CN-122017565-A - Backup power supply health condition analysis method, system, equipment and medium based on AI deep learning and battery big data processing

CN122017565ACN 122017565 ACN122017565 ACN 122017565ACN-122017565-A

Abstract

The invention discloses a backup power supply health condition analysis method, a system, equipment and a medium based on AI deep learning and battery big data processing, which belong to the technical field of backup power supply health condition analysis and comprise the steps of collecting operation parameters of single batteries and battery packs in real time through cooperation of sampling plates and sensors, synchronously recording time stamps and equipment identifiers of the parameters in the collecting process, measuring the internal resistance of the batteries, switching a detection method to calibrate the internal resistance data by combining discharge current when the internal resistance of the batteries is too small, establishing a backup power supply health criterion comprising a plurality of criteria, establishing a battery health condition assessment model and a fault prediction model based on a deep learning algorithm, dynamically assessing the battery health level and predicting the occurrence period of faults, comparing the collected data with a criterion threshold in real time, and triggering an alarm mechanism when any criterion is detected to meet abnormal conditions. The invention realizes the differential monitoring and health management of the storage battery, the lithium battery and the super capacitor backup power supply.

Inventors

HAO LIPING
DU JUAN
LI ZHIYU
LAN HAI
KOU JIN
ZHANG LIBO
GAO DAN
WANG CHAO
YUAN JINMING
YANG QIANG
XIANG XUAN
TAN YANYI
MAO YARU
ZHOU ZHONGQIANG
TAO YONGWEI
Pi Dinggui
TIAN YU
WANG LEI
ZHOU BING
FU TONGFU

Assignees

贵州电网有限责任公司

Dates

Publication Date: 20260512
Application Date: 20251231

Claims (10)

1. The backup power supply health condition analysis method based on AI deep learning and battery big data processing is characterized by comprising the following steps of, The operation parameters of the single battery and the battery pack are collected in real time through the cooperative work of the sampling plate and the sensor, and the time stamp and the equipment identifier of each parameter are synchronously recorded in the collection process; Measuring the internal resistance of the battery by adopting a detection method, calculating the internal resistance of the battery according to the real-time change values of the discharge voltage and the current, switching the detection method when the internal resistance of the battery is too small, and calibrating the internal resistance data by combining the discharge current; Establishing a backup power health criterion comprising a plurality of criterion types; Constructing a battery health state evaluation model and a fault prediction model, dynamically evaluating the battery health state and predicting the occurrence period of the fault; and comparing the acquired data with the criterion threshold in real time, triggering an alarm mechanism when any criterion is detected to meet the abnormal condition, and synchronously recording abnormal parameters, occurrence time and equipment position information.
2. The method for analyzing the health condition of the backup power supply based on AI deep learning and battery big data processing of claim 1, wherein the operation parameters of the single battery and the battery pack are collected in real time through the cooperation of the sampling plate and the sensor, and the time stamp and the equipment identifier of each parameter are synchronously recorded in the collection process, The battery is connected with the sensor, and analog-to-digital conversion is carried out on battery data through the sampling plate, and analog signals are converted into digital signals; the non-differential monitoring of the single battery is realized through the storage of the digital signals; Collecting operation parameters of the single batteries and the battery pack in real time, and comprehensively covering the operation state characteristics of the batteries; And after the equipment is electrified, the sensor zero drift compensation and the temperature and humidity self-adaptive gain adjustment are automatically executed, and parameter correction is carried out by combining the pre-calibrated temperature change parameters through the comparison of the detection of the ambient temperature and humidity and the battery running temperature.
3. The method for analyzing the health condition of the backup power supply based on AI deep learning and battery big data processing of claim 2, wherein the detecting method is used for measuring the internal resistance of the battery, calculating the internal resistance of the battery according to the real-time change values of the discharge voltage and the current, and when the internal resistance of the battery is too small, switching the detecting method and calibrating the internal resistance data in combination with the discharge current, comprising, Dividing the battery pack into a plurality of loops, sequentially discharging according to a loop sequence when measuring the internal resistance, and synchronously collecting a discharge curve of the battery; Calculating initial internal resistance of the battery according to the discharge voltage drop acquisition data; collecting the real-time change values of the discharge voltage and the current, analyzing and calculating the internal resistance of the storage battery, and automatically switching the detection method when the internal resistance of the storage battery is too small; And controlling a discharge load to ensure continuous and stable discharge current, acquiring data values after multiple measurements, and calculating an average value of acquired data by a processor to deduce stable internal resistance data.
4. The method for analyzing the health of the backup power supply based on AI deep learning and battery big data processing of claim 3, wherein the establishing comprises a plurality of criteria type backup power supply health criteria including, Respectively establishing health criteria for different types of backup power supplies; the type of criteria in the health criteria includes a criterion based on battery operating parameters; Setting a corresponding threshold range for each criterion type; and (3) comprehensively evaluating the health condition of the backup power supply by using the combination of the health criteria.
5. The method for analyzing the health status of the backup power supply based on AI deep learning and battery big data processing of claim 4, wherein constructing a battery health status assessment model and a fault prediction model, dynamically assessing the battery health status and predicting the occurrence period of the fault comprises, Transmitting the data to a temporary database, automatically screening the data to remove error data, and transmitting the rest data to a core database; Calculating the operation data of the storage battery through a balance algorithm, and carrying out parameter correction on the operation data and preset data; comparing the corrected data with the standard storage battery operation data to obtain a final operation condition; Alarming the storage battery with abnormal internal resistance and abnormal capacity; when the calculated value and the actual data are not reached, recording the actual data for carrying out data comparison and research again, carrying out difference calculation on the error data and the actual data, adjusting by combining the original set data, recording the research and judgment result and updating the preset database.
6. The method for analyzing the health condition of the backup power supply based on AI deep learning and battery big data processing of claim 5, wherein when the calculated value and the actual value are not reached, recording the actual data for performing data comparison and research again, performing difference calculation on the error data and the actual data, adjusting by combining the original setting data, recording the research result and updating a preset database, Judging the test value of the operation condition and the actual operation condition for real-time comparison; recording actual data into alarm information of judgment errors, which is not actually reached by the calculated value, and clicking to return to verification; performing difference value calculation on error data of calculation reference and recorded actual data; Increasing and decreasing by combining the original setting data; Automatically recording a research and judgment result, and inputting the result into a preset database to replace original setting data; and when the same-class storage battery operation data appears again, starting new data to carry out comparison correction.
7. The method for analyzing the health condition of the backup power supply based on AI deep learning and battery big data processing of claim 6, wherein the real-time comparison of the collected data and the criterion threshold value, when any criterion is detected to meet the abnormal condition, triggering an alarm mechanism, synchronously recording the abnormal parameters, the occurrence time and the equipment position information, comprises, When the model predicts that the state of health of the battery is abnormal and the life decay is close to a critical value or fails, the system automatically sends out an alarm prompt and battery replacement early warning; continuously carrying out alarm prompt on a monitoring platform, and manually confirming that an alarm party can reset alarm information by an operation and maintenance management platform and related personnel; based on the battery health state evaluation result, establishing a personalized maintenance scheme for each battery; and encrypting the acquired data, the health state evaluation result and the alarm data through the communication module and uploading the encrypted data to the master station.
8. The backup power supply health condition analysis system based on AI deep learning and battery big data processing is applied to the backup power supply health condition analysis method based on AI deep learning and battery big data processing according to any one of claims 1-7, and is characterized by comprising a data acquisition module, an internal resistance detection calibration module, a health criterion construction module, an analysis module and an alarm triggering module; the data acquisition module is used for acquiring the operation parameters of the single battery and the battery pack in real time through the cooperative work of the sampling plate and the sensor, and synchronously recording the time stamp and the equipment identifier of each parameter in the acquisition process; The internal resistance detection and calibration module is used for measuring the internal resistance of the battery by adopting a detection method, calculating the internal resistance of the battery according to the real-time change values of the discharge voltage and the current, and when the internal resistance of the battery is too small, switching the detection method and calibrating the internal resistance data by combining the discharge current; The health criterion construction module is used for establishing a backup power health criterion comprising a plurality of criterion types; The analysis module is used for constructing a battery health state evaluation model and a fault prediction model, dynamically evaluating the battery health state and predicting the occurrence period of the fault; the alarm triggering module is used for comparing the acquired data with the criterion threshold in real time, and when any criterion is detected to meet the abnormal condition, the system triggers an alarm mechanism and synchronously records the abnormal parameters, the occurrence time and the equipment position information.
9. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that the processor, when executing the computer program, implements the steps of the backup power health analysis method based on AI deep learning and battery big data processing of any of claims 1 to 7.
10. A computer-readable storage medium, on which a computer program is stored, characterized in that the computer program, when being executed by a processor, implements the steps of the backup power health analysis method based on AI deep learning and battery big data processing as claimed in any one of claims 1 to 7.

Description

Backup power supply health condition analysis method, system, equipment and medium based on AI deep learning and battery big data processing Technical Field The invention relates to the technical field of backup power supply health condition analysis, in particular to a backup power supply health condition analysis method, system, equipment and medium based on AI deep learning and battery big data processing. Background The power distribution network is a national economy basic industry, and the stable and reliable power supply of the power distribution network directly influences the economic development and the life of residents. The backup power supply of the distribution network terminal is used as key equipment for guaranteeing the continuous operation of the distribution network, and the health condition and the operation and maintenance quality of the backup power supply are important. The current distribution network operation and maintenance is digitalized and intelligently converted, and the traditional backup power supply operation and maintenance mode is difficult to adapt to the high-efficiency and accurate operation and maintenance requirements. The operation and maintenance of the existing backup power supply mainly comprises manual inspection and post-maintenance, and the capability of real-time monitoring and intelligent prejudgment is lacking. Monitoring data fragmentation, and a non-unified integration and analysis platform cannot realize battery full life cycle state tracing and trend prediction. The lack of an internal resistance online accurate detection technology is that internal resistance is a core index reflecting the health condition of the battery. Most of the existing on-line monitoring devices only monitor voltage, but ignore internal resistance changes. Few devices with an internal resistance test function usually adopt a single alternating current injection method or an instantaneous heavy current discharge method, are easily interfered by line noise, have poor measurement accuracy when the battery is in a floating state, and are difficult to accurately reflect the degradation inflection point inside the battery. Disclosure of Invention In view of the above problems, the present invention provides a backup power health analysis method, system, device and medium based on AI deep learning and battery big data processing. Therefore, the invention solves the technical problems that how to solve the operation and maintenance of the existing backup power supply mainly comprises manual inspection and post-maintenance, and lacks the capability of real-time monitoring and intelligent prejudging. Monitoring data fragmentation, and a non-unified integration and analysis platform cannot realize battery full life cycle state tracing and trend prediction. The technical scheme includes that the backup power supply health condition analysis method based on AI deep learning and battery big data processing includes the steps of collecting operation parameters of single batteries and battery packs in real time through cooperation of sampling plates and sensors, synchronously recording time stamps and equipment identifiers of the parameters in the collecting process, measuring the internal resistance of the batteries by adopting a detection method, calculating the internal resistance of the batteries according to real-time change values of discharge voltage and current, switching the detection method to calibrate internal resistance data when the internal resistance of the batteries is too small, combining the discharge current to calibrate the internal resistance data, establishing backup power supply health criteria comprising a plurality of criteria, constructing a battery health state assessment model and a fault prediction model, dynamically assessing the battery health state and predicting the fault occurrence period, comparing the collected data with a criterion threshold in real time, triggering an alarm mechanism when any criterion is detected to meet abnormal conditions, and synchronously recording abnormal parameters, occurrence time and equipment position information. The method comprises the steps of synchronously recording time stamps and equipment identifications of all parameters in the acquisition process through the cooperation of an acquisition plate and a sensor, performing analog-to-digital conversion on battery data through the acquisition plate to convert analog signals into digital signals, realizing indifferent monitoring of the single batteries through storage of the digital signals, acquiring the operating parameters of the single batteries and the battery pack in real time, automatically executing sensor null drift compensation and temperature and humidity self-adaptive gain adjustment after the equipment is powered on, and performing parameter correction by combining pre-calibrated temperature change parameters through comparison of detection of ambient temperature