A method and system for fault detection and error correction in electronic products

Abstract

This application provides a method and system for fault detection and error correction in electronic products. The method includes: upon receiving a batch switching command or when the device is in a preset idle period, initiating an integrity verification process for calibration parameters stored internally in the smart terminal device; in the integrity verification process, reading the calibration parameters stored internally in the smart terminal device; generating a verification value reflecting the data status of the calibration parameters based on the read calibration parameters; comparing the generated verification value with a pre-stored reference verification value; determining whether the calibration parameters deviate based on the comparison result; when a deviation is determined, outputting deviation information of the calibration parameters; and performing a calibration parameter recovery operation based on the deviation information of the calibration parameters. This application can proactively and promptly detect potential deviations in the calibration parameters internally in the smart terminal device, improving the stability and reliability of the device.

Description

Technical Field

[0001] This application relates to the field of electronic product fault detection and error correction technology, and more specifically, to a method and system for electronic product fault detection and error correction. Background Technology

[0002] In existing industrial automated production lines, intelligent terminal equipment often experiences deviations in production parameters due to sensor reading drift, communication interference, or internal processing anomalies during long-term operation. However, existing fault detection mechanisms suffer from problems such as lag, high false alarm rates, and limited correction strategies. In particular, when the firmware of intelligent terminal equipment contains subtle defects, such as the internal automatic calibration program making erroneous, minor modifications to the stored sensor calibration reference values ​​due to minute calculation errors when processing specific production batch switching commands, these modifications are difficult to detect in a single instance. However, as production continues and batch switching commands are repeatedly executed, these minor erroneous modifications accumulate, eventually leading to a significant deviation in the calibration reference values ​​recorded internally by the system. This deviation in calibration reference values ​​manifests externally as sensor physical aging or drift, meaning that even after technicians replace the sensors with brand-new, fully functional ones, the equipment will still trigger alarms again after a period of operation due to data errors caused by firmware defects, creating a ghost fault that cannot be eradicated by simple hardware replacement. This type of fault not only causes significant disruption to the stable operation of the production line but also substantially increases maintenance costs.

[0003] Meanwhile, existing systems commonly employ signal smoothing techniques such as moving average calculations to eliminate short-term random noise. However, when faced with persistent and systematic deviations in the aforementioned sensor data, this smoothing process actually averages out the erroneous bias, further masking the true fault signal and prolonging the fault latency. When the accumulated deviation finally exceeds the alarm threshold, because the fault signal has been masked by long-term smoothing, the system cannot accurately trace back to the true starting point and evolution of the fault. It can only perform a preset single error correction action based on the current instantaneous data, failing to address the root cause of the fault, leading to recurring faults or poor correction results. Summary of the Invention

[0004] This application discloses a method and system for fault detection and error correction of electronic products, which aims to solve the problems of ghost faults caused by the accumulation of minor errors in internal calibration parameters during long-term operation of smart terminal devices, which are difficult to diagnose and eliminate, as well as the problems of traditional fault detection mechanisms being lagging behind, having a high false alarm rate, and having a single correction strategy.

[0005] The technical solution of this application is as follows: In a first aspect, this application discloses a method for fault detection and error correction of electronic products, used in smart terminal devices, comprising: Upon receiving a batch switching instruction or when the device is in a preset idle period, initiate the integrity verification process for the calibration parameters stored internally in the smart terminal device. In the integrity verification process, the calibration parameters stored inside the smart terminal device are read; Based on the read calibration parameters, a verification value reflecting the status of the calibration parameter data is generated; The generated verification value is compared with the pre-stored reference verification value; Based on the comparison results, determine whether there is any deviation in the calibration parameters; When a deviation in the calibration parameters is detected, the deviation information of the calibration parameters is output; and Based on the deviation information of the calibration parameters, perform the calibration parameter recovery operation.

[0006] Secondly, this application also discloses an electronic product fault detection and error correction system for smart terminal devices, the system comprising: The instruction receiving module is used to receive batch switching instructions or determine whether the device is in a preset idle period and to initiate the integrity verification process of the calibration parameters stored inside the smart terminal device. The parameter reading module is used to read the calibration parameters stored inside the smart terminal device during the integrity verification process. The verification value generation module is used to generate a verification value that reflects the status of the calibration parameter data based on the read calibration parameters. The verification value comparison module is used to compare the generated verification value with the pre-stored reference verification value; The deviation judgment module is used to determine whether there is a deviation in the calibration parameters based on the comparison results; The information output module is used to output deviation information of the calibration parameters when a deviation is detected; and The recovery operation module is used to perform a recovery operation on the calibration parameters based on the deviation information of the calibration parameters.

[0007] Compared with the prior art, this application has at least the following beneficial effects: This application can proactively and promptly detect potential deviations in the internal calibration parameters of smart terminal devices, especially during batch switching or idle periods, avoiding the lag of traditional passive detection and enabling targeted recovery operations. It effectively solves the ghost failure problem caused by the accumulation of calibration parameter errors, and improves the stability and reliability of the device. Attached Figure Description

[0008] Figure 1This is a flowchart illustrating a method for fault detection and error correction of electronic products provided in this application.

[0009] Figure 2 This is a schematic diagram of the structure of an electronic product fault detection and error correction system provided in this application. Detailed Implementation

[0010] The technical solutions in this application will now be clearly and completely described in conjunction with the accompanying drawings.

[0011] This application proposes a method for fault detection and error correction in electronic products, such as... Figure 1 As shown, it includes the following steps: Upon receiving a batch switching instruction or when the device is in a preset idle period, initiate the integrity verification process for the calibration parameters stored internally in the smart terminal device. In the integrity verification process, the calibration parameters stored inside the smart terminal device are read; Based on the read calibration parameters, a verification value reflecting the status of the calibration parameter data is generated; The generated verification value is compared with the pre-stored reference verification value; Based on the comparison results, determine whether there is any deviation in the calibration parameters; When a deviation in the calibration parameters is detected, the deviation information of the calibration parameters is output; and Based on the deviation information of the calibration parameters, perform the calibration parameter recovery operation.

[0012] This application proactively triggers the integrity verification of calibration parameters at specific times (batch switching or idle periods), and combines this with a verification value comparison mechanism to promptly detect and correct potential deviations in calibration parameters, thereby effectively avoiding ghost failures and significantly improving the stability and reliability of smart terminal devices.

[0013] The intelligent terminal devices mentioned in this application refer to electronic devices with certain computing, storage, and communication capabilities, such as industrial controllers, intelligent sensors, and edge computing devices. These devices internally store calibration parameters for normal operation and performance optimization. These calibration parameters can include sensor zero-point offset, gain coefficient, linearity compensation value, communication protocol configuration, and motor control parameters, which directly affect the measurement accuracy, control effect, and overall performance of the equipment. Batch switching instructions refer to instructions issued by the host system or operators during the production process when material batches, product models, or production processes change. These instructions are usually accompanied by the reconfiguration of internal equipment parameters. Preset idle periods refer to time periods when the equipment has low workload or no critical tasks are executed, such as during non-peak production periods, maintenance windows, or system self-test cycles.

[0014] Upon receiving a batch switching command or when the device is in a preset idle period, an integrity verification process for the calibration parameters stored internally in the smart terminal device is initiated. This step can be implemented in several ways: the command parsing module inside the smart terminal device continuously listens for batch switching commands from the host system; once the command is received, it sends a signal to the system's core control unit to initiate the verification process. Alternatively, the clock management module and task scheduling module inside the smart terminal device can work together to automatically trigger the integrity verification process for the calibration parameters when the current time is detected to be in a preset idle period (e.g., 2 AM to 4 AM daily) and no high-priority tasks are running.

[0015] In the integrity verification process, calibration parameters stored internally in the smart terminal device are read. This step can be implemented in several ways: the system can directly access the specific memory address storing the calibration parameters through the memory management unit and load the data into the processor register or cache. Alternatively, the calibration parameter configuration file stored in non-volatile memory (such as flash memory or EEPROM) can be read through a file system interface.

[0016] Based on the read calibration parameters, a check value reflecting the status of the calibration parameter data is generated. This step can be implemented in several ways: First, a Cyclic Redundancy Check (CRC) algorithm is used to calculate a fixed-length check code for all read calibration parameter data blocks. Second, a hash algorithm (such as SHA-256) is used to process the calibration parameter data and generate a unique digital fingerprint as the check value.

[0017] The generated checksum is compared with a pre-stored reference checksum. This step can be implemented by comparing the currently generated checksum bit by bit with the reference checksum stored in the device's secure storage area (such as a hardware security module or encrypted storage area). Alternatively, software logic can be used to input the two checksums into a comparator and determine whether they are completely identical.

[0018] Based on the comparison results, determine whether the calibration parameters deviate. This step can be implemented in the following ways: if the currently generated verification value is inconsistent with the reference verification value, then determine that the calibration parameters have deviated. Alternatively, if the comparison results show that the two verification values ​​are different, the system immediately marks the calibration parameters as deviating.

[0019] When a deviation from the calibration parameters is detected, the deviation information is output. This step can be implemented by: displaying a deviation warning message through the device's user interface (such as an LCD screen), indicating the specific parameter type or area of ​​deviation; or by uploading the deviation information to a higher-level monitoring system or SCADA system via a communication interface (such as Modbus or Ethernet / IP) to trigger a remote alarm.

[0020] Based on the deviation information of the calibration parameters, perform a calibration parameter recovery operation. This step can be implemented in several ways: loading preset, verified default calibration parameters from the device's internal backup storage area to overwrite the currently deviated parameters; or, by communicating with a host system to request the download of the latest, correct calibration parameter file and writing it to the device's internal storage.

[0021] This application's method for fault detection and error correction in electronic products proactively initiates an integrity verification process for calibration parameters stored internally in the smart terminal device when receiving a batch switching command or during a preset idle period, thus avoiding the lag in fault detection in traditional methods. In the integrity verification process, calibration parameters are read and a verification value is generated, which is then compared with a pre-stored reference verification value to accurately determine if there is a deviation in the calibration parameters. When a deviation is detected, the system outputs deviation information and performs a calibration parameter recovery operation, thereby promptly correcting potential errors and effectively preventing ghost faults caused by cumulative deviations in calibration parameters. This method can not only identify deviations caused by external factors such as sensor physical aging or drift, but more importantly, it can detect and correct the cumulative problem of minor modifications to calibration parameters caused by internal logic errors such as firmware defects. This is difficult to achieve with traditional fault detection methods based on signal smoothing processing. Through this proactive and comprehensive verification and correction mechanism, this application significantly improves the operational stability and data reliability of smart terminal devices, reduces maintenance costs, and avoids the impact on production efficiency caused by recurring faults.

[0022] Compared to existing technologies, the core innovation of this application lies in its proactive and preventative calibration parameter integrity verification mechanism. Existing technologies generally employ signal smoothing techniques such as moving average calculations to eliminate short-term noise. However, this method can mask the true fault when faced with systematic and cumulative calibration parameter deviations, leading to prolonged fault latency and making it impossible to trace the root cause. This application, on the other hand, triggers integrity verification of the calibration parameters themselves at critical moments (batch switching or idle periods), directly detecting the accuracy of the parameter data, rather than relying solely on external signal changes. This method can effectively identify and correct the cumulative problem of minor calibration parameter modifications caused by internal logic errors such as firmware defects, thereby eradicating ghost faults. Furthermore, when a calibration parameter deviation is detected, this application can output detailed deviation information and perform targeted recovery operations, which is more accurate and efficient than the single error correction strategy in existing technologies. Therefore, this application demonstrates significant progress in terms of the timeliness, accuracy, and effectiveness of fault detection and error correction.

[0023] In practical applications, calibration parameters typically comprise multiple sub-parameters with different functions and importance. If only a single check value is generated to reflect the overall state of all calibration parameters, it may be impossible to accurately identify which specific calibration parameter has deviated, or to differentiate the handling based on the importance of different parameters. This could lead to difficulties in effectively locating the root cause when a deviation is detected, or to treating deviations in non-critical parameters in the same way during recovery operations, thereby reducing the efficiency and accuracy of fault detection and error correction.

[0024] In this regard, this application further proposes that the step of generating a verification value reflecting the status of the calibration parameter data based on the read calibration parameters includes: The read calibration parameters are divided into multiple preset parameter blocks, and each parameter block corresponds to a pre-stored calibration sub-parameter with a specific function or importance; Each parameter block is independently checked to obtain multiple block check values. The check values ​​of multiple blocks are compared with their respective pre-stored calibration sub-parameters; Based on the comparison results, determine whether there is any deviation in each parameter block; and Based on the deviation status of each parameter block and its preset importance, a verification value is generated that comprehensively reflects the overall data status of the calibration parameters.

[0025] Specifically, dividing the read calibration parameters into multiple preset parameter blocks refers to dividing the calibration parameters into several independent, manageable units based on their logical function, data type, storage location, or degree of impact on device performance. For example, the calibration parameters of a smart terminal device can be divided into display calibration blocks, camera calibration blocks, sensor calibration blocks, power management calibration blocks, etc. Each parameter block contains one or a group of calibration sub-parameters, which are functionally closely related and may have specific functions or importance. For example, the camera calibration block may contain white balance parameters, focus parameters, distortion correction parameters, etc., which are crucial to image quality.

[0026] The independent verification value calculation for each parameter block refers to using an independent verification algorithm (such as cyclic redundancy check, MD5 hash, or simple XOR check) to calculate the data integrity verification value for each divided parameter block, thereby obtaining multiple block verification values. These block verification values ​​can independently reflect the data status of their respective parameter blocks.

[0027] Furthermore, comparing multiple block verification values ​​with their respective pre-stored calibration sub-parameters means comparing each currently calculated block verification value with the reference verification value pre-stored under normal device operation, corresponding to that parameter block. Here, the pre-stored calibration sub-parameters can be understood as the reference verification value for each parameter block.

[0028] Therefore, based on the comparison results, determining whether each parameter block has a deviation means that if the current check value of a certain block is inconsistent with its calibration sub-parameter, then the parameter block is judged to have a data deviation. This block-by-block judgment method can accurately locate the specific parameter block where the deviation has occurred.

[0029] Finally, based on the deviation status of each parameter block and its preset importance, a comprehensive verification value reflecting the overall data status of the calibration parameters is generated. This means that after determining which parameter blocks have deviations, the system comprehensively evaluates these deviations by combining the pre-set importance level of each block (e.g., critical parameter blocks, general parameter blocks, and non-critical parameter blocks). For example, different weights can be assigned to blocks of different importance levels, and then a comprehensive verification value is calculated based on the number of deviating blocks, their importance weights, and the degree of deviation. This verification value can more comprehensively and accurately reflect the overall data status of the calibration parameters and highlight the impact of deviations in critical parameters.

[0030] This application addresses the limitation of a single verification value being insufficient to accurately pinpoint the root cause of a problem by refining calibration parameters into multiple parameter blocks and performing independent verification and deviation assessment on each block. Specifically, when the calibration parameters stored internally in the smart terminal device are read, they are first logically divided into parameter blocks with specific functions or importance. Then, a verification value is independently calculated for each parameter block and compared with a preset reference value, thereby accurately identifying which parameter block(s) have deviated from their data. This refined detection mechanism allows the system to not only know that a deviation exists but also pinpoint its location. Furthermore, by combining the preset importance of each parameter block, the deviation status is weighted to generate a comprehensive verification value. This value not only reflects the overall deviation situation but also highlights the impact of deviations in key parameter blocks on the overall performance and risk of the device, thus providing more instructive information for subsequent fault diagnosis and recovery operations.

[0031] Through the above technical solutions, this application can significantly improve the accuracy and efficiency of fault detection and error correction in electronic products. Firstly, by dividing and independently verifying parameter blocks, precise location of calibration parameter deviations can be achieved, avoiding blind restoration of the entire calibration parameter set, thus saving system resources and time. Secondly, by introducing an importance assessment of parameter blocks, the system can respond differently based on the severity and scope of the deviation. For example, it can prioritize handling deviations in critical parameter blocks or only restore specific affected blocks, avoiding unnecessary global restoration operations. Therefore, this application not only improves the diagnostic capabilities of fault detection but also optimizes the error correction strategy, enabling smart terminal devices to operate more stably and reliably, and extending their service life.

[0032] In practical applications, smart terminal devices often have multiple operating modes. The impact of each calibration parameter block on device performance and risk may vary significantly depending on the mode. Using only a single preset importance configuration may fail to accurately assess the deviation risk of calibration parameters in a specific operating mode, thus affecting the accuracy and timeliness of fault detection.

[0033] In response, this application further proposes a more refined method for generating check values, which dynamically adjusts the importance configuration of parameter blocks to adapt to the needs of smart terminal devices in different operating modes.

[0034] The steps described above for generating a comprehensive verification value that reflects the overall data status of the calibration parameters based on the deviation status of each parameter block and its preset importance include: Inside the smart terminal device, multiple sets of parameter block importance configurations are pre-configured. Each set of importance configurations includes a weight value for each parameter block. The weight value reflects the degree of impact of each parameter block on the overall performance and risk of the device under a specific operating mode. When a smart terminal device receives an operation mode switching command, it selects and activates the parameter block importance configuration that matches the current operation mode from a pre-configured set of parameter block importance configurations based on the mode identifier contained in the command. Based on the weight values ​​corresponding to the importance configuration of the activated parameter blocks, the deviation status of each parameter block is weighted to obtain a verification value that comprehensively reflects the overall data status of the calibration parameters.

[0035] Specifically, the multiple sets of parameter block importance configurations refer to a set of parameter block importance settings pre-stored within the smart terminal device, customized for different operating modes. Each set of importance configurations contains a series of weight values, which are assigned to each parameter block to quantify the impact of that parameter block on the overall performance, functional stability, and potential risks of the device in a specific operating mode. For example, in high-performance mode, parameter blocks related to processor frequency and memory management may be assigned higher weights, while in low-power mode, parameter blocks related to display brightness and network connectivity may be assigned higher weights. Operating mode switching instructions can be understood as signals triggered by user operations, application requests, or internal system events, instructing the smart terminal device to change its current operating state. These instructions typically contain a mode identifier to uniquely identify the target operating mode. When the smart terminal device receives such an instruction, the system, based on the mode identifier, precisely selects and activates the set of configurations that best matches the current operating mode from the pre-configured multiple sets of parameter block importance configurations. Subsequently, the system uses the corresponding weight values ​​in this activated configuration to weight the deviations of each parameter block. This means that the degree of deviation of each parameter block will be combined with its importance (i.e., weight value) in the current operating mode to more accurately reflect its impact on the overall data status of the device, and finally generate a verification value that comprehensively reflects the overall data status of the calibration parameters.

[0036] This application solves the problem that a single preset importance configuration cannot adapt to changing operating modes by pre-configuring multiple sets of parameter block importance settings within the smart terminal device and dynamically selecting and activating the configuration matching the current operating mode based on the operating mode switching command received by the device. Specifically, when the device enters different operating modes, such as switching from standby mode to high-performance gaming mode, the system activates the corresponding parameter block importance configuration based on the mode identifier. This configuration includes weight values ​​for each parameter block, which accurately reflect the degree of impact of each parameter block on the overall performance and potential risks of the device in the current operating mode. Therefore, when weighting the deviation state of each parameter block, the deviation of key parameter blocks in the current mode can be highlighted more accurately, so that the generated comprehensive verification value can more realistically reflect the data state of calibration parameters in a specific operating mode. This dynamic adjustment mechanism ensures the sensitivity and accuracy of fault detection, avoiding over-response to deviations of unimportant parameters in non-critical modes or under-response to deviations of important parameters in critical modes.

[0037] Through the above technical solution, this application can dynamically adjust the importance assessment of calibration parameter blocks according to different operating modes of smart terminal devices, thereby significantly improving the accuracy and adaptability of fault detection. Compared with a scheme using a fixed importance configuration, this application can more accurately identify the parameter deviations that have the greatest impact on device performance in specific operating modes, avoiding false alarms or missed alarms. This not only optimizes resource utilization efficiency and reduces unnecessary recovery operations, but also ensures that potential calibration parameter problems can be detected and corrected in a timely and effective manner during critical operating modes of the device, thereby improving the overall stability of smart terminal devices and user experience.

[0038] In the actual production process of smart terminal devices, factors such as material characteristics, environmental conditions, and production process parameters may be in dynamic flux. If the risk of parameter blocks is assessed solely based on preset importance, it may fail to reflect the actual risk sensitivity in the current production process in a timely and accurate manner. This can lead to inaccurate judgments of calibration parameter deviations, affecting the efficiency and accuracy of fault detection and error correction.

[0039] To this end, this application further proposes a step for generating a verification value that comprehensively reflects the overall data status of the calibration parameters based on the deviation status of each parameter block and its preset importance, including: Continuously collect physical response parameters related to the current production process of intelligent terminal equipment; Analyze the changing trends of the collected physical response parameters to identify minute changes in material or environmental factors; Based on the identified minute changes, assess the actual risk sensitivity of each parameter block; Based on the actual risk sensitivity assessed, adjust the weight values ​​corresponding to each parameter block in the activated parameter block importance configuration; and Based on the adjusted weight values, the deviation status of each parameter block is weighted to obtain a comprehensive verification value that reflects the overall data status of the calibration parameters.

[0040] Specifically, continuously collecting physical response parameters related to the current production process of smart terminal equipment refers to acquiring physical quantity data related to the production environment, material status, or equipment operation in real time during the manufacturing process of smart terminal equipment. These physical response parameters may include, but are not limited to, temperature, humidity, pressure, vibration, current, voltage, and light intensity, with the aim of obtaining objective data reflecting the real-time status of the production process.

[0041] Analyzing the changing trends of the collected physical response parameters to identify minute changes in materials or environmental factors can be understood as performing time-series analysis, statistical analysis, or pattern recognition on the continuously collected physical response parameters to detect any abnormal fluctuations, drifts, or abrupt changes. These trends may indicate small but potentially critical changes in material batches, production environments, or equipment conditions. The aim is to extract potential risk signals that affect the stability of calibration parameters from massive amounts of real-time data.

[0042] Assessing the actual risk sensitivity of each parameter block based on identified minute changes involves analyzing the trends in physical response parameters, combined with a pre-set risk model or expert experience, to determine whether the impact of each calibration parameter block on the overall equipment performance and reliability has changed under current production conditions. For example, if there are abnormal fluctuations in ambient temperature, the risk sensitivity of calibration parameter blocks related to temperature compensation may increase significantly. The aim is to dynamically correlate real-time production status with the importance of calibration parameters.

[0043] Furthermore, adjusting the weight values ​​of each parameter block in the activated parameter block importance configuration based on the actual risk sensitivity assessed means dynamically modifying the weight coefficients used to calculate the comprehensive verification value based on the real-time assessed actual risk sensitivity. For example, for parameter blocks with increased risk sensitivity, their weight values ​​will be increased accordingly to give them a more important role in the comprehensive verification value calculation. The aim is to ensure that the verification value calculation more accurately reflects the real risks in the current production environment.

[0044] Based on the adjusted weight values, the deviation status of each parameter block is weighted to obtain a comprehensive verification value that reflects the overall data status of the calibration parameters. This means that after obtaining the dynamically adjusted weight values, they are applied to the deviation status of each parameter block, and weighted summation or other weighted operations are performed to ultimately generate a verification value that can more accurately and in real time reflect the overall data status of the calibration parameters.

[0045] This application addresses the static nature of calibration parameter importance assessment by introducing continuous collection and analysis of physical response parameters related to the current production process of intelligent terminal equipment. Specifically, when minute changes occur in the production environment, materials, or equipment status, these changes are promptly identified through the changing trends of physical response parameters. Based on these identified minute changes, the system can dynamically assess the actual risk sensitivity of each parameter block, thereby avoiding assessment biases that may result from relying solely on preset importance. By converting this real-time assessed risk sensitivity into weight values ​​and adjusting the weights in the parameter block importance configuration accordingly, the calculation of the comprehensive verification value can more accurately reflect the true risks under current production conditions. For example, when an abnormal increase in production line temperature is detected, the weight of the parameter block related to temperature sensor calibration is dynamically increased. Even if the deviation is small, it will occupy a more significant position in the comprehensive verification value, prompting the system to pay attention to potential fault risks earlier. This dynamic adjustment mechanism ensures that the fault detection and error correction methods can adapt to the constantly changing production environment, improving the early warning capability for potential problems.

[0046] Through the above technical solution, this application enables more accurate and real-time detection of deviations in calibration parameters of electronic products. Since the weight values ​​are dynamically adjusted based on the physical response parameters during actual production, the generated comprehensive verification value more accurately reflects the true data status and potential risks of the calibration parameters. This significantly improves the sensitivity and accuracy of fault detection, reduces false alarms or missed alarms caused by static weight evaluation, and effectively avoids equipment performance degradation or product quality problems due to deviations in calibration parameters. Furthermore, this solution makes the fault detection and error correction method for intelligent terminal devices more environmentally adaptable and robust, better able to cope with the challenges of complex and ever-changing production environments.

[0047] In actual production environments, production processes or material batches may change frequently. If the methods for collecting physical response parameters cannot adapt to these changes in a timely and accurate manner, the collected data may not match the current production status, thereby affecting the accuracy of subsequent risk sensitivity assessments.

[0048] In this regard, this application further proposes that the steps for continuously collecting physical response parameters related to the current production process of the smart terminal device include: Inside the smart terminal device, multiple sets of parameter acquisition configurations are pre-stored. Each set of parameter acquisition configurations includes a list of key physical response parameters for a specific production process or material batch, as well as the corresponding acquisition frequency. When the smart terminal device receives a production process switching instruction or a material batch change instruction, it selects and activates the parameter acquisition configuration that matches the current production process from a pre-stored set of parameter acquisition configurations based on the process or batch identifier contained in the instruction. Based on the parameter list and acquisition frequency specified in the activated parameter acquisition configuration, adjust the sensor data acquisition module inside the smart terminal device to continuously acquire physical response parameters related to the current production process of the smart terminal device.

[0049] Specifically, multiple sets of parameter acquisition configurations can be pre-stored within the intelligent terminal device. Each set of configurations is designed for a specific production process or material batch, containing a list of key physical response parameters for that particular process or batch, along with the corresponding acquisition frequency. For example, the key physical response parameters (such as temperature, humidity, pressure, current, and voltage) and their optimal acquisition frequencies may differ for different product models or production lines. When the intelligent terminal device receives a production process switching command or a material batch change command, the command includes a pattern identifier indicating the current production process or material batch. Based on this identifier, the system selects and activates the parameter acquisition configuration that matches the current production process from the pre-stored multiple sets of configurations. Once the corresponding parameter acquisition configuration is activated, the sensor data acquisition module within the intelligent terminal device adjusts according to the parameter list and acquisition frequency specified in the configuration. This means that the sensors will only acquire the most relevant parameters required for the current production process and at the most appropriate frequency, thereby ensuring the validity and real-time nature of the data.

[0050] This application's solution addresses the limitations of traditional fixed or universal data acquisition methods in variable production environments by introducing a dynamic parameter acquisition configuration management mechanism. Specifically, by pre-storing parameter acquisition configurations for different production processes or material batches, and upon receiving a corresponding switching command, the solution dynamically selects and activates the matching configuration based on the identifier in the command. This allows the sensor data acquisition module to adjust its acquisition strategy in real time. This ensures that, under any production condition, the acquired physical response parameters are critical data highly relevant to the current production process, avoiding the risk of acquiring irrelevant data or missing crucial data. This adaptive acquisition method provides a more accurate and reliable data foundation for subsequent analysis of physical response parameter change trends, identification of minute changes in material or environmental factors, and assessment of the actual risk sensitivity of each parameter block.

[0051] The above technical solution significantly improves the targeting and accuracy of physical response parameter acquisition, enabling it to flexibly adapt to dynamic changes in intelligent terminal equipment under different production processes or material batches. This not only avoids resource waste caused by collecting irrelevant data, but more importantly, ensures the validity and real-time nature of the collected data, thus providing high-quality input for subsequent risk sensitivity assessment and weight adjustment. Consequently, it allows for more precise identification of minute changes in the production process, more accurate assessment of the actual risk sensitivity of each parameter block, and ultimately improves the reliability of overall calibration parameter data status verification and the timeliness and accuracy of fault detection.

[0052] Suppose a smart terminal device needs to adapt to two different production processes: Process A (e.g., for producing high-end models) and Process B (e.g., for producing standard models). For Process A, a set of parameter acquisition configurations is pre-configured, including high-frequency acquisition requirements for temperature, humidity, specific voltages, and currents. For Process B, another set of parameter acquisition configurations is configured, which may only focus on temperature and pressure, with a relatively lower acquisition frequency. When the production line switches from Process A to Process B, the smart terminal device receives a production process switch command containing an identifier for Process B. Based on this identifier, the system automatically selects and activates parameter acquisition configuration B. Subsequently, the internal sensor data acquisition module immediately adjusts, ceasing the acquisition of voltage and current parameters specific to Process A, and acquiring temperature and pressure data at the lower frequency specified in configuration B. The reverse is also true. In this way, it ensures that the sensor data acquisition is always precisely matched to the current production process, thus providing the most relevant and effective data support for subsequent calibration parameter integrity verification.

[0053] In practical applications, sensors used to acquire physical response parameters may be affected by various factors, such as environmental interference, sensor aging, or intermittent failures, leading to abnormal patterns in their output data, such as occasional spikes, periodic fluctuations, or persistent drift. If these inherent sensor abnormalities are not addressed, the analysis of trends in physical response parameters may be distorted, affecting the accuracy of assessing the actual risk sensitivity of each parameter block. Ultimately, the generated comprehensive verification value may fail to accurately reflect the overall data status of the calibration parameters, thereby reducing the reliability of fault detection and error correction.

[0054] In this regard, this application further proposes a step for continuously collecting physical response parameters related to the current production process of the smart terminal device, including: Inside the smart terminal device, an independent monitoring unit is configured for each physical response parameter. The monitoring unit includes a data buffer and an abnormal pattern recognition module. The data buffer is used to store continuously acquired values ​​of physical response parameters; The abnormal pattern recognition module is used to analyze the continuously acquired values ​​of physical response parameters stored in the data buffer and identify whether there are preset abnormal patterns of the sensor itself. Abnormal patterns include occasional spikes, periodic fluctuations, or continuous drifts. When the abnormal pattern recognition module identifies an abnormal pattern in the sensor itself, it marks the abnormal data and initiates a data correction process. The data correction process compensates for or removes the abnormal data according to the type of abnormal pattern. The calibrated physical response parameters are used as the true physical response parameters reflecting the current production process of the smart terminal equipment, and the true physical response parameters are continuously collected.

[0055] Specifically, for each physical response parameter that needs to be monitored within the smart terminal device, such as temperature, humidity, voltage, current, or vibration, an independent monitoring unit is configured. This monitoring unit is designed to contain two core components: a data buffer and an anomaly pattern recognition module. The data buffer is used to temporarily store the physical response parameter values ​​continuously collected by the sensors. For example, it can be a first-in-first-out (FIFO) queue or a circular buffer to store sensor data from a recent period, allowing the anomaly pattern recognition module to perform historical data analysis.

[0056] Furthermore, the anomaly pattern recognition module is configured to perform real-time or near-real-time analysis on the continuously acquired values ​​of physical response parameters stored in the data buffer. Its core function is to identify whether preset sensor-specific anomaly patterns exist within these data. These anomaly patterns can specifically include: sporadic spikes, where data values ​​suddenly and significantly increase or decrease within a short period, then quickly return to normal; periodic fluctuations, where data values ​​exhibit regular periodic changes, but their amplitude or frequency does not match the fluctuation pattern under normal operating conditions; and persistent drift, where data values ​​slowly and continuously deviate from their normal baseline or trend over a relatively long period. The anomaly pattern recognition module can be implemented using various algorithms, such as statistical methods (e.g., Z-score, moving average), machine learning models (e.g., support vector machines, neural networks), or rule-based expert systems.

[0057] When the anomaly pattern recognition module successfully identifies an anomaly pattern within the sensor itself, the corresponding anomalous data is immediately marked. Subsequently, a data correction process is initiated. This process performs corresponding correction operations based on the specific type of the identified anomaly pattern. For example, for occasional spikes, median filtering, mean filtering, or directly removing the spike data point followed by interpolation can be used; for periodic fluctuations, Fourier transform can be used for spectral analysis to identify and filter out anomalous frequency components, or adaptive filtering techniques can be used for compensation; for persistent drift, the drift amount can be calculated and linear or nonlinear compensation can be performed by comparing it with historical normal data trends. Through these correction operations, the anomalous data is corrected, thereby obtaining physical response parameters that more closely approximate the actual situation.

[0058] Ultimately, the physical response parameters, after undergoing the data correction process, are considered to accurately reflect the current production process of the smart terminal equipment. This corrected data will be continuously collected and used for subsequent analysis, such as analyzing trends, assessing risk sensitivity, and adjusting weights in the aforementioned scheme.

[0059] This application's solution effectively addresses the issue of inaccurate data caused by abnormal sensor patterns by introducing an independent monitoring unit into the continuous acquisition of physical response parameters. Specifically, each physical response parameter is equipped with a data buffer and an anomaly pattern recognition module. The data buffer stores continuous acquisition values, providing sufficient data samples for analysis by the anomaly pattern recognition module. The anomaly pattern recognition module, based on preset algorithms and models, accurately identifies abnormal sensor patterns such as occasional spikes, periodic fluctuations, or persistent drifts. Once an anomaly is identified, the data correction process compensates for or removes the data according to the anomaly type, thereby transforming the raw, potentially biased sensor data into corrected physical response parameters that are closer to the true values. This preprocessing and correction of the raw acquisition data ensures the accuracy and reliability of the physical response parameters upon which subsequent analysis relies. This results in more accurate identification of minute changes in materials or environmental factors, more accurate assessment of the actual risk sensitivity of each parameter block, and ultimately ensures that the comprehensive verification value truly and effectively reflects the overall data status of the calibration parameters.

[0060] Through the above technical solution, this application can significantly improve the robustness and accuracy of fault detection and error correction methods for intelligent terminal equipment. By introducing anomaly pattern recognition and data correction mechanisms during the physical response parameter acquisition stage, data distortion caused by sensor anomalies (such as sporadic spikes, periodic fluctuations, or continuous drift) is effectively avoided. This ensures the accuracy of subsequent analysis of physical response parameter change trends, making the identification of minute changes in materials or environmental factors more reliable, thereby improving the accuracy of the actual risk sensitivity assessment of each parameter block. Consequently, the weight values ​​of parameter blocks can be adjusted more accurately, so that the final comprehensive verification value can more realistically and effectively reflect the overall data status of calibration parameters, thus significantly improving the sensitivity of fault detection and the reliability of error correction, reducing false alarm and false negative rates, and providing a more solid technical guarantee for the stable operation of intelligent terminal equipment.

[0061] The following is a specific example to illustrate this.

[0062] Suppose that during the production process of smart terminal devices, a critical temperature sensor needs to be continuously monitored to assess the risk sensitivity of a certain calibration parameter block. Without the solution proposed in this application, if the temperature sensor suddenly generates an intermittent spike due to environmental electromagnetic interference, or if its readings continuously drift due to sensor aging, then the trend analysis based on these inaccurate temperature data will be incorrect. This could lead to inaccurate risk sensitivity assessment of the relevant calibration parameter block, thereby affecting the final calibration value.

[0063] After adopting the scheme of this application, the temperature sensor data is sent to its independent monitoring unit. The data buffer continuously stores the acquired temperature values. The anomaly pattern recognition module analyzes these continuously acquired values ​​in real time. For example, when the temperature data is detected to suddenly jump from 25°C to 80°C and then quickly return to 25°C within a very short period, the anomaly pattern recognition module will identify it as an occasional spike. At this time, the data correction process will be initiated, and the spike data point may be removed and compensated using the average or interpolation of the normal data before and after it. As another example, if the temperature reading is detected to slowly rise from 25°C to 28°C over several hours, while the actual ambient temperature does not change significantly, the anomaly pattern recognition module will identify it as a persistent drift. The data correction process may compensate subsequent data in real time according to a preset drift model, bringing it back to the normal trend.

[0064] In this way, even if the sensor itself malfunctions, the calibrated temperature data can still accurately reflect the actual temperature conditions during production. These calibrated real physical response parameters are then used to analyze trends, assess the actual risk sensitivity of relevant calibration parameter blocks, and adjust weights accordingly to ultimately generate a more reliable and accurate comprehensive verification value. This significantly improves the accuracy and reliability of the entire fault detection and error correction method.

[0065] In some embodiments, the step of the above-described abnormal pattern recognition module for analyzing the continuously acquired values ​​of physical response parameters stored in the data buffer and identifying whether there is a preset sensor-specific abnormal pattern may include: Inside the smart terminal device, multiple sets of abnormal pattern recognition configurations are pre-stored. Each set of abnormal pattern recognition configurations contains a set of recognition parameters for a specific sensor type or abnormal pattern category. When the abnormal pattern recognition module receives data collected from a specific sensor, it selects and activates the abnormal pattern recognition configuration that matches the sensor type from a pre-stored set of abnormal pattern recognition configurations, based on the sensor type. Based on the set of identification parameters specified in the activated anomaly pattern recognition configuration, the continuously collected values ​​of physical response parameters stored in the data buffer are analyzed to identify whether there are any preset sensor-specific anomaly patterns.

[0066] The pre-storing of multiple sets of abnormal pattern recognition configurations refers to the pre-setting and saving of various rule sets or parameter sets for identifying abnormal sensor patterns in the internal storage unit of the smart terminal device. These configurations can be customized according to different sensor characteristics, operating environments, or possible types of abnormal patterns. For example, one set of recognition parameters may be configured for a temperature sensor, while another set may be configured for a pressure sensor, or different recognition thresholds and algorithm parameters may be configured for sporadic spike patterns and persistent drift patterns. The set of recognition parameters may include, but is not limited to, thresholds, time window sizes, statistical model parameters, frequency analysis ranges, etc. These parameters collectively define how to determine whether the continuously acquired values ​​of a physical response parameter conform to a certain pre-set abnormal pattern.

[0067] Furthermore, when the anomaly pattern recognition module receives data from a specific sensor, it selects and activates the anomaly pattern recognition configuration that matches the sensor type from a pre-stored set of configurations. This means that before processing the sensor data, the anomaly pattern recognition module first identifies the sensor type from which the data originates. Based on this type information, the system dynamically selects the most suitable configuration for anomaly recognition from a pre-set configuration library. This dynamic selection mechanism ensures the targeted nature and effectiveness of the recognition process. For example, when data is received from an accelerometer, the system activates the recognition configuration for accelerometer anomaly patterns (such as high-frequency noise or zero-point drift).

[0068] Therefore, the solution in this application presets and dynamically selects corresponding recognition configurations for different sensor types or anomaly pattern categories. This enables the anomaly pattern recognition module to analyze the specific sensor characteristics and potential abnormal behavior patterns using the most suitable recognition parameters and algorithms. This mechanism avoids misjudgments or omissions that may occur when using a single general recognition method, thereby improving the accuracy and robustness of anomaly pattern recognition. For example, some sensors may exhibit normal periodic fluctuations under specific operating conditions, which may be misjudged as anomalies if a general threshold is used; however, by using a matched recognition configuration, anomalies exceeding the normal fluctuation range can be accurately identified.

[0069] Through the above technical solution, this application can significantly improve the accuracy and adaptability of sensor abnormal pattern recognition. By configuring a dedicated set of recognition parameters for different sensor types or abnormal pattern categories, the system can more precisely distinguish between normal fluctuations and genuine anomalies, effectively reducing false alarm and false negative rates. This customized recognition method enables intelligent terminal devices to more reliably monitor sensor status in complex production environments, providing a more accurate basis for subsequent data correction and fault detection, thereby ensuring the overall effectiveness of electronic product fault detection and error correction methods.

[0070] In practical applications, sensor data may be affected by a variety of complex factors, such as environmental noise, instantaneous disturbances in the production process, or the aging of the sensor itself. These factors may make the characteristics of abnormal patterns unclear or confused with other normal fluctuations, making it difficult for traditional abnormal pattern recognition methods based on simple thresholds or rules to accurately and robustly identify various types of abnormal patterns of the sensor itself, such as occasional spikes, periodic fluctuations, or continuous drifts. This may lead to false alarms or missed alarms, affecting the accuracy and timeliness of fault detection.

[0071] In response, this application further proposes the aforementioned abnormal pattern recognition module, which analyzes the continuously acquired values ​​of physical response parameters stored in the data buffer and identifies whether there are preset sensor-specific abnormal patterns. The steps include: Inside the smart terminal device, a time series analysis module is configured for each physical response parameter; The time series analysis module performs time series decomposition on the continuously collected values ​​of physical response parameters stored in the data buffer, separating out trend components, periodic components, and residual components. The time series analysis module performs statistical characteristic analysis on the residual components, and calculates the fluctuation range and deviation of the residual components. When the fluctuation range or deviation of the residual component exceeds the preset threshold, it is determined that the physical response parameter has an occasional spike or a continuous drift mode. The time series analysis module performs spectral analysis on the periodic components to identify the frequency characteristics of whether a preset periodic fluctuation pattern exists. When the spectrum analysis results show that there is a frequency characteristic of a preset periodic fluctuation pattern, it is determined that the physical response parameters have a periodic fluctuation pattern. The time series analysis module compares the trend components with a preset normal trend pattern to determine whether the trend components deviate significantly from the normal trend pattern. When the trend component deviates significantly from the normal trend pattern, it is determined that the physical response parameter exhibits a persistent drift pattern.

[0072] Specifically, the time series analysis module can be understood as a data processing unit configured to execute complex time series data analysis algorithms. This module can be implemented using software algorithms on a dedicated processor, digital signal processor (DSP), or general-purpose microcontroller (MCU) within a smart terminal device. Its purpose is to reveal hidden patterns and anomalies in the raw data collected by sensors through in-depth analysis.

[0073] Time series decomposition refers to the process of breaking down continuously acquired values ​​of physical response parameters into trend components, periodic components, and residual components. The trend component reflects the long-term trend of the data, such as the drift caused by sensor aging over time; the periodic component reflects recurring patterns in the data, such as regular fluctuations caused by periodic equipment operation or environmental factors; and the residual component represents random fluctuations or transient anomalies after removing the influence of trends and periods. This decomposition can be achieved using methods such as moving averages, exponential smoothing, or more complex seasonal trend decomposition (STL) algorithms.

[0074] Statistical analysis of the residual components is performed to calculate their fluctuation range and deviation, aiming to identify sporadic spikes and persistent drift. The fluctuation range can refer to the difference between the maximum and minimum values ​​of the residual component, or its standard deviation; the deviation can refer to the deviation of the residual component from the zero mean. When these statistics exceed preset thresholds, it indicates an anomaly. For example, a sudden spike can cause the residual component to significantly exceed the normal fluctuation range within a short period, while persistent drift may manifest as the residual component deviating from the zero mean for an extended period.

[0075] Furthermore, spectral analysis of the periodic components aims to identify the frequency characteristics of a pre-defined periodic fluctuation pattern. Spectral analysis, such as through a Fast Fourier Transform (FFT), converts a time-domain signal into a frequency-domain signal, revealing the various frequency components contained in the signal and their intensities. If the spectral analysis results show a significant energy concentration at a specific frequency, and this frequency matches a pre-defined periodic fluctuation pattern (e.g., the vibration frequency caused by a malfunction of a rotating component), then a periodic fluctuation pattern can be identified.

[0076] Furthermore, the trend component is compared with a preset normal trend pattern to determine whether there is a significant deviation between the trend component and the normal trend pattern. The normal trend pattern can be a trend curve set through historical data learning or expert experience under normal device operation. If there is a persistent difference between the current trend component and the normal trend pattern that exceeds the allowable range, such as the sensor output value being consistently higher or lower than expected, it indicates the existence of a persistent drift pattern, which may indicate a slow decline in sensor performance or a long-term deviation of calibration parameters.

[0077] This application's solution overcomes the limitations of traditional methods in identifying complex anomaly patterns by introducing a time series analysis module to refine the continuously acquired values ​​of physical response parameters. Specifically, time series decomposition effectively separates different types of changes (trend, periodic, random) in the original signal, making the identification of each anomaly pattern more focused and accurate. Statistical analysis of the residual components effectively captures instantaneous, high-amplitude anomalies such as occasional spikes, as well as slow, cumulative anomalies such as persistent drifts, because these anomalies exhibit significant statistical characteristics in the residual components after removing trend and periodic effects. Simultaneously, spectral analysis of the periodic components accurately identifies periodic fluctuation patterns hidden in noise, which are often associated with specific mechanical faults or environmental interference. Furthermore, comparing trend components with normal trend patterns allows for timely detection of long-term, gradual deviations in sensor or system parameters, crucial for predictive maintenance and preventing potential failures. It is precisely this multi-dimensional, hierarchical analysis method that enables the anomaly pattern recognition module to more comprehensively and accurately identify preset sensor anomaly patterns.

[0078] Through the above technical solution, this application can significantly improve the accuracy and robustness of intelligent terminal devices in identifying abnormal patterns in sensors. Compared to methods that rely solely on simple threshold judgments or single feature recognition, this solution uses time series decomposition to break down complex sensor data into more easily analyzable trend, periodic, and residual components. This allows for the application of the most suitable analysis method for different types of anomalies (occasional spikes, periodic fluctuations, and persistent drifts). Consequently, it can effectively reduce false alarm and false negative rates, ensuring accurate detection of potential sensor faults or calibration parameter deviations at an early stage, providing a reliable basis for subsequent calibration processes. This refined anomaly identification capability is of great significance for ensuring the long-term stable operation of electronic products and improving the quality control level of the production process.

[0079] Suppose that a smart terminal device integrates a temperature sensor during the production process to monitor the temperature of key components in real time. Under normal circumstances, the temperature sensor readings will show a certain trend (e.g., the overall temperature of the equipment rises slightly as production batches proceed), as well as possible periodic fluctuations (e.g., due to the cyclical operation of the heating / cooling system).

[0080] If the temperature sensor exhibits sporadic spikes, such as a sudden surge in readings due to external transient interference, the time series analysis module, after decomposing the residual components, will find that the residual components exceed the preset fluctuation range threshold within a short period of time, thus accurately identifying it as a sporadic spike mode.

[0081] If the sensor exhibits persistent drift, such as due to sensor aging causing its readings to be consistently high or low, the time series analysis module will compare the trend component with the preset normal temperature trend pattern. If a significant deviation is found between the two, it will be determined to be a persistent drift pattern.

[0082] If a component malfunctions, causing regular periodic fluctuations in temperature, such as the temperature rising and falling every so often, the time series analysis module will perform spectral analysis on the periodic components and identify significant frequency features corresponding to the fault frequency, thus determining it as a periodic fluctuation pattern.

[0083] In this way, the time series analysis module can accurately identify different types of sensor anomalies, providing strong technical support for fault detection and error correction of smart terminal devices.

[0084] In actual production processes, transient disturbances may occur, such as brief mechanical vibrations, power fluctuations, or environmental noise. These disturbances are not inherently abnormal to the sensor itself, but the resulting transient data fluctuations may be misinterpreted as abnormalities in the residual components, leading to false alarms or inaccurate fault detection. Failure to address these issues may affect the accuracy of fault detection and the stability of the system.

[0085] In response, this application further proposes the aforementioned time series analysis module, which performs statistical characteristic analysis on the residual components and calculates the fluctuation range and deviation of the residual components, including the following steps: Inside the smart terminal device, a transient disturbance recognition unit is configured for each physical response parameter; The instantaneous disturbance identification unit performs real-time analysis on the residual components to identify whether there are preset instantaneous disturbance patterns in the production process. When the transient disturbance identification unit identifies a transient disturbance pattern in the production process, it adjusts the calculation method or threshold of the fluctuation range and deviation degree of the residual components according to the characteristics of the transient disturbance pattern. Calculate the fluctuation range and deviation of the residual components based on the adjusted calculation method or threshold.

[0086] Specifically, the transient disturbance identification unit can be understood as a dedicated processing module within the intelligent terminal device, configured to monitor and analyze the residual component data stream in real time. This unit can identify whether there are features in the residual components that match transient disturbance patterns in the production process using preset algorithms or models, such as those based on short-time Fourier transform, wavelet analysis, or machine learning classifiers. Transient disturbance patterns in the production process refer to non-continuous, predictable, or explainable data fluctuation patterns caused by transient external or internal factors during normal production, such as transient shocks during equipment startup or shutdown, or brief interferences during specific operation switching. Its purpose is to distinguish these normal disturbances from genuine sensor anomalies or malfunctions.

[0087] When the transient disturbance identification unit detects a transient disturbance pattern in the production process, it dynamically adjusts the calculation method or threshold for the fluctuation range and deviation degree of the residual components based on the specific characteristics of the pattern. For example, if the detected disturbance is a brief, known-amplitude impact, the system can temporarily relax the judgment criteria for residual components exceeding the threshold, or adopt a statistical calculation method that is insensitive to transient spikes, such as the absolute deviation of the median instead of the standard deviation. In practical applications, this adjustment can be a preset set of rules or an adaptive adjustment strategy learned from historical data. The purpose is to avoid misjudging normal transient disturbances as anomalies while ensuring sensitivity to genuine anomalies.

[0088] This application's solution effectively solves the misjudgment problem that may occur in the presence of transient disturbances in traditional methods by introducing a transient disturbance identification unit and dynamically adjusting the calculation method or threshold of the residual component statistical analysis based on the identified transient disturbance patterns in the production process. Specifically, after the time series analysis module decomposes the residual components, the transient disturbance identification unit monitors these residual components in real time in parallel. Once a specific transient disturbance pattern in the production process is identified, such as a brief data spike caused by changes in the external environment or equipment operation, the unit immediately triggers an adjustment mechanism. This adjustment mechanism can modify the calculation logic or judgment threshold of the residual component fluctuation range and deviation degree based on the characteristics of the disturbance pattern, such as its duration, amplitude, or frequency. It is precisely because of this dynamic adaptability that the system can avoid misjudging normal transient disturbances as sensor anomalies or malfunctions, thereby improving the accuracy and robustness of fault detection.

[0089] Through the above technical solution, this application can significantly improve the accuracy and reliability of electronic product fault detection and error correction methods. By introducing a transient disturbance identification unit into the residual component statistical analysis, and dynamically adjusting the calculation method or threshold according to the identified transient disturbance patterns in the production process, the system can effectively distinguish between normal data fluctuations caused by transient disturbances in the production process and genuine sensor anomalies or malfunctions. This avoids false alarms caused by transient disturbances, reduces unnecessary calibration parameter recovery operations, and thus improves the system's operating efficiency and stability. Furthermore, this solution makes the fault detection system more adaptable to complex and changing production environments, ensuring accurate fault diagnosis under various operating conditions.

[0090] The following is a specific example to illustrate this.

[0091] Suppose that on a production line for smart terminal devices, a temperature sensor continuously collects data. A time series analysis module decomposes this temperature data, obtaining trend components, periodic components, and residual components. At a certain moment, an auxiliary heating device on the production line briefly activates, causing a tiny, transient increase in ambient temperature for a very short time, which quickly returns to normal. This transient increase will appear as a brief spike in the residual component of the temperature sensor.

[0092] Without a transient disturbance identification unit, this spike might be misjudged by the time series analysis module as the residual component fluctuation range or deviation exceeding a threshold, thus incorrectly determining that the temperature sensor has an occasional spike anomaly. However, in the solution of this application, the transient disturbance identification unit configured inside the intelligent terminal device analyzes the residual component in real time. This unit pre-stores the characteristics of the transient disturbance mode of the auxiliary heating equipment startup process, such as its typical duration, amplitude range, and frequency of occurrence. When a spike in the residual component is identified that matches the preset mode, the transient disturbance identification unit immediately adjusts the threshold of the residual component statistical analysis. For example, it can temporarily increase the threshold for judging occasional spikes, or use a statistic that is insensitive to transient spikes for calculation. Through this adjustment, even if the residual component has a brief spike, it will not be misjudged as an anomaly, thereby avoiding unnecessary fault alarms and calibration parameter recovery operations. At the same time, for real anomalies not caused by transient disturbances in the production process, the system can still maintain its original sensitivity for accurate detection.

[0093] Based on the same inventive concept, this application discloses an electronic product fault detection and error correction system for smart terminal devices, such as... Figure 2 As shown, the system includes: The instruction receiving module 1 is used to receive batch switching instructions or determine whether the device is in a preset idle period and to initiate the integrity verification process of the calibration parameters stored inside the smart terminal device. Parameter reading module 2 is used to read calibration parameters stored inside the smart terminal device during the integrity verification process; The verification value generation module 3 is used to generate a verification value that reflects the status of the calibration parameter data based on the read calibration parameters. The verification value comparison module 4 is used to compare the generated verification value with the pre-stored reference verification value; Deviation judgment module 5 is used to determine whether there is a deviation in the calibration parameters based on the comparison results; Information output module 6 is used to output deviation information of the calibration parameters when a deviation is detected; and The recovery operation module 7 is used to perform a recovery operation on the calibration parameters based on the deviation information of the calibration parameters.

[0094] The system provided in this application can automate and periodically detect and correct the calibration parameters of smart terminal devices, effectively solving the ghost failure problem caused by the accumulation of calibration parameter errors, improving the stability and reliability of the equipment, and reducing manual maintenance costs.

[0095] The above are merely embodiments of this application and are not intended to limit the scope of protection of this application.

Claims

1. A method for fault detection and error correction in electronic products, used in smart terminal devices, characterized in that, include: Upon receiving a batch switching instruction or when the device is in a preset idle period, the integrity verification process for the calibration parameters stored internally in the smart terminal device is initiated. In the integrity verification process, the calibration parameters stored inside the smart terminal device are read; Based on the read calibration parameters, a verification value reflecting the status of the calibration parameter data is generated; The generated verification value is compared with the pre-stored reference verification value; Based on the comparison results, determine whether the calibration parameters deviate. When it is determined that the calibration parameter deviates, the deviation information of the calibration parameter is output; as well as Based on the deviation information of the calibration parameters, a calibration parameter recovery operation is performed.

2. The method for fault detection and error correction of electronic products according to claim 1, characterized in that, The step of generating a verification value reflecting the state of the calibration parameter data based on the read calibration parameters includes: The read calibration parameters are divided into multiple preset parameter blocks, and each parameter block corresponds to a pre-stored calibration sub-parameter with a specific function or importance; Each parameter block is independently checked to obtain multiple block check values. The multiple block verification values ​​are compared with their respective pre-stored calibration sub-parameters; Based on the comparison results, determine whether there is any deviation in each parameter block; and Based on the deviation status of each parameter block and its preset importance, a verification value is generated that comprehensively reflects the overall data status of the calibration parameters.

3. The method for fault detection and error correction of electronic products according to claim 2, characterized in that, The step of generating a verification value that comprehensively reflects the overall data status of the calibration parameters based on the deviation status of each parameter block and its preset importance includes: Inside the smart terminal device, multiple sets of parameter block importance configurations are pre-configured. Each set of importance configurations includes a weight value for each parameter block. The weight value reflects the degree of impact of each parameter block on the overall performance and risk of the device under a specific operating mode. When the smart terminal device receives an operation mode switching instruction, it selects and activates a parameter block importance configuration that matches the current operation mode from the pre-configured multiple sets of parameter block importance configurations according to the mode identifier contained in the instruction. Based on the weight values ​​corresponding to the importance configuration of the activated parameter blocks, the deviation status of each parameter block is weighted to obtain a verification value that comprehensively reflects the overall data status of the calibration parameters.

4. The method for fault detection and error correction of electronic products according to claim 2, characterized in that, The step of generating a verification value that comprehensively reflects the overall data status of the calibration parameters based on the deviation status of each parameter block and its preset importance includes: Continuously collect physical response parameters related to the current production process of the intelligent terminal device; Analyze the changing trends of the collected physical response parameters to identify minute changes in the material or environmental factors; Based on the identified minute changes, assess the actual risk sensitivity of each parameter block; Based on the actual risk sensitivity assessed, adjust the weight values ​​corresponding to each parameter block in the activated parameter block importance configuration; and Based on the adjusted weight values, the deviation status of each parameter block is weighted to obtain a verification value that comprehensively reflects the overall data status of the calibration parameters.

5. The method for fault detection and error correction of electronic products according to claim 4, characterized in that, The step of continuously collecting physical response parameters related to the current production process of the intelligent terminal device includes: The intelligent terminal device has multiple sets of parameter acquisition configurations pre-stored inside. Each set of parameter acquisition configurations includes a list of key physical response parameters for a specific production process or material batch and the corresponding acquisition frequency. When the intelligent terminal device receives a production process switching instruction or a material batch change instruction, it selects and activates a parameter acquisition configuration that matches the current production process from the pre-stored multiple sets of parameter acquisition configurations, based on the process or batch identifier contained in the instruction. Based on the parameter list and acquisition frequency specified in the activated parameter acquisition configuration, the sensor data acquisition module inside the smart terminal device is adjusted to continuously acquire physical response parameters related to the current production process of the smart terminal device.

6. The method for fault detection and error correction of electronic products according to claim 4, characterized in that, The step of continuously collecting physical response parameters related to the current production process of the intelligent terminal device includes: Inside the intelligent terminal device, an independent monitoring unit is configured for each physical response parameter. The monitoring unit includes a data buffer and an abnormal pattern recognition module. The data buffer is used to store the continuously collected values ​​of the physical response parameters; The abnormal pattern recognition module is used to analyze the continuously collected values ​​of physical response parameters stored in the data buffer and identify whether there are preset abnormal patterns of the sensor itself. The abnormal patterns include occasional spikes, periodic fluctuations, or continuous drifts. When the abnormal pattern recognition module identifies an abnormal pattern in the sensor itself, it marks the abnormal data and initiates a data correction process. The data correction process compensates for or removes the abnormal data according to the type of the abnormal pattern. The corrected physical response parameters are used as the true physical response parameters reflecting the current production process of the smart terminal device, and the true physical response parameters are continuously collected.

7. The method for fault detection and error correction of electronic products according to claim 6, characterized in that, The abnormal pattern recognition module is used to analyze the continuously acquired values ​​of physical response parameters stored in the data buffer and identify whether there are preset sensor abnormal patterns. The steps include: Inside the intelligent terminal device, multiple sets of abnormal pattern recognition configurations are pre-stored, and each set of abnormal pattern recognition configurations contains a set of recognition parameters for a specific sensor type or abnormal pattern category. When the abnormal pattern recognition module receives data collected from a specific sensor, it selects and activates an abnormal pattern recognition configuration that matches the sensor type from the pre-stored multiple sets of abnormal pattern recognition configurations, based on the sensor type. Based on the set of identification parameters specified in the activated abnormal pattern recognition configuration, the continuously collected values ​​of physical response parameters stored in the data buffer are analyzed to identify whether there are any preset sensor abnormal patterns.

8. The method for fault detection and error correction of electronic products according to claim 6, characterized in that, The abnormal pattern recognition module is used to analyze the continuously acquired values ​​of physical response parameters stored in the data buffer and identify whether there are preset sensor abnormal patterns. The steps include: Inside the intelligent terminal device, a time series analysis module is configured for each physical response parameter; The time series analysis module performs time series decomposition on the continuously collected values ​​of physical response parameters stored in the data buffer, separating out trend components, periodic components, and residual components. The time series analysis module performs statistical characteristic analysis on the residual components and calculates the fluctuation range and deviation of the residual components. When the fluctuation range or deviation of the residual component exceeds a preset threshold, it is determined that the physical response parameter has an occasional spike or a continuous drift mode. The time series analysis module performs spectral analysis on the periodic components to identify the frequency characteristics of whether a preset periodic fluctuation pattern exists. When the spectrum analysis results show that there is a frequency characteristic of a preset periodic fluctuation pattern, it is determined that the physical response parameter has a periodic fluctuation pattern. The time series analysis module compares the trend component with a preset normal trend pattern to determine whether the trend component deviates significantly from the normal trend pattern. When the trend component deviates significantly from the normal trend pattern, it is determined that the physical response parameter exhibits a persistent drift pattern.

9. The method for fault detection and error correction of electronic products according to claim 8, characterized in that, The time series analysis module performs statistical characteristic analysis on the residual components and calculates the fluctuation range and deviation of the residual components, including the following steps: Inside the intelligent terminal device, a transient disturbance identification unit is configured for each physical response parameter; The instantaneous disturbance identification unit performs real-time analysis on the residual components to identify whether there is a preset instantaneous disturbance pattern in the production process. When the instantaneous disturbance identification unit identifies the instantaneous disturbance pattern of the production process, it adjusts the calculation method or threshold of the fluctuation range and deviation degree of the residual component according to the characteristics of the instantaneous disturbance pattern. The fluctuation range and deviation of the residual component are calculated based on the adjusted calculation method or threshold.

10. An electronic product fault detection and error correction system for smart terminal devices, characterized in that, The system includes: The instruction receiving module is used to receive batch switching instructions or determine whether the device is in a preset idle period, and to initiate the integrity verification process of the calibration parameters stored inside the smart terminal device. The parameter reading module is used to read the calibration parameters stored inside the smart terminal device during the integrity verification process. The verification value generation module is used to generate a verification value that reflects the status of the calibration parameter data based on the read calibration parameters. The verification value comparison module is used to compare the generated verification value with a pre-stored reference verification value; The deviation judgment module is used to determine whether the calibration parameters deviate based on the comparison results; An information output module is used to output deviation information of the calibration parameter when it is determined that there is a deviation; and The recovery operation module is used to perform a recovery operation on the calibration parameters based on the deviation information of the calibration parameters.

Images