A motherboard intelligent power supply method and system based on fault diagnosis feedback

Abstract

This invention belongs to the field of motherboard power supply technology. It discloses a motherboard intelligent power supply method and system based on fault diagnosis feedback. The method involves determining the motherboard's historical power supply data and storing it in a preset fault database. Within this database, a collection area, an isolation area, and a trust area are defined. A collection channel is established between the isolation area and the collection area, and a verification channel is established between the isolation area and the trust area. The data within the corresponding collection area is monitored and verified within the collection channel, and the data that passes the monitoring and verification is input as an anomaly parameter to the isolation area. Furthermore, this invention solves the problem that when existing motherboards are undergoing fault diagnosis, if the fault is a maliciously created fault, the fault database is easily affected by the malicious fault, leading to inaccurate power supply strategies output.

Description

Technical Field

[0001] This invention relates to the field of motherboard power supply technology, and more specifically, to a motherboard intelligent power supply method and system based on fault diagnosis feedback. Background Technology

[0002] As computer systems become increasingly complex, the motherboard, as a core hardware platform, integrates a variety of key components, including the central processing unit (CPU), memory controller, chipset, power management unit (PMU), and various peripheral interfaces. When diagnosing faults on existing motherboards, if the fault is a maliciously created fault, the fault database can be easily affected by the malicious fault, causing the output power supply strategy to lose accuracy.

[0003] In view of this, the present invention proposes a motherboard intelligent power supply method and system based on fault diagnosis feedback to solve the above problems. Summary of the Invention

[0004] To overcome the aforementioned deficiencies of the prior art and to achieve the above objectives, the present invention provides the following technical solution: a motherboard intelligent power supply method based on fault diagnosis feedback, comprising the following steps: S1. Determine the historical power supply data of the motherboard and store the historical power supply data in the preset fault database. Set up a collection area, an isolation area, and a trust area in the fault database. S2. Establish a collection channel between the isolation zone and the collection zone, and establish a verification channel between the isolation zone and the trust zone; S3. Monitor and verify the data in the corresponding collection area within the collection channel, and input the data that has passed the monitoring and verification as anomaly parameters into the isolation area. S4. In the verification channel, perform correlation verification on the abnormal parameters in the corresponding isolation area, input the abnormal parameters that meet the correlation verification to the trust area, the trust area diagnoses the abnormal parameters and outputs the corresponding power supply strategy, and executes the power supply strategy using the virtual twin model set in the trust area.

[0005] Furthermore, the steps of determining the motherboard's historical power supply data and storing the historical power supply data in a preset fault database, and setting up a collection area, an isolation area, and a trust area within the fault database, include: The fault database is divided into a collection area, an isolation area, and a trust area; Determine the component information and topology of the motherboard, and divide the motherboard into multiple component zones according to the component information and topology. Determine the historical power supply data corresponding to each component area, and pre-store multiple component areas and their corresponding historical power supply data in the trust area of ​​the fault database; wherein, the historical power supply data includes fault parameters and power supply strategies.

[0006] Furthermore, the steps of establishing a collection channel between the isolation zone and the collection zone, and establishing a verification channel between the isolation zone and the trust zone, include: Within the collection area, isolation area, and trust area, the same number of collection circles, isolation circles, and trust circles are configured according to the number of component areas, with one component area corresponding to one collection circle, isolation circle, and trust circle; Historical power supply data corresponding to multiple component areas are stored in their respective trust circles. The trust circles, isolation circles, and collection circles are sequentially associated. The collection circle is used to collect data of the component areas within its corresponding trust circle. Establish collection channels between the corresponding collection zones and isolation zones; Establish a verification channel between the corresponding isolation zone and trust zone.

[0007] Furthermore, the step of monitoring and verifying the data in the corresponding collection area within the collection channel, and inputting the data that has passed the monitoring and verification as anomaly parameters into the isolation area includes: Determine the component area corresponding to the collection circle, and deploy multiple monitoring points in the corresponding component area. The data monitored by the multiple monitoring points are input into the collection channel corresponding to the collection circle. Each monitoring point monitors one type of data on the component area. A screening gate is set at the node connecting the collection channel and the isolation zone. A limit frame and a retention frame are configured for each monitoring point of the screening gate. The data monitored by the corresponding monitoring point are filtered using the limit frame. Data exceeding the limit frame is input into the isolation zone, and data within the limit frame is temporarily stored in the retention frame.

[0008] Furthermore, the step of configuring a limiting frame and a retention frame corresponding to each monitoring point for the screening gate, and using the limiting frames to filter the data monitored at the corresponding monitoring points, includes: Record the data value of each monitoring point and the corresponding values ​​of multiple adjacent time points in chronological order to form the data fluctuation sequence of that monitoring point. A dynamic confidence interval threshold is calculated based on the data fluctuation sequence of each monitoring point. The dynamic confidence interval threshold includes a numerical range threshold and a volatility threshold. The limiting box is configured as follows: for the input real-time value, combined with the real-time data fluctuation sequence of the monitoring point, it is determined whether the real-time value and the real-time data fluctuation sequence exceed the corresponding dynamic confidence interval threshold. Real-time values ​​and real-time data fluctuation sequences that exceed the limit frame are treated as abnormal parameters and input into the isolation area, while real-time values ​​and real-time data fluctuation sequences that do not exceed the limit frame are temporarily stored in the corresponding retention frame.

[0009] Furthermore, the step of correlating and verifying abnormal parameters within the corresponding isolation zone in the verification channel includes: An association gate is set at the node connecting the verification channel and the trust circle. The association gate is configured with a soft frame, a hard frame, an extended frame, and a projected frame. Among them, the first type of association rule set, which is based on the dynamic coupling characteristics between the data corresponding to multiple monitoring points with nonlinear influence relationships, is used as the soft frame; the second type of rigid constraint condition set, which is based on physical laws or linear influence relationships, is used as the hard frame; the third comprehensive condition set, which is formed by taking the union of the judgment results of the soft frame and the hard frame, is used as the extended frame; and the fourth type of system-level association rule set, which is based on the functional dependence, signal transmission, or thermal / electrical / control coupling relationship between different component areas, is used as the projected frame. The abnormal parameters in the isolation area are copied and distributed to the soft box, hard box, and extended box, and matched and judged respectively. If the abnormal parameter meets the judgment condition of at least one of the soft box, hard box, or extended box, it is input into the projection box for further verification. The abnormal parameter is considered to meet the association verification only when the abnormal parameter meets the conditions of the projection box at the same time; otherwise, it does not meet the conditions.

[0010] Furthermore, the step of inputting the abnormal parameters that meet the correlation verification into the trust zone, diagnosing the abnormal parameters and outputting the corresponding power supply strategy, and executing the power supply strategy using the virtual twin model set in the trust zone includes: The abnormal parameters that pass the correlation verification are taken as trusted faults and the trusted faults are input into the corresponding trust circle. The trust circle is the power supply strategy corresponding to the trusted fault output. In the trusted zone, a virtual twin model of the motherboard is set up in the corresponding mapping. The component area corresponding to the virtual twin model is divided into multiple jurisdictional zones, and one jurisdictional zone corresponds to one component area. A main control point is set up in each jurisdiction, and the main control point is communicatively connected to the control center of the main board. The power supply parameters in the power supply strategy are determined, the main control point in the corresponding jurisdiction is activated based on the power supply parameters, and the corresponding secondary control points are configured in the jurisdiction according to the power supply parameters as the main control points; wherein, the secondary control points are communicatively associated with the virtual components in the jurisdiction corresponding to the power supply parameters, and the secondary control points are associated with the power supply parameters one-to-one. The configured secondary control point executes the control signals corresponding to the associated power supply parameters, and the secondary control point is destroyed after it completes the execution. When a new power supply strategy is generated, the main control point is copied for the new power supply strategy until the new secondary control point corresponding to the new power supply strategy is completed and destroyed.

[0011] Furthermore, the step of determining the power supply parameters in the power supply strategy and activating the main control point within the corresponding jurisdiction based on the power supply parameters includes: The power supply strategy is broken down into multiple power supply parameters and filled into a preset allocation table. In the allocation table, each table corresponds to a power supply parameter. A receiving table is established corresponding to the allocation table, and each table in the receiving table corresponds to the location and jurisdiction of the virtual component; Establish a one-to-one communication channel between each table in the allocation table and each table in the receiving table; When the power supply parameters are filled into the table in the allocation table, the communication channel transmits the power supply parameters to the corresponding table in the receiving table to complete the activation of the corresponding main control point. When a new power supply strategy is generated, a new allocation table and a receiving table are added, and the allocation table and receiving table corresponding to the completed power supply strategy are destroyed.

[0012] The present invention also includes a motherboard intelligent power supply system based on fault diagnosis feedback, which includes: The configuration module is used to determine the motherboard's historical power supply data and store the historical power supply data in a preset fault database. Within the fault database, a collection area, an isolation area, and a trust area are set up. Establish a module, which is used to establish a collection channel between the quarantine zone and the collection zone, and to establish a verification channel between the quarantine zone and the trusted zone; The verification module is used to monitor and verify the data in the corresponding collection area within the collection channel, and input the data that passes the monitoring and verification as anomaly parameters into the isolation area. The update module is used to perform correlation verification on abnormal parameters in the corresponding isolation area within the verification channel. Abnormal parameters that meet the correlation verification are input to the trust area. The trust area diagnoses the abnormal parameters and outputs the corresponding power supply strategy. The power supply strategy is executed using the virtual twin model set in the trust area.

[0013] The technical effects and advantages of this invention, which is a motherboard intelligent power supply method and system based on fault diagnosis feedback, are as follows: 1. By rationally deploying monitoring points in the component area, comprehensive data on motherboard component operation can be obtained, providing rich information for fault diagnosis; by setting monitoring verification and correlation verification, the complex relationships between the corresponding data of each component on the motherboard can be comprehensively and deeply considered, avoiding misjudgment of fault due to a single abnormal data; it can accurately identify abnormal parameters that are truly related to motherboard faults, avoid externally maliciously generated abnormal parameters, improve the accuracy and reliability of fault diagnosis, and facilitate the output of accurate power supply strategies; 2. By communicating with the main control point and the control center of the motherboard, the main control point replicates the secondary control point based on the power supply parameters, and a communication channel is established between the allocation table and the receiving table, parameters can be accurately transmitted, the main control point corresponding to the power supply parameters can be automatically activated, the control signals corresponding to the power supply parameters can be executed one-to-one with high execution efficiency, the power supply strategy can be accurately located to the components, and the system stability can be improved. Attached Figure Description

[0014] Figure 1 This is a flowchart illustrating the motherboard intelligent power supply method based on fault diagnosis feedback according to the present invention.

[0015] Figure 2 This is a schematic diagram illustrating the combination of the collection area, isolation area, and trust area in this invention.

[0016] Figure 3 This is a schematic diagram of the motherboard intelligent power supply system based on fault diagnosis feedback according to the present invention. Detailed Implementation

[0017] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0018] Please see Figure 1 and Figure 2 As shown, the motherboard intelligent power supply method based on fault diagnosis feedback described in this embodiment includes the following steps: S1. Determine the historical power supply data of the motherboard and store the historical power supply data in the preset fault database. Set up a collection area, an isolation area, and a trust area in the fault database. S2. Establish a collection channel between the isolation zone and the collection zone, and establish a verification channel between the isolation zone and the trust zone; S3. Monitor and verify the data in the corresponding collection area within the collection channel, and input the data that has passed the monitoring and verification as anomaly parameters into the isolation area. S4. In the verification channel, perform correlation verification on the abnormal parameters in the corresponding isolation area, input the abnormal parameters that meet the correlation verification to the trust area, the trust area diagnoses the abnormal parameters and outputs the corresponding power supply strategy, and executes the power supply strategy using the virtual twin model set in the trust area.

[0019] In this embodiment, by rationally deploying monitoring points in the component area, comprehensive data on motherboard component operation can be acquired, providing rich information for fault diagnosis. Preliminary screening of monitoring data is achieved by setting filtering gates, limiting frames, and retention frames. By recording data fluctuation sequences and calculating dynamic confidence interval thresholds, the normal fluctuation range of data can be reflected more comprehensively and accurately, rather than relying solely on single numerical values. By combining real-time values ​​with data fluctuation sequences for judgment, the accuracy and reliability of abnormal data detection are improved, reducing misjudgments and omissions of abnormal parameters. Through correlation verification, the complex relationships between various data on the motherboard can be comprehensively and deeply considered, avoiding misjudgments of faults due to single data analysis. Abnormal parameters truly related to motherboard faults can be identified more accurately, avoiding maliciously generated abnormal parameters, improving the accuracy and reliability of fault diagnosis, and providing a more solid foundation for subsequent fault handling. This solves the problem that in existing motherboard fault diagnosis, if the fault is a maliciously generated fault, the fault database is easily affected by the malicious fault, leading to inaccurate power supply strategies output.

[0020] In one embodiment, the step of determining the motherboard's historical power supply data and storing the historical power supply data in a preset fault database, and setting up a collection area, an isolation area, and a trust area within the fault database, includes: The fault database is divided into a collection area, an isolation area, and a trust area; Determine the component information and topology of the motherboard, and divide the motherboard into multiple component zones according to the component information and topology. Determine the historical power supply data corresponding to each component area, and pre-store multiple component areas and their corresponding historical power supply data in the trust area of ​​the fault database; wherein, the historical power supply data includes fault parameters and power supply strategies; It should be noted that, for example, motherboard component information includes the CPU, memory, hard drive, power supply module, etc., and the topology refers to the connection methods and data transmission paths between them. Based on this information, the motherboard is divided into multiple component areas, such as the CPU component area, memory component area, hard drive component area, etc. Historical power supply data for each component area over a period of time is collected. By determining the corresponding component information and topology of the motherboard, the motherboard is divided into multiple component areas according to the components and topology. This allows each component area to be categorized with its corresponding historical power supply data, facilitating the location of faulty components corresponding to faulty parameters. By setting up collection areas, isolation areas, and trust areas in the fault database, it is easy to monitor and verify the collected abnormal parameters, preventing maliciously set abnormal parameters from being added to the fault database, which could cause the motherboard to output power supply strategies due to abnormal parameters. This reduces the risk of erroneous power supply strategies and ensures the reliability of the power supply strategy.

[0021] In one embodiment, the steps of establishing a collection channel between the isolation zone and the collection zone, and establishing a verification channel between the isolation zone and the trust zone, include: Within the collection area, isolation area, and trust area, the same number of collection circles, isolation circles, and trust circles are configured according to the number of component areas, with one component area corresponding to one collection circle, isolation circle, and trust circle; Historical power supply data corresponding to multiple component areas are stored in their respective trust circles. The trust circles, isolation circles, and collection circles are sequentially associated. The collection circle is used to collect data of the component areas within its corresponding trust circle. Establish collection channels between the corresponding collection zones and isolation zones; Establish a verification channel between the corresponding isolation zone and trust zone; It should be noted that by storing the historical power supply data corresponding to multiple component areas into their respective trust circles, and linking the trust circles, isolation circles, and collection circles in sequence, the collection circles can be used to monitor and collect the data of the component areas within the corresponding trust circles.

[0022] In one embodiment, the step of monitoring and verifying the data in the corresponding collection area within the collection channel, and inputting the data that has passed the monitoring and verification as anomaly parameters into the isolation area includes: Determine the component area corresponding to the collection circle, and deploy multiple monitoring points in the corresponding component area. The data monitored by the multiple monitoring points are input into the collection channel corresponding to the collection circle. Each monitoring point monitors one type of data on the component area. A screening gate is set at the node connecting the collection channel and the isolation zone. A limit frame and a retention frame are configured for each monitoring point of the screening gate. The data monitored by the corresponding monitoring point are filtered using the limit frame. Data exceeding the limit frame is input into the isolation zone, and data within the limit frame is temporarily stored in the retention frame. In one embodiment, the step of configuring a limiting frame and a retention frame corresponding to each monitoring point for the screening gate, and using the limiting frames to filter the data monitored by the corresponding monitoring points, includes: Record the data value of each monitoring point and the corresponding values ​​of multiple adjacent time points in chronological order to form the data fluctuation sequence of that monitoring point. A dynamic confidence interval threshold is calculated based on the data fluctuation sequence of each monitoring point. The dynamic confidence interval threshold includes a numerical range threshold and a volatility threshold. The limiting box is configured as follows: for the input real-time value, combined with the real-time data fluctuation sequence of the monitoring point, it is determined whether the real-time value and the real-time data fluctuation sequence exceed the corresponding dynamic confidence interval threshold. Real-time values ​​and real-time data fluctuation sequences that exceed the limit frame are treated as abnormal parameters and input into the isolation area, while real-time values ​​and real-time data fluctuation sequences that do not exceed the limit frame are temporarily stored in the corresponding retention frame. It should be noted that the most recent N data points are taken, and the mean ± 2 standard deviations are calculated as the data's numerical range threshold. The data volatility threshold is determined by the moving average and standard deviation of the absolute differences between adjacent points. Data within the retention frame can be overwritten by subsequent newly monitored data and data volatility sequences after a certain period of time. For example, the temperature monitoring point in the CPU component area records data once per minute, recording temperature data from 10:00 to 10:30, forming a data volatility sequence. The calculated numerical range threshold for this monitoring point under normal conditions is set to 30℃~70℃, and the volatility threshold is a temperature change rate not exceeding 10%~50% per half hour. At 10:15, A temperature value of 75℃ was detected. Upon examining the temperature data at several nearby time points (10:14 and 10:16), the values ​​were 68℃ and 70℃ respectively. Combining this with the data fluctuation sequence during this period, it was found that the value exceeded the threshold range. Therefore, this value was determined to be outside the defined bounding box and was input as an anomaly parameter into the isolation zone. Furthermore, by setting a dynamic confidence interval threshold, data exceeding the dynamic confidence interval threshold can be treated as anomaly parameters and input into the isolation zone for subsequent correlation verification. By combining real-time values ​​with data fluctuation sequences for judgment, the accuracy and reliability of anomaly parameter detection are improved, reducing the occurrence of false positives and false negatives.

[0023] In one embodiment, the step of performing correlation verification on abnormal parameters within the corresponding isolation zone in the verification channel includes: An association gate is set at the node connecting the verification channel and the trust circle. The association gate is configured with a soft frame, a hard frame, an extended frame, and a projected frame. Among them, the first type of association rule set, which is based on the dynamic coupling characteristics between the data corresponding to multiple monitoring points with nonlinear influence relationships, is used as the soft frame; the second type of rigid constraint condition set, which is based on physical laws or linear influence relationships, is used as the hard frame; the third comprehensive condition set, which is formed by taking the union of the judgment results of the soft frame and the hard frame, is used as the extended frame; and the fourth type of system-level association rule set, which is based on the functional dependence, signal transmission, or thermal / electrical / control coupling relationship between different component areas, is used as the projected frame. The abnormal parameters in the isolation area are copied and distributed to the soft box, hard box, and extended box, and matched and judged respectively. If the abnormal parameter meets the judgment condition of at least one of the soft box, hard box, or extended box, it is input into the projection box for further verification. The abnormal parameter is considered to meet the association verification only when the abnormal parameter meets the conditions of the projection box at the same time; otherwise, it does not meet the conditions. It should be noted that, for example, taking the motherboard area as an example, when the temperature sensor reports an abnormally high temperature value (such as 75℃, i.e., an abnormal parameter), and the current sensor simultaneously detects a significant increase in operating current; analysis reveals that there is a non-linear growth trend between temperature and current (e.g., leakage current increases exponentially with increasing temperature), which conforms to the non-linear coupling relationship defined by the soft frame; under the current stable power supply voltage conditions, the measured current change is still within the reasonable deviation range of the theoretically calculated value (based on I=V / R(T)). The model satisfies the linear constraints of the hard bounding box. Combining the two, the abnormal combination falls within the third comprehensive condition set covered by the extended bounding box. Further monitoring shows that the abnormal parameter also causes fluctuations in the output voltage of the motherboard's power management unit and frequency reduction behavior of adjacent chips, which conforms to the fourth type of system-level association rule set in the projection box. Therefore, the abnormal parameter is determined to pass the association verification and can proceed to subsequent fault diagnosis steps. Furthermore, through association verification, the complex relationships between various data on the motherboard can be comprehensively and deeply considered, avoiding misjudgment of faults due to single data judgments. It can more accurately identify abnormal parameters that are truly related to motherboard faults, avoid externally maliciously generated abnormal parameters, improve the accuracy and reliability of fault diagnosis, and provide a more solid foundation for subsequent fault handling.

[0024] In one embodiment, the step of inputting the abnormal parameters that meet the correlation verification into the trust zone, the trust zone diagnosing the abnormal parameters and outputting the corresponding power supply strategy, and executing the power supply strategy using the virtual twin model set in the trust zone includes: The abnormal parameters that pass the correlation verification are taken as trusted faults and the trusted faults are input into the corresponding trust circle. The trust circle is the power supply strategy corresponding to the trusted fault output. In the trusted zone, a virtual twin model of the motherboard is set up in the corresponding mapping. The component area corresponding to the virtual twin model is divided into multiple jurisdictional zones, and one jurisdictional zone corresponds to one component area. A main control point is set up in each jurisdiction, and the main control point is communicatively connected to the control center of the main board. The power supply parameters in the power supply strategy are determined, the main control point in the corresponding jurisdiction is activated based on the power supply parameters, and the corresponding secondary control points are configured in the jurisdiction according to the power supply parameters as the main control points; wherein, the secondary control points are communicatively associated with the virtual components in the jurisdiction corresponding to the power supply parameters, and the secondary control points are associated with the power supply parameters one-to-one. The configured secondary control point executes the control signals corresponding to the associated power supply parameters, and the secondary control point is destroyed after it completes the execution. When a new power supply strategy is generated, the main control point is copied for the new power supply strategy until the new secondary control point corresponding to the new power supply strategy is completed and destroyed.

[0025] In one embodiment, the step of determining the power supply parameters in the power supply strategy and activating the main control point in the corresponding jurisdiction based on the power supply parameters includes: The power supply strategy is broken down into multiple power supply parameters and filled into a preset allocation table. In the allocation table, each table corresponds to a power supply parameter. A receiving table is established corresponding to the allocation table, and each table in the receiving table corresponds to the location and jurisdiction of the virtual component; Establish a one-to-one communication channel between each table in the allocation table and each table in the receiving table; When the power supply parameters are filled into the table in the allocation table, the communication channel transmits the power supply parameters to the corresponding table in the receiving table to complete the activation of the corresponding main control point. When a new power supply strategy is generated, a new allocation table and a receiving table are added, and the allocation table and receiving table corresponding to the completed power supply strategy are destroyed.

[0026] It should be noted that by inputting trusted faults into the corresponding trust circle, a trained machine learning fault diagnosis algorithm can be used to diagnose trusted faults on the motherboard, i.e., output the power supply strategy for trusted faults; alternatively, the matching of abnormal parameters with historical power supply data can be used to output the power supply strategy with the highest matching degree; thus, it is possible to avoid abnormal parameters maliciously generated by external parties, improve the accuracy and reliability of fault diagnosis, and at the same time improve the reliability and stability of the entire fault diagnosis feedback and motherboard power supply. When executing a power supply strategy using a virtual twin model: For example, the power supply strategy includes adjusting CPU voltage and memory frequency. This adjustment is broken down into two power supply parameters: a 1.2V CPU voltage and a 1600MHz memory frequency. These two parameters are filled into the CPU voltage table and the memory frequency table in a preset allocation table, respectively. The table position for each power supply parameter in the allocation table is fixed. When the allocation table transmits the power supply parameters to the corresponding table in the receiving table, the main control point within the corresponding jurisdiction is activated. After the main control point is activated, it simultaneously replicates the same number of secondary control points based on the virtual components corresponding to the power supply parameters. Each table in the allocation table, after being filled with the corresponding power supply parameters, is associated with a corresponding effective execution time based on the power supply strategy. The effective execution time is transmitted to the corresponding table in the receiving table along with the transmitted power supply parameters. The control point transmits the control signals corresponding to the power supply parameters to the motherboard's control center for execution based on the allocation table and the effective execution time associated with the table. The effective control time includes the start time and validity period of the control signal. Furthermore, through communication between the main control point and the motherboard's control center, the main control point replicating the secondary control point based on the power supply parameters, and establishing a communication channel between the allocation and receiving tables, parameters can be accurately transmitted. This automatically activates the main control point corresponding to the power supply parameters, facilitating one-to-one execution of the control signals corresponding to the power supply parameters with high efficiency. It also allows for precise positioning of the power supply strategy to components, improving system stability. When a new power supply strategy is generated, a new allocation and receiving table is created, and existing allocation and receiving tables corresponding to completed power supply strategies are destroyed. This prevents the accumulation of secondary control points, avoids control logic confusion, and ensures the lightweight and stability of the virtual twin model.

[0027] Please see Figure 3 As shown, the motherboard intelligent power supply system based on fault diagnosis feedback described in this embodiment includes: The configuration module is used to determine the motherboard's historical power supply data and store the historical power supply data in a preset fault database. Within the fault database, a collection area, an isolation area, and a trust area are set up. Establish a module, which is used to establish a collection channel between the quarantine zone and the collection zone, and to establish a verification channel between the quarantine zone and the trusted zone; The verification module is used to monitor and verify the data in the corresponding collection area within the collection channel, and input the data that passes the monitoring and verification as anomaly parameters into the isolation area. The update module is used to perform correlation verification on abnormal parameters in the corresponding isolation area within the verification channel. Abnormal parameters that meet the correlation verification are input to the trust area. The trust area diagnoses the abnormal parameters and outputs the corresponding power supply strategy. The power supply strategy is executed using the virtual twin model set in the trust area.

[0028] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed in this invention can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0029] In the several embodiments provided by this invention, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of units is only one method, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or units may be electrical, mechanical, or other forms.

[0030] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in the present invention should be included within the scope of protection of the present invention.

[0031] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A motherboard intelligent power supply method based on fault diagnosis feedback, characterized in that, Includes the following steps: Determine the motherboard's historical power supply data and store it in a preset fault database. Set up a collection area, an isolation area, and a trusted area within the fault database. Establish a collection channel between the quarantine zone and the collection zone, and establish a verification channel between the quarantine zone and the trust zone; The data in the corresponding collection area is monitored and verified within the collection channel, and the data that passes the monitoring and verification is input into the isolation area as anomaly parameters. Within the verification channel, abnormal parameters in the corresponding isolation area are correlated and verified. Abnormal parameters that meet the correlation verification are input to the trust area. The trust area diagnoses the abnormal parameters and outputs the corresponding power supply strategy. The power supply strategy is executed using the virtual twin model set in the trust area.

2. The motherboard intelligent power supply method based on fault diagnosis feedback according to claim 1, characterized in that, The steps of determining the motherboard's historical power supply data and storing it in a preset fault database, and setting up a collection area, an isolation area, and a trust area within the fault database, include: The fault database is divided into a collection area, an isolation area, and a trust area; Determine the component information and topology of the motherboard, and divide the motherboard into multiple component zones according to the component information and topology. Determine the historical power supply data corresponding to each component area, and pre-store multiple component areas and their corresponding historical power supply data in the trust area of ​​the fault database; wherein, the historical power supply data includes fault parameters and power supply strategies.

3. The motherboard intelligent power supply method based on fault diagnosis feedback according to claim 2, characterized in that, The steps of establishing a collection channel between the isolation zone and the collection zone, and establishing a verification channel between the isolation zone and the trust zone, include: Within the collection area, isolation area, and trust area, the same number of collection circles, isolation circles, and trust circles are configured according to the number of component areas, with one component area corresponding to one collection circle, isolation circle, and trust circle; Store the historical power supply data corresponding to multiple component areas into their respective trust circles; Establish collection channels between the corresponding collection zones and isolation zones; Establish a verification channel between the corresponding isolation zone and trust zone.

4. The motherboard intelligent power supply method based on fault diagnosis feedback according to claim 3, characterized in that, The step of monitoring and verifying the data in the corresponding collection area within the collection channel, and inputting the data that has passed the monitoring and verification as anomaly parameters into the isolation area includes: Determine the component area corresponding to the collection circle, and deploy multiple monitoring points in the corresponding component area. The data monitored by the multiple monitoring points are input into the collection channel corresponding to the collection circle. Each monitoring point monitors one type of data on the component area. A screening gate is set at the node connecting the collection channel and the isolation zone. A limit frame and a retention frame are configured for each monitoring point of the screening gate. The data monitored by the corresponding monitoring point are filtered using the limit frame. Data exceeding the limit frame is input into the isolation zone, and data within the limit frame is temporarily stored in the retention frame.

5. The motherboard intelligent power supply method based on fault diagnosis feedback according to claim 4, characterized in that, The step of configuring a limiting frame and a retention frame for each monitoring point of the screening gate, and using the limiting frames to filter the data monitored by the corresponding monitoring points, includes: Record the data value of each monitoring point and the corresponding values ​​of multiple adjacent time points in chronological order to form the data fluctuation sequence of that monitoring point. A dynamic confidence interval threshold is calculated based on the data fluctuation sequence of each monitoring point. The dynamic confidence interval threshold includes a numerical range threshold and a volatility threshold. The limiting box is configured as follows: for the input real-time value, combined with the real-time data fluctuation sequence of the monitoring point, it is determined whether the real-time value and the real-time data fluctuation sequence exceed the corresponding dynamic confidence interval threshold. Real-time values ​​and real-time data fluctuation sequences that exceed the limit frame are treated as abnormal parameters and input into the isolation area, while real-time values ​​and real-time data fluctuation sequences that do not exceed the limit frame are temporarily stored in the corresponding retention frame.

6. The motherboard intelligent power supply method based on fault diagnosis feedback according to claim 5, characterized in that, The step of correlating and verifying abnormal parameters in the corresponding isolation zone within the verification channel includes: An association gate is set at the node connecting the verification channel and the trust circle. The association gate is configured with a soft frame, a hard frame, an extended frame, and a projected frame. Among them, the first type of association rule set, which is based on the dynamic coupling characteristics between the data corresponding to multiple monitoring points with nonlinear influence relationships, is used as the soft frame; the second type of rigid constraint condition set, which is based on physical laws or linear influence relationships, is used as the hard frame; the third comprehensive condition set, which is formed by taking the union of the judgment results of the soft frame and the hard frame, is used as the extended frame; and the fourth type of system-level association rule set, which is based on the functional dependence, signal transmission, or thermal / electrical / control coupling relationship between different component areas, is used as the projected frame. The abnormal parameters in the isolation area are copied and distributed to the soft box, hard box, and extended box, and matched and judged respectively. If the abnormal parameter meets the judgment condition of at least one of the soft box, hard box, or extended box, it is input into the projection box for further verification. The abnormal parameter is considered to meet the association verification only when the abnormal parameter meets the conditions of the projection box at the same time; otherwise, it does not meet the conditions.

7. The motherboard intelligent power supply method based on fault diagnosis feedback according to claim 6, characterized in that, The steps of inputting the abnormal parameters that meet the correlation verification into the trust zone, diagnosing the abnormal parameters and outputting the corresponding power supply strategy, and executing the power supply strategy using the virtual twin model set in the trust zone include: The abnormal parameters that pass the correlation verification are taken as trusted faults and the trusted faults are input into the corresponding trust circle. The trust circle is the power supply strategy corresponding to the trusted fault output. In the trusted zone, a virtual twin model of the motherboard is set up in the corresponding mapping. The component area corresponding to the virtual twin model is divided into multiple jurisdictional zones, and one jurisdictional zone corresponds to one component area. A main control point is set up in each jurisdiction, and the main control point is communicatively connected to the control center of the main board. The power supply parameters in the power supply strategy are determined, the main control point in the corresponding jurisdiction is activated based on the power supply parameters, and the corresponding secondary control points are configured in the jurisdiction according to the power supply parameters as the main control points; wherein, the secondary control points are communicatively associated with the virtual components in the jurisdiction corresponding to the power supply parameters, and the secondary control points are associated with the power supply parameters one-to-one. The configured secondary control point executes the control signals corresponding to the associated power supply parameters, and the secondary control point is destroyed after it completes the execution. When a new power supply strategy is generated, the main control point is copied for the new power supply strategy until the new secondary control point corresponding to the new power supply strategy is completed and destroyed.

8. The motherboard intelligent power supply method based on fault diagnosis feedback according to claim 7, characterized in that, The step of determining the power supply parameters in the power supply strategy and activating the main control point in the corresponding jurisdiction based on the power supply parameters includes: The power supply strategy is broken down into multiple power supply parameters and filled into a preset allocation table. In the allocation table, each table corresponds to a power supply parameter. A receiving table is established corresponding to the allocation table, and each table in the receiving table corresponds to the location and jurisdiction of the virtual component; Establish a one-to-one communication channel between each table in the allocation table and each table in the receiving table; When the power supply parameters are filled into the table in the allocation table, the communication channel transmits the power supply parameters to the corresponding table in the receiving table to complete the activation of the corresponding main control point. When a new power supply strategy is generated, a new allocation table and a receiving table are added, and the allocation table and receiving table corresponding to the completed power supply strategy are destroyed.

9. A motherboard intelligent power supply system based on fault diagnosis feedback, applied to the motherboard intelligent power supply method based on fault diagnosis feedback as described in claims 1-8, characterized in that, include: The configuration module is used to determine the motherboard's historical power supply data and store the historical power supply data in a preset fault database. Within the fault database, a collection area, an isolation area, and a trust area are set up. Establish a module, which is used to establish a collection channel between the quarantine zone and the collection zone, and to establish a verification channel between the quarantine zone and the trusted zone; The verification module is used to monitor and verify the data in the corresponding collection area within the collection channel, and input the data that passes the monitoring and verification as anomaly parameters into the isolation area. The update module is used to perform correlation verification on abnormal parameters in the corresponding isolation area within the verification channel. Abnormal parameters that meet the correlation verification are input to the trust area. The trust area diagnoses the abnormal parameters and outputs the corresponding power supply strategy. The power supply strategy is executed using the virtual twin model set in the trust area.

Images