CN-122017504-A - Photoelectric tube high-voltage excitation type solar blind ultraviolet discharge detection method and system

CN122017504ACN 122017504 ACN122017504 ACN 122017504ACN-122017504-A

Abstract

The invention relates to the technical field of photoelectric tube detection, in particular to a photoelectric tube high-voltage excitation type solar blind ultraviolet discharge detection method and system, comprising the steps of acquiring original voltage sequence data acquired by a high-speed transimpedance sampling circuit in real time; the method comprises the steps of positioning feature points of each suspected pulse according to original voltage sequence data, extracting pulse waveform features, obtaining a physical confidence factor of each suspected pulse according to falling edge attenuation characteristics, calculating effective discharge energy flow in unit time based on the physical confidence factor, calculating a discharge stability coefficient according to a coupling relation between a time-dependent entropy value and the effective discharge energy flow, and calculating a dynamic risk index according to the effective discharge energy flow, the physical confidence factor and the discharge stability coefficient.

Inventors

WANG BAOYI
LU YONG
DING ZEYU
YANG GUANG

Assignees

杭州众电智能技术有限公司

Dates

Publication Date: 20260512
Application Date: 20260413

Claims (10)

1. A photoelectric tube high-voltage excitation type solar blind ultraviolet discharge detection method is characterized by comprising the following steps: Acquiring original voltage sequence data acquired by a high-speed transimpedance sampling circuit in real time, wherein the original voltage sequence data is generated by converting a received solar blind ultraviolet light signal into an electric pulse signal under direct current bias provided by a high-voltage excitation module by a back-illuminated solar blind ultraviolet light tube with a built-in high-voltage electric structure and performing transimpedance sampling; Locating feature points of each suspected pulse according to the original voltage sequence data, and extracting pulse waveform features, wherein the pulse waveform features comprise peak points, rising edge slope distribution and falling edge attenuation characteristics; calculating an actual measurement attenuation constant of each suspected pulse according to the falling edge attenuation characteristic, and comparing the actual measurement attenuation constant with a pre-stored hardware inherent ionization quenching constant to obtain a physical confidence factor of each suspected pulse; Screening suspected pulses according to the physical confidence factor, determining candidate effective pulses, carrying out energy quantization processing on each candidate effective pulse, and carrying out energy deconstruction and correction on the candidate effective pulses with nonlinear stacks according to the rising edge slope distribution to obtain effective discharge energy flow in unit time; Acquiring time distribution characteristics of adjacent candidate effective pulses, performing regular quantitative analysis on the time distribution characteristics to obtain time-dependent entropy values, and further calculating a discharge stability coefficient according to a coupling relation between the time-dependent entropy values and the effective discharge energy flow; and constructing a current state vector according to the effective discharge energy flow, the physical confidence factor and the discharge stability coefficient, and calculating to obtain a dynamic risk index according to the characteristic distance of the current state vector in a preset insulation degradation space and the change slope of a historical state vector.
2. The method for detecting solar blind ultraviolet discharge according to claim 1, wherein the step of locating the feature point of each suspected pulse according to the original voltage sequence data, extracting pulse waveform features including peak point, rising edge slope distribution and falling edge attenuation characteristics comprises: performing waveform change rate analysis on the original voltage sequence data, positioning a starting point and an ending point of each suspected pulse based on zero crossing point characteristics of the waveform change rate, and determining a waveform interval of each suspected pulse according to the starting point and the ending point; in each waveform interval, positioning the peak value point of the suspected pulse with sub-sampling point precision based on local extremum analysis, and recording the peak value amplitude and peak value moment corresponding to the peak value point; And extracting rising edge data before the peak point according to the peak point and the waveform interval to obtain rising edge slope distribution characteristics, and intercepting falling edge data points in a preset time window after the peak point according to the peak point to construct a pulse attenuation sequence for representing pulse attenuation characteristics.
3. The method of claim 1, wherein the step of calculating an actual measurement decay constant of each suspected pulse according to the falling edge decay characteristic and comparing the actual measurement decay constant with a pre-stored hardware intrinsic ionization quenching constant to obtain a physical confidence factor of each suspected pulse comprises: performing exponential decay model fitting on the pulse decay sequence to obtain an actual measurement decay constant of the suspected pulse; Acquiring a direct current bias voltage applied to a photoelectric tube by a high-voltage excitation module at the current moment, and reading a hardware inherent ionization quenching constant of the photoelectric tube under the current direct current bias voltage from prestored parameters; calculating the deviation value of the actually measured attenuation constant and the inherent ionization quenching constant of the hardware; And mapping the deviation value into a physical confidence factor between 0 and 1 according to a preset deviation-confidence coefficient mapping relation, wherein the mapping relation satisfies that the smaller the deviation value is, the larger the physical confidence factor is, and when the deviation value exceeds a preset range, the physical confidence factor tends to be 0.
4. The method for detecting solar blind ultraviolet discharge according to claim 1, wherein the steps of screening suspected pulses according to the physical confidence factor, determining candidate effective pulses, performing energy quantization processing on each candidate effective pulse, and performing energy deconstructing and correcting on the candidate effective pulses having nonlinear stacks according to the rising edge slope distribution, so as to obtain an effective discharge energy flow in unit time include: screening suspected pulses with the physical confidence coefficient factor larger than a preset confidence coefficient threshold as candidate effective pulses; Performing energy quantization processing on each candidate effective pulse based on a waveform interval of each candidate effective pulse, and acquiring a feedback resistance value of a pre-stored high-speed trans-impedance sampling circuit and an internal ionization gain multiple of a photoelectric tube under the current direct current bias so as to perform dimension conversion and gain correction on the voltage time integral value to obtain an apparent charge quantity of the candidate effective pulse; analyzing the rising edge slope distribution of each candidate effective pulse, detecting whether waveform distortion characteristics representing superposition of a plurality of avalanche processes exist in the rising edge slope distribution, judging that nonlinear superposition exists in the candidate effective pulse if the waveform distortion characteristics exist, and dividing the corresponding pulse waveform into a plurality of sub-pulse units according to the waveform distortion characteristics; Respectively carrying out energy quantization processing on each sub-pulse unit, and correcting by combining hardware inherent parameters to obtain corresponding sub-unit charge quantity; comparing and verifying the subunit charge quantity with the apparent charge quantity of the corresponding candidate effective pulse to obtain the subunit charge quantity passing verification; and acquiring the corrected effective charge quantity of the subunit according to the verified charge quantity of the subunit and the physical confidence factor of the corresponding candidate effective pulse, and accumulating to obtain the effective discharge energy flow in unit time.
5. The solar blind ultraviolet discharge detection method according to claim 1, wherein the step of obtaining the time distribution characteristics of the adjacent candidate effective pulses, performing regular quantitative analysis on the time distribution characteristics to obtain a time-dependent entropy value, and further calculating a discharge stability coefficient according to a coupling relation between the time-dependent entropy value and the effective discharge energy flow comprises: Constructing a time interval sequence of adjacent pulses according to the time stamp of the candidate effective pulse; carrying out probability distribution statistics on the time interval sequence, and calculating a time-dependent entropy value of the time interval sequence based on an information entropy theory, wherein the time-dependent entropy value is used for quantifying the regularity degree of pulse time distribution; And calculating a discharge stability coefficient according to the coupling relation between the time-dependent entropy value and the effective discharge energy flow, wherein the discharge stability coefficient is used for comprehensively representing the energy intensity and the time regularity of the discharge process.
6. The solar blind ultraviolet discharge detection method according to claim 1, wherein the step of constructing a current state vector according to the effective discharge energy flow, the physical confidence factor and the discharge stability factor, and calculating a dynamic risk index according to a characteristic distance of the current state vector in a preset insulation degradation space and a change slope of a historical state vector comprises: calculating the average value of the physical confidence factors of all candidate effective pulses in unit time; Normalizing the effective discharge energy flow, the average value of the physical confidence factor and the discharge stability coefficient to construct a current state vector; Acquiring a standard state vector set in a preset insulation degradation space, wherein the standard state vectors respectively represent different risk levels; determining an instantaneous dangerous value of the current state according to the similarity between the current state vector and each standard state vector; Reading state vector sequences of a plurality of previous time windows from a historical database, and analyzing the evolution trend of the state vector sequences to obtain the change rate of the risk; and calculating a dynamic risk index based on the instantaneous risk value and the change rate, wherein the dynamic risk index is used for reflecting the current discharge risk and the development trend thereof.
7. The method for detecting solar blind ultraviolet discharge according to claim 1, wherein after the step of calculating the dynamic risk index, the method further comprises: triggering the alarm of the corresponding level according to the relation between the dynamic risk index and a plurality of preset risk thresholds, and uploading corresponding state information.
8. The method for detecting solar blind ultraviolet discharge according to claim 7, wherein after the step of calculating the dynamic risk index, the method further comprises performing an adaptive feedback adjustment procedure: counting the energy contribution duty ratio of low-confidence pulses in unit time in real time, and taking the energy contribution duty ratio as the noise base level of a current time window, wherein the low-confidence pulses refer to suspected pulses with physical confidence factors not larger than a preset confidence threshold; Comprehensively judging the type of the current abnormal state according to the change trend of the statistical characteristics of the physical confidence factor, the change trend of the dynamic risk index and the noise base level; And generating a feedback adjustment instruction based on the type of the current abnormal state, and sending the feedback adjustment instruction to a high-voltage excitation module so as to finely adjust the direct-current bias voltage value of the photoelectric tube.
9. A photoelectric tube high-voltage excitation type solar blind ultraviolet discharge detection system is characterized by comprising: The ultraviolet detection unit adopts a back-illuminated solar blind ultraviolet photoelectric tube with a built-in high-voltage ionization structure, and the photoelectric tube converts a received solar blind ultraviolet signal into avalanche electric pulse with an inherent ionization quenching characteristic under the preset direct-current bias provided by the high-voltage excitation module; the high-voltage excitation module is electrically connected with the ultraviolet detection unit and is used for providing an adjustable direct-current bias voltage for the built-in high-voltage ionization structure so as to control the internal ionization gain multiplying power; The high-speed transimpedance sampling circuit is directly connected with the output end of the ultraviolet detection unit and is used for capturing the avalanche electric pulse and converting the avalanche electric pulse into original voltage sequence data; The edge computing core is connected with the high-speed transimpedance sampling circuit and internally provided with a nonvolatile memory, wherein the memory is pre-stored with hardware physical characteristic parameters of the photoelectric tube, and the hardware physical characteristic parameters at least comprise inherent ionization quenching constants of the photoelectric tube.
10. The photocell high voltage excitation type solar blind ultraviolet discharge detection system according to claim 9, wherein the edge calculation core is configured to run the following logic modules in real time: the signal acquisition module is used for acquiring original voltage sequence data acquired by the high-speed transimpedance sampling circuit in real time; The data extraction module is used for positioning the characteristic point of each suspected pulse according to the original voltage sequence data and extracting pulse waveform characteristics; The pulse evaluation module is used for calculating the actual measurement attenuation constant of each suspected pulse according to the falling edge attenuation characteristic, and comparing the actual measurement attenuation constant with a pre-stored hardware inherent ionization quenching constant to acquire a physical confidence factor of each suspected pulse; The screening and reconstructing module is used for screening suspected pulses according to the physical confidence factor, determining candidate effective pulses, carrying out energy quantization processing on each candidate effective pulse, and carrying out energy deconstruction and correction on the candidate effective pulses with nonlinear stacks according to the rising edge slope distribution to obtain effective discharge energy flow in unit time; The stability analysis module is used for acquiring time distribution characteristics of adjacent candidate effective pulses, performing regular quantitative analysis on the time distribution characteristics to obtain a time-dependent entropy value, and further calculating a discharge stability coefficient according to a coupling relation between the time-dependent entropy value and the effective discharge energy flow; The risk assessment module is used for constructing a current state vector according to the effective discharge energy flow, the physical confidence factor and the discharge stability coefficient, and calculating to obtain a dynamic risk index according to the characteristic distance of the current state vector in a preset insulation degradation space and the change slope of a historical state vector; the alarm communication module is used for triggering the alarms of the corresponding levels according to the relation between the dynamic risk indexes and a plurality of preset risk thresholds and uploading corresponding state information; And the feedback adjustment module is used for generating a feedback adjustment instruction according to the change trend of the physical confidence coefficient factor and the dynamic risk index and sending the feedback adjustment instruction to the high-voltage excitation module so as to finely adjust the direct-current bias voltage value of the photoelectric tube.

Description

Photoelectric tube high-voltage excitation type solar blind ultraviolet discharge detection method and system Technical Field The invention relates to the technical field of photoelectric tube detection, in particular to a photoelectric tube high-voltage excitation type solar blind ultraviolet discharge detection method and system. Background In the partial discharge monitoring of the power system, solar blind ultraviolet detection has extremely high signal to noise ratio because the solar blind ultraviolet detection avoids solar spectrum interference. The back-illuminated solar blind ultraviolet photoelectric tube integrates a high-voltage ionization structure in the tube body, and generates secondary ionization avalanche of photo-generated electrons by utilizing a strong electric field generated by high-voltage excitation, so that the high-magnification gain of signals is realized on a physical level, and the system can capture weak discharge signals without a complex external multistage amplifying circuit. However, in this high-voltage excitation operation mode, the physical process inside the photocell becomes complicated. On one hand, the high-pressure electric field can induce random avalanche noise excited by hot electrons while amplifying ultraviolet signals, and the noise is similar to real ultraviolet pulse in time domain morphology, and on the other hand, as the partial discharge intensity is increased, the time density of photons reaching a photosurface is increased, so that nonlinear pulse stacking effect occurs in the internal ionization avalanche process. The existing detection method mostly adopts pulse counting or amplitude accumulation, thermal noise and weak signals are difficult to distinguish from physical mechanisms, and real discharge energy cannot be accurately quantized when pulse stacking is serious, so that contradiction exists between detection sensitivity and quantization precision. Disclosure of Invention The invention mainly aims to provide a photoelectric tube high-voltage excitation type solar blind ultraviolet discharge detection method and system, and aims to solve the problems of 'avalanche noise interference' of a high-voltage excitation type photoelectric tube under extremely weak signals and 'pulse stacking quantification misalignment' under strong discharge. By excavating the physical attenuation characteristic of the avalanche process in the ionization tube, a detection method based on ionization dynamics consistency test and energy flow time sequence evolution analysis is constructed, the deep extraction of discharge signals and the accurate quantification of the risk degree are realized, and meanwhile, the extreme simplicity of a hardware architecture is maintained. The invention provides a photoelectric tube high-voltage excitation type solar blind ultraviolet discharge detection method, which comprises the following steps: Acquiring original voltage sequence data acquired by a high-speed transimpedance sampling circuit in real time, wherein the original voltage sequence data is generated by converting a received solar blind ultraviolet light signal into an electric pulse signal under direct current bias provided by a high-voltage excitation module by a back-illuminated solar blind ultraviolet light tube with a built-in high-voltage electric structure and performing transimpedance sampling; Locating feature points of each suspected pulse according to the original voltage sequence data, and extracting pulse waveform features, wherein the pulse waveform features comprise peak points, rising edge slope distribution and falling edge attenuation characteristics; calculating an actual measurement attenuation constant of each suspected pulse according to the falling edge attenuation characteristic, and comparing the actual measurement attenuation constant with a pre-stored hardware inherent ionization quenching constant to obtain a physical confidence factor of each suspected pulse; Screening suspected pulses according to the physical confidence factor, determining candidate effective pulses, carrying out energy quantization processing on each candidate effective pulse, and carrying out energy deconstruction and correction on the candidate effective pulses with nonlinear stacks according to the rising edge slope distribution to obtain effective discharge energy flow in unit time; Acquiring time distribution characteristics of adjacent candidate effective pulses, performing regular quantitative analysis on the time distribution characteristics to obtain time-dependent entropy values, and further calculating a discharge stability coefficient according to a coupling relation between the time-dependent entropy values and the effective discharge energy flow; and constructing a current state vector according to the effective discharge energy flow, the physical confidence factor and the discharge stability coefficient, and calculating to obtain a dynamic risk index according to