Intelligent supervision system and method applied to VCSEL chip

Abstract

The invention discloses an intelligent supervision system and method applied to VCSEL chips, and relates to the technical field of chip intelligent supervision, and the method comprises the steps: obtaining historical test records corresponding to a plurality of VCSEL chips belonging to the same specification, and extracting a marking test record from the test records; taking a chip corresponding to the marked test record as a marked chip to obtain a first health value of the marked chip; according to the first health value, target values of the marked chips are obtained, and then a first set is established; obtaining all target sampling chips obtained through a current sampling rule, and establishing a second set; and according to the first set and the second set, judging whether to carry out early warning prompt on the current sampling rule. Whether the sampling rule is reasonable or not is judged by analyzing the historical test record of the chip in combination with the actual state and the numerical value of each performance test, so that the reliability of sample sampling is improved, and the difference between the test result and the real overall level is reduced.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent chip supervision, and in particular to an intelligent supervision system and method applied to a VCSEL chip. Background Art

[0002] Vertical Cavity Surface Emitting Laser (VCSEL) chips are widely used in 3D sensing, optical communications and other fields due to their unique advantages. Common applications include fingerprint recognition, face recognition, defect detection, etc. However, with the expansion of application scenarios and the growth of demand, higher requirements are placed on the performance, stability and reliability of VCSEL chips. Traditional regulatory methods are difficult to meet, and intelligent supervision has emerged. VCSEL chips need to be tested after they are manufactured, but currently test sampling mostly relies on manual or simple random sampling, which is not only subjective, but also does not consider the actual state of the chip and various characteristic parameters during sampling, resulting in a lack of representativeness of the sample and an inability to fully demonstrate the performance characteristics of the chip, resulting in a deviation between the final test results and the actual overall level. Summary of the invention

[0003] The purpose of the present invention is to provide an intelligent monitoring system and method for VCSEL chips to solve the problems raised in the prior art.

[0004] To achieve the above object, the present invention provides the following technical solutions: An intelligent monitoring method for VCSEL chips includes the following steps: Step S100: obtaining historical test records corresponding to several VCSEL chips of the same specification, where the test record is a record formed after various performance tests of the chip are performed, analyzing the values ​​corresponding to each performance test in the test record, and extracting a marked test record from the test record; Step S200: taking the chip corresponding to the marking test record as the marking chip, detecting the electrical characteristic items of the marking chip during the actual operation after the test is completed, and obtaining the first health value of the marking chip; obtaining the value of each performance indicator of the marking chip during the actual operation, and obtaining the target value of the marking chip according to the first health value; and then establishing the first set based on the target value of each marking chip; Step S300: Acquire all target sampling chips obtained by the current sampling rule, and analyze the values ​​of various performance tests of the target sampling chips during the actual operation after the test is completed to establish a second set; In this scheme, the current sampling rules are chip sampling methods determined according to actual conditions, including but not limited to manual sampling, random sampling, etc., and this current sampling rule is exactly what this scheme needs to study and analyze to see whether it is reasonable.

[0005] Step S400: Calculate the warning coefficient of the current sampling method according to the first set and the second set, and determine whether to issue a warning prompt for the current sampling rule according to the warning coefficient.

[0006] Furthermore, step S100 includes: Step S110: Acquire several historical test records, where the test records are records formed after various performance tests of the chip, wherein the performance test of the chip includes optical performance test, electrical performance test and thermal performance test; number each tested chip, extract the chip number, test time and performance value after various performance tests corresponding to each test record; Obtain the normal value range of each performance item of the chip; if the performance value corresponding to a certain performance item in a certain test record is within the normal value range corresponding to the certain performance item, the test result of the certain performance item in the certain test record is deemed to be qualified; Step S120: Obtain the normal value range of the chip leakage current [0,A 1 ], where A 1 is the maximum current value preset by the system; extract several test moments in the test time corresponding to a test record R, take the chip corresponding to a test record R as C, use a measuring instrument with leakage current measurement function to capture the leakage current value of chip C at each test moment, and mark the test moment when the leakage current value is within the normal value range; Leakage current refers to a tiny current that appears on a path where no current should flow when the chip is in a non-conducting state. When the leakage current is large, it means that the chip is more harmful. The marking test record is a record of normal performance after measurement. In this embodiment, the leakage current is used for judgment, which improves the reliability of the extraction of the marking test record. If the ratio of the number of marked test moments to the total number of test moments is greater than the preset ratio threshold, and the test results of each performance in a test record R are qualified, then the test record R is used as the marked test record, and all marked test records are obtained.

[0007] Further, step S200 includes: Step S210: Obtain the actual running device E corresponding to a certain marking chip X X , Extraction equipment E X The device startup and shutdown time recorded in the system log of , and the duration between adjacent startup and shutdown times is accumulated to obtain the total power-on time of chip X during operation; install chip X on device E XThe time above is taken as the starting point, and the time when the total power-on time reaches the set running time is taken as the detection time; For chip X, starting from the detection time, the subsequent period is P X The values ​​of the electrical characteristic items at each moment in the period P are detected to obtain the normal fluctuation range [a, b] of the electrical characteristic items of chip X, where a is the minimum fluctuation value and b is the maximum fluctuation value. X The values ​​of the electrical characteristics items are added within the normal fluctuation range, as P 1 , and then get the first health value of chip X: D X =P 1 / P X ; Step S220: For time period P X The optical performance item, electrical performance item and thermal performance item of the chip are detected at each moment, and the moment when the value of the optical performance item is greater than the preset first value threshold is marked first, and the first marking time is summarized to obtain the first marking time length F 1 , a second mark is performed when the value of the electrical performance item is greater than a preset second value threshold, and the second marking time is summarized to obtain a second marking duration F 2 , the third marking is performed when the value of the thermal performance item is greater than the preset third value threshold, and the third marking time is summarized to obtain the third marking time F 3 ; According to F 1 、F 2 、F 3 , period P X and the first health value D X , the target value of chip X is , k 1 is the first correlation coefficient; obtain the field type of the equipment actually put into operation corresponding to each marking chip, and add the target values ​​of the marking chips corresponding to the equipment with the same field type to get the average value as the first field value corresponding to each field type, and normalize the first field value, and establish the first set according to each normalized first field value.

[0008] It should be noted that in actual use, the damage degree of the chip and the use of each performance on the chip are different in the field types corresponding to each device. The field types here include fingerprint recognition, face recognition, defect detection, etc., so it is necessary to analyze the chips that are actually put into use after they are manufactured. Among them, the damage degree is determined by the health value D X Determine, the usage is determined by the sum of the marking time; when the chip health value D XThe smaller the value is, and the larger the sum of the marking durations is, the larger the target value of the chip is. In this solution, the first domain value is determined by the target value. When the first domain value corresponding to a certain domain type is large, it means that the domain type should be paid more attention to, that is, when sampling, the chips belonging to this domain type should have more sampling numbers compared to the other domain types. This also explains the analysis of the number of chips of different domain types in the following step S300, and the reliability of judging whether the current sampling rule is reasonable through the first set and the second set in step S400.

[0009] Furthermore, step S300 includes: Step S310: Obtain the actual running device E corresponding to a target sampling chip Y Y According to step S210, the detection time is obtained, and the detection time is taken as the starting point, and the subsequent period is P Y The value of the electrical characteristic item at each moment in the period P is detected, and the Y The values ​​of the electrical characteristics items are added within the normal fluctuation range, as P 2 , and then get the second health value D of chip Y Y =P 2 / P Y ; Step S320: According to the field type of the device corresponding to each target sampling chip, the total number of various field types is obtained; the total number of all chips corresponding to a certain field type is taken as N, and the second health values ​​of all chips are added and averaged to obtain M, and then the second field value of a certain field type is obtained as D=K 2 *N / M, K 2 The second correlation coefficient is normalized, and the second domain value is normalized, and the second set is established in descending order of the normalized second domain values.

[0010] Furthermore, step S400 includes: extracting elements of the same field type as the second set in the first set, and establishing a third set according to the sorting order of the extracted elements in the second set; then calculating the Pearson correlation coefficient between the two sets based on the values ​​of each element in the second set and the third set, and using the Pearson correlation coefficient as a warning coefficient. If the warning coefficient is less than a preset warning threshold, a warning prompt is issued for the current sampling rule.

[0011] An intelligent monitoring system for VCSEL chips, comprising a marking test record extraction module, a first set establishment module, a second set establishment module and an early warning prompt module; Marked test record extraction module: used to obtain the test records corresponding to a number of VCSEL chips of the same specification in history. The test records are the records formed after testing the various performances of the chips. Analyze the numerical values of each performance test in the test records, and extract the marked test records from the test records; First set establishment module: used to take the chips corresponding to the marked test records as marked chips, detect the electrical characteristic items during the actual operation process after the marked chips complete the test, and obtain the first health value of the marked chips; obtain the numerical values of each performance index of the marked chips during the actual operation process, and based on the first health value, obtain the target values of the marked chips; furthermore, based on the target values of each marked chip, establish the first set; Second set establishment module: used to obtain all target sampling chips obtained through the current sampling rule, and analyze the numerical values of various performance tests during the actual operation process after the target sampling chips complete the test, and establish the second set; Early warning prompt module: used to calculate the early warning coefficient of the current sampling method according to the first set and the second set, and judge whether to give an early warning prompt for the current sampling rule according to the early warning coefficient.

[0012] Furthermore, the marked test record extraction module includes a test record extraction unit and a marked test record extraction unit; Test record extraction unit: used to obtain a number of historical test records. The test records are the records formed after testing the various performances of the chips; obtain the normal numerical range of each performance of the chips, and judge the test results of each performance in the test records; Marked test record extraction unit: used to analyze the numerical values of each performance test corresponding in the test records, and extract the marked test records from the test records.

[0013] Furthermore, the second set establishment module includes a second health value calculation unit and a second set establishment unit; Second health value calculation unit: used to obtain the equipment actually put into operation corresponding to a certain target sampling chip, and obtain the detection time, and then calculate the second health value of a certain target sampling chip; Second set establishment unit: used to obtain the total number of various field types according to the field type of the equipment corresponding to each target sampling chip; furthermore, obtain the second field value of various field types, and then establish the second set.

[0014] Compared with the prior art, the beneficial effects of the present invention are as follows: the present invention provides an intelligent supervision system and method for VCSEL chips, including: obtaining the test records corresponding to several VCSEL chips of the same specification in history, extracting the marked test records from the test records; using the chip corresponding to the marked test record as the marked chip, and obtaining the first health value of the marked chip; and according to the first health value, obtaining the target value of the marked chip, and then establishing a first set; obtaining all target sampling chips obtained by the current sampling rules, and establishing a second set; according to the first set and the second set, judging whether to give an early warning prompt for the current sampling rules. The present invention judges whether the sampling rules are reasonable by analyzing the historical test records of the chip in combination with the actual status and the numerical values ​​of various performance tests, thereby improving the reliability of sample sampling and reducing the difference between the test results and the true overall level. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 A schematic diagram of a process of an intelligent monitoring method for VCSEL chips according to the present invention; Figure 2 This is a structural diagram of an intelligent monitoring system applied to VCSEL chips in the present invention. DETAILED DESCRIPTION

[0016] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0017] Example: Figure 1 As shown, the present invention provides a technical solution for an intelligent supervision method for VCSEL chips, comprising the following steps: Step S100: obtaining historical test records corresponding to several VCSEL chips of the same specification, where the test record is a record formed after various performance tests of the chip are performed, analyzing the values ​​corresponding to each performance test in the test record, and extracting a marked test record from the test record; Step S110: Acquire several historical test records, where the test records are records formed after various performance tests of the chip, wherein the performance test of the chip includes optical performance test, electrical performance test and thermal performance test; number each tested chip, extract the chip number, test time and performance value after various performance tests corresponding to each test record; Obtain the normal value range of each performance item of the chip; if the performance value corresponding to a certain performance item in a certain test record is within the normal value range corresponding to the certain performance item, the test result of the certain performance item in the certain test record is deemed to be qualified; Step S120: Obtain the normal value range of the chip leakage current [0,A 1 ], where A 1 is the maximum current value preset by the system; extract several test moments in the test time corresponding to a test record R, take the chip corresponding to a test record R as C, use a measuring instrument with leakage current measurement function to capture the leakage current value of chip C at each test moment, and mark the test moment when the leakage current value is within the normal value range; Leakage current refers to the tiny current that appears on a path where no current should flow when the chip is in a non-conducting state. When the leakage current is large, it means that the chip is more dangerous. The mark test record is a record of normal performance after measurement. In this embodiment, the leakage current is used for judgment, which improves the reliability of the extraction of the mark test record.

[0018] If the ratio of the number of marked test moments to the total number of test moments is greater than the preset ratio threshold, and the test results of each performance in a test record R are qualified, then the test record R is used as the marked test record, and all marked test records are obtained.

[0019] Step S200: taking the chip corresponding to the marking test record as the marking chip, detecting the electrical characteristic items of the marking chip during the actual operation after the test is completed, and obtaining the first health value of the marking chip; obtaining the value of each performance indicator of the marking chip during the actual operation, and obtaining the target value of the marking chip according to the first health value; and then establishing the first set based on the target value of each marking chip; Step S210: Obtain the actual running device E corresponding to a certain marking chip X X , Extraction equipment E X The device startup and shutdown time recorded in the system log of , and the duration between adjacent startup and shutdown times is accumulated to obtain the total power-on time of chip X during operation; install chip X on device E X The time above is taken as the starting point, and the time when the total power-on time reaches the set running time is taken as the detection time; For chip X, starting from the detection time, the subsequent period is P X The values ​​of the electrical characteristic items at each moment in the period P are detected to obtain the normal fluctuation range [a, b] of the electrical characteristic items of chip X, where a is the minimum fluctuation value and b is the maximum fluctuation value. XThe values ​​of the electrical characteristics items are added within the normal fluctuation range, as P 1 , and then get the first health value of chip X: D X =P 1 / P X ; In this embodiment, the electrical characteristic item is one of temperature, voltage or current. Because temperature, voltage and current have a great influence on the performance of the chip, high and low temperatures will accelerate chip aging and damage, and too high or too low voltage and current will also cause chip damage, thereby affecting the chip life. Therefore, when the electrical characteristic item is not within the normal fluctuation range, it means that the chip health value is low.

[0020] Step S220: For time period P X The optical performance item, electrical performance item and thermal performance item of the chip are detected at each moment, and the moment when the value of the optical performance item is greater than the preset first value threshold is marked first, and the first marking time is summarized to obtain the first marking time length F 1 , a second mark is performed when the value of the electrical performance item is greater than a preset second value threshold, and the second marking time is summarized to obtain a second marking duration F 2 , a third mark is performed when the value of the thermal performance item is greater than a preset third value threshold, and the third mark time is summarized to obtain a third mark duration F 3 ; In this embodiment, since the chip is a VCSEL, the optical performance item is the power of light emitted by the chip; the electrical performance item is the current passing through the chip; and the thermal performance item is the value of a temperature sensor deployed on the chip.

[0021] According to F 1 、F 2 、F 3 , period P X and the first health value D X , the target value of chip X is , k 1 is the first correlation coefficient; obtain the field type of the equipment actually put into operation corresponding to each marking chip, and add the target values ​​of the marking chips corresponding to the equipment with the same field type to get the average value as the first field value corresponding to each field type, and normalize the first field value, and establish the first set according to each normalized first field value.

[0022] It should be noted that in actual use, the damage degree of the chip caused by the field type corresponding to each device and the usage of each performance on the chip are different. The field types here include fingerprint recognition, face recognition, defect detection, etc. Therefore, it is necessary to analyze the chips that have been manufactured and are actually put into use. Among them, the damage degree is determined by the health value D X and the usage is determined by the sum of the marked durations; when the health value D of the chip X is smaller and the sum of the marked durations is larger, it indicates that the target value of the chip is larger. In this solution, the first field value is determined by the target value. When the first field value corresponding to a certain field type is larger, it means that more attention should be paid to this field type. That is, when sampling, for the chips belonging to this field type, compared with the other field types, there should be more sampling quantities. This also indicates the reliability of analyzing the number of chips of different field types in step S300 below and judging whether the current sampling rule is reasonable through the first set and the second set in step S400.

[0023] Step S300: Obtain all target sampling chips obtained through the current sampling rule, and after the target sampling chips complete the test, analyze the numerical values of various performance tests during the actual operation process, and establish a second set; Step S310: Obtain the device E actually put into operation corresponding to a certain target sampling chip Y Y , and according to step S210, obtain the detection time. Taking the detection time as the starting point, detect the numerical values of the electrical characteristic items at each moment in the subsequent time period P Y . Add up the moments when the numerical values of the electrical characteristic items are within the normal fluctuation range within the time period P Y , and use it as P 2 . Furthermore, obtain the second health value D of chip Y Y =P 2 / P Y ; Step S320: According to the field type of the device corresponding to each target sampling chip, obtain the total quantity of various field types; and take the total quantity of all chips corresponding to a certain field type as N, and add up the second health values of all chips and find the average value to get M. Furthermore, obtain the second field value of a certain field type as D = K 2 *N / M, where K 2 is the second correlation coefficient, and normalize the second field value. Establish a second set in the order from largest to smallest of the normalized second field values.

[0024] Step S400: Calculate the warning coefficient of the current sampling method according to the first set and the second set, and judge whether to give a warning prompt for the current sampling rule according to the warning coefficient.

[0025] Extract elements of the same field type as those in the second set from the first set, and establish a third set according to the sorting order of the extracted elements in the second set; then, based on the values ​​of each element in the second set and the third set, calculate the Pearson correlation coefficient between the two sets, and use the Pearson correlation coefficient as a warning coefficient. If the warning coefficient is less than the preset warning threshold, a warning prompt is issued for the current sampling rule.

[0026] The calculation of the Pearson correlation coefficient belongs to the prior art, and the actual calculation steps are not repeated here. The value range of the Pearson correlation coefficient is r∈[-1,1]. When r=1, it means that there is a completely positive linear correlation between the two variables; when r=0, it means that there is no linear correlation between the two variables; when r=1, it means that there is a completely negative linear correlation between the two variables. In this embodiment, the warning threshold is 0.5, that is, when the value of the Pearson correlation coefficient is less than 0.5, it is considered that the current sampling rule is unreasonable, and a warning prompt is issued for the current sampling rule, and the relevant technical personnel are informed.

[0027] The present invention also provides an intelligent monitoring system applied to VCSEL chips, including a mark test record extraction module, a first set establishment module, a second set establishment module and an early warning prompt module; Marked test record extraction module: used to obtain the test records corresponding to several VCSEL chips of the same specification. The test record is a record formed after the various performance tests of the chip are performed. The numerical value of each performance test corresponding to the test record is analyzed, and the marked test record is extracted from the test record; The first set establishment module is used to use the chip corresponding to the marking test record as the marking chip, detect the electrical characteristic items of the marking chip during the actual operation after the test is completed, and obtain the first health value of the marking chip; obtain the value of each performance indicator of the marking chip during the actual operation, and obtain the target value of the marking chip according to the first health value; and then establish the first set based on the target value of each marking chip; The second set establishment module is used to obtain all target sampling chips obtained by the current sampling rules, and to analyze the values ​​of various performance tests of the target sampling chips during the actual operation after the test is completed, so as to establish the second set; Early warning prompt module: used to calculate the early warning coefficient of the current sampling method according to the first set and the second set, and determine whether to issue an early warning prompt for the current sampling rule according to the early warning coefficient.

[0028] Wherein, the marked test record extraction module includes a test record extraction unit and a marked test record extraction unit; Test record extraction unit: used to obtain several historical test records, which are records formed after testing various performances of the chip; obtain the normal value range of each performance of the chip, and judge the test results of each performance in the test record; Marked test record extraction unit: used for analyzing the numerical value of each performance test corresponding to the test record, and extracting the marked test record from the test record.

[0029] The second set establishment module includes a second health value calculation unit and a second set establishment unit; A second health value calculation unit: used to obtain the device actually put into operation corresponding to a certain target sampling chip, and obtain the detection time, and then calculate the second health value of the certain target sampling chip; The second set establishing unit is used to obtain the total quantity of various field types according to the field type of the device corresponding to each target sampling chip; and then obtain the second field value of each field type, and then establish the second set.

[0030] It will be apparent to those skilled in the art that the invention is not limited to the details of the exemplary embodiments described above and that the invention can be implemented in other specific forms without departing from the spirit or essential features of the invention. Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description, and it is intended that all variations falling within the meaning and scope of the equivalent elements of the claims be included in the invention. Any reference numeral in a claim should not be considered as limiting the claim to which it relates.

Claims

1. An intelligent monitoring method for VCSEL chips, characterized in that: The following steps are involved: Step S100: obtaining historical test records corresponding to several VCSEL chips of the same specification, wherein the test records are records formed after various performances of the chip are tested, analyzing the values ​​corresponding to each performance test in the test records, and extracting marked test records from the test records; Step S200: taking the chip corresponding to the marking test record as the marking chip, detecting the electrical characteristic items of the marking chip during the actual operation after the test is completed, and obtaining the first health value of the marking chip; obtaining the value of each performance indicator of the marking chip during the actual operation, and obtaining the target value of the marking chip according to the first health value; and then establishing the first set based on the target value of each marking chip; Step S300: Acquire all target sampling chips obtained by the current sampling rule, and analyze the values ​​of various performance tests of the target sampling chips during the actual operation after the test is completed to establish a second set; Step S400: Calculate the warning coefficient of the current sampling method according to the first set and the second set, and determine whether to issue a warning prompt for the current sampling rule according to the warning coefficient.

2. The intelligent monitoring method for VCSEL chips according to claim 1, characterized in that: Step S100 includes: Step S110: Acquire several historical test records, wherein the test records are records formed after various performance tests of the chip are performed, wherein the performance test of the chip includes an optical performance test, an electrical performance test, and a thermal performance test; number each tested chip, and extract the chip number, test time, and performance values ​​after various performance tests corresponding to each test record; Obtain the normal value range of each performance of the chip; if the performance value corresponding to a certain performance in a certain test record is within the normal value range corresponding to the certain performance, the test result of the certain performance in the certain test record is deemed to be qualified; Step S120: obtaining a normal value range [0, A1] of the leakage current of the chip, where A1 is the maximum current value preset by the system; extracting several test moments in the test time corresponding to a test record R, taking the chip corresponding to the test record R as C, using a measuring instrument with a leakage current measurement function, capturing the leakage current value of chip C at each test moment, and marking the test moment when the leakage current value is within the normal value range; If the ratio of the number of marked test moments to the total number of test moments is greater than a preset ratio threshold, and the test results of each performance item in the test record R are qualified, then the test record R is used as a marked test record, and then all marked test records are obtained.

3. The intelligent monitoring method for VCSEL chips according to claim 2 is characterized in that: Step S200 includes: Step S210: Obtain the actual equipment E that is put into operation corresponding to a certain marking chip X X , Extraction equipment E X The device startup and shutdown time recorded in the system log of , and the duration between adjacent startup and shutdown times is accumulated to obtain the total power-on time of chip X during operation; install chip X on device E X The time above is taken as the starting point, and the time when the total power-on time reaches the set running time is taken as the detection time; For chip X, starting from the detection time, the subsequent period is P X The values ​​of the electrical characteristic items at each moment in the period P are detected to obtain the normal fluctuation range [a, b] of the electrical characteristic items of chip X, where a is the minimum fluctuation value and b is the maximum fluctuation value. X The values ​​of the electrical characteristic items within the normal fluctuation range are added together as P1, and then the first health value of the chip X is obtained: D X =P1 / P X ; Step S220: For time period P X The values ​​of the optical performance item, the electrical performance item and the thermal performance item of the chip at each moment are detected, and the moment when the value of the optical performance item is greater than the preset first numerical threshold is marked for the first time, and the first marked moments are summarized to obtain the first marked duration F1, the moment when the value of the electrical performance item is greater than the preset second numerical threshold is marked for the second time, and the second marked moments are summarized to obtain the second marked duration F2, and the moment when the value of the thermal performance item is greater than the preset third numerical threshold is marked for the third time, and the third marked moments are summarized to obtain the third marked duration F3; According to F1, F2, F3, time period P X and the first health value D X , the target value of chip X is , k1 is the first correlation coefficient; obtain the field type of the equipment actually put into operation corresponding to each marker chip, and add the target values ​​of the marker chips corresponding to the equipment with the same field type to get the average value as the first field value corresponding to each field type, and normalize the first field value, and establish the first set according to each normalized first field value.

4. The intelligent monitoring method for VCSEL chips according to claim 3 is characterized in that: Step S300 includes: Step S310: Obtain the actual running device E corresponding to a target sampling chip Y Y According to step S210, the detection time is obtained, and the detection time is taken as the starting point, and the subsequent period is P Y The value of the electrical characteristic item at each moment in the period P is detected, and the Y The values ​​of the electrical characteristic items are added within the normal fluctuation range as P2, and then the second health value D of the chip Y is obtained. Y =P2 / P Y ; Step S320: According to the field type of the device corresponding to each target sampling chip, the total number of various field types is obtained; and the total number of all chips corresponding to a certain field type is taken as N, and the second health values ​​of all the chips are added and averaged to obtain M, and then the second field value of the certain field type is obtained as D=K2*N / M, K2 is the second correlation coefficient, and the second field value is normalized, and the second set is established in the order of the normalized second field values ​​from large to small.

5. The intelligent monitoring method for VCSEL chips according to claim 1, characterized in that: Step S400 includes: extracting elements of the same field type as the second set in the first set, and establishing a third set according to the sorting order of the extracted elements in the second set; then calculating the Pearson correlation coefficient between the two sets based on the values ​​of each element in the second set and the third set, and using the Pearson correlation coefficient as a warning coefficient. If the warning coefficient is less than a preset warning threshold, a warning prompt is issued for the current sampling rule.

6. An intelligent monitoring system, used to execute an intelligent monitoring method for VCSEL chips as described in any one of claims 1 to 5, characterized in that: The system includes a marking test record extraction module, a first set establishment module, a second set establishment module and an early warning prompt module; Marked test record extraction module: used to obtain historical test records corresponding to several VCSEL chips of the same specification, wherein the test record is a record formed after various performance tests of the chip are performed, and the numerical value of each performance test corresponding to the test record is analyzed, and the marked test record is extracted from the test record; The first set establishment module is used to use the chip corresponding to the marking test record as the marking chip, detect the electrical characteristic items of the marking chip during the actual operation after the test is completed, and obtain the first health value of the marking chip; obtain the value of each performance indicator of the marking chip during the actual operation, and obtain the target value of the marking chip according to the first health value; and then establish the first set based on the target value of each marking chip; The second set establishment module is used to obtain all target sampling chips obtained by the current sampling rules, and to analyze the values ​​of various performance tests of the target sampling chips during the actual operation after the test is completed, so as to establish the second set; Early warning prompt module: used to calculate the early warning coefficient of the current sampling method according to the first set and the second set, and determine whether to issue an early warning prompt for the current sampling rule according to the early warning coefficient.

7. An intelligent monitoring system according to claim 6, characterized in that: The marked test record extraction module includes a test record extraction unit and a marked test record extraction unit; Test record extraction unit: used to obtain several historical test records, wherein the test records are records formed after various performances of the chip are tested; obtain the normal value range of each performance of the chip, and judge the test results of each performance in the test record; Marked test record extraction unit: used for analyzing the numerical value of each performance test corresponding to the test record, and extracting the marked test record from the test record.

8. An intelligent monitoring system according to claim 7, characterized in that: The second set establishing module includes a second health value calculating unit and a second set establishing unit; A second health value calculation unit: used to obtain the device actually put into operation corresponding to a certain target sampling chip, and obtain the detection time, and then calculate the second health value of the certain target sampling chip; The second set establishing unit is used to obtain the total quantity of various field types according to the field type of the device corresponding to each target sampling chip; and then obtain the second field value of each field type, and then establish the second set.

Images