CN114781128A - Chip life prediction system and method using artificial intelligence technology - Google Patents
Chip life prediction system and method using artificial intelligence technology Download PDFInfo
- Publication number
- CN114781128A CN114781128A CN202210294933.0A CN202210294933A CN114781128A CN 114781128 A CN114781128 A CN 114781128A CN 202210294933 A CN202210294933 A CN 202210294933A CN 114781128 A CN114781128 A CN 114781128A
- Authority
- CN
- China
- Prior art keywords
- chip
- aging
- curve
- tested
- service life
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 73
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 25
- 238000005516 engineering process Methods 0.000 title claims abstract description 19
- 230000032683 aging Effects 0.000 claims abstract description 204
- 238000004088 simulation Methods 0.000 claims abstract description 63
- 238000012545 processing Methods 0.000 claims abstract description 35
- 230000011218 segmentation Effects 0.000 claims abstract description 18
- 230000008859 change Effects 0.000 claims description 31
- 238000004364 calculation method Methods 0.000 claims description 15
- 238000012795 verification Methods 0.000 claims description 15
- 230000008569 process Effects 0.000 claims description 14
- 238000004458 analytical method Methods 0.000 claims description 12
- 238000012216 screening Methods 0.000 claims description 5
- 238000005457 optimization Methods 0.000 claims description 3
- 230000007774 longterm Effects 0.000 description 4
- 230000009471 action Effects 0.000 description 3
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 238000002360 preparation method Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/23—Clustering techniques
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2119/00—Details relating to the type or aim of the analysis or the optimisation
- G06F2119/04—Ageing analysis or optimisation against ageing
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Data Mining & Analysis (AREA)
- Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Evolutionary Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Life Sciences & Earth Sciences (AREA)
- Computer Hardware Design (AREA)
- Geometry (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a system and a method for predicting the service life of a chip by utilizing an artificial intelligence technology, which comprises the following steps of S100: respectively acquiring simulated aging track curves of chips of different specifications and models when life data are obtained before delivery; respectively carrying out long-time segmentation stage processing on the simulation aging track curves of different chips; step S200: performing curve fitting on the basis of the aging characteristic values of the chips at different use time periods to obtain an aging characteristic value track curve of each chip; step S300: performing initial matching judgment to obtain a matching judgment result, wherein the matching judgment result comprises the need of carrying out life prediction time length adjustment on the chip to be detected, the need of carrying out life prediction time length adjustment on the chip to be detected and the need of carrying out re-matching on a simulation aging track curve on the chip to be detected; step S400: checking the matching judgment result; step S500: and matching and predicting the service life of the chip to be tested according to the verified matching judgment result.
Description
Technical Field
The invention relates to the technical field of artificial intelligence equipment service life prediction, in particular to a system and a method for predicting the service life of a chip by utilizing an artificial intelligence technology.
Background
In the prior art, the service life of a chip is predicted, a corresponding service life aging track curve is obtained according to the use condition of a historical chip, and the service life of the newly used data with the same specification and model is predicted by consistently referring to the service life aging track curve obtained in the past; in the actual use process, the use scenes of the chip are complex and various, because the actual working environments of different chips are different, and the influence on the service life of the chip caused by different external environments or different manual operation habits is different; the phenomenon of advanced aging of some chips relative to other chips with the same specification and model is caused by long-term severe environment or long-term artificial improper operation; some chips may have a phenomenon that the performance of the chips is fundamentally destroyed due to a long-term harsh environment or a long-term artificial improper operation, so that the results obtained by predicting the chips according to the original aging curve track are often inaccurate and have no reference value in these times.
Disclosure of Invention
The present invention is directed to a system and a method for predicting a lifetime of a chip using an artificial intelligence technique, so as to solve the problems set forth in the background art.
In order to solve the technical problems, the invention provides the following technical scheme: a chip life prediction method using artificial intelligence technology is characterized in that the prediction method comprises the following steps:
step S100: the method comprises the steps that a prediction system respectively obtains simulation aging track curves of chips of different specifications when the chips obtain service life data before leaving a factory; the simulation aging track curve is a track curve which is obtained by setting the attenuation change of chip performance parameter values of the chip under the standard supply condition along with the use duration; respectively carrying out service time division stage processing on simulation aging track curves of different chips to respectively obtain the segmentation periods of the service time division stage processing of the different chips and the aging characteristic values of the different chips corresponding to each segmentation period;
step S200: performing curve fitting on the basis of the aging characteristic values of the chips at different use time periods to obtain an aging characteristic value track curve of each chip;
step S300: the prediction system performs initial matching judgment based on the information of the chip to be tested in combination with the corresponding simulated aging track curve and the aging characteristic value track curve to obtain a matching judgment result, wherein the matching judgment result comprises the need of performing life prediction duration adjustment on the chip to be tested, the need of performing life prediction duration adjustment on the chip to be tested and the need of performing simulated aging track curve re-matching on the chip to be tested;
step S400: verifying a matching judgment result of the chip to be detected, which needs to be subjected to life prediction duration adjustment, and a matching judgment result of the chip to be detected, which needs to be subjected to simulation aging trajectory curve re-matching;
step S500: and matching and predicting the service life of the chip to be tested according to the verified matching judgment result.
Further, step S100 includes:
step S101: let the service life of the chip of the ith specification and model be Ti，TiThe unit of (A); based on TiRandomly setting a plurality of divided time periods tiA kind of tiThe setting mode of the numerical value corresponds to a using time length dividing method, tiHas the unit of A-1(ii) a Are respectively based on tiThe service life T of the chip of the ith specification and modeliN using time length stages are obtained through division; calculating A in each interval in k using time length stages-2The rate of change in performance p obtained per unit; wherein A is-1Represents the next level unit of A; a. the-2Represents the next two-level unit of A, A-1The next level of units of (a);
step S102: in the xth usage duration division method, the number of performance change rates in the kth usage duration stage is QKThe j individual performance change rate in the k usage period is
The average of all the performance change rates in the kth age period is
Step S103: calculating the deviation square sum in each use duration stage in the x-th use duration division method
Calculating the deviation square sum of all the service duration stages in the x-th service duration division method
Calculating the variance goodness of fit of the xth usage duration division method
Step S104: calculating the corresponding variance goodness of fit of different using time length division methods of chips with different specifications according to the steps S101-S103, and selecting the using time length division method with the variance goodness of fit closest to 1 as the optimal division method for each specification type of chip; taking the division time period corresponding to the optimal division method as a segmentation period for carrying out service duration segmentation stage processing on the chip of the specification type to obtain a plurality of service duration stages, and taking the average value of all performance change rates in each service duration stage as an aging characteristic value corresponding to each segmentation period;
the step of using time division for carrying out the simulated aging track curve on various chips is to obtain the average aging characteristics of various chips in different time periods; because the chip is a long-life component, but the brand-new chip and the chip used for a period of time are different from the chip used for a half period of time in aging speed, in the application, the method for using time division stage processing on different chips uses clustering thinking, so that the similarity of the performance change rates in the same using time period is the largest in the finally determined division method, and the dissimilarity of the performance change rates between different using time periods is the largest; according to the performance change rate of each service duration stage, stage locking during the service life prediction of the chip can be realized, so that the finally obtained chip service life prediction result is more accurate.
Further, step S200 includes:
step S201: recording the finally obtained segment period as h and the number of the use duration stages as EhThe aging characteristic value of the ith usage duration stage is Ri(ii) a The time range of the ith use duration stage is recorded as [ (i-1) h, ih](ii) a The time range of each service duration stage is used as an abscissa, and the aging characteristic value of each service duration stage is used as an ordinate to obtain EhA plurality of coordinate points, each coordinate point being of the form: [ [ (i-1) h, ih]，Ri]；
Step S202: respectively corresponding to each chip EhCarrying out curve fitting on the coordinate points to obtain a track curve of the aging characteristic value of each chip in different service time periods;
the obtained track curve of the aging characteristic values of each chip in different service time periods is a curve obtained after the simulation aging track curve of the chip is subjected to the chemical-phase processing, and the curve can better reflect different service life aging characteristics of different chips and can also reflect the performance between different chips, and the curve is similar to the situation that the initial aging characteristic values of a chip with better performance are probably close to each other in the initial service time period of the chip with the better performance relative to a chip with poorer performance, but the time required for the chip with the better performance to enter the next aging characteristic value from the initial aging characteristic value is longer than the time required for the chip with the poorer performance to enter the next aging characteristic value from the initial aging characteristic value; that is, the time period for the chip with good performance to maintain the initial aging characteristic value is longer than the time period for the chip with poor performance to maintain the initial aging characteristic value; or the time required for entering the next aging characteristic value from the initial aging characteristic value is close, but the difference between the initial aging characteristic values is large, namely the initial aging characteristic value of the chip with good performance is far smaller than that of the chip with poor performance; therefore, the track curve of the aging characteristic value of each specification type chip can be used as an intuitive characteristic curve for distinguishing different performance chips with different specifications and types.
Further, step S300 includes:
step S301: initially matching a simulation aging track curve L1 and an aging characteristic value track curve L2 according to the specification and the model of the chip to be tested; acquiring the actual use time Ta of the chip to be tested until the current life prediction time, the segment period Ra of the actual performance parameter values Ya and Ta until the current life prediction time in the curve L2, and the aging characteristic value Ua corresponding to the segment period Ra; obtaining a standard performance parameter value Yb in a curve L1 according to Ta;
step S302: obtaining the use time length Tb in the curve L1 according to Ya, and obtaining the belonging segment period Rb and the aging characteristic value Ub corresponding to the belonging segment period Rb of the Tb in the curve L2; obtaining a standard performance parameter value Yb in a curve L1 according to Ta;
step S302: if Ya is larger than Yb, judging that the service life prediction duration of the chip to be tested needs to be adjusted; if Ya is equal to Yb, judging that the service life prediction duration adjustment of the chip to be tested is not needed; if Ya<Yb and Ra ≠ Rb or Ra ≠ Rb, i.e.
The value of (c) is contained in the (Ua, Ub), and the service life prediction duration adjustment of the chip to be tested is judged; if Ya<Yb, and Ra ≠ Rb, i.e.
The value of (c) is not contained in the (Ua, Ub), and the re-matching of the simulation aging track curve of the chip to be tested is judged;
ya is greater than Yb, namely that the actual performance parameter value Ya is greater than the performance parameter value Yb obtained according to the standard simulation aging trajectory curve, namely that the current chip is in a healthy state, and the service life is longer than the service life obtained by estimating the service life according to the actual time length of putting into use probably because the use frequency is less and keeping the chip properly; the Ya-Yb means that the actual performance parameter value Ya is the same as the performance parameter value Yb obtained according to the standard simulation aging track curve, namely the current chip is in a healthy state, and the chip completely changes the performance parameter value according to the change rule of the standard simulation aging track curve; ya < Yb means that the actual performance parameter value Ya is smaller than the performance parameter value Yb obtained according to a standard simulation aging trajectory curve, namely the current chip is possibly in an unhealthy state, and the service life of the chip is shorter than the service life obtained by estimating the service life according to the actual use time because the use frequency is high and the storage is not proper, namely the chip with the same specification and model is used for a shorter time after being made clear, but the aging characteristic value of the chip is equal to the aging characteristic value which is obtained after being used for a longer time; if Ra is Rb or Ra is not equal to Rb, the chip is proved to have an advanced aging phenomenon, but the performance parameter value is changed according to the change rule of a standard simulation aging track curve, and only jump-stage aging occurs; ya is less than Yb, and Ra is not equal to Rb, namely that the chip is really aged in advance, but the performance parameter value of the chip is not changed according to the change rule of a standard simulation aging track curve, the performance change is probably caused due to high use frequency and improper storage, the aging rule of the chip is changed to be the same as the aging rule of a chip with another specification and model and poor performance, the service life prediction is inaccurate according to the standard simulation aging track curve matched according to the specification and model, and the simulation aging track curve of the chip to be tested needs to be matched again.
Further, step S400 includes:
step S401: capturing a plurality of attention operation items which can cause damage to the performance of a chip in the actual use process based on artificial intelligence, wherein the history of the attention operation items is the same as the specification and model of the chip to be detected;
step S402: if a matching judgment result that the life prediction duration needs to be adjusted occurs when the life of a certain chip to be tested is predicted, a historical operation record of the certain chip to be tested between the life predictions is retrieved, and if a plurality of cautionary operation items are implemented in the historical operation record, the verification result is passed;
step S403: if a matching judgment result that the simulated aging trajectory curve needs to be re-matched appears when the service life of a certain chip to be tested is predicted, a historical operation record of the certain chip to be tested between service life predictions is retrieved, and if a plurality of cautionary operation items are implemented in the historical operation record and the implementation frequency is greater than the implementation frequency of the plurality of cautionary operation items implemented in the step S402 or the duration of the cautionary operation items in any one implementation record is greater than the duration of the same cautionary operation items implemented in the step S402, the verification result is passed.
Further, the process of matching and predicting the service life of the chip to be tested in step S500 includes:
when Ya is larger than Yb, obtaining the use time Tb in the curve L1 according to Ya, taking the use time Tb as the actual calculation time for carrying out the service life prediction on the chip to be tested, namely adjusting the time of the use time Ta of the chip to be tested along the curve L1 to obtain a service life prediction result by carrying out the service life prediction from Tb;
the current chip is in a healthy state, and the chip is probably due to the reason that the use frequency is low and the chip is properly kept, the service life is longer than the service life obtained by estimating the service life according to the actual time length put into use, and the accurate service life calculation of the chip needs to be moved forward;
when Ya is Yb, predicting the service life of the chip to be tested from Ta along a curve L1 to obtain a service life prediction result;
the above means that the chip is in a healthy state at present, and the chip completely changes the performance parameter value according to the change rule of the standard simulation aging track curve;
when Ya<Yb, and Ra ═ Rb or Ra ≠ Rb,
when the service life of the chip to be tested is predicted, the service life Tb is used as the actual starting time of the service life prediction of the chip to be tested, namely the service life Ta of the chip to be tested is adjusted along the curve L1 to be the service life prediction result obtained by predicting the service life from Tb;
the chip is subjected to an advanced aging phenomenon, but the performance parameter value is changed according to the change rule of a standard simulation aging track curve, only jump-stage aging occurs, and the accurate service life needs to be moved backwards;
when Ya<Yb, and Ra ≠ Rb,
respectively substituting the actual service time Ta of the chip to be tested till the current life prediction time and the actual performance parameter value Ya of the chip to be tested till the current life prediction time into the simulated aging track curve L1 of the chip of the i-th specification modeliAnd the simulated aging trajectory curve L1iCorresponding aging characteristic value trajectory curve L2iThe preparation method comprises the following steps of (1) performing; according to Ta in the curve L1iTo obtain corresponding standard performance parameter value Ybi(ii) a Screening out Yb satisfyingi-obtaining a set of simulated aging trajectory curves S1 for the plurality of simulated aging trajectory curves having deviation values smaller than the deviation threshold obtained between Ya; obtaining the total performance parameter value Yq of the chip to be tested when leaving the factory, and calculating
Respectively corresponding ages in the simulation aging track curve set S1Curve of characteristic value trace L2iScreening out aging characteristic value and
obtaining a simulation aging track curve set S2 by the equal simulation aging track curves, and respectively obtaining
Judging the simulation aging track curve which is farthest away from the Ta within the range of the segmentation period as the simulation aging track curve matched with the chip to be tested for life prediction again;
the chip is subjected to an advanced aging phenomenon, but performance parameter values are not changed according to the change rule of a standard simulation aging track curve, and performance conversion occurs due to the fact that the chip is high in use frequency and cannot be stored properly; the simulated aging track curve matched with the life prediction is found again, firstly, the simulated aging track curve with higher curve matching degree is found according to Ya and Ta which are actually measured and obtained, then, the aging characteristic values obtained according to the aging characteristic value track curves corresponding to the simulated aging track curves are further matched, namely, the chip with the highest matching degree of the simulated aging track curve and the aging characteristic value track curve is found for the chip to be tested, the performance of the chip with the other specification type accords with the actual performance of the chip to be tested after the performance of the chip is changed, and the life prediction of the chip to be tested is carried out according to the standard simulated aging track curve of the chip with the specification type, so that the result is more accurate.
In order to better realize the method, a chip life prediction system utilizing an artificial intelligence technology is also provided, and the prediction system comprises: the system comprises an aging data acquisition module, an aging period stage processing module, a characteristic value processing module, an initial matching judgment module, a verification module and a matching prediction module;
the aging data acquisition module is used for acquiring a simulation aging track curve when chips of different specifications and models obtain service life data before leaving a factory;
the aging period division stage processing module is used for receiving the data in the aging data acquisition module and carrying out long-time division stage processing on the simulation aging track curves of different chips;
the characteristic value processing module is used for receiving data in the aging period stage dividing processing module, acquiring aging characteristic values of the chips in different service time periods and carrying out curve fitting to obtain an aging characteristic value track curve of each chip;
the initial matching judgment module is used for performing initial matching judgment by combining the information of the chip to be tested with the corresponding simulation aging track curve and the aging characteristic value track curve to obtain a matching judgment result;
the verification module is used for receiving the data in the initial matching judgment module and verifying the data;
and the matching prediction module is used for receiving the data in the verification module and completing the service life matching and prediction of the chip to be tested.
Further, the aging period stage processing module comprises: the method comprises the steps of dividing a time period setting unit, a calculating unit and a optimizing unit;
the device comprises a division time period setting unit, a comparison unit and a comparison unit, wherein the division time period setting unit is used for randomly setting a plurality of division time periods respectively based on the service lives of chips with different specifications and models;
the calculation unit is used for calculating the performance change rate, the deviation sum of squares and the variance goodness of fit involved in the aging period division stage processing;
and the optimization unit is used for receiving the data in the calculation unit and respectively finishing the selection of the optimal division time periods of the chips with different specifications and models.
Further, the initial matching judgment module comprises: the device comprises a data capturing unit, a judging unit and a matching analysis unit;
the data capturing unit is used for capturing all judgment data required in the process of predicting the service life of the chip to be tested;
the judging unit is used for receiving the data in the data capturing unit and carrying out matching judgment based on the data;
and the matching analysis unit is used for receiving the data in the judgment unit and obtaining a matching analysis result of the to-be-detected chip needing to be subjected to service life prediction duration adjustment and a matching analysis result of the to-be-detected chip needing to be subjected to re-matching of the simulation aging trajectory curve based on the data.
Compared with the prior art, the invention has the following beneficial effects: in the process of predicting the service life of the chip, the influence of the actual working environment or the manual operation habit of the chip on the service life of the chip, which is caused by the complicated and various use scenes of the chip or the influence of the fundamental damage of the performance, on the service life of the chip is considered in the actual use process of the chip, the accurate service life starting time is positioned or the proper service life prediction related curve track is matched again in the process of predicting the service life of the chip to be tested, so that the accuracy and the reliability are improved for the finally obtained service life prediction result of the chip.
Drawings
The accompanying drawings, which are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate embodiments of the invention and together with the description serve to explain the principles of the invention and not to limit the invention. In the drawings:
FIG. 1 is a flow chart of a method for predicting the life of a chip by using artificial intelligence technology according to the present invention;
FIG. 2 is a schematic diagram of a chip life prediction system using artificial intelligence technology according to the present invention;
FIG. 3 is a diagram of a chip life prediction method using artificial intelligence technology according to an embodiment of the present invention;
FIG. 4 is a second embodiment of the method for predicting the lifetime of a chip using artificial intelligence technology according to the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be obtained by a person skilled in the art without making any creative effort based on the embodiments in the present invention, belong to the protection scope of the present invention.
Referring to fig. 1-4, the present invention provides a technical solution: a chip life prediction method by using an artificial intelligence technology is characterized by comprising the following steps:
step S100: the method comprises the steps that a prediction system respectively obtains simulation aging track curves of chips of different specifications when the chips obtain service life data before leaving a factory; the simulation aging track curve is a track curve in which the performance parameter value of the chip obtained by the chip under the standard supply condition is set to be changed along with the attenuation of the use time; respectively carrying out service time division stage processing on the simulation aging track curves of different chips to respectively obtain the segmentation periods of the service time division stage processing of the different chips and the aging characteristic values of the different chips corresponding to each segmentation period;
wherein, step S100 includes:
step S101: let the service life of the chip of the ith specification and model be Ti，TiThe unit of (A); based on TiRandomly setting a plurality of division time periods tiA kind of tiThe setting mode of the numerical value corresponds to a method for dividing the use duration, tiHas the unit of A-1(ii) a Are respectively based on tiThe service life T of the chip of the ith specification and modeliN stages of using time length are obtained through division; calculating A in each interval in k using time length stages-2The rate of change in performance p obtained per unit; wherein, A-1Represents the next level unit of A; a. the-2Represents the next two-level unit of A, A-1The next level of units of (a);
for example, T in this embodimentiThe unit of (a) is year, then tiUnit A of-1Is a month, A-2Is day; can also be caused by TiThe unit of (1) is month, then tiUnit A of-1Is day, A-2Is hour;
step S102: in the xth usage duration division method, the number of performance change rates in the kth usage duration stage is QKThe j individual performance change rate in the k usage period is
The average of all the performance change rates in the kth service life period is
Step S103: calculating the deviation square sum in each use duration stage in the x-th use duration division method
Calculating the deviation square sum of all the service time length stages in the x-th service time length division method
Calculating the goodness of fit of the variance of the xth usage duration division method
Step S104: calculating the corresponding variance goodness of fit of different using time length division methods of chips with different specifications according to the steps S101-S103, and selecting the using time length division method with the variance goodness of fit closest to 1 as the optimal division method for each specification type of chip; taking the division time period corresponding to the optimal division method as a segmentation period for carrying out service duration segmentation stage processing on the chip of the specification type to obtain a plurality of service duration stages, and taking the average value of all performance change rates in each service duration stage as an aging characteristic value corresponding to each segmentation period;
step S200: performing curve fitting on the aging characteristic values of the chips at different using time periods to obtain an aging characteristic value track curve of each chip;
wherein, step S200 includes:
step S201: recording the finally obtained segment period as h and the number of the use duration stages as EhThe aging characteristic value of the ith service life stage is Ri(ii) a The time range of the ith using duration stage is recorded as [ (i-1) h, ih](ii) a At each time of use duration stageThe interval range is an abscissa, and the aging characteristic value of each service time period stage is an ordinate to obtain EhA plurality of coordinate points, each coordinate point being of the form: [ [ (i-1) h, ih]，Ri]；
Step S202: respectively corresponding to each chip EhCarrying out curve fitting on the coordinate points to obtain a track curve of the aging characteristic value of each chip in different service time periods;
for example, for a chip of one model specification, the total life duration of the chip is 15 years as shown by a simulation aging track curve when the chip obtains life data before leaving the factory, and the optimal division method is that the chip is a use duration stage every 3 years, so that the number of the use duration stages is 5, namely (0,3), (3,6), (6,9), (9,12) and (12, 15); in a first duration phase (0,3) its aging characteristic value is 0.1%, in a second duration phase (3,6) its aging characteristic value is 0.14%, in a third duration phase (6,9) its aging characteristic value is 0.19%, in a fourth duration phase (9,12) its aging characteristic value is 0.21%, in a fifth duration phase (12,15) its aging characteristic value is 0.25%; its trace plot 3 of the aging characteristic values is shown;
step S300: the prediction system performs initial matching judgment based on the information of the chip to be tested in combination with the corresponding simulated aging track curve and the aging characteristic value track curve to obtain a matching judgment result, wherein the matching judgment result comprises the need of performing life prediction duration adjustment on the chip to be tested, the need of performing life prediction duration adjustment on the chip to be tested and the need of performing simulated aging track curve re-matching on the chip to be tested;
wherein, step S300 includes:
step S301: setting a simulation aging track curve L1 and an aging characteristic value track curve L2 which are initially matched according to the specification and the model of a chip to be tested; acquiring the actual use time Ta of the chip to be tested until the current life prediction time, the segment period Ra of the actual performance parameter values Ya and Ta until the current life prediction time in the curve L2, and the aging characteristic value Ua corresponding to the segment period Ra; obtaining a standard performance parameter value Yb in a curve L1 according to Ta;
step S302: obtaining the use time length Tb in the curve L1 according to Ya, and obtaining the belonged segment period Rb and the aging characteristic value Ub corresponding to the belonged segment period Rb of the Tb in the curve L2; obtaining a standard performance parameter value Yb in a curve L1 according to Ta;
step S302: if Ya is larger than Yb, judging that the service life prediction duration of the chip to be tested needs to be adjusted; if Ya is Yb, the service life prediction time length of the chip to be tested does not need to be adjusted; if Ya<Yb and Ra ≠ Rb or Ra ≠ Rb, i.e.
The value of (c) is contained in the (Ua, Ub), and the service life prediction duration adjustment of the chip to be tested is judged; if Ya<Yb, and Ra ≠ Rb, i.e.
The value of (c) is not contained in the (Ua, Ub), and the re-matching of the simulation aging track curve of the chip to be tested is judged;
step S400: verifying a matching judgment result of the chip to be detected, which needs to be subjected to life prediction duration adjustment, and a matching judgment result of the chip to be detected, which needs to be subjected to simulation aging trajectory curve re-matching;
wherein, step S400 includes:
step S401: capturing a plurality of attention operation items which can cause damage to the performance of a chip in the actual use process based on artificial intelligence, wherein the history of the attention operation items is the same as the specification and model of the chip to be detected;
step S402: if a matching judgment result that the life prediction duration needs to be adjusted occurs when the life of a certain chip to be detected is predicted, calling a historical operation record of the certain chip to be detected during the life prediction, and if a plurality of notice operation items are implemented in the historical operation record, the verification result is passed;
step S403: if a matching judgment result that the simulation aging trajectory curve needs to be re-matched appears when the service life of a certain chip to be tested is predicted, calling a historical operation record of the certain chip to be tested between service life predictions, and if a plurality of cautionary operation items are implemented in the historical operation record and the implementation times are greater than the implementation times of the plurality of cautionary operation items implemented in the step S402 or the duration of the cautionary operation items in any one implementation record is greater than the duration of the same cautionary operation items implemented in the step S402, the verification result is passed;
step S500: matching and predicting the service life of the chip to be tested according to the verified matching judgment result;
the process of matching and predicting the service life of the chip to be tested in the step S500 includes:
when Ya is larger than Yb, obtaining the use time Tb in the curve L1 according to Ya, taking the use time Tb as the actual starting time for carrying out the life prediction on the chip to be detected, namely adjusting the time of the use time Ta of the chip to be detected along the curve L1 to obtain a life prediction result by carrying out the life prediction on the chip from Tb;
for example, Yb is 76, Ta is 3 years, if Ya of the chip to be tested is 78, then according to fig. 4, t1 corresponds to the performance parameter value 78, and t1 is zero 3 months in two years; adjusting the service life of the chip to be tested along a curve L1 for 3 years to obtain a life prediction result by predicting the service life of the chip from two years and 3 months, wherein if the total life duration of the chip to be tested is 15 years as shown by a simulation aging track curve when the chip to be tested obtains life data before leaving a factory, the residual life of the chip to be tested is 12 years and 9 months;
when Ya is Yb, predicting the service life of the chip to be tested from Ta along a curve L1 to obtain a service life prediction result;
when Ya<Yb, and Ra ═ Rb or Ra ≠ Rb,
when the service life of the chip to be tested is predicted, the service life Tb is used as the actual starting time of the service life prediction of the chip to be tested, namely the service life Ta of the chip to be tested is adjusted along the curve L1 to be the service life prediction result obtained by predicting the service life from Tb;
for example, if Yb is 76 and Ta is 3 years, and Ya of the chip under test is 73, then according to fig. 4, t2 corresponds to the performance parameter value 73,t2 is three years for zero 3 months; ra ≠ Rb or Ra ≠ Rb,
the method comprises the steps that the service life of a chip to be tested is adjusted to be 3 years along a curve L1, the service life is predicted from three years to zero 3 months to obtain a service life prediction result, if the total service life of the chip to be tested is 15 years as shown by a simulation aging track curve when the chip to be tested obtains service life data before leaving a factory, the residual service life of the chip to be tested is 11 years to zero 9 months;
when Ya<Yb, and Ra ≠ Rb,
then, respectively substituting the actual service duration Ta of the chip to be tested up to the current life prediction time and the actual performance parameter value Ya of the chip to be tested up to the current life prediction time into the simulation aging track curve L1 of the chip with the model of the i & ltth & gt specification modeliAnd the simulated aging trajectory curve L1iCorresponding aging characteristic value trajectory curve L2iPerforming the following steps; according to Ta in the curve L1iTo obtain corresponding standard performance parameter value Ybi(ii) a Screening out Yb satisfyingi-obtaining a set of simulated aging trajectory curves S1 for the plurality of simulated aging trajectory curves having deviation values smaller than the deviation threshold obtained between Ya; obtaining the total performance parameter value Yq of the chip to be tested when leaving the factory, and calculating
Aging characteristic value track curves L2 respectively corresponding to the simulation aging track curve set S1iThe aging characteristic value and is screened out
Obtaining a simulation aging trajectory curve set S2 by the equal simulation aging trajectory curves, and respectively obtaining the simulation aging trajectory curves
Judging the simulation aging track curve which is farthest away between the range of the segment period and Ta as the segment period to which the equal aging characteristic value belongs toAnd carrying out life prediction matched simulation aging track curve.
In order to better realize the method, a chip life prediction system utilizing an artificial intelligence technology is also provided, and the prediction system comprises: the system comprises an aging data acquisition module, an aging period stage processing module, a characteristic value processing module, an initial matching judgment module, a verification module and a matching prediction module;
the aging data acquisition module is used for acquiring a simulation aging track curve when chips of different specifications and models obtain service life data before leaving a factory;
the aging period stage processing module is used for receiving the data in the aging data acquisition module and carrying out use time stage processing on the simulation aging track curves of different chips;
wherein, the aging cycle stage processing module comprises: the method comprises the steps of dividing a time period setting unit, a calculating unit and a optimizing unit;
the device comprises a division time period setting unit, a comparison unit and a comparison unit, wherein the division time period setting unit is used for randomly setting a plurality of division time periods respectively based on the service lives of chips with different specifications and models;
the calculation unit is used for completing the calculation of performance change rate, the calculation of deviation square sum and the calculation of variance goodness of fit, wherein the performance change rate, the deviation square sum and the variance goodness of fit are involved in the aging period division stage processing;
the characteristic value processing module is used for receiving data in the aging period stage dividing processing module, acquiring aging characteristic values of the chips in different service time periods and carrying out curve fitting to obtain an aging characteristic value track curve of each chip;
the initial matching judgment module is used for performing initial matching judgment by combining the information of the chip to be tested with the corresponding simulation aging track curve and the aging characteristic value track curve to obtain a matching judgment result;
the verification module is used for receiving the data in the initial matching judgment module and verifying the data;
the matching prediction module is used for receiving the data in the verification module and completing the service life matching and prediction of the chip to be tested;
the optimization unit is used for receiving the data in the calculation unit and respectively finishing the selection of the optimal division time periods of the chips with different specifications and models;
wherein, the initial matching judgment module comprises: the device comprises a data capturing unit, a judging unit and a matching analysis unit;
the data capturing unit is used for capturing all judgment data required in the process of predicting the service life of the chip to be tested;
the judging unit is used for receiving the data in the data capturing unit and performing matching judgment based on the data;
and the matching analysis unit is used for receiving the data in the judgment unit and obtaining a matching analysis result of the to-be-detected chip needing to be subjected to service life prediction duration adjustment and a matching analysis result of the to-be-detected chip needing to be subjected to re-matching of the simulation aging trajectory curve based on the data.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: although the present invention has been described in detail with reference to the foregoing embodiments, it will be apparent to those skilled in the art that modifications may be made to the embodiments described above, or equivalents may be substituted for elements thereof. Any modification, equivalent replacement, or improvement made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (9)
1. A chip life prediction method using artificial intelligence technology is characterized by comprising the following steps:
step S100: the method comprises the steps that a prediction system respectively obtains simulation aging track curves of chips of different specifications when the chips obtain service life data before leaving a factory; the simulation aging track curve is a track curve which is obtained by setting the attenuation change of chip performance parameter values of a chip under a standard supply condition along with the use duration; respectively carrying out service time division stage processing on the simulation aging track curves of different chips to respectively obtain the segmentation periods of the service time division stage processing of the different chips and the aging characteristic values of the different chips corresponding to the segmentation periods;
step S200: performing curve fitting on the aging characteristic values of the chips at different using time periods to obtain an aging characteristic value track curve of each chip;
step S300: the prediction system performs initial matching judgment based on information of the chip to be tested and combined with the corresponding simulated aging track curve and the aging characteristic value track curve to obtain a matching judgment result, wherein the matching judgment result comprises that the life prediction time length of the chip to be tested needs to be adjusted, the life prediction time length of the chip to be tested does not need to be adjusted, and the simulated aging track curve of the chip to be tested needs to be re-matched;
step S400: verifying a matching judgment result of the chip to be detected, which needs to be subjected to life prediction duration adjustment, and a matching judgment result of the chip to be detected, which needs to be subjected to simulation aging trajectory curve re-matching;
step S500: and matching and predicting the service life of the chip to be tested according to the verified matching judgment result.
2. The method for predicting the lifetime of a chip using artificial intelligence technology according to claim 1, wherein said step S100 comprises:
step S101: if the service life of the chip of the ith specification model is Ti，TiThe unit of (a); based on TiRandomly setting a plurality of divided time periods tiA kind of tiThe setting mode of the numerical value corresponds to a using time length dividing methodMethod, tiHas a unit of-1(ii) a Are based on t respectivelyiThe service life T of the chip of the ith specification and modeliN using time length stages are obtained through division; calculating A in each interval in k using time length stages-2The rate of change in performance p obtained per unit; wherein A is-1Represents the next level unit of A; a. the-2Represents the next two-level unit of A, A-1The next level of units of (a);
step S102: in the xth usage duration division method, the number of performance change rates in the kth usage duration stage is QKThe j individual performance change rate in the k usage period is
The average of all the performance change rates in the kth service life period is
Step S103: calculating the deviation square sum in each use duration stage in the x-th use duration division method
Calculating the deviation square sum of all the service duration stages in the x-th service duration division method
Calculating the variance goodness of fit of the xth usage duration division method
Step S104: calculating the variance goodness of fit corresponding to different using time length division methods of chips with different specifications and models according to the steps S101 to S103, and selecting the using time length division method with the variance goodness of fit closest to 1 as an optimal division method for each specification and model; and taking the division time period corresponding to the optimal division method as a segmentation period for carrying out service duration segmentation stage processing on the chip of the specification type to obtain a plurality of service duration stages, and taking the average value of all performance change rates in each service duration stage as an aging characteristic value corresponding to each segmentation period.
3. The method for predicting lifetime of a chip using artificial intelligence technique according to claim 2, wherein said step S200 comprises:
step S201: recording the finally obtained segment period as h and the number of the use duration stages as EhThe aging characteristic value of the ith usage duration stage is Ri(ii) a The time range of the ith using duration stage is recorded as [ (i-1) h, ih](ii) a The time range of each service duration stage is used as an abscissa, and the aging characteristic value of each service duration stage is used as an ordinate to obtain EhA plurality of coordinate points, each coordinate point being of the form: [ [ (i-1) h, ih]，Ri]；
Step S202: respectively corresponding to each chip EhAnd performing curve fitting on the coordinate points to obtain a track curve of the aging characteristic value of each chip in different service time periods.
4. The method for predicting lifetime of a chip using artificial intelligence technology according to claim 1, wherein said step S300 comprises:
step S301: setting a simulation aging track curve L1 and an aging characteristic value track curve L2 which are initially matched according to the specification and the model of the chip to be tested; acquiring the actual service time Ta of the chip to be tested until the current life prediction time, the segment period Ra of the actual performance parameter values Ya and Ta in the curve L2 until the current life prediction time, and the aging characteristic value Ua corresponding to the segment period Ra; obtaining a standard performance parameter value Yb in a curve L1 according to Ta;
step S302: obtaining the use time length Tb in the curve L1 according to Ya, and obtaining the belonging segment period Rb and the aging characteristic value Ub corresponding to the belonging segment period Rb of the Tb in the curve L2; obtaining a standard performance parameter value Yb in a curve L1 according to Ta;
step (ii) ofS302: if Ya is larger than Yb, judging that the service life prediction duration of the chip to be tested needs to be adjusted; if Ya is equal to Yb, judging that the service life prediction duration adjustment of the chip to be tested is not needed; if Ya < Yb, and Ra ≠ Rb or Ra ≠ Rb, i.e.
The value of (c) is contained in the (Ua, Ub), and the service life prediction duration adjustment of the chip to be tested is judged; if Ya < Yb, and Ra ≠ Rb, i.e.
The value of (c) is not included in (Ua, Ub), and it is determined that the re-matching of the simulated aging trace curve needs to be performed on the chip to be tested.
5. The method for predicting the lifetime of a chip using artificial intelligence technology according to claim 1, wherein said step S400 comprises:
step S401: capturing a plurality of attention operation items which can cause damage to the performance of a chip in the actual use process based on artificial intelligence, wherein the history of the attention operation items is the same as the specification and model of the chip to be detected;
step S402: if a matching judgment result that the life prediction duration needs to be adjusted occurs when the life of a certain chip to be tested is predicted, a historical operation record of the certain chip to be tested between the life predictions is retrieved, and if a plurality of cautionary operation items are implemented in the historical operation record, a verification result is passed;
step S403: if a matching judgment result that the simulated aging trajectory curve needs to be re-matched appears when the service life of a certain chip to be tested is predicted, a historical operation record of the certain chip to be tested between service life predictions is retrieved, and if a plurality of cautionary operation items are implemented in the historical operation record and the implementation frequency is greater than the implementation frequency of the plurality of cautionary operation items implemented in the step S402 or the duration of the cautionary operation items in any implementation record is greater than the duration of the same cautionary operation items implemented in the step S402, the verification result is passed.
6. The method for predicting the service life of a chip by using the artificial intelligence technology as claimed in claim 4, wherein the step S500 of matching and predicting the service life of the chip to be tested comprises the steps of:
when Ya is larger than Yb, obtaining the use time Tb in a curve L1 according to Ya, taking the use time Tb as the actual calculation time for carrying out the service life prediction on the chip to be detected, namely adjusting the time of the use time Ta of the chip to be detected along the curve L1 to obtain a service life prediction result from the time Tb by carrying out the service life prediction;
when Ya is Yb, carrying out life prediction on the chip to be tested from Ta along a curve L1 to obtain a life prediction result;
when Ya is less than Yb, and Ra is Rb or Ra is not equal to Rb,
when the service life Tb is used as the actual starting time for predicting the service life of the chip to be tested, the service life Ta of the chip to be tested along the curve L1 is adjusted to be the service life prediction result obtained by predicting the service life from Tb;
when Ya is less than Yb and Ra is not equal to Rb,
respectively substituting the actual service time Ta of the chip to be tested till the current life prediction time and the actual performance parameter value Ya of the chip to be tested till the current life prediction time into the simulated aging track curve L1 of the chip of the i-th specification modeliAnd the simulated aging trajectory curve L1iCorresponding aging characteristic value track curve L2iPerforming the following steps; according to Ta at curve L1iObtaining the corresponding standard performance parameter value Ybi(ii) a Screening out the Ybi-obtaining a set of simulated aging trajectory curves S1 for the plurality of simulated aging trajectory curves having deviation values smaller than the deviation threshold obtained between Ya; obtaining the total performance parameter value Yq of the chip to be tested when leaving the factory, and calculating
Aging characteristic value track curves L2 corresponding to the simulation aging track curve set S1iScreening out aging characteristic value and
obtaining a simulation aging track curve set S2 by the equal simulation aging track curves, and respectively obtaining
And judging the simulation aging track curve which is farthest away from the Ta within the range of the segmentation period as the simulation aging track curve which is matched with the chip to be tested for life prediction again.
7. An artificial intelligence technology-based chip life prediction system applied to the artificial intelligence technology-based chip life prediction method according to any one of claims 1 to 6, the prediction system comprising: the system comprises an aging data acquisition module, an aging period stage processing module, a characteristic value processing module, an initial matching judgment module, a verification module and a matching prediction module;
the aging data acquisition module is used for acquiring a simulation aging track curve when chips of different specifications and models obtain service life data before leaving a factory;
the aging period stage processing module is used for receiving the data in the aging data acquisition module and carrying out use time stage processing on simulation aging track curves of different chips;
the characteristic value processing module is used for receiving the data in the aging period stage dividing processing module, acquiring the aging characteristic values of the chips in different service time periods and carrying out curve fitting to obtain an aging characteristic value track curve of each chip;
the initial matching judgment module is used for performing initial matching judgment by combining the information of the chip to be tested with the corresponding simulation aging track curve and the aging characteristic value track curve to obtain a matching judgment result;
the verification module is used for receiving the data in the initial matching judgment module and verifying the data;
and the matching prediction module is used for receiving the data in the verification module and completing the service life matching and prediction of the chip to be tested.
8. The system of claim 7, wherein the aging period staging module comprises: the device comprises a time period setting unit, a calculating unit and a preferred unit;
the dividing time period setting unit is used for randomly setting a plurality of dividing time periods respectively based on the service lives of chips with different specifications and models;
the calculation unit is used for completing the calculation of the performance change rate, the calculation of the sum of squares of deviation and the calculation of the goodness of fit of the variance, which are involved in the aging period division stage processing;
and the optimization unit is used for receiving the data in the calculation unit and respectively finishing the selection of the optimal division time periods of chips with different specifications.
9. The system of claim 7, wherein the initial matching judgment module comprises: the device comprises a data capturing unit, a judging unit and a matching analysis unit;
the data capturing unit is used for capturing all judgment data required in the process of predicting the service life of the chip to be tested;
the judging unit is used for receiving the data in the data capturing unit and performing matching judgment based on the data;
and the matching analysis unit is used for receiving the data in the judgment unit and obtaining a matching analysis result of the to-be-detected chip needing to be subjected to life prediction duration adjustment and a matching analysis result of the to-be-detected chip needing to be subjected to re-matching of the simulated aging trajectory curve based on the data.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210294933.0A CN114781128A (en) | 2022-03-24 | 2022-03-24 | Chip life prediction system and method using artificial intelligence technology |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202210294933.0A CN114781128A (en) | 2022-03-24 | 2022-03-24 | Chip life prediction system and method using artificial intelligence technology |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114781128A true CN114781128A (en) | 2022-07-22 |
Family
ID=82425816
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202210294933.0A Pending CN114781128A (en) | 2022-03-24 | 2022-03-24 | Chip life prediction system and method using artificial intelligence technology |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114781128A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN117765374A (en) * | 2023-11-15 | 2024-03-26 | 知码芯(无锡)通讯技术有限公司 | Data analysis system and method based on artificial intelligence |
| WO2024187572A1 (en) * | 2023-03-16 | 2024-09-19 | 中国科学院微电子研究所 | Service life prediction and repair method for resistive random access memory chip |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112327124A (en) * | 2020-09-30 | 2021-02-05 | 河海大学 | Method for monitoring thermal fatigue aging of IGBT module and method for non-uniform aging |
| US20210325277A1 (en) * | 2020-04-15 | 2021-10-21 | Aktiebolaget Skf | Method for predicting the remaining service life of a machine |
| CN113534774A (en) * | 2021-06-28 | 2021-10-22 | 长沙理工大学 | Fault prediction method, system and medium for subway brake system |
| CN113987998A (en) * | 2021-10-25 | 2022-01-28 | 西安交通大学 | Full-control type power semiconductor module power circulation method based on electric power working condition |
| CN113987932A (en) * | 2021-10-26 | 2022-01-28 | 电子科技大学 | MOSFET service life prediction method based on time series model |
-
2022
- 2022-03-24 CN CN202210294933.0A patent/CN114781128A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20210325277A1 (en) * | 2020-04-15 | 2021-10-21 | Aktiebolaget Skf | Method for predicting the remaining service life of a machine |
| CN112327124A (en) * | 2020-09-30 | 2021-02-05 | 河海大学 | Method for monitoring thermal fatigue aging of IGBT module and method for non-uniform aging |
| CN113534774A (en) * | 2021-06-28 | 2021-10-22 | 长沙理工大学 | Fault prediction method, system and medium for subway brake system |
| CN113987998A (en) * | 2021-10-25 | 2022-01-28 | 西安交通大学 | Full-control type power semiconductor module power circulation method based on electric power working condition |
| CN113987932A (en) * | 2021-10-26 | 2022-01-28 | 电子科技大学 | MOSFET service life prediction method based on time series model |
Non-Patent Citations (2)
| Title |
|---|
| 李劲松;张惠娟;杨忠;李小明;张辉斌;: "基于退化曲线相似性的剩余使用寿命估计方法", 应用科技, no. 05, 15 December 2017 (2017-12-15) * |
| 杨凌云;: "FLASH芯片寿命试验设计及测试方法", 电子技术与软件工程, no. 04, 7 February 2018 (2018-02-07) * |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| WO2024187572A1 (en) * | 2023-03-16 | 2024-09-19 | 中国科学院微电子研究所 | Service life prediction and repair method for resistive random access memory chip |
| CN117765374A (en) * | 2023-11-15 | 2024-03-26 | 知码芯(无锡)通讯技术有限公司 | Data analysis system and method based on artificial intelligence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN114781128A (en) | Chip life prediction system and method using artificial intelligence technology | |
| CN110874674B (en) | Abnormality detection method, device and equipment | |
| US8468161B2 (en) | Determining a seasonal effect in temporal data | |
| CN110909306B (en) | Business abnormality detection method and device, electronic equipment and storage equipment | |
| CN111368134B (en) | Traffic data processing method and device, electronic equipment and storage medium | |
| US20190087248A1 (en) | Anomaly detection and automated analysis using weighted directed graphs | |
| CN105653521A (en) | Data verification method and device | |
| US20100169166A1 (en) | Data quality tests for use in a causal product demand forecasting system | |
| CN115907181A (en) | Urban rail transit passenger flow prediction method and system | |
| CN118211721B (en) | Building energy consumption prediction method based on artificial intelligence | |
| CN110532122B (en) | Fault analysis method and system, electronic equipment and storage medium | |
| CN105429792A (en) | User behavior flow obtaining method and device and user behavior analysis method and system | |
| CN116051185B (en) | Advertisement position data abnormality detection and screening method | |
| CN116089402B (en) | Block chain asset data intelligent processing system | |
| CN102214129B (en) | System and method for managing internal storage capacity limit test report | |
| CN113127804B (en) | Method and device for determining number of vehicle faults, computer equipment and storage medium | |
| CN110020744A (en) | Dynamic prediction method and its system | |
| US20190196445A1 (en) | Method and system for sensing fine changes in processing/equipment measurement data | |
| CN108761170A (en) | A kind of NAND reference voltages measurement method, system, equipment and storage medium | |
| CN112561114B (en) | Bayonet flow prediction method and device, electronic equipment and storage medium | |
| CN114021302A (en) | Method, device and system for evaluating service life of power transmission line and storage medium | |
| CN114077917A (en) | Method, device, equipment and storage medium for determining event prediction model | |
| CN118114187B (en) | Method, system, device and storage medium for detecting abnormal production line of flexible board | |
| CN117893256B (en) | Big data-based app user intelligent management system | |
| JP7287093B2 (en) | Learning program, learning method and learning device |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |