Search

CN-120999518-B - Ceramic module ferrule series connection surge cutting-off method and surge protector thereof

CN120999518BCN 120999518 BCN120999518 BCN 120999518BCN-120999518-B

Abstract

The invention discloses a ceramic module core insert serial connection surge cutting method and a surge protector thereof, belonging to the technical field of surge protection, comprising the steps of constructing a surge protection loop formed by connecting a plurality of ceramic module core insert units in series, wherein each core insert unit is internally provided with a surge absorption assembly and a thermal tripping assembly; when a surge event occurs, the nonlinear capacitive reactance migration factor and the longitudinal equivalent coupling impedance mutation rate of the surge absorbing assembly in each ferrule unit are detected, whether a set cut-off threshold is reached or not is judged according to the nonlinear capacitive reactance migration factor and the longitudinal equivalent coupling impedance mutation rate, a thermal tripping action is triggered to realize module-level cut-off, the position and the response time of the cut-off event are recorded to a memory guide layer in a coding mode for subsequent non-contact scanning and identification, after the system is recovered, the cut-off arm can be automatically reset in an external excitation mode to recover a conducting state, and the surge protection method and the surge protection device realize multiple technical effects of accurate surge response, traceable event and module reusability and are suitable for surge protection in high-density electronic equipment and intelligent power systems.

Inventors

  • ZHENG TIERU
  • ZHENG TIEYING
  • XU XUYANG
  • HAN DANDAN

Assignees

  • 浙江泰科天唯电气有限公司

Dates

Publication Date
20260508
Application Date
20250826

Claims (6)

  1. 1. The ceramic module ferrule series surge cutting-off method is characterized by comprising the following steps: A surge protection loop formed by connecting a plurality of ceramic module core inserting units in series is constructed, wherein each ceramic module core inserting unit contains a surge absorption component and a thermal tripping component; when a surge event occurs, the nonlinear capacitive reactance migration factor and the longitudinal equivalent coupling impedance mutation rate of the surge absorbing assembly in each ceramic module core inserting unit are detected, and whether the set cut-off threshold is reached is judged, wherein the nonlinear capacitive reactance migration factor is defined by measuring the instantaneous voltage V (t) and the current I (t) at two ends of a certain module by using a high-frequency sampler, and a series of data points at discrete moments are acquired In the formula (I), in the formula (II), Representing the ith sampling point voltage; Representing the i-th sample point current; Representing sampling time, calculating equivalent capacitive reactance sequence Using the basic relationship of capacitance: calculating the voltage change rate by numerical differentiation The expression is: And (3) carrying out the following steps: Calculating a nonlinear capacitive reactance migration factor NCD-F, and defining the nonlinear capacitive reactance migration factor as a derivative of capacitive reactance with respect to time; recording the position and response time code of the cut-off event to a memory guiding layer in the ceramic module for automatic diagnosis, position tracking and local module replacement; after the system is recovered, the cut-off arm with the automatic reset function can realize conduction recovery through external triggering in a surge-free state.
  2. 2. The method for series surge cutting off of ceramic module core insert according to claim 1, wherein the longitudinal equivalent coupling impedance mutation rate defines complex impedance of the core insert along the current axial path Mutation amplitude Rate in extremely short time The expression is: 。
  3. 3. The method for cutting off serial surges of ceramic module ferrules according to claim 2, wherein nonlinear capacitive reactance transfer factors and longitudinal equivalent coupling impedance mutation rates are converted into comprehensive feature vectors, the comprehensive feature vectors are used as inputs of a machine learning model, the machine learning model predicts surge boundary instability predicted value labels of each group of comprehensive feature vectors to serve as prediction targets, the sum of prediction errors of all the surge boundary instability predicted value labels is minimized to serve as a training target, the machine learning model is trained, model training is stopped until the sum of the prediction errors reaches convergence, and surge boundary instability predicted values are determined according to model output results, wherein the machine learning model is a polynomial regression model.
  4. 4. The method for cutting ceramic module insert core serial connection surge according to claim 3, wherein the obtained surge boundary instability predicted value is compared with a preset threshold value, if the surge boundary instability predicted value is larger than or equal to the preset threshold value, the insert core unit is judged to be in a surge out-of-control edge state, a TTU thermal tripping assembly is started immediately, module-level cutting operation is implemented, the breakpoint position is synchronously recorded in a module identification code for subsequent inspection and replacement, and if the surge boundary instability predicted value is smaller than the preset threshold value, adjustment is not needed.
  5. 5. The method for series surge shutdown of ceramic module ferrules according to claim 1 wherein the position and response time codes of the shutdown event are recorded to a memory guiding layer within the ceramic module, specifically: When the thermal tripping component triggers a cutting action, the driving response detection circuit acquires the position code and the triggering time stamp of the current core inserting unit; Converting the cut-off state signal into a double-bit pulse code signal, and injecting the double-bit pulse code signal into the memory guide layer through the coupling piezoelectric channel; the memory guide layer adopts a ceramic layer structure with ferroelectric hysteresis characteristic, and physical writing of event information is realized by utilizing controllable change of local polarization orientation; and in the system inspection stage, the polarization state is identified through non-contact electric field scanning, so that the cutting-off position and response time information are obtained through decoding.
  6. 6. The method for series surge cutting off of ceramic module ferrules according to claim 5, wherein the automatic diagnosis, position tracking and local module replacement comprises: periodically activating a non-contact electric field scanning array by a system control unit, and reading the polarization state of a memory guide layer in each ceramic module ferrule unit to obtain the marking information of the break event; matching the polarization data with the module address index matrix to reconstruct the specific module position and triggering time of the surge cutting event; generating a maintenance prompt instruction according to the cut-off position, and indicating a module to be replaced through a photoelectric label or a graphical interface; after the replacement is completed, the flag is cleared or reset by a double sweep confirmation.

Description

Ceramic module ferrule series connection surge cutting-off method and surge protector thereof Technical Field The invention relates to the technical field of surge protection, in particular to a ceramic module ferrule series surge cutting method and a surge protector thereof. Background Conventional surge protection devices typically employ a parallel configuration to absorb surge energy through a varistor or gas discharge tube. However, under high-frequency surge impact, the structure has the problems of rapid heat accumulation, rapid device aging, low protection precision and the like, and particularly in serial multi-module arrangement, if a certain stage is damaged and not separated in time, the whole group of protection failures are easy to cause. In the prior art, although a series cut-off method is also proposed, passive elements such as fuses, thermal fragments and the like are generally relied on, the response time is dependent on the thermal inertia of the device, and the nanosecond surge response requirement is difficult to adapt. In addition, most of the prior art cannot realize the position memory detachment or state tracking cut-off logic among the modules, so that the system is difficult to maintain and the misjudgment rate is high. Therefore, there is an urgent need for a serial surge cutting method with dynamic state memory capability, capable of constructing a recoverable separation path inside a ceramic module, which does not depend on an external electric control system, but combines three functions of surge response judgment, action triggering and module cutting in a structural body, so as to solve the problems of slow response, non-adjustable structure, non-resettable structure and the like in the prior art, and improve the accuracy and reliability of multistage surge protection. Disclosure of Invention The invention aims to provide a ceramic module ferrule series surge cutting method and a surge protector thereof, which solve the defects in the background technology. In order to achieve the purpose, the invention provides the following technical scheme that the ceramic module ferrule series surge cutting method comprises the following steps: A surge protection loop formed by connecting a plurality of ceramic module core inserting units in series is constructed, wherein each ceramic module core inserting unit contains a surge absorption component and a thermal tripping component; When a surge event occurs, detecting nonlinear capacitive reactance migration factors and longitudinal equivalent coupling impedance mutation rates of surge absorption components in each ceramic module core-insert unit, and judging whether a set cutting threshold is reached; recording the position and response time code of the cut-off event to a memory guiding layer in the ceramic module for automatic diagnosis, position tracking and local module replacement; after the system is recovered, the cut-off arm with the automatic reset function can realize conduction recovery through external triggering in a surge-free state. Preferably, the nonlinear capacitive reactance migration factor is defined by measuring the instantaneous voltage V (t) and current I (t) at two ends of a certain module by a high-frequency sampler, and acquiring a series of data points at discrete momentsIn the formula (I), in the formula (II),Representing the ith sampling point voltage; Representing the i-th sample point current; Representing sampling time, calculating equivalent capacitive reactance sequence Using the basic relationship of capacitance: calculating the voltage change rate by numerical differentiation The expression is: And (3) carrying out the following steps: and calculating a nonlinear capacitive reactance migration factor NCD-F, and defining the nonlinear capacitive reactance migration factor as the derivative of capacitive reactance with respect to time. Preferably, the longitudinal equivalent coupling impedance mutation rate defines the complex impedance of the insertion module along the current axial pathMutation amplitude Rate in extremely short timeThe expression is:。 Preferably, the nonlinear capacitive reactance transition factor and the longitudinal equivalent coupling impedance mutation rate are converted into comprehensive feature vectors, the comprehensive feature vectors are used as input of a machine learning model, each group of comprehensive feature vector prediction surge boundary instability predicted value labels are used as prediction targets, the sum of prediction errors of all surge boundary instability predicted value labels is minimized to be used as a training target, the machine learning model is trained until the sum of the prediction errors reaches convergence, model training is stopped, and the surge boundary instability predicted value is determined according to model output results, wherein the machine learning model is a polynomial regression model. Preferably, the obtained surge boundary i