Performance management system and method based on artificial intelligence chip

Abstract

The invention discloses a performance management system and method based on an artificial intelligence chip, and relates to the technical field of chip performance management, and the method comprises the steps: collecting multi-dimensional performance data in real time through a built-in monitoring unit of the artificial intelligence chip; preprocessing the collected multi-dimensional performance data, and converting the preprocessed multi-dimensional performance data into a structured performance data set; constructing a fault prediction model by using the performance data set to obtain a fault prediction value; dividing the fault into different early warning levels according to the fault prediction value; based on different early warning levels, an automatic optimization feedback mechanism is started, loads are dynamically adjusted, and resource allocation is optimized; through real-time data monitoring, accurate fault prediction and a dynamic response mechanism, the reliability and performance of the system are remarkably improved, the continuity of key services is ensured, the risk of fault occurrence is reduced, and powerful support is provided for safe operation of equipment.

Description

Technical Field

[0001] The present invention relates to the technical field of chip performance management, and in particular to a performance management system and method based on an artificial intelligence chip. Background Art

[0002] With the rapid development of smart devices and Internet of Things technologies, fault prediction and performance management based on artificial intelligence have gradually become a research hotspot. Traditional fault monitoring methods mostly use static data analysis, which is not only difficult to capture the operating status of the equipment in real time, but also easily leads to delayed response to sudden faults. In addition, existing technologies have limitations in processing multi-dimensional performance data, and often cannot effectively integrate key indicators such as temperature, power consumption and load, which affects the accuracy and reliability of fault prediction. Summary of the invention

[0003] In view of the above existing problems, the present invention is proposed.

[0004] Therefore, the present invention provides a performance management system and method based on artificial intelligence chips to solve the problem that existing fault prediction methods lack real-time performance and dynamic response capabilities.

[0005] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0006] In a first aspect, an embodiment of the present invention provides a performance management method based on an artificial intelligence chip, which includes collecting multi-dimensional performance data in real time through a monitoring unit built into the artificial intelligence chip;

[0007] Preprocessing the collected multidimensional performance data, and converting the preprocessed multidimensional performance data into a structured performance data set;

[0008] Using the performance data set, a fault prediction model is constructed to obtain the fault prediction value;

[0009] Classify faults into different warning levels according to the fault prediction value;

[0010] Based on different warning levels, the automatic optimization feedback mechanism is activated to dynamically adjust the load and optimize resource allocation.

[0011] As a preferred solution of the performance management method based on artificial intelligence chip described in the present invention, wherein: the monitoring unit includes a temperature sensor, a current sensor and a load sensor;

[0012] The multi-dimensional performance data includes temperature, power consumption and load.

[0013] As a preferred solution of the performance management method based on artificial intelligence chip described in the present invention, the collected multidimensional performance data is preprocessed, and the preprocessed multidimensional performance data is converted into a structured performance data set, including the following steps:

[0014] The collected multi-dimensional performance data is cleaned, noise and outliers are removed, and then standardized;

[0015] Determine the format of the structured data set and add a timestamp to each piece of multidimensional performance data;

[0016] Each multi-dimensional performance data after adding a timestamp is integrated to generate a performance data set.

[0017] As a preferred solution of the performance management method based on artificial intelligence chip described in the present invention, wherein: using the performance data set to build a fault prediction model to obtain a fault prediction value, including the following steps:

[0018] According to the performance data set, set the temperature threshold and power consumption threshold of the AI ​​chip under normal operating conditions;

[0019] Select the random forest model, which is an ensemble learning method consisting of multiple decision trees;

[0020] extracting a ratio of temperature to a temperature threshold and a ratio of power consumption to a power consumption threshold from the performance data set;

[0021] The calculated ratio is used to train the random forest model to obtain a trained random forest model, predict the possibility of a fault, and obtain a fault prediction value, which is expressed as:

[0022]

[0023] Where F is the fault prediction value, t is the current time point, τ is the time variable, erf is the error function, λ is the time decay rate, σ is the standard deviation, T is the temperature, L is the load, W is the power consumption, α is the temperature adjustment coefficient, T c is the normal threshold of temperature, β is the load adjustment coefficient, γ is the constant term, W n is the normal threshold for power consumption.

[0024] As a preferred solution of the performance management method based on artificial intelligence chip described in the present invention, wherein: based on the fault prediction value, the fault warning threshold is calculated, which is expressed as:

[0025] T1 = μ-σ';

[0026] T2=μ+k·σ';

[0027] Wherein, T1 is the first fault warning threshold, T2 is the second fault warning threshold, μ is the mean of the fault prediction value, k is the safety factor, and σ' is the standard deviation of the fault prediction value F;

[0028] When the fault prediction value is less than or equal to the first fault warning threshold, it is judged as the first warning, indicating that the performance data is within a safe range and users can use it with confidence;

[0029] When the fault prediction value is between the first fault warning threshold and the second fault warning threshold, it is judged as the second warning, indicating that the performance data is slightly abnormal. It is recommended to conduct real-time monitoring and prepare to take measures;

[0030] When the fault prediction value is greater than the second fault warning threshold, it is judged as the third warning, indicating that the performance data is seriously abnormal and facing a fault. Check and take repair measures immediately.

[0031] As a preferred solution of the performance management method based on artificial intelligence chip described in the present invention, wherein: based on different warning levels, starting the automatic optimization feedback mechanism includes the following steps:

[0032] When in safe range, maintain current load and resource configuration;

[0033] When there is a slight abnormality, the monitoring frequency of temperature, load and power consumption is increased, and the system load is reduced according to historical data and current load conditions. The operator is informed of the current warning status, recommended measures and risks through the user interface;

[0034] When performance data is seriously abnormal, the preset emergency response mechanism is automatically triggered to immediately reduce the load, dynamically adjust the allocation of computing resources according to the current state of the system, start the troubleshooting program, check the system performance data, identify potential causes of failures, automatically collect fault logs and prepare related data.

[0035] As a preferred solution of the performance management method based on artificial intelligence chip described in the present invention, dynamic adjustment of load and optimization of resource allocation refers to using machine learning algorithms to analyze historical data, identify load patterns and resource requirements, optimize resource allocation strategies, adaptively adjust load and resource configuration strategies according to real-time performance data and warning levels, regularly evaluate the effectiveness of the automatic optimization mechanism, make continuous improvements based on system performance and fault occurrence, collect feedback from operators on the automatic optimization feedback mechanism, feed back new data and feedback information into the fault prediction model, and conduct continuous learning and optimization.

[0036] In a second aspect, the present invention provides a performance management system based on an artificial intelligence chip, including a data collection module responsible for collecting multi-dimensional performance data in real time through a monitoring unit built into the artificial intelligence chip;

[0037] The data preliminary processing module is responsible for preprocessing the collected multi-dimensional performance data and converting the preprocessed multi-dimensional performance data into a structured performance data set;

[0038] The fault prediction model module is responsible for using the performance data set to build a fault prediction model and obtain the fault prediction value;

[0039] The warning level classification module is responsible for classifying faults into different warning levels according to the fault prediction value;

[0040] The automatic optimization feedback mechanism module is responsible for starting the automatic optimization feedback mechanism based on different warning levels, dynamically adjusting the load and optimizing resource allocation.

[0041] In a third aspect, an embodiment of the present invention provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program is executed by the processor, any step of the performance management method based on an artificial intelligence chip as described in the first aspect of the present invention is implemented.

[0042] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, it implements any step of the performance management method based on an artificial intelligence chip as described in the first aspect of the present invention.

[0043] The beneficial effects of the present invention are as follows: by collecting multi-dimensional performance data in real time through the built-in monitoring unit of the artificial intelligence chip, the invention effectively solves the problem of insufficient real-time monitoring and dynamic response capabilities of existing fault prediction technologies. The monitoring unit has high-frequency sampling capabilities and can collect key parameters such as temperature, current and load multiple times per second, thereby ensuring the real-time and accuracy of the data. On this basis, the fault prediction model constructed using the generated structured data set comprehensively considers the nonlinear effects and time decay effects of multi-dimensional performance indicators, enabling the system to more comprehensively assess the risk of failure. This model can not only identify potential faults in a timely manner, but also start a dynamic optimization feedback mechanism according to different warning levels. Through intelligent adjustment of load and resources, the system can increase the monitoring frequency in the event of minor anomalies, and automatically reduce the load or shut down non-critical tasks in the event of serious anomalies, thereby ensuring the stable operation of the system. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0045] Figure 1 This is a flow chart of the performance management method based on artificial intelligence chip in Example 1.

[0046] Figure 2 This is a diagram showing the classification of fault levels in Example 1. DETAILED DESCRIPTION

[0047] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the accompanying drawings.

[0048] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0049] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0050] Example 1, reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides a performance management method based on an artificial intelligence chip, comprising the following steps:

[0051] S1. Collect multi-dimensional performance data in real time through the monitoring unit built into the artificial intelligence chip.

[0052] The monitoring unit includes a temperature sensor, a current sensor and a load sensor. The sampling frequency of the monitoring unit is set. Usually, data is collected multiple times per second (for example, 10 times per second) to improve the real-time performance of the data.

[0053] Multi-dimensional performance data includes temperature, power consumption and load.

[0054] It should be noted that by setting up temperature, current and load sensors and collecting data at a frequency of multiple times per second, high-frequency monitoring of the equipment status is achieved. This process ensures the real-time nature of the data and can reflect the operating status of the system in a timely manner, thereby providing a reliable raw data basis for subsequent fault prediction. This high-frequency sampling capability enables the system to quickly capture changes, thereby reducing the risk of failure.

[0055] S2. Preprocess the collected multidimensional performance data, and convert the preprocessed multidimensional performance data into a structured performance data set.

[0056] The collected multi-dimensional performance data is cleaned to remove noise and outliers, and then standardized to ensure that each indicator has the same dimension for easy comparison;

[0057] Determine the format of the structured data set, usually in the form of JSON, CSV or database table, and add a timestamp to each multi-dimensional performance data to record the exact time of data collection. The timestamp is usually in ISO8601 format (such as 2024-09-13T12:00:00Z) to ensure global standardization;

[0058] Each multi-dimensional performance data after adding a timestamp is integrated to generate a performance data set.

[0059] It should be noted that the accuracy and consistency of the data are ensured through data cleaning, noise removal and outlier removal. Standardization makes the indicators comparable and facilitates subsequent analysis. The generated structured performance data set can be formatted in a unified format (such as JSON, CSV), which enhances the accessibility and compatibility of the data. This process greatly improves the data quality and ensures the reliability of subsequent model analysis.

[0060] S3. Use the performance data set to build a fault prediction model and obtain a fault prediction value.

[0061] Based on the performance data set, set the temperature threshold and power consumption threshold of the AI ​​chip under normal operating conditions. If the temperature exceeds the temperature threshold, it may mean overheating risk. If the power consumption threshold is exceeded, it may cause overload or other power-related failures.

[0062] Select the random forest model, which is an ensemble learning method consisting of multiple decision trees;

[0063] extracting a ratio of temperature to a temperature threshold and a ratio of power consumption to a power consumption threshold from the performance data set;

[0064] The calculated ratio is used to train the random forest model. During the training process, the data needs to be divided into a training set and a test set to evaluate the performance of the model. The trained random forest model, i.e., the fault prediction model, is obtained to predict the possibility of a fault and obtain the fault prediction value, which is expressed as:

[0065]

[0066] Among them, F is the fault prediction value, a model for comprehensively evaluating the system fault risk, t is the current time point, which indicates the length of time from the start of the model to the current moment, and is used to calculate the comprehensive impact of the performance data during this period on the fault risk, τ is the time variable, which indicates the time point used in the integration process, ranging from 0 to t, and is used to evaluate the impact of temperature, load and power consumption on the fault risk during the time period, erf represents the error function, and the error function expression is is the time decay factor, which reflects the cumulative effect of fault risk over time. λ is the time decay rate. The larger the value, the more obvious the cumulative effect of fault risk. σ is the standard deviation, which reflects the influence of time fluctuation and can capture the uncertainty of fault risk within a specific time. T is the temperature, which reflects the thermal state of the system. Too high temperature may cause fault. L is the load, which represents the current workload of the system. Too high load may cause performance degradation or fault. W is the power consumption, which represents the energy consumption of the system. Abnormal changes in power consumption are usually related to fault risk. α is the temperature adjustment coefficient, which controls the nonlinear influence of temperature on fault prediction. The larger the value, the more significant the influence of temperature on fault prediction. T c is the normal threshold of temperature, which is used to normalize the temperature data so that it is in the range of 0 to 1. β is the load adjustment coefficient, which reflects the influence of load on fault prediction. The larger the value, the more significant the influence of load on fault prediction. γ is a constant term, which ensures that the denominator is not zero, reflecting the basic influence of power consumption on fault prediction. n is the normal threshold of power consumption, used to normalize power consumption, It is the root mean square of power consumption, reflecting the fluctuation and stability of power consumption.

[0067] The derivation process includes the following steps,

[0068] The fault prediction model expression is designed according to each data in the performance data set, aiming to comprehensively evaluate the fault risk through performance data (temperature, load, power consumption, etc.) and form an initial fault prediction model, which is expressed as,

[0069] F 0 =T+L+W;

[0070] Among them, F 0 is the initial failure prediction value, which represents the preliminary assessment of the system failure risk;

[0071] Considering the dimensional influence of different indicators, the nonlinear influence of performance indicators on fault risk, the load term is introduced into the exponential function to enhance its influence on fault prediction, the stability influence of time factors and power consumption, normalization, power function, exponential function, integration and root mean square are introduced into the initial fault prediction model to form the introduced fault prediction model, which is expressed as,

[0072]

[0073] Among them, F 1 For the fault prediction value after introduction, the impact of multiple factors on the fault risk is considered comprehensively;

[0074] In the introduced fault prediction model, the time decay factor is added to reflect the cumulative effect of fault risk and the error function is added to take into account uncertainty to form the final fault prediction model;

[0075] It should be noted that the basis for designing the above fault prediction model formula is to comprehensively consider the impact of multiple factors on the risk of system failure, and to achieve accurate prediction of faults through mathematical models. First of all, temperature, load and power consumption are key parameters that affect equipment performance and reliability. By normalizing the temperature data, the dimension effect can be eliminated, making different indicators comparable. At the same time, the introduction of temperature adjustment coefficients and load adjustment coefficients enables the model to flexibly reflect the relative importance of each factor in fault prediction.

[0076] Secondly, the historical data is cumulatively evaluated in an integral form to capture the long-term impact of temperature and load on the failure risk over a period of time. In addition, the root mean square of power consumption is used to measure the volatility of power consumption to ensure that the model can reflect the fundamental role of power consumption stability in fault prediction.

[0077] The introduction of time decay factor and error function in the model takes into account the cumulative effect and uncertainty of fault risk over time, making the prediction more dynamic. This comprehensive model design not only improves the accuracy and adaptability of fault prediction, but also provides a scientific basis for system maintenance and fault prevention, ensuring the safe and stable operation of equipment.

[0078] It should be noted that by establishing a fault prediction model that comprehensively considers temperature, load, and power consumption, the system can dynamically evaluate the risk of faults. The introduction of nonlinear functions and time decay factors enhances the flexibility and prediction ability of the model. This model design enables the system to issue early warnings before potential faults occur, significantly improving the accuracy of fault identification.

[0079] S4. Classify faults into different warning levels according to the fault prediction value.

[0080] Based on the fault prediction value, the fault warning threshold is calculated and expressed as,

[0081] T1 = μ-σ';

[0082] T2=μ+k·σ';

[0083] Wherein, T1 is the first fault warning threshold, T2 is the second fault warning threshold, μ is the mean of the fault prediction value, which indicates the average fault prediction level in the historical data, k is the safety factor, which can be set according to the requirements, and a positive number greater than 0 (such as 1.5 or 2) is usually selected to improve the safety of the threshold, and σ' is the standard deviation of the fault prediction value F, which reflects the fluctuation degree of the prediction value;

[0084] It should be noted that the basis for designing the above fault warning threshold formula is to use statistical principles to reflect the distribution characteristics of fault prediction data through the mean and standard deviation of the fault prediction value, so as to ensure that the system can accurately identify different levels of fault risks. The first fault warning threshold is set as the mean minus one standard deviation, thereby defining normal fluctuations within a safe range; while the second fault warning threshold is to increase the sensitivity and safety of the warning by adding a safety factor multiplied by the standard deviation based on the mean. This design not only takes into account the volatility of historical data, but also adjusts the safety factor so that the system can adapt to the needs of different application scenarios, thereby effectively preventing potential failures and ensuring the stable operation and safety of the equipment.

[0085] When the fault prediction value is less than or equal to the first fault warning threshold, it is judged as the first warning, indicating that the performance data is within a safe range and the user can use it with confidence;

[0086] When the fault prediction value is between the first fault warning threshold and the second fault warning threshold, it is judged as the second warning, indicating that the performance data is slightly abnormal. It is recommended to conduct real-time monitoring and prepare to take measures;

[0087] When the fault prediction value is greater than the second fault warning threshold, it is judged as the third warning, indicating that the performance data is seriously abnormal and facing a fault. Check immediately and take repair measures.

[0088] It should be noted that by calculating the fault warning threshold, the system can take corresponding measures according to different risk levels. This hierarchical warning mechanism ensures that the system can operate safely when the risk of failure is low, and timely monitoring or intervention measures can be taken when the risk increases. This mechanism makes maintenance work more efficient, prioritizes high-risk situations, and ensures system stability.

[0089] S5. Based on different warning levels, start the automatic optimization feedback mechanism to dynamically adjust the load and optimize resource allocation.

[0090] When in safe range, maintain current load and resource configuration;

[0091] When there are minor anomalies, the monitoring frequency of temperature, load and power consumption is increased. Based on historical data and current load conditions, it is recommended to appropriately reduce the system load. The operator is informed of the current warning status, recommended measures and possible risks through the user interface or notification system;

[0092] When performance data is seriously abnormal, the preset emergency response mechanism is automatically triggered to immediately reduce the load, shut down non-critical tasks or services, prioritize system stability, and dynamically adjust the allocation of computing resources (such as CPU and memory) according to the current state of the system. Resources can be transferred from low-priority tasks to high-priority tasks to ensure the stable operation of critical services. The troubleshooting program is started to check system performance data, identify potential causes of failures, automatically collect fault logs and prepare relevant data for subsequent analysis and repair.

[0093] Dynamically adjusting load and optimizing resource allocation means using machine learning algorithms to analyze historical data, identify load patterns and resource requirements, optimize resource allocation strategies, adaptively adjust load and resource allocation strategies based on real-time performance data and warning levels, regularly evaluate the effectiveness of automatic optimization mechanisms, make continuous improvements based on system performance and fault occurrence, collect operator feedback on the automatic optimization feedback mechanism for improvement, and feed new data and feedback information back to the fault prediction model for continuous learning and optimization.

[0094] It should be noted that when the system identifies an anomaly, it can automatically adjust the load and resource allocation to ensure the stable operation of key services. This dynamic adjustment capability enables the system to respond to changes in a timely manner, avoid the expansion of potential failures, and improve overall operational efficiency. At the same time, continuous improvement using machine learning algorithms ensures that the system can adapt to new data patterns and enhances the system's adaptive capabilities.

[0095] This embodiment also provides a performance management system based on an artificial intelligence chip, including: a data collection module, responsible for collecting multi-dimensional performance data in real time through a monitoring unit built into the artificial intelligence chip;

[0096] The data preliminary processing module is responsible for preprocessing the collected multi-dimensional performance data and converting the preprocessed multi-dimensional performance data into a structured performance data set;

[0097] The fault prediction model module is responsible for using the performance data set to build a fault prediction model and obtain the fault prediction value;

[0098] The warning level classification module is responsible for classifying faults into different warning levels according to the fault prediction value;

[0099] The automatic optimization feedback mechanism module is responsible for starting the automatic optimization feedback mechanism based on different warning levels, dynamically adjusting the load and optimizing resource allocation.

[0100] This embodiment also provides a computer device, which is suitable for the performance management method based on artificial intelligence chip, including: a memory and a processor; the memory is used to store computer executable instructions, and the processor is used to execute computer executable instructions to implement the performance management method based on artificial intelligence chip proposed in the above embodiment.

[0101] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0102] This embodiment also provides a storage medium on which a computer program is stored. When the program is executed by a processor, the performance management method based on an artificial intelligence chip as proposed in the above embodiment is implemented; the storage medium can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (Static Random Access Memory, referred to as SRAM), electrically erasable programmable read-only memory (Electrically Erasable Programmable Read-Only Memory, referred to as EEPROM), erasable programmable read-only memory (Erasable Programmable Read-Only Memory, referred to as EPROM), programmable read-only memory (Programmable Red-Only Memory, referred to as PROM), read-only memory (Read-Only Memory, referred to as ROM), magnetic storage, flash memory, disk or optical disk.

[0103] In summary, the present invention collects multi-dimensional performance data in real time through the built-in monitoring unit of the artificial intelligence chip, and the invention effectively solves the problem of insufficient real-time monitoring and dynamic response capabilities of existing fault prediction technologies. The monitoring unit has high-frequency sampling capabilities and can collect key parameters such as temperature, current and load multiple times per second, thereby ensuring the real-time and accuracy of the data. On this basis, the fault prediction model constructed using the generated structured data set comprehensively considers the nonlinear effects and time decay effects of multi-dimensional performance indicators, enabling the system to more comprehensively evaluate the risk of failure. This model can not only identify potential faults in a timely manner, but also start a dynamic optimization feedback mechanism according to different warning levels. Through intelligent adjustment of load and resources, the system can increase the monitoring frequency in the case of minor anomalies, and automatically reduce the load or shut down non-critical tasks in the case of serious anomalies, thereby ensuring the stable operation of the system.

[0104] Example 2, referring to Table 1, is the second example of the present invention. In order to further verify the technical solution of the present invention, experimental simulation data of the performance management method based on artificial intelligence chip are provided.

[0105] In order to verify the effectiveness of the proposed fault prediction model, an industrial equipment was selected as the experimental object, and the equipment type was a high-load motor. First, a monitoring unit with a built-in artificial intelligence chip was installed, including a temperature sensor, a current sensor, and a load sensor to ensure that multi-dimensional performance data can be collected in real time. The sampling frequency of the monitoring unit was set to 10 times per second to improve the real-time nature of the data. The collected data included the temperature (unit: ℃), power consumption (unit: kW), and load (unit: %) of the equipment. After one week of monitoring operation, the collected raw data was preprocessed to remove noise and outliers to ensure the accuracy of the data. Subsequently, the data was standardized into a structured performance data set, stored in CSV format, and a timestamp in ISO8601 format was added to each data to ensure global standardization of the data.

[0106] Next, the fault prediction model is constructed using the preprocessed data. First, the initial fault prediction value F is calculated. 0 Then, polynomial and nonlinear factors are introduced to generate the fault prediction value F after introduction 1 , and finally form the final model F for comprehensive evaluation of system failure risk. According to the model, the fault warning thresholds T1 and T2 are calculated, and the warning levels are divided according to different fault prediction values. During the implementation process, the automatic optimization feedback mechanism is started at different warning levels to dynamically adjust the load and resource allocation of the equipment. The experiment records parameters such as temperature, power consumption and load at different warning levels, analyzes their impact on equipment stability, and provides data support for subsequent optimization.

[0107] The details are shown in Table 1 below:

[0108] Table 1 Experimental record table

[0109]

[0110]

[0111] Through the experimental data in Table 1, we can clearly observe the performance of the fault prediction model under different working conditions. Initial fault prediction value F 0 The change in the subsequent fault prediction value F shows the stability and risk assessment of the system during operation. The data in the table shows that when the temperature and load of the equipment gradually increase, the fault prediction value F also increases accordingly, indicating that the system is highly sensitive to potential faults. Especially when reaching the third warning level, the prediction value increases significantly, prompting the operator to take timely measures.

[0112] Compared with other technologies, this model shows significant advantages in the accuracy and real-time performance of fault prediction. Existing technologies usually rely on a single indicator for fault monitoring, while the present invention comprehensively considers multi-dimensional performance data, making fault prediction more comprehensive and accurate. By dynamically adjusting the load and optimizing the allocation of resources, the system can automatically respond when the risk of failure increases, greatly improving the safety and stability of the equipment. This multi-dimensional fault prediction capability and real-time early warning mechanism ensures that the equipment can continue to operate safely and efficiently in complex working environments, providing a solution that significantly improves the shortcomings of existing technologies.

[0113] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. A performance management method based on artificial intelligence chip, characterized by: include, Collect multi-dimensional performance data in real time through the monitoring unit built into the AI ​​chip; Preprocessing the collected multidimensional performance data, and converting the preprocessed multidimensional performance data into a structured performance data set; Using the performance data set, a fault prediction model is constructed to obtain the fault prediction value; Classify faults into different warning levels according to the fault prediction value; Based on different warning levels, the automatic optimization feedback mechanism is activated to dynamically adjust the load and optimize resource allocation.

2. The performance management method based on artificial intelligence chip according to claim 1, characterized in that: The monitoring unit includes a temperature sensor, a current sensor and a load sensor; The multi-dimensional performance data includes temperature, power consumption and load.

3. The performance management method based on artificial intelligence chip according to claim 2, characterized in that: Preprocessing the collected multidimensional performance data and converting the preprocessed multidimensional performance data into a structured performance data set includes the following steps: The collected multi-dimensional performance data is cleaned, noise and outliers are removed, and then standardized; Determine the format of the structured data set and add a timestamp to each piece of multidimensional performance data; Each multi-dimensional performance data after adding a timestamp is integrated to generate a performance data set.

4. The performance management method based on artificial intelligence chip according to claim 3, characterized in that: Using the performance data set, a fault prediction model is constructed to obtain the fault prediction value, including the following steps: According to the performance data set, set the temperature threshold and power consumption threshold of the AI ​​chip under normal operating conditions; Select the random forest model, which is an ensemble learning method consisting of multiple decision trees; extracting a ratio of temperature to a temperature threshold and a ratio of power consumption to a power consumption threshold from the performance data set; The calculated ratio is used to train the random forest model to obtain a trained random forest model, predict the possibility of a fault, and obtain a fault prediction value, which is expressed as: Where F is the fault prediction value, t is the current time point, τ is the time variable, erf is the error function, λ is the time decay rate, σ is the standard deviation, T is the temperature, L is the load, W is the power consumption, α is the temperature adjustment coefficient, T c is the normal threshold of temperature, β is the load adjustment coefficient, γ is the constant term, W n is the normal threshold for power consumption.

5. The performance management method based on artificial intelligence chip according to claim 4, characterized in that: According to the fault prediction value, the fault is divided into different warning levels, including the following steps: Based on the fault prediction value, the fault warning threshold is calculated and expressed as, T1 = μ-σ'; T2=μ+k·σ'; Wherein, T1 is the first fault warning threshold, T2 is the second fault warning threshold, μ is the mean of the fault prediction value, k is the safety factor, and σ' is the standard deviation of the fault prediction value F; When the fault prediction value is less than or equal to the first fault warning threshold, it is judged as the first warning, indicating that the performance data is within a safe range and users can use it with confidence; When the fault prediction value is between the first fault warning threshold and the second fault warning threshold, it is judged as the second warning, indicating that the performance data is slightly abnormal. It is recommended to conduct real-time monitoring and prepare to take measures; When the fault prediction value is greater than the second fault warning threshold, it is judged as the third warning, indicating that the performance data is seriously abnormal and facing a fault. Check and take repair measures immediately.

6. The performance management method based on artificial intelligence chip according to claim 5, characterized in that: Based on different warning levels, the automatic optimization feedback mechanism is started, including the following steps: When in safe range, maintain current load and resource configuration; When there is a slight abnormality, the monitoring frequency of temperature, load and power consumption is increased, and the system load is reduced according to historical data and current load conditions. The operator is informed of the current warning status, recommended measures and risks through the user interface; When performance data is seriously abnormal, the preset emergency response mechanism is automatically triggered to immediately reduce the load, dynamically adjust the allocation of computing resources according to the current state of the system, start the troubleshooting program, check the system performance data, identify potential causes of failures, automatically collect fault logs and prepare related data.

7. The performance management method based on artificial intelligence chip according to claim 6, characterized in that: Dynamically adjusting load and optimizing resource allocation means using machine learning algorithms to analyze historical data, identify load patterns and resource requirements, optimize resource allocation strategies, adaptively adjust load and resource configuration strategies based on real-time performance data and warning levels, regularly evaluate the effectiveness of automatic optimization mechanisms, make continuous improvements based on system performance and fault occurrence, collect feedback from operators on the automatic optimization feedback mechanism, feed new data and feedback information back to the fault prediction model, and conduct continuous learning and optimization.

8. A performance management system based on an artificial intelligence chip, based on the performance management method based on an artificial intelligence chip according to any one of claims 1 to 7, characterized in that: include, The data collection module is responsible for collecting multi-dimensional performance data in real time through the monitoring unit built into the AI ​​chip; The data preliminary processing module is responsible for preprocessing the collected multi-dimensional performance data and converting the preprocessed multi-dimensional performance data into a structured performance data set; The fault prediction model module is responsible for using the performance data set to build a fault prediction model and obtain the fault prediction value; The warning level classification module is responsible for classifying faults into different warning levels according to the fault prediction value; The automatic optimization feedback mechanism module is responsible for starting the automatic optimization feedback mechanism based on different warning levels, dynamically adjusting the load and optimizing resource allocation.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the performance management method based on the artificial intelligence chip described in any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the performance management method based on an artificial intelligence chip according to any one of claims 1 to 7 are implemented.

Images