vibration energy density of the structural vibration at the coordinate point , is the magnetic energy dissipation density of the magnetic flux disturbance at the coordinate point , and are both weight values;

[0031] Optionally, the comparison and optimization unit compares the unit-time coupling energy consumption of the coupling excitation region with the total unit-time energy consumption of the dry-type transformer to calculate the unit-time energy consumption ratio. The unit-time energy consumption ratio is used to reflect the influence degree of the energy consumption in the coupling excitation region on the total energy consumption in the dry-type transformer. If the unit-time energy consumption ratio exceeds the pre-set ratio threshold, it is determined that the coupling excitation region at the corresponding time point enters the abnormal dissipation critical state, and the coupling excitation region is marked as the abnormal state of structural looseness driving magnetic flux drift, and transmitted to the background operation platform to execute the structural adjustment mechanism for the coupling excitation region. The structural adjustment mechanism includes the reinforcement suggestion for the iron core support point and the local reapplication of the structural bonding material.

[0032] The beneficial effects of the present invention:

[0033] (1)By generating a two-dimensional frequency-energy response spectrum through the local structure analysis subsystem and combining modal response offset detection, the structural excitation characteristic region (i.e., the loosening hot zone) caused by core loosening can be effectively identified, further enhancing the ability to recognize the hidden degradation behavior of early mechanical structures. The perturbation analysis subsystem further identifies the flux density jump points based on the loosening hot zone, comprehensively determines whether the magnetic flux is in a saturated state and whether it is driven by structural perturbations, so as to accurately extract the drift trajectory of the saturated zone and obtain the drift region, realizing the causal association modeling between magnetic flux behavior and structural variation. Through the dissipation coupling subsystem, spatial overlap analysis is carried out between the structural vibration excitation zone and the magnetic flux drift zone to construct a coupling excitation zone, and the energy ratio between the coupling energy consumption per unit time in this zone and the total energy consumption of the system is evaluated in combination with the eigenvalue vector set. If the energy consumption ratio is too high, it can be determined that this zone has entered the abnormal dissipation critical state and trigger a dynamic structure adjustment mechanism suggestion. Compared with the traditional single-modal monitoring method, this system can realize the joint monitoring and linkage feedback judgment of the three dimensions of structure, magnetic flux, and energy consumption, with high hidden danger identification sensitivity, strong path tracing ability, and high energy consumption abnormal source localization accuracy, thus improving the energy efficiency management and reliability guarantee level during the operation of dry-type transformers.

[0034] (2)The perturbation analysis subsystem realizes the causal coupling diagnosis ability between the magnetic flux behavior and the structural state of dry-type transformers by constructing a magnetic flux perturbation analysis mechanism centered on structural anomalies. The zone unit expands one grid outward based on the identified loosening hot zone to construct a spatio-temporal field coverage area, effectively covering the magnetic flux distribution range that may be affected by structural perturbations, providing an accurate spatial boundary for magnetic flux behavior analysis. The jump unit constructs a magnetic flux density spatio-temporal field based on the local magnetic flux density signal within this spatio-temporal field coverage area, extracts the time series at each position and calculates the first derivative, and uses the slope mutation discrimination method to identify the maximum derivative point, which is determined as the instantaneous magnetic flux jump position. This method can identify early magnetic flux dynamic abnormal signals. By combining whether the jump point is within the loosening hot zone, the distinction between the magnetic flux saturation behavior driven by structural perturbations and the magnetic flux perturbations from non-structural sources is realized, so as to effectively judge whether the current magnetic flux saturation phenomenon is induced by core structure loosening, improving the accuracy of fault attribution. This mechanism can not only avoid false alarms but also enhance the analysis coherence of the structure-magnetic flux physical chain. The trajectory unit further extracts the spatial coordinates of the saturated zone changing with time based on the time series of the above jump points to form a drift trajectory vector, which is used to track the dynamic migration path of the saturated zone.

[0035] (3) Through the quantitative calculation of the overlap degree, the present invention determines the coupling degree between the structural excitation and the magnetic flux disturbance in the physical space. This analysis process enhances the physical logic support for fault mechanism identification, can effectively screen out the coupling excitation areas where the structure-magnetic flux behavior is highly consistent, and provides a clear regional basis for subsequent energy consumption analysis. The quantitative analysis unit further extracts the structural vibration energy density and the magnetic flux disturbance energy density of each coordinate point in the coupling excitation area from the eigenvalue vector set, and constructs a coupling energy consumption index per unit time through weighted summation, enabling the system to evaluate the specific contribution of local coupling dissipation inside the dry-type transformer to the overall energy consumption based on the multi-modal physical response, and being able to quantitatively judge the composite energy consumption process caused by the interaction between structural fatigue and magnetic flux saturation in a certain area. Description of the Drawings

[0036] To more clearly illustrate the technical solutions in the present application or the prior art, the following will briefly introduce the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings in the following description are some embodiments of the present application.

[0037] Figure 1 It is a block diagram of a comprehensive monitoring system for the energy consumption of a dry-type transformer based on multi-modal sensing according to the present invention. Detailed Embodiments

[0038] To make the objectives, technical solutions, and advantages of the present application clearer, the following will clearly and completely describe the technical solutions in the present application in conjunction with the drawings in the present application. Obviously, the described embodiments are some, but not all, of the embodiments of the present application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present application without creative efforts shall fall within the scope of protection of the present application.

[0039] In the scenario of photovoltaic inverter grid connection, the dry-type transformer often has frequent fluctuations in current input, causing the magnetic flux density of the iron core to break through the saturation threshold within a short time, inducing abnormal eddy currents and increasing energy consumption. Although the surface temperature rise is not significant, local fatigue or insulation degradation of the iron core may still occur in the long term. Existing systems mostly rely on single current or temperature signals and lack the perception and identification of the hidden energy consumption mechanism of saturation area drift. Therefore, the present invention proposes a comprehensive monitoring system for the energy consumption of a dry-type transformer based on multi-modal sensing.

[0040] Embodiment 1

[0041] As Figure 1 shown, the embodiment of the present invention proposes a comprehensive monitoring system for the energy consumption of a dry-type transformer based on multi-modal sensing, including:

[0042] A multi-modal perception subsystem, which pre-deploys multi-modal sensing devices inside the dry-type transformer to collect the operating state of the dry-type transformer to generate an eigenvalue vector set;

[0043] The local structure analysis subsystem constructs a two-dimensional frequency-energy response map, determines the triggering of the mechanical response mechanism, and identifies the loose hot zone;

[0044] The perturbation analysis subsystem, based on the loose hot zone, identifies the jump points where mutations occur, and according to the positions of the jump points, filters out the jump points where the saturation zone is shifted by structural perturbations, and obtains the drift region;

[0045] The dissipation coupling subsystem combines the drift region and the loose hot zone to identify the coupled excitation zone, and combines the eigenvalue vector set to evaluate the influence degree of the energy consumption in the coupled excitation zone on the total energy consumption in the dry-type transformer, and realizes the execution of the dynamic structure adjustment mechanism.

[0046] In this embodiment, by deploying multi-source heterogeneous sensing devices in the dry-type transformer and combining modal information such as structure and magnetic flux, the joint sensing and analysis of micro-perturbed structure changes, magnetic flux perturbations, and local energy consumption anomalies during operation are realized. Compared with the traditional single-parameter monitoring method, this system can identify latent abnormal operating states such as latent structure loosening, saturation zone shift, and energy consumption drift in the non-fault state, and through the analysis of the energy consumption contribution ratio, judge whether this area has a substantial impact on the system operation efficiency, so as to trigger the structure adjustment mechanism, such as strengthening the support points or locally reapplying the binder, to achieve the systematic optimization goals of early diagnosis, precise intervention, and delayed degradation.

[0047] For example, in a dry-type transformer in the power supply system of a certain data center, the transformer operates in a high-load fluctuation scenario. Although the traditional monitoring means can provide temperature rise alarms, during a night server group switching process, this system detected a slight shift in the modal frequency of a certain area of the iron core through a three-dimensional vibration sensor array and a magnetic flux density probe, and then the magnetic flux density had a slope jump hour by hour at that place. The perturbation analysis subsystem further identified that the magnetic flux value at the jump point in this area had reached the saturation critical point and was accompanied by spatial drift. The dissipation coupling subsystem superimposed this drift trajectory and the hot spot map, and confirmed that this point was the coupled excitation zone, and the proportion of the coupled energy consumption per unit time rose to 34.2% of the total system power fluctuation. The system immediately marked it as an abnormal state of loose-driven magnetic flux drift and fed back the structure adjustment suggestion to the operation and maintenance platform. During this process, the equipment did not enter the alarm state, but the system identified the trend of energy consumption anomalies in advance, avoiding potential insulation deterioration and energy efficiency waste, and further improving the equipment stability and the safety margin of the life cycle.

[0048] Among them, when the short-time magnetic flux density exceeds the saturation point of the iron core, abnormal eddy currents will be generated, but there is no obvious temperature rise on the surface. Therefore, the traditional method of using temperature rise alarms may have the phenomenon of untimely detection of abnormalities.

[0049] By constructing a multi-modal dry-type transformer monitoring system integrating perception, reasoning, and feedback, it is possible to dynamically identify the problem of flux saturation area drift, visualize the hidden losses with "no surface temperature rise" but "increased internal energy consumption", achieve real-time dynamic risk assessment and adaptive optimization feedback, and intervene in advance to adjust control means to prevent local overheating and insulation performance degradation.

[0050] Embodiment 2

[0051] Please refer to Figure 1 , specifically: The multi-modal perception subsystem includes a deployment unit and a collection unit;

[0052] The deployment unit is used to deploy multi-modal sensing devices inside the dry-type transformer, including electromagnetic field density probes, three-dimensional vibration sensor arrays, and thermal imaging probes, and integrate a unified clock reference module in each modal sensing device to achieve timing alignment of data acquisition;

[0053] The clock reference module is used to provide high-precision time synchronization signals for perception devices such as electromagnetic field density probes and three-dimensional vibration sensor arrays. Through the unified clock reference provided by this module, each sensor can trigger data acquisition at the same sampling moment and generate data frames with global timestamps, ensuring that multi-modal perception information is comparable and fusible in the same time domain, and laying a data foundation for subsequent modal collaborative analysis and energy consumption behavior modeling.

[0054] The electromagnetic field density probes will be installed at the intersection areas of multiple flux conduction paths outside the core structure of the dry-type transformer to collect local instantaneous flux density values and their spatial gradient distribution characteristics; the three-dimensional vibration sensor array will deploy high-sensitivity acceleration sensors along the vibration conduction sensitive areas such as the connection parts between the core frame and the fixing parts to collect multi-axial structural vibration response signals, and then extract indicators such as modal frequency, amplitude response, and energy distribution;

[0055] The collection unit is used to use multi-modal sensing devices to collect vibration acceleration signals, flux density, structural vibration energy density, and flux perturbation energy density at each position in real time. After data preprocessing, a set of eigenvalue vectors is generated.

[0056] Data preprocessing includes noise removal, missing value filling, and data smoothing operations. Among them, the methods for filling missing values include mean filling, median filling, interpolation filling, and regression filling;

[0057] The local structure analysis subsystem includes a graph generation unit, a mechanism trigger unit, and an identification unit;

[0058] Spectrum generation unit: Vibration acceleration signals are collected in real time by a three-dimensional vibration sensor array to form a time-domain vibration data stream. The time-domain vibration data stream is subjected to signal preprocessing. A time segmentation window processing is performed on the preprocessed time-domain vibration data stream. Each time segment is weighted by a window function (such as a Hamming window), and a fast Fourier transform is performed on the time-domain vibration data stream within each windowed time segment (i.e., the time-domain vibration data within a fixed-length time segment) to obtain spectral results for multiple time periods. By splicing the spectral results for multiple time periods on the time axis, a two-dimensional frequency-energy response spectrum is generated;

[0059] Performing time segmentation window processing means dividing the preprocessed time-domain vibration data stream into multiple fixed-length time segments (e.g., 1 second). Each segment is weighted by a window function (such as a Hamming window) to reduce the spectral leakage phenomenon, and a fast Fourier transform is performed on the signal for each windowed time period (i.e., the time-domain vibration data within a fixed-length time segment) to obtain spectral results for multiple time segments. By splicing the spectral results for multiple time segments on the time axis, a three-dimensional array of frequency-time-energy is generated, and based on the three-dimensional array of frequency-time-energy, a two-dimensional frequency-energy response spectrum is generated;

[0060] Signal preprocessing: The following processing operations are performed on the original vibration signal to remove noise and redundant information and improve the accuracy of spectral calculation: removing the DC component; using moving average or median filtering for low-frequency denoising; using a band-pass filter to limit the frequency band range (such as 5–500 Hz) to match the core resonance frequency band.

[0061] Mechanism trigger unit: Modal resonance peaks under each frequency band condition are extracted from the two-dimensional frequency-energy response spectrum, compared with the initial baseline modal frequency, the modal shift ratio is calculated and obtained. A continuous determination window is set. If the modal shift ratio exceeds the preset shift threshold within the continuous determination window, it indicates that the modal shift is abnormal, and at this time, the mechanical response mechanism is triggered;

[0062] A frequency band refers to a series of frequency components obtained after FFT transformation within a time segment;

[0063] Extracting the modal resonance peak under each frequency band condition means extracting the frequency point with the maximum energy in the corresponding time period, and this point is the modal resonance peak of this time period;

[0064] The initial baseline modal frequency is the resonance modal frequency corresponding to the corresponding sensor point or measurement area when the system has no structural looseness. The steps to obtain it are as follows:

[0065] 1. Select a time window when the equipment is in a structurally complete state just after commissioning or maintenance;

[0066] 2. Start the multi-modal perception system (especially the three-dimensional vibration sensor);

[0067] 3. Collect the natural running vibration signals for a period of time (without obvious impact or disturbance);

[0068] 4. Perform FFT analysis on the collected time-domain vibration data to extract the resonance peak frequencies;

[0069] 5. Store the frequency peak sequence as the modal baseline library, that is, obtain the baseline modal frequencies;

[0070] The modal shift ratio is used to measure whether there is an obvious frequency drift in the current mode compared with the initial one, and it is specifically obtained through the following formula: , where is the modal shift ratio, is the modal resonance peak, is the initial baseline modal frequency;

[0071] The identification unit receives the mechanical response mechanism, constructs the structural looseness index under different time segment conditions by using the weighted sum of the vibration energy concentration degree and the modal shift ratio, and makes a stabilized judgment through the time-domain averaging method to obtain the stabilized result. Compare the stabilized result with the preset looseness threshold. If the stabilized result exceeds the looseness threshold, the corresponding position is taken as the looseness hot zone, and a spatial heat map is generated according to the looseness hot zone.

[0072] The vibration energy concentration degree is the ratio of the main frequency energy to the full frequency band energy, which is used to reflect the energy focusing degree of a certain frequency band. The specific acquisition method is: , where is the vibration energy concentration degree, f is the frequency number, is the energy corresponding to the main peak frequency, indicating the frequency point with the largest energy on the frequency axis in the fast Fourier transform the corresponding amplitude square, is the total energy of all frequencies in the current time segment, is the vibration energy density on the spectral component corresponding to the frequency f. If significantly increases, such as changing from the original 10% to 40%, it indicates that the vibration energy of the system is abnormally focused, which is common in the resonance coupling of loose parts.

[0073] The structural looseness index is a time-series quantization index with spatial attributes, which reflects whether there is modal frequency shift and abnormal vibration energy aggregation at a certain spatial position and a certain time period. The specific acquisition method is: Structural looseness index = vibration energy concentration degree multiplied by the corresponding weight value + modal shift ratio multiplied by the corresponding weight value;

[0074] During actual operation, dry-type transformers often face disturbances caused by the frequent start-stop of external equipment during operation. These disturbances are conducted to the iron core through the infrastructure, causing the loosening of its tiny structure, and thus there are loosening hot spots.

[0075] The spatial thermal map will mark the positions of each sensor in the three-dimensional vibration sensor array that serves as a loosening hot spot;

[0076] The time-domain averaging method for stabilization judgment, that is, calculating the average value of the structural loosening index under different frequency band conditions; the stabilization result is to take the average value of the structural loosening index at the corresponding position within the monitoring period.

[0077] The main purpose of the local structure analysis subsystem is to identify whether the iron core has a modal frequency drift (i.e., a change in modal response) due to micro-mechanical loosening, to determine whether this drift causes a redistribution of vibration energy, and then to construct a structural loosening index.

[0078] In this embodiment, the integrated monitoring system for the energy consumption of dry-type transformers based on multi-modal sensing proposed by the present invention has a good modular structure and a timing coupling mechanism, and can collect, dynamically analyze, and identify abnormal behaviors of structures and magnetic fluxes in multiple physical modes from the source, and realize the rapid discrimination of abnormal energy consumption states and the output of structural adjustment suggestions.

[0079] First, the multi-modal perception subsystem constructs a sensing network covering multiple physical dimensions by deploying electromagnetic field density probes, three-dimensional vibration sensor arrays, structural acoustic arrays, and thermal imaging probes. All sensors are integrated with a unified clock reference module, which can provide high-precision synchronous clock signals for different types of modal acquisitions, enabling each modal data to have a consistent global time stamp, thus ensuring the time consistency and fusibility between cross-modal information in subsequent analyses.

[0080] For example, when a certain part of the iron core loosens due to mechanical fatigue, this area will exhibit characteristics such as enhanced vibration, magnetic flux disturbance, and local heating. These weak signals can be observed simultaneously through multi-modal synchronous acquisition, providing a basis for subsequent abnormal behavior determination. Secondly, the local structure analysis subsystem constructs a frequency-energy two-dimensional response map from the three-dimensional vibration acceleration signal through the map generation unit. Its core logic is as follows:

[0081] 1) Perform preprocessing operations on the original vibration signal, such as removing DC and filtering;

[0082] 2) Segment by a fixed length (such as 1 second) and apply a window (Hamming window) to reduce spectral leakage;

[0083] 3) Perform a fast Fourier transform (FFT) on each segment of data, splice and construct a three-dimensional frequency-time-energy array, and project it to generate a two-dimensional spectrum. This spectrum is the key basis for the system to identify mechanical modal changes and can intuitively display the evolution process of vibration modal resonance peaks.

[0084] The mechanism trigger unit then calculates the modal offset ratio based on the modal frequency changes extracted from the spectrum. If it continuously exceeds the offset threshold, the mechanical response mechanism can be triggered. This process essentially converts the frequency-domain behavior into a structural response evidence chain, realizing the intelligent conversion from "data" to "event".

[0085] Furthermore, based on the triggering mechanism, the recognition unit fuses the modal offset ratio and the vibration energy concentration, constructs a structural looseness index through weighted combination, and adopts a time-domain averaging strategy on multiple frequency bands and multiple time windows to stably output the results, forming a stable determination result of structural looseness. For example: if a certain sensor point continuously shows high energy concentration and modal offset in multiple frequency bands, its corresponding position will be identified as a loosening hot spot and marked in the spatial heat map. The spatial heat map can reflect the state space distribution of each sensor in the multi-modal sensor array, realizing the cross-dimensional projection from time-frequency features to spatial abnormal hot spots and forming an intuitive visual expression of the structural state.

[0086] In summary, through a series of continuous and closed-loop processing procedures such as the above-mentioned deployment acquisition, frequency-domain analysis, modal response recognition, and stabilization evaluation, the present invention can effectively solve problems existing in traditional dry-type transformer monitoring, such as single modality, time misalignment, and lag in early anomaly recognition. The system can identify and give early warnings in various non-ideal operating states, and is particularly suitable for identifying energy consumption anomalies caused by minor looseness, resonance excitation, and magnetic flux drift, effectively extending the equipment life and improving the operation reliability and energy efficiency level.

[0087] Embodiment 3

[0088] Please refer to Figure 1 , specifically: The disturbance analysis subsystem includes a zone unit, a jump unit, a disturbance unit, and a trajectory unit;

[0089] The zone unit takes the loosening hot spot as the center and expands a grid range of one circle outward as the boundary zone, combines the boundary zone with the loosening hot spot to generate a spatio-temporal field coverage area;

[0090] The jump unit constructs a local magnetic flux density spatio-temporal field within the spatio-temporal field coverage area and uses the slope mutation discrimination method to identify the instantaneous jump position, aiming to find the position and moment where the magnetic flux density suddenly changes with time (i.e., the slope sharply rises or falls);

[0091] The specific content is:

[0092] According to the spatio-temporal field of the local magnetic flux density, time series are extracted at each position to calculate the first-order time derivative (slope) of the magnetic flux density;

[0093] Extract the maximum first-order time derivative of the magnetic flux density from the time series. If the maximum first-order time derivative of the magnetic flux density exceeds the mean value of the maximum first-order time derivative of the magnetic flux density at the corresponding position within the historical period, it is determined that a mutation jump occurs at the corresponding position and is marked as a jump point;

[0094] According to the jump points, record the jump point information, including time, position, and magnetic flux density.

[0095] For the perturbation unit, a preset critical threshold of magnetic flux density saturation is set. If the magnetic flux density at the jump point exceeds the critical threshold of magnetic flux density saturation and is within the loose heat zone, it is determined that the saturation zone is shifted due to the drive of structural perturbation; if the magnetic flux density at the jump point exceeds the critical threshold of magnetic flux density saturation but is not within the loose heat zone, it is determined that the generation of the saturation zone is a non-mechanical drive source;

[0096] The shift of the saturation zone driven by structural perturbation reflects that inside the transformer core where the original magnetic flux distribution is uniform or relatively stable, due to structural looseness or micro-deformation at a certain place, the geometry or magnetic resistance of the magnetic flux path changes, thus causing the spatial migration of the position of the local magnetic flux density aggregation point (i.e., the saturation zone).

[0097] In the case of a healthy structure, the position of the saturation zone is relatively stable. However, when the structure is perturbed, such as uneven clamping force of the iron core or loosening of fasteners, it will cause non-linear redistribution of the magnetic circuit, shifting the saturation point originally located in a certain area to another place.

[0098] On the contrary, if the magnetic flux density at the jump point does not exceed the critical threshold of magnetic flux density saturation, it indicates that the current jump point does not belong to the saturation zone;

[0099] For the trajectory unit, according to the jump points where the saturation zone is shifted due to the drive of structural perturbation, determine the migration path of the saturation zone in space over time and obtain the drift trajectory vector.

[0100] The drift trajectory vector records the coverage area of the movement trajectory of the saturation zone in space over time, and at each moment, the position coordinates of the saturation zone;

[0101] The saturation zone is the position where the magnetic flux density exceeds the critical threshold of magnetic flux density saturation;

[0102] In the dry-type transformer core, when the core material enters the magnetic saturation zone, the non-linear eddy current dissipation in the saturation zone increases, the magnetic flux distribution is broken, turning into uneven local focusing, the local eddy current is significantly enhanced, generating hot spots and local energy consumption. In the long-term operation, it will cause serious problems such as thermal fatigue, insulation breakdown, and performance degradation.

[0103] The local magnetic flux density spatio-temporal field is a three-dimensional functional relationship, representing the magnetic flux density distribution evolving over time at a certain structural region (i.e., spatial position) (x, y). By deploying magnetic flux density probes at various positions within the loose area of the structure (transformer), at each sampling moment, the magnetic flux density values at all position points are recorded, and then through interpolation, a continuous magnetic flux density spatio-temporal field function (which can use spatio-temporal convolution or spatio-temporal interpolation algorithms) is reconstructed, ultimately forming the local magnetic flux density spatio-temporal field;

[0104] In this embodiment, the proposed perturbation analysis subsystem of the present invention realizes the spatio-temporal analysis and modeling of the magnetic flux perturbation behavior in dry-type transformers and the structural coupling traceability judgment through the integration area unit, jump unit, perturbation unit, and trajectory unit. This system can establish a highly sensitive and accurately tracking magnetic flux drift recognition mechanism starting from the loose hot zone in a way that combines spatial stratification and time evolution, effectively enhancing the perception and interpretation ability of local non-linear energy consumption anomalies.

[0105] Logically, the area unit takes the identified loose hot zone as the center and expands one grid range outward as the boundary area. Through the combination of the boundary area and the loose hot zone, a spatio-temporal field coverage area is constructed to define the spatial boundary of subsequent perturbation analysis, focusing on key areas while reducing data redundancy and improving calculation efficiency. For example: when the three-dimensional vibration sensor array detects an obvious structural loose hot zone at the position of the sensor in the upper right corner of the core of the dry-type transformer, the system automatically takes this area as the core and forms a boundary area within the surrounding grids, and combines them to form a coverage area for magnetic flux perturbation analysis.

[0106] In this coverage area, the jump unit constructs a magnetic flux density time series curve based on each coordinate point, and identifies local magnetic flux sharp change points by calculating its first-order time derivative (i.e., magnetic flux slope). Further, the maximum slope value is extracted from each time series and compared with the maximum value mean at this position in the historical period. If it exceeds the significant level, it can be determined that a mutation jump occurs at this position, which is marked as a jump point, and its time, position, and magnetic flux density are recorded.

[0107] For example: at a sensing node (x3, y4) in the coverage area, the magnetic flux slope at the current moment reaches 1.7 T / s, while the historical maximum average slope is only 0.9 T / s. Based on this, the system automatically marks this point as a jump point, indicating that an abnormal magnetic flux perturbation occurs at this position during the current period.

[0108] The perturbation unit compares the magnetic flux density at each jump point with a preset critical threshold of magnetic flux saturation. If the value exceeds the critical threshold and the spatial position is within the loose heat zone, it is determined that the saturation behavior is driven by structural perturbation; if its position is not within the loose heat zone, it is considered that the saturation source is caused by non-mechanical driving factors. Conversely, if the magnetic flux density at the jump point does not reach the saturation threshold, it is not regarded as a saturation zone. For example: The magnetic flux threshold is set to 1.5 T. If the magnetic flux value at the jump point is 1.8 T and it is located in the structural heat zone, it indicates that the formation of the magnetic flux saturation zone is triggered by structural perturbation, and it can be further determined that it is a structure-driven magnetic flux drift.

[0109] During the continuous operation of the system, the trajectory unit tracks the spatial positions and occurrence times of all jump points identified as structure-driven, and constructs a drift trajectory vector that evolves over time. This vector records the coordinate migration path of the magnetic flux saturation zone at each moment in sequence form, forming a continuous motion trajectory, and then maps the dynamic evolution area of the saturation zone. For example: The distribution of jump points collected by the system continuously for 10 minutes forms a trajectory sequence (x3, y4) → (x4, y5) → (x5, y6). The system automatically constructs the corresponding drift trajectory vector, showing that the magnetic flux saturation zone shifts from the center of the transformer to the corner, warning of a possible trend of structural support instability.

[0110] In summary, the perturbation analysis subsystem further realizes the closed-loop identification of the causal chain between structural behavior and magnetic flux perturbation, can accurately locate complex magnetic flux perturbation phenomena to the structural loosening source, and supports the reconstruction of the evolution trajectory and visual tracking of the magnetic flux saturation zone; enhances the system's real-time identification ability for hidden abnormal states, provides a decision-making basis for subsequent structural adjustment and heat dissipation compensation, and avoids the evolution of structural cumulative faults into sudden damages.

[0111] Embodiment 4

[0112] Please refer to Figure 1 , specifically: The dissipation coupling subsystem includes a coupling analysis unit, a quantitative analysis unit, and a comparison and optimization unit;

[0113] The coupling analysis unit determines the drift region according to the drift trajectory vector, and superimposes the drift region on the spatial thermal map to obtain the overlap degree. The overlap degree is used to reflect whether there is a high degree of consistency between the structural vibration excitation and the magnetic flux anomaly, and it is a possible source area of energy consumption surge. Specifically: , where is the overlap degree, is the drift region, is the part of the spatial thermal map with a loose heat zone, is the spatially overlapping region, is the spatially union region; if the overlap degree exceeds the preset overlap threshold, the spatially overlapping region is marked as the coupling excitation zone.

[0114] The drift region is the area covered by the spatio-temporal trajectory formed by multiple jump points during time evolution; the drift trajectory vector is a set of spatially ordered jump points in time series, usually expressed as a vector sequence, and the drift region is the area swept by this set of vectors in space;

[0115] The quantitative analysis unit extracts features from the eigenvalue vector set to obtain the structural vibration energy density and magnetic flux perturbation energy density at different positions in the dry-type transformer. Combining with the coupling excitation region, it analyzes the energy consumption behavior caused by structural vibration excitation in the coupling excitation region. By using the weighted summation algorithm for the structural vibration energy density and magnetic flux perturbation energy density in the coupling excitation region, it evaluates the unit-time coupling energy consumption in the current dry-type transformer coupling excitation region, specifically: , where is the unit-time coupling energy consumption of the coupling excitation region, is the position coordinate point in the coupling excitation region, is the coupling excitation region, is the unit-time vibration energy density of the structural vibration at the coordinate point , is the magnetic energy dissipation density of the magnetic flux perturbation at the coordinate point , and are both weight values used to adjust the contributions of the two to the coupling energy consumption;

[0116] The structural vibration energy density represents the local structural vibration energy consumption trend, and its monitoring and calculation methods are as follows:

[0117] By using a three-dimensional micro-acceleration sensor array installed on the surface of the transformer core to collect vibration signals a(t) in real time and analyzing the position of each sensor point, the structural vibration energy density can be estimated by the unit mass vibration energy formula, specifically: , where is the equivalent mass (which can be approximately taken as the unit mass or obtained through a finite element model), is the velocity vector obtained by integrating the acceleration signal; , , , are the acceleration signals on different axes;

[0118] can be simplified to: , where and are the edge time points within each time period;

[0119] The magnetic flux perturbation energy density represents the heat loss caused by local magnetic flux perturbation, and its monitoring and calculation methods are as follows:

[0120] Use a magnetic flux density probe or Hall effect sensor array, deployed outside the core or in the core opening area, to collect the magnetic flux density and its rate of change in real time. The local eddy current loss caused by the magnetic flux disturbance can be used to approximate the magnetic flux disturbance energy density by the following expression: , is the rate of change of magnetic flux density, is the material conductivity (transformer core material parameter), D is the core thickness, It is a proportional symbol in mathematics; this is an expression that approximately estimates the eddy current energy consumption density per unit volume, reflecting that the more drastic the change in magnetic flux density per unit time, the stronger the conductivity of the material, and the thicker the conductive path, the greater the local electromagnetic dissipation (eddy current heat) generated.

[0121] Specifically, the coupling energy consumption per unit time in the coupling excitation zone is used to quantitatively evaluate the comprehensive energy consumption behavior caused by structural disturbances and magnetic flux anomalies in the coupling excitation zone, and to determine their impact on the system operation efficiency. This is more practical than looking at magnetic flux or vibration alone, and is an energy causal determination method that integrates multi-modal physical fields.

[0122] In this embodiment, the dissipative coupling subsystem accurately identifies the coupling excitation area of magnetic flux anomaly induced by structural disturbance by constructing the spatial overlapping relationship between the drift area and the structural loose hot zone, and further realizes the quantitative evaluation of coupling energy consumption. Specifically, the coupling analysis unit analyzes the spatial coverage area (i.e., the drift area) formed by the drift trajectory vector, and superimposes it with the spatial thermal map generated by the vibration identification unit in the early stage, so as to calculate the spatial overlap between the two, and then determine the possibility of coupling excitation. For example, when the structural vibration caused by the looseness of the core in a certain area of the dry-type transformer is identified as a loose hot zone in the spatial thermal map, and the magnetic flux behavior analysis in the area identifies multiple jump points that evolve over time, and finally the drift trajectory formed by these jump points covers the hot zone. At this time, the overlap between the two increases further, which means that there is a structural-flux coupling driving mechanism in the area, which has a potential high energy consumption risk.

[0123] The system automatically marks the area as a coupling excitation area. On this basis, the quantitative analysis unit extracts the multimodal feature vector set in the coupling excitation area, obtains the structural vibration energy density and magnetic flux disturbance energy density of each coordinate point, and adjusts its contribution to the total energy consumption with the weight parameter. Finally, the coupling energy consumption per unit time is calculated by weighted summation. This indicator has more physical consistency and fault root cause significance than the traditional monitoring of magnetic flux anomalies or vibration intensity. The logic of this step is to construct a coupling space using vibration and magnetic flux information, extract and attribute the energy consumption sources of the coupling space, and finally output a quantifiable coupling energy consumption indicator as a basis for judgment.

[0124] Compared with existing methods, this system can quickly determine during actual operation whether a sharp increase in energy consumption is caused by the coupling effect, avoiding misjudgment, and can quantitatively evaluate the impact of a certain area on the total energy consumption of the entire transformer to assist in operation and maintenance decisions, such as performing structural reinforcement or adjusting the operating conditions. Therefore, this method provides the ability to trace the linkage between structural disturbances and flux offsets, enhances the causal clarity of fault identification, and provides a dual quantitative basis of spatial positioning and energy consumption intensity, improving the scientificity of maintenance priority assessment.

[0125] Embodiment 5

[0126] Please refer to Figure 1 , specifically: The comparison and optimization unit compares the unit-time coupling energy consumption in the coupling excitation area with the total unit-time energy consumption of the dry-type transformer to calculate and obtain the unit-time energy consumption ratio. The unit-time energy consumption ratio is used to reflect the degree of influence of the energy consumption in the coupling excitation area on the total energy consumption in the dry-type transformer. If the unit-time energy consumption ratio exceeds the pre-set ratio threshold, it indicates that this coupling area has a significant impact on system fluctuations, determines that the coupling excitation area at the corresponding time point enters the abnormal dissipation critical state, and marks the coupling excitation area as the abnormal state of structural looseness driving flux drift, so as to transmit it to the background operation platform and execute the structural adjustment mechanism for the coupling excitation area. The structural adjustment mechanism includes suggestions for strengthening the iron core support points and locally reapplying the structural bonding material.

[0127] The total unit-time energy consumption of the dry-type transformer is obtained by installing a power sensor at the input end of the transformer (such as the low-voltage distribution side);

[0128] In this embodiment, the comparison and optimization unit, based on the quantitative comparison mechanism between the unit-time energy consumption in the coupling excitation area and the total unit-time energy consumption of the entire dry-type transformer, realizes the real-time identification of the abnormal influence degree of the coupling area on system energy consumption and the triggering of dynamic regulation. By constructing the unit-time energy consumption ratio index, this unit enables the system to clearly judge whether the composite energy consumption formed by structural vibration and flux disturbance in the coupling excitation area reaches the critical threshold that substantially affects the operating efficiency of the transformer system. When the unit-time coupling energy consumption ratio exceeds the preset threshold, the system automatically determines the corresponding coupling excitation area as the abnormal dissipation critical state, and further marks it as the abnormal state of structural looseness driving flux drift, and transmits it to the background operation platform through the command interface to drive the precise activation of the structural adjustment mechanism.

[0129] The structure adjustment mechanism can automatically recommend reinforcement of the iron core support points or local repair solutions for the structural bonding materials at the corresponding positions based on the identified abnormal state positions, thereby forming a complete closed-loop path from perception → analysis → identification → response without relying on manual inspection. This process has high timeliness and intelligence, effectively reducing the long-term energy efficiency degradation problems caused by the failure to respond in a timely manner to structural micro-loosening, and improving the overall life cycle management level of the equipment. Therefore, this comparison and optimization unit not only improves the analytical ability of the dry-type transformer system for complex coupled abnormal energy consumption, but also realizes the closed-loop control logic for predictive maintenance of structural faults, with significant beneficial effects such as strong intelligent identification, fast response mechanism, and high effectiveness of structural regulation, and is applicable to the safe operation and maintenance of dry-type transformers under high operating loads or critical application scenarios.

[0130] Some of the data in the above formula are numerically calculated after removing their dimensions, and the content not described in detail in this specification belongs to the prior art well-known to those skilled in the art.

[0131] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present application and are not intended to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features; and these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. An integrated monitoring system for the energy consumption of dry-type transformers based on multimodal sensing, characterized in that: including, a multi-modal perception subsystem that pre-deploys multi-modal sensing devices inside a dry-type transformer to collect the operating status of the dry-type transformer and generate a set of eigenvalue vectors; a local structure analysis subsystem that constructs a two-dimensional frequency-energy response map, determines the triggering of the mechanical response mechanism, and identifies loose hot spots; a perturbation analysis subsystem that, based on the loose hot spots, identifies jump points that undergo sudden jumps, and according to the positions of the jump points, filters out the jump points where the saturation region is shifted by structural perturbations to obtain a drift region; a dissipation coupling subsystem that combines the drift region and the loose hot spots to identify a coupled excitation region, and combines the set of eigenvalue vectors to evaluate the impact of the energy consumption in the coupled excitation region on the total energy consumption inside the dry-type transformer, thereby implementing the execution of the dynamic structure adjustment mechanism.

2. The comprehensive monitoring system for the energy consumption of dry-type transformers based on multi-modal sensing according to claim 1, wherein: The multi-modal perception subsystem includes a deployment unit and a collection unit; The deployment unit is used to deploy multi-modal sensing devices inside the dry-type transformer, including electromagnetic field density probes, a three-dimensional vibration sensor array, and thermal imaging probes, and integrates a unified clock reference module in each modal sensing device to achieve timing alignment of data collection; The collection unit is used to use the multi-modal sensing devices to collect vibration acceleration signals, magnetic flux density, structural vibration energy density, and magnetic flux perturbation energy density at various positions in real time. After data preprocessing, a set of eigenvalue vectors is generated.

3. The comprehensive monitoring system for the energy consumption of dry-type transformers based on multi-modal sensing according to claim 2, characterized in that: The local structure analysis subsystem includes a map generation unit, a mechanism trigger unit, and an identification unit; The map generation unit collects vibration acceleration signals in real time by the three-dimensional vibration sensor array to form a time-domain vibration data stream, performs signal preprocessing on the time-domain vibration data stream, performs a time-segment window process on the preprocessed time-domain vibration data stream, weights each time segment using a window function, and performs a fast Fourier transform on the time-domain vibration data stream within each windowed time segment to obtain spectral results for multiple time periods. By splicing the spectral results for multiple time periods on the time axis, a two-dimensional frequency-energy response map is generated; The mechanism trigger unit extracts the modal resonance peaks under each frequency band condition from the two-dimensional frequency-energy response map, compares them with the initial baseline modal frequency, calculates the modal offset ratio, sets a continuous determination window. If the modal offset ratio exceeds the preset offset threshold within the continuous determination window, it indicates that the modal offset is abnormal, and at this time, the mechanical response mechanism is triggered; The identification unit receives the mechanical response mechanism, constructs a structural looseness index under different time segment conditions by weighted summation of the vibration energy concentration degree and the modal offset ratio, and performs a stabilized judgment through time-domain averaging to obtain a stabilized result. The stabilized result is compared with the preset looseness threshold. If the stabilized result exceeds the looseness threshold, the corresponding position is taken as a loose hot spot, and a spatial thermal map is generated based on the loose hot spot.

4. The comprehensive monitoring system for the energy consumption of dry-type transformers based on multi-modal sensing according to claim 3, wherein: The perturbation analysis subsystem includes a region unit, a jump unit, a perturbation unit, and a trajectory unit; The region unit takes the loose hot spot as the center and expands a grid range one circle outward as the boundary region, combines the boundary region and the loose hot spot to generate a spatio-temporal field coverage region; Jump unit, within the spatio-temporal field coverage area, constructs a local magnetic flux density spatio-temporal field, and uses the slope mutation discrimination method to identify the instantaneous jump position. The specific content is as follows: According to the local magnetic flux density spatio-temporal field, a time series is extracted at each position to calculate the first-order time derivative of the magnetic flux density; The maximum first-order time derivative of the magnetic flux density is extracted from the time series. If the maximum first-order time derivative of the magnetic flux density exceeds the mean value of the maximum first-order time derivative of the magnetic flux density at the corresponding position during the historical period, it is determined that a mutation jump occurs at the corresponding position and is marked as a jump point; According to the jump point, the jump point information is recorded, including time, position, and magnetic flux density.

5. The comprehensive monitoring system for the energy consumption of dry-type transformers based on multi-modal sensing according to claim 4, characterized in that: Disturbance unit, preset the critical threshold of magnetic flux density saturation. If the magnetic flux density at the jump point exceeds the critical threshold of magnetic flux density saturation and is in the loose heat zone, it is determined that the saturation zone is shifted due to the drive of structural disturbance; if the magnetic flux density at the jump point exceeds the critical threshold of magnetic flux density saturation but is not in the loose heat zone, it is determined that the generation of the saturation zone is a non-mechanical drive source; Trajectory unit, according to the jump point where the saturation zone is shifted due to the drive of structural disturbance, determines the migration path of the saturation zone in space over time, and obtains the drift trajectory vector.

6. The comprehensive monitoring system for the energy consumption of dry-type transformers based on multi-modal sensing according to claim 5, characterized in that: The dissipation coupling subsystem includes a coupling analysis unit, a quantitative analysis unit, and a comparison and optimization unit; The coupling analysis unit determines the drift region according to the drift trajectory vector, and superimposes the drift region on the spatial heat map to obtain the overlap degree. The overlap degree is used to reflect whether there is a high degree of consistency between the structural vibration excitation and the magnetic flux anomaly. Specifically: , where is the overlap degree, is the drift region, is the part with loose heat zones in the spatial heat map, is the region of spatial overlap, is the region of spatial union; if the overlap degree exceeds the preset overlap threshold, the region of spatial overlap is marked as the coupling excitation region.

7. The comprehensive monitoring system for the energy consumption of dry-type transformers based on multi-modal sensing according to claim 6, wherein: A quantitative analysis unit extracts features from the eigenvalue vector set to obtain the structural vibration energy density and magnetic flux perturbation energy density at different positions inside the dry-type transformer. Combining with the coupled excitation region, it analyzes the energy consumption behavior caused by structural vibration excitation in the coupled excitation region. By using the weighted summation algorithm for the structural vibration energy density and magnetic flux perturbation energy density in the coupled excitation region, it evaluates the unit-time coupling energy consumption in the current coupled excitation region of the dry-type transformer, specifically as follows: , where is the unit-time coupling energy consumption in the coupled excitation region, is the position coordinate point in the coupled excitation region, is the coupled excitation region, is the unit-time vibration energy density of the structural vibration at the coordinate point , is the magnetic energy dissipation density of the magnetic flux perturbation at the coordinate point , and are both weight values.

8. The comprehensive monitoring system for the energy consumption of dry-type transformers based on multi-modal sensing according to claim 7, characterized in that: Comparison and optimization unit, compares the unit-time coupling energy consumption in the coupling excitation area with the total unit-time energy consumption of the dry-type transformer to calculate and obtain the unit-time energy consumption ratio. The unit-time energy consumption ratio is used to reflect the influence degree of the energy consumption in the coupling excitation area on the total energy consumption in the dry-type transformer. If the unit-time energy consumption ratio exceeds the pre-set ratio threshold, it is determined that the coupling excitation area at the corresponding time point enters the abnormal dissipation critical state, and the coupling excitation area is marked as the abnormal state of structural loosening driving magnetic flux drift, and is transmitted to the background operation platform to execute the structure adjustment mechanism for the coupling excitation area. The structure adjustment mechanism includes suggestions for strengthening the iron core support points and local reapplication of the structural bonding material.

Images