GIS partial discharge signal adaptive gain control method based on UHF built-in sensor

Abstract

The present application relates to the technical field of sensor regulation, and particularly relates to a GIS partial discharge signal adaptive gain control method based on a UHF built-in sensor, the present application constructs a historical event data set based on a plurality of event time stamps and event time stamp associated power frequency phase angles determined by each trigger event, performs timing analysis on the historical event data set to determine a trigger rhythm parameter, determines a signal state and filters a risk working condition period, determines whether to call a UHF time domain curve of each risk working condition period based on the signal state, marks a period tendency state according to a homologous tendency parameter of the risk working condition period, and determines a regulation mode of a current partial discharge signal gain according to a neighboring disturbance parameter. The present application realizes identification of a signal state, adaptively adjusts a gain execution mode according to different signal tendency states, and improves the reliability of GIS partial discharge signal adaptive gain control.

Description

Technical Field

[0001] This invention relates to the field of sensor control technology, and in particular to an adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor. Background Technology

[0002] Gas-insulated switchgear (GIS) has significant advantages such as small footprint, high operational reliability, and low maintenance workload, and has been widely used in high-voltage and ultra-high-voltage power transmission and transformation fields. However, during long-term operation, insulation defects introduced during manufacturing, transportation, and installation may lead to partial discharge phenomena. If not detected and handled in time, these defects can cause insulation breakdown accidents, seriously threatening the safe operation of the power grid. UHF detection, due to its excellent anti-interference performance, high sensitivity, ability to identify defect types, and location of discharge sources, has become an important means of GIS insulation condition monitoring. Built-in sensors offer higher coupling efficiency compared to external sensors. Efficiency and stronger anti-interference capabilities can improve the reliability and authenticity of signals from the source. However, UHF signals have a very large dynamic range of amplitude and severe attenuation due to the L-shaped structure. Fixed gain is difficult to balance the sensitivity of micro signals and the prevention of saturation of strong signals. Traditional automatic gain control cannot distinguish between internal and external signal sources and perform homogeneous amplitude equalization. Under strong interference, it raises the noise floor to suppress partial discharge. Fixed time constant is prone to gain oscillation or pulse clipping. Existing systems have insufficient anti-interference capabilities, frequent false alarms and missed alarms, poor real-time performance, and are disconnected from gain control. Therefore, improving the reliability of adaptive gain control for GIS partial discharge signals is an urgent technical problem to be solved.

[0003] For example, Chinese patent application publication number CN117554754A discloses an optimization method and system for a built-in ultra-high frequency sensor in GIS, relating to the field of sensor optimization technology. The method includes collecting data from GIS equipment, partial discharge pulse current, and sensors; establishing a calculation model for a circular microstrip antenna; calculating the effective radius of the circular microstrip antenna; comparing the physical radius and effective radius of the circular microstrip antenna and obtaining the comparison result; and optimizing the sensor based on the comparison result. By calculating the effective radius of the circular microstrip antenna and comparing the physical radius with the effective radius, the sensor model or location is optimized to fully utilize its performance and meet the performance requirements for detecting partial discharge generated by gas-insulated switchgear (GIS).

[0004] The following problems still exist in the existing technology: Existing technologies do not consider the contradiction between the dynamic range of signal amplitude and severe transmission attenuation, and it is also difficult to distinguish the source attributes of external interference and internal partial discharge. Existing technologies cannot quickly identify the signal state and adaptively adjust the gain execution mode according to different signal tendency states, which affects the reliability of adaptive gain control of GIS partial discharge signals. Summary of the Invention

[0005] To address this issue, the present invention provides an adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor, which overcomes the problem that existing technologies cannot quickly identify signal states and adaptively adjust the gain execution mode according to different signal tendency states, thus affecting the reliability of adaptive gain control of GIS partial discharge signals.

[0006] To achieve the above objectives, this invention provides an adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor, comprising: Based on each triggering event within a historical monitoring period, several event timestamps and the associated power frequency phase angles of the event timestamps are determined to construct a historical event data set; The triggering event is determined based on the UHF signal; Based on the historical event data set, several phase filling parameters are determined respectively, and time series analysis is performed on the historical event data set to determine the trigger rhythm parameters. The time series analysis includes calling the timestamps of each event in the historical event data set to determine several interval durations. The signal state is determined and the risky operating period is screened based on the phase fill parameter and the trigger rhythm parameter of the same operating condition period; Based on the signal state, determine whether to call the UHF time-domain curves of each of the risk conditions to determine the common source trend parameters. Mark the time period trend state according to the common source trend parameters of the risk conditions. The time period trend state includes strong dominant trend state and weak dominant trend state. The time-period tendency of each risk condition within the nearest sliding time window is obtained to determine the neighboring disturbance parameters. Based on the neighboring disturbance parameters, the current partial discharge signal gain is determined to perform defensive control or detection control.

[0007] Furthermore, the process of identifying triggering events and constructing a historical event dataset includes, The event is triggered based on the instantaneous amplitude of the UHF signal during the operating period; Acquire several trigger events during the same working condition period within the historical monitoring cycle; The event timestamps and power frequency phase angles of each triggered event are stored in a single data set to determine the historical event data set associated with the operating period.

[0008] Furthermore, the process of determining several phase filling parameters based on the historical event data set includes, Obtain several power frequency phase angles from a set of historical event data associated with the operating time period; The phase fill parameters for the operating period are determined based on the phase angle of each power frequency.

[0009] Furthermore, the process of performing time-series analysis on the historical event dataset to determine triggering rhythm parameters includes, Call the timestamps of each event in the historical event data set associated with the operating period; Arrange the timestamps of each event in chronological order; Obtain the interval between each adjacent event timestamp; The trigger rhythm parameters are determined based on the duration of each interval.

[0010] Furthermore, the process of determining signal status and screening for high-risk operating periods includes, If the phase fill parameter and the trigger rhythm parameter of the same working condition period meet the risk conditions, then the signal state is determined to be a risk signal state, and the working condition period that meets the risk conditions is selected as a risk working condition period. The risk condition is that the phase fill parameter does not exceed the preset phase fill reference value, and the trigger rhythm parameter does not exceed the preset trigger rhythm reference value.

[0011] Furthermore, the process of determining whether to invoke the UHF time-domain curves for each of the aforementioned risky operating conditions includes, If the signal status is the risk signal status, then the UHF time domain curves for each of the risk operating conditions will be called.

[0012] Furthermore, the process of marking the trend state during a time period includes, If the homogeneity tendency parameter during a risky operating period exceeds the preset homogeneity tendency benchmark value, then the risky operating period is marked as a strong dominant tendency state. If the homogeneous tendency parameter during the risky working period does not exceed the preset homogeneous tendency benchmark value, then the risky working period is marked as a weak dominant tendency state. The homogeneous tendency parameter is determined based on several UHF time-domain curves obtained during the same risky operating condition period.

[0013] Furthermore, the process of determining the neighboring disturbance parameters based on the time-prone state of each risk condition within the nearest neighbor sliding time window includes, Obtain the number of risky working conditions with a strong dominant tendency and the number of risky working conditions with a weak dominant tendency within the nearest sliding time window, respectively. The proximity disturbance parameters are determined based on the number of risky operating periods in the strongly dominant tendency state and the number of risky operating periods in the weakly dominant tendency state.

[0014] Furthermore, the process of determining the current partial discharge signal gain control method includes, If the proximity disturbance parameter exceeds the preset proximity disturbance reference value, the current partial discharge signal gain control mode is determined to be defensive control. If the proximity disturbance parameter does not exceed the preset proximity disturbance reference value, then the current partial discharge signal gain control mode is determined to be detection control.

[0015] Furthermore, the defense control includes reducing the gain factor of the front-stage variable gain amplifier and increasing the response time constant. The reduction in the gain factor of the front-stage variable gain amplifier is positively correlated with the amplitude characteristic quantity, and the increase in the response time constant is positively correlated with the homogeneity tendency parameter. The detection control includes increasing the gain of the front-stage variable gain amplifier and decreasing the response time constant. The increase in the gain of the front-stage variable gain amplifier is negatively correlated with the amplitude characteristic quantity, and the decrease in the response time constant is negatively correlated with the homogeneity tendency parameter. The amplitude characteristic quantity is determined based on the UHF time-domain curves of each risk condition period within the nearest sliding time window.

[0016] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention constructs a historical event data set by determining several event timestamps and the associated power frequency phase angles of each triggering event; determines phase filling parameters based on the historical event data set; performs time series analysis on the historical event data set to determine trigger rhythm parameters; determines the signal state and filters risky operating periods based on the phase filling parameters and trigger rhythm parameters of the same operating period; determines whether to call the UHF time domain curves of each risky operating period based on the signal state; marks the period tendency state based on the homogeneous trend parameters of the risky operating periods; and determines the current partial discharge signal gain control method based on the adjacent disturbance parameters. Thus, the present invention realizes the identification of the signal state and adaptively adjusts the gain execution method according to different signal tendency states, thereby improving the reliability of adaptive gain control of GIS partial discharge signals.

[0017] In particular, this invention determines the signal state and filters high-risk operating periods based on the phase fill parameter and trigger rhythm parameter of the same operating period. It can be understood that by constructing a historical event data set, the phase fill parameter and trigger rhythm parameter are used to identify the pollution risk of interference. At the same time, based on the dual joint judgment of the phase fill parameter and trigger rhythm parameter, most of the clean operating periods can be filtered out in the initial screening stage, which greatly reduces the period that needs to be called for fine analysis of the original waveform in the later stage, reduces the storage and computing power overhead of the system, provides a feasible path for the engineering deployment of long-term online monitoring, and thus realizes the identification of signal state and improves the reliability of adaptive gain control of GIS partial discharge signal.

[0018] In particular, this invention marks the time-period tendency state based on the same-source tendency parameter of the risky operating period. It can be understood that the UHF time-domain curve is called only under the risk signal state, so that the reading and processing of the original waveform data is only performed on the selected suspicious time periods. This avoids continuous waveform storage and overlap calculation of all historical data or real-time data streams. While ensuring accurate waveform morphology confirmation of the risky period, it significantly reduces storage occupation and computing resource overhead. At the same time, by introducing the same-source tendency parameter as a quantitative indicator of waveform morphology, it can effectively distinguish between a single external fixed interference source and real partial discharge or complex aliasing signals with variable waveforms. The time-period tendency state is clearly marked, providing a reliable basis for differentiated selection of gain control. This avoids the risk of misjudgment caused by the lack of morphology confirmation after relying solely on statistical features for initial screening, and improves the accuracy and reliability of interference identification. Thus, the marking of the time-period tendency state is realized, which improves the reliability of adaptive gain control of GIS partial discharge signals.

[0019] In particular, this invention determines the current partial discharge signal gain to perform defensive control when the neighboring disturbance parameter exceeds the preset neighboring disturbance reference value. It can be understood that by triggering defensive control when the neighboring disturbance parameter exceeds the preset neighboring disturbance reference value, active identification and pre-defense against persistent strong external interference are achieved. Based on the analysis of multiple risky operating conditions within the neighboring sliding time window, the gain and response speed are pre-adjusted in a feedforward manner before the strong interference causes saturation clipping of the ADC, avoiding permanent loss of waveform information caused by amplifier saturation. At the same time, through the driving mechanism that the gain reduction amplitude is positively correlated with the amplitude characteristic quantity and the response time constant growth amplitude is positively correlated with the same source tendency parameter, fine control is achieved. Under the premise of ensuring the safe operation of the amplifier, the usable dynamic range of the effective signal is preserved to the maximum extent. Thus, the gain execution mode is adaptively adjusted according to different signal tendency states, improving the reliability of adaptive gain control of GIS partial discharge signals.

[0020] In particular, this invention determines the current partial discharge signal gain and performs detection control under the condition that the neighboring disturbance parameter does not exceed the preset neighboring disturbance reference value. It can be understood that by triggering detection control when the neighboring disturbance parameter does not exceed the preset neighboring disturbance reference value, active adaptation and high-sensitivity detection of suspected partial discharge or complex weak signal environments are achieved. Based on the judgment that the weak dominant tendency period is dominant within the neighboring sliding time window, the gain of the front-stage variable gain amplifier is actively increased and the response time constant is shortened, so that the monitoring system enters a highly alert detection state for weak transient pulses. This provides signal link conditions for capturing weak partial discharge signals generated by early insulation defects. Through the driving mechanism that the gain increase amplitude is negatively correlated with the amplitude characteristic quantity and the response time constant decrease amplitude is negatively correlated with the same source tendency parameter, fine control is achieved. Under the premise of maximizing the detection sensitivity of weak signals, noise amplification and system instability caused by excessive amplification are avoided. This provides a reliable data basis for early warning of GIS insulation status. Furthermore, it realizes the adaptive adjustment of the gain execution mode according to different signal tendency states, improving the reliability of adaptive gain control of GIS partial discharge signals. Attached Figure Description

[0021] Figure 1 This is a flowchart illustrating the steps of the adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor, as described in an embodiment of the present invention. Figure 2 A flowchart illustrating the logic of determining signal status and filtering high-risk operating periods in an embodiment of the present invention; Figure 3 A flowchart illustrating the logic for determining whether to invoke the UHF time-domain curves for each risky operating condition period in an embodiment of the present invention; Figure 4 This is a flowchart illustrating the logic of marking time period tendency states in an embodiment of the present invention. Detailed Implementation

[0022] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0023] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0024] It should be noted that in the description of this invention, the terms "upper," "lower," "inner," "outer," etc., which indicate the direction or positional relationship, are based on the direction or positional relationship shown in the drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0025] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation" and "connection" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0026] Please see Figure 1 The diagram illustrates the steps of an adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor, according to an embodiment of the present invention. The method of this embodiment includes: Step S100: Based on each triggering event within the historical monitoring period, determine several event timestamps and the power frequency phase angle associated with the event timestamps to construct a historical event data set; The triggering event is determined based on the UHF signal; In this embodiment of the invention, the historical monitoring period refers to the time backtracking length used to construct a historical event data set. Preferably, those skilled in the art can flexibly set it according to the actual application scenario, or it can be set based on the equipment operation cycle. Newly commissioned equipment is set to a shorter cycle, and long-term operating equipment is set to a duration that matches the maintenance cycle or operation and maintenance inspection cycle. It can also be set based on a dynamic sliding window method, automatically adjusting the cycle length according to the density of triggering events, thereby improving the response sensitivity to changes in the interference environment while ensuring a sufficient statistical sample size.

[0027] Step S200: Determine several phase filling parameters according to the historical event data set, and perform time series analysis on the historical event data set to determine trigger rhythm parameters. The time series analysis includes calling the timestamps of each event in the historical event data set to determine several interval durations. Step S300: Determine the signal status and filter risky operating periods based on the phase fill parameter and the trigger rhythm parameter for the same operating period; Step S400: Based on the signal state, determine whether to call the UHF time-domain curves of each of the risk conditions to determine the common source trend parameters, and mark the time period trend state according to the common source trend parameters of the risk conditions. The time period trend state includes strong dominant trend state and weak dominant trend state. Step S500: Obtain the time-period tendency state of each risk condition period within the nearest sliding time window to determine the neighboring disturbance parameters, and determine the current partial discharge signal gain to perform defensive control or detection control based on the neighboring disturbance parameters.

[0028] Specifically, the process of determining triggering events and constructing a historical event data set includes: marking triggering events based on the instantaneous amplitude of UHF signals during the operating period; acquiring several triggering events during the same operating period within the historical monitoring cycle; and storing the event timestamp and power frequency phase angle of each triggering event into a single data set to determine the historical event data set associated with the operating period.

[0029] In this embodiment of the invention, the process of marking trigger events includes: acquiring several instantaneous amplitude values ​​of UHF signals, and marking events whose instantaneous amplitude values ​​exceed a preset instantaneous amplitude threshold as trigger events. In this embodiment, the preset instantaneous amplitude threshold is dynamically determined based on the system noise floor level.

[0030] It is understood that the preset instantaneous amplitude threshold can be pre-set in the system by those skilled in the art based on the voltage level of the GIS equipment, the sensitivity of the UHF sensor, the gain of the preamplifier, and the level of electromagnetic noise in the field, through on-site calibration experiments. Specifically, during the power outage maintenance window of the GIS equipment, all high-voltage equipment can be shut down while the monitoring system is kept running normally. The UHF signal collected at this time is a pure noise floor signal. The mean amplitude and standard deviation of the amplitude under this condition are recorded, and the preset instantaneous amplitude threshold is obtained accordingly.

[0031] In this embodiment of the invention, the operating period can be set by those skilled in the art based on the equipment load status, dividing the process of the equipment load current changing from operation to shutdown or from low load to high load into different operating periods, so as to ensure that the historical data within the same operating period have the same load boundary conditions.

[0032] Specifically, the process of determining several phase filling parameters based on the historical event data set includes: obtaining several power frequency phase angles within the historical event data set associated with the operating period; and determining the phase filling parameters for the operating period based on each power frequency phase angle.

[0033] In this embodiment of the invention, the process of determining the phase fill parameter includes: obtaining the standard deviation and mean of each power frequency phase angle within the historical event data set; determining the phase fill parameter based on the ratio of the standard deviation to the mean, thereby determining the relative relationship between the breadth and concentration of the phase distribution of triggering events on the power frequency cycle. A larger phase fill parameter results in a more dispersed and uniform distribution of triggering events on the power frequency cycle, with more phase intervals where triggering events occur; a smaller phase fill parameter results in a more concentrated and clustered distribution of triggering events on the power frequency cycle, with triggering events only appearing in one or more specific phase intervals, and a lower degree of phase fill.

[0034] It is understandable that real partial discharges are driven by power frequency alternating electric fields and only occur in specific polarity phase intervals, with relatively small phase fill parameters. External interferences such as communication signals and random noise are unrelated to power frequency electric fields and can occur at any position on the time axis, resulting in larger phase fill parameters.

[0035] Specifically, the process of performing time-series analysis on the historical event data set to determine the trigger rhythm parameters includes: calling the timestamps of each event in the historical event data set associated with the working period; arranging the timestamps of each event in chronological order; obtaining the interval duration between each adjacent event timestamp; and determining the trigger rhythm parameters based on each interval duration.

[0036] In this embodiment of the invention, the process of determining the trigger rhythm parameter includes: determining the variance of each interval duration as the trigger rhythm parameter, so as to determine the regularity or uniformity of the occurrence of the trigger event on the time axis. The smaller the trigger rhythm parameter, the more consistent and uniform the interval duration between adjacent triggers, and the more regular the occurrence of the trigger event on the time axis. The larger the trigger rhythm parameter, the more drastic the change in the interval duration between adjacent triggers, and the more random and sudden the trigger event.

[0037] It is understandable that external fixed interference sources are driven by crystal oscillator clocks, and their signal transmission intervals are controlled by hardware timers, resulting in relatively small triggering rhythm parameters; while real partial discharge is caused by the random breakdown process of insulation defects, and the breakdown time is affected by random factors such as the state of the insulation medium, voltage phase, and ambient temperature and humidity, resulting in relatively large triggering rhythm parameters.

[0038] Please see Figure 2 As shown, it is a logic flowchart of determining the signal status and filtering the risky operating conditions in an embodiment of the present invention. The process of determining the signal status and filtering the risky operating conditions includes, If the phase fill parameter and the trigger rhythm parameter of the same working condition period meet the risk conditions, then the signal state is determined to be a risk signal state, and the working condition period that meets the risk conditions is selected as a risk working condition period. If the phase fill parameter and the trigger rhythm parameter do not meet the risk conditions during the same working condition period, the signal state is determined to be a non-risk signal state, and no risk working condition period screening is performed. The risk condition is that the phase fill parameter does not exceed the preset phase fill reference value, and the trigger rhythm parameter does not exceed the preset trigger rhythm reference value.

[0039] In this embodiment of the invention, the process of determining the preset phase fill reference value includes: determining the preset phase fill reference value based on the phase fill parameter distribution characteristics of a clean operating condition period with only background noise, no external interference, and no partial discharge signal within a historical monitoring period, and a typical external interference period, to ensure that the phase fill difference between external interference and the normal clean environment can be effectively distinguished. When the preset phase fill reference value is set too high, although the external interference signal has a high degree of phase fill, if its phase fill parameter is lower than the excessively high reference value, it will lead to missed detection, causing subsequent waveform verification and defense gain adjustment to fail to trigger normally, weakening the system's anti-interference early warning capability; when the preset phase fill reference value is set too low, a small number of randomly triggered events in the clean operating condition period will have a phase fill parameter higher than the excessively low reference value due to the accidental dispersion of phase distribution, causing the system to frequently misjudge risk-free periods as risky operating conditions, increasing unnecessary subsequent waveform calls and calculation overhead, and reducing system operating efficiency. Preferably, the preset phase fill reference value can be dynamically adjusted, and the preset phase fill reference value can be periodically updated to adapt to background noise drift and environmental changes.

[0040] It is understandable that setting the preset phase fill reference value in the transition range between the lower limit of the phase fill parameter distribution of typical external interference signals and the upper limit of the phase fill parameter distribution during clean operating conditions allows the initial screening to be based on the effective identification of interference risks, while taking into account both the utilization efficiency of computing resources and the recall rate of risk identification.

[0041] In this embodiment of the invention, the process of determining the preset trigger rhythm reference value includes: determining the preset trigger rhythm reference value based on the variance distribution characteristics of the trigger interval duration within a typical external interference period and the variance distribution characteristics of the trigger interval duration within a real partial discharge period. This ensures that the difference between the regular emission characteristics of external fixed interference and the random burst characteristics of partial discharge signals can be effectively distinguished. If the preset trigger rhythm reference value is set too high, it will cause missed detections and weaken the system's ability to identify external interference; if the preset trigger rhythm reference value is set too low, some partial discharge signals with weak randomness or sporadic noise may be misjudged as interference risk periods by the system because the trigger interval happens to show a low variance in a limited sample, resulting in a waste of computational resources for subsequent waveform homogeneity analysis.

[0042] It is understandable that setting the preset trigger rhythm baseline value within the transition range between the upper limit of the trigger rhythm parameter distribution of typical external fixed interference and the lower limit of the trigger rhythm parameter distribution of partial discharge signal can avoid misscreening non-interference signals with weak randomness as risk periods and ensure that the initial screening achieves a balance between high recall and low false alarm rate.

[0043] Please see Figure 3 The diagram shown is a flowchart illustrating the logic of determining whether to invoke the UHF time-domain curves for each risky operating condition period according to an embodiment of the present invention. The process of determining whether to invoke the UHF time-domain curves for each risky operating condition period includes: If the signal status is the risk signal status, then the UHF time domain curves for each of the risk conditions are determined to be invoked. If the signal status is the non-risk signal status, then it is determined that the UHF time domain curves for each of the risky operating conditions will not be invoked.

[0044] In this embodiment of the invention, the process of determining the UHF time-domain curve includes: taking the occurrence time of each trigger event as the origin, and obtaining a curve composed of discrete sampling point sequences of instantaneous voltage changes over time by truncating preset lengths forward and backward. Those skilled in the art can flexibly set the preset lengths according to actual application scenarios. Preferably, it can be set based on the typical time-domain width of a UHF pulse, so that the truncated time-domain curve can completely contain the main waveform information of a single trigger pulse, avoiding the loss of key waveform features due to excessively short truncations or the introduction of invalid noise data due to excessively long truncations, and minimizing the amount of data reading and processing while ensuring the accuracy of overlap calculation.

[0045] Please see Figure 4 As shown, it is a logic flowchart of marking time period trend states according to an embodiment of the present invention. The process of marking time period trend states includes, If the homogeneity tendency parameter during a risky operating period exceeds the preset homogeneity tendency benchmark value, then the risky operating period is marked as a strong dominant tendency state. If the homogeneous tendency parameter during the risky working period does not exceed the preset homogeneous tendency benchmark value, then the risky working period is marked as a weak dominant tendency state. The homogeneous tendency parameter is determined based on several UHF time-domain curves obtained during the same risky operating condition period.

[0046] In this embodiment of the invention, the process of determining the common-origin trend parameter includes: determining the average overlap of several UHF time-domain curves within the same risk condition period as the common-origin trend parameter, so as to determine the consistency of the waveform morphology of multiple UHF time-domain curves within the same risk condition period. The larger the common-origin trend parameter, the higher the consistency of the waveform morphology of each UHF time-domain curve within the same period; the smaller the common-origin trend parameter, the more obvious the difference in waveform morphology of each UHF time-domain curve within the same period.

[0047] It is understandable that the transmitted signal of the external fixed interference source is controlled by the crystal oscillator clock and the fixed digital modulation algorithm. The UHF pulses generated by the same interference source at different times have high repeatability and consistency in the time domain. However, the actual partial discharge is caused by the random breakdown process at the insulation defect. The start time, breakdown voltage, discharge amount and corresponding pulse rise edge and decay oscillation characteristics of each discharge are all affected by random factors such as space charge accumulation, ambient temperature and air pressure. The waveform of each discharge pulse shows similar but different random fluctuation characteristics.

[0048] In this embodiment of the invention, the method for determining the degree of overlap is not limited. For example, the cosine similarity method can be used to determine it. The cosine similarity is calculated after the UHF time-domain curve segment is vectorized, and the obtained cosine similarity is determined as the degree of overlap. Of course, other methods can also be used, which will not be elaborated here.

[0049] In this embodiment of the invention, the process of determining the preset common-source tendency reference value includes: determining the preset common-source tendency reference value based on the overlap distribution characteristics of each UHF time-domain curve during a typical external fixed interference period and the overlap distribution characteristics of each UHF time-domain curve during a real partial discharge period, so as to ensure that the difference between the waveform characteristics of the external fixed interference and the waveform characteristics of the partial discharge signal can be effectively distinguished. If the preset common-source tendency reference value is set too high, the strong interference period that should trigger defense regulation will not be correctly identified, weakening the system's anti-saturation protection capability. If the preset common-source tendency reference value is set too low, the partial discharge period that should maintain high sensitivity detection will be misjudged as a strong dominant tendency state, triggering defense regulation and reducing gain, causing weak partial discharge signals to be suppressed or even missed, resulting in loss of detection timeliness and effectiveness.

[0050] It is understandable that setting the preset co-originating reference value in the transition range between the lower limit of the waveform overlap distribution of typical external fixed interference and the upper limit of the waveform overlap distribution of real partial discharge signal can avoid misjudging real partial discharge signal with random waveform fluctuations as interference, and maintain a high sensitivity detection capability for weak partial discharge signal while ensuring accurate triggering of defense control.

[0051] Specifically, the process of determining the neighboring disturbance parameter based on the time-period tendency of each risk condition period within the nearest neighbor sliding time window includes obtaining the number of risk condition periods with strong dominant tendency and the number of risk condition periods with weak dominant tendency within the nearest neighbor sliding time window; and determining the neighboring disturbance parameter based on the number of risk condition periods with strong dominant tendency and the number of risk condition periods with weak dominant tendency.

[0052] In this embodiment of the invention, the process of determining the proximity disturbance parameter includes: obtaining the sum of the number of risky operating conditions with a strong dominant tendency and the number of risky operating conditions with a weak dominant tendency; determining the ratio of the number of risky operating conditions with a strong dominant tendency to the sum of the number of time periods as the proximity disturbance parameter; and determining the proportion of risky operating conditions with a strong dominant tendency. The larger the proximity disturbance parameter, the more time periods within the nearest neighbor sliding time window are marked as strong dominant tendencies, indicating that external fixed interference has consistently and stably dominated in recent history, and the confidence level is high that the current time is in a strong interference environment. The smaller the proximity disturbance parameter, the lower the proportion of strong dominant tendencies within the nearest neighbor sliding time window, the more weak dominant tendencies dominate, the more complex and variable signal sources or suspected partial discharges have been in recent history, and the confidence level is high that the current time is in a clean or complex signal environment.

[0053] Specifically, the nearest neighbor sliding time window is determined based on multiple risky operating condition periods prior to the current time, and the nearest neighbor sliding time window includes multiple risky operating condition periods prior to the current time.

[0054] Specifically, the process of determining the current partial discharge signal gain control method includes, If the proximity disturbance parameter exceeds the preset proximity disturbance reference value, the current partial discharge signal gain control mode is determined to be defensive control. If the proximity disturbance parameter does not exceed the preset proximity disturbance reference value, then the current partial discharge signal gain control mode is determined to be detection control.

[0055] In this embodiment of the invention, the process of determining the preset neighboring disturbance reference value includes: determining the preset neighboring disturbance reference value based on the distribution characteristics of neighboring disturbance parameters during periods when strong interference continuously dominates and the distribution characteristics of neighboring disturbance parameters during periods when the proportion of periods with a strong dominant tendency within the window is relatively low. This ensures that the difference between two different environmental states—one where interference is continuously dominant and the other where the signal is complex and variable—can be effectively distinguished, thereby achieving accurate switching of the gain control mode. If the preset neighboring disturbance reference value is set too small, it is easy to misjudge a transitional state within the window, where there are periods with a certain strong dominant tendency but the overall situation does not yet constitute continuous strong interference, as a strong interference environment. This leads to premature entry into the defense mode, unnecessary reduction of gain and increase of response time constant, weakening the detection sensitivity to possible weak partial discharge signals, and increasing the risk of missed detection. If the preset neighboring disturbance reference value is set too large, it will cause the amplifier to repeatedly enter the saturation state under continuous strong signal impact, resulting in the permanent loss of key waveform information. Furthermore, the gain oscillation caused by an excessively small response time constant may cause severe distortion of the output signal, losing its value for subsequent analysis.

[0056] It is understandable that setting the preset neighboring disturbance baseline value within the transition range between the upper limit of the neighboring disturbance parameter distribution during periods of strong dominance with a low proportion and the lower limit of the neighboring disturbance parameter distribution during periods of strong interference duration can ensure effective prevention of saturation and oscillation under strong interference conditions, while avoiding damage to the sensitivity and signal integrity of the monitoring system due to switching too early or too late.

[0057] Specifically, the defense control includes reducing the gain factor of the front-stage variable gain amplifier and increasing the response time constant. The reduction in the gain factor of the front-stage variable gain amplifier is positively correlated with the amplitude characteristic quantity, and the increase in the response time constant is positively correlated with the homogeneity tendency parameter. The detection control includes increasing the gain of the front-stage variable gain amplifier and decreasing the response time constant. The increase in the gain of the front-stage variable gain amplifier is negatively correlated with the amplitude characteristic quantity, and the decrease in the response time constant is negatively correlated with the homogeneity tendency parameter. The amplitude characteristic quantity is determined based on the UHF time-domain curves of each risk condition period within the nearest sliding time window.

[0058] In this embodiment, the process of determining the amplitude characteristic quantity includes: determining the average peak voltage of the UHF time-domain curve of each risk condition period within the nearest sliding time window as the amplitude characteristic quantity, so as to determine the actual intrusion intensity level of the UHF signal in the recent risk period.

[0059] Understandably, the larger the amplitude characteristic, the stronger the signal intrusion and the greater the threat to ADC saturation. Therefore, a larger gain attenuation is required to pull the signal back to the linear operating range, achieving on-demand saturation prevention. This avoids both insufficient attenuation leading to saturation clipping and excessive attenuation wasting the effective bit width of the ADC. The larger the homogeneity parameter, the more fixed the signal waveform, and the less the AGC needs to quickly track signal changes. In this case, a significant increase in the response time constant makes the AGC sluggish, effectively avoiding gain oscillation caused by rapid tracking of continuous wave interference, ensuring a stable output signal.

[0060] Understandably, the smaller the amplitude characteristic, the weaker the signal and the further it is from the optimal quantization range of the ADC. Therefore, a larger gain boost is needed to bring the weak signal into the high-resolution region of the ADC, making full use of the ADC's effective bit width and avoiding signal detail loss due to insufficient quantization bit width utilization. The smaller the homogeneity parameter, the more scattered and random the waveform, and the more it conforms to the random breakdown characteristics of real partial discharge. The rise time of the partial discharge pulse can be as short as nanoseconds, requiring the AGC to have an extremely fast response speed to track the pulse amplitude change in time and avoid distortion caused by the clipping of the first peak due to response lag. Therefore, the lower the homogeneity parameter, the greater the reduction in response time constant, in order to match the steep rise time of the partial discharge pulse.

[0061] All technologies not mentioned in the above embodiments are applicable to existing technologies. It is understood that no specific limitation is made to any preset parameter or critical parameter in the embodiments of the present invention, and the above values ​​are not limited thereto. Those skilled in the art can adjust the preset parameters or critical parameters accordingly based on actual needs, analysis of historical data, or equipment usage.

[0062] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0063] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor, characterized in that, include: Based on each triggering event within a historical monitoring period, several event timestamps and the associated power frequency phase angles of the event timestamps are determined to construct a historical event data set; The triggering event is determined based on the UHF signal; Based on the historical event data set, several phase filling parameters are determined respectively, and time series analysis is performed on the historical event data set to determine the trigger rhythm parameters. The time series analysis includes calling the timestamps of each event in the historical event data set to determine several interval durations. The signal state is determined and the risky operating period is screened based on the phase fill parameter and the trigger rhythm parameter of the same operating condition period; Based on the signal state, determine whether to call the UHF time-domain curves of each of the risk conditions to determine the common source trend parameters. Mark the time period trend state according to the common source trend parameters of the risk conditions. The time period trend state includes strong dominant trend state and weak dominant trend state. The time-period tendency of each risk condition within the nearest sliding time window is obtained to determine the neighboring disturbance parameters. Based on the neighboring disturbance parameters, the current partial discharge signal gain is determined to perform defensive control or detection control.

2. The adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor according to claim 1, characterized in that, The process of identifying triggering events and building a historical event dataset includes, The event is triggered based on the instantaneous amplitude of the UHF signal during the operating period; Acquire several trigger events during the same working condition period within the historical monitoring cycle; The event timestamps and power frequency phase angles of each triggered event are stored in a single data set to determine the historical event data set associated with the operating period.

3. The adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor according to claim 2, characterized in that, The process of determining several phase filling parameters based on the historical event data set includes: Obtain several power frequency phase angles from a set of historical event data associated with the operating time period; The phase fill parameters for the operating period are determined based on the phase angle of each power frequency.

4. The adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor according to claim 3, characterized in that, The process of performing time-series analysis on the historical event dataset to determine triggering rhythm parameters includes, Call the timestamps of each event in the historical event data set associated with the operating period; Arrange the timestamps of each event in chronological order; Obtain the interval between each adjacent event timestamp; The trigger rhythm parameters are determined based on the duration of each interval.

5. The adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor according to claim 4, characterized in that, The process of determining signal status and screening for high-risk operating conditions includes, If the phase fill parameter and the trigger rhythm parameter of the same working condition period meet the risk conditions, then the signal state is determined to be a risk signal state, and the working condition period that meets the risk conditions is selected as a risk working condition period. The risk condition is that the phase fill parameter does not exceed the preset phase fill reference value, and the trigger rhythm parameter does not exceed the preset trigger rhythm reference value.

6. The adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor according to claim 5, characterized in that, The process of determining whether to invoke the UHF time-domain curves for each of the aforementioned risky operating conditions includes, If the signal status is the risk signal status, then the UHF time domain curves for each of the risk operating conditions will be called.

7. The adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor according to claim 6, characterized in that, The process of marking time-prone states includes, If the homogeneity tendency parameter during a risky operating period exceeds the preset homogeneity tendency benchmark value, then the risky operating period is marked as a strong dominant tendency state. If the homogeneous tendency parameter during the risky working period does not exceed the preset homogeneous tendency benchmark value, then the risky working period is marked as a weak dominant tendency state. The homogeneous tendency parameter is determined based on several UHF time-domain curves obtained during the same risky operating condition period.

8. The adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor according to claim 7, characterized in that, The process of determining the neighboring disturbance parameters based on the time-prone state of each risk condition within the nearest neighbor sliding time window includes: Obtain the number of risky working conditions with a strong dominant tendency and the number of risky working conditions with a weak dominant tendency within the nearest sliding time window, respectively. The proximity disturbance parameters are determined based on the number of risky operating periods in the strongly dominant tendency state and the number of risky operating periods in the weakly dominant tendency state.

9. The adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor according to claim 8, characterized in that, The process of determining the current partial discharge signal gain control method includes, If the proximity disturbance parameter exceeds the preset proximity disturbance reference value, the current partial discharge signal gain control mode is determined to be defensive control. If the proximity disturbance parameter does not exceed the preset proximity disturbance reference value, then the current partial discharge signal gain control mode is determined to be detection control.

10. The adaptive gain control method for GIS partial discharge signals based on a UHF built-in sensor according to claim 9, characterized in that, The defense control includes reducing the gain factor of the front-stage variable gain amplifier and increasing the response time constant. The reduction in the gain factor of the front-stage variable gain amplifier is positively correlated with the amplitude characteristic quantity, and the increase in the response time constant is positively correlated with the homogeneity tendency parameter. The detection control includes increasing the gain of the front-stage variable gain amplifier and decreasing the response time constant. The increase in the gain of the front-stage variable gain amplifier is negatively correlated with the amplitude characteristic quantity, and the decrease in the response time constant is negatively correlated with the homogeneity tendency parameter. The amplitude characteristic quantity is determined based on the UHF time-domain curves of each risk condition period within the nearest sliding time window.

Images