Method and system for predicting and locating fault development stage based on multiple parameters of partial discharge

Abstract

The disclosure provides a fault development stage prediction positioning method and system based on partial discharge multi-parameters, relates to the technical field of power equipment state monitoring and fault positioning, and comprises the following steps: firstly, a multi-parameter synchronous acquisition experiment platform is built, a multi-parameter characteristic item set is constructed, then an improved Apriori algorithm with sliding window time sequence confidence constraint is introduced to mine the characteristic item set, the correlation strength between each characteristic quantity and the discharge stage is quantitatively evaluated, strong correlation characteristics are screened, and a discharge development stage deduction rule library is established; in the positioning link, a plurality of groups of candidate discharge source coordinates are independently solved, the candidate coordinates and the reference propagation time thereof constitute a four-dimensional data point set, and after outliers are removed by using a density-based clustering algorithm, the clustering center is taken as the final discharge source position output. Compared with the traditional single physical quantity positioning means, the disclosure effectively solves the problems of fuzzy mapping of partial discharge signal characteristics and fault degree and limited positioning accuracy.

Description

Technical Field

[0001] This disclosure relates to the field of power equipment condition monitoring and fault location technology, specifically to a method and system for predicting and locating fault development stages based on multi-parameter partial discharge. Background Technology

[0002] The statements in this section are merely background information relating to this disclosure and do not necessarily constitute prior art.

[0003] Partial discharge is a significant cause of insulation degradation in power transformers, gradually evolving into spark discharge and even high-energy arc discharge over time, ultimately leading to transformer combustion and explosion accidents. Therefore, detecting and analyzing partial discharge signals to assess the insulation condition of transformers is crucial for ensuring the safe and stable operation of power systems. However, in practical engineering applications, current transformer partial discharge detection and location technologies still have the following limitations: (i) The mapping relationship between partial discharge characteristic parameters and the severity of equipment insulation faults is ambiguous. Traditional detection methods often rely on amplitude thresholds or phase spectra of a single detection means for qualitative judgment, making it difficult to construct accurate predictive models from multi-dimensional signal characteristics to apparent discharge quantity and even discharge development stages. This makes it impossible for maintenance personnel to quantitatively assess and provide trend warnings of the degree of insulation degradation inside the equipment based on monitoring data.

[0004] (II) The spatial positioning accuracy of the internal discharge source of the transformer remains insufficient. Although UHF, ultrasonic, and optical detection technologies have made significant progress in recent years, positioning methods based on a single physical quantity have inherent limitations. For example, UHF electromagnetic waves are easily reflected and diffracted by metal components during propagation; ultrasonic signals exhibit significant multipath propagation in complex media and are easily interfered with by mechanical vibration; optical detection is limited by its effective line-of-sight range, making it difficult to achieve large-area coverage. Single detection methods cannot effectively overcome measurement errors introduced by factors such as strong electromagnetic interference, mechanical vibration, and obstruction by internal components, resulting in a large deviation between the estimated coordinates of the discharge source and the actual fault location.

[0005] (III) In recent years, the concept of multi-parameter fusion detection for partial discharge has provided a new approach to solving the above problems. By simultaneously acquiring electromagnetic, acoustic, and optical signals, the complementary advantages of different physical quantities in spatiotemporal propagation characteristics are fully utilized. However, current multi-parameter analysis methods still mainly rely on simple feature comparison and lack in-depth mining and effective utilization of the correlation rules of multi-source heterogeneous signals. Summary of the Invention

[0006] To address the aforementioned issues, this disclosure proposes a method and system for predicting and locating fault development stages based on multi-parameter partial discharge. Based on multi-parameter partial discharge data of transformers, an improved Apriori algorithm with temporal confidence constraints is introduced to mine feature association rules and establish a rule base for predicting discharge development stages. Using UHF electrical signals as a time reference, the arrival time differences between light and electricity and between different UHF signals are calculated separately, and multiple sets of candidate discharge source coordinates are independently solved. The candidate coordinates and their reference propagation times form a four-dimensional spatiotemporal data point set. After removing outliers using density clustering, the cluster center is used as the final spatial location of the discharge source, achieving stage assessment and high-precision location of transformer insulation defects.

[0007] According to some embodiments, the present disclosure adopts the following technical solutions: Fault development stage prediction and localization methods based on multi-parameter partial discharge include: Acquire multi-parameter signals of partial discharge in transformers; Based on the multi-parameter signal of partial discharge in transformers, a multi-parameter feature set is constructed; An improved Apriori association rule algorithm is used to mine multi-parameter feature itemsets, quantify the correlation strength between each parameter feature and the discharge stage, and select strongly correlated features. These features are then compared with a preset fault development stage prediction rule base to obtain the current fault development stage. A global time reference is established, and the optical-electric arrival time difference between each optical sensor and the global time reference, as well as the arrival time difference between each ultra-high frequency sensor, are calculated respectively. The time difference of arrival (TDOA) positioning algorithm is used to independently calculate the spatial coordinates of the discharge source under different sensor combinations, and obtain multiple candidate positioning results and reference propagation time. A four-dimensional spatial-temporal data point set is constructed based on multiple candidate positioning results and reference propagation time. A density-based clustering algorithm is used to perform cluster analysis on the four-dimensional data point set. After removing outliers, the coordinates of the cluster center are used as the final spatial location of the discharge source and output.

[0008] As one embodiment, acquiring the multi-parameter signal of partial discharge in a transformer includes: A multi-parameter synchronous acquisition experimental platform for partial discharge of transformers was built. The location and type of partial discharge defects were preset at different locations on the transformer. Pulse current signals, ultra-high frequency electromagnetic wave signals, ultrasonic signals and optical signals generated by partial discharge under multiple power frequency cycles were collected and recorded.

[0009] As one embodiment, the locations and types of the pre-set partial discharge defects include: (a) surface discharge defects of oil-paper insulation at the end of high-voltage windings: conductive contaminants are pre-placed on the surface of the insulating paperboard at the end of the high-voltage windings to simulate surface creepage; (b) air gap discharge defects between turns: multiple layers of insulating paper are embedded between adjacent winding coils and air gaps are reserved to simulate air gap discharge inside the inter-turn insulation; (c) floating potential discharge defects of metal components: unreliably grounded metal components are set at the core clamps to simulate floating potential discharge; (d) free metal particle defects in insulating oil: tiny metal shavings are placed at the bottom of the transformer tank or in the winding oil passages to simulate discharge caused by free particles in the oil; (e) discharge defects of high-voltage leads to the tank tip: thin copper wires are tied to the surface of the high-voltage leads so that their ends point towards the tank wall to simulate high-voltage conductor tip discharge to ground.

[0010] As one embodiment, the construction of a multi-parameter feature itemset based on the multi-parameter signal of transformer partial discharge includes: Using the power frequency cycle as the basic unit, the maximum amplitude, average amplitude, and number of discharge pulses of ultrasonic, ultra-high frequency, and optical signals are extracted within each cycle; for the pulse current method, the maximum apparent discharge quantity within each power frequency cycle is calculated. Discretize the multi-parameter features to construct a discrete encoding scheme for the multi-parameter feature set of partial discharge; By using discretization encoding, the original continuous, high-dimensional multi-physical quantity waveform data is converted into structured itemsets, resulting in multi-parameter feature itemsets.

[0011] As one embodiment, the method of mining multi-parameter feature itemsets using an improved Apriori association rule algorithm, quantifying the correlation strength between each parameter feature and the discharge stage, and selecting strongly correlated features includes: The multi-parameter feature itemset is grouped according to the defect type label to form a sub-itemet corresponding to each preset insulation defect; Perform frequent itemset mining within each sub-itemset, and count the co-occurrence frequency of any feature item in the sub-itemset; Based on the co-occurrence frequency, the conditional support and conditional confidence of the feature terms under the defect type label for the discharge stage are calculated, and the conditional lift is further calculated. A sliding window time series confidence assessment is introduced to evaluate the temporal stability of the association rules, quantify the association strength between each feature and the discharge stage, and select the feature with the most significant association as the strongly associated feature.

[0012] As one embodiment, the method of introducing a sliding window temporal confidence assessment to evaluate the temporal stability of association rules includes: Arrange the items in the sub-item set in chronological order of collection time, and set the window size to [value missing]. KGiven a continuous power frequency cycle, define a period window. For candidate rules, calculate their value on the [number]th [period]. i Local confidence level within a window; Simultaneously, the dynamic characteristics of the trend enhancement factor capturing rule confidence increasing over time are defined, and the final time series confidence is a weighted synthesis of the window stability term and the trend enhancement term.

[0013] As one embodiment, the quantification of the correlation strength between each characteristic quantity and the discharge stage, and the selection of the characteristic quantity with the most significant correlation as the strongly correlated characteristic quantity, includes: Define the overall correlation strength between characteristic terms and the discharge stage under defect type conditions; Based on the comprehensive correlation strength, all feature items in the current defect subset are sorted in descending order, and feature items with the highest intensity value or exceeding the preset intensity threshold are selected as the strongly correlated feature quantity with the largest correlation with the target discharge stage under this defect type. Based on this, discharge stage prediction rules are formed and stored in the rule base.

[0014] As one embodiment, the step of setting a global time reference and calculating the optical-electrical arrival time difference between each optical sensor and the global time reference, as well as the arrival time difference between each ultra-high frequency sensor, includes: Based on the relationship that signal energy is proportional to the square of signal amplitude, the signal voltage amplitude function is transformed into an energy accumulation function. Then, the average power of the signal is introduced. Considering that the signal amplitude of noise is very small, the average energy is much lower than the average power before the partial discharge signal arrives. Therefore, the energy curve will gradually decrease before the partial discharge signal arrives. When the partial discharge signal arrives, the signal amplitude is very large. At this time, the average energy is higher than the average power, and the energy curve stops decreasing and begins to gradually rise, thus generating an inflection point of minimum value. The time corresponding to this inflection point is regarded as the arrival time of the sensor's partial discharge signal.

[0015] As one embodiment, the step of independently calculating the spatial coordinates of the discharge source under different sensor combinations using the time difference of arrival (TDOA) positioning algorithm includes: Define the equivalent wave speeds of electromagnetic waves and light; based on the signal's movement from the partial discharge source to the... i The time difference between the individual sensor and the reference sensor is used to establish the spherical equation: The particle swarm optimization algorithm was used to solve the problem, and multiple candidate localization results were obtained.

[0016] As one embodiment, the construction of a four-dimensional space-time data point set based on multiple candidate localization results and a reference propagation time includes: Four-dimensional data points are constructed based on the candidate positioning results obtained independently by optical-electronic positioning and UHF positioning methods, as well as the reference propagation time estimates corresponding to different sensor combinations for each method.

[0017] As one embodiment, the density-based clustering algorithm performs cluster analysis on the four-dimensional data point set, removes outliers, and outputs the final discharge power source spatial location using the cluster center coordinates, including: Based on a four-dimensional data point set, a weighted Euclidean distance is defined between any two points; The density-based DBSCAN clustering algorithm is adopted to cluster the four-dimensional data point set by setting the neighborhood radius and the minimum number of neighborhood points. Isolated points that do not belong to any cluster are marked as outliers and removed. For the maximum density cluster, its cluster center is calculated by the arithmetic mean of all points in the cluster, and finally used as the output of the spatial location of the fused discharge source.

[0018] According to some embodiments, the present disclosure adopts the following technical solutions: A fault development stage prediction and location system based on multi-parameter partial discharge includes: The signal acquisition module is used to acquire multi-parameter signals of partial discharge in the transformer. The feature extraction module is used to construct a multi-parameter feature set based on the multi-parameter signal of partial discharge of transformer; The association mining and prediction module is used to mine multi-parameter feature itemsets using the improved Apriori association rule algorithm, quantify the association strength between each parameter feature and the discharge stage, and screen out strongly associated features. These features are then compared with the preset fault development stage prediction rule library to obtain the current fault development stage. The time difference calculation module is used to set a global time reference and calculate the optical-electric arrival time difference between each optical sensor and the global time reference, as well as the arrival time difference between each ultra-high frequency sensor. An independent calculation module is used to independently calculate the spatial coordinates of the discharge source under different sensor combinations using the time difference of arrival (TDOA) positioning algorithm, and obtain multiple candidate positioning results and reference propagation time. The fault location module is used to construct a four-dimensional space-time data point set based on multiple candidate location results and reference propagation time. It uses a density-based clustering algorithm to perform cluster analysis on the four-dimensional data point set, removes outliers, and outputs the final discharge power source spatial location using the coordinates of the cluster center.

[0019] According to some embodiments, the present disclosure adopts the following technical solutions: A computer program product includes a computer program that, when executed by a processor, implements the fault development stage prediction and localization method based on partial discharge multi-parameters.

[0020] According to some embodiments, the present disclosure adopts the following technical solutions: A non-transitory computer-readable storage medium is provided for storing computer instructions, which, when executed by a processor, implement the aforementioned fault development stage prediction and location method based on multi-parameter partial discharge.

[0021] According to some embodiments, the present disclosure adopts the following technical solutions: An electronic device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the fault development stage prediction and location method based on partial discharge multi-parameters.

[0022] Compared with the prior art, the beneficial effects of this disclosure are as follows: The fault development stage prediction and localization method based on multi-parameter partial discharge disclosed herein achieves accurate prediction of the discharge development stage of transformer insulation defects and high-precision localization of the discharge source by introducing an improved Apriori association rule algorithm to deeply mine the multi-parameter feature itemset and adopting a four-dimensional spatiotemporal density clustering localization strategy that integrates the reference propagation time.

[0023] The fault development stage prediction and location method based on partial discharge multi-parameters disclosed herein improves the Apriori algorithm by introducing a sliding window time sequence confidence constraint within a continuous power frequency cycle. This enables the quantitative screening of feature combinations that have a stable and strong correlation with each discharge stage under a specific defect type from a large number of multi-parameter features, effectively solving the problem of fuzzy mapping relationship between partial discharge feature parameters and the severity of equipment insulation faults in traditional methods.

[0024] The fault development stage prediction and location method based on partial discharge multi-parameters disclosed herein constructs a four-dimensional data point set by independently solving the candidate spatial coordinates obtained by different location methods and their corresponding reference propagation time in the location stage. A density-based clustering algorithm is used for spatiotemporal joint clustering. By introducing time dimension constraints, outliers caused by multipath propagation, local convergence anomalies and other reasons are effectively eliminated. The method makes full use of the complementary information of different physical quantity location results in spatiotemporal consistency.

[0025] The fault development stage prediction and location method based on multi-parameter partial discharge disclosed herein can fully adapt to the complex insulation structure and multi-physics field propagation environment inside the transformer. While improving the reliability of partial discharge fault development stage prediction, it significantly improves the accuracy of spatial location of the discharge source. It has important engineering application value and promotion prospects for ensuring the safe and stable operation of power transformers. Attached Figure Description

[0026] The accompanying drawings, which form part of this disclosure, are used to provide a further understanding of this disclosure. The illustrative embodiments of this disclosure and their descriptions are used to explain this disclosure and do not constitute an undue limitation of this disclosure.

[0027] Figure 1 This is a flowchart illustrating the implementation of the fault development stage prediction and location method based on multi-parameter partial discharge in this embodiment of the present disclosure. Figure 2 The energy accumulation curve of the U2 signal from the ultra-high frequency sensor is shown in the embodiment of this disclosure. Detailed Implementation

[0028] The present disclosure will be further described below with reference to the accompanying drawings and embodiments.

[0029] It should be noted that the following detailed descriptions are illustrative and intended to provide further explanation of this disclosure. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this disclosure pertains.

[0030] It should be noted that the terminology used herein is for the purpose of describing particular embodiments only and is not intended to limit the exemplary embodiments according to this disclosure. As used herein, the singular form is intended to include the plural form as well, unless the context clearly indicates otherwise. Furthermore, it should be understood that when the terms “comprising” and / or “including” are used in this specification, they indicate the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0031] Example 1 One embodiment of this disclosure provides a method for predicting and locating fault development stages based on multi-parameter partial discharge, the method steps including: Step 1: Acquire multi-parameter signals of partial discharge in the transformer; Step 2: Construct a multi-parameter feature set based on the multi-parameter signal of partial discharge in the transformer; Step 3: Use the improved Apriori association rule algorithm to mine the multi-parameter feature itemset, quantify the correlation strength between each parameter feature and the discharge stage, and screen out the strongly correlated features. Compare them with the preset fault development stage prediction rule base to obtain the current fault development stage. Step 4: Set a global time reference, and calculate the optical-electrical arrival time difference between each optical sensor and the global time reference, as well as the arrival time difference between each ultra-high frequency sensor; Step 5: Use the time difference of arrival (TDOA) positioning algorithm to independently calculate the spatial coordinates of the discharge source under different sensor combinations, and obtain multiple candidate positioning results and reference propagation time; Step 6: Construct a four-dimensional space-time data point set based on multiple candidate positioning results and reference propagation time. Perform cluster analysis on the four-dimensional data point set using a density-based clustering algorithm. After removing outliers, use the coordinates of the cluster center as the final spatial location of the discharge source and output it.

[0032] As one embodiment, the fault development stage prediction and location method based on multi-parameter partial discharge disclosed herein first establishes a transformer partial discharge multi-parameter signal acquisition platform to acquire pulse current, UHF, ultrasonic, and optical signals of partial discharge, and extracts feature parameters of each signal; then, it uses an improved Apriori algorithm with introduced time-series confidence constraints to mine feature association rules and establish a discharge development stage prediction rule base; using the UHF electrical signal as a time reference, it calculates the arrival time difference between light and electricity and between each UHF signal, and independently solves multiple sets of candidate discharge source coordinates; the candidate coordinates and their reference propagation time constitute a four-dimensional spatiotemporal data point set, and after removing outliers using density clustering, the cluster center is used as the final spatial location of the discharge source, realizing stage assessment and high-precision location of transformer insulation defects. Specific implementation details are as follows: Step 1: Build a multi-parameter synchronous acquisition experimental platform for partial discharge of transformers. Preset the location and type of partial discharge defects at different locations on the transformer, and collect and record the pulse current signal, ultra-high frequency electromagnetic wave signal, ultrasonic signal and optical signal generated by partial discharge under multiple power frequency cycles. Preferably, the location and type of the partial discharge defect preset in step 1 include: (a) Surface discharge defect of oil-paper insulation at the end of high voltage winding: Conductive contaminants are pre-placed on the surface of the insulating paperboard at the end of high voltage winding to simulate surface creepage; (b) Inter-turn insulation air gap discharge defect: Multiple layers of insulating paper are embedded between adjacent winding coils and an air gap is reserved to simulate the internal air gap discharge of the inter-turn insulation; (c) Floating potential discharge defect of metal components: Unreliably grounded metal components are installed at the core clamping parts to simulate floating potential discharge; (d) Free metal particle defects in insulating oil: tiny metal particles are placed at the bottom of the transformer tank or in the winding oil passage to simulate the discharge caused by free particles in the oil; (e) High-voltage lead to box tip discharge defect: A thin copper wire is tied to the surface of the high-voltage lead so that its end points to the box wall to simulate the high-voltage conductor to ground tip discharge.

[0033] The above-mentioned defect types cover the typical insulation defect forms inside the transformer, and are located in key areas such as windings, core structural components, leads and insulating oil channels. They can comprehensively excite multi-parameter response signals with different discharge physical characteristics.

[0034] Step 2: Extract the feature parameters of the partial discharge multi-parameter signal waveform, and discretize and encode the feature parameters to construct a multi-parameter feature itemset; Specifically, the extraction method for the characteristic parameters of UHF ultrasound, UHF, optical, and pulse current of partial discharge multi-parameter signals is as follows: taking the power frequency cycle as the basic unit, extract the maximum amplitude, average amplitude, and number of discharge pulses of ultrasound, UHF, and optical signals in each cycle; for the pulse current method, calculate the maximum apparent discharge in each power frequency cycle.

[0035] Furthermore, the multi-parameter features are discretized and encoded to construct a discretized encoding of the multi-parameter feature set of partial discharge, specifically as follows: First, the acquired partial discharge signal waveforms are segmented using the power frequency period as the basic unit. Then, each characteristic parameter is divided into equal-width or equal-frequency components within its global value range. k Each interval is assigned a unique symbolic identifier; for example, the maximum amplitude of the UHF frequency is divided and encoded as UHF_1 to UHF_1. k The maximum amplitude of the ultrasound is encoded as AE_1 to AE_ k The peak intensity of the optical signal is encoded as OPT_1 to OPT_ k .

[0036] At the same time, a fixed type label is assigned to each type of insulation defect preset in step 1. D The following defects were identified: surface discharge defects in the oil-paper insulation at the high-voltage winding ends, air gap discharge defects in the inter-turn insulation, floating potential discharge defects in metal components, free metal particle defects in the insulating oil, and discharge defects at the tips of the high-voltage lead wires to the enclosure. These defects were identified using DEF_ surface DEF_ void DEF_ floating DEF_ particle DEF_ tip express.

[0037] Based on the measured apparent discharge range, the discharge development stages are divided into: w Each level corresponds to STAGE_1 to STAGE_ w The process involves several steps: first, defining the stage labels; second, using a single power frequency cycle as an itemset unit, and third, combining the discretized symbolic identifiers of all feature parameters within that cycle with the corresponding defect type label and discharge stage label to form an itemset. The itemset units for all power frequency cycles constitute a multi-parameter feature itemset for partial discharge. Through this discretization and encoding process, the originally continuous, high-dimensional multi-physical quantity waveform data is transformed into structured itemsets, enabling the subsequent improved Apriori algorithm to mine the statistical correlation between feature parameters and discharge stages and defect types within a unified symbol space.

[0038] Step 3: Use the improved Apriori association rule algorithm to mine the multi-parameter feature itemset, evaluate the temporal stability of the association rule by introducing a sliding window time series confidence, quantify the association strength between each feature and the discharge stage, and select the feature with the most significant association relationship. Specifically, firstly, the improved Apriori association rule algorithm is used to mine multi-parameter feature itemsets. The specific process includes: First, the multi-parameter feature itemset constructed in step 2 is categorized by defect type label. D Grouping is performed to form a subset corresponding to each preset insulation defect. Let the total number of itemsets be . N This includes defect labels. D The number of itemsets is N D ,Right now .

[0039] Secondly, in each defect subset Perform frequent itemset mining internally. For any feature item... f (with discharge stage label) S Statistics on its The co-occurrence frequency in [the context]. Specifically, the following three basic count values ​​are defined: : Feature Item f The number of times it appears in the defective subset; Stage Tags S The number of times it appears in the defective subset; : Feature Item f With stage labels S The number of times they occur simultaneously.

[0040] Based on the above count values, calculate the feature terms. f In defects D Discharge stage under certain conditions S Conditional support Support D Conditional confidence Conf D : To measure characteristic terms f The occurrence of the discharge phase S The degree of promotion that occurs is further calculated based on the conditional lift. First, the calculation stage. S In defectsD Baseline occurrence probability under certain conditions Support D ( S ): The lift is defined as the ratio of the conditional confidence level to the baseline probability. when When, it indicates the characteristics f The emergence of this positively promoted the stage S The occurrence; when When, it indicates that the two are independent; when When, it indicates the characteristics The emergence of this inhibited the stage S The occurrence of.

[0041] Furthermore, a sliding window time-series confidence assessment is introduced to evaluate the temporal stability of association rules. Specifically: subset The items are arranged in chronological order of collection time, and the window size is set to [size missing]. K A total of [number] continuous power frequency cycles are generated, sliding with a step size of 1 cycle. One window. For candidate rules. Calculate its in the first i Local confidence level within a window : in, This is an indicator function; it takes a value of 1 when the condition within the parentheses is true, and 0 otherwise. Set the confidence threshold. Define window stability metrics For local confidence levels not lower than The proportion of the window: Simultaneously define trend enhancement factors This is used to capture the dynamic characteristic of rule confidence increasing over time. Final time series confidence TC ( R It is a weighted composite of the window stability term and the trend enhancement term: in This is the trend weighting coefficient, used to adjust the contribution of the trend term to the time series confidence level.

[0042] Preferably, in step 3, the correlation strength between each characteristic quantity and the discharge stage is quantitatively evaluated, and the characteristic quantity with the strongest correlation is selected. The specific steps are as follows: Define feature terms f In defect type D Under conditions and discharge phase S Comprehensive correlation strength for: Sort all feature items within the current defect subset in descending order based on the overall correlation strength, and select those with the highest intensity value or those exceeding a preset intensity threshold. The characteristic terms are selected as the feature quantities most closely related to the target discharge stage under this defect type, and based on this, a form like... The rules for predicting the discharge stage are stored in the rule base. Through the above operations, the improved Apriori algorithm can not only quantify and filter out features with stable and strong correlations with the discharge stage from a large number of multi-parameter features, but also effectively filter out spurious correlations caused by occasional interference through time-series confidence constraints, providing a reliable criterion for predicting the severity of transformer partial discharge during online monitoring.

[0043] Step 4: Select the UHF signal with the highest signal-to-noise ratio as the global time reference, and calculate the optical-electrical arrival time difference between each optical sensor and the time reference, as well as the arrival time difference between each UHF sensor; As a preferred method, the calculation method for the time difference of arrival in step 4 is as follows: Based on the relationship that signal energy is proportional to the square of signal amplitude, the signal voltage amplitude function is transformed into an energy accumulation function; then, the average power of the signal is introduced. P Considering the very small amplitude of the noise signal, the average energy is much lower than the average power before the partial discharge signal arrives. Therefore, the energy curve before the partial discharge signal arrives... E It will gradually decrease. When the partial discharge signal arrives, the signal amplitude is very large, and at this time the average energy is higher than the average power, as shown in the energy curve. E The signal stops decreasing and begins to gradually rise, eventually reaching a minimum at an inflection point. The time corresponding to this inflection point is considered the arrival time of the partial discharge signal from the sensor. The calculation formula is as follows: In the formula, t n Indicates the first n The time corresponding to each sampling point E ( t n )express t n The cumulative energy value corresponding to each moment. u (t i )express t i The magnitude of the signal at any given time. N s This represents the total number of sampling points. Find... E ( t n The time corresponding to the minimum t n This is the arrival time of the sensor signal.

[0044] The arrival time difference between sensors is the difference in arrival times of the signals from the two sensors.

[0045] Step 5: Use the time difference of arrival (TDOA) positioning algorithm to independently calculate the spatial coordinates of the discharge source under different methods and sensor combinations, and obtain multiple candidate positioning results and reference propagation time; This disclosure employs a time-of-arrival (TOA) positioning algorithm to independently calculate the spatial coordinates of the discharge source under different methods and sensor combinations. The calculation method is as follows: Let... v = 2×10 8 m / s is the equivalent wave speed of electromagnetic waves and light; Δ t i1 For the signal from the partial discharge power supply to the first i The time difference between the individual sensor and the reference sensor can be used to establish a spherical equation: In the formula, the first i Coordinates of each sensor S i ( x i , y i , z i ), partial discharge coordinates ( x , y , z ) and partial discharge power supply to the reference sensor S 1 reference propagation time t 1 represents the quantity to be determined.

[0046] Considering that traditional methods are difficult to directly solve nonlinear equation systems, the current common approach is to transform them into optimization problems based on the least squares principle, and then use optimization algorithms such as particle swarm optimization to solve them. Since the traditional time difference of arrival method requires at least four sensors to obtain a candidate positioning result, when there are a large number of ultra-high frequency sensors or optical sensors, multiple candidate positioning results can be obtained.

[0047] Step 6: Construct a four-dimensional space-time data point set by combining the multiple candidate positioning results and their corresponding estimated reference propagation times. Perform cluster analysis on the four-dimensional data point set using a density-based clustering algorithm. After removing outliers, use the coordinates of the cluster center as the final output of the power source spatial location.

[0048] Preferably, the multiple candidate positioning results and their corresponding estimated reference propagation times are constructed into a four-dimensional space-time data point set, specifically: The candidate positioning results obtained independently by the optical-electronic positioning and UHF positioning methods are denoted as follows: L j =( x j , y j , z j ), and the corresponding baseline propagation time estimates for different sensor combinations under each method are denoted as . T j To form four-dimensional data points Q j = ( x j , y j , z j , T j ).

[0049] Preferably, a density-based clustering algorithm is used to perform cluster analysis on the four-dimensional data point set. After removing outliers, the coordinates of the cluster centers are used as the final output of the power source spatial location. Specifically: The candidate positioning results obtained independently from different sensor combinations of optical-electronic positioning and UHF positioning methods are denoted as follows: The baseline propagation time estimates for each method under different sensor combinations are denoted as follows: T j ,constitute γ Four-dimensional data points , j = 1, 2, …, γ Define any two points and The weighted Euclidean distance between them is : in, and These are the weighting coefficients for the spatial coordinate dimension and the time dimension, respectively, satisfying... To balance the contributions of spatial location differences and propagation time differences to the clustering results, a density-based DBSCAN clustering algorithm is employed, with a set neighborhood radius. ε and minimum neighborhood number MinPts For four-dimensional data point sets Clustering is performed, and isolated points that do not belong to any cluster are marked as outliers and removed; for the highest density clusters retained... Its cluster center Calculated by the arithmetic mean of all points within the cluster, i.e. , , Ultimately This serves as the spatial location output of the fused discharge source. Through the aforementioned four-dimensional spatiotemporal joint density clustering, the complementary spatial consistency information of optical-electric positioning and UHF positioning under different sensor combinations can be fully utilized, effectively eliminating outlier solutions and significantly improving the accuracy and robustness of the final positioning result.

[0050] Example 2 One embodiment of this disclosure provides a fault development stage prediction and location method based on partial discharge multi-parameters. Taking a full-scale three-phase transformer with a voltage level of 110 kV and dimensions of 4.8 × 2.4 × 2.8 m pre-set with five typical insulation defects as an example, the implementation method of this disclosure based on partial discharge multi-parameter fault development stage prediction and location method is specifically illustrated. The implementation flowchart is as follows. Figure 1 As shown.

[0051] Step 1: Build a multi-parameter synchronous acquisition experimental platform for partial discharge of transformers, preset insulation defects and acquire signals.

[0052] With one corner of the transformer tank as the origin of a spatial rectangular coordinate system, the length, width, and height directions are the x, y, and z axes, respectively. Four ultra-high frequency sensors (numbered U1~U4), four ultrasonic sensors (numbered A1~A4), and four fluorescent fiber optic sensors (numbered O1~O4) are installed on the transformer tank wall. The coordinates of the UHF sensors U1~U4 are (1.20, 0.00, 1.80), (3.60, 0.00, 1.60), (0.00, 1.20, 2.10), and (4.80, 1.80, 1.40), respectively; the coordinates of the ultrasonic sensors A1~A4 are (1.00, 0.00, 1.50), (3.80, 0.00, 1.40), (0.00, 1.50, 1.80), and (4.80, 1.20, 1.20), respectively; the coordinates of the optical sensors O1~O4 are (1.50, 0.80, 0.50), (3.00, 1.60, 0.80), (0.00, 1.90, 2.30), and (4.10, 0.70), respectively. 2.80); The pulse current is obtained at the end screen of the bushing through the detection impedance.

[0053] Following the preferred scheme in step 1, five typical insulation defects were pre-set inside the transformer: (a) surface discharge defect at the end of the high-voltage winding (labeled DEF_surface); (b) inter-turn insulation air gap discharge defect (labeled DEF_void); (c) floating potential discharge defect in the core clamp (labeled DEF_floating); (d) free metal particle defect in the insulating oil (labeled DEF_particle); and (e) discharge defect between the high-voltage lead and the tank tip (labeled DEF_tip). For each defect, pulse current, ultra-high frequency, ultrasonic, and optical signal waveform data were continuously collected for 500 power frequency cycles under different test voltages.

[0054] Step 2: Extract feature parameters and perform discretization encoding to construct a multi-parameter feature itemset. Using the power frequency cycle as the basic unit, extract the feature parameters of each signal within each cycle. For UHF and ultrasonic signals, extract the maximum amplitude, average amplitude, and number of discharge pulses; for optical signals, extract the peak light intensity; and for pulsed current signals, calculate the maximum apparent discharge. Each feature parameter is divided into k = 5 intervals of equal width within its global range, and each interval is assigned a discretized code. For example, the global range of the maximum amplitude of the UHF signal is [0, 3] V, which is divided into 5 intervals and encoded as UHF_A1 to UHF_A5; the maximum amplitude of the ultrasound signal is encoded as AE_A1 to AE_A5; and the peak intensity of the optical signal is encoded as OPT_A1 to OPT_A5. The apparent discharge quantity is... , , It is divided into three discharge stages, which are labeled STAGE_1, STAGE_2, and STAGE_3 respectively.

[0055] Using a single power frequency cycle as an item set unit, the discretized identifiers corresponding to all characteristic parameters within that cycle are combined with the defect type label and the discharge stage label to form an item set record. For example, under the point discharge defect (DEF_TIP), the item set record for the 12th power frequency cycle is {UHF_A2, UHF_AVG2, AE_A1, AE_AVG1, OPT_A1, DEF_tip,STAGE_1}; and for the 187th cycle it is {UHF_A4, UHF_AVG3, AE_A2, AE_AVG2, OPT_A2, DEF_tip,STAGE_2}.

[0056] The itemsets from all test cycles constitute a complete multi-parameter feature set of partial discharge.

[0057] Step 3: Use the improved Apriori algorithm to mine association rules and quantify the association strength between features and the discharge stage.

[0058] First, the feature itemset is grouped according to the defect type label D to obtain each defect subset. For example, a subset of a tip discharge defect (D = DEF_tip) contains ND = 500 itemset records.

[0059] The co-occurrence frequency of each feature term f and the discharge stage label S was statistically analyzed. Taking feature term f = OPT_A2 (the second interval of optical signal peak intensity) and target discharge stage S = STAGE_2 as an example, the statistical results were obtained. (Number of times the feature appears) (Number of times a phase occurs) (Number of simultaneous occurrences). Calculate conditional lift: A sliding window time-series confidence level is introduced, with a window size K = 3 and a confidence threshold θc = 0.70. The local confidence sequence of this rule within each sliding window is calculated, and the stability index is statistically obtained. Trend enhancement factor Taking the trend weighting coefficient α = 0.5, calculate the time series confidence score: The final overall association strength is: The correlation strength between all feature terms and STAGE_2 under the tip discharge defect was calculated and sorted using the same method. The top 5 feature terms were OPT_A2, UHF_A4, AE_A2, UHF_AVG3, and AE_AVG2.

[0060] It can be seen that under the condition of tip discharge defects, the optical signal characteristic OPT_A2 (i.e., the peak light intensity is at a medium level) is most closely correlated with the intermediate discharge stage STAGE_2. This means that when the system detects the optical signal intensity entering the second interval, it can be inferred with high confidence that the tip discharge has developed to the intermediate stage. The above-selected strongly correlated features and their corresponding rules are then analyzed. Stored in the rule base. Similarly, strong correlation features of each discharge stage can be mined for each defect type to form a complete rule base for predicting fault development stages.

[0061] Step 4: Calculate the arrival time difference between each sensor.

[0062] The UHF sensor U2, with the highest signal-to-noise ratio, was selected as the global time reference. Taking a typical tip discharge event as an example, the arrival time of the signals from each sensor was calculated using the energy accumulation method. Figure 2 The energy accumulation curve of the U2 sensor signal is given. The time tU2 = 467.4 ns, corresponding to the minimum inflection point of the curve, is the arrival time of its signal. The arrival times of the other sensors are obtained in the same way.

[0063] Step 5: Independently calculate the coordinates of candidate power sources for optical-electric positioning and UHF positioning.

[0064] Optical-electric positioning and ultra-high frequency positioning are calculated independently, respectively. Equivalent wave velocities of electromagnetic waves and light are taken. Taking UHF positioning as an example, with U2 as the reference sensor, a system of spherical equations is established using four UHF sensors and transformed into an optimization problem, which is then solved using the particle swarm optimization algorithm. Similarly, for optical-electronic positioning, U2 is used as the reference sensor, and three of the four optical sensors are selected for positioning.

[0065] There are four ultra-high frequency sensors, which can obtain one set of candidate positioning results; the combination of optical sensor and ultra-high frequency reference sensor can obtain four sets of candidate results.

[0066] Step 6: Four-dimensional spatiotemporal density clustering fusion localization.

[0067] The five candidate positioning results and their baseline propagation times are used to construct five four-dimensional data points, and spatial weights are assigned. Time weight The DBSCAN algorithm is used to set the neighborhood radius. Minimum number of neighborhood points Cluster analysis showed that multiple data points were grouped into the same cluster. There are no outliers. Calculate the coordinates of the cluster centers: The final output location of the fused discharge source is (2.043, 0.320, 1.783) m. Compared with the actual preset location of the tip discharge defect (2.15, 0.30, 1.60) m, the positioning error is only 0.213 m.

[0068] This embodiment fully demonstrates the entire process of this disclosure, from multi-parameter signal acquisition, feature discretization itemset construction, improved Apriori association rule mining to four-dimensional spatiotemporal clustering localization, and verifies the effectiveness and superiority of the method in both discharge stage prediction and high-precision localization.

[0069] Example 3 One embodiment of this disclosure provides a fault development stage prediction and location system based on partial discharge multi-parameters, including: The signal acquisition module is used to acquire multi-parameter signals of partial discharge in the transformer. The feature extraction module is used to construct a multi-parameter feature set based on the multi-parameter signal of partial discharge of transformer; The association mining and prediction module is used to mine multi-parameter feature itemsets using the improved Apriori association rule algorithm, quantify the association strength between each parameter feature and the discharge stage, and screen out strongly associated features. These features are then compared with the preset fault development stage prediction rule library to obtain the current fault development stage. The time difference calculation module is used to set a global time reference and calculate the optical-electric arrival time difference between each optical sensor and the global time reference, as well as the arrival time difference between each ultra-high frequency sensor. An independent calculation module is used to independently calculate the spatial coordinates of the discharge source under different sensor combinations using the time difference of arrival (TDOA) positioning algorithm, and obtain multiple candidate positioning results and reference propagation time. The fault location module is used to construct a four-dimensional space-time data point set based on multiple candidate location results and reference propagation time. It uses a density-based clustering algorithm to perform cluster analysis on the four-dimensional data point set, removes outliers, and outputs the final discharge power source spatial location using the coordinates of the cluster center.

[0070] Example 4 One embodiment of this disclosure provides a computer program product, including a computer program that, when executed by a processor, implements the fault development stage prediction and localization method based on partial discharge multi-parameters.

[0071] Example 5 One embodiment of this disclosure provides a non-transitory computer-readable storage medium for storing computer instructions. When these computer instructions are executed by a processor, they implement the fault development stage prediction and location method based on partial discharge multi-parameters.

[0072] Example 6 One embodiment of this disclosure provides an electronic device, including a processor, a memory, and a computer program; wherein the processor is connected to the memory, and the computer program is stored in the memory. When the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to implement the fault development stage prediction and location method based on partial discharge multi-parameters.

[0073] In further embodiments, the following is also provided: A server that can be used to execute the methods provided in the above embodiments. Specifically: A server includes a Central Processing Unit (CPU), system memory comprising Random Access Memory (RAM) and Read Only Memory (ROM), and a system bus connecting the system memory and the CPU. The server also includes a basic input / output system (I / O system) to facilitate information transfer between various components within the computer, and mass storage devices for storing the operating system, applications, and other program modules.

[0074] A basic input / output system includes a display for showing information and input devices such as a mouse and keyboard for user input. Both the display and the input devices are connected to the central processing unit via an input / output controller connected to the system bus. The basic input / output system may also include an input / output controller for receiving and processing input from multiple other devices such as a keyboard, mouse, or electronic stylus. Similarly, the input / output controller also provides output to a display screen, printer, or other types of output devices.

[0075] Mass storage devices are connected to the central processing unit via a mass storage controller (not shown) connected to the system bus. The mass storage devices and their associated computer-readable media provide non-volatile storage for the server. That is, mass storage devices may include computer-readable media (not shown) such as hard disks or CD-ROM (CompactDisc Read-Only Memory) drives.

[0076] Computer-readable media can include computer storage media and communication media. Computer storage media includes volatile and non-volatile, removable and non-removable media implemented using any method or technology for storing information such as computer-readable instructions, data structures, program modules, or other data. Computer storage media include RAM, ROM, EPROM (Erasable Programmable Read Only Memory), EEPROM (Electrically Erasable Programmable Read Only Memory), flash memory or other solid-state storage technologies, CD-ROM, DVD (Digital Versatile Disc) or other optical storage, magnetic tape cassettes, magnetic tape, disk storage, or other magnetic storage devices. Of course, those skilled in the art will understand that computer storage media are not limited to the above-mentioned types. The aforementioned system memories and mass storage devices can be collectively referred to as memory.

[0077] According to various embodiments of this disclosure, the server can also connect to and operate on a remote computer on a network, such as the Internet. That is, the server can connect to a network via a network interface unit connected to the system bus, or it can use a network interface unit to connect to other types of networks or remote computer systems (not shown).

[0078] The aforementioned memory also includes one or more programs, which are stored in the memory and configured to be executed by the CPU.

[0079] One embodiment provides a terminal that can be used to perform the methods provided in the above embodiments. The terminal may be a portable mobile terminal, such as a smartphone, tablet computer, MP3 player (Moving Picture Experts Group Audio Layer III), MP4 player (Moving Picture Experts Group Audio Layer IV), laptop computer, or desktop computer. The terminal may also be referred to by other names such as user terminal, portable terminal, laptop terminal, desktop terminal, etc.

[0080] Typically, a terminal includes a processor and memory.

[0081] The processor may include one or more processing cores, such as a quad-core processor or an octa-core processor. The processor may be implemented using at least one of the following hardware forms: DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), or PLA (Programmable Logic Array). The processor may also include a main processor and coprocessors. The main processor, also known as the CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, the processor may integrate a GPU (Graphics Processing Unit), which is responsible for rendering and drawing the content to be displayed on the screen. In some embodiments, the processor may also include an AI (Artificial Intelligence) processor, which handles computational operations related to machine learning.

[0082] The memory may include one or more computer-readable storage media, which may be non-transitory. The memory may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory are used to store at least one instruction, which is executed by a processor to implement the sound reverberation method provided in the method embodiments of this application.

[0083] In some embodiments, the terminal may also optionally include: a peripheral device interface and at least one peripheral device. The processor, memory, and peripheral device interface can be connected via a bus or signal line. Each peripheral device can be connected to the peripheral device interface via a bus, signal line, or circuit board. Specifically, the peripheral device includes at least one of: a radio frequency circuit, a display screen, a camera assembly, an audio circuit, a positioning assembly, or a power supply.

[0084] Peripheral device interfaces can be used to connect at least one I / O (Input / Output) related peripheral device to the processor and memory. In some embodiments, the processor, memory, and peripheral device interface are integrated on the same chip or circuit board; in some other embodiments, any one or two of the processor, memory, and peripheral device interface can be implemented on separate chips or circuit boards, which is not limited in this embodiment.

[0085] Radio frequency (RF) circuits are used to receive and transmit RF signals, also known as electromagnetic signals. RF circuits communicate with communication networks and other communication devices via electromagnetic signals. RF circuits convert electrical signals into electromagnetic signals for transmission, or convert received electromagnetic signals back into electrical signals. Optionally, RF circuits include: antenna systems, RF transceivers, one or more amplifiers, tuners, oscillators, digital signal processors, codec chipsets, user identity module cards, etc. RF circuits can communicate with other terminals through at least one wireless communication protocol. These wireless communication protocols include, but are not limited to: the World Wide Web, metropolitan area networks, intranets, various generations of mobile communication networks (2G, 3G, 4G, and 5G), wireless local area networks, and / or WiFi (Wireless Fidelity) networks. In some embodiments, the RF circuit may also include circuitry related to NFC (Near Field Communication), which is not limited in this application.

[0086] The display screen is used to display the UI (User Interface). This UI can include graphics, text, icons, videos, and any combination thereof. When the display screen is a touch screen, it also has the ability to collect touch signals on or above the surface of the display. These touch signals can be input as control signals to a processor for processing. In this case, the display screen can also be used to provide virtual buttons and / or a virtual keyboard, also known as soft buttons and / or a soft keyboard. In some embodiments, there can be one display screen, which serves as the front panel of the terminal; in other embodiments, there can be at least two display screens, respectively disposed on different surfaces of the terminal or in a folded design; in still other embodiments, the display screen can be a flexible display screen, disposed on a curved or folded surface of the terminal. Furthermore, the display screen can be configured as a non-rectangular, irregular shape, i.e., a non-rectangular screen. The display screen can be made of materials such as LCD (Liquid Crystal Display) and OLED (Organic Light-Emitting Diode).

[0087] A camera assembly is used to capture images or videos. Optionally, the camera assembly includes a front-facing camera and a rear-facing camera. Typically, the front-facing camera is located on the front panel of the terminal, and the rear-facing camera is located on the back of the terminal. In some embodiments, there are at least two rear-facing cameras, which are any one of a main camera, a depth-sensing camera, a wide-angle camera, and a telephoto camera, to achieve background blurring by fusion of the main camera and the depth-sensing camera, panoramic shooting by fusion of the main camera and the wide-angle camera, VR (Virtual Reality) shooting, or other fusion shooting functions. In some embodiments, the camera assembly may also include a flash. The flash can be a single-color temperature flash or a dual-color temperature flash. A dual-color temperature flash refers to a combination of a warm-light flash and a cool-light flash, which can be used for light compensation at different color temperatures.

[0088] The audio circuitry may include a microphone and a speaker. The microphone is used to collect sound waves from the user and the environment, converting them into electrical signals that are input to a processor for processing, or to radio frequency (RF) circuitry for voice communication. For stereo sound acquisition or noise reduction purposes, multiple microphones may be used, positioned at different locations on the terminal. The microphone may also be an array microphone or an omnidirectional microphone. The speaker is used to convert electrical signals from the processor or RF circuitry into sound waves. The speaker may be a traditional diaphragm speaker or a piezoelectric ceramic speaker. When the speaker is a piezoelectric ceramic speaker, it can convert electrical signals not only into audible sound waves but also into inaudible sound waves for purposes such as distance measurement. In some embodiments, the audio circuitry may also include a headphone jack.

[0089] The positioning component is used to determine the current geographical location of the terminal to enable navigation or LBS (Location Based Service). The positioning component can be based on the US GPS (Global Positioning System), China's BeiDou system, or Russia's Galileo system.

[0090] The power supply is used to power the various components in the terminal. The power supply can be alternating current (AC), direct current (DC), a disposable battery, or a rechargeable battery. When the power supply includes a rechargeable battery, it can be a wired or wirelessly rechargeable battery. A wired rechargeable battery is charged via a wired connection, while a wirelessly rechargeable battery is charged via a wireless coil. The rechargeable battery can also be used to support fast charging technology.

[0091] In some embodiments, the terminal further includes one or more sensors. These one or more sensors include, but are not limited to, accelerometers, gyroscopes, pressure sensors, fingerprint sensors, optical sensors, and proximity sensors.

[0092] An accelerometer can detect the magnitude of acceleration along the three axes of a coordinate system established by the terminal. For example, an accelerometer can be used to detect the components of gravitational acceleration along the three axes. The processor can then control the touchscreen to display the user interface in either landscape or portrait view based on the gravitational acceleration signals acquired by the accelerometer. Accelerometers can also be used for collecting motion data in games or for other applications.

[0093] The gyroscope sensor can detect the terminal's orientation and rotation angle. It can work in conjunction with an accelerometer to capture the user's 3D movements on the terminal. Based on the data collected by the gyroscope sensor, the processor can perform the following functions: motion sensing (e.g., changing the UI based on the user's tilt), image stabilization during shooting, game control, and inertial navigation.

[0094] The pressure sensor can be located on the side bezel of the terminal and / or under the touchscreen display. When the pressure sensor is located on the side bezel, it can detect the user's grip signal on the terminal, and the processor can perform left / right hand recognition or quick operation based on the grip signal collected by the pressure sensor. When the pressure sensor is located under the touchscreen display, the processor can control the operable controls on the UI interface based on the user's pressure on the touchscreen display. Operable controls include at least one of button controls, scroll bar controls, icon controls, or menu controls.

[0095] A fingerprint sensor is used to collect a user's fingerprint. The processor identifies the user based on the fingerprint collected by the sensor, or vice versa. When the user's identity is verified as trusted, the processor authorizes the user to perform relevant sensitive operations, including unlocking the screen, viewing encrypted information, downloading software, making payments, and changing settings. The fingerprint sensor can be located on the front, back, or side of the terminal. When the terminal has physical buttons or a manufacturer's logo, the fingerprint sensor can be integrated with those buttons or the logo.

[0096] An optical sensor is used to collect ambient light intensity. In one embodiment, the processor can control the display brightness of the touch screen based on the ambient light intensity collected by the optical sensor. Specifically, when the ambient light intensity is high, the display brightness of the touch screen is increased; when the ambient light intensity is low, the display brightness of the touch screen is decreased. In another embodiment, the processor can also dynamically adjust the shooting parameters of the camera assembly based on the ambient light intensity collected by the optical sensor.

[0097] A proximity sensor, also known as a distance sensor, is typically located on the front panel of a terminal. It is used to detect the distance between the user and the front of the terminal. In one embodiment, when the proximity sensor detects that the distance between the user and the front of the terminal is gradually decreasing, the processor controls the touchscreen display to switch from a screen-on state to a screen-off state; conversely, when the proximity sensor detects that the distance between the user and the front of the terminal is gradually increasing, the processor controls the touchscreen display to switch from a screen-off state to a screen-on state.

[0098] Those skilled in the art will understand that the structure shown does not constitute a limitation on the terminal, and may include more or fewer components than shown, or combine certain components, or employ different component arrangements.

[0099] An electronic device includes a memory and a processor, as well as computer instructions stored in the memory and running on the processor. When executed by the processor, the computer instructions perform the method described in Embodiment 1. For brevity, further details are omitted here.

[0100] It should be understood that in this embodiment, the processor can be a central processing unit (CPU), or it can be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or any conventional processor, etc.

[0101] Memory may include read-only memory and random access memory, and provides instructions and data to the processor. A portion of memory may also include non-volatile random access memory. For example, memory may also store information about the device type.

[0102] A computer-readable storage medium for storing computer instructions, which, when executed by a processor, perform the method described in Embodiment 1.

[0103] The method in Embodiment 1 can be directly implemented by a hardware processor, or implemented by a combination of hardware and software modules within the processor. The software modules can reside in readily available storage media in the art, such as random access memory, flash memory, read-only memory, programmable read-only memory, electrically erasable programmable memory, or registers. This storage medium is located in memory; the processor reads information from the memory and, in conjunction with its hardware, completes the steps of the above method.

[0104] The computer storage medium of this embodiment can be any combination of one or more computer-readable media. The computer-readable medium can be a computer-readable signal medium or a computer-readable storage medium. For example, a computer-readable storage medium can be—but is not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media (a non-exhaustive list) include: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this document, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0105] Computer-readable signal media may include data signals propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, capable of sending, propagating, or transmitting programs for use by or in connection with an instruction execution system, apparatus, or device.

[0106] Program code contained on a computer-readable medium may be transmitted using any suitable medium, including—but not limited to—wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0107] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as "C" or similar languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0108] A computer program product includes a computer program that, when executed by a processor, implements the method described in Embodiment 1.

[0109] This disclosure also provides at least one computer program product tangibly stored on a non-transitory computer-readable storage medium. The computer program product includes computer-executable instructions, such as instructions included in program modules, which execute in a device on a target real or virtual processor to perform the processes / methods described above. Typically, program modules include routines, programs, libraries, objects, classes, components, data structures, etc., that perform specific tasks or implement specific abstract data types. In various embodiments, the functionality of program modules can be combined or divided among program modules as needed. The machine-executable instructions for the program modules can execute within a local or distributed device. In a distributed device, the program modules can reside in both local and remote storage media.

[0110] Computer program code used to implement the methods of this disclosure may be written in one or more programming languages. This computer program code may be provided to a processor of a general-purpose computer, a special-purpose computer, or other programmable data processing apparatus, such that when executed by the computer or other programmable data processing apparatus, the program code causes the functions / operations specified in the flowcharts and / or block diagrams to be performed. The program code may be executed entirely on a computer, partially on a computer, as a stand-alone software package, partially on a computer and partially on a remote computer, or entirely on a remote computer or server.

[0111] In the context of this disclosure, computer program code or related data may be carried on any suitable carrier to enable a device, apparatus, or processor to perform the various processes and operations described above. Examples of carriers include signals, computer-readable media, and the like. Examples of signals may include electrical, optical, radio, sound, or other forms of propagation signals, such as carrier waves, infrared signals, etc.

[0112] Those skilled in the art will recognize that the units and algorithm steps described in conjunction with the embodiments herein can be implemented in electronic hardware or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0113] While the specific embodiments of this disclosure have been described above in conjunction with the accompanying drawings, this is not intended to limit the scope of protection of this disclosure. Those skilled in the art should understand that various modifications or variations that can be made by those skilled in the art without creative effort based on the technical solutions of this disclosure are still within the scope of protection of this disclosure.

Claims

1. A fault development stage prediction and location method based on multi-parameter partial discharge, characterized in that, include: Acquire multi-parameter signals of partial discharge in transformers; Based on the multi-parameter signal of partial discharge in transformers, a multi-parameter feature itemset is constructed; An improved Apriori association rule algorithm is used to mine multi-parameter feature itemsets, quantify the correlation strength between each parameter feature and the discharge stage, and select strongly correlated features. These features are then compared with a preset fault development stage prediction rule base to obtain the current fault development stage. A global time reference is established, and the optical-electric arrival time difference between each optical sensor and the global time reference, as well as the arrival time difference between each ultra-high frequency sensor, are calculated respectively. The time difference of arrival (TDOA) positioning algorithm is used to independently calculate the spatial coordinates of the discharge source under different sensor combinations, and obtain multiple candidate positioning results and reference propagation time. A four-dimensional spatial-temporal data point set is constructed based on multiple candidate positioning results and reference propagation time. A density-based clustering algorithm is used to perform cluster analysis on the four-dimensional data point set. After removing outliers, the coordinates of the cluster center are used as the final spatial location of the discharge source and output.

2. The fault development stage prediction and location method based on multi-parameter partial discharge as described in claim 1, characterized in that, The acquisition of multi-parameter signals of partial discharge in the transformer includes: A multi-parameter synchronous acquisition experimental platform for partial discharge of transformers was built. The location and type of partial discharge defects were preset at different locations on the transformer. Pulse current signals, ultra-high frequency electromagnetic wave signals, ultrasonic signals and optical signals generated by partial discharge under multiple power frequency cycles were collected and recorded.

3. The fault development stage prediction and location method based on multi-parameter partial discharge as described in claim 2, characterized in that, The locations and types of the pre-set partial discharge defects include: (a) surface discharge defects of oil-paper insulation at the end of high-voltage windings: conductive contaminants are pre-placed on the surface of the insulating paperboard at the end of the high-voltage windings to simulate surface creepage; (b) air gap discharge defects between turns: multiple layers of insulating paper are embedded between adjacent winding coils and air gaps are reserved to simulate air gap discharge inside the inter-turn insulation; (c) floating potential discharge defects of metal components: unreliably grounded metal components are set at the core clamps to simulate floating potential discharge; (d) free metal particle defects in insulating oil: tiny metal shavings are placed at the bottom of the transformer tank or in the winding oil passages to simulate discharge caused by free particles in the oil; (e) discharge defects of high-voltage leads to the tank tip: thin copper wires are tied to the surface of the high-voltage leads so that their ends point to the tank wall to simulate high-voltage conductor tip discharge to ground.

4. The fault development stage prediction and location method based on multi-parameter partial discharge as described in claim 1, characterized in that, The construction of a multi-parameter feature itemset based on the multi-parameter signal of transformer partial discharge includes: Using the power frequency cycle as the basic unit, the maximum amplitude, average amplitude, and number of discharge pulses of ultrasonic, ultra-high frequency, and optical signals are extracted within each cycle; for the pulse current method, the maximum apparent discharge quantity within each power frequency cycle is calculated. Discretize the multi-parameter features to construct a discrete encoding scheme for the multi-parameter feature set of partial discharge; By using discretization encoding, the original continuous, high-dimensional multi-physical quantity waveform data is converted into structured itemsets, resulting in multi-parameter feature itemsets.

5. The fault development stage prediction and location method based on multi-parameter partial discharge as described in claim 1, characterized in that, The improved Apriori association rule algorithm is used to mine multi-parameter feature itemsets, quantify the correlation strength between each parameter feature and the discharge stage, and screen out strongly correlated features, including: The multi-parameter feature itemset is grouped according to the defect type label to form a sub-itemet corresponding to each preset insulation defect; Perform frequent itemset mining within each sub-itemset, and count the co-occurrence frequency of any feature item in the sub-itemset; Based on the co-occurrence frequency, the conditional support and conditional confidence of the feature terms under the defect type label for the discharge stage are calculated, and the conditional lift is further calculated. A sliding window time series confidence assessment is introduced to evaluate the temporal stability of the association rules, quantify the association strength between each feature and the discharge stage, and select the feature with the most significant association as the strongly associated feature.

6. The fault development stage prediction and location method based on multi-parameter partial discharge as described in claim 5, characterized in that, The method of introducing a sliding window time-series confidence assessment to evaluate the temporal stability of association rules includes: Arrange the items in the sub-item set in chronological order of collection time, and set the window size to [value missing]. K Given a continuous power frequency cycle, define a period window. For candidate rules, calculate their value on the [number]th [period]. i Local confidence level within a window; Simultaneously, the dynamic characteristics of the trend enhancement factor capturing rule confidence increasing over time are defined, and the final time series confidence is a weighted synthesis of the window stability term and the trend enhancement term.

7. The fault development stage prediction and location method based on multi-parameter partial discharge as described in claim 5, characterized in that, The process quantifies the correlation strength between each characteristic quantity and the discharge stage, and selects the characteristic quantity with the most significant correlation as the strongly correlated characteristic quantity, including: Define the overall correlation strength between characteristic terms and the discharge stage under defect type conditions; Based on the comprehensive correlation strength, all feature items in the current defect subset are sorted in descending order, and feature items with the highest intensity value or exceeding the preset intensity threshold are selected as the strongly correlated feature quantity with the largest correlation with the target discharge stage under this defect type. Based on this, discharge stage prediction rules are formed and stored in the rule base.

8. The fault development stage prediction and location method based on multi-parameter partial discharge as described in claim 1, characterized in that, The process of setting a global time reference and calculating the optical-electrical arrival time difference between each optical sensor and the global time reference, as well as the arrival time difference between each ultra-high frequency sensor, includes: Based on the relationship that signal energy is proportional to the square of signal amplitude, the signal voltage amplitude function is transformed into an energy accumulation function. Then, the average power of the signal is introduced. Considering that the signal amplitude of noise is very small, the average energy is much lower than the average power before the partial discharge signal arrives. Therefore, the energy curve will gradually decrease before the partial discharge signal arrives. When the partial discharge signal arrives, the signal amplitude is very large. At this time, the average energy is higher than the average power, and the energy curve stops decreasing and begins to gradually rise, thus generating an inflection point of minimum value. The time corresponding to this inflection point is regarded as the arrival time of the sensor's partial discharge signal.

9. The fault development stage prediction and location method based on multi-parameter partial discharge as described in claim 1, characterized in that, The method of independently calculating the spatial coordinates of the discharge source under different sensor combinations using the time difference of arrival positioning algorithm includes: Define the equivalent wave speeds of electromagnetic waves and light; based on the signal from the partial discharge source to the... i The time difference between the individual sensor and the reference sensor is used to establish the spherical equation: The particle swarm optimization algorithm was used to solve the problem, and multiple candidate localization results were obtained.

10. The fault development stage prediction and location method based on multi-parameter partial discharge as described in claim 1, characterized in that, The construction of a four-dimensional space-time data point set based on multiple candidate localization results and a reference propagation time includes: Four-dimensional data points are constructed based on the candidate positioning results obtained independently by optical-electronic positioning and UHF positioning methods, as well as the reference propagation time estimates corresponding to different sensor combinations for each method.

11. The fault development stage prediction and location method based on multi-parameter partial discharge as described in claim 1, characterized in that, The density-based clustering algorithm performs cluster analysis on the four-dimensional data point set, removes outliers, and outputs the final discharge power source spatial location using the cluster center coordinates, including: Based on a four-dimensional data point set, a weighted Euclidean distance is defined between any two points; The density-based DBSCAN clustering algorithm is adopted to cluster the four-dimensional data point set by setting the neighborhood radius and the minimum number of neighborhood points. Isolated points that do not belong to any cluster are marked as outliers and removed. For the maximum density cluster, its cluster center is calculated by the arithmetic mean of all points in the cluster, and finally used as the output of the spatial location of the fused discharge source.

12. A fault development stage prediction and location system based on multi-parameter partial discharge, characterized in that, include: The signal acquisition module is used to acquire multi-parameter signals of partial discharge in the transformer. The feature extraction module is used to construct a multi-parameter feature set based on the multi-parameter signal of partial discharge of transformer; The association mining and prediction module is used to mine multi-parameter feature itemsets using the improved Apriori association rule algorithm, quantify the association strength between each parameter feature and the discharge stage, and screen out strongly associated features. These features are then compared with the preset fault development stage prediction rule library to obtain the current fault development stage. The time difference calculation module is used to set a global time reference and calculate the optical-electric arrival time difference between each optical sensor and the global time reference, as well as the arrival time difference between each ultra-high frequency sensor. An independent calculation module is used to independently calculate the spatial coordinates of the discharge source under different sensor combinations using the time difference of arrival (TDOA) positioning algorithm, and obtain multiple candidate positioning results and reference propagation time. The fault location module is used to construct a four-dimensional space-time data point set based on multiple candidate location results and reference propagation time. It uses a density-based clustering algorithm to perform cluster analysis on the four-dimensional data point set, removes outliers, and outputs the final discharge power source spatial location using the coordinates of the cluster center.

13. A computer program product, comprising a computer program, characterized in that, When the computer program is executed by the processor, it implements the fault development stage prediction and location method based on partial discharge multi-parameters as described in any one of claims 1-11.

14. A non-transitory computer-readable storage medium, characterized in that, The non-transitory computer-readable storage medium is used to store computer instructions, which, when executed by a processor, implement the fault development stage prediction and location method based on partial discharge multi-parameters as described in any one of claims 1-11.

15. An electronic device, characterized in that, include: The device includes a processor, a memory, and a computer program; wherein the processor is connected to the memory, the computer program is stored in the memory, and when the electronic device is running, the processor executes the computer program stored in the memory to enable the electronic device to perform the fault development stage prediction and location method based on partial discharge multi-parameters as described in any one of claims 1-11.

Images