Chip fault detection method and device
By obtaining the chip's data distribution information and the change characteristics of the fault data, combining feature extraction and similarity calculation, predicting chip failure information, the problems of fault detection accuracy and low efficiency in the existing technology are solved, and efficient and accurate chip failures are achieved Test.
Patent Information
- Application Number
- CN202411348513.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-09-26
- Publication Date
- 2025-01-10
AI Technical Summary
The existing chip fault detection methods rely on the accuracy and simulation accuracy of the convolutional network, resulting in low accuracy and efficiency of fault detection.
By obtaining the chip's data distribution information and the change characteristics of the fault data, combining feature extraction strategies and similarity calculations, the chip's fault information is predicted, and fault warning information is generated, the cost of fault detection is reduced and the efficiency and accuracy of fault detection is improved.
It realizes more efficient and accurate chip failure detection, avoids the tedious process of neural networks and simulation, and improves the timeliness and comprehensiveness of fault prediction.
Smart Images
Figure CN119293437A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the field of chip detection technology, and in particular to a chip fault detection method and device. Background Art
[0002] At present, with the continuous upgrading of chip hardware, the chip's anti-interference ability, computing efficiency, service life, and scope of application are constantly improving. However, due to the diversity of the external environment and the richness of operating conditions, the application frequency of chips is increasing, and long-term operation under high voltage and high current is accompanied by frequent opening and closing actions, which puts higher requirements on the reliability of chips. In order to avoid chip failures during actual application, which may lead to the failure of running tasks and delay the execution of normal tasks, it is often necessary to predict whether the chip will fail and detect the chip's operating data in real time to avoid chip failures and ensure the normal operation of the chip.
[0003] Existing fault detection methods use convolutional networks to predict chip anomalies or simulate monitoring data to identify chip fault points and fault information. However, the data simulation process is complex and the detection accuracy depends on the accuracy of the convolutional network and the simulation accuracy, resulting in low accuracy and efficiency of fault detection. Summary of the invention
[0004] The present invention provides a chip fault detection method and device to reduce the cost of fault detection and improve the efficiency and accuracy of fault detection.
[0005] In order to solve the above technical problems, an embodiment of the present invention provides a chip fault detection method, comprising:
[0006] Acquire data distribution information of the chip based on first operation detection information of the chip, and generate fault data distribution information of each detection type based on first fault report information of the chip;
[0007] Acquire the fault data change characteristics of each detection type based on the fault data distribution information;
[0008] The fault information of the chip is predicted based on the data distribution information, the fault data change characteristics and the fault identification strategy.
[0009] Furthermore, the step of acquiring data distribution information of the chip based on the first operation detection information of the chip includes:
[0010] splitting the first operation detection information into second operation detection data corresponding to the detection type based on the detection type, and matching the detection target based on the detection type;
[0011] The detection target data of each detection type is acquired based on the second operation detection data, and the data distribution information of the chip is acquired based on the detection target data.
[0012] Further, the generating of the fault data distribution information of each detection type based on the first fault report information of the chip includes:
[0013] Acquire first fault report information of the chip, and split the first fault report information into second fault report information corresponding to the fault type;
[0014] Obtain the fault type of the second fault report information and fault log information of each detection type;
[0015] Fault data before the fault occurs is acquired based on the fault log information of the detection type, and fault data distribution information is acquired based on the fault data.
[0016] Furthermore, the acquiring of the fault data change characteristics of each detection type based on the fault data distribution information includes:
[0017] Acquire the first data change feature and feature type of each detection type based on a preset database;
[0018] Acquire the second data change feature and feature type of each detection type based on the feature extraction strategy and the fault distribution information;
[0019] A first similarity between the first data change feature and the second data change feature is calculated based on the feature type, and a fault data change feature is acquired based on the first similarity and a preset similarity threshold.
[0020] Further, the predicting chip fault information based on the data distribution information, the fault data change characteristics and the fault identification strategy includes:
[0021] Perform feature extraction on the data distribution information based on a feature extraction strategy to obtain chip data change features and feature types of each detection type;
[0022] Calculate a second similarity between the chip and the fault feature based on the fault data change feature and the chip data change feature;
[0023] The fault frequency of the chip relative to the fault type is calculated, and the fault probability is calculated based on the fault frequency, the second similarity and the fault identification strategy to generate fault information, wherein the fault information includes the second similarity, the fault type and the corresponding fault probability.
[0024] Furthermore, after predicting the fault information of the chip based on the data distribution information, the fault data change characteristics and the fault identification strategy, it also includes:
[0025] A warning fault type and a target fault feature type are acquired based on the fault information, and a warning is issued based on the warning fault type and the target fault feature type.
[0026] Further, the acquiring of a warning fault type and a target fault characteristic type based on the fault information, and issuing a warning based on the warning fault type and the target fault characteristic type, includes:
[0027] Acquire a warning fault type based on a preset probability threshold, the warning fault type being a fault type whose fault probability is greater than the probability threshold;
[0028] Acquire a target fault feature type based on a preset second similarity threshold, wherein the target fault feature type is a fault feature of the early warning fault type whose second similarity of the fault feature to the chip is greater than a second similarity threshold;
[0029] Fault warning information is generated based on the warning fault type and the target fault feature type, and a warning is issued based on the fault warning information.
[0030] In a second aspect, the present invention provides a chip fault detection device, including: a data distribution recognition module, a feature extraction module and a prediction module;
[0031] The data distribution identification module is used to obtain data distribution information of the chip based on the first operation detection information of the chip, and generate fault data distribution information of each detection type based on the first fault report information of the chip;
[0032] The feature extraction module is used to obtain the fault data change characteristics of each detection type based on the fault data distribution information;
[0033] The prediction module is used to predict the fault information of the chip based on the data distribution information, the fault data change characteristics and the fault identification strategy.
[0034] In a third aspect, the present invention provides a terminal device, comprising a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, wherein the processor implements the chip fault detection method when executing the computer program.
[0035] In a fourth aspect, the present invention provides a computer-readable storage medium, the computer-readable storage medium including a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the chip fault detection method. BRIEF DESCRIPTION OF THE DRAWINGS
[0036] Figure 1A schematic diagram of a chip fault detection method provided in an embodiment of the present invention;
[0037] Figure 2 A schematic diagram of the structure of a chip fault detection device provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0038] The specific implementation of the present invention is further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present invention, but are not intended to limit the scope of the present invention.
[0039] The terms "first" and "second" and the like in the specification, claims and drawings of this application are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device comprising a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to these processes, methods, products or devices.
[0040] Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present application. The appearance of the phrase in various locations in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive with other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.
[0041] Example 1
[0042] See also Figure 1 , Figure 1 The flowchart of a chip fault detection method provided in an embodiment of the present invention is shown in FIG. The chip fault detection method provided in an embodiment of the present invention includes steps 101 to 103, which are as follows:
[0043] Step 101: acquiring data distribution information of a chip based on first operation detection information of the chip, and generating fault data distribution information of each detection type based on first fault report information of the chip;
[0044] In this embodiment, the step of acquiring data distribution information of the chip based on the first operation detection information of the chip includes:
[0045] splitting the first operation detection information into second operation detection data corresponding to the detection type based on the detection type, and matching the detection target based on the detection type;
[0046] Based on the second operation detection data, detection target data of each detection type is acquired, and based on the detection target data, data distribution information of the chip is acquired, wherein the detection target data is the operation detection data corresponding to the detection target.
[0047] In this embodiment, the first operation detection information of the chip is collected, and the operation detection information refers to various dynamic states and performance data obtained through monitoring and testing during the operation of the chip. The operation detection information includes but is not limited to basic state information, performance monitoring information and power consumption information, wherein the basic state information includes the chip operating temperature, voltage and temperature; the performance monitoring information includes clock frequency, processor load, memory usage, etc., and the power consumption information includes dynamic power consumption and static power consumption.
[0048] In this embodiment, the first operation detection information is split into second operation detection data corresponding to the detection type according to the detection type, and the detection type includes: power consumption, status, performance, etc.; for example, if the detection type is power consumption, the corresponding second operation detection information includes power consumption information.
[0049] In this embodiment, after the first operation detection information is split into the second operation detection information of each detection type, the detection target corresponding to each detection type is queried in the database; wherein the database stores the detection targets corresponding to different detection types, and the stored information is the information input into the database by the staff.
[0050] In this embodiment, for each detection type, second running detection data based on the detection type is obtained, the detection target data at each moment of the detection type is identified, and the detection target data at each moment is sorted and processed in chronological order to obtain initial chip data distribution information; the initial chip data distribution information is smoothed to obtain chip data distribution information of the detection type, wherein the detection target data is the running data corresponding to the detection target, and the smoothing method is a smoothing method based on the moving average method.
[0051] In this embodiment, by splitting the running detection information into various detection types and then identifying the chip data distribution information of each detection type respectively, the comprehensiveness of the analysis is improved. Finally, through smoothing processing, the random errors and noise in the data are reduced, making the data smoother, thereby better reflecting the real trends and characteristics of the data.
[0052] In this embodiment, the generating of the fault data distribution information of each detection type based on the first fault report information of the chip includes:
[0053] Acquire first fault report information of the chip, and split the first fault report information into second fault report information corresponding to the fault type;
[0054] Obtain the fault type of the second fault report information and fault log information of each detection type;
[0055] Fault data before the fault occurs is acquired based on the fault log information of the detection type, and fault data distribution information is acquired based on the fault data.
[0056] In this embodiment, the first fault report information is the historical fault report information of the chip. The historical fault report information of the chip refers to detailed information about the occurrence of the fault recorded during the use and testing of the chip, including the fault type, fault occurrence time, fault point, operating data in the period before the fault occurred, fault detection type and its corresponding fault log information, fault cause, fault handling and number of faults, etc.
[0057] In this embodiment, based on the historical fault report information of the chip, the historical fault report information is divided into second fault report information corresponding to the fault type according to the fault type, and fault log information corresponding to each detection type is obtained according to the second fault report information. The fault data at each moment before the fault occurs is filtered according to the fault log information, and the fault data at each moment is distributed and arranged according to the time sequence of each moment to obtain the fault data distribution information of the detection type.
[0058] In this embodiment, by screening the sub-fault data at each time before the fault occurs, similarity identification can be performed with the chip data distribution information of each detection type before the fault occurs, so as to determine whether the fault type may exist.
[0059] In this embodiment, after the fault report information is split into fault logs of each detection type, the sub-fault data at each moment before the fault occurs is screened, thereby improving the accuracy of the acquired fault data for subsequent identification of possible fault types.
[0060] Step 102: acquiring fault data variation characteristics of each detection type based on the fault data distribution information;
[0061] In this embodiment, the acquiring of the fault data change characteristics of each detection type based on the fault data distribution information includes:
[0062] Acquire the first data change feature and feature type of each detection type based on a preset database;
[0063] Acquire the second data change feature and feature type of each detection type based on the feature extraction strategy and the fault distribution information;
[0064] A first similarity between the first data change feature and the second data change feature is calculated based on the feature type, and a fault data change feature is acquired based on the first similarity and a preset first similarity threshold.
[0065] In this embodiment, the first data change feature of each detection type and the feature type of each first data change feature are queried in the database, and the first data change feature is a regular data change feature of each detection type under normal operating conditions.
[0066] In this embodiment, the feature extraction strategy is a distribution information feature extraction strategy, specifically a histogram feature extraction method, which performs feature extraction on the fault distribution information to obtain the second data change feature and feature type of the detection type.
[0067] In this embodiment, a first similarity between the first data change feature and the second data change feature is calculated for each feature type, and a fault data change feature is acquired based on the first similarity and a preset first similarity threshold.
[0068] In this embodiment, the cosine similarity algorithm is used to calculate the first similarity between the first data change feature of each feature type and the second data change feature of each feature type, and the second data change feature corresponding to the first similarity lower than the preset first similarity threshold is screened as the fault data change feature, and the fault type corresponding to the fault data change feature is set as the fault feature type.
[0069] In this embodiment, the detection types including the fault data change characteristics are used as the abnormality detection types corresponding to the fault types.
[0070] In this embodiment, the data change characteristics of each detection type are extracted by feature extraction, thereby improving the extraction accuracy of the data change characteristics. Then, the similarity between two features is identified by a similarity algorithm, thereby identifying the fault data change characteristics in each detection type. Compared with traditional simulation recognition, the recognition efficiency is improved while ensuring the recognition accuracy.
[0071] Step 103: Predict chip fault information based on the data distribution information, the fault data change characteristics and the fault identification strategy.
[0072] In this embodiment, the prediction of chip fault information based on the data distribution information, the fault data change characteristics and the fault identification strategy includes:
[0073] Perform feature extraction on the data distribution information based on a feature extraction strategy to obtain chip data change features and feature types of each detection type;
[0074] Calculate a second similarity between the chip and the fault feature based on the fault data change feature and the chip data change feature;
[0075] The fault frequency of the chip relative to the fault type is calculated, and the fault probability is calculated based on the fault frequency, the second similarity and the fault identification strategy to generate fault information, wherein the fault information includes the second similarity, the fault type and the corresponding fault probability.
[0076] In this embodiment, feature extraction is performed on the data distribution information based on a distribution feature extraction strategy to obtain chip data change features and feature types of each detection type. The distribution feature extraction strategy is a histogram feature extraction method.
[0077] In this embodiment, for each fault type, the second similarity between the chip and the fault type is calculated based on the fault data variation characteristics of each fault feature type of the fault type and the chip data variation characteristics of each fault feature type.
[0078] In this embodiment, the similarity between the fault data variation feature and the chip data variation feature is calculated by a cosine similarity algorithm, and this is used as the second similarity between the chip and the fault type.
[0079] In this embodiment, based on the historical fault report information of the chip, the number of faults of the chip in the fault type is obtained, and the fault frequency of the chip in the fault type is calculated; and based on the fault frequency of the fault type and the second similarity between the chip and the fault type, the fault probability of each fault type corresponding to the chip is calculated through a fault probability algorithm. The fault frequency is the ratio value obtained by dividing the number of faults of each fault type in the historical fault report information by the number of all faults.
[0080] In this embodiment, the fault probability algorithm is to preset a ratio value between the fault frequency and the fault probability, and then obtain the initial fault probability of each fault type by multiplying the fault frequency of each fault type by the ratio value. Then, the terminal multiplies the second similarity between the chip and the fault type by the initial fault probability to obtain the failure probability of each fault type corresponding to the chip.
[0081] In this embodiment, when identifying the fault probability, the fault probability of each fault type corresponding to the chip is analyzed simultaneously through the fault frequency and similarity, thereby improving the practicality and accuracy of the analysis.
[0082] In this embodiment, after predicting the fault information of the chip based on the data distribution information, the fault data change characteristics and the fault identification strategy, the method further includes:
[0083] A warning fault type and a target fault feature type are acquired based on the fault information, and a warning is issued based on the warning fault type and the target fault feature type.
[0084] In this embodiment, based on the fault types corresponding to the fault probabilities greater than the preset probability threshold and the current data distribution information of each abnormal detection type corresponding to each fault type, the fault warning information of the chip is generated for warning.
[0085] In this embodiment, the acquiring of the warning fault type and the target fault feature type based on the fault information, and issuing a warning based on the warning fault type and the target fault feature type, includes:
[0086] Acquire a warning fault type based on a preset probability threshold, the warning fault type being a fault type whose fault probability is greater than the probability threshold;
[0087] Acquire a target fault feature type based on a preset second similarity threshold, wherein the target fault feature type is a fault feature of the early warning fault type whose second similarity of the fault feature to the chip is greater than a second similarity threshold;
[0088] Fault warning information is generated based on the warning fault type and the target fault feature type, and a warning is issued based on the fault warning information.
[0089] In this embodiment, the terminal uses the fault type corresponding to the fault probability greater than the preset probability threshold as the warning fault type, and based on the second similarity between the warning fault type and each fault feature type between the chips, screens the target fault feature type greater than the second similarity threshold.
[0090] In this embodiment, fault warning information is generated based on the warning fault type and the target fault feature type, and a warning is performed based on the fault warning information, including generating a failure probability cause of the chip in the warning fault type based on the chip data change characteristics of each target fault feature type, and importing the warning fault type, target fault feature type and failure probability cause into a warning template to generate fault warning information for warning.
[0091] In this embodiment, based on the chip data change characteristics of the chip in each target fault feature type, the fault probability cause of the chip in the warning fault type is generated. The fault probability cause is the chip data change information corresponding to each chip data change characteristic, for example, the chip data fluctuation range corresponding to the chip data fluctuation range characteristic, the chip data fluctuation range trend corresponding to the chip data fluctuation range trend characteristic, etc.
[0092] In this embodiment, according to the order of the failure probability of the chip in each warning fault type from large to small, the failure probability of the chip in each warning fault type and the cause of the failure probability of the chip in each warning fault type are respectively filled into the fault warning template to obtain the fault warning information of the chip. The fault warning template is preset in the terminal and includes three fields: fault type, fault probability, and fault cause. The terminal fills the failure probability of the chip in each warning fault type and the cause of the failure probability of the chip in each warning fault type into the fault warning template according to each field to obtain the fault warning information of the chip.
[0093] In this embodiment, after screening the warning fault types with higher failure probabilities, the faults are sorted and filled in to obtain the chip fault warning information, thereby improving the comprehensiveness and accuracy of obtaining the chip fault warning information.
[0094] In this embodiment, by identifying the fault data distribution information of each detection type before each fault type in the fault report information of different chips is generated, the fault data change characteristics of each abnormal detection type corresponding to each fault type are analyzed, and then based on the chip data distribution information of each detection type in the operation detection information of the chip collected within a preset time period, the failure probability of each fault type corresponding to the chip is identified, thereby generating the chip fault warning information. This scheme does not need to go through the cumbersome and inefficient process of neural network or simulation, but instead comprehensively identifies the feature similarity of each detection type and each fault type, which not only ensures the accuracy of the predicted fault warning information, but also has a simple algorithm and better recognition efficiency, thereby improving the prediction timeliness of the predicted fault warning information, thereby comprehensively improving the efficiency of chip fault prediction.
[0095] Please refer to Figure 2 , Figure 2 A schematic diagram of the structure of a chip fault detection device provided by an embodiment of the present invention includes: a data distribution identification module 201, a feature extraction module 202 and a prediction module 203;
[0096] The data distribution identification module 201 is used to obtain data distribution information of the chip based on the first operation detection information of the chip, and generate fault data distribution information of each detection type based on the first fault report information of the chip;
[0097] The feature extraction module 202 is used to obtain the fault data change characteristics of each detection type based on the fault data distribution information;
[0098] The prediction module 203 is used to predict the fault information of the chip based on the data distribution information, the fault data change characteristics and the fault identification strategy.
[0099] In this embodiment, the feature extraction module includes:
[0100] splitting the first operation detection information into second operation detection data corresponding to the detection type based on the detection type, and matching the detection target based on the detection type;
[0101] Based on the second operation detection data, detection target data of each detection type is acquired, and based on the detection target data, data distribution information of the chip is acquired, wherein the detection target data is the operation detection data corresponding to the detection target.
[0102] In this embodiment, the data distribution identification module includes:
[0103] Acquire first fault report information of the chip, and split the first fault report information into second fault report information corresponding to the fault type;
[0104] Obtain the fault type of the second fault report information and fault log information of each detection type;
[0105] Fault data before the fault occurs is acquired based on the fault log information of the detection type, and fault data distribution information is acquired based on the fault data.
[0106] In this embodiment, the feature extraction module includes:
[0107] Acquire the first data change feature and feature type of each detection type based on a preset database;
[0108] Acquire the second data change feature and feature type of each detection type based on the feature extraction strategy and the fault distribution information;
[0109] A first similarity between the first data change feature and the second data change feature is calculated based on the feature type, and a fault data change feature is acquired based on the first similarity and a preset first similarity threshold.
[0110] In this embodiment, the prediction module includes:
[0111] Perform feature extraction on the data distribution information based on a feature extraction strategy to obtain chip data change features and feature types of each detection type;
[0112] Calculate a second similarity between the chip and the fault feature based on the fault data change feature and the chip data change feature;
[0113] The fault frequency of the chip relative to the fault type is calculated, and the fault probability is calculated based on the fault frequency, the second similarity and the fault identification strategy to generate fault information, wherein the fault information includes the second similarity, the fault type and the corresponding fault probability.
[0114] In this embodiment, an early warning module is also included, and the early warning module includes:
[0115] A warning fault type and a target fault feature type are acquired based on the fault information, and a warning is issued based on the warning fault type and the target fault feature type.
[0116] In this embodiment, the early warning module includes:
[0117] Acquire a warning fault type based on a preset probability threshold, the warning fault type being a fault type whose fault probability is greater than the probability threshold;
[0118] Acquire a target fault feature type based on a preset second similarity threshold, wherein the target fault feature type is a fault feature of the early warning fault type whose second similarity of the fault feature to the chip is greater than a second similarity threshold;
[0119] Fault warning information is generated based on the warning fault type and the target fault feature type, and a warning is issued based on the fault warning information.
[0120] In this embodiment, by identifying the fault data distribution information of each detection type before each fault type in the historical fault report information of different chips is generated, the fault data change characteristics of each abnormal detection type corresponding to each fault type are analyzed, and then based on the chip data distribution information of each detection type in the operation detection information of the chip collected within a preset time, the fault probability of each fault type corresponding to the chip is identified, thereby generating the fault warning information of the chip, so that the scheme does not need to go through the cumbersome and inefficient process of neural network or simulation, but through the comprehensive identification of the feature similarity of each detection type and each fault type, not only the accuracy of the predicted fault warning information is ensured, but also the algorithm is simple, the recognition efficiency is better, and the prediction timeliness of the predicted fault warning information is improved. Thereby, the efficiency of chip fault prediction is comprehensively improved.
[0121] The present invention also provides a terminal device, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor, and the processor implements the chip fault detection method when executing the computer program.
[0122] The present invention also provides a computer-readable storage medium, which includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the chip fault detection method.
[0123] In an embodiment of the present invention, a multi-device access platform processing device is also provided, including a processor, a memory, and a computer program stored in the memory and configured to be executed by the processor. When the processor executes the computer program, the above-mentioned multi-device access platform processing method is implemented.
[0124] In an embodiment of the present invention, a computer-readable storage medium is also provided, the computer-readable storage medium including a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute the above-mentioned multi-device access platform processing method.
[0125] Exemplarily, the computer program may be divided into one or more modules, one or more modules are stored in a memory and executed by a processor to implement the present invention. One or more modules may be a series of computer program instruction segments capable of implementing specific functions, and the instruction segments are used to describe the execution process of the computer program in the multi-device access platform processing device.
[0126] The multi-device access platform processing device may be a computing device such as a desktop computer, a notebook, a PDA, and a cloud server. The multi-device access platform processing device may include, but is not limited to, a processor, a memory, and a display. Those skilled in the art will appreciate that the above components are merely examples of the multi-device access platform processing device and do not constitute a limitation on the multi-device access platform processing device. It may include more or fewer components than the components, or a combination of certain components, or different components. For example, the multi-device access platform processing device may also include input and output devices, network access devices, buses, and the like.
[0127] The processor may be a central processing unit (CPU), other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor, etc. The processor is the control center of the multi-device access platform processing device, and uses various interfaces and lines to connect various parts of the entire multi-device access platform processing device.
[0128] The memory can be used to store computer programs and / or modules. The processor realizes various functions of the multi-device access platform processing device by running or executing the computer programs and / or modules stored in the memory and calling the data stored in the memory. The memory can mainly include a program storage area and a data storage area, wherein the program storage area can store an operating system, an application required for at least one function (such as a sound playback function, a text conversion function, etc.), etc.; the data storage area can store data created according to the use of the mobile phone (such as audio data, text message data, etc.), etc. In addition, the memory can include a high-speed random access memory, and can also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (Smart Med ia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.
[0129] Among them, if the module based on the processing of the multi-device access platform is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the present invention implements all or part of the processes in the above-mentioned embodiment method, and can also be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a computer-readable storage medium, and the computer program can implement the steps of the above-mentioned various method embodiments when executed by the processor. Among them, the computer program includes computer program code, and the computer program code can be in source code form, object code form, executable file or some intermediate form. The computer-readable medium may include: any entity or device capable of carrying computer program code, recording medium, U disk, mobile hard disk, disk, optical disk, computer memory, read-only memory (ROM, Read-Only Memory), random access memory (RAM, RandomAccess Memory), electric carrier signal, telecommunication signal and software distribution medium, etc. Ordinary technicians in this field can understand and implement it without paying creative labor.
[0130] The specific embodiments described above further illustrate the purpose, technical solutions and beneficial effects of the present invention. It should be understood that the above description is only a specific embodiment of the present invention and is not intended to limit the scope of protection of the present invention. It is particularly pointed out that for those skilled in the art, any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the scope of protection of the present invention.
Claims
1. A chip fault detection method, characterized in that: include: Acquire data distribution information of the chip based on first operation detection information of the chip, and generate fault data distribution information of each detection type based on first fault report information of the chip; Acquire the fault data change characteristics of each detection type based on the fault data distribution information; The fault information of the chip is predicted based on the data distribution information, the fault data change characteristics and the fault identification strategy.
2. A chip fault detection method as claimed in claim 1, characterized in that: The step of acquiring data distribution information of the chip based on first operation detection information of the chip includes: splitting the first operation detection information into second operation detection data corresponding to the detection type based on the detection type, and matching the detection target based on the detection type; Based on the second operation detection data, detection target data of each detection type is acquired, and based on the detection target data, data distribution information of the chip is acquired, wherein the detection target data is the operation detection data corresponding to the detection target.
3. A chip fault detection method as claimed in claim 2, characterized in that: The generating of the fault data distribution information of each detection type based on the first fault report information of the chip includes: Acquire first fault report information of the chip, and split the first fault report information into second fault report information corresponding to the fault type; Obtain the fault type of the second fault report information and fault log information of each detection type; Fault data before the fault occurs is acquired based on the fault log information of the detection type, and fault data distribution information is acquired based on the fault data.
4. A chip fault detection method as claimed in claim 3, characterized in that: The acquiring the fault data change characteristics of each detection type based on the fault data distribution information includes: Acquire the first data change feature and feature type of each detection type based on a preset database; Acquire the second data change feature and feature type of each detection type based on the feature extraction strategy and the fault distribution information; A first similarity between the first data change feature and the second data change feature is calculated based on the feature type, and a fault data change feature is acquired based on the first similarity and a preset first similarity threshold.
5. A chip fault detection method as claimed in claim 4, characterized in that: The method of predicting chip fault information based on the data distribution information, the fault data change characteristics and the fault identification strategy includes: Perform feature extraction on the data distribution information based on a feature extraction strategy to obtain chip data change features and feature types of each detection type; Calculate a second similarity between the chip and the fault feature based on the fault data change feature and the chip data change feature; The fault frequency of the chip relative to the fault type is calculated, and the fault probability is calculated based on the fault frequency, the second similarity and the fault identification strategy to generate fault information, wherein the fault information includes the second similarity, the fault type and the corresponding fault probability.
6. A chip fault detection method as claimed in claim 5, characterized in that: After predicting the fault information of the chip based on the data distribution information, the fault data change characteristics and the fault identification strategy, the method further includes: A warning fault type and a target fault feature type are acquired based on the fault information, and a warning is issued based on the warning fault type and the target fault feature type.
7. A chip fault detection method as claimed in claim 6, characterized in that: The acquiring the warning fault type and the target fault feature type based on the fault information, and issuing a warning based on the warning fault type and the target fault feature type, includes: Acquire a warning fault type based on a preset probability threshold, the warning fault type being a fault type whose fault probability is greater than the probability threshold; Acquire a target fault feature type based on a preset second similarity threshold, wherein the target fault feature type is a fault feature of the early warning fault type whose second similarity of the fault feature to the chip is greater than a second similarity threshold; Fault warning information is generated based on the warning fault type and the target fault feature type, and a warning is issued based on the fault warning information.
8. A chip fault detection device, characterized in that: include: Data distribution identification module, feature extraction module and prediction module; The data distribution identification module is used to obtain data distribution information of the chip based on the first operation detection information of the chip, and generate fault data distribution information of each detection type based on the first fault report information of the chip; The feature extraction module is used to obtain the fault data change characteristics of each detection type based on the fault data distribution information; The prediction module is used to predict the fault information of the chip based on the data distribution information, the fault data change characteristics and the fault identification strategy.
9. A terminal device, characterized in that: The invention comprises a processor, a memory and a computer program stored in the memory and configured to be executed by the processor, and when the processor executes the computer program, a chip fault detection method as claimed in any one of claims 1 to 7 is implemented.
10. A computer-readable storage medium, characterized in that: The computer-readable storage medium includes a stored computer program, wherein when the computer program is running, the device where the computer-readable storage medium is located is controlled to execute a chip fault detection method as described in any one of claims 1 to 7.