Storage chip aging test management method and system

Abstract

A method and system for managing aging test of storage chips are disclosed. First, the test status information of the laboratory is collected. Then, based on the test status information, whether the test environment of the laboratory is normal is determined. Then, in response to the test environment of the laboratory being normal, a monitoring server is built. The monitoring server is used for device client access. The monitoring server records basic data. Finally, the loading and unloading equipment and the test equipment are connected to the monitoring server. After the test is connected, the storage chip is subjected to aging test. In this way, the accuracy and reliability of the aging test of the storage chip can be improved.

Description

Technical Field

[0001] The present application relates to the field of intelligent test management, and more specifically, to a storage chip aging test management method and system. Background Art

[0002] Memory chips are electronic components used to store data. They have the advantages of high speed, large capacity, and low power consumption, and are widely used in various electronic devices. In order to ensure the performance and reliability of memory chips, they need to be subjected to aging tests before entering the market, that is, they need to be run for a long time under extreme environments such as high temperature and high pressure to detect whether they have defects or failures. Aging tests are an important part of the memory chip production process, which directly affects the quality and cost of memory chips.

[0003] However, the current memory chip aging test is mainly based on engineering testing and verification in the laboratory. For large-scale mass-produced memory chips, due to the complexity of chip types, aging test boards (there are several positions on a test board where materials can be loaded, and the status of each needs to be recorded), loading conditions, and test procedures, conventional offline management by engineers can no longer meet the needs of production line testing. In addition, since the test environment in the laboratory is difficult to maintain stability, that is, due to changes in factors such as temperature, humidity, air conditioning, and instruments in the laboratory, the results of the aging test may be inaccurate or unreliable.

[0004] Therefore, an optimized memory chip aging test management solution is desired. Summary of the invention

[0005] In order to solve the above technical problems, the present application is proposed. The present application provides a memory chip aging test management method and system, which can improve the accuracy and reliability of the memory chip aging test.

[0006] According to one aspect of the present application, a memory chip aging test management method is provided, which includes:

[0007] Collect laboratory test status information;

[0008] Based on the test status information, determining whether the test environment of the laboratory is normal;

[0009] In response to the test environment of the laboratory being normal, building a monitoring server, the monitoring server being used for device client access, wherein the monitoring server has basic data recorded; and

[0010] The loading and unloading equipment and the testing equipment are connected to the monitoring server, and after the test connection, the storage chip is subjected to an aging test.

[0011] According to another aspect of the present application, a memory chip aging test management system is provided, which includes:

[0012] Information collection module, used to collect laboratory test status information;

[0013] An environment analysis module, used to determine whether the test environment of the laboratory is normal based on the test status information;

[0014] A monitoring server building module, used for building a monitoring server in response to the test environment of the laboratory being normal, wherein the monitoring server is used for device client access, wherein the monitoring server has basic data recorded; and

[0015] The aging test module is used to connect the loading and unloading equipment and the testing equipment to the monitoring server, and after the test connection, perform an aging test on the storage chip.

[0016] Compared with the prior art, the storage chip aging test management method and system provided by the present application first collects the test status information of the laboratory, then, based on the test status information, determines whether the test environment of the laboratory is normal, and then, in response to the normal test environment of the laboratory, sets up a monitoring server, which is used for device client access, wherein the monitoring server records basic data, and finally, connects the loading and unloading equipment and the test equipment to the monitoring server, and after the test is connected, performs an aging test on the storage chip. In this way, the accuracy and reliability of the storage chip aging test can be improved. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present application, the following briefly introduces the drawings required for use in the description of the embodiments. The following drawings are not intentionally scaled to the actual size, and the focus is on illustrating the main purpose of the present application.

[0018] Figure 1 The present invention is a flowchart of a method for managing aging test of a memory chip according to an embodiment of the present application.

[0019] Figure 2 Flow chart of sub-step S120 of the memory chip aging test management method according to an embodiment of the present application.

[0020] Figure 3 Schematic diagram of the architecture of sub-step S120 of the memory chip aging test management method according to an embodiment of the present application.

[0021] Figure 4 This is a flowchart of sub-step S123 of the memory chip aging test management method according to an embodiment of the present application.

[0022] Figure 54 is a block diagram of a memory chip aging test management system according to an embodiment of the present application.

[0023] Figure 6 This is an application scenario diagram of the memory chip aging test management method according to an embodiment of the present application. DETAILED DESCRIPTION

[0024] The technical solutions in the embodiments of the present application will be described clearly and completely below in conjunction with the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present application, not all embodiments. Based on the embodiments of the present application, all other embodiments obtained by ordinary technicians in this field without creative work also fall within the scope of protection of the present application.

[0025] As shown in this application and claims, unless the context clearly indicates an exception, the words "a", "an", "an" and / or "the" do not refer to the singular and may also include the plural. Generally speaking, the terms "include" and "comprise" only indicate the inclusion of the steps and elements that have been clearly identified, and these steps and elements do not constitute an exclusive list. The method or device may also include other steps or elements.

[0026] Although the present application makes various references to certain modules in the system according to the embodiments of the present application, any number of different modules can be used and run on the user terminal and / or server. The modules are only illustrative, and different aspects of the system and method can use different modules.

[0027] Flowcharts are used in the present application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed accurately in order. On the contrary, various steps may be processed in reverse order or simultaneously as required. Meanwhile, other operations may also be added to these processes, or a certain step or several steps of operations may be removed from these processes.

[0028] Below, the exemplary embodiments according to the present application will be described in detail with reference to the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present application, rather than all the embodiments of the present application, and it should be understood that the present application is not limited to the exemplary embodiments described here.

[0029] Specifically, in the technical solution of the present application, a memory chip aging test management method is proposed. Figure 1 FIG. 1 is a flow chart of a method for managing a memory chip aging test according to an embodiment of the present application. Figure 1As shown, the storage chip aging test management method according to the embodiment of the present application includes the steps of: S110, collecting the test status information of the laboratory; S120, determining whether the test environment of the laboratory is normal based on the test status information; S130, in response to the test environment of the laboratory being normal, building a monitoring server, the monitoring server being used for device client access, wherein the monitoring server records basic data; and, S140, connecting the loading and unloading equipment and the test equipment to the monitoring server, and performing an aging test on the storage chip after the test is connected.

[0030] Wherein, in step S110, the test status information includes test status, test schedule, cleaning status, test instruments, laboratory temperature data, humidity data, operating status of laboratory temperature control air conditioners, and operating status of laboratory equipment. Wherein, in step S130, the monitoring server records basic data, and the basic data includes device type, device ID, aging test board ID, type and distribution array of aging test boards, test product ID, and program required for test products.

[0031] In particular, in the process of performing an aging test on a memory chip, the device communication instructions of the loading and unloading equipment in the loading stage are monitored, and based on the device communication instructions, the batch information of the loaded samples to be tested and the status information of the aging test board are obtained from the monitoring server, wherein the type of the aging test board matches the batch information, and is responded to the loading and unloading equipment according to the protocol content and format. In response to a correct response to the communication protocol content, the loading and unloading equipment automatically executes a loading procedure for the aging test board, and after the execution of the loading procedure is completed, the status of the aging test board is updated to the monitoring server, and the loading procedure of the next aging test board is executed until the loading of the batch of samples to be tested is completed. In response to an error in the communication protocol content response, the loading and unloading equipment cannot proceed to the next step.

[0032] Furthermore, the device communication instructions of the loading and unloading equipment in the unloading stage are monitored, and based on the device communication instructions, the batch information of the unloaded samples to be tested and the updated status of the aging test board are obtained from the monitoring server, and the unloading and unloading equipment is responded to according to the protocol content and format. In response to the correct response of the communication protocol content, the loading and unloading equipment automatically executes the unloading procedure for the aging test board, and after the unloading procedure is completed, the unloading procedure of the next aging test board is executed until the unloading of the batch of samples to be tested is completed.

[0033] Accordingly, considering that in the above-mentioned memory chip aging test process, it is difficult to keep the test environment of the laboratory stable, that is, due to changes in factors such as temperature, humidity, air conditioning, and instruments in the laboratory, the results of the aging test may be inaccurate or unreliable. Based on this, in order to improve the efficiency and accuracy of the memory chip aging test, the technical concept of the present application is that when performing the memory chip aging test, it is necessary to manage and monitor the test status information of the laboratory, and introduce data processing and semantic understanding algorithms at the back end to perform semantic association analysis of these test status information, so as to determine whether the test environment of the laboratory is normal, so as to ensure the accuracy and reliability of the memory chip aging test. In this way, it is possible to automatically detect the normality of the laboratory test environment based on the test status information, and take corresponding measures to manage and adjust in time when the environment is abnormal, thereby improving the efficiency and accuracy of the memory chip aging test and ensuring the quality and reliability of the memory chip.

[0034] Figure 2 Flow chart of sub-step S120 of the memory chip aging test management method according to an embodiment of the present application. Figure 3 FIG. 1 is a schematic diagram of the architecture of sub-step S120 of the memory chip aging test management method according to an embodiment of the present application. Figure 2 and Figure 3 As shown, according to the storage chip aging test management method of the embodiment of the present application, based on the test status information, determining whether the test environment of the laboratory is normal includes: S121, encoding each data item in the test status information to obtain a plurality of test status information item encoding feature vectors; S122, performing context test status semantic association encoding on the plurality of test status information item encoding feature vectors to obtain a plurality of context test status information item encoding feature vectors; S123, performing semantic importance weighted analysis based on the overall feature distribution on each context test status information item encoding feature vector in the plurality of context test status information item encoding feature vectors to obtain a global semantic encoding feature of the test status information; and, S124, determining whether there is any abnormality in the laboratory based on the global semantic encoding feature of the test status information.

[0035] Specifically, in the technical solution of the present application, in the process of determining whether the test environment of the laboratory is normal based on the test status information, since the test status information includes test status, test scheduling time, cleaning status, test instruments, laboratory temperature data, humidity data, the operating status of the laboratory temperature control air conditioner and the operating status of the laboratory equipment, each data item in the test status information has semantic information of different data types. Therefore, in order to be able to map different types of data items to feature spaces of the same dimension, and to perform semantic understanding of each data item in the test status information, and to facilitate subsequent data processing and analysis, in the technical solution of the present application, each data item in the test status information is encoded to obtain multiple test status information item encoding feature vectors. In particular, the encoding process can be designed according to specific data items and requirements. Common encoding methods include one-hot encoding, label encoding, hash encoding, etc. These encoding methods can convert different types of data into numerical or vector representations to meet subsequent data processing and analysis requirements. By encoding each data item in the test status information, complex status information can be converted into a feature vector that is easy to analyze, which is convenient for subsequent data analysis and modeling, thereby achieving management and monitoring of the laboratory test environment.

[0036] Next, it is considered that the coding features of each data item in the test status information have an associated relationship with the laboratory environment state. Therefore, in order to capture the association and context information between the coding features of each data item in the test status information, so as to improve the accuracy of laboratory environment anomaly detection, in the technical solution of the present application, the multiple test status information item coding feature vectors are further passed through a context encoder based on a converter module to obtain multiple context test status information item coding feature vectors. Such a feature representation can capture the association and context information between the coding features of each data item in the test status information, which is conducive to better reflecting the overall semantics and correlation of the test status information, thereby improving the accuracy of laboratory environment anomaly detection to ensure the effect of storage chip aging test.

[0037] Correspondingly, in step S122, contextual test state semantic association encoding is performed on the multiple test state information item encoding feature vectors to obtain multiple contextual test state information item encoding feature vectors, including: passing the multiple test state information item encoding feature vectors through a context encoder based on a converter module to obtain the multiple contextual test state information item encoding feature vectors.

[0038] It should be understood that when the abnormal state detection of the laboratory is actually performed, different data items in the test state information have different importance. That is to say, in the multiple context test state information item encoding feature vectors, each feature vector represents the encoding feature of a context test state information item. These feature vectors have different importance and contribution. Therefore, in order to more fully capture the global semantics and importance of the test state information, so as to better represent the overall semantics of the test state information, thereby improving the accuracy of the abnormal detection of the laboratory environment state, in the technical solution of the present application, the semantic weight value of each context test state information item encoding feature vector in the multiple context test state information item encoding feature vectors relative to the feature distribution of the multiple context test state information item encoding feature vectors is further calculated to obtain a sequence of semantic weight values, and based on the sequence of semantic weight values, the multiple context test state information item encoding feature vectors are fused to obtain the global semantic encoding feature vector of the test state information. It should be understood that by calculating the semantic weight value, the importance of the encoding feature of each context test state information item in the overall feature distribution can be determined, and based on this importance, the encoding features of each context test state information item are weighted and fused. In this way, the obtained global semantic coding feature vector of the test state information integrates the important semantic features of each context test state information item, and can better reflect the importance and overall associated semantics of the test state information.

[0039] Accordingly, in step S123, if Figure 4 As shown, a semantic importance weighted analysis based on the overall feature distribution is performed on each of the multiple contextual test state information item encoding feature vectors to obtain a global semantic encoding feature of the test state information, including: S1231, calculating the semantic weight value of each of the multiple contextual test state information item encoding feature vectors relative to the overall feature distribution of the multiple contextual test state information item encoding feature vectors to obtain a sequence of semantic weight values; and, S1232, based on the sequence of semantic weight values, fusing the multiple contextual test state information item encoding feature vectors to obtain a global semantic encoding feature vector of the test state information as the global semantic encoding feature of the test state information.

[0040] Wherein, in step S1231, the semantic weight value of each contextual test state information item encoding feature vector in the multiple contextual test state information item encoding feature vectors relative to the overall feature distribution of the multiple contextual test state information item encoding feature vectors is calculated to obtain a sequence of semantic weight values, including: calculating the semantic weight value of each contextual test state information item encoding feature vector in the multiple contextual test state information item encoding feature vectors relative to the overall feature distribution of the multiple contextual test state information item encoding feature vectors using the following weight formula to obtain the sequence of semantic weight values; wherein, the weight formula is:

[0041]

[0042] Where A is 1×N w The matrix, N w is the dimension of each contextual trial state information item encoding feature vector in the plurality of contextual trial state information item encoding feature vectors, and B is 1×N h The matrix, N h is the dimension of the feature vector encoding each of the contextual trial state information items, σ is the Sigmoid function, and w i is the i-th contextual trial state information item encoding feature vector among the plurality of contextual trial state information item encoding feature vectors, h i is the i-th context trial state information item encoding feature vector among multiple context trial state information item encoding feature vectors, s i is a sequence of semantic weight values.

[0043] Then, the global semantic coding feature vector of the test status information is passed through a classifier to obtain a classification result, which is used to indicate whether the laboratory is abnormal. In other words, the classification process is performed by using the weighted fusion feature information based on the importance of the coding features of each data item in the test status information to determine whether the laboratory test environment is normal, and take corresponding measures to manage and adjust the environment in a timely manner when an abnormality is detected, thereby ensuring the accuracy and reliability of the memory chip aging test.

[0044] Accordingly, in step S124, based on the global semantic coding features of the test status information, it is determined whether there is an abnormality in the laboratory, including: passing the global semantic coding feature vector of the test status information through a classifier to obtain a classification result, and the classification result is used to indicate whether there is an abnormality in the laboratory.

[0045] Specifically, the global semantic encoding feature vector of the test status information is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether there is an abnormality in the laboratory, including: using the fully connected layer of the classifier to fully connect encode the global semantic encoding feature vector of the test status information to obtain an encoded classification feature vector; and inputting the encoded classification feature vector into the Softmax classification function of the classifier to obtain the classification result.

[0046] That is, in the technical solution of the present application, the labels of the classifier include the presence of an abnormality in the laboratory (first label), and the absence of an abnormality in the laboratory (second label), wherein the classifier determines to which classification label the characteristic vector representing the consistency of the mixture spreading state belongs through a soft maximum function. It is worth noting that the first label p1 and the second label p2 here do not contain artificially set concepts. In fact, during the training process, the computer model does not have the concept of "whether there is an abnormality in the laboratory". It only has two classification labels and the probability of the output feature under these two classification labels, that is, the sum of p1 and p2 is one. Therefore, the classification result of whether there is an abnormality in the laboratory is actually converted into a binary class probability distribution that conforms to the laws of nature through the classification label. In essence, the physical meaning of the natural probability distribution of the label is used, rather than the linguistic text meaning of "whether there is an abnormality in the laboratory".

[0047] It should be understood that the role of the classifier is to use the given categories and known training data to learn classification rules and classifiers, and then classify (or predict) unknown data. Logistic regression and SVM are often used to solve binary classification problems. For multi-class classification problems, logistic regression or SVM can also be used, but multiple binary classifications are required to form a multi-classification. However, this is prone to errors and is not efficient. Commonly used multi-classification methods include the Softmax classification function.

[0048] Furthermore, in the technical solution of the present application, the memory chip aging test management method further includes a training step: for training the context encoder based on the converter module and the classifier.

[0049] In a specific example, the training step includes: collecting training test state information of the laboratory; encoding each data item in the training test state information to obtain a plurality of training test state information item encoding feature vectors; passing the plurality of training test state information item encoding feature vectors through the context encoder based on the converter module to obtain a plurality of training context test state information item encoding feature vectors; calculating the training semantic weight value of each training context test state information item encoding feature vector in the plurality of training context test state information item encoding feature vectors relative to the overall feature distribution of the plurality of training context test state information item encoding feature vectors to obtain a sequence of training semantic weight values; based on the sequence of training semantic weight values, fusing the plurality of training context test state information item encoding feature vectors to obtain a global semantic encoding feature vector of training test state information; performing information convergence optimization on the global semantic encoding feature vector of the training test state information to obtain an optimized global semantic encoding feature vector of training test state information; passing the optimized global semantic encoding feature vector of training test state information through the classifier to obtain a classification loss function value; and, training the context encoder based on the converter module and the classifier based on the classification loss function value.

[0050] In particular, in the technical solution of the present application, each of the plurality of training context test state information item encoding feature vectors expresses the contextual long-range semantic features of the test state information item, so the contextual long-range semantic feature encoding features of each of the individual test state information items will have a better semantic information convergence effect, that is, each of the training context test state information item encoding feature vectors has self-space convergence. However, when calculating the semantic attention weight value of each of the plurality of training context test state information item encoding feature vectors relative to the overall feature distribution of the plurality of training context test state information item encoding feature vectors, and fusing the plurality of training context test state information item encoding feature vectors based on the sequence of the semantic weight values ​​to obtain the global semantic encoding feature vector of the training test state information, the self-space convergence weight attributable to each of the training context test state information item encoding feature vectors will result in the discretization of the semantic space information game between each of the training context test state information item encoding feature vectors, thereby affecting the semantic information expression effect of the global semantic encoding feature vector of the training test state information.

[0051] Based on this, the applicant of the present application preferably performs information convergence optimization on the global semantic encoding feature vector of the training trial state information when the global semantic encoding feature vector of the training trial state information is trained through a classifier.

[0052] Accordingly, in one example, performing information convergence optimization on the global semantic coding feature vector of the training test state information to obtain an optimized global semantic coding feature vector of the training test state information includes: performing information convergence optimization on the global semantic coding feature vector of the training test state information to obtain the optimized global semantic coding feature vector of the training test state information using the following optimization formula; wherein the optimization formula is:

[0053]

[0054] Among them, v i is the ith eigenvalue of the global semantic encoding feature vector of the training trial state information, L is a scale hyperparameter, log is a logarithmic function with base 2, and v i ′ is the i-th eigenvalue of the global semantic encoding feature vector of the optimized training test state information.

[0055] Specifically, when the global semantic coding feature vector of the training test state information is trained through a classifier, when the classifier acts on the global semantic coding feature vector of the training test state information, due to the dense characteristics of the feature generation itself, the position information game discretization between the eigenvalues ​​at each position of the global semantic coding feature vector of the training test state information will produce a large-scale information game, resulting in the generated solution being unable to converge to a Nash equilibrium on the basis of the game, especially in the presence of large-scale non-perfect game discretization information in units of the feature vector encoding the training context test state information item. In this way, by performing an equivalent convergence of information game equalization on the global semantic coding feature vector of the training test state information based on the vector information self-controlled equilibrium neighborhood of the global semantic coding feature vector of the training test state information, convergence can be promoted through self-game of the eigenvalues ​​in the local neighborhood, thereby improving the accuracy of the classification result obtained by the global semantic coding feature vector of the training test state information through the classifier. In this way, the laboratory test environment can be automatically detected for anomalies based on the test status information, and corresponding measures can be taken in time to manage and adjust the environment when an abnormality is detected, thereby improving the efficiency and accuracy of the memory chip aging test and ensuring the quality and reliability of the memory chip.

[0056] In summary, the memory chip aging test management method based on the embodiment of the present application is explained, which can improve the accuracy and reliability of the memory chip aging test.

[0057] Figure 5 FIG. 1 is a block diagram of a memory chip aging test management system 100 according to an embodiment of the present application. Figure 5As shown, according to the storage chip aging test management system 100 of the embodiment of the present application, it includes: an information collection module 110, which is used to collect the test status information of the laboratory; an environment analysis module 120, which is used to determine whether the test environment of the laboratory is normal based on the test status information; a monitoring server construction module 130, which is used to build a monitoring server in response to the normal test environment of the laboratory, and the monitoring server is used for device client access, wherein the monitoring server records basic data; and an aging test module 140, which is used to connect the loading and unloading equipment and the test equipment to the monitoring server, and after the test is connected, perform aging test on the storage chip.

[0058] Here, those skilled in the art can understand that the specific functions and operations of each module in the above-mentioned memory chip aging test management system 100 have been described in the above reference. Figures 1 to 4 The present invention has been described in detail in the description of the memory chip aging test management method, and therefore, its repeated description will be omitted.

[0059] As described above, the memory chip aging test management system 100 according to the embodiment of the present application can be implemented in various wireless terminals, such as a server with a memory chip aging test management algorithm. In one example, the memory chip aging test management system 100 according to the embodiment of the present application can be integrated into the wireless terminal as a software module and / or a hardware module. For example, the memory chip aging test management system 100 can be a software module in the operating system of the wireless terminal, or can be an application developed for the wireless terminal; of course, the memory chip aging test management system 100 can also be one of the many hardware modules of the wireless terminal.

[0060] Alternatively, in another example, the memory chip aging test management system 100 and the wireless terminal may also be separate devices, and the memory chip aging test management system 100 may be connected to the wireless terminal via a wired and / or wireless network and transmit interactive information in accordance with an agreed data format.

[0061] Figure 6 FIG. 1 is an application scenario diagram of the memory chip aging test management method according to an embodiment of the present application. Figure 6 As shown, in this application scenario, first, the laboratory test status information is collected (for example, Figure 6 D), and then, the test status information is input to a server (for example, Figure 6 In S) shown in , the server is capable of using the storage chip aging test management algorithm to process the test status information to obtain a classification result indicating whether there is an abnormality in the laboratory.

[0062] Furthermore, the memory chip aging test management method of the embodiment of the present application further includes: a loading monitoring response step, a test monitoring response step, and a unloading monitoring response step.

[0063] Specifically, in the loading monitoring response step, this step is completed in the loading and unloading equipment (Load / Unload). Before this step, the monitoring server (i.e., Burn-in Board Server) enters the basic data in advance (device type, device ID, aging test board ID, aging test board type and distribution array, test product ID, test product required program, etc.), and obtains the batch, quantity and other basic information of the product to be tested from the manufacturing execution system.

[0064] Monitor the device communication instructions of the loading and unloading equipment (Load / Unload) in the loading stage, obtain the information of the batch to be loaded and the status of the aging test board from the monitoring server according to the communication instructions of the device (the model of the aging test board matches the batch to be loaded, and the aging board is in an available state), and respond correctly to the monitored loading and unloading equipment (Load / Unload) according to the protocol content and format, and the loading and unloading equipment (Load / Unload) continues to the next step. If the communication protocol content response is wrong (part of the response content is missing or the format is abnormal), the loading and unloading equipment (Load / Unload) cannot continue to the next step, and prompts the abnormality and the cause of the abnormality. The production line operator handles the problem according to the abnormal prompt.

[0065] After the monitoring server responds correctly, the loading and unloading equipment (Load / Unload) automatically loads the aging test board. After the board is loaded, the board status and the loading information on the board will be updated to the monitoring server. At the same time, the loading and unloading equipment (Load / Unload) loads the next test board until the loading of this batch of samples is completed.

[0066] In the test monitoring response step, this step is completed in the test equipment (Burn-in Test equipment). Before this, the operator needs to put the aging test board that has been loaded into the test equipment. The device communication instructions of the test phase of the monitoring test equipment are obtained from the monitoring server for the batch that needs to have been loaded, the status of the corresponding aging test board that has been loaded, and the test program.

[0067] Operate according to the operating manual of the test equipment, monitor the loading information of the test equipment, and update the loading information to the monitoring server to ensure that the loading information on the monitoring server is the latest. Monitor the test conditions and test progress of the test equipment until the test is completed, upload the test results to the monitoring server, and generate the unloading information of these test boards in the batch in the monitoring server according to the test results.

[0068] In the unloading monitoring response step, the device communication instructions of the unloading and loading equipment in the unloading stage are monitored, and the batch information of the unloaded samples to be tested and the updated status of the aging test board are obtained from the monitoring server based on the device communication instructions, and the unloading and loading equipment is responded to according to the protocol content and format. In response to the correct response of the communication protocol content, the unloading and loading equipment automatically executes the unloading procedure for the aging test board, and after the unloading procedure is completed, the unloading procedure of the next aging test board is executed until the unloading of the batch of samples to be tested is completed.

[0069] Specifically, the unloading monitoring response step is completed in the loading and unloading equipment (Load / Unload). The equipment communication instructions of the unloading stage of the loading and unloading equipment (Load / Unload) are monitored, and the information of the batch that needs to be unloaded and the status of the aging test board (aging board means that there is material on the board, and the batch information and test results are already available) are obtained from the monitoring server, and the correct response is given to the monitored loading and unloading equipment (Load / Unload) according to the protocol content and format. The loading and unloading equipment (Load / Unload) continues to the next step. If the communication protocol content response is wrong (part of the response content is missing or the format is abnormal), the loading and unloading equipment (Load / Unload) cannot continue to the next step, and prompts the abnormality and the cause of the abnormality. The production line operator handles the problem according to the abnormal prompt.

[0070] After the correct response, the loading and unloading equipment (Load / Unload) automatically unloads this aging test board. After this board is finished, the next test board is unloaded until the unloading of this batch of samples is completed.

[0071] In addition, it will be appreciated by those skilled in the art that various aspects of the present application may be illustrated and described by a number of patentable categories or situations, including any new and useful process, machine, product or combination of substances, or any new and useful improvements thereto. Accordingly, various aspects of the present application may be performed entirely by hardware, entirely by software (including firmware, resident software, microcode, etc.), or by a combination of hardware and software. The above hardware or software may all be referred to as "data blocks", "modules", "engines", "units", "components" or "systems". In addition, various aspects of the present application may be represented as a computer product located in one or more computer-readable media, which includes computer-readable program code.

[0072] Unless otherwise defined, all terms (including technical and scientific terms) used herein have the same meaning as commonly understood by those of ordinary skill in the art to which this application belongs. It should also be understood that terms such as those defined in common dictionaries should be interpreted as having a meaning consistent with their meaning in the context of the relevant technology, and should not be interpreted in an idealized or extremely formal sense, unless explicitly defined as such herein.

[0073] The above is an explanation of the present application and should not be considered as a limitation thereof. Although several exemplary embodiments of the present application are described, it will be readily understood by those skilled in the art that many modifications may be made to the exemplary embodiments without departing from the novel teachings and advantages of the present application. Therefore, all of these modifications are intended to be included within the scope of the present application as defined by the claims. It should be understood that the above is an explanation of the present application and should not be considered to be limited to the specific embodiments disclosed, and modifications to the disclosed embodiments and other embodiments are intended to be included within the scope of the appended claims. The present application is defined by the claims and their equivalents.

Claims

1. A memory chip aging test management method, characterized in that: include: Collect laboratory test status information; Based on the test status information, determining whether the test environment of the laboratory is normal; In response to the test environment of the laboratory being normal, building a monitoring server, the monitoring server being used for device client access, wherein the monitoring server has basic data recorded; and Connect the loading and unloading equipment and the testing equipment to the monitoring server, and after the test connection, perform an aging test on the storage chip; The test status information includes test status, test schedule, cleaning status, test instruments, laboratory temperature data, humidity data, operating status of laboratory temperature control air conditioners, and operating status of laboratory equipment; Wherein, based on the test status information, determining whether the test environment of the laboratory is normal includes: Encoding each data item in the test state information to obtain a plurality of test state information item encoding feature vectors; Performing contextual test state semantic association coding on the plurality of test state information item coding feature vectors to obtain a plurality of contextual test state information item coding feature vectors; Performing a semantic importance weighted analysis based on the overall feature distribution on each of the plurality of contextual test state information item encoding feature vectors to obtain a global semantic encoding feature of the test state information; and Based on the global semantic coding features of the test status information, determine whether there is an abnormality in the laboratory.

2. The memory chip aging test management method according to claim 1, characterized in that: Performing contextual test state semantic association encoding on the plurality of test state information item encoding feature vectors to obtain a plurality of contextual test state information item encoding feature vectors, including: The plurality of trial state information item encoding feature vectors are passed through a context encoder based on a converter module to obtain the plurality of context trial state information item encoding feature vectors.

3. The memory chip aging test management method according to claim 2, characterized in that: Performing a semantic importance weighted analysis based on the overall feature distribution on each of the plurality of contextual test state information item encoding feature vectors to obtain a global semantic encoding feature of the test state information, including: Calculating the semantic weight value of each contextual test state information item encoding feature vector in the plurality of contextual test state information item encoding feature vectors relative to the overall feature distribution of the plurality of contextual test state information item encoding feature vectors to obtain a sequence of semantic weight values; and Based on the sequence of semantic weight values, the plurality of contextual test state information item encoding feature vectors are fused to obtain a test state information global semantic encoding feature vector as the test state information global semantic encoding feature.

4. The memory chip aging test management method according to claim 3, characterized in that: Calculating the semantic weight value of each contextual test state information item encoding feature vector in the plurality of contextual test state information item encoding feature vectors relative to the overall feature distribution of the plurality of contextual test state information item encoding feature vectors to obtain a sequence of semantic weight values, comprising: Calculating the semantic weight value of each contextual test state information item encoding feature vector in the plurality of contextual test state information item encoding feature vectors relative to the overall feature distribution of the plurality of contextual test state information item encoding feature vectors using the following weight formula to obtain a sequence of the semantic weight values; Wherein, the weight formula is: in, Yes 1× The matrix of is the dimension of each contextual trial state information item encoding feature vector in the plurality of contextual trial state information item encoding feature vectors, Yes 1× The matrix of is the dimension of the encoding feature vector of each of the contextual trial state information items, is the Sigmoid function, is the first of the plurality of contextual trial state information item encoding feature vectors context trial state information item encoding feature vector, is the first of multiple contextual trial state information items in the encoding feature vector context trial state information item encoding feature vector, is a sequence of semantic weight values.

5. The memory chip aging test management method according to claim 4, characterized in that: Based on the global semantic coding features of the test status information, determine whether there is an abnormality in the laboratory, including: The global semantic encoding feature vector of the test status information is passed through a classifier to obtain a classification result, and the classification result is used to indicate whether there is an abnormality in the laboratory.

6. The memory chip aging test management method according to claim 5, characterized in that: It also includes a training step: used to train the context encoder based on the converter module and the classifier.

7. The memory chip aging test management method according to claim 6, characterized in that: The training step comprises: Collect training test status information from the laboratory; Encoding each data item in the training test state information to obtain a plurality of training test state information item encoding feature vectors; Passing the plurality of training trial state information item encoding feature vectors through the context encoder based on the converter module to obtain a plurality of training context trial state information item encoding feature vectors; Calculating the training semantic weight value of each training context test state information item encoding feature vector in the plurality of training context test state information item encoding feature vectors relative to the overall feature distribution of the plurality of training context test state information item encoding feature vectors to obtain a sequence of training semantic weight values; Based on the sequence of the training semantic weight values, the plurality of training context test state information item encoding feature vectors are merged to obtain a training test state information global semantic encoding feature vector; Performing information convergence optimization on the global semantic coding feature vector of the training test state information to obtain an optimized global semantic coding feature vector of the training test state information; Passing the optimized training test state information global semantic encoding feature vector through the classifier to obtain a classification loss function value; and The transformer module-based context encoder and the classifier are trained based on the classification loss function value.

8. A memory chip aging test management system, used to execute the memory chip aging test management method according to claim 1, characterized in that: include: Information collection module, used to collect laboratory test status information; An environment analysis module, used to determine whether the test environment of the laboratory is normal based on the test status information; A monitoring server building module, used for building a monitoring server in response to the test environment of the laboratory being normal, wherein the monitoring server is used for device client access, wherein the monitoring server has basic data recorded; and The aging test module is used to connect the loading and unloading equipment and the testing equipment to the monitoring server, and after the test connection, perform an aging test on the storage chip.

Images