CN-121479627-B - Early warning method driven by historical fault heat dissipation data of optical storage and charging system based on AI

CN121479627BCN 121479627 BCN121479627 BCN 121479627BCN-121479627-B

Abstract

The invention discloses an AI-based early warning method driven by historical fault heat dissipation data of an optical storage and filling system, which relates to the technical field of optical storage and filling system fault early warning, and comprises the following steps of integrating and recording historical operation data of the optical storage and filling system; the method comprises the steps of calculating the heating value and the heat dissipation efficiency of the optical storage and filling system, analyzing the heat dissipation influence relation, calibrating the heat dissipation influence relation through historical environment data to obtain a heat dissipation calibration influence relation, analyzing a fault judgment standard of the heat dissipation efficiency based on historical operation data, monitoring the heat dissipation state of the optical storage and filling system in real time, and predicting the fault of the optical storage and filling system based on the heat dissipation calibration influence relation and the fault judgment standard.

Inventors

ZENG XU
LI ZONGQI

Assignees

深圳驿普乐氏科技有限公司

Dates

Publication Date: 20260512
Application Date: 20260109

Claims (7)

1. The early warning method driven by the historical fault heat dissipation data of the optical storage and filling system based on the AI is characterized by comprising the following steps: integrating and recording historical data of the light storage and charging system, wherein the historical data comprises historical operation data, historical environment data and historical heat dissipation data; Calculating the heating value and the heat dissipation efficiency of the optical storage and filling system, and simultaneously analyzing the change relation between the heating value and the heat dissipation efficiency, and naming the change relation as a heat dissipation influence relation; Calibrating the heat dissipation influence relation through historical environment data to obtain a heat dissipation calibration influence relation, and analyzing a fault judgment standard of heat dissipation efficiency based on historical operation data; the heat radiation state of the optical storage and filling system is monitored in real time, and meanwhile, fault prediction is carried out on the optical storage and filling system based on a heat radiation calibration influence relation and a fault judgment standard; calculating the heating value and the heat dissipation efficiency of the optical storage and filling system, and simultaneously analyzing the change relation between the temperature before heat dissipation and the heat dissipation efficiency, wherein the heat dissipation influence relation is named as the following substeps: Calculating the pre-heat dissipation temperature and heat dissipation efficiency of the optical storage and filling system according to the historical operation data and the historical heat dissipation data; Analyzing the change relation between the temperature before heat dissipation and the heat dissipation efficiency to obtain a heat dissipation influence relation; The method for calculating the pre-heat dissipation temperature and the heat dissipation efficiency of the optical storage and charging system through the historical operation data and the historical heat dissipation data comprises the following substeps: the temperature before heat dissipation and the heat dissipation capacity of the optical storage and filling system can be obtained according to a heat conduction formula Q=m×c×delta T, wherein Q represents the heat quantity, m represents the mass of an object, c represents the specific heat capacity of the object, and delta T represents the temperature change; Calculating the difference between the temperature of the air inlet and the temperature of the air inlet to obtain the temperature change of the heat radiation system and the light storage and filling system after heat exchange, and recording the temperature change as delta T1; The method comprises the steps of obtaining the mass of air and the specific heat capacity of the air in a region where an optical storage and filling system is located, respectively marking as m1 and c1, substituting DeltaT 1, m1 and c1 into Delta T, m and c in a heat conduction formula respectively, solving to obtain heat dissipation capacity, and marking as QS; the parts capable of generating heat in the optical storage and filling system are named as heating parts, and the mass and specific heat capacity of the heating parts in the optical storage and filling system are acquired and respectively recorded as m2 and c2; substituting QS into Q in a heat conduction formula, substituting m2 and c2 into m and c in the heat conduction formula, solving to obtain a heat dissipation temperature difference of the heating component, and recording as TC; Adding TC and system temperature to obtain the temperature before heat dissipation of the optical storage and filling system, marking as TA, and calculating TC/TA to obtain heat dissipation efficiency; Analyzing the change relation between the temperature before heat dissipation and the heat dissipation efficiency to obtain a heat dissipation influence relation comprises the following substeps: Establishing a two-dimensional coordinate system with the temperature before heat dissipation as an X axis and the heat dissipation efficiency as a Y axis, naming the two-dimensional coordinate system as a temperature heat dissipation trend chart, and recording the heat dissipation efficiency into the temperature heat dissipation trend chart according to the temperature before heat dissipation; and performing function fitting on the temperature heat dissipation trend graph, and naming the fitted function as a temperature heat dissipation trend function, wherein the temperature heat dissipation trend function is the heat dissipation influence relation.
2. The AI-based optical storage and charging system historical fault heat dissipation data driven early warning method of claim 1, wherein the historical operating data comprises historical system temperatures in the optical storage and charging system, the historical environmental data comprises historical environmental temperatures, the historical heat dissipation data comprises historical air outlet volumes, air inlet temperatures and air outlet temperatures in the optical storage and charging system, and in addition, the historical operating data is divided into historical normal operating data and historical abnormal operating data according to whether the optical storage and charging system fails.
3. The AI-based optical storage and charging system historical fault heat dissipation data driving early warning method according to claim 2, wherein the heat dissipation influence relation is calibrated through historical environment data to obtain a heat dissipation calibration influence relation, and the fault judgment standard for analyzing heat dissipation efficiency based on historical operation data comprises the following sub-steps: Calibrating the heat dissipation influence relation through historical environment data to obtain a heat dissipation calibration influence relation; and analyzing a fault judgment standard of the heat radiation efficiency based on the heat radiation calibration influence relation and combining the historical operation data.
4. The AI-based optical storage and charging system historical fault heat dissipation data driven early warning method of claim 3, wherein the heat dissipation influence relationship is calibrated through historical environmental data, and the heat dissipation calibration influence relationship is obtained by the following sub-steps: the coordinate points in the temperature heat dissipation trend chart are named as temperature heat dissipation trend points, and the environmental temperature in the historical operation data of the temperature heat dissipation trend points is obtained and named as influence temperature; The values of X and Y of the temperature heat dissipation trend points are respectively marked as PX and PY, PX is substituted into the temperature heat dissipation trend function, the efficiency before calibration is obtained through solving, and the efficiency is marked as PCE; Calculating PY/PCE, and naming the calculation result as an efficiency calibration parameter, and recording the calculation result as EP; establishing a two-dimensional coordinate system by taking the influence temperature as a horizontal axis and the efficiency calibration parameter as a vertical axis, naming the two-dimensional coordinate system as a heat dissipation calibration analysis chart, and recording the EP into the heat dissipation calibration analysis chart according to the influence temperature; and performing function fitting on the heat radiation calibration analysis chart, and naming the function obtained by fitting as a heat radiation calibration function, wherein the heat radiation calibration function is the heat radiation calibration influence relation.
5. The AI-based optical storage-system historical-failure heat dissipation data-driven early warning method of claim 4, wherein the failure determination criteria based on heat dissipation calibration influence relationships and analyzing heat dissipation efficiency in combination with historical operating data includes the sub-steps of: the coordinate points which are obtained by analyzing and constructing the historical normal operation data and the historical abnormal operation data in the heat radiation calibration analysis chart are respectively named as historical normal points and historical abnormal points; constructing a curve of a heat radiation calibration function in the heat radiation calibration analysis chart, and naming the curve as a heat radiation calibration curve; Moving the heat radiation calibration curve vertically downwards until all the historical normal points are above the heat radiation calibration curve, so as to obtain a heat radiation calibration normal boundary; Vertically moving the heat radiation calibration curve downwards until the heat radiation calibration curve is intersected with the historical abnormal point for the first time to obtain a heat radiation calibration abnormal boundary; the method comprises the steps of (1) naming a region between a heat radiation calibration normal boundary and a heat radiation calibration abnormal boundary as a fuzzy region, copying a heat radiation calibration abnormal boundary and renaming the heat radiation calibration abnormal boundary as a fuzzy auxiliary line; Moving the fuzzy auxiliary line vertically from the heat radiation calibration abnormal boundary to the heat radiation calibration normal boundary, counting the number of the historical abnormal points above the fuzzy auxiliary line and the historical normal points below the fuzzy auxiliary line in real time, respectively recording the numbers as QU and QD, Counting the total number of the historical normal points and the historical abnormal points in the fuzzy area, calculating (QU+QD)/F by the symbol F, and naming the calculation result as error probability; And continuously moving the fuzzy auxiliary line, calculating error probability, and naming the fuzzy auxiliary line with the minimum error probability as a fault judgment standard.
6. The AI-based optical storage and charging system history fault heat dissipation data driven early warning method of claim 5, wherein the real-time monitoring of the heat dissipation state of the optical storage and charging system and the fault prediction of the optical storage and charging system based on the heat dissipation calibration influence relationship and the fault judgment criteria comprise the following sub-steps: On the premise that a self-checking system of the optical storage and filling system does not find a fault, acquiring the system temperature, the environment temperature, the air outlet volume, the air inlet temperature and the air outlet temperature of the optical storage and filling system in real time, and sequentially designating the system real-time temperature, the environment real-time temperature, the air outlet real-time volume, the air inlet real-time temperature and the air outlet real-time temperature as TK; Calculating the real-time pre-heat dissipation temperature and heat dissipation efficiency of the optical storage and charging system according to the real-time temperature of the system, the real-time volume of the air outlet, the real-time temperature of the air inlet and the real-time temperature of the air outlet, and sequentially named as the pre-heat dissipation real-time temperature and the heat dissipation real-time efficiency, and recording the heat dissipation real-time efficiency as RE; based on the real-time temperature before heat dissipation, the temperature heat dissipation trend function, the heat dissipation real-time efficiency and the environment real-time temperature, whether the optical storage and charging system has fault risks or not is analyzed.
7. The AI-based optical storage and retrieval system historical fault heat dissipation data driven pre-warning method of claim 6, wherein analyzing whether a fault risk exists in the optical storage and retrieval system based on pre-heat dissipation real-time temperature, a temperature heat dissipation trend function, heat dissipation real-time efficiency, and ambient real-time temperature includes the sub-steps of: substituting the real-time temperature before heat dissipation into a temperature heat dissipation trend function, solving to obtain reference efficiency, marking the reference efficiency as BE, calculating RE/BE, and marking the calculation result as H; Substituting the coordinates (TK, H) into a heat radiation calibration analysis chart and naming the heat radiation calibration analysis chart as a heat radiation real-time calibration point, judging whether the heat radiation real-time calibration point is above a fault judgment standard, if so, outputting a system normal signal, and if not, outputting a system abnormal signal; And if the system abnormal signal is output, sending early warning information to the maintenance end.

Description

Early warning method driven by historical fault heat dissipation data of optical storage and charging system based on AI Technical Field The invention relates to the technical field of fault early warning of an optical storage and filling system, in particular to an early warning method driven by historical fault heat dissipation data of an optical storage and filling system based on AI. Background The optical storage and filling system fault early warning technology is an active protection technology which is characterized in that early abnormal symptoms are identified before faults occur through real-time monitoring, data analysis and intelligent algorithms, and an alarm is sent out in advance, and the core aim is to realize the transition from post-maintenance to pre-prevention. The prior art of fault early warning of the optical storage and filling system generally only carries out fault prediction on the operation parameters of the optical storage and filling system, but does not carry out all-round fault prediction on the optical storage and filling system, because the optical storage and filling system is also provided with a heat dissipation system, the efficiency of the heat dissipation system has close relation with the health degree of the optical storage and filling system, if the optical storage and filling system is abnormal, the temperature of the optical storage and filling system is inevitably abnormal, the heat dissipation system at the moment can also be influenced, the prior art of fault early warning of the optical storage and filling system mostly adopts a threshold value judging mode when finally predicting, if the optical storage and filling system is abnormal, but does not exceed the threshold value, the optical storage and filling system cannot be accurately predicted on the basis of the heat dissipation system, for example, in the patent application with the publication number of CN118657509A, the scheme is disclosed as an operation and maintenance processing method and a system of the optical storage and filling system, the scheme is that the optical storage and filling system only carries out fault prediction on data such as the current voltage of the optical storage and filling system, meanwhile, the method is finally judging through the operation and maintenance limit value, a certain blind area is not capable of carrying out all-round fault prediction on the optical storage and filling system, and the optical storage and filling system cannot be accurately predicted on the optical storage and filling system when the optical storage and filling system cannot be completely predicted. Disclosure of Invention The invention aims to solve at least one of the technical problems in the prior art to a certain extent, by integrating and recording historical operation data of the optical storage and filling system, wherein the historical operation data comprises historical operation data, historical environment data and historical heat dissipation data, then the temperature before heat dissipation and heat dissipation efficiency of the optical storage and filling system are calculated through the historical operation data and the historical heat dissipation data, the change relation between the temperature before heat dissipation and the heat dissipation efficiency is analyzed to obtain a heat dissipation influence relation, then the heat dissipation influence relation is calibrated through the historical environment data to obtain a heat dissipation calibration influence relation, then the heat dissipation state of the optical storage and filling system is monitored in real time based on the heat dissipation calibration influence relation and a fault judgment standard for analyzing the heat dissipation efficiency by combining the historical operation data, and meanwhile, the fault prediction of the optical storage and filling system is carried out based on the heat dissipation calibration influence relation and the fault judgment standard, so that the problem that the fault early warning technology of the existing optical storage and filling system is not comprehensive and accurate enough, and the optical storage and filling system cannot be maintained in time is solved. In order to achieve the above purpose, the present application provides an AI-based early warning method for driving historical fault heat dissipation data of an optical storage and charging system, comprising the following steps: integrating and recording historical operation data of the light storage and charging system, wherein the historical operation data comprises historical operation data, historical environment data and historical heat dissipation data; Calculating the heating value and the heat dissipation efficiency of the optical storage and filling system, and simultaneously analyzing the change relation between the heating value and the heat dissipation efficiency, and naming the change relation as a heat