Transformer oil treatment whole-process intelligent monitoring and quality tracing platform

Abstract

The invention relates to the technical field of intelligent control, in particular to a transformer oil treatment whole-process intelligent monitoring and quality tracing platform which comprises an architecture mapping module, a path tracking module, a risk pre-judgment module, an offset evaluation module, a coefficient correction module, an interval reconstruction module and a regulation and control filing module. A transmission framework is obtained; dynamically tracking the operation parameters to obtain a real-time transmission path diagram; performing step-by-step cumulative effect quantization on the real-time transmission path diagram, and performing risk advanced pre-judgment on the real-time transmission path diagram to obtain overrun nodes; performing high-frequency sampling on adjacent downstream links of the overrun node to obtain a pre-judgment accuracy rate; iteratively correcting the step-by-step amplification coefficient to obtain an optimized amplification coefficient; performing threshold screening on the optimized amplification coefficient to obtain a parameter stable interval; active regulation and control are carried out on transformer oil affairs to obtain a traceability file; according to the invention, the efficiency of intelligent monitoring and quality tracing of transformer oil treatment can be improved.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology, and in particular to an intelligent monitoring and quality traceability platform for the entire process of transformer oil handling. Background Technology

[0002] The current intelligent monitoring system for transformer oil handling lacks a global flow correlation mapping capability for monitoring the operating parameters of each handling stage. It cannot build a complete parameter transmission architecture. Dynamic tracking of parameters can only achieve single-point data collection, making it difficult to form a continuous real-time transmission path diagram. At the same time, there are no quantitative analysis methods for the cumulative effect of parameter fluctuations at each level, making it impossible to predict risks in the handling process in advance. It can only take passive measures after parameters exceed the limits, resulting in a significant lag in risk control during the oil handling process and failing to avoid oil quality problems caused by parameter anomalies from the source.

[0003] Existing technologies for quality traceability in transformer oil handling lack a mechanism for assessing the degree of deviation of risk prediction nodes. They cannot accurately obtain prediction accuracy through high-frequency sampling, thus failing to effectively iteratively correct the step-by-step amplification coefficient of fluctuation quantification. The reconstruction of parameter stability intervals also fails to incorporate differentiated threshold screening based on the historical response characteristics of fluctuation paths, making it difficult to define collaborative operating ranges based on path sensitivity. Furthermore, the records of oil handling control processes and parameter stability intervals lack standardized temporal alignment and encapsulation, and the traceability identification binding of the entire process data is incomplete, resulting in low efficiency in quality traceability and an inability to achieve verifiable, traceable, and accurate correlation of the entire oil handling process data. Therefore, improving the efficiency of intelligent monitoring and quality traceability of the entire transformer oil handling process has become an urgent problem to be solved. Summary of the Invention

[0004] To achieve the above objectives, the present invention provides an intelligent monitoring and quality traceability platform for the entire transformer oil handling process. The platform includes an architecture mapping module, a path tracing module, a risk prediction module, a deviation assessment module, a coefficient correction module, an interval reconstruction module, and a control and archiving module, wherein: The architecture mapping module is used to perform flow-direction association mapping on the operating parameters of transformer oil to obtain the transmission architecture of the operating parameters; The path tracing module is used to dynamically track the running parameters based on the transmission architecture to obtain a real-time transmission path diagram of the running parameters. The risk prediction module is used to quantify the cumulative effect of the parameter fluctuation amplitude of the real-time transmission path map, obtain the step-by-step amplification coefficient of the real-time transmission path map, and based on the step-by-step amplification coefficient, perform risk prediction on the real-time transmission path map to obtain the over-limit nodes of the real-time transmission path map. The offset evaluation module is used to perform high-frequency sampling of the adjacent downstream links of the over-limit node, and based on the sampling data, evaluate the degree of offset of the over-limit node to obtain the prediction accuracy of the over-limit node. The coefficient correction module is used to iteratively correct the step-by-step amplification coefficient based on the prediction accuracy to obtain the optimized amplification coefficient of the step-by-step amplification coefficient. The interval reconstruction module is used to perform threshold filtering on the optimized amplification coefficient and to reconstruct the operating range of the associated fluctuation path of the filtered coefficient to obtain the parameter stability interval of the associated fluctuation path. The control and archiving module is used to actively control the transformer oil condition based on the stable range, and record and integrate the control process to obtain the traceability file of the transformer oil condition.

[0005] In a preferred embodiment, when the architecture mapping module performs flow-direction association mapping on the operating parameters of transformer oil to obtain the transmission architecture of the operating parameters, it is specifically used for: The operating parameters of the transformer oil service at each processing stage are synchronously captured on a time axis to obtain a time-series capture record of the operating parameters; Based on the time-series capture records, a sliding window matching is performed on the running parameters to obtain the time delay correlation of the running parameters; Based on the aforementioned time delay correlation, the forward and reverse flow directions of the operating parameters are determined to obtain the flow direction identifier of the operating parameters; Based on the flow direction identifier, the topology of the running parameters is reconstructed to obtain the transmission architecture of the running parameters.

[0006] In a preferred embodiment, when the path tracing module performs dynamic tracing of the running parameters based on the transmission architecture to obtain a real-time transmission path graph of the running parameters, it is specifically used for: Based on the aforementioned transmission architecture, the operational parameters are subjected to sliding window fluctuation identification to obtain fluctuation segments of the operational parameters. The start and end times of the wave segment are interleaved and compared to obtain the wave propagation direction of the wave segment; Based on the wave propagation direction, a similarity measurement is performed on the waveform profile of the wave segment to obtain the waveform matching degree of the wave segment; Based on the waveform matching degree, the wave segments are merged and associated across nodes to obtain the wave propagation trajectory segments of the running parameters; Based on the aforementioned transmission architecture, the wave propagation trajectory segments are stitched together to obtain a real-time transmission path diagram of the operational parameters.

[0007] In a preferred embodiment, when the risk prediction module performs a step-by-step cumulative effect quantification on the parameter fluctuation amplitude of the real-time transmission path diagram to obtain the step-by-step amplification coefficient of the real-time transmission path diagram, and performs risk advance prediction on the real-time transmission path diagram based on the step-by-step amplification coefficient to obtain the over-limit nodes of the real-time transmission path diagram, it is specifically used for: The parametric fluctuation sequence of the independent path in the real-time transmission path diagram is subjected to fluctuation mode decomposition to obtain the fast fluctuation component and slow trend component of the independent path. Based on the rapidly changing wave components, the wave propagation phase of the independent path is compared with the lead-lag phase to obtain the wave propagation phase difference sequence of the independent path. Based on the wave propagation phase difference sequence, coherence is determined for the independent path to obtain the wave in-phase superposition segment and the wave out-of-phase cancellation segment of the independent path. The parametric fluctuation amplitude of the superimposed wave segment is recursively accumulated node by node to obtain the step-by-step amplification coefficient of the independent path; Based on the stepwise amplification coefficient, the fluctuation amplitude of the terminal node of the independent path is extrapolated to obtain the expected fluctuation peak value of the terminal node. The end nodes whose expected fluctuation peak exceeds the preset risk threshold are marked as over-limit nodes in the real-time transmission path diagram.

[0008] In a preferred embodiment, when the risk prediction module performs node-by-node recursive accumulation of the parametric fluctuation amplitude of the overlapping fluctuation segment to obtain the step-by-step amplification coefficient of the independent path, it is specifically used for: Amplitude correlation matching is performed on the parametric fluctuation amplitude of the superimposed wave phase segment to obtain the wave peak-valley pairing sequence of the superimposed wave phase segment; Based on the peak-valley pairing sequence, the inter-node transmission characteristics of the independent path are decoupled to obtain the instantaneous transmission gain factor of the independent path. Based on the instantaneous transfer gain factor, the transfer gain loss of the independent path is traced to obtain the current cumulative adsorbent consumption of the independent path. The correlation analysis between the instantaneous transfer gain factor and the wave transfer phase difference sequence is performed to obtain the in-phase superposition contribution coefficient of the independent path; Based on the instantaneous transfer gain factor and the in-phase superposition contribution coefficient, the step-by-step amplification factor of the independent path is calculated, wherein the formula for calculating the step-by-step amplification factor is: ; In the formula, For the independent path, the first The amplification factor of each node at each level, For the independent path, the first The amplification factor of each node at each level, For the independent path, the first The instantaneous transfer gain factor of each node, The node number of the independent path. For the independent path, the first The current cumulative adsorbent consumption of each node. For the independent path, the first The preset adsorbent saturation capacity of each node, For the independent path, the first The in-phase superposition contribution coefficient of each node, The preset waveform steepness factor, For the independent path, the first The ratio of the wave energy of the in-phase superposition segment to the out-of-phase cancellation segment at each node. It is a natural constant. For the independent path, the first The number of sub-paths contained within the in-phase overlay segment at each node. For the independent path, the first The sequence number of the sub-path within the in-phase overlapping section at each node. For the independent path, the first At the node, the first Normalized fluctuation amplitude of the strip path.

[0009] In a preferred embodiment, when the offset evaluation module performs high-frequency sampling of the adjacent downstream links of the out-of-limit node and evaluates the degree of offset of the out-of-limit node based on the sampled data to obtain the prediction accuracy of the out-of-limit node, it is specifically used for: Real-time parametric interruption capture is performed on the adjacent downstream links of the over-limit node to obtain the dense sampling sequence of the adjacent downstream links; Peak and valley extreme value retrieval is performed on the dense sampling sequence to obtain the fluctuation peak and valley records of the adjacent downstream links; The peak and valley times of the fluctuation peak and valley records are aligned and mapped with the risk occurrence time window of the over-limit node to obtain the pairing sequence of the over-limit node and the fluctuation peak and valley records. Based on the measured results of the out-of-limit nodes, the matching sequence is judged to obtain the prediction accuracy of the out-of-limit nodes.

[0010] In a preferred embodiment, when the coefficient correction module performs iterative correction of the step-by-step amplification coefficient based on the prediction accuracy to obtain the optimized amplification coefficient, it is specifically used for: The correlation between the step-by-step amplification coefficient and the prediction accuracy is analyzed to obtain the coefficient term to be corrected for the step-by-step amplification coefficient; Based on the correlation parameters of the stepwise amplification coefficient, reverse deviation compensation is performed on the coefficient term to be corrected to obtain the single correction value of the coefficient term to be corrected. Based on the single correction value, the coefficient to be corrected is incrementally updated, and the updated amplification coefficient is re-predicted for risk, so as to obtain the prediction accuracy update value of the updated amplification coefficient. The updated prediction accuracy values ​​are selectively retained to obtain the optimized amplification coefficient of the step-by-step amplification coefficient.

[0011] In a preferred embodiment, when the interval reconstruction module performs threshold filtering on the optimized amplification coefficient and reconstructs the operating range of the associated fluctuation path of the filtered coefficient to obtain the parameter stability interval of the associated fluctuation path, it is specifically used for: By backtracking the historical response characteristics of the associated fluctuation path of the optimized amplification factor, the historical stability index of the associated fluctuation path is obtained. Based on the historical stability index, the path-specific boundary of the optimized amplification coefficient is delineated to obtain the differentiated screening threshold for each associated fluctuation path. Based on the differentiated screening threshold, the correlation fluctuation paths are classified by fluctuation sensitivity to obtain high-sensitivity paths and low-sensitivity paths of the correlation fluctuation paths. The operating parameter range of the high-sensitivity path is reconstructed through collaborative operation to obtain the parameter stability range of the associated fluctuation path.

[0012] In a preferred embodiment, when the interval reconstruction module performs cooperative operating range reconstruction on the operating parameter interval of the highly sensitive path to obtain the parameter stability interval of the associated fluctuation path, it is specifically used for: Based on the high-sensitivity path, the optimized amplification coefficients are normalized to obtain the sensitivity weight ratio of each optimized amplification coefficient. Based on the sensitivity weight ratio, the collaborative compression coefficient of the high-sensitivity path is calculated, wherein the formula for calculating the collaborative compression coefficient is: ; In the formula, For the first The collaborative compression coefficient of the high-sensitivity path, For the first The optimized amplification factor for a high-sensitivity path. For the first A highly sensitive path, The total number of the highly sensitive paths. For the first A highly sensitive path, For the wave propagation phase difference sequence, the first... The highly sensitive path and the first The transmission coupling strength of a highly sensitive path, For the first The optimized amplification factor for a highly sensitive path; Based on the cooperative compression coefficient, the original operating parameter range of the high-sensitivity path is nonlinearly compressed to obtain the compressed range of the high-sensitivity path. The compressed interval and the original interval of the low-sensitivity path are reconstructed by boundary nesting to obtain the parameter stable interval of the associated fluctuation path.

[0013] In a preferred embodiment, when the control and archiving module performs active control of the transformer oil condition based on the parameter stability range and records and integrates the control process to obtain a traceability file for the transformer oil condition, it is specifically used for: Based on the stable range of the parameters, real-time control commands are issued for the transformer oil system to obtain the sequence of actions to be performed on the transformer oil system. Based on the sequence of actions, the operating conditions of the transformer oil purification process are adjusted to obtain the operating condition adjustment record of the purification process. The operating condition adjustment record and the stable interval are time-aligned and encapsulated to obtain the associated mapping file of the operating condition adjustment record and the stable interval; Based on the associated mapping file, the entire process record of the transformer oil service is traceable and bound to a traceability identifier to obtain the traceability file of the transformer oil service.

[0014] Compared with the prior art, the present invention has the following beneficial effects: 1. This invention achieves intelligent monitoring of transformer oil handling parameters throughout the entire process through multi-module collaboration. First, it completes the flow association mapping and dynamic tracking of operating parameters, accurately constructs the transmission architecture and real-time transmission path diagram, and can also quantify the cumulative effect of parameter fluctuations and predict over-limit nodes in advance. At the same time, it evaluates the prediction accuracy through high-frequency sampling, iteratively corrects the amplification coefficient, reconstructs the parameter stability range based on path characteristics, and carries out active control, which greatly improves the accuracy, foresight, and initiative of parameter monitoring in the oil handling process, making the operation and management of oil handling more scientific and effectively improving the overall efficiency of intelligent monitoring.

[0015] 2. This invention establishes a standardized transformer oil quality traceability system. After oil regulation is completed based on the parameter stability range, the regulation execution actions and operating condition adjustment records are sequentially aligned and encapsulated. Furthermore, a unique traceability identifier is bound to the entire process processing record, forming a complete and related traceability archive. This achieves systematic integration and accurate traceability of the entire oil processing process data, improves the integrity and efficiency of quality traceability, and makes the quality information of the entire oil processing process verifiable and traceable, ensuring the standardized implementation of quality traceability work. Attached Figure Description

[0016] Figure 1 This is a platform architecture diagram of an intelligent monitoring and quality traceability platform for the entire transformer oil handling process provided in an embodiment of the present invention; The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0017] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments belong to some, but not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] The terminology used in the embodiments of this invention is for the purpose of describing particular embodiments only and is not intended to limit the invention. The singular forms “said” and “the” as used in the embodiments of this invention and the appended claims are also intended to include the plural forms, and “multiple” generally includes at least two unless the context clearly indicates otherwise.

[0019] Depending on the context, the word "if" or "if" as used here can be interpreted as "when," "when," "in response to determination," or "in response to detection." Similarly, depending on the context, the phrase "if determination" or "if detection (of the stated condition or event)" can be interpreted as "when determination," "in response to determination," "when detection (of the stated condition or event)," or "in response to detection (of the stated condition or event)."

[0020] Furthermore, the timing of the steps in the following method embodiments is merely an example and not a strict limitation.

[0021] In practice, the server-side equipment deployed by the intelligent monitoring and quality traceability platform for the entire transformer oil handling process may consist of one or more devices. This platform can be implemented as a business instance, a virtual machine, or hardware devices. For example, it can be implemented as a business instance deployed on one or more devices in a cloud node. Simply put, it can be understood as software deployed on a cloud node, providing intelligent monitoring and quality traceability services for the entire transformer oil handling process to various user terminals. Alternatively, it can be implemented as a virtual machine deployed on one or more devices in a cloud node, with application software installed to manage various user terminals. Or, it can also be implemented as a server composed of numerous identical or different types of hardware devices, with one or more devices configured to provide intelligent monitoring and quality traceability services for the entire transformer oil handling process to various user terminals.

[0022] In terms of implementation, the intelligent monitoring and quality traceability platform for the entire transformer oil handling process and the user terminal are mutually compatible. That is, if the intelligent monitoring and quality traceability platform for the entire transformer oil handling process is implemented as an application installed on a cloud service platform, then the user terminal is a client that establishes a communication connection with the application; or if the intelligent monitoring and quality traceability platform for the entire transformer oil handling process is implemented as a website, then the user terminal is implemented as a webpage; or if the intelligent monitoring and quality traceability platform for the entire transformer oil handling process is implemented as a cloud service platform, then the user terminal is implemented as a mini-program in an instant messaging application.

[0023] like Figure 1 The diagram shown is a platform architecture diagram of an intelligent monitoring and quality traceability platform for the entire transformer oil handling process provided in an embodiment of the present invention.

[0024] The intelligent monitoring and quality traceability platform 10 for the entire transformer oil handling process described in this invention can be located on a cloud server. In terms of implementation, it can be one or more service devices, or an application installed on the cloud (e.g., a mobile service operator's server, server cluster, etc.), or it can be developed into a website. Depending on the implemented functions, the intelligent monitoring and quality traceability platform 10 for the entire transformer oil handling process may include an architecture mapping module 11, a path tracing module 12, a risk prediction module 13, an offset evaluation module 14, a coefficient correction module 15, an interval reconstruction module 16, and a control and archiving module 17. The modules described in this invention can also be called units, referring to a series of computer program segments that can be executed by an electronic device's processor and perform a fixed function, stored in the electronic device's memory.

[0025] In this embodiment of the invention, the intelligent monitoring and quality traceability platform for the entire transformer oil handling process can be implemented independently and can call other modules. This "calling" can be understood as one module connecting to multiple modules of another type and providing corresponding services to those connected modules. In the intelligent monitoring and quality traceability platform for the entire transformer oil handling process provided by this embodiment of the invention, the applicable scope of the platform architecture can be adjusted by adding modules and directly calling them without modifying the program code, achieving cluster-based horizontal expansion to quickly and flexibly expand the platform. In practical applications, the above modules can be set in the same device or different devices, or they can be set in virtual devices, such as service instances in a cloud server.

[0026] The following describes the components and specific workflow of the intelligent monitoring and quality traceability platform for the entire transformer oil handling process, using specific embodiments as examples: The architecture mapping module 11 is used to perform flow-direction association mapping on the operating parameters of transformer oil to obtain the transmission architecture of the operating parameters. In this embodiment of the invention, when the architecture mapping module performs flow-direction association mapping on the operating parameters of transformer oil to obtain the transmission architecture of the operating parameters, it is specifically used for: The operating parameters of the transformer oil service at each processing stage are synchronously captured on a time axis to obtain a time-series capture record of the operating parameters; Based on the time-series capture records, a sliding window matching is performed on the running parameters to obtain the time delay correlation of the running parameters; Based on the aforementioned time delay correlation, the forward and reverse flow directions of the operating parameters are determined to obtain the flow direction identifier of the operating parameters; Based on the flow direction identifier, the topology of the running parameters is reconstructed to obtain the transmission architecture of the running parameters.

[0027] For each specific step in transformer oil handling, a unified time measurement standard is set. Real-time data of operating parameters in each step are collected synchronously according to this standard. During the collection process, the time stamps corresponding to each parameter data are completely preserved. The operating parameter data of all processing steps are organized in an orderly manner according to the time stamp sequence, and finally a complete record containing complete time information and parameter data of each step is formed. This record is the time sequence capture record of operating parameters.

[0028] Based on the time stamps corresponding to each processing stage in the timing capture record, a fixed-length window range is set. This window is then sequentially shifted along the timeline of the timing capture record at fixed steps. At each window's stopping position, all operational parameter data for each processing stage within the window are precisely extracted. The occurrence time and data characteristics of parameter data from different processing stages within the same window are compared one by one to clarify the time delay between parameter data of each stage. The time delay information between all processing stages and the corresponding parameter correlation information are systematically integrated. The integrated overall information is the time delay correlation of the operational parameters.

[0029] Based on the time delay information between each processing stage as clearly defined in the time delay correlation, the order of transmission of operational parameter data in each stage is accurately sorted out and clarified. Combined with the actual process logic of transformer oil handling, the specific direction of parameter data transmission from one processing stage to another is determined one by one, clearly distinguishing the forward and reverse flow of parameter transmission. The corresponding parameter transmission direction information is accurately marked for each processing stage, and a unique direction identification information is set for the parameter transmission relationship between each group of processing stages. All marked direction information and set identification information are integrated as a whole, and the integrated information is the flow direction identification of the operational parameters.

[0030] Based on the parameter transmission direction information corresponding to each processing stage in the flow direction identifier, the parameter connection relationship between all processing stages is comprehensively sorted out. According to the actual direction of parameter transmission, each processing stage is arranged as an independent node in an orderly manner. The parameter transmission relationship between each node is represented by a line, and the direction of the line is completely consistent with the direction information corresponding to the flow direction identifier. All nodes representing processing stages and lines representing parameter transmission relationships are systematically sorted out and integrated, and finally an overall structure that can clearly and completely show the direction of parameter transmission and connection relationship between each processing stage is formed. This structure is the transmission architecture of the operating parameters.

[0031] The beneficial effects include the synchronous collection and standardized organization of operating parameters in all processing stages of transformer oil handling, accurate identification of the time delay correlation between parameters in each stage, clear determination and marking of the parameter transmission direction, and finally the construction of a transmission architecture that can completely and clearly show the parameter transmission direction and connection relationship between each processing stage. This makes the flow correlation mapping of transformer oil handling operating parameters more accurate and orderly, and provides a clear and reliable basis for subsequent dynamic tracking of parameters.

[0032] The path tracing module 12 is used to dynamically track the running parameters based on the transmission architecture to obtain a real-time transmission path diagram of the running parameters. In this embodiment of the invention, when the path tracing module performs dynamic tracing of the running parameters based on the transmission architecture to obtain a real-time transmission path graph of the running parameters, it is specifically used for: Based on the aforementioned transmission architecture, the operational parameters are subjected to sliding window fluctuation identification to obtain fluctuation segments of the operational parameters. The start and end times of the wave segment are interleaved and compared to obtain the wave propagation direction of the wave segment; Based on the wave propagation direction, a similarity measurement is performed on the waveform profile of the wave segment to obtain the waveform matching degree of the wave segment; Based on the waveform matching degree, the wave segments are merged and associated across nodes to obtain the wave propagation trajectory segments of the running parameters; Based on the aforementioned transmission architecture, the wave propagation trajectory segments are stitched together to obtain a real-time transmission path diagram of the operational parameters.

[0033] Based on the established operational parameter transmission architecture, the acquisition dimensions and temporal correlation attributes of parameter data for each node are clarified. A fixed-length sliding window is set and continuously translated on the temporal data axis of the operational parameters according to equal intervals. At each stop position of the window, all operational parameter data of each node within the corresponding time period are extracted. The extracted parameter data is compared with the benchmark stable value of the node parameter time by time, and all parameter data segments that deviate from the benchmark stable value are extracted. Each segment of parameter data that continuously deviates from the benchmark stable value is independently divided and organized according to the node and time series to which it belongs. Each segment of parameter data after organization is the fluctuation segment of the operational parameter.

[0034] A comprehensive time series analysis is performed on all fluctuation segments obtained from each node. The start and end times of each fluctuation segment are extracted and clearly and uniquely labeled. Fluctuation segments between different nodes are paired in an orderly manner according to the node connection relationship of the transmission architecture. The start and end times of the paired fluctuation segments are compared and contrasted to analyze the sequential connection relationship between the time series of fluctuation segments of the previous node and the time series of fluctuation segments of the next node. Based on the sequential connection relationship, the specific direction of parametric fluctuation transmission from one node to another is determined. The specific directions of parametric fluctuation transmission between all nodes are systematically summarized. All the summarized direction information is the fluctuation transmission direction of the fluctuation segment.

[0035] Based on the determined wave propagation direction, wave segments corresponding to each node propagating sequentially along this direction are identified. The wave segment of the starting node in the wave propagation direction is used as the reference segment. The waveform contour features of the reference wave segment are extracted, including the temporal features and data change features of the waveform's rising edge duration, falling edge duration, peak position, and trough position. Then, the waveform contour features of the same dimension of the wave segments corresponding to subsequent nodes are extracted sequentially. The waveform contour features of the wave segments of subsequent nodes are compared with the waveform contour features of the reference wave segment dimension by dimension. The degree of fit between the wave segments of each node and the reference wave segment is determined based on the fit of the features in each dimension. The degree of fit is quantified and labeled according to a unified standard. The quantified result after labeling is the waveform matching degree of the wave segment.

[0036] Based on the waveform matching degree corresponding to each node's fluctuation segment, all fluctuation segments that meet the preset matching requirements along the fluctuation transmission direction are selected. These qualified fluctuation segments are then seamlessly connected in time according to the node order in the fluctuation transmission direction. All connected fluctuation segments on the same fluctuation transmission path are merged into a whole parameter fluctuation transmission data segment. At the same time, the original node identifiers and timing information of each node's fluctuation segment in this data segment are completely preserved. Each merged whole parameter fluctuation transmission data segment is the fluctuation propagation trajectory segment of the running parameter.

[0037] Based on the transmission architecture of operational parameters, the connection relationships of all nodes and the overall path layout in the architecture are comprehensively sorted out. All obtained fluctuation propagation trajectory fragments are accurately classified according to their respective transmission paths. The temporal sequence and node connection relationship of the fluctuation propagation trajectory fragments on each transmission path are checked for continuity. The fluctuation propagation trajectory fragments that meet the continuity requirements after verification are completely spliced ​​together according to the node connection relationship and temporal sequence. After splicing, complete parameter fluctuation propagation trajectories on each transmission path are formed. The complete fluctuation propagation trajectories on all transmission paths are accurately matched and integrated with the node and path layout of the transmission architecture. The real-time transmission path and trajectory characteristics of parameter fluctuations between nodes are displayed in a visual form. The overall visualization formed after integration is the real-time transmission path diagram of operational parameters.

[0038] The beneficial effects include the ability to accurately identify fluctuation segments of operating parameters based on the transmission architecture, clearly define the direction of fluctuation transmission, quantify the degree of waveform contour fit to obtain waveform matching degree, and merge and associate these segments to form fluctuation propagation trajectory segments. Finally, these segments are spliced ​​and fused to form a real-time transmission path diagram, enabling precise dynamic tracking of operating parameters and fully displaying the real-time transmission status of parameter fluctuations. This provides a clear and accurate path basis for subsequent risk prediction.

[0039] The risk prediction module 13 is used to quantify the cumulative effect of the parameter fluctuation amplitude of the real-time transmission path map step by step, obtain the step-by-step amplification coefficient of the real-time transmission path map, and make risk prediction of the real-time transmission path map based on the step-by-step amplification coefficient to obtain the over-limit nodes of the real-time transmission path map. In this embodiment of the invention, when the risk prediction module performs a step-by-step cumulative effect quantification on the parameter fluctuation amplitude of the real-time transmission path diagram to obtain the step-by-step amplification coefficient of the real-time transmission path diagram, and performs risk advance prediction on the real-time transmission path diagram based on the step-by-step amplification coefficient to obtain the over-limit nodes of the real-time transmission path diagram, it is specifically used for: The parametric fluctuation sequence of the independent path in the real-time transmission path diagram is subjected to fluctuation mode decomposition to obtain the fast fluctuation component and slow trend component of the independent path. Based on the rapidly changing wave components, the wave propagation phase of the independent path is compared with the lead-lag phase to obtain the wave propagation phase difference sequence of the independent path. Based on the wave propagation phase difference sequence, coherence is determined for the independent path to obtain the wave in-phase superposition segment and the wave out-of-phase cancellation segment of the independent path. The parametric fluctuation amplitude of the superimposed wave segment is recursively accumulated node by node to obtain the step-by-step amplification coefficient of the independent path; Based on the stepwise amplification coefficient, the fluctuation amplitude of the terminal node of the independent path is extrapolated to obtain the expected fluctuation peak value of the terminal node. The end nodes whose expected fluctuation peak exceeds the preset risk threshold are marked as over-limit nodes in the real-time transmission path diagram.

[0040] When the risk prediction module performs node-by-node recursive accumulation of the parametric fluctuation amplitude of the overlapping fluctuation segment to obtain the step-by-step amplification coefficient of the independent path, it is specifically used for: Amplitude correlation matching is performed on the parametric fluctuation amplitude of the superimposed wave phase segment to obtain the wave peak-valley pairing sequence of the superimposed wave phase segment; Based on the peak-valley pairing sequence, the inter-node transmission characteristics of the independent path are decoupled to obtain the instantaneous transmission gain factor of the independent path. Based on the instantaneous transfer gain factor, the transfer gain loss of the independent path is traced to obtain the current cumulative adsorbent consumption of the independent path. The correlation analysis between the instantaneous transfer gain factor and the wave transfer phase difference sequence is performed to obtain the in-phase superposition contribution coefficient of the independent path; Based on the instantaneous transfer gain factor and the in-phase superposition contribution coefficient, the step-by-step amplification factor of the independent path is calculated, wherein the formula for calculating the step-by-step amplification factor is: ; In the formula, For the independent path, the first The amplification factor of each node at each level, For the independent path, the first The amplification factor of each node at each level, For the independent path, the first The instantaneous transfer gain factor of each node, The node number of the independent path. For the independent path, the first The current cumulative adsorbent consumption of each node. For the independent path, the first The preset adsorbent saturation capacity of each node, For the independent path, the first The in-phase superposition contribution coefficient of each node, The preset waveform steepness factor, For the independent path, the first The ratio of the wave energy of the in-phase superposition segment to the out-of-phase cancellation segment at each node. It is a natural constant. For the independent path, the first The number of sub-paths contained within the in-phase overlay segment at each node. For the independent path, the first The sequence number of the sub-path within the in-phase overlapping section at each node. For the independent path, the first At the node, the first Normalized fluctuation amplitude of the strip path.

[0041] Extract the complete parameter fluctuation sequence of each independent path in the real-time transmission path diagram, organize it into continuous parameter change data segments according to time sequence, perform hierarchical decomposition on the data segments, separate the parts of parameter fluctuation with fast change rate and organize them into continuous component data, which is the fast change fluctuation component of the independent path, and separate the parts of parameter fluctuation with slow change rate and fixed change trend and organize them into continuous component data, which is the slow change trend component of the independent path.

[0042] Extract the wave phase characteristics at each node in the rapidly changing wave component. According to the node transmission order of the independent path, compare the wave phase characteristics of the previous node with the wave phase characteristics of the next node one by one to clarify the leading and lagging relationship of the wave phase between each node. Record the phase difference between each group of nodes in an orderly manner to form a continuous difference record sequence. This sequence is the wave transmission phase difference sequence of the independent path.

[0043] Based on the phase difference values ​​between each node in the wave propagation phase difference sequence, the propagation intervals of each segment of the independent path are judged one by one. The propagation interval with a phase difference value of zero is defined as the interval with the same wave phase, which is the wave phase superposition segment of the independent path. The propagation interval with a phase difference value of a fixed out-of-phase value is defined as the interval with opposite wave phase, which is the out-of-phase cancellation segment of the independent path.

[0044] Extract all peak and trough data of parametric fluctuation amplitude of each node in the wave phase superposition section, match the peak and trough data one by one according to time sequence, associate each peak with its adjacent previous and next trough, and mark the node position and time sequence information corresponding to each peak and trough. Organize all associated peak and trough data in order of transmission to form a complete matching data sequence, which is the peak and trough pairing sequence of the wave phase superposition section.

[0045] Based on the peak and valley amplitude data of each node in the peak and valley pairing sequence, the transmission and change characteristics of parametric fluctuations between nodes are sorted out. The influence of the characteristics of each node on the transmission of fluctuations is removed, and only the gain change characteristics in the transmission process of parametric fluctuations between nodes are extracted. The gain change characteristics at each node are quantified and labeled according to time sequence and node order to form continuous feature data. This data is the instantaneous transmission gain factor of independent path.

[0046] Based on the gain change characteristics of each node in the instantaneous transfer gain factor, and compared with the standard gain characteristics under the condition of no adsorbent consumption, the correlation between gain change and adsorbent consumption at each node is analyzed. The consumption of adsorbent at each node is traced according to the degree of gain change. The cumulative consumption data of adsorbent at each node is accurately statistically recorded. This recorded data is the current cumulative adsorbent consumption of an independent path.

[0047] All data of instantaneous transfer gain factor and all data of wave transfer phase difference sequence are precisely matched according to node position and time sequence. The correlation characteristics of the two sets of data at each node are analyzed. The contribution characteristics of instantaneous transfer gain at each node to wave in-phase superposition are extracted. The contribution characteristics are quantified and labeled according to node order to form continuous feature data. This data is the in-phase superposition contribution coefficient of independent path.

[0048] The instantaneous transfer gain factor and the in-phase superposition contribution coefficient are matched one-to-one according to the node position. The two data at each node are fused together according to the transfer order between each node. Starting from the starting node of the independent path, the cumulative amplification characteristic data of the parameter fluctuation amplitude are sorted out and recorded node by node. The cumulative amplification characteristic data at each node are arranged into a continuous sequence according to the node order. This sequence is the step-by-step amplification coefficient of the independent path.

[0049] Extract the cumulative amplification characteristic data of each node in the step-by-step amplification coefficient, sort out the amplification change law from the starting node to the end node according to the node transmission order of the independent path, and deduce the parameter fluctuation amplitude at the end node based on the law to obtain the corresponding peak data. This data is the expected fluctuation peak value of the end node.

[0050] The expected fluctuation peak of each independent path terminal node is compared with the preset risk threshold one by one. Terminal nodes whose expected fluctuation peak value exceeds the preset risk threshold are clearly marked. All marked terminal nodes are integrated as a whole. The integrated node set is the over-limit node of the real-time transmission path map.

[0051] The stepwise amplification factor of the previous node in the independent path is taken from the calculation result of the stepwise amplification factor of the previous node. The instantaneous transfer gain factor is taken from the processing result of the decoupling of transfer characteristics between nodes in the independent path. The current cumulative consumption of adsorbent is taken from the source tracing result of transfer gain loss based on the instantaneous transfer gain factor. The preset adsorbent saturation capacity is a pre-set fixed value. The in-phase superposition contribution coefficient is taken from the correlation analysis result of the instantaneous transfer gain factor and the wave transfer phase difference sequence. The preset waveform steepness factor is a pre-set fixed value. The wave energy ratio of the in-phase superposition section and the anti-phase cancellation section is taken from the comparison result of the wave energy of the two types of sections. The number of sub-paths contained in the in-phase superposition section is taken from the statistical result of the sub-paths of the in-phase superposition section. The normalized wave amplitude of the sub-path is taken from the normalization processing result of the wave amplitude of the corresponding sub-path.

[0052] This calculation method is used to quantify the stepwise amplification coefficient of the current node in an independent path. It consists of two parts: the first part reflects the cumulative transmission effect of the amplification coefficient of the previous node after correction by the current node gain and adsorbent consumption; the second part reflects the new fluctuation effect contributed by the in-phase superposition. It fully reflects the degree of stepwise cumulative amplification of parameter fluctuations among nodes in an independent path, providing a quantitative basis for advance risk prediction.

[0053] When the current cumulative consumption of adsorbent approaches the preset adsorbent saturation capacity, the cumulative transmission effect of the first part will gradually weaken. When the ratio of the fluctuation energy of the same-phase superposition section to the anti-phase cancellation section gradually increases, the same-phase superposition contribution effect of the second part will gradually strengthen. Overall, the stepwise amplification coefficient will gradually accumulate and change with the node transmission sequence, intuitively reflecting the amplification degree of parameter fluctuation on the independent path.

[0054] The beneficial effects include the ability to accurately decompose the parametric fluctuation sequence of independent paths to obtain fast-changing fluctuation components and slow-changing trend components, clarify the phase difference sequence of fluctuation transmission through phase comparison, and accurately identify the in-phase superposition segment and the anti-phase cancellation segment of fluctuations. Simultaneously, it can achieve precise matching of peak and trough amplitudes, decouple the instantaneous transmission gain factor between nodes, trace the current cumulative consumption of adsorbent, and analyze the in-phase superposition contribution coefficient. This enables precise quantification of the step-by-step cumulative effect of parametric fluctuation amplitude, obtaining a reliable step-by-step amplification coefficient. Based on this coefficient, the expected peak fluctuation value of the terminal node is accurately extrapolated, accurately marking the over-limit node, achieving advanced risk prediction in the transformer oil handling process, providing precise node basis for subsequent risk management, and making the risk prediction results more accurate and targeted.

[0055] The offset evaluation module 14 is used to perform high-frequency sampling of the adjacent downstream links of the over-limit node, and based on the sampling data, evaluate the degree of offset of the over-limit node to obtain the prediction accuracy of the over-limit node. In this embodiment of the invention, when the offset evaluation module performs high-frequency sampling on the adjacent downstream links of the over-limit node and evaluates the degree of offset of the over-limit node based on the sampled data to obtain the prediction accuracy of the over-limit node, it is specifically used for: Real-time parametric interruption capture is performed on the adjacent downstream links of the over-limit node to obtain the dense sampling sequence of the adjacent downstream links; Peak and valley extreme value retrieval is performed on the dense sampling sequence to obtain the fluctuation peak and valley records of the adjacent downstream links; The peak and valley times of the fluctuation peak and valley records are aligned and mapped with the risk occurrence time window of the over-limit node to obtain the pairing sequence of the over-limit node and the fluctuation peak and valley records. Based on the measured results of the out-of-limit nodes, the matching sequence is judged to obtain the prediction accuracy of the out-of-limit nodes.

[0056] For all adjacent downstream processing stages corresponding to the over-limit node, a unified high-frequency sampling time interval is set. According to this interval, the operating parameters of each downstream stage are continuously extracted in an interrupted manner. Each extraction completely retains the actual data of the parameter and the corresponding acquisition time. All parameter data extracted from each downstream stage are continuously and orderly organized according to the order of acquisition time, and finally a continuous sequence containing complete acquisition time information and parameter data of each stage is formed. This sequence is the dense sampling sequence of adjacent downstream stages.

[0057] By traversing all parameter data in the dense sampling sequence, the parameter values ​​of adjacent data points are compared one by one according to the acquisition time sequence. The peak points where the parameter values ​​change from rising to falling and the valley points where the parameter values ​​change from falling to rising are accurately identified. The parameter values ​​and acquisition times corresponding to all peak points and valley points are extracted. The numerical and temporal information of these peak and valley values ​​are systematically organized according to the acquisition time sequence. The complete information record formed after organization is the fluctuation peak and valley record of the adjacent downstream links.

[0058] The start and end times of the risk occurrence window of the over-limit node are clearly defined. A unified time axis is built based on this time range and clearly marked with time scale. The collection time corresponding to each peak and valley point in the fluctuation peak and valley records is extracted. These peak and valley times are accurately mapped to the risk occurrence time window range of the over-limit node according to the scale of the time axis. The peak and valley point information that has been matched on the time axis is correlated one-to-one with the risk occurrence time window information of the over-limit node. At the same time, the original parameter data of the peak and valley points are completely preserved. All the correlated information is organized into a continuous sequence according to the time axis. This sequence is the pairing sequence of the over-limit node and the fluctuation peak and valley records.

[0059] Extract relevant information on risk prediction of over-limit nodes and actual measurement results obtained from monitoring the actual operation of the node. Check the consistency of each related information in the paired sequence with the actual measurement results of the over-limit node in chronological order. Completely count the number of information in the paired sequence that matches the actual measurement results and the total number of information in the paired sequence. Determine the degree of conformity between the risk prediction results of the over-limit node and the actual operation based on the correspondence between the number of matching information and the total number of information. Mark and record the degree of conformity according to a unified standard. The result of the completed marking and recording is the prediction accuracy rate of the over-limit node.

[0060] The beneficial effects include enabling standardized high-frequency sampling of adjacent downstream links of the over-limit node, accurately acquiring dense sampling sequences and extracting fluctuation peak and valley records, forming paired sequences through time axis alignment mapping, completing compliance judgment by combining the measured results of the over-limit node, accurately obtaining the prediction accuracy rate, realizing a scientific assessment of the degree of prediction deviation of the over-limit node, providing real and reliable actual data basis for the subsequent iterative correction of the step-by-step amplification coefficient, and ensuring the rationality and pertinence of the coefficient correction.

[0061] The coefficient correction module 15 is used to iteratively correct the step-by-step amplification coefficient based on the prediction accuracy to obtain the optimized amplification coefficient of the step-by-step amplification coefficient. In this embodiment of the invention, when the coefficient correction module performs iterative correction of the step-by-step amplification coefficient based on the prediction accuracy to obtain the optimized amplification coefficient of the step-by-step amplification coefficient, it is specifically used for: The correlation between the step-by-step amplification coefficient and the prediction accuracy is analyzed to obtain the coefficient term to be corrected for the step-by-step amplification coefficient; Based on the correlation parameters of the stepwise amplification coefficient, reverse deviation compensation is performed on the coefficient term to be corrected to obtain the single correction value of the coefficient term to be corrected. Based on the single correction value, the coefficient to be corrected is incrementally updated, and the updated amplification coefficient is re-predicted for risk, so as to obtain the prediction accuracy update value of the updated amplification coefficient. The updated prediction accuracy values ​​are selectively retained to obtain the optimized amplification coefficient of the step-by-step amplification coefficient.

[0062] We sort out all the coefficient items included in the step-by-step amplification coefficient and the corresponding prediction accuracy data of the over-limit nodes. We match each coefficient item with the corresponding prediction accuracy one by one according to the node order of the independent path. We analyze and sort out the correlation characteristics between the coefficient item value and the prediction accuracy, identify the coefficient items that have a direct deviation related to the prediction accuracy, and systematically organize all the identified coefficient items according to the nodes and paths. The sorted set of coefficient items is the coefficient item to be corrected for the step-by-step amplification coefficient.

[0063] Extract the actual monitoring data of the operating parameters of each related link in the transformer oil handling corresponding to the coefficient item to be corrected, and comprehensively sort out the actual change characteristics and transmission status of the parameters of each related link. Compare the current value of the coefficient item to be corrected with the actual change characteristics of the parameters of the related links to clarify the direction and specific deviation between the two. According to the actual deviation, perform reverse numerical compensation and adjustment on the coefficient item to be corrected. Standardize and normalize the adjusted coefficient item value. The normalized coefficient item value is the single correction value of the coefficient item to be corrected.

[0064] The single correction value is assigned to the corresponding coefficient item to be corrected, completing the incremental numerical update of the coefficient item to be corrected. The updated coefficient item to be corrected is integrated with the uncorrected coefficient item to form a new amplification coefficient. Based on the original complete process and operation method of risk advance prediction, the risk of transformer oil handling operation parameters is re-predicted based on the updated amplification coefficient. The result of the risk re-prediction is recorded and the corresponding prediction accuracy rate is determined according to the original evaluation standard. The accuracy rate obtained by this determination is the updated prediction accuracy rate update value of the updated amplification coefficient.

[0065] Each time the amplification factor is updated, the predicted accuracy value is compared with the accuracy value obtained in the previous round. The set of amplification factors with the better predicted accuracy update value is retained. The incremental update of the coefficient to be corrected, the risk re-prediction based on the updated amplification factor, and the comparison and retention of the predicted accuracy are continuously performed until the predicted accuracy update value no longer changes. The set of amplification factors that is finally retained is the optimized amplification factor of the step-by-step amplification factor.

[0066] The beneficial effects include the ability to accurately analyze the correlation between the progressive amplification coefficient and the prediction accuracy, accurately locate the coefficient item to be corrected, and complete reverse deviation compensation by combining the actual monitoring data of the parameters in the related links, thus obtaining an accurate single correction value. Accuracy update values ​​are obtained through incremental updates and risk re-prediction, and coefficient iterative optimization is completed through continuous selection and retention, ultimately yielding an optimized amplification coefficient. This ensures that the amplification coefficient closely matches the actual operation of oilfield processing, improving coefficient accuracy and providing a reliable quantitative basis for subsequent parameter stability range reconstruction.

[0067] The interval reconstruction module 16 is used to perform threshold screening on the optimized amplification coefficient and to reconstruct the operating range of the associated fluctuation path of the screened coefficient to obtain the parameter stability interval of the associated fluctuation path. In this embodiment of the invention, when the interval reconstruction module performs threshold filtering on the optimized amplification coefficient and reconstructs the operating range of the associated fluctuation path of the filtered coefficient to obtain the parameter stability interval of the associated fluctuation path, it is specifically used for: By backtracking the historical response characteristics of the associated fluctuation path of the optimized amplification factor, the historical stability index of the associated fluctuation path is obtained. Based on the historical stability index, the path-specific boundary of the optimized amplification coefficient is delineated to obtain the differentiated screening threshold for each associated fluctuation path. Based on the differentiated screening threshold, the correlation fluctuation paths are classified by fluctuation sensitivity to obtain high-sensitivity paths and low-sensitivity paths of the correlation fluctuation paths. The operating parameter range of the high-sensitivity path is reconstructed through collaborative operation to obtain the parameter stability range of the associated fluctuation path.

[0068] When the interval reconstruction module performs collaborative operation range reconstruction on the operating parameter interval of the highly sensitive path to obtain the parameter stability interval of the associated fluctuation path, it is specifically used for: Based on the high-sensitivity path, the optimized amplification coefficients are normalized to obtain the sensitivity weight ratio of each optimized amplification coefficient. Based on the sensitivity weight ratio, the collaborative compression coefficient of the high-sensitivity path is calculated, wherein the formula for calculating the collaborative compression coefficient is: ; In the formula, For the first The collaborative compression coefficient of the high-sensitivity path, For the first The optimized amplification factor for a high-sensitivity path. For the first A highly sensitive path, The total number of the highly sensitive paths. For the first A highly sensitive path, For the wave propagation phase difference sequence, the first... The highly sensitive path and the first The transmission coupling strength of a highly sensitive path, For the first The optimized amplification factor for a highly sensitive path; Based on the cooperative compression coefficient, the original operating parameter range of the high-sensitivity path is nonlinearly compressed to obtain the compressed range of the high-sensitivity path. The compressed interval and the original interval of the low-sensitivity path are reconstructed by boundary nesting to obtain the parameter stable interval of the associated fluctuation path.

[0069] All associated fluctuation paths corresponding to the optimized amplification factor are sorted out, historical operation data of each path are extracted for the whole time period, and the parameter response characteristics of each path under different operating conditions are fully collected, covering the frequency of fluctuation, duration of fluctuation, and magnitude of parameter change. The historical response characteristics of each path are systematically organized and analyzed, and quantitative labeling is performed based on the stable performance of each path in historical operation. The quantitative result after labeling is the historical stability index of the associated fluctuation path.

[0070] The historical stability index of each associated fluctuation path is bound to the corresponding optimization amplification coefficient. Based on the actual operating characteristics of each path, a unique numerical boundary range is defined for the optimization amplification coefficient of each associated fluctuation path. This numerical boundary range serves as the screening criterion for the optimization amplification coefficient of the corresponding path. The screening criteria corresponding to each of all associated fluctuation paths are systematically organized, and the organized path-specific screening criteria are the differentiated screening thresholds for each associated fluctuation path.

[0071] The actual values ​​of the optimization amplification coefficients of each associated fluctuation path are compared one by one with the corresponding differential screening thresholds. The paths are distinguished according to the width of the range of the differential screening threshold values. The associated fluctuation paths with narrow screening threshold values ​​are grouped into one category, and the associated fluctuation paths with wide screening threshold values ​​are grouped into another category. The set of paths grouped into the former category is the high-sensitivity path of the associated fluctuation path, and the set of paths grouped into the latter category is the low-sensitivity path of the associated fluctuation path.

[0072] Extract the optimization amplification coefficient values ​​corresponding to all high-sensitivity paths, analyze the overall distribution range of each value, and convert the optimization amplification coefficient values ​​of each high-sensitivity path according to a unified standard so that all converted values ​​are within the same fixed value range. Calculate the proportion of the optimization amplification coefficient of each high-sensitivity path in the total value of all high-sensitivity path coefficients based on the converted values, and accurately label and record each proportion. The result after labeling and recording is the sensitivity weight proportion of each optimization amplification coefficient.

[0073] By combining the sensitivity weight ratio of each high-sensitivity path with the actual operating characteristics of the corresponding path, the interrelationship and influence between the high-sensitivity paths are comprehensively analyzed. Based on the sensitivity weight ratio of each path and the interrelationship and influence between paths, a corresponding compression ratio value is determined for each high-sensitivity path. This value can reflect the compression characteristics of the corresponding path after collaboration with other high-sensitivity paths. The determined compression ratio values ​​are the collaborative compression coefficients of each high-sensitivity path.

[0074] Extract the original operating parameter ranges for each high-sensitivity path, clarify the specific values ​​of the upper and lower boundaries of each original operating parameter range, match and apply the cooperative compression coefficients corresponding to each high-sensitivity path to the original operating parameter range of the corresponding path, and nonlinearly adjust the upper and lower boundary values ​​of the original operating parameter ranges according to the proportion corresponding to the cooperative compression coefficients. The adjustment process conforms to the actual operating parameter change characteristics of the path. The adjusted boundary values ​​are normalized to form a new parameter range. The normalized new parameter range is the compressed range of the high-sensitivity path.

[0075] Extract the compressed intervals of all high-sensitivity paths and the original intervals of all low-sensitivity paths, and clarify the upper and lower boundary values ​​and overall coverage of each interval. Combined with the actual operation requirements of transformer oil handling, arrange all compressed intervals and original intervals in a hierarchical nested manner to ensure that the boundaries of each interval are mutually compatible and without numerical conflicts. Integrate all the nested intervals to form a complete parameter interval system that includes all related fluctuation paths and is adapted to the actual operation of oil handling. This system is the parameter stability interval of related fluctuation paths.

[0076] No. The optimized amplification factor for each high-sensitivity path is taken from the optimized amplification factor obtained after iterative correction of the successive amplification factors. The total number of high-sensitivity paths is taken from the number of high-sensitivity paths obtained after classifying the fluctuation sensitivity. The optimized amplification factor for the high-sensitivity path is taken from the optimized amplification factor obtained after iterative correction of the successive amplification factors. The highly sensitive path and the first The transmission coupling strength of the highly sensitive path is derived from the analysis results of the transmission coupling relationship between paths in the wave transmission phase difference sequence.

[0077] This calculation method is used to obtain the first... The cooperative compression coefficient of the high-sensitivity path consists of two parts. The first part reflects the... The proportion of the high-sensitivity path optimization amplification factor in the total amplification factor of all high-sensitivity paths is shown in the second part, which reflects the exclusion of the first path. After the path itself, the proportion of the sum of the products of the coupling strength and the optimization amplification factor of other paths in the sum of the optimization amplification factors of all high-sensitivity paths is corrected. The resulting collaborative compression factor is used to perform nonlinear compression on the original operating parameter range of the high-sensitivity path, providing a basis for reconstructing the parameter stability range.

[0078] When the When the proportion of the optimization amplification factor of the first high-sensitivity path in the sum of the optimization amplification factors of all high-sensitivity paths increases, the value of the first part will increase. When the propagation coupling strength between a high-sensitivity path and other high-sensitivity paths increases, the value of the correction term in the second part decreases, and the overall cooperative compression coefficient increases with the increase in the third part. The compression factor increases with the increase of the proportion of the path optimization amplification factor, and decreases with the increase of the coupling strength between paths, which directly reflects the combined effect of the path's own weight and the coupling effect on the degree of compression.

[0079] The beneficial effects include the ability to comprehensively trace the historical response characteristics of associated fluctuation paths and obtain a historical stability index. Based on this index, differentiated screening thresholds are set for each path, enabling precise classification of the fluctuation sensitivity of associated fluctuation paths and distinguishing between high-sensitivity and low-sensitivity paths. Simultaneously, the optimization amplification coefficient of high-sensitivity paths can be normalized to obtain the sensitivity weight ratio. Based on this, a collaborative compression coefficient is determined, and nonlinear compression of the original interval of high-sensitivity paths is performed. The compressed interval is then hierarchically nested with the original interval of low-sensitivity paths, ultimately integrating to form a parameter stability interval adapted to the actual operational requirements of transformer oil handling. This interval adapts to the operating characteristics of all associated fluctuation paths, providing a precise and realistic parameter range basis for subsequent proactive oil handling control.

[0080] The control and archiving module 17 is used to actively control the transformer oil condition based on the parameter stability range, and record and integrate the control process to obtain the traceability file of the transformer oil condition; In this embodiment of the invention, when the control and archiving module performs active control of the transformer oil condition based on the parameter stability range and records and integrates the control process to obtain the traceability file of the transformer oil condition, it is specifically used for: Based on the stable range of the parameters, real-time control commands are issued for the transformer oil, resulting in the execution sequence of the parameter control mechanism. Based on the sequence of actions, the operating conditions of the transformer oil purification process are adjusted to obtain the operating condition adjustment record of the purification process. The operating condition adjustment record and the stable interval are time-aligned and encapsulated to obtain the associated mapping file of the operating condition adjustment record and the stable interval; Based on the associated mapping file, the entire process record of the transformer oil service is traceable and bound to a traceability identifier to obtain the traceability file of the transformer oil service.

[0081] Based on the clearly defined operating parameter ranges of each associated fluctuation path within the parameter stability range, the operating parameter values ​​of each processing stage of transformer oil are monitored in real time. The monitored parameter values ​​are compared one by one with the corresponding value ranges within the parameter stability range. For processing stages that exceed the parameter stability range, control instructions adapted to that stage are generated. The control instructions are accurately issued to the corresponding parameter control mechanisms according to the execution sequence of the control operations. At the same time, all specific operational actions performed by the parameter control mechanisms after receiving the instructions are fully recorded. All operational actions are organized into a continuous action sequence according to time sequence and execution logic. This sequence is the execution action sequence of the parameter control mechanism.

[0082] Extract each specific action from the sequence of actions, clarify the corresponding transformer oil purification process step, specific operation method, and preset execution time for each action, and carry out targeted operating condition adjustments for the corresponding steps of the purification process in chronological order according to the sequence of actions. During the adjustment process, record the original operating conditions before the adjustment, the actual adjustment operation content, the final operating conditions after the adjustment, and the specific time of adjustment completion for each step. Systematically sort out all the adjustment-related information of the purification process steps according to the time sequence and step affiliation. The complete information record after sorting out is the operating condition adjustment record of the purification process.

[0083] A unified timeline covering the entire transformer oil handling process is established, with clear and continuous time scales marked on the timeline. The actual execution time corresponding to each adjustment item in the operating condition adjustment record is extracted, and the official effective time corresponding to each parameter range in the parameter stability interval is extracted. The execution time of the operating condition adjustment record and the effective time of the parameter stability interval are precisely matched according to the timeline scale. Each operating condition adjustment item is associated and bound with the corresponding parameter range in the parameter stability interval. All information with completed time alignment and content association is encapsulated in chronological order. The encapsulated complete information set is the association mapping file between the operating condition adjustment record and the stability interval.

[0084] A unique traceability identifier is generated for the entire process of transformer oil handling for this batch. All information in the associated mapping file is extracted, and the entire process record of transformer oil handling from the start to the end of the process is collected, covering the operating parameter monitoring data, basic processing operation content, start and end time of each process, etc. The unique traceability identifier is comprehensively and deeply bound to the associated mapping file and the entire process record, ensuring that every processing record and control-related information can accurately correspond to the traceability identifier. All information that has been bound to the traceability identifier is integrated and sorted out in an orderly manner to form a complete file containing the traceability identifier, the entire process record, the operating condition adjustment record, and the parameter stability range association information. This file is the traceability file for transformer oil handling.

[0085] The beneficial effects include the ability to accurately issue control commands based on the stable range of parameters, forming a standardized sequence of execution actions, completing the adjustment of the purification process conditions and recording them completely, establishing a correlation mapping between the adjustment of conditions and the stable range through time-series alignment and encapsulation, and finally binding a unique traceability identifier to form a complete traceability file. This achieves the accuracy and full-process traceability of transformer oil control, ensures that the quality of oil treatment is traceable, and provides a reliable basis for subsequent quality control and problem tracing.

[0086] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0087] This application embodiment can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence is the theory, method, technology, and application platform that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0088] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. A fully intelligent monitoring and quality traceability platform for transformer oil handling, characterized in that: The platform includes an architecture mapping module, a path tracing module, a risk prediction module, an offset assessment module, a coefficient correction module, an interval reconstruction module, and a control and archiving module, wherein: The architecture mapping module is used to perform flow-direction association mapping on the operating parameters of transformer oil to obtain the transmission architecture of the operating parameters; The path tracing module is used to dynamically track the running parameters based on the transmission architecture to obtain a real-time transmission path diagram of the running parameters. The risk prediction module is used to quantify the cumulative effect of the parameter fluctuation amplitude of the real-time transmission path map, obtain the step-by-step amplification coefficient of the real-time transmission path map, and based on the step-by-step amplification coefficient, perform risk prediction on the real-time transmission path map to obtain the over-limit nodes of the real-time transmission path map. The offset evaluation module is used to perform high-frequency sampling of the adjacent downstream links of the over-limit node, and based on the sampling data, evaluate the degree of offset of the over-limit node to obtain the prediction accuracy of the over-limit node. The coefficient correction module is used to iteratively correct the step-by-step amplification coefficient based on the prediction accuracy to obtain the optimized amplification coefficient of the step-by-step amplification coefficient. The interval reconstruction module is used to perform threshold filtering on the optimized amplification coefficient and to reconstruct the operating range of the associated fluctuation path of the filtered coefficient to obtain the parameter stability interval of the associated fluctuation path. The control and archiving module is used to actively control the transformer oil condition based on the stable range, and record and integrate the control process to obtain the traceability file of the transformer oil condition.

2. The intelligent monitoring and quality traceability platform for the entire transformer oil treatment process as described in claim 1, characterized in that, When the architecture mapping module performs flow-direction association mapping on the operating parameters of transformer oil services to obtain the transmission architecture of the operating parameters, it is specifically used for: The operating parameters of the transformer oil service at each processing stage are synchronously captured on a time axis to obtain a time-series capture record of the operating parameters; Based on the time-series capture records, a sliding window matching is performed on the running parameters to obtain the time delay correlation of the running parameters; Based on the aforementioned time delay correlation, the forward and reverse flow directions of the operating parameters are determined to obtain the flow direction identifier of the operating parameters; Based on the flow direction identifier, the topology of the running parameters is reconstructed to obtain the transmission architecture of the running parameters.

3. The intelligent monitoring and quality traceability platform for the entire transformer oil treatment process as described in claim 1, characterized in that, When the path tracing module performs dynamic tracing of the running parameters based on the transmission architecture to obtain a real-time transmission path graph of the running parameters, it is specifically used for: Based on the aforementioned transmission architecture, the operational parameters are subjected to sliding window fluctuation identification to obtain fluctuation segments of the operational parameters. The start and end times of the wave segment are interleaved and compared to obtain the wave propagation direction of the wave segment; Based on the wave propagation direction, a similarity measurement is performed on the waveform profile of the wave segment to obtain the waveform matching degree of the wave segment; Based on the waveform matching degree, the wave segments are merged and associated across nodes to obtain the wave propagation trajectory segments of the running parameters; Based on the aforementioned transmission architecture, the wave propagation trajectory segments are stitched together to obtain a real-time transmission path diagram of the operational parameters.

4. The intelligent monitoring and quality traceability platform for the entire transformer oil treatment process as described in claim 1, characterized in that, The risk prediction module, when performing step-by-step cumulative effect quantification of the parameter fluctuation amplitude of the real-time transmission path diagram to obtain the step-by-step amplification coefficient of the real-time transmission path diagram, and based on the step-by-step amplification coefficient, performing risk advance prediction on the real-time transmission path diagram to obtain the over-limit nodes of the real-time transmission path diagram, is specifically used for: The parametric fluctuation sequence of the independent path in the real-time transmission path diagram is subjected to fluctuation mode decomposition to obtain the fast fluctuation component and slow trend component of the independent path. Based on the rapidly changing wave components, the wave propagation phase of the independent path is compared with the lead-lag phase to obtain the wave propagation phase difference sequence of the independent path. Based on the wave propagation phase difference sequence, coherence is determined for the independent path to obtain the wave in-phase superposition segment and the wave out-of-phase cancellation segment of the independent path. The parametric fluctuation amplitude of the superimposed wave segment is recursively accumulated node by node to obtain the step-by-step amplification coefficient of the independent path; Based on the stepwise amplification coefficient, the fluctuation amplitude of the terminal node of the independent path is extrapolated to obtain the expected fluctuation peak value of the terminal node. The end nodes whose expected fluctuation peak exceeds the preset risk threshold are marked as over-limit nodes in the real-time transmission path diagram.

5. The intelligent monitoring and quality traceability platform for the entire transformer oil treatment process as described in claim 4, characterized in that, When the risk prediction module performs node-by-node recursive accumulation of the parametric fluctuation amplitude of the overlapping fluctuation segment to obtain the step-by-step amplification coefficient of the independent path, it is specifically used for: Amplitude correlation matching is performed on the parametric fluctuation amplitude of the superimposed wave phase segment to obtain the wave peak-valley pairing sequence of the superimposed wave phase segment; Based on the peak-valley pairing sequence, the inter-node transmission characteristics of the independent path are decoupled to obtain the instantaneous transmission gain factor of the independent path. Based on the instantaneous transfer gain factor, the transfer gain loss of the independent path is traced to obtain the current cumulative adsorbent consumption of the independent path. The correlation analysis between the instantaneous transfer gain factor and the wave transfer phase difference sequence is performed to obtain the in-phase superposition contribution coefficient of the independent path; Based on the instantaneous transfer gain factor and the in-phase superposition contribution coefficient, the step-by-step amplification factor of the independent path is calculated, wherein the formula for calculating the step-by-step amplification factor is: ； In the formula, For the independent path, the first The amplification factor of each node at each level, For the independent path, the first The amplification factor of each node at each level, For the independent path, the first The instantaneous transfer gain factor of each node, The node number of the independent path. For the independent path, the first The current cumulative adsorbent consumption of each node. For the independent path, the first The preset adsorbent saturation capacity of each node, For the independent path, the first The in-phase superposition contribution coefficient of each node, The preset waveform steepness factor, For the independent path, the first The ratio of the wave energy of the in-phase superposition segment to the out-of-phase cancellation segment at each node. It is a natural constant. For the independent path, the first The number of sub-paths contained within the in-phase overlay segment at each node. For the independent path, the first The sequence number of the sub-path within the in-phase overlapping section at each node. For the independent path, the first At the node, the first Normalized fluctuation amplitude of the strip path.

6. The intelligent monitoring and quality traceability platform for the entire transformer oil treatment process as described in claim 1, characterized in that, The offset evaluation module, when performing high-frequency sampling of the adjacent downstream links of the out-of-limit node and evaluating the degree of offset of the out-of-limit node based on the sampled data to obtain the prediction accuracy of the out-of-limit node, is specifically used for: Real-time parametric interruption capture is performed on the adjacent downstream links of the over-limit node to obtain the dense sampling sequence of the adjacent downstream links; Peak and valley extreme value retrieval is performed on the dense sampling sequence to obtain the fluctuation peak and valley records of the adjacent downstream links; The peak and valley times of the fluctuation peak and valley records are aligned and mapped with the risk occurrence time window of the over-limit node to obtain the pairing sequence of the over-limit node and the fluctuation peak and valley records. Based on the measured results of the out-of-limit nodes, the matching sequence is judged to obtain the prediction accuracy of the out-of-limit nodes.

7. The intelligent monitoring and quality traceability platform for the entire transformer oil treatment process as described in claim 1, characterized in that, When the coefficient correction module performs iterative correction of the step-by-step amplification coefficient based on the prediction accuracy to obtain the optimized amplification coefficient, it is specifically used for: The correlation between the step-by-step amplification coefficient and the prediction accuracy is analyzed to obtain the coefficient term to be corrected for the step-by-step amplification coefficient; Based on the correlation parameters of the stepwise amplification coefficient, reverse deviation compensation is performed on the coefficient term to be corrected to obtain the single correction value of the coefficient term to be corrected. Based on the single correction value, the coefficient to be corrected is incrementally updated, and the updated amplification coefficient is re-predicted for risk, so as to obtain the prediction accuracy update value of the updated amplification coefficient. The updated prediction accuracy values ​​are selectively retained to obtain the optimized amplification coefficient of the step-by-step amplification coefficient.

8. The intelligent monitoring and quality traceability platform for the entire transformer oil treatment process as described in claim 1, characterized in that, When the interval reconstruction module performs threshold filtering on the optimized amplification coefficient and reconstructs the operating range of the associated fluctuation path of the filtered coefficient to obtain the parameter stability interval of the associated fluctuation path, it is specifically used for: By backtracking the historical response characteristics of the associated fluctuation path of the optimized amplification factor, the historical stability index of the associated fluctuation path is obtained. Based on the historical stability index, the path-specific boundary of the optimized amplification coefficient is delineated to obtain the differentiated screening threshold for each associated fluctuation path. Based on the differentiated screening threshold, the correlation fluctuation paths are classified by fluctuation sensitivity to obtain high-sensitivity paths and low-sensitivity paths of the correlation fluctuation paths. The operating parameter range of the high-sensitivity path is reconstructed through collaborative operation to obtain the parameter stability range of the associated fluctuation path.

9. The intelligent monitoring and quality traceability platform for the entire transformer oil treatment process as described in claim 8, characterized in that, When the interval reconstruction module performs collaborative operation range reconstruction on the operating parameter interval of the highly sensitive path to obtain the parameter stability interval of the associated fluctuation path, it is specifically used for: Based on the high-sensitivity path, the optimized amplification coefficients are normalized to obtain the sensitivity weight ratio of each optimized amplification coefficient. Based on the sensitivity weight ratio, the collaborative compression coefficient of the high-sensitivity path is calculated, wherein the formula for calculating the collaborative compression coefficient is: ； In the formula, For the first The collaborative compression coefficient of the high-sensitivity path, For the first The optimized amplification factor for a high-sensitivity path. For the first A highly sensitive path, The total number of the highly sensitive paths. For the first A highly sensitive path, For the wave propagation phase difference sequence, the first... The highly sensitive path and the first The transmission coupling strength of a highly sensitive path, For the first The optimized amplification factor for a highly sensitive path; Based on the cooperative compression coefficient, the original operating parameter range of the high-sensitivity path is nonlinearly compressed to obtain the compressed range of the high-sensitivity path. The compressed interval and the original interval of the low-sensitivity path are reconstructed by boundary nesting to obtain the parameter stable interval of the associated fluctuation path.

10. The intelligent monitoring and quality traceability platform for the entire transformer oil treatment process as described in claim 1, characterized in that, When the control and archiving module performs active control of the transformer oil condition based on the parameter stability range and records and integrates the control process to obtain the traceability file of the transformer oil condition, it is specifically used for: Based on the stable range of the parameters, real-time control commands are issued for the transformer oil, resulting in the execution sequence of the parameter control mechanism. Based on the sequence of actions, the operating conditions of the transformer oil purification process are adjusted to obtain the operating condition adjustment record of the purification process. The operating condition adjustment record and the stable interval are time-aligned and encapsulated to obtain the associated mapping file of the operating condition adjustment record and the stable interval; Based on the associated mapping file, the entire process record of the transformer oil service is traceable and bound to a traceability identifier to obtain the traceability file of the transformer oil service.

Images