An embedded control system fault self-healing method and system

Abstract

The application discloses an embedded control system fault self-recovery method and system, and belongs to the technical field of embedded systems. The method comprises the following steps: setting and dynamically adjusting a fault threshold value based on historical data and current working conditions; triggering a pre-warning when a real-time parameter exceeds the dynamic threshold value, and confirming by comparing main and redundant sensor data to distinguish between sensor abnormalities and real equipment faults; intelligently diagnosing based on a knowledge base using methods such as fault tree analysis to determine the fault type, location and severity; the system matches candidate self-recovery strategies from a strategy library, selects and executes the optimal strategy by calculating the value and execution priority of each strategy, and completes self-recovery control through software and hardware execution mechanisms. The application significantly improves the accuracy and reliability of fault detection through dynamic fault threshold values and redundant data confirmation, effectively avoiding false positives; precise positioning of faults and optimal self-recovery strategy selection are achieved.

Description

Technical Field

[0001] This invention belongs to the field of automatic control, and in particular relates to a method and system for self-healing faults in embedded control systems. Background Technology

[0002] Currently, embedded control systems are widely used in industrial control, smart devices, and consumer electronics. However, due to the complex and variable environment in which embedded systems operate, they may be affected by various interferences and faults, such as hardware failures, software errors, and communication failures. These faults may lead to performance degradation, functional failure, or even safety incidents. Therefore, improving the reliability and fault tolerance of embedded control systems, and achieving automatic fault detection, diagnosis, and self-healing, has become a hot research topic.

[0003] Existing fault handling methods for embedded control systems primarily rely on manual intervention, such as restarting the equipment, upgrading software, or manually locating fault points. These methods are not only inefficient but also ill-suited for handling sudden or complex fault situations. Furthermore, some methods may require operation while the equipment is offline, severely impacting production efficiency and system availability. Another common fault handling technique is rule-based diagnostic systems, which rely on a pre-defined rule base to reason about and judge faults. However, building and updating the rule base requires extensive expertise and experience and is difficult to cover all possible fault scenarios, thus limiting the accuracy and generalization ability of the diagnosis.

[0004] Existing fault handling methods for embedded control systems mainly rely on manual intervention. For example, after a fault is detected by a monitoring system, it is manually investigated and repaired. This method is inefficient, cannot address faults in a timely manner, and may lead to prolonged system downtime, resulting in significant economic losses. Although there are some automatic fault detection and diagnosis methods, these methods often can only detect simple faults and lack effective self-healing capabilities for complex fault scenarios. Summary of the Invention

[0005] To address the aforementioned technical problems, this invention proposes a technical solution for a self-healing method for embedded control systems. The purpose of this invention is to provide an automated system and method for self-healing faults in embedded control systems, thereby resolving the issues of existing embedded control system fault handling methods that rely on manual intervention, are inefficient, and lack sufficient self-healing capabilities.

[0006] The first aspect of this invention discloses a fault self-healing method for an embedded control system, the method comprising:

[0007] Step S1: Based on historical normal operation data, calculate the initial fault thresholds corresponding to various preset operating conditions using the 3σ principle;

[0008] During real-time system operation, the current operating condition is identified, and a dynamic fault threshold is obtained by dynamically adjusting the initial fault threshold based on the operating condition.

[0009] When the number of consecutive preset number of times the collected real-time feature parameters exceed the dynamic fault threshold, a fault warning is triggered.

[0010] Furthermore, by comparing the consistency between the main sensor data and at least one set of redundant sensor data, the fault warning is confirmed to distinguish between sensor abnormalities and equipment failures, and confirmed fault information is generated.

[0011] Step S2: Based on a preset knowledge base, at least one diagnostic method among fault tree analysis, expert system or support vector machine is used to diagnose the confirmed fault information to determine the type, location and severity of the fault, and generate diagnostic results;

[0012] Step S3: Receive the diagnostic results and match multiple candidate self-healing strategies from a preset strategy library;

[0013] For each candidate self-healing strategy, a strategy value is calculated based on the probability that it can resolve the current fault after execution and the reward it can bring.

[0014] Based on the value of the strategy, and combined with a set of preset dynamic decision weights, the execution priority of each candidate self-healing strategy is calculated.

[0015] In addition, select the candidate self-healing strategy with the highest execution priority and generate the corresponding self-healing control instructions;

[0016] Step S4: Execute the self-healing control instructions through a hardware actuator or software control program to complete the fault self-healing;

[0017] Step S5: Evaluate the execution effect of the fault self-healing, and update the diagnosis results, the candidate self-healing strategies and the evaluation results to the knowledge base and / or the strategy base.

[0018] Preferably, in step S1, the dynamic adjustment of the initial fault threshold based on the operating condition to obtain a dynamic fault threshold includes:

[0019] The mean and standard deviation of real-time feature parameters are calculated using the sliding window technique;

[0020] Dynamic fault threshold = initial fault threshold × [1 + 0.1 × (real-time mean - interval mean) / interval standard deviation].

[0021] Preferably, in step S2, the diagnostic result includes:

[0022] Fault types: hardware faults, software faults, and communication faults;

[0023] Fault severity: minor, moderate, severe, fatal, and catastrophic.

[0024] Preferably, in step S3, the candidate self-healing strategy includes:

[0025] Redundancy switching strategy :

[0026] Perform the following actions:

[0027] If the main device fails, it will switch to the hot standby device.

[0028] If the hot standby fails, the cold standby device will be activated.

[0029] If the cold backup fails, the software fault tolerance mode will be enabled.

[0030] Fault Repair Strategy :

[0031] Perform the following actions:

[0032] : Reset the faulty module;

[0033] Configuration parameter verification;

[0034] Functional self-test;

[0035] Link Reconstruction Strategy :

[0036] Perform the following actions:

[0037] : Real-time acquisition of system resource status matrix;

[0038] The optimal reconstruction scheme is found using a genetic algorithm.

[0039] Generates a reconstructed sequence containing 15 control instructions.

[0040] Preferably, in step S3, the strategy value is calculated based on the probability that its execution can resolve the current fault and the resulting reward, including:

[0041]

[0042] in, Indicates when a certain strategy is adopted Troubleshooting s The value function; Representation strategy; Indicates when a certain strategy is adopted At that time, the action is executed after the fault s. The probability of; Indicates a fault; A represents the action to be performed; A represents the set of actions to be performed under a certain strategy. Indicates when a certain strategy is adopted Perform actions under fault s Expected returns at that time.

[0043] Preferably, in step S3,

[0044] in, Indicates when a certain strategy is adopted Perform actions under fault s Instant rewards , To perform the action Expected value of fault self-healing; Indicates when a certain strategy is adopted In the fault s Next action Future returns at that time The range of values ​​for the discount factor is: ; Indicates the execution of an action The subsequent failure; Indicates the execution of an action The probability of; This indicates the action to be performed under fault s. A malfunction occurred later. The probability of.

[0045] Preferably, in step S3, calculating the execution priority of each candidate self-healing strategy based on the strategy value and a set of preset dynamic decision weights includes:

[0046] Based on fault severity v, execution time t, and redundant resource status System operation continuity and the probability of self-healing failure during execution. Quantification yields process quantities

[0047] =

[0048] By multiplying the dynamic decision weights by the process variables, the execution priority is obtained, i.e.

[0049] Priority =

[0050] Where w=[ [] indicates dynamic decision weight.

[0051] A second aspect of this invention discloses an embedded control system fault self-healing system, the system comprising:

[0052] The fault detection module is configured to calculate the initial fault threshold corresponding to various preset operating conditions based on historical normal operation data and using the 3σ principle.

[0053] During real-time system operation, the current operating condition is identified, and a dynamic fault threshold is obtained by dynamically adjusting the initial fault threshold based on the operating condition.

[0054] When the number of consecutive preset number of times the collected real-time feature parameters exceed the dynamic fault threshold, a fault warning is triggered.

[0055] Furthermore, by comparing the consistency between the main sensor data and at least one set of redundant sensor data, the fault warning is confirmed to distinguish between sensor abnormalities and equipment failures, and confirmed fault information is generated.

[0056] The fault diagnosis module is configured to diagnose the confirmed fault information based on a preset knowledge base, using at least one of the following diagnostic methods: fault tree analysis, expert system, or support vector machine, to determine the type, location, and severity of the fault, and generate a diagnostic result.

[0057] The self-healing decision module is configured to receive the diagnostic results and match multiple candidate self-healing strategies from a preset strategy library.

[0058] For each candidate self-healing strategy, a strategy value is calculated based on the probability that it can resolve the current fault after execution and the reward it can bring.

[0059] Based on the value of the strategy, and combined with a set of preset dynamic decision weights, the execution priority of each candidate self-healing strategy is calculated.

[0060] In addition, select the candidate self-healing strategy with the highest execution priority and generate the corresponding self-healing control instructions;

[0061] The self-healing execution module is configured to execute the self-healing control instructions through a hardware actuator or a software control program to complete fault self-healing.

[0062] A third aspect of this invention discloses an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the self-healing method for an embedded control system according to any one of the first aspects of this disclosure.

[0063] A fourth aspect of this invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a fault self-healing method for an embedded control system according to any one of the first aspects of this disclosure.

[0064] As can be seen, the solution proposed in this invention achieves automatic detection, diagnosis, and self-healing of system faults, reducing manual intervention, improving the efficiency and accuracy of fault handling, and enabling timely restoration of normal system operation, reducing system downtime and economic losses. It can accurately detect and diagnose complex fault scenarios, improving the system's fault tolerance and reliability. The self-healing decision module formulates appropriate self-healing strategies and solutions based on fault diagnosis results, and the execution module implements specific self-healing operations, achieving automated fault handling and improving the system's intelligence level. The knowledge base module can be continuously updated and improved, accumulating more fault handling experience and knowledge through data mining and machine learning methods, enhancing the system's self-learning and adaptive capabilities, enabling it to adapt to different application scenarios and fault conditions. Attached Figure Description

[0065] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0066] Figure 1 This is a flowchart of a fault self-healing method for an embedded control system according to an embodiment of the present invention;

[0067] Figure 2 This is a schematic diagram of an embedded control system fault self-healing system according to an embodiment of the present invention;

[0068] Figure 3 This is a structural diagram of an embedded control system fault self-healing system according to an embodiment of the present invention;

[0069] Figure 4 This is a structural diagram of an electronic device according to an embodiment of the present invention. Detailed Implementation

[0070] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, 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.

[0071] The purpose of this invention is to provide an automated fault self-healing system and method for embedded control systems, addressing the problems of existing embedded control systems' fault handling relying on manual intervention, low efficiency, and insufficient self-healing capability. Specifically, it includes the following steps:

[0072] Data acquisition: The data acquisition module collects the operating data of various devices and sensors in the embedded control system in real time and transmits the data to the data processing module;

[0073] Data preprocessing and feature extraction: The data processing module preprocesses the collected data, including filtering, noise reduction, and normalization, and then extracts key feature parameters that can reflect the system's operating status.

[0074] Fault Detection: Based on the extracted feature parameters, the fault detection module uses multiple fault detection algorithms to monitor the system's operating status in real time and determine whether a fault has occurred in the system. If a fault is detected, a fault alarm is issued and the fault information is transmitted to the fault diagnosis module.

[0075] Fault Diagnosis: After receiving fault information, the fault diagnosis module analyzes and diagnoses the fault by combining the knowledge information in the knowledge base module, determines the type, location and severity of the fault, and transmits the diagnosis results to the self-healing decision module.

[0076] Self-healing decision: Based on the fault diagnosis results, the self-healing decision module selects appropriate self-healing strategies and solutions from the knowledge base module, generates specific control instructions and operation steps, and transmits them to the execution module;

[0077] Self-healing execution: The execution module executes control commands and operation steps to achieve self-healing of faults, including but not limited to repairing faulty equipment, switching redundant equipment, and reconstructing the system structure;

[0078] Effectiveness Evaluation and Knowledge Update: After the self-healing process is completed, the system's operational status is evaluated to determine whether the self-healing effect is satisfactory. If self-healing is successful, the fault handling process and results are recorded in the knowledge base module for future reference. If self-healing fails, an alarm is issued to prompt the user for manual intervention, and relevant information is recorded.

[0079] The first aspect of this invention discloses a fault self-healing method for embedded control systems. Figure 1 This is a flowchart of a fault self-healing method for an embedded control system according to an embodiment of the present invention, as shown below. Figure 1 As shown, the method includes:

[0080] Step S1: Fault detection and confirmation, performed by the fault detection module;

[0081] Based on historical normal operation data, the initial fault thresholds corresponding to various preset operating conditions are calculated using the 3σ principle.

[0082] During real-time system operation, the current operating condition is identified, and a dynamic fault threshold is dynamically adjusted based on the initial fault threshold of the operating condition to obtain a dynamic fault threshold.

[0083] When the number of consecutive preset number of times the collected real-time feature parameters exceed the dynamic fault threshold, a fault warning is triggered.

[0084] Furthermore, by comparing the consistency between the main sensor data and at least one set of redundant sensor data, the fault warning is confirmed to distinguish between sensor abnormalities and equipment failures, and confirmed fault information is generated.

[0085] In step S1, the dynamic adjustment of the initial fault threshold based on the operating condition to obtain a dynamic fault threshold includes:

[0086] The mean and standard deviation of real-time feature parameters are calculated using the sliding window technique;

[0087] Dynamic fault threshold = initial fault threshold × [1 + 0.1 × (real-time mean - interval mean) / interval standard deviation].

[0088] Specifically as follows:

[0089] The fault detection algorithm in the fault detection module is based on an adaptive fault detection algorithm with dynamic thresholds. This algorithm builds a dynamic threshold model through training on historical data, solving the problem that traditional fixed thresholds cannot adapt to complex working conditions. The specific implementation steps are as follows:

[0090] Step S11: Threshold Model Construction Stage

[0091] 1. Collect characteristic parameters (such as amplitude / frequency / phase of physical quantities such as voltage, current, and temperature) for three complete working cycles during normal operation of the system, with a sample size of no less than 2000 sets;

[0092] 2. Calculate the initial threshold range using the 3σ principle: Upper threshold = mean + 3 times standard deviation, lower threshold = mean - 3 times standard deviation;

[0093] 3. Establish a working condition mapping table, divide the equipment into 4 operating ranges according to the equipment load rate (0-25% / 25%-50% / 50%-75% / 75%-100%), and build dynamic threshold models for each range;

[0094] Step S12: Real-time detection phase

[0095] 1. The real-time characteristic parameters output by the data processing module are first used for operating condition identification to determine the current load range;

[0096] 2. The real-time mean and standard deviation are calculated using a sliding window technique (window size set to 50 sampling periods);

[0097] 3. Dynamically Adjusted Threshold: Current threshold = Initial threshold × [1 + 0.1 × (Real-time mean - Interval mean) / Interval standard deviation] (Adaptive adjustment coefficient);

[0098] 4. When the feature parameter exceeds the dynamic threshold range for 10 consecutive sampling periods, a fault warning is triggered;

[0099] Step S12: Fault Location Assistance

[0100] 1. Simultaneously collect data from three sets of redundant sensors for consistency verification;

[0101] 2. If the main sensor data exceeds the threshold but the redundant sensor data is normal, mark it as "sensor abnormal";

[0102] 3. If the data from multiple sensors all exceed the threshold and meet the operating condition correlation characteristics (such as abnormal motor temperature accompanied by increased current), it is determined to be "equipment failure"; when an abnormality is detected in the system, the fault detection module will issue a fault alarm and transmit the fault information to the fault diagnosis module.

[0103] S2: Fault diagnosis, performed by the fault diagnosis module.

[0104] Based on a pre-defined knowledge base, at least one diagnostic method among fault tree analysis, expert system, or support vector machine is used to diagnose the confirmed fault information in order to determine the type, location, and severity of the fault and generate diagnostic results.

[0105] In step S2, the diagnostic results include:

[0106] Fault types: hardware faults, software faults, and communication faults;

[0107] Fault severity: minor, moderate, severe, fatal, and catastrophic.

[0108] Specifically as follows:

[0109] The fault diagnosis module employs fault tree analysis, expert systems, and support vector machines, combined with the system's historical data and knowledge base, to accurately locate and diagnose faults, determining their type, location, and severity. Simultaneously, the fault diagnosis module transmits the diagnostic results to the self-healing decision module.

[0110] Step S3: Self-healing decision, executed by the self-healing decision module;

[0111] Receive the diagnostic results and match multiple candidate self-healing strategies from a preset strategy library;

[0112] For each candidate self-healing strategy, a strategy value is calculated based on the probability that it can resolve the current fault after execution and the reward it can bring.

[0113] Based on the value of the strategy, and combined with a set of preset dynamic decision weights, the execution priority of each candidate self-healing strategy is calculated.

[0114] In addition, select the candidate self-healing strategy with the highest execution priority and generate the corresponding self-healing control instructions;

[0115] In step S3, the candidate self-healing strategy includes:

[0116] Redundancy switching strategy :

[0117] Perform the following actions:

[0118] If the main device fails, it will switch to the hot standby device.

[0119] If the hot standby fails, the cold standby device will be activated.

[0120] If the cold backup fails, the software fault tolerance mode will be enabled.

[0121] Fault Repair Strategy :

[0122] Perform the following actions:

[0123] : Reset the faulty module;

[0124] Configuration parameter verification;

[0125] Function self-test;

[0126] Link Reconstruction Strategy :

[0127] Perform the following actions:

[0128] : Real-time acquisition of system resource status matrix;

[0129] The optimal reconstruction scheme is found using a genetic algorithm.

[0130] Generates a reconstructed sequence containing 15 control instructions.

[0131] In step S3, a strategy value is calculated based on the probability that its execution can resolve the current fault and the resulting reward, including:

[0132] in, Indicates when a certain strategy is adopted Troubleshooting s The value function; Representation strategy; Indicates when a certain strategy is adopted At that time, the action is executed after the fault s. The probability of; Indicates a fault; A represents the action to be performed; A represents the set of actions to be performed under a certain strategy. Indicates when a certain strategy is adopted Perform actions under fault s Expected returns at that time.

[0133] In step S3, calculating the execution priority of each candidate self-healing strategy based on the strategy value and a set of preset dynamic decision weights includes:

[0134] Based on fault severity v, execution time t, and redundant resource status System operation continuity and the probability of self-healing failure during execution. Quantification yields process quantities

[0135] =

[0136] By multiplying the dynamic decision weights by the process variables, the execution priority is obtained, i.e.

[0137] Priority =

[0138] Where w=[ [] indicates dynamic decision weight.

[0139] Specifically as follows:

[0140] The self-healing decision module selects the most suitable self-healing strategy based on the type and severity of the fault, and generates specific control instructions and operation steps, which are then transmitted to the execution module. The specific implementation process is as follows:

[0141] Step S31: Fault Characteristic Analysis

[0142] 1. Parse and receive diagnostic results from the fault diagnosis module:

[0143] Fault type U: Hardware failure Software malfunction , communication failure

[0144] Fault severity V: (slight), (medium), (serious), (fatal), (Catastrophic)

[0145] A collection of faults

[0146] in:

[0147] : Representative fault type

[0148] : Represents the severity of the fault

[0149] Step S32: Classification of the strategy library

[0150] 1. Redundancy handover strategy :

[0151] Perform the following actions:

[0152] Main equipment failure → Switch to hot standby equipment (<10ms switching time);

[0153] Hot standby failure → Activate cold standby device (requires 3 seconds initialization);

[0154] Cold backup failure → Activate software fault tolerance mode (performance degradation of 50%).

[0155] 2. Fault Repair Strategy :

[0156] Perform the following actions:

[0157] Faulty module reset (wait 500ms response time);

[0158] Configuration parameter validation (comparing with standard parameters in the knowledge base);

[0159] Functional self-test (perform verification of 3 core functions);

[0160] 3. Link Reconstruction Strategy :

[0161] Perform the following actions:

[0162] Real-time acquisition of system resource status matrix (including 8 types of hardware resources and 5 types of software resources);

[0163] Genetic algorithms are used to find the optimal reconstruction scheme (the fitness function includes three elements: reconstruction time, resource utilization, and reliability).

[0164] Generate a reconstructed sequence containing 15 control instructions (executed in order of priority).

[0165] According to the fault Adopt corresponding strategies At that time, perform the action The probability of:

[0166]

[0167] in:

[0168] : Execute action probability

[0169] , : =1,2,3 corresponds to , , , =1,2,3 corresponds to , ,

[0170] According to the fault Adopt corresponding strategies At that time, the action is performed. The reward obtained, i.e., the strategy used At that time, the fault Take action below The value equals the immediate reward plus the product of the decayed probability of the next possible failure and its corresponding value:

[0171]

[0172] in: To perform the action Expected value of fault self-healing

[0173] : Indicates when a certain strategy is adopted Perform actions under fault s Expected return at time

[0174] : Indicates when a certain strategy is adopted Perform actions under fault s Instant returns

[0175] : Perform action The subsequent failure

[0176] : Indicates when a certain strategy is adopted Perform actions under fault s Future returns at that time The range of values ​​for the discount factor is: ; Indicates the execution of an action The subsequent failure; Indicates the execution of an action The probability of; This indicates the action to be performed under fault s. A malfunction occurred later. The probability of.

[0177] Step S33: Calculate the value function for decision-making based on the fault type, fault severity, strategy type, and execution action:

[0178]

[0179] Introduction get:

[0180]

[0181] : Representation strategy

[0182] : Indicates when a certain strategy is adopted At that time, the value function for resolving fault s

[0183] : Indicates when a certain strategy is adopted At that time, the action is executed after the fault s. probability

[0184] :Fault

[0185] : Execute action

[0186] The set of actions performed under a given strategy

[0187] Fault Set

[0188] The fault is calculated based on the above formula. In strategy Next action value .

[0189] Based on fault severity v, execution time t, and redundant resource status System operation continuity and the probability of self-healing failure during execution. Quantification

[0190] =

[0191] Step S34: Dynamic Weight Decision

[0192] Establish an evaluation matrix w containing 5 decision factors:

[0193]

[0194] Calculate the priority of the actions to be executed by the strategy:

[0195] Priority =

[0196] Send it to the execution module.

[0197] Step S4: Self-healing execution, performed by the self-healing execution module.

[0198] The self-healing control instructions are executed through a hardware actuator or a software control program to complete the fault self-healing.

[0199] Specifically, the execution module includes hardware actuators and software control programs, such as hardware devices like relays, motors, and switches, as well as corresponding control algorithms and programs; through the actions of the execution module, the faulty equipment is repaired, reconstructed, or switched, so that the system can return to normal operation.

[0200] Furthermore, the method includes step S5: evaluating the effectiveness of the fault self-healing process and updating the diagnostic results, candidate self-healing strategies, and evaluation results to the knowledge base and / or the strategy base. The effectiveness evaluation and knowledge update are performed by the effectiveness evaluation and knowledge update module; specifically, the knowledge base module continuously updates and improves the system's fault detection, diagnosis, and self-healing capabilities through data mining and machine learning methods; simultaneously, the knowledge base module also provides knowledge support and reference for the fault diagnosis module and the self-healing decision-making module.

[0201] In summary, the embedded control system fault self-healing automation system and method provided by this invention achieves automatic detection, diagnosis, and self-healing of system faults, reducing manual intervention, improving the efficiency and accuracy of fault handling, and enabling timely restoration of normal system operation, reducing system downtime and economic losses. The system employs multiple fault detection and diagnosis algorithms, combined with knowledge information from the knowledge base module, to accurately detect and diagnose complex fault scenarios, improving the system's fault tolerance and reliability. The self-healing decision module formulates appropriate self-healing strategies and schemes based on the fault diagnosis results, and the execution module implements the specific self-healing operations, achieving automated fault handling and improving the system's intelligence level. The knowledge base module can be continuously updated and improved, accumulating more fault handling experience and knowledge through data mining and machine learning methods, enhancing the system's self-learning and adaptive capabilities, enabling it to adapt to different application scenarios and fault conditions.

[0202] The embedded control system fault self-healing automation system and method provided by this invention realizes automatic detection, diagnosis, and self-healing of system faults, reducing manual intervention, improving the efficiency and accuracy of fault handling, and enabling timely restoration of normal system operation, reducing system downtime and economic losses. It can accurately detect and diagnose complex fault scenarios, improving the system's fault tolerance and reliability. The self-healing decision module formulates appropriate self-healing strategies and schemes based on fault diagnosis results, and the execution module implements specific self-healing operations, achieving automated fault handling and improving the system's intelligence level. The knowledge base module can be continuously updated and improved, accumulating more fault handling experience and knowledge through data mining and machine learning methods, improving the system's self-learning and adaptive capabilities, enabling it to adapt to different application scenarios and fault conditions.

[0203] The second aspect of this invention discloses a fault self-healing system for embedded control systems. Figure 3 This is a structural diagram of an embedded control system fault self-healing system according to an embodiment of the present invention; as shown below. Figure 3 As shown, the system 100 includes:

[0204] The fault detection module 101 is configured to calculate the initial fault threshold corresponding to various preset operating conditions based on historical normal operation data.

[0205] During real-time system operation, the current operating condition is identified, and a dynamic fault threshold is dynamically adjusted based on the initial fault threshold of the operating condition to obtain a dynamic fault threshold.

[0206] When the number of consecutive preset number of times the collected real-time feature parameters exceed the dynamic fault threshold, a fault warning is triggered.

[0207] Furthermore, by comparing the consistency between the main sensor data and at least one set of redundant sensor data, the fault warning is confirmed to distinguish between sensor abnormalities and equipment failures, and confirmed fault information is generated.

[0208] The fault diagnosis module 102 is configured to diagnose the confirmed fault information based on a preset knowledge base, using at least one of the following diagnostic methods: fault tree analysis, expert system, or support vector machine, to determine the type, location, and severity of the fault, and generate a diagnostic result.

[0209] The self-healing decision module 103 is configured to receive the diagnostic results and match multiple candidate self-healing strategies from a preset strategy library.

[0210] For each candidate self-healing strategy, a strategy value is calculated based on the probability that it can resolve the current fault after execution and the reward it can bring.

[0211] Based on the value of the strategy, and combined with a set of preset dynamic decision weights, the execution priority of each candidate self-healing strategy is calculated.

[0212] In addition, select the candidate self-healing strategy with the highest execution priority and generate the corresponding self-healing control instructions;

[0213] The self-healing execution module 104 is configured to execute the self-healing control instructions through a hardware actuator or a software control program to complete fault self-healing.

[0214] In some specific embodiments,

[0215] like Figure 2As shown, the self-healing fault system for embedded control systems provided by this invention can be implemented based on existing embedded hardware platforms, such as ARM processors and DSP processors. The data acquisition module connects to the controlled device and sensors via sensors and interface circuits to collect operational data. The data processing module, fault detection module, fault diagnosis module, self-healing decision-making module, self-healing execution module, and knowledge base module can be implemented in software on the processor, utilizing the processor's computing and storage capabilities for data processing and decision-making. The execution module connects to the controlled device via a hardware actuator to achieve self-healing operations. The human-machine interaction module can use devices such as a touchscreen, keyboard, and display to enable user interaction with the system. The communication module can use hardware devices such as network cards, serial port chips, and wireless communication modules to achieve data transmission and interaction.

[0216] The system's software employs a modular design, with modules communicating and exchanging data via interfaces. The data acquisition module controls sensors and interface circuits through driver programs to achieve real-time data acquisition. The data processing module uses digital signal processing algorithms to preprocess and extract features from the data. The fault detection and fault diagnosis modules utilize corresponding algorithm libraries to implement fault detection and diagnosis functions. The self-healing decision module generates control commands based on fault diagnosis results and self-healing strategies in the knowledge base. The execution module controls hardware actuators through control algorithms and driver programs to achieve self-healing operations. The knowledge base module stores knowledge information using a database or file system and provides corresponding interfaces for other modules to access and update it.

[0217] Taking an embedded control system for an industrial robot as an example, the specific application of this invention is illustrated. During operation, industrial robots may experience motor failures, sensor failures, communication failures, etc. The data acquisition module collects real-time data on the current, voltage, and temperature of the motors at each joint of the robot, as well as the position and speed of the sensors, and transmits this data to the data processing module for preprocessing and feature extraction. The fault detection module analyzes the data using threshold detection and neural network detection methods to determine if a fault exists. If an abnormal motor current is detected, a fault alarm is issued. The fault diagnosis module determines, through fault tree analysis, that a short circuit fault in the motor windings is the cause. The self-healing decision module selects a strategy to switch to a redundant motor, generates control commands, and the execution module controls the relay to switch to the redundant motor, allowing the robot to resume normal operation. Simultaneously, the fault handling process and results are recorded in the knowledge base module for future reference.

[0218] The key technical point of this invention is that it provides an embedded control system fault self-healing system. This system integrates multiple modules, including data acquisition, processing, fault detection, diagnosis, self-healing decision-making, execution, knowledge base, human-computer interaction, and communication, to achieve automatic detection, diagnosis, and self-healing of system faults. Specifically, the data acquisition module is responsible for real-time acquisition of operational data from various devices and sensors in the system; the data processing module preprocesses and extracts features from the acquired data; the fault detection module uses various algorithms to monitor the system's operating status in real time; the fault diagnosis module accurately analyzes and locates detected faults; the self-healing decision-making module formulates appropriate self-healing strategies based on the diagnostic results; the execution module implements the specific self-healing operations; the knowledge base module provides knowledge support and reference for the system's fault detection, diagnosis, and self-healing; the human-computer interaction module enables interaction between humans and the system; and the communication module ensures communication between various modules within the system and between the system and external devices.

[0219] This invention describes the overall architecture of an automated fault self-healing system for embedded control systems, the functions and designs of its various modules, and the methods and processes for automatic fault detection, diagnosis, and self-healing implemented by the system. Specifically, the scope of protection sought by this invention covers, but is not limited to, the specific design, implementation methods, and collaborative mechanisms of the data acquisition module, data processing module, fault detection module, fault diagnosis module, self-healing decision module, execution module, knowledge base module, human-computer interaction module, and communication module, as well as the specific algorithms and processes for automatic fault detection, diagnosis, and self-healing implemented by the system. Furthermore, this invention also seeks protection for the application of this system in the field of embedded control system fault handling, and the technical effects of the system in improving fault handling efficiency, accuracy, and intelligence.

[0220] A third aspect of this invention discloses an electronic device. The electronic device includes a memory and a processor. The memory stores a computer program, and when the processor executes the computer program, it implements the steps of the self-healing method for an embedded control system according to any one of the first aspects of this invention.

[0221] Figure 4 This is a structural diagram of an electronic device according to an embodiment of the present invention, such as... Figure 4As shown, the electronic device includes a processor, memory, communication interface, display screen, and input device connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, Near Field Communication (NFC), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input device can be a touch layer covering the display screen, buttons, a trackball, or a touchpad mounted on the device's casing, or an external keyboard, touchpad, or mouse.

[0222] Those skilled in the art will understand that Figure 4 The structure shown is merely a structural diagram of the part related to the technical solution of this disclosure and does not constitute a limitation on the electronic device to which the solution of this application is applied. The specific electronic device may include more or fewer components than shown in the figure, or combine certain components, or have different component arrangements.

[0223] A fourth aspect of this invention discloses a computer-readable storage medium. The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of a fault self-healing method for an embedded control system according to any one of the first aspects of this invention.

[0224] Please note that the technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments have been described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification. The above embodiments only illustrate several implementation methods of this application, and their descriptions are relatively specific and detailed, but they should not be construed as limiting the scope of the invention patent. It should be pointed out that for those skilled in the art, several modifications and improvements can be made without departing from the concept of this application, and these all fall within the protection scope of this application. Therefore, the protection scope of this patent application should be determined by the appended claims.

Claims

1. A self-healing method for embedded control systems, characterized in that, The method includes: Step S1: Based on historical normal operation data, calculate the initial fault thresholds corresponding to various preset operating conditions; During real-time system operation, the current operating condition is identified, and a dynamic fault threshold is dynamically adjusted based on the initial fault threshold of the operating condition to obtain a dynamic fault threshold. When the number of consecutive preset number of times the collected real-time feature parameters exceed the dynamic fault threshold, a fault warning is triggered. Furthermore, by comparing the consistency between the main sensor data and at least one set of redundant sensor data, the fault warning is confirmed to distinguish between sensor abnormalities and equipment failures, and confirmed fault information is generated. Step S2: Based on a preset knowledge base, at least one diagnostic method among fault tree analysis, expert system or support vector machine is used to diagnose the confirmed fault information to determine the type, location and severity of the fault, and generate diagnostic results; Step S3: Receive the diagnostic results and match multiple candidate self-healing strategies from a preset strategy library; For each candidate self-healing strategy, a strategy value is calculated based on the probability that it can resolve the current fault after execution and the reward it can bring. Based on the value of the strategy, and combined with a set of preset dynamic decision weights, the execution priority of each candidate self-healing strategy is calculated. In addition, select the candidate self-healing strategy with the highest execution priority and generate the corresponding self-healing control instructions; Step S4: Execute the self-healing control instructions through a hardware actuator or software control program to complete the fault self-healing; In step S3, a strategy value is calculated based on the probability that its execution can resolve the current fault and the resulting reward, including: in, Indicates when a certain strategy is adopted Troubleshooting s The value function; Representation strategy; Indicates when a certain strategy is adopted At that time, the action is executed after the fault s. The probability of; Indicates a fault; A represents the action to be performed; A represents the set of actions to be performed under a certain strategy. Indicates when a certain strategy is adopted Perform actions under fault s Expected returns at that time; In step S3, the expected return Calculated using the following formula: in, Indicates when a certain strategy is adopted Perform actions under fault s Instant rewards , To perform the action Expected value of fault self-healing; Indicates when a certain strategy is adopted In the fault s Next action Future returns at that time The range of values ​​for the discount factor is: ; Indicates the execution of an action The subsequent failure; Indicates the execution of an action The probability of; Indicates a fault s Next action A malfunction occurred later. The probability of.

2. The self-healing method for embedded control systems according to claim 1, characterized in that, In step S1, the dynamic adjustment of the initial fault threshold based on the operating condition to obtain a dynamic fault threshold includes: The mean and standard deviation of real-time feature parameters are calculated using the sliding window technique; Dynamic fault threshold = initial fault threshold × [1 + 0.1 × (real-time mean - interval mean) / interval standard deviation].

3. The self-healing method for embedded control systems according to claim 1, characterized in that, In step S2, the diagnostic results include: Fault types: hardware faults, software faults, and communication faults; Fault severity: minor, moderate, severe, fatal, and catastrophic.

4. A fault self-healing method for an embedded control system according to claim 1 or 3, characterized in that, In step S3, the candidate self-healing strategy includes: Redundancy switching strategy : Perform the following actions: If the main device fails, it will switch to the hot standby device. If the hot standby fails, the cold standby device will be activated. If the cold backup fails, the software fault tolerance mode will be enabled. Fault Repair Strategy : Perform the following actions: : Reset the faulty module; Configuration parameter verification; Functional self-test; Link Reconstruction Strategy : Perform the following actions: : Real-time acquisition of system resource status matrix; The optimal reconstruction scheme is found using a genetic algorithm. Generates a reconstructed sequence containing 15 control instructions.

5. The self-healing method for embedded control systems according to claim 1, characterized in that, In step S3, calculating the execution priority of each candidate self-healing strategy based on the strategy value and a set of preset dynamic decision weights includes: Based on fault severity v, execution time t, and redundant resource status System operation continuity and the probability of self-healing failure during execution. Quantification yields process quantities = By multiplying the dynamic decision weights by the process variables, the execution priority is obtained, i.e. Priority = Where w=[ [] indicates dynamic decision weight.

6. A fault self-healing system for embedded control systems, characterized in that, The system employs the method described in any one of claims 1-5, and the system comprises: The fault detection module is configured to calculate the initial fault threshold corresponding to various preset operating conditions based on historical normal operation data. During real-time system operation, the current operating condition is identified, and a dynamic fault threshold is dynamically adjusted based on the initial fault threshold of the operating condition to obtain a dynamic fault threshold. When the number of consecutive preset number of times the collected real-time feature parameters exceed the dynamic fault threshold, a fault warning is triggered. Furthermore, by comparing the consistency between the main sensor data and at least one set of redundant sensor data, the fault warning is confirmed to distinguish between sensor abnormalities and equipment failures, and confirmed fault information is generated. The fault diagnosis module is configured to diagnose the confirmed fault information based on a preset knowledge base, using at least one of the following diagnostic methods: fault tree analysis, expert system, or support vector machine, to determine the type, location, and severity of the fault, and generate a diagnostic result. The self-healing decision module is configured to receive the diagnostic results and match multiple candidate self-healing strategies from a preset strategy library. For each candidate self-healing strategy, a strategy value is calculated based on the probability that it can resolve the current fault after execution and the reward it can bring. Based on the value of the strategy, and combined with a set of preset dynamic decision weights, the execution priority of each candidate self-healing strategy is calculated. In addition, select the candidate self-healing strategy with the highest execution priority and generate the corresponding self-healing control instructions; The self-healing execution module is configured to execute the self-healing control instructions through a hardware actuator or a software control program to complete fault self-healing.

7. An electronic device, characterized in that, The electronic device includes a memory and a processor. The memory stores a computer program. When the processor executes the computer program, it implements the steps of the self-healing method for embedded control system faults according to any one of claims 1 to 5.

8. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the self-healing method for an embedded control system according to any one of claims 1 to 5.

Images