Artificial intelligence assisted data error correction and recovery method and system

Abstract

The invention discloses an artificial intelligence-assisted data error correction and recovery method and system, and relates to the technical field of data error correction and recovery. Comprising the steps of obtaining data information needing to be transmitted, obtaining a target data item, and obtaining storage distribution information of the target data item; cutting the target data item for the first time based on the storage allocation information to obtain a data cutting item which is used for representing the cutting number of the target data item. According to the method, the data information is refined in the first step and then refined in the second step, the data information is split into the multiple parts again, the feature information of each part is obtained and compared with the initial information, and when the comparison result is different, the different segments are replaced, so that the error correction and recovery effects on the data information are achieved, and the error correction and recovery efficiency is improved. The data information is a plurality of fragments obtained through double refinement, so that the efficiency is higher when the fragments are subjected to error correction and recovery.

Description

Technical Field

[0001] The present invention relates to the technical field of data error correction and recovery, and specifically provides an artificial intelligence-assisted data error correction and recovery method and system. Background Art

[0002] Data error correction and recovery is a key technology in computer storage systems. It refers to the process of detecting and correcting data errors through specific algorithms and mechanisms, and restoring the original information when data is lost or damaged. Its core goal is to ensure the integrity and availability of data, which is particularly crucial in scenarios such as storage medium failures, human errors, or network attacks. The artificial intelligence-assisted data error correction and recovery method refers to using artificial intelligence technology to assist in data error correction and recovery, improving the error correction efficiency and the level of intelligence and automation in recovery, thereby enhancing the integrity and reliability of data.

[0003] The flash memory data error correction method and device with the patent publication number CN105740088A, when receiving a data read instruction, first performs error correction processing using the row check data of the data block to be read, and when the error correction of the data block to be read fails using the row check data, uses the column check data corresponding to the data block to be read for error correction processing, and then combines the row check data to perform error correction processing on the data to be read, thereby restoring the original data of the data to be read, increasing the number of error bits that the flash memory storage device can correct, and improving the flash memory data recovery rate.

[0004] When transmitting data using the above and similar technical solutions, due to the influence of different coefficients such as the number of erase and write cycles, temperature, and write amplification factor of different storage hard disks, their physical aging degrees are inconsistent. As the number of erase and write cycles increases, charge leakage and interference between units in the hard disk flash memory cells intensify, causing the error rate to increase exponentially. At this time, conventional error correction schemes such as fixed-strength LDPC codes will greatly increase the data error correction and recovery time in the later stage of aging, leading to a soaring data recovery failure rate. Especially when the data volume is large, it will further exacerbate the data error correction and recovery time, thereby increasing the failure rate of data recovery. Summary of the Invention

[0005] The purpose of the present invention is to provide an artificial intelligence-assisted data error correction and recovery method and system to solve the problems raised in the above background art.

[0006] To achieve the above purpose, the present invention provides the following technical solutions: An artificial intelligence-assisted data error correction and recovery method and system, including: Obtain the data information to be transmitted to obtain the target data item, and obtain the storage allocation information of the target data item; Perform the first cut on the target data item based on the storage allocation information to obtain a data cut item, where the data cut item is used to represent the number of cuts of the target data item; Create a hard disk aging degree quantification formula, and calculate the aging degree of the storage allocation information through artificial intelligence calculation to obtain a target aging set; Classify the aging degree to obtain an aging level item. At the same time, judge the aging level of the storage allocation information through a judgment method to obtain a target level item. Set cut-off values based on the aging level item respectively, and obtain the corresponding cut-off values based on the target level item to obtain a cut-off comparison set, where the cut-off comparison set corresponds to the storage allocation information respectively; Based on the data cut item, obtain the corresponding storage target information to obtain a storage target set, where the storage target set is used to represent the separate storage of the target data item cut into different quantities; Based on the cut-off comparison set, perform a second cut on the data cut item corresponding to the cut-off comparison set, and then obtain a cut-off plan for the data cut item at different storage positions to obtain a target cut set. Based on the target cut set, respectively perform comparison error correction and recovery reservation on the data cut item through a matching method, so as to realize the error correction and rapid recovery of the storage partition data at different aging degrees with the assistance of artificial intelligence.

[0007] Furthermore, the hard disk aging degree quantification formula includes: ; Where is the aging degree, is the Euler's constant, represents the erasure attenuation coefficient, represents the current cumulative erasure count, represents the maximum erasure count, represents the temperature sensitivity coefficient, represents the hard disk operating temperature, represents the reference temperature, represents the data write volume, represents the main control optimization factor, represents the cumulative power-on time, represents the time decay weight, represents the remaining available capacity, represents the total capacity.

[0008] Furthermore, the method for obtaining the aging level item includes: Set at least two aging degree range data to obtain an aging range item; Sort the aging range items in ascending order to obtain an aging sorting item, and set at least two evaluation levels corresponding to the aging sorting item respectively, and then obtain an aging level item.

[0009] Further, the method for setting the segmentation value includes: Based on the aging level items, the corresponding level values are set respectively. The corresponding level values are frequency values, and at least two corresponding level items are obtained. Based on the target level item, the corresponding corresponding level item is obtained to get the target corresponding item. The average value of the target corresponding item is obtained to get the corresponding average item. Using the corresponding average item as the initial score value, an increment value is set. Based on the combination results of the increment value and the initial score value respectively, score division items are obtained. The score division items are respectively corresponding to the segmentation value and the target level item.

[0010] Further, the method for obtaining the target segmentation set includes: Based on the segmentation comparison set, the data cutting items corresponding to the segmentation comparison set are obtained respectively, and the segmentation values matching the data cutting items are obtained. The data fragment of the data cutting item is obtained. Based on the matching segmentation value, the data fragment is equally divided and cut. The data cutting item is cut into data sub-fragments corresponding to the segmentation value. One data cutting item includes at least one data sub-fragment. The data sub-fragments included in the data cutting item are combined to obtain the target segmentation set.

[0011] Further, the matching method includes: Based on the target segmentation set, the target data item is divided into comparison segments, and the initial segment feature information of the target data item is obtained to get the initial information item. The segment feature information of the target segmentation set is obtained to get the segmentation information item. The segmentation information item and the initial information item corresponding to the segmentation information item are compared to determine whether the segmentation information item is abnormal. When the segmentation information item is abnormal, the abnormal segment information is obtained to get the abnormal segment item. Based on the abnormal segment item, the corresponding segment in the initial information item is obtained to get the replacement segment item, and then the abnormal segment item is replaced and restored.

[0012] Further, the method for obtaining the data cutting item includes: The quantity information of the storage allocation information is obtained to get the storage quantity item. The storage quantity item represents different types of storage modules for storing the target data item. Based on the quantity information of the storage quantity item, the target cutting quantity is obtained. Based on the target cutting quantity, the target data item is cut into the corresponding number of cuts to get the data cutting item.

[0013] Further, the method for obtaining the storage target set includes: Based on the importance sorting of the data cutting items, the data sorting item is obtained. The data sorting item is used to represent the sorting result of the importance of the data cutting items from high to low. Based on the comparison and matching result between the target level item and the data sorting item, obtain the comparison storage level of the data cutting item, and obtain the corresponding storage target set information based on the comparison storage level.

[0014] Furthermore, an artificial intelligence-assisted data error correction and recovery system uses the above-mentioned artificial intelligence-assisted data error correction and recovery method, including: Information acquisition module: Acquire the data information to be transmitted, obtain the target data item, acquire the storage allocation information of the target data item, and perform the first cut on the target data item based on the storage allocation information to obtain the data cutting item; Aging calculation module: Create a quantization formula for the hard disk aging degree, calculate the aging degree of the storage allocation information to obtain the target aging set, classify the aging degree to obtain the aging level item, and at the same time judge the aging level of the storage allocation information through a judgment method to obtain the target level item; Cutting and storage module: Set cut values respectively based on the aging level item, obtain the corresponding cut values based on the target level item to obtain the cut comparison set. The cut comparison set corresponds to the storage allocation information respectively. Based on the data cutting item, obtain the corresponding storage target information to obtain the storage target set. The storage target set is used to represent the separate storage of the target data items cut into different quantities. Based on the cut comparison set, perform a secondary cut on the data cutting item corresponding to the cut comparison set to obtain the splitting scheme of the data cutting item at different storage positions, and obtain the target cut set; Error correction and recovery module: Based on the target cut set, respectively perform comparison error correction and recovery reservation on the data cutting item through a matching method.

[0015] Compared with the prior art, the beneficial effects of the present invention are: The artificial intelligence-assisted data error correction and recovery method and system initially cut the data information to be transmitted and store it in different storage intervals respectively, which is the first step of refining the data information. Then, the refined data information is further refined in the second step, split into multiple parts again, and the characteristic information of each part is obtained and compared with the initial information. When the comparison result shows a difference, the different segment is replaced, thus realizing the error correction and recovery effect of the data information. Since the data information is obtained as multiple segments through two layers of refinement, the efficiency is higher when correcting and recovering the segments.

[0016] Meanwhile, the aging calculation is performed on the storage area where the data information is about to be stored through the created quantization formula for the hard disk aging degree. According to the aging calculation results, different numbers of cut values are matched for each storage area, and the data information stored in each storage area is subdivided to different degrees according to the cut values. Among them, the storage area with a higher aging degree needs to be more carefully subdivided to facilitate quick error correction and recovery response, thereby achieving the effect of error correction and recovery according to the actual aging state of each storage area. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 It is a schematic diagram of the overall process of the present invention; Figure 2 It is a schematic diagram of the relationship between the target data item and the storage quantity item of the present invention; Figure 3 It is a schematic diagram of the acquisition process of the score division item of the present invention; Figure 4 It is a schematic diagram of the acquisition relationship of the cut value of the present invention; Figure 5 It is a schematic diagram of the acquisition of the data sub - fragment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0018] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts belong to the scope of protection of the present invention.

[0019] During the data transmission process, the difference in the aging degree of storage hard disks is an important issue that cannot be ignored. Due to different hard disks being affected by factors such as different numbers of erase-write cycles, temperature, and write amplification factor, their physical aging degrees are often different. As the number of erase-write cycles increases, phenomena such as charge leakage and interference between cells in the hard disk flash memory units will intensify, resulting in an exponential increase in the error rate. Conventional error correction schemes need to spend a large amount of time on data error correction and recovery in the later stage of aging. This is because the sharp increase in the error rate makes the error correction process extremely complex and requires a large amount of calculation and iteration. On the other hand, long-term error correction and recovery operations will greatly increase the failure rate of data recovery. Especially in the case of a large amount of data, the time required for error correction and recovery will be further extended, thereby further increasing the failure rate of data recovery. The problems brought about by this aging difference are not only reflected in the error correction efficiency but also in the data reliability. Since the aging speeds of different hard disks are different, even if the same error correction scheme is adopted, there will be differences in the final data reliability. This difference will pose a great challenge to data management and maintenance, making it difficult to accurately evaluate and predict data reliability. And an artificial intelligence-assisted data error correction and recovery method provided by this application, through preliminary cutting of the data information to be transmitted, stores it in different storage intervals respectively, conducts the first refinement of the data information, and then conducts the second refinement of the refined data information, splits it into multiple parts again, and obtains the feature information of each part, compares it with the initial information, and when there is a difference in the comparison result, replaces the different segments, thereby achieving the effect of data information error correction and recovery. Since the data information is obtained as multiple segments through two-fold refinement, the efficiency is more efficient when correcting and recovering the segments among them, as Figure 1 shown, including steps S100 - S600.

[0020] Step S100: Obtain the data information to be transmitted to get the target data item, and obtain the storage allocation information of the target data item.

[0021] It should be noted that when transmitting data, relevant information of the transmitted data is obtained. For example, when packing and transporting data such as videos and pictures, the relevant information obtained at this time is information such as the numbers and names of the videos and pictures, and then the target data item is obtained. At the same time, the storage allocation information of the target data item is obtained. For example, when storing the target data item in a certain hard disk, the storage allocation information obtained at this time is the information of the storage hard disk.

[0022] Step S200: Based on the storage allocation information, conduct the first cut on the target data item to obtain the data cut item.

[0023] It should be noted that the data cutting item is used to represent the number of cuts of the target data item. The method for obtaining the data cutting item includes: obtaining the quantity information of the storage allocation information to obtain the storage quantity item, and the storage quantity item represents the different types of storage modules used to store the target data item; based on the quantity information of the storage quantity item, obtaining the target cutting amount, and cutting the target data item by a corresponding amount based on the target cutting amount to obtain the data cutting item.

[0024] In the specific implementation process, Figure 2 As shown, ten videos need to be stored now. At this time, the target data items are ten videos. There are two storage allocation information items, namely disk A and disk B. At this time, the storage quantity item is 2, which is used to represent different storage modules for storing ten videos. At this time, the target cutting amount is also 2. The target data item is cut based on the target cutting amount, that is, the ten videos are divided into two parts, which are stored in disk A and disk B respectively to obtain data cutting items.

[0025] Step S300: creating a quantitative formula for the degree of hard disk aging.

[0026] It should be noted that the aging degree of storage allocation information is calculated through artificial intelligence calculation methods to obtain the target aging set. The quantification formula for the hard disk aging degree includes: ; in For the degree of aging, is Euler's constant, Indicates the erase attenuation coefficient, with a value of 0.01, which is used to control the slope of the curve. Indicates the current cumulative number of erase and write times. Indicates the maximum number of erase and write times. The maximum number of erase and write times for different hard drives is different. The maximum number of erase and write times for SLC is about 100,000 times, the maximum number of erase and write times for MLC is about 3,000-5,000 times, and the maximum number of erase and write times for TLC is about 1,000-3,000 times. Indicates the temperature sensitivity coefficient, with a value of 1.5–2.0, indicating that high temperature accelerates aging. Indicates the hard disk operating temperature. Indicates the reference temperature. The set reference temperature is 25℃. Indicates the amount of data written. Indicates the master control optimization factor, which is 5–10 for advanced master control and 2–3 for low-end master control. Indicates the cumulative power-on time. represents the time decay weight, ranging from 0.05 to 0.1, reflecting non-write aging. Indicates the remaining available capacity. Indicates the total capacity.

[0027] It should be noted that since the number of erase / write cycles is a core parameter, and flash memory stores data essentially through the tunneling effect of the oxide layer, each erase / write operation will damage the oxide layer. Therefore, the formula for quantifying the aging degree of the hard disk includes the condition of the number of erase / write cycles. At the same time, high temperature will exacerbate electron migration and the generation of interface states, and high temperature accelerates the formation of interface defects in silicon oxides, hindering electron transport. Therefore, the formula for quantifying the aging degree of the hard disk also includes the condition of temperature. Moreover, write amplification represents efficiency loss. When data is updated, old blocks need to be moved, resulting in additional writes. And when the main controller distributes writes to different blocks, it increases the address mapping operation. Therefore, space pressure is also an important factor in hard disk aging. Fourth, even when not powered on, the electrons trapped in the floating gate will leak through quantum tunneling. The γ parameter is correlated with temperature through the TDDB equation. Therefore, the cumulative power-on time also needs to be considered as a factor in hard disk aging. Finally, when the available space is low, not only will the performance decrease, but the garbage collection frequency will increase exponentially. Therefore, the remaining space is also a factor in hard disk aging.

[0028] In a specific implementation process, it is now necessary to store a video in the AA hard disk. The cumulative number of erase / write cycles of the AA hard disk during use is obtained as 800 times, and the AA hard disk is TLC with a maximum number of erase / write cycles of 1500 times. At the same time, the operating temperature is obtained as 45 °C, the data write volume is 3, the cumulative power-on time of this hard disk is 2.5 years, and the remaining space is 50 GB, while the total space of this hard disk is 512 GB. At this time, based on the obtained data, is 800, is 1500, is 45, is 3, is 2.5, is 50, is 512, and at the same time the erase / write attenuation coefficient takes a value of 0.01, the temperature sensitivity coefficient takes a value of 1.8, the main controller optimization factor takes a value of 5, the time decay weight takes 0.07. At this time, according to the formula for quantifying the aging degree of the hard disk: ; The calculation is as follows: ; The final calculation result is 0.62, that is, the aging degree of this hard disk is 62%.

[0029] Step S400: Classify the aging degree to obtain the aging level items.

[0030] It should be noted that the method for obtaining the aging level items includes: setting three aging degree range data to obtain three aging range items; sorting the aging range items in ascending order to obtain aging sorting items, setting at least two evaluation levels corresponding to the aging sorting items respectively, and then obtaining the aging level items.

[0031] Specifically, set three aging degree range data, namely range 1, range 2, and range 3, with their data being 0-33%, 33%-66%, and 66%-100% respectively. At this time, sort range 1, range 2, and range 3 in ascending order, and the sorting result is range 1 > range 2 > range 3, thus obtaining the aging sorting items. Set three evaluation levels, namely low-level aging, medium-level aging, and high-level aging, corresponding to the aging sorting items respectively, and then obtain the aging level items.

[0032] Judge the aging level of the storage allocation information through the judgment method to obtain the target level items, and set the cut-off values respectively based on the aging level items.

[0033] It should be noted that as Figure 3 shown, the method for setting the cut-off values includes: based on the aging level items, set the level corresponding values respectively. The level corresponding values start from 1 and increase by 1 successively, and correspond to the aging level items. The level corresponding values are the number of times values, obtaining at least two level corresponding items; based on the target level items, obtain the corresponding level corresponding items to get the target corresponding items, obtain the average value of the target corresponding items to get the corresponding average item, use the corresponding average item as the initial score, set the increment value, and the increment value is also 1. Based on the combination results of the increment value and the initial score respectively, obtain the score division items, and the score division items are used as the cut-off values corresponding to the target level items respectively.

[0034] In the specific implementation process, as Figure 4As shown, it is now necessary to store 100 videos. At this time, the target data items are 100, which are stored in 4 hard disks respectively, namely disk a, disk b, disk c, and disk d. Through the set hard disk aging degree quantization formula, the aging degree of disk a is calculated to be 54%, the aging degree of disk b is 24%, the aging degree of disk c is 44%, and the aging degree of disk d is 74%. Since the set aging level items are low-level aging, medium-level aging, and high-level aging, and the corresponding data are 0 - 33%, 33% - 66%, and 66% - 100% respectively, so both disk a and disk c belong to medium-level aging, disk b belongs to low-level aging, and disk d belongs to high-level aging. Since the set level corresponding values start from 1 and increase by 1 one by one, and correspond to the aging level items, the level corresponding values for low-level aging, medium-level aging, and high-level aging are 1, 2, and 3 respectively. At this time, the target corresponding items for disk a, disk b, disk c, and disk d are 2, 1, 2, and 3 respectively. The obtained corresponding average value is 2. At this time, taking 2 as the initial score, according to the set increment value, the increment value is 1, and the three score division items obtained are 2, 3, and 4, which correspond to low-level aging, medium-level aging, and high-level aging respectively, that is, the cut values for disk a, disk b, disk c, and disk d are 3, 2, 3, and 4 respectively.

[0035] Based on the target level item, obtain the corresponding cut value to get the cut comparison set.

[0036] It should be noted that the cut comparison set corresponds to the storage allocation information respectively.

[0037] Step S500: Based on the data cutting item, obtain the corresponding storage target information to get the storage target set.

[0038] It should be noted that the storage target set is used to represent the separate storage of target data items cut into different quantities. The acquisition method of the storage target set includes: performing importance sorting based on the data cutting item to obtain the data sorting item, and the data sorting item is used to represent the sorting result of the importance of the data cutting item from high to low; based on the comparison and matching result of the target level item and the data sorting item, obtain the comparison storage level of the data cutting item, and based on the comparison storage level, obtain the corresponding storage target set information.

[0039] Step S600: Based on the cut comparison set, perform secondary cutting on the data cutting item corresponding to the cut comparison set, and then obtain the cut scheme of the data cutting item at different storage positions to get the target cut set.

[0040] It should be noted that the method for obtaining the target segmentation set includes: based on the segmentation comparison set, respectively obtaining the data segmentation items corresponding to the segmentation comparison set to obtain the segmentation values matching the data segmentation items; obtaining the data segments of the data segmentation items, and based on the matching segmentation values, equally dividing the data segments to cut the data segmentation items into data sub-segments corresponding to the number of segmentation values. A data segmentation item includes at least one data sub-segment, and the data sub-segments included in the data segmentation item are combined to obtain the target segmentation set.

[0041] In a specific implementation process, such as Figure 5 shown, when 100 videos need to be stored and are respectively stored in 4 hard disks, namely disk a, disk b, disk c, and disk d. It is obtained that both disk a and disk c are in medium aging, disk b is in low aging, and disk d is in high aging. The segmentation values of disk a, disk b, disk c, and disk d are 3, 2, 3, and 4 respectively. At this time, the videos allocated to disk a need to be cut into 3 parts, the videos allocated to disk b need to be cut into 2 parts, the videos allocated to disk c need to be cut into 3 parts, and the videos allocated to disk d need to be cut into 4 parts.

[0042] Based on the target segmentation set, the data segmentation items are respectively compared, corrected, and restored and reserved through the matching method.

[0043] It should be noted that with the assistance of artificial intelligence, the data error correction and rapid recovery of storage partitions in different aging degrees are realized through the matching method. The matching method includes: based on the target segmentation set, dividing the target data items into comparison segments and obtaining the initial segment feature information of the target data items to obtain the initial information items. Since according to the target segmentation set, the target data items are secondarily segmented, that is, the data after preliminary segmentation is secondarily segmented and further split into comparison segments; obtaining the segment feature information of the target segmentation set to obtain the segmentation information items, that is, the feature information of each segment, comparing the segmentation information items with the corresponding initial information items, and determining whether the segmentation information items are abnormal. When the segmentation information items are abnormal, obtaining the abnormal segment information to obtain the abnormal segment items, and based on the abnormal segment items, obtaining the corresponding segments in the initial information items to obtain the replacement segment items, and then replacing and restoring the abnormal segment items.

[0044] An artificial intelligence-assisted data error correction and recovery system that uses the above-mentioned artificial intelligence-assisted data error correction and recovery method, including: an information acquisition module: acquiring the data information to be transmitted to obtain target data items, acquiring the storage allocation information of the target data items, and performing a first cut on the target data items based on the storage allocation information to obtain data cut items; an aging calculation module: creating a quantization formula for the aging degree of the hard disk, calculating the aging degree of the storage allocation information to obtain a target aging set, classifying the aging degree to obtain an aging level item, and at the same time determining the aging level of the storage allocation information through a determination method to obtain a target level item; a split storage module: respectively setting split values based on the aging level item, obtaining the corresponding split values based on the target level item to obtain a split comparison set, the split comparison set corresponding to the storage allocation information respectively, obtaining the corresponding storage target information based on the data cut items to obtain a storage target set, the storage target set being used to represent separately storing the target data items cut into different quantities, and based on the split comparison set, performing a second cut on the data cut items corresponding to the split comparison set to further obtain a split scheme for the data cut items at different storage positions to obtain a target split set; an error correction and recovery module: based on the target split set, respectively performing comparison error correction and recovery reservation on the data cut items through a matching method.

[0045] Although the embodiments of the present invention have been shown and described, for those of ordinary skill in the art, it can be understood that various changes, modifications, substitutions, and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended embodiments and their equivalents.

Claims

1. An artificial intelligence-assisted data error correction and recovery method, comprising: Obtaining data information to be transmitted to obtain target data items, and obtaining storage allocation information of the target data items; Performing a first cut on the target data items based on the storage allocation information to obtain data cut items, where the data cut items are used to represent the number of cuts of the target data items; It is characterized in that: Creating a hard disk aging degree quantization formula, and calculating the aging degree of the storage allocation information through an artificial intelligence calculation method to obtain a target aging set; Dividing the aging degree into levels to obtain aging level items, and at the same time determining the aging level of the storage allocation information through a determination method to obtain target level items. Setting cut values respectively based on the aging level items, obtaining corresponding cut values based on the target level items to obtain a cut comparison set, and the cut comparison set corresponds to the storage allocation information respectively; Based on the data cut items, obtaining corresponding storage target information to obtain a storage target set, where the storage target set is used to represent the separate storage of the target data items cut into different quantities; Based on the cut comparison set, performing a second cut on the data cut items corresponding to the cut comparison set, and then obtaining a cut scheme for the data cut items at different storage positions to obtain a target cut set. Based on the target cut set, respectively performing comparison error correction and recovery reservation on the data cut items through a matching method, so as to realize the error correction and rapid recovery of storage partition data at different aging degrees under the assistance of artificial intelligence.

2. The artificial intelligence-assisted data error correction and recovery method according to claim 1, wherein: The hard disk aging degree quantization formula includes: ； Among them is the aging degree, is the Euler's constant, represents the erasure attenuation coefficient, represents the current cumulative number of erasures, represents the maximum number of erasures, represents the temperature sensitivity coefficient, represents the hard disk operating temperature, represents the reference temperature, represents the data write volume, represents the main control optimization factor, represents the cumulative power-on time, represents the time decay weight, represents the remaining available capacity, represents the total capacity.

3. An artificial intelligence-assisted data error correction and recovery method according to claim 1, characterized in that: The method for obtaining the aging level items includes: Setting at least two aging degree range data to obtain aging range items; Sorting the aging range items in ascending order to obtain aging sorting items, setting at least two evaluation levels corresponding to the aging sorting items respectively, and then obtaining aging level items.

4. An artificial intelligence-assisted data error correction and recovery method according to claim 1, characterized in that: The method for setting the cut values includes: Setting level corresponding values respectively based on the aging level items, where the level corresponding values are number of times values, to obtain at least two level corresponding items; Based on the target level items, obtaining the corresponding level corresponding items to obtain target corresponding items, obtaining the average value of the target corresponding items to obtain a corresponding average item, using the corresponding average item as the initial score, setting an increment value, and obtaining score division items based on the combined results of the increment value and the initial score respectively, and the score division items are used as cut values corresponding to the target level items respectively.

5. An artificial intelligence-assisted data error correction and recovery method according to claim 1, characterized in that: The method for obtaining the target cut set includes: Based on the cut comparison set, respectively obtaining the data cut items corresponding to the cut comparison set to obtain cut values matching the data cut items; Obtaining data segments of the data cut items, and equally dividing the data segments based on the matching cut values, cutting the data cut items into data sub-segments corresponding to the cut values, where one data cut item includes at least one data sub-segment, and the data sub-segments included in the data cut items are combined to obtain a target cut set.

6. An artificial intelligence-assisted data error correction and recovery method according to claim 1, characterized in that: The matching method includes: Based on the target cut set, dividing the target data items into comparison segments and obtaining the initial segment feature information of the target data items to obtain initial information items; Obtain the segment feature information of the target cut set to get the cut information item. Compare the cut information item with the initial information item corresponding to the cut information item to determine whether the cut information item is abnormal. When the cut information item is abnormal, obtain the abnormal segment information to get the abnormal segment item. Based on the abnormal segment item, obtain the corresponding segment in the initial information item to get the replacement segment item, and then replace and recover the abnormal segment item.

7. An artificial intelligence-assisted data error correction and recovery method according to claim 1, characterized in that: The method for obtaining the data cutting item includes: Obtain the quantity information of the storage allocation information to get the storage quantity item. The storage quantity item represents different types of storage modules used to store the target data item; Based on the quantity information of the storage quantity item, obtain the target cutting amount. Based on the target cutting amount, perform corresponding quantity of cutting on the target data item to get the data cutting item.

8. An artificial intelligence-assisted data error correction and recovery method according to claim 1, characterized in that: The method for obtaining the storage target set includes: Perform importance sorting based on the data cutting item to get the data sorting item. The data sorting item is used to represent the sorting result of the importance of the data cutting item from high to low; Based on the comparison and matching result of the target level item and the data sorting item, obtain the comparison storage level of the data cutting item. Based on the comparison storage level, obtain the corresponding storage target set information.

9. An artificial intelligence-assisted data error correction and recovery system, characterized in that: An artificial intelligence-assisted data error correction and recovery method according to any one of claims 1-8 is used, including: Information acquisition module: Obtain the data information to be transmitted to get the target data item. Obtain the storage allocation information of the target data item, and perform the first cut on the target data item based on the storage allocation information to get the data cutting item; Aging calculation module: Create a quantization formula for the hard disk aging degree, calculate the aging degree of the storage allocation information to get the target aging set, divide the aging degree into levels to get the aging level item, and at the same time determine the aging level of the storage allocation information through a determination method to get the target level item; Segmented storage module: Set segmentation values respectively based on the aging level item, obtain the corresponding segmentation values based on the target level item to get the segmentation comparison set. The segmentation comparison set corresponds to the storage allocation information respectively. Based on the data cutting item, obtain the corresponding storage target information to get the storage target set. The storage target set is used to represent the separate storage of the target data item cut into different quantities. Based on the segmentation comparison set, perform a second cut on the data cutting item corresponding to the segmentation comparison set, and then obtain the segmentation scheme of the data cutting item at different storage positions to get the target cut set; Error correction and recovery module: Based on the target cut set, perform comparison error correction and recovery reservation on the data cutting item respectively through a matching method.

Images