Integrated work terminal operation security control method and system

Abstract

The application provides an integrated operation terminal running safety control method and system, and relates to the technical field of integrated operation terminal running safety control. The method comprises the following steps: establishing a first power consumption reference and a first completion time reference of a processor when performing a pre-designed calculation task; monitoring the actual power consumption and the actual completion time of the processor when performing a current calculation task in real time; determining the processor overheating risk according to the actual power consumption, the actual completion time, the first power consumption reference and the first completion time reference; and adjusting the working strategy of the processor according to the processor overheating risk. The integrated operation terminal running safety control method and system provided by the application guarantee the safety and stability of the integrated operation terminal under long-term high-load operation, and significantly improve the reliability and service life of the equipment.

Description

Technical Field

[0001] This application relates to the field of integrated operation terminal operation safety control technology, and in particular to an integrated operation terminal operation safety control method and system. Background Technology

[0002] In modern industrial production environments, integrated work terminals are widely used for the automated monitoring and control of critical processes, aiming to improve production efficiency and ensure operational accuracy. These terminals typically require continuous operation for extended periods, sometimes even weeks or months, performing data acquisition and processing tasks without interruption. Under this sustained high-load operating mode, the processor inside the terminal continuously generates a significant amount of heat. In particular, the thermal grease used to fill the tiny gaps between the processor chip and the heatsink in the processor's cooling module gradually ages due to the heat generated during processor operation and the thermal cycling caused by ambient temperature fluctuations. This aging process manifests as the slow evaporation or separation of the silicone oil component within the thermal grease, causing its filler particles to gradually settle, thus altering the overall structure of the thermal grease and resulting in localized degradation of its thermal conductivity. This localized degradation of the thermal grease's performance directly leads to a decrease in heat transfer efficiency. Heat that should have been quickly dissipated from the processor begins to accumulate locally inside the processor, forming one or more localized hot spots.

[0003] At this point, although the temperature sensor inside the all-in-one terminal may be placed within the overall processor package or a specific location on the motherboard, the detected value reflects the overall temperature of the processor package or the ambient temperature, rather than the actual local hotspot temperature inside the processor core. This means that the "overall temperature" sensed by the system may still be within the normal range, but a tiny area inside the processor chip is actually overheating far beyond the design threshold. Because the overall temperature measured by the system's internal temperature sensor has not reached the preset processor throttling threshold, the system's strategy for adjusting the processor's operating frequency and power consumption based on the processor temperature fails to intervene in time. These strategies typically rely on a preset temperature-frequency adjustment correspondence, i.e., a "mapping table," which aims to reduce heat generation by lowering the processor frequency to protect the processor. However, because the local hotspot problem caused by the aging of thermal paste is not accurately captured by the overall temperature sensor, this mapping table cannot identify and respond to this local overheating phenomenon. Therefore, the processor continues to operate at its normal or near-normal frequency, continuously working in a locally overheated state, which not only accelerates the processor's own performance degradation but also poses a threat to the safe operation of the device. Summary of the Invention

[0004] This application aims to address at least one of the technical problems existing in the prior art. To this end, this application proposes a method and system for safe operation control of an integrated work terminal, which ensures the safety and stability of the integrated work terminal under long-term high-load operation, and significantly improves the reliability and service life of the equipment.

[0005] In a first aspect, the embodiments of this application provide an integrated operation terminal operation security control method, including:

[0006] Establish a first power consumption benchmark and a first completion time benchmark for the processor when executing a preset computing task;

[0007] Real-time monitoring of the processor's actual power consumption and actual completion time while executing the current computing task;

[0008] The processor overheating risk is determined based on the actual power consumption, the actual completion time, the first power consumption benchmark, and the first completion time benchmark.

[0009] Adjust the processor's operating strategy based on the risk of processor overheating.

[0010] According to some embodiments of this application, the step of determining the processor overheating risk based on the actual power consumption and the actual completion time, and the first power consumption reference and the first completion time reference, includes:

[0011] Within a preset time window, the processor is brought into a preset stable working state;

[0012] The ambient temperature, actual power consumption, and actual time to complete the preset task are obtained by the processor under the stable operating state.

[0013] Based on the ambient temperature under the stable operating state, the actual power consumption, and the actual time, the first power consumption reference and the first completion time reference are corrected to obtain the second power consumption reference and the second completion time reference.

[0014] The processor overheating risk is determined based on the second power consumption benchmark and the second completion time benchmark.

[0015] According to some embodiments of this application, the step of determining the processor overheating risk based on the second power consumption reference and the second completion time reference includes:

[0016] Get the current ambient temperature of the processor;

[0017] The criticality of obtaining the current production status of the processor;

[0018] Based on the current ambient temperature, the criticality of the current production condition, the actual power consumption, and the actual completion time, the abnormal power consumption threshold and the abnormal completion time threshold are dynamically adjusted to obtain the corrected abnormal power consumption threshold and the corrected abnormal completion time threshold.

[0019] The processor overheating risk is determined based on the second power consumption benchmark, the second completion time benchmark, the corrected power consumption anomaly threshold, and the corrected completion time anomaly threshold.

[0020] According to some embodiments of this application, the step of adjusting the processor's operating strategy based on the processor overheating risk includes:

[0021] When there is a risk of processor overheating, monitor real-time data from the production process to assess the criticality of the current production conditions to the real-time performance of computing.

[0022] Based on the criticality of the production conditions, the processor's operating strategy is adjusted, wherein...

[0023] When the production condition is of high real-time criticality, the priority of the processor's working strategy is reduced to the lowest level, and the execution of internal processor adjustment actions is restricted.

[0024] When the production condition is of normal real-time criticality, the priority of the processor's working strategy is restored to the normal level.

[0025] According to some embodiments of this application, the step of adjusting the processor operating strategy based on the criticality of the production condition further includes:

[0026] Obtain key control commands and timestamp the key control commands;

[0027] Calculate the actual end-to-end delay of the key control command based on the timestamp to obtain initial delay data;

[0028] The initial delay data is smoothed to obtain smoothed delay data;

[0029] The processor's operating strategy is adjusted based on the smoothed latency data.

[0030] According to some embodiments of this application, the step of calculating the actual end-to-end delay of the critical control instruction based on the timestamp to obtain initial delay data includes:

[0031] When sending critical control commands, a command sending timestamp is generated and embedded in the command data packet;

[0032] When the instruction data packet passes through an intermediate network node, a local timestamp is recorded, and the local timestamp is synchronized with the instruction sending timestamp carried in the instruction data packet using a network time synchronization protocol to obtain the intermediate network node timestamp.

[0033] Upon receiving the instruction data packet, record its local receiving timestamp;

[0034] After the instruction is executed, record the local execution completion timestamp;

[0035] Based on the local reception timestamp, the local execution completion timestamp, the instruction sending timestamp, and the intermediate network node timestamp, the actual end-to-end delay of the critical control instruction is calculated to obtain initial delay data.

[0036] According to some embodiments of this application, the step of smoothing the initial delay data to obtain smoothed delay data includes:

[0037] Obtain the type of the critical control command;

[0038] Select the smoothing algorithm and corresponding parameters according to the type of the key control command;

[0039] The initial delay data is smoothed according to the smoothing algorithm and the corresponding parameters to obtain smoothed delay data.

[0040] According to some embodiments of this application, after the step of smoothing the initial delay data according to the smoothing algorithm and corresponding parameters, the method further includes:

[0041] Monitor the residual fluctuation between the smoothed delay data and the initial delay data;

[0042] Monitor the rate of change of delayed data after smoothing;

[0043] When the residual fluctuation or the rate of change exceeds a preset threshold, the smoothing algorithm or its parameters corresponding to the instruction are adjusted according to the characteristics of the instruction. The residual fluctuation or the rate of change is monitored according to the adjusted smoothing algorithm or its parameters until the residual fluctuation and the rate of change return to the preset threshold range.

[0044] According to some embodiments of this application, when multiple smoothed latency data are close to a preset threshold, the step of adjusting the processor's operating strategy based on the smoothed latency data includes:

[0045] Identify the type of key control command corresponding to the smoothed delayed data;

[0046] Based on the type, obtain a preset weight for the degree of impact of the instruction on production safety;

[0047] Calculate the comprehensive risk value of potential abnormal delays based on the instruction type corresponding to the delay data and its preset weight;

[0048] Based on the comprehensive risk value, potential abnormal delays are classified into risk levels.

[0049] Adjust the processor's operating strategy based on the aforementioned risk classification.

[0050] Secondly, embodiments of this application provide an integrated operational terminal safety control system, comprising:

[0051] The benchmark establishment module establishes a first power consumption benchmark and a first completion time benchmark for the processor when executing a preset computing task;

[0052] The real-time monitoring module is used to monitor the processor's actual power consumption and actual completion time when executing the current computing task.

[0053] The risk assessment module determines the processor overheating risk based on the actual power consumption, the actual completion time, the first power consumption benchmark, and the first completion time benchmark.

[0054] The strategy adjustment module adjusts the processor's operating strategy based on the risk of processor overheating.

[0055] The technical solution according to the embodiments of this application has at least the following beneficial effects:

[0056] This application provides a method for safe operation control of an integrated workstation. By establishing a first power consumption benchmark and a first completion time benchmark for the processor when executing a preset computing task, and by monitoring the actual power consumption and actual completion time of the processor in real time while executing the current computing task, the method can compare the actual operating data with the preset benchmark. Based on this, the application can accurately determine whether the processor has an overheating risk based on the difference between the actual power consumption and actual completion time and the first power consumption benchmark and the first completion time benchmark. Once an overheating risk is identified, the application can dynamically adjust the processor's operating strategy according to the degree of risk. This method overcomes the problem in existing technologies that rely solely on a general temperature sensor and cannot effectively capture local hot spots inside the processor, avoiding the failure to respond in time to local overheating caused by the aging of thermal grease, which leads to processor performance degradation and system instability. Through real-time, dynamic monitoring and strategy adjustment, this application effectively ensures the safety and stability of the integrated workstation under long-term high-load operation, significantly improving the reliability and service life of the equipment. Attached Figure Description

[0057] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0058] Figure 1 A flowchart illustrating an integrated work terminal operation safety control method provided in one embodiment of this application;

[0059] Figure 2 This is a schematic diagram of an integrated operation terminal safety control system provided in one embodiment of this application. Detailed Implementation

[0060] To make the objectives, technical methods, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0061] It should be noted that the meaning of "multiple" (or "more than") in the description of the embodiments of this application refers to two or more, and "greater than," "less than," "exceeding," etc. are understood to exclude the number itself, while "above," "below," "within," etc. are understood to include the number itself. If "first," "second," etc. are used in the description, they are only for the purpose of distinguishing technical features and should not be construed as indicating or implying relative importance or implicitly indicating the number of technical features indicated or the order of the technical features indicated.

[0062] In the description of this application, unless otherwise expressly defined, terms such as "setup," "installation," and "connection" should be interpreted broadly, and those skilled in the art can reasonably determine the specific meaning of the above terms in this application in conjunction with the specific content of the technical solution.

[0063] Based on the above, this application proposes an integrated operation terminal operation safety control method and system, aiming to improve the reliability and accuracy of integrated operation terminal operation safety control.

[0064] See Figure 1 , Figure 1 This is a flowchart illustrating an integrated workstation operation security control method according to an embodiment of this application. The integrated workstation operation security control method provided in this embodiment includes, but is not limited to, steps S110 to S140, which are described below.

[0065] S110. Establish a first power consumption reference and a first completion time reference for the processor when executing a preset computing task;

[0066] S120: Real-time monitoring of the processor's actual power consumption and actual completion time while executing the current computing task;

[0067] S130. Determine the processor overheating risk based on the actual power consumption, actual completion time, first power consumption benchmark, and first completion time benchmark;

[0068] S140. Adjust the processor's operating strategy based on the risk of processor overheating.

[0069] It should be noted that an integrated workstation refers to an industrial-grade device that integrates multiple functions such as computing, control, and communication. It is typically used in automated production lines and smart factories, requiring stable operation over extended periods. The processor is the core computing unit of this terminal, responsible for executing various computational tasks. Power consumption and completion time benchmarks are pre-set reference values ​​for power consumption and time when executing a specific task under normal processor health conditions, used for subsequent comparative analysis. Actual power consumption and actual completion time are the real-time power consumption and time data measured by the processor while executing the current task. Processor overheating risk refers to the possibility that the processor's internal temperature may exceed a safe threshold, potentially caused by localized hotspots or poor overall heat dissipation. Adjusting the operating strategy involves changing parameters such as the processor's operating frequency, voltage, and task scheduling priority to reduce power consumption and heat generation, thereby mitigating or eliminating the overheating risk.

[0070] The core of the integrated operating terminal operation safety control method proposed in this application lies in effectively addressing the local overheating problem caused by the failure of traditional heat dissipation mechanisms through refined monitoring of the processor's operating status and intelligent strategy adjustment.

[0071] First, regarding the feature of "establishing a first power consumption benchmark and a first completion time benchmark for the processor when executing preset computing tasks," the establishment of this benchmark is the basis for subsequent assessment of processor overheating risk. One implementation is to have the processor execute a series of preset, representative computing tasks in a controlled environment after the initial deployment or maintenance of the integrated workstation. During the execution of these tasks, the instantaneous power consumption of the processor is continuously recorded using a high-precision power consumption measurement device (e.g., a digital power meter integrated in the power management unit), and its average power consumption is calculated as the first power consumption benchmark. Simultaneously, the start and end times of each preset task are accurately recorded using a system timer or task scheduler, thereby calculating the task completion time, and its average value is used as the first completion time benchmark. For example, a benchmark test program can be set up, which includes various typical industrial control algorithms (such as PID control, data filtering, image processing, etc.), and the power consumption and completion time are recorded when the processor is running at its rated frequency and voltage while executing these algorithms. Another implementation is that, before the processor leaves the factory, the manufacturer provides power consumption and completion time data obtained under standard test conditions as the initial first power consumption benchmark and first completion time benchmark. This data can be stored in the terminal's non-volatile memory for loading and use during system startup.

[0072] Secondly, regarding the feature of "real-time monitoring of the processor's actual power consumption and actual completion time while executing the current computing task," continuous real-time monitoring of the processor's operating data is necessary to promptly detect abnormal processor states. One implementation involves an integrated power management unit (PMU) within the all-in-one terminal, equipped with a power sensor capable of acquiring instantaneous processor power consumption data at a high sampling rate (e.g., once per millisecond). This data is transmitted via an internal bus to the main control unit for processing, calculating the average actual power consumption within a specific time window. Simultaneously, the operating system or real-time operating system (RTOS) task scheduler records the start and end timestamps of the currently executing computing task, thereby calculating the task's actual completion time. For example, when the processor executes a data acquisition task, the system records the total time elapsed from task initiation to data processing completion and uses this as the actual completion time. Another implementation involves using an external sensor array, such as deploying current and voltage sensors along the processor's power supply path, to measure the processor's input power in real time and use this as the actual power consumption. The task completion time can then be determined by monitoring task-related output signals or status registers.

[0073] Secondly, regarding the feature of "determining processor overheating risk based on actual power consumption, actual completion time, a first power consumption benchmark, and a first completion time benchmark," this is a key step in this application to determine the processor's health status. One implementation involves the system comparing the real-time monitored actual power consumption with a first power consumption benchmark and calculating a power consumption deviation rate. Simultaneously, it compares the actual completion time with a first completion time benchmark and calculates a time deviation rate. When either the power consumption deviation rate or the time deviation rate exceeds a preset threshold, the processor is considered to have an overheating risk. For example, if the actual power consumption is more than 15% higher than the first power consumption benchmark, or the actual completion time is more than 10% longer than the first completion time benchmark, the system will determine that the processor may have an overheating risk. This is because when the processor experiences localized overheating, the leakage current of its internal transistors increases, leading to increased power consumption; simultaneously, overheating also causes a decrease in processor performance, thereby prolonging task completion time. Another implementation involves weighted averaging of the power consumption deviation rate and the time deviation rate to obtain a comprehensive risk index. When this risk index exceeds a preset threshold, the processor is determined to have an overheating risk. For example, based on historical data analysis, different weights can be assigned to power consumption deviation and time deviation to more accurately reflect the risk of overheating.

[0074] Finally, regarding the feature of "adjusting processor operating strategies based on processor overheating risk," once an overheating risk is identified, the system needs to take timely measures to prevent further damage. One implementation is that when an overheating risk is detected, the system immediately triggers a frequency and voltage reduction strategy. Specifically, by adjusting the processor's clock frequency and core voltage, its operating speed and power consumption are reduced, thereby reducing heat generation. For example, the processor frequency can be reduced from 2.0GHz to 1.5GHz, while the core voltage is reduced from 1.2V to 1.0V. This strategy can quickly reduce the processor temperature, but may have some impact on computing performance. Another implementation is that the system dynamically adjusts task scheduling priorities based on the level of overheating risk. For example, the priority of non-critical tasks is reduced, or even suspended, to ensure that critical tasks receive sufficient computing resources and allow the processor more time to dissipate heat. Simultaneously, additional cooling fans or liquid cooling systems can be activated to enhance heat dissipation.

[0075] The integrated operational terminal safety control method proposed in this application establishes a first power consumption benchmark and a first completion time benchmark for the processor when executing a preset computing task, and monitors the actual power consumption and actual completion time of the processor in real time when executing the current computing task, thereby more accurately capturing changes in the processor's operating state. Traditional methods often rely on data from a single temperature sensor, which may not accurately reflect local hot spots inside the processor. This application, by combining the two key indicators of power consumption and task completion time, can more sensitively detect the risk of local overheating of the processor caused by factors such as aging of thermal paste.

[0076] Specifically, when the thermal paste in an all-in-one terminal ages, the heat conduction efficiency inside the processor decreases, leading to a rise in localized temperature. This localized overheating increases the processor's leakage current, resulting in increased actual power consumption. Simultaneously, overheating also affects the processor's transistor performance, causing it to take longer to complete the same computational task, i.e., an extended actual completion time. This application effectively identifies these anomalies by comparing the real-time monitored actual power consumption and actual completion time with preset first power consumption benchmarks and first completion time benchmarks. For example, if the actual power consumption is significantly higher than the first power consumption benchmark, or the actual completion time is significantly longer than the first completion time benchmark, it can be determined that the processor is at risk of overheating.

[0077] Once an overheating risk is identified in the processor, this application adjusts the processor's operating strategy according to the degree of risk. This strategy adjustment is dynamic and intelligent, and can include various methods such as frequency and voltage reduction, task scheduling optimization, and activation of auxiliary cooling. Compared with the single frequency reduction strategy in existing technologies that only rely on temperature thresholds, the strategy adjustment in this application is more flexible and refined. For example, in the case of a minor overheating risk, the scheduling of non-critical tasks can be adjusted first, while in the case of a severe overheating risk, more aggressive frequency and voltage reduction measures can be taken. This risk assessment and strategy adjustment mechanism based on multi-dimensional data analysis can effectively prevent the processor from continuing to operate under localized overheating conditions, thereby extending the processor's lifespan, ensuring the long-term stable operation of the integrated workstation, and significantly improving the reliability and safety of equipment in industrial production environments.

[0078] In one embodiment of this application, regarding the above-mentioned S130, the processor overheating risk is determined based on the actual power consumption, the actual completion time, the first power consumption reference, and the first completion time reference, including but not limited to steps S210 to S240, which are described in turn below.

[0079] S210. Within a preset time window, the processor enters a preset stable working state.

[0080] S220 obtains the processor's ambient temperature, actual power consumption, and actual time to complete a preset task under stable operating conditions.

[0081] S230. Based on the ambient temperature, actual power consumption and actual time under stable working conditions, the first power consumption reference and the first completion time reference are corrected to obtain the second power consumption reference and the second completion time reference.

[0082] S240: Determine the processor overheating risk based on the second power consumption benchmark and the second completion time benchmark.

[0083] Specifically, within a preset time window, the processor is guided into a pre-defined stable operating state. This stable operating state refers to the state in which the processor's power consumption, temperature, and performance are relatively stable when performing specific, repeatable computational tasks. For example, the processor can execute a standardized diagnostic procedure or a light-load task to ensure that its operating states are comparable. The preset time window is designed to provide a sufficiently long period of time for the processor to reach and maintain its stable operating state, thereby obtaining accurate performance data.

[0084] During the processor's stable operation, the system acquires its ambient temperature, actual power consumption, and the actual time taken to complete a preset task. Ambient temperature refers to the air temperature surrounding the processor; actual power consumption refers to the instantaneous or average power consumption of the processor in this stable state; and actual time refers to the time taken for the processor to complete the preset task. These data are key metrics for evaluating the processor's current performance and environmental conditions.

[0085] In practical applications, the initially established first power consumption benchmark and first completion time benchmark are corrected based on the ambient temperature, actual power consumption, and actual time obtained under stable operating conditions, thus obtaining a second power consumption benchmark and a second completion time benchmark. The correction process can employ various algorithms, such as weighted averaging, exponential smoothing, or machine learning-based models, to reflect the processor's true performance baseline under current environmental and operating conditions. The second power consumption benchmark and the second completion time benchmark are dynamically adjusted reference values ​​that more closely reflect the current actual situation, used for subsequent overheating risk assessment.

[0086] Therefore, the determination of processor overheating risk will be based on the revised second power consumption benchmark and second completion time benchmark. This means that the system no longer relies solely on the initial static benchmark, but instead uses a dynamic benchmark that can adapt to environmental changes and processor performance degradation to evaluate the processor's operating status, thereby improving the accuracy of risk assessment.

[0087] In some preferred embodiments, the all-in-one terminal triggers a baseline correction process every 24 hours or when it first reaches a stable operating state after system startup. Specifically, the system suspends non-critical tasks, causing the processor to enter a preset low-load cyclic test mode, such as executing a 5-minute CPU floating-point operation test. During this period, temperature sensors record the processor core temperature and ambient temperature, power sensors record the average power consumption, and the system accurately measures the time required to complete the test task. Assume the initial first power consumption baseline is 50W and the first completion time baseline is 100ms. During a correction process, the system detects a 5°C increase in ambient temperature under stable operating conditions, an actual power consumption of 55W, and an actual completion time of 110ms. Based on this data, the system uses a preset correction algorithm (e.g., a linear regression model based on temperature and power consumption) to correct the first power consumption baseline to 53W (second power consumption baseline) and the first completion time baseline to 105ms (second completion time baseline). Subsequent overheating risk assessments will be based on these corrected second baselines, thus more accurately reflecting the processor's true performance boundaries under current environmental conditions. For example, an overheating risk is only identified when the processor's actual power consumption exceeds 53W plus a dynamic threshold, or when the actual completion time exceeds 105ms plus a dynamic threshold, thus avoiding misjudgments caused by environmental changes.

[0088] This application's solution effectively addresses the potential deviations that may occur in traditional static benchmarks during long-term operation by introducing a periodic benchmark correction mechanism. Specifically, by allowing the processor to enter a stable operating state within a preset time window and acquiring its ambient temperature, actual power consumption, and the actual time to complete a preset task in that state, the true performance of the processor under the current operating environment can be captured. This real-time acquired data is used to dynamically correct the initially established first power consumption benchmark and first completion time benchmark, thereby generating a more timely and accurate second power consumption benchmark and second completion time benchmark. It is precisely because of this dynamically corrected benchmark that subsequent processor overheating risk assessment can more accurately reflect the actual operating status of the processor, avoiding misjudgments or omissions caused by outdated benchmarks.

[0089] Through the above technical solution, this application can significantly improve the accuracy and adaptability of processor overheating risk assessment in integrated workstations. Compared with solutions that rely solely on static benchmarks, this application introduces a dynamic benchmark correction mechanism, enabling the system to adapt in real time to environmental changes, processor aging, and performance drift caused by long-term operation. This not only reduces the possibility of false or missed overheating risks, effectively avoiding the risks of system performance degradation, instability, or even hardware damage caused by processor overheating, but also ensures that the integrated workstation can operate continuously, safely, and efficiently under various complex operating conditions, thus improving the overall system reliability.

[0090] In one embodiment of this application, regarding the above-described S240, determining the processor overheating risk based on the second power consumption reference and the second completion time reference, including but not limited to steps S310 to S340, each step is described below.

[0091] S310: Obtain the current ambient temperature of the processor;

[0092] S320, the criticality of obtaining the current production status of the processor;

[0093] S330: Based on the current ambient temperature, the criticality of the current production condition, the actual power consumption and the actual completion time, dynamically adjust the abnormal power consumption threshold and the abnormal completion time threshold to obtain the corrected abnormal power consumption threshold and the corrected abnormal completion time threshold.

[0094] S340. Determine the processor overheating risk based on the second power consumption benchmark, the second completion time benchmark, the corrected power consumption abnormal threshold, and the corrected completion time abnormal threshold.

[0095] Specifically, obtaining the processor's current ambient temperature refers to collecting real-time temperature data of the processor's operating environment through an ambient temperature sensor integrated inside or outside the all-in-one terminal. This ambient temperature data is a crucial factor affecting the processor's heat dissipation efficiency and its own temperature; its level directly relates to the processor's temperature rise rate and eventual stable temperature at the same power consumption.

[0096] The criticality of acquiring the processor's current production status refers to assessing the level of real-time performance, reliability, or security requirements of the production task currently being executed by the integrated workstation. For example, production status can be categorized into different levels such as high criticality (e.g., emergency braking, precision control), medium criticality (e.g., routine production process monitoring), and low criticality (e.g., data logging, non-real-time report generation). This criticality level can be determined by analyzing the type of control command currently being executed, system load, user configuration, or preset production plans.

[0097] In practical applications, based on the current ambient temperature, the criticality of the current production condition, the actual power consumption, and the actual completion time, the abnormal power consumption threshold and the abnormal completion time threshold are dynamically adjusted to obtain corrected abnormal power consumption threshold and corrected abnormal completion time threshold. The purpose is to make the risk assessment criteria more adaptable. Specifically, when the current ambient temperature is high, to avoid processor overheating, the abnormal power consumption threshold and abnormal completion time threshold are tightened, meaning the allowable fluctuation range of power consumption and completion time is reduced. When the current production condition is highly critical, to ensure the real-time performance and reliability of the task, these thresholds are also tightened to detect potential performance degradation or overheating risks earlier. Conversely, when the ambient temperature is low or the production condition is less critical, these thresholds can be appropriately relaxed. This dynamic adjustment can be implemented based on a preset lookup table, empirical formula, or machine learning model. The corrected abnormal power consumption threshold and the corrected abnormal completion time threshold are the assessment criteria obtained after the above dynamic adjustment and are used for the current specific operating condition.

[0098] Furthermore, based on the second power consumption benchmark, the second completion time benchmark, the corrected power consumption anomaly threshold, and the corrected completion time anomaly threshold, the processor overheating risk is determined. The purpose is to assess the risk using more accurate benchmarks and judgment thresholds more suited to the current operating conditions. Specifically, the deviation between the processor's actual power consumption and the second power consumption benchmark, and the deviation between the processor's actual completion time and the second completion time benchmark, are compared with the corrected power consumption anomaly threshold and the corrected completion time anomaly threshold, respectively. If either deviation exceeds its corresponding correction threshold, it can be determined that the processor has an overheating risk.

[0099] In some preferred embodiments, a specific example is given below. Assume the integrated workstation operates in two different scenarios:

[0100] Scenario 1: The ambient temperature is high (e.g., 40°C), and a critical control task with extremely high real-time requirements is being performed (e.g., precise trajectory control of a robotic arm). In this situation, the system will acquire information about the high current ambient temperature and the high-criticality production conditions. Based on this information, the abnormal power consumption threshold and the abnormal completion time threshold will be dynamically tightened. For example, if the second power consumption benchmark is 50W and the second completion time benchmark is 100ms, under normal conditions, the abnormal threshold might be set to ±10% (i.e., power consumption exceeding 55W or falling below 45W, and time exceeding 110ms or falling below 90ms are considered abnormal). However, under high ambient temperature and high-criticality conditions, the corrected power consumption abnormal threshold might be adjusted to ±5% (i.e., power consumption exceeding 52.5W or falling below 47.5W, and time exceeding 105ms or falling below 95ms are considered abnormal), and the corrected completion time abnormal threshold will also be tightened accordingly. At this time, even a small deviation in the processor's actual power consumption or completion time is more likely to be identified as an overheating risk, thereby triggering corresponding strategy adjustments.

[0101] Scenario 2: The ambient temperature is low (e.g., 10°C), and a non-critical data logging task is being performed. In this case, the system will acquire the low current ambient temperature and low-criticality production conditions. Based on this information, the power consumption anomaly threshold and the completion time anomaly threshold will be dynamically relaxed. For example, the power consumption anomaly threshold may be adjusted to ±15% (i.e., power consumption exceeding 57.5W or falling below 42.5W, and completion time exceeding 115ms or falling below 85ms are considered abnormal), and the completion time anomaly threshold will also be relaxed accordingly. In this case, even if the processor power consumption or completion time fluctuates relatively significantly, as long as it remains within the relaxed threshold range, it will not be immediately judged as an overheating risk, thus allowing the processor to operate within a wider performance range and avoiding unnecessary performance limitations.

[0102] As can be seen from the above examples, the solution proposed in this application can intelligently adjust the risk assessment criteria according to the actual operating conditions, making the operation safety control of the integrated operation terminal more refined and intelligent.

[0103] Through the above technical solutions, the accuracy and adaptability of processor overheating risk assessment in integrated workstations have been significantly improved. Specifically, by dynamically adjusting the abnormal power consumption and completion time thresholds based on real-time acquisition and utilization of current ambient temperature and the criticality of production conditions, false alarms or missed alarms caused by fixed thresholds can be avoided. In harsh environments or when the task is highly critical, stricter thresholds can provide earlier risk warnings, thus gaining valuable time for intervention and effectively ensuring the safe operation of the integrated workstation and the reliable completion of tasks. Conversely, in favorable environments or when the task is less critical, appropriately relaxed thresholds can reduce unnecessary performance limitations and improve system efficiency and resource utilization. This adaptive risk assessment mechanism allows integrated workstations to better balance the relationship between performance, energy consumption, and safety.

[0104] In one embodiment of this application, regarding the above-mentioned S140, adjusting the processor's operating strategy according to the processor overheating risk, including but not limited to steps S410 to S440, each step will be described in turn below.

[0105] S410: When there is a risk of processor overheating, monitor real-time data of the production process and assess the criticality of the current production conditions to the real-time performance of computing.

[0106] S420 adjusts the processor's operating strategy based on the criticality of the production process, including...

[0107] When the production process is highly real-time critical, the priority of the processor's working strategy is reduced to the lowest level, and the execution of internal processor adjustment actions is restricted.

[0108] When the production environment is characterized by normal real-time criticality, the priority of the processor's operating strategy will be restored to the normal level.

[0109] In some preferred embodiments, a specific example is given below. Assume an integrated work terminal is applied to an automated production line, which includes multiple sensors and actuators. When the processor detects a risk of overheating, the system first assesses the current production conditions.

[0110] For example, if the production line is executing an emergency stop command or performing a robot path planning task with extremely high time accuracy requirements, the production condition is identified as having high real-time criticality. In this situation, the system will lower the priority of the processor's working strategy to the lowest level and restrict the processor from performing internal adjustment actions such as downclocking or core shutdown to ensure that the emergency stop command can be responded to immediately or that the robot path planning task can be completed with maximum performance, avoiding delays caused by thermal management and thus ensuring production safety.

[0111] On the other hand, if the production line is performing routine data logging, equipment status monitoring, or non-urgent parameter adjustment tasks, this production condition is identified as having normal real-time criticality. In this case, the processor's operating strategy priority will be restored to normal levels. The processor can appropriately reduce its frequency or adjust fan speed based on its temperature and load to effectively control the temperature without affecting the normal execution of routine tasks. In this way, the integrated workstation can maximize its ability to meet the real-time requirements of different production conditions while ensuring its own operational safety.

[0112] This application's solution addresses the conflict between traditional processor operating strategies and real-time production requirements by introducing an assessment of the real-time criticality of production conditions. Because the system can intelligently identify the importance of current tasks when the processor faces overheating risks, the adjustment of the processor operating strategy is no longer a single-dimensional response, but rather a comprehensive consideration of both equipment safety and production continuity. When production conditions have extremely high real-time requirements, by reducing the priority of the processor operating strategy and limiting internal processor adjustments, the supply of computing resources for critical tasks is ensured, avoiding performance bottlenecks or delays caused by thermal management measures. Conversely, under normal real-time conditions, the processor operating strategy is allowed to function normally, thereby effectively controlling processor temperature and extending equipment lifespan.

[0113] In one embodiment of this application, the above step S420, which adjusts the processor working strategy according to the criticality of the production conditions, also includes, but is not limited to, steps S510 to S540. Each step will be described in turn below.

[0114] S510: Obtain key control instructions and timestamp the key control instructions;

[0115] S520: Calculate the actual end-to-end delay of key control commands based on timestamps to obtain initial delay data;

[0116] S530. Smooth the initial delay data to obtain smoothed delay data;

[0117] S540: Adjust the processor's operating strategy based on the smoothed latency data.

[0118] Acquiring critical control commands refers to the system identifying and intercepting commands that are essential to the operational safety or core functions of the integrated terminal. These commands typically have strict real-time requirements, such as emergency stop commands, critical sensor data acquisition commands, or actuator control commands. The identification of these commands can be achieved through preset command type lists, command priority tags, or pattern matching based on command content.

[0119] The purpose of timestamping the critical control instructions is to accurately record key time points in the instruction's lifecycle. Timestamps can include the instruction's sending time, transmission time at network nodes, reception time, and execution completion time. By embedding timestamp information into the instruction data packet, basic data can be provided for subsequent delay calculations. For example, a dedicated field can be added to the header or footer of the instruction data packet to store the timestamp.

[0120] The actual end-to-end delay of the critical control command is calculated based on the timestamp, yielding initial delay data. The purpose of this initial delay data is to quantify the total time elapsed from the issuance of the critical command to its completion. The actual end-to-end delay can be understood as the time consumed from the issuance of the command from its source to the completion of the action by the target actuator. The initial delay data is the raw delay value calculated directly from the timestamp and may contain instantaneous fluctuations or noise.

[0121] The actual end-to-end delay is smoothed to obtain smoothed delay data. The purpose of this smoothing process is to eliminate instantaneous noise and outliers in the initial delay data, thereby obtaining more stable delay information that better reflects the true trend. Various algorithms can be used for smoothing, such as moving average, exponential smoothing, and Kalman filtering. The specific algorithm and its parameters are determined based on the characteristics of the key control commands and the system's sensitivity to delay data. The smoothed delay data can more accurately reflect the system's real-time performance over a period of time.

[0122] Based on the smoothed latency data, the processor's operating strategy is adjusted. The aim is to dynamically and finely adjust the processor's operating strategy based on a precise perception of the real-time performance of critical instructions. When the smoothed latency data indicates that the latency is close to or exceeds a preset threshold, the system can take corresponding measures, such as adjusting the processor frequency, core allocation, task scheduling priority, and processor operating strategy, to ensure timely response to critical instructions. This adjustment is based on actual performance feedback and provides a more accurate and timely response compared to relying solely on the criticality of production conditions.

[0123] In some preferred embodiments, this application is implemented as follows: Assume the integrated operation terminal is performing a high-precision robotic arm control task, where "emergency stop" and "path correction" instructions are identified as critical control instructions. When the system detects a risk of processor overheating, in addition to adjusting the processor's operating strategy based on production condition assessment (e.g., the robotic arm is currently performing a delicate operation, which is of high real-time criticality), it will further monitor the actual latency of these critical control instructions.

[0124] Specifically, when a "path correction" command is sent, the system timestamps it, recording the transmission time. After the command is transmitted over the network and received by the robotic arm controller, the reception time is recorded. The completion time of the robotic arm's path correction action is also recorded. Based on these timestamps, the actual end-to-end delay of the command is calculated. For example, if the initial delays of multiple consecutive "path correction" commands are 50ms, 52ms, 48ms, and 55ms, these data are input into a moving average filter for smoothing, resulting in a smoothed delay, such as 51ms.

[0125] The system continuously monitors these smoothed latency data. If the smoothed latency data begins to gradually increase, for example, from 51ms to 60ms, and approaches the preset latency threshold (e.g., 65ms), the system will dynamically adjust the processor's operating strategy based on this trend, even if the processor overheating risk has not yet reached the highest level. For example, it can moderately increase the processor's clock frequency, or prioritize tasks related to robotic arm control, or even temporarily limit the resource usage of some non-critical background tasks to ensure that the latency of the "path correction" instruction does not exceed a safe range. This strategy adjustment based on actual latency feedback can more precisely and promptly ensure the real-time performance of critical operations, thereby improving the operational safety and control accuracy of the integrated operation terminal.

[0126] This application's solution overcomes the limitations of relying solely on macroscopic assessments of criticality in production conditions by introducing monitoring and analysis of the actual end-to-end latency of critical control instructions. Specifically, when the processor faces overheating risks and needs to adjust its operating strategy, the solution no longer relies solely on a rough judgment based on whether the production condition is of high or normal real-time criticality. Instead, it further acquires critical control instructions crucial to system operational safety. By timestamping these instructions, the system can accurately calculate the actual end-to-end latency from instruction issuance to execution completion, thus obtaining initial latency data. This initial latency data, after smoothing, effectively filters out transient fluctuations and noise, providing more stable and reliable smoothed latency data that truly reflects the real-time performance of critical instructions. Therefore, processor operating strategy adjustments are no longer based on a static or macroscopic assessment, but rather on dynamic, actual performance feedback from critical instructions. For example, even if a production condition is assessed as having normal real-time criticality, if smoothed latency data shows that the latency of critical control instructions is approaching or exceeding a safety threshold, the system can still adjust the processor's operating strategy in a timely and targeted manner. This could involve appropriately increasing processor resource allocation or adjusting task scheduling priorities to ensure the real-time performance of critical instructions, thereby avoiding potential risks caused by accumulated latency. This feedback mechanism based on actual latency data makes processor operating strategy adjustments more refined and intelligent, and can more effectively ensure the operational safety of integrated workstations.

[0127] In one embodiment of this application, the above-mentioned S520 calculates the actual end-to-end delay of the key control instruction based on the timestamp mark and obtains the initial delay data, including but not limited to steps S610 to S650, which are described in turn below.

[0128] S610. When sending critical control commands, generate a command sending timestamp and embed the command sending timestamp into the command data packet;

[0129] S620. When the instruction data packet passes through the intermediate network node, record the local timestamp and use the network time synchronization protocol to synchronize and calibrate the local timestamp with the instruction sending timestamp carried in the instruction data packet to obtain the intermediate network node timestamp.

[0130] S630. After receiving the instruction data packet, record its local receiving timestamp;

[0131] S640. After completing the instruction execution, record the local execution completion timestamp;

[0132] S650: Calculate the actual end-to-end delay of the critical control command based on the local reception timestamp, local execution completion timestamp, instruction sending timestamp, and intermediate network node timestamps to obtain initial delay data.

[0133] The instruction transmission timestamp refers to the time stamp generated at the precise moment when the integrated operation terminal issues a critical control instruction. This timestamp is embedded in the instruction data packet carrying the critical control instruction to ensure that the instruction always carries its original transmission time information during transmission.

[0134] Furthermore, as the instruction data packet is transmitted in the network and passes through any intermediate network node, such as a router or switch, the intermediate network node records its local timestamp. To ensure the accuracy and consistency of the timestamp, a network time synchronization protocol (such as Network Time Protocol NTP or Precision Time Protocol PTP) is used to synchronize and calibrate the local timestamp with the instruction sending timestamp carried in the instruction data packet, thereby obtaining a unified intermediate network node timestamp.

[0135] Specifically, when the target device receives a command data packet, it records its local reception timestamp. After the critical control commands carried by the command data packet have been fully executed, the target device also records a local execution completion timestamp.

[0136] Therefore, by comprehensively utilizing the instruction sending timestamp, intermediate network node timestamp, local receiving timestamp, and local execution completion timestamp, the actual end-to-end delay of the entire process from the sending end to the execution end of the critical control instruction can be accurately calculated, thereby obtaining the initial delay data.

[0137] This application's solution constructs a comprehensive time chain by accurately recording and synchronizing timestamps at multiple key points in the lifecycle of critical control commands (including command transmission, intermediate network node transmission, command reception, and command execution completion). This multi-point timestamp capture mechanism, combined with calibration using a network time synchronization protocol, ensures that all timestamps are compared under a unified time benchmark. By calculating the time differences between these timestamps, the time consumed by the command in each stage, such as network transmission, node processing, and terminal execution, can be meticulously decomposed, thereby achieving accurate quantification of the actual end-to-end latency of critical control commands. This method avoids the errors that may arise from single-point-of-time measurements and can reveal the specific sources of latency, providing reliable data support for subsequent latency analysis and strategy adjustments.

[0138] In one embodiment of this application, the process of smoothing the initial delay data to obtain smoothed delay data in step S530 includes, but is not limited to, steps S710 to S730, which will be described in turn below.

[0139] S710, Obtain the type of critical control command;

[0140] S720. Select the smoothing algorithm and corresponding parameters according to the type of key control instructions;

[0141] S730. Smooth the initial delay data according to the smoothing algorithm and corresponding parameters. Smooth the delay data.

[0142] Specifically, identifying the type of critical control command refers to determining the category to which the currently processed critical control command belongs. Critical control commands can be categorized based on their function, data characteristics, real-time requirements, or impact on the production process. For example, command types may include, but are not limited to: emergency stop commands, motion control commands, sensor data acquisition commands, and status report commands. Each command type may have different data packet sizes, transmission frequencies, latency tolerances, and accuracy requirements for smoothing the processing results.

[0143] The selection of a smoothing algorithm and corresponding parameters based on the type of key control command refers to dynamically matching the most suitable smoothing algorithm and its configuration parameters for different types of key control commands. For example, for motion control commands with extremely high real-time requirements and small data fluctuations, exponential smoothing or moving average methods can be selected, with a small smoothing coefficient set to quickly respond to delay changes; while for status reporting commands with large data fluctuations or high randomness, Kalman filtering or weighted moving average methods can be used, with a larger smoothing window or more complex parameters set to better filter out noise. Selecting appropriate algorithms and parameters aims to ensure that the smoothed delay data accurately reflects the actual delay trend while effectively suppressing instantaneous noise and outliers.

[0144] In practical applications, smoothing the initial latency data according to the smoothing algorithm and corresponding parameters results in smoothed latency data. Smoothed latency data refers to applying a specific smoothing algorithm and its parameters, selected based on the instruction type, to the original, unprocessed initial latency data. In this way, the original latency data, which may contain noise or transient fluctuations, can be optimized to generate more stable and representative smoothed latency data. This smoothed latency data will be used for subsequent processor operating strategy adjustments to ensure the accuracy and effectiveness of the adjustments.

[0145] In some preferred embodiments, this application is implemented as follows:

[0146] Suppose that the integrated operation terminal needs to process two main types of key control commands: motion control commands and environmental sensor data acquisition commands. Motion control commands have extremely high real-time requirements, and their latency fluctuations are usually small, but they need to respond quickly; while environmental sensor data acquisition commands have a relatively low transmission frequency, and the data fluctuations may be larger, but they have a slightly higher tolerance for instantaneous delays.

[0147] When the system acquires the initial latency data of the motion control command, due to its high real-time performance and low fluctuation characteristics, the system selects exponential smoothing based on the command type and sets a small smoothing coefficient (e.g., 0.2). This ensures that the smoothed latency data quickly reflects the latest latency changes, allowing the processor's operating strategy to adjust in a timely manner to maintain the accuracy of motion control.

[0148] When the system acquires the initial latency data of the environmental sensor data acquisition command, considering the potentially large fluctuations, the system selects a weighted moving average method based on the command type and sets a relatively large smoothing window (e.g., containing the most recent 10 data points) and corresponding weight distribution. In this way, smoothing can effectively filter out instantaneous noise in the data, providing a more stable and reliable latency trend and avoiding unnecessary processor strategy adjustments triggered by occasional fluctuations.

[0149] By dynamically adjusting the smoothing algorithm and parameters according to the instruction type, the system can provide the most optimized and accurate latency data assessment for the characteristics of different key control instructions, thereby making the adjustment of the processor's working strategy more refined and effective.

[0150] This application's solution addresses the limitations of a single smoothing method when handling diverse instruction latency data by introducing the identification of key control instruction types. Because different types of key control instructions possess different characteristics and sensitivities to latency data, a uniform smoothing strategy cannot adequately address all situations. By acquiring the instruction type, the system can selectively choose the most suitable smoothing algorithm and parameters. For example, for instructions requiring fast response, selecting a fast-response smoothing algorithm avoids latency evaluation lag caused by over-smoothing; while for instructions requiring more noise filtering, selecting a more powerful filtering algorithm improves the accuracy of latency data. This dynamic adaptability ensures that the smoothed latency data more realistically and accurately reflects the actual latency status of the current system, thus providing a more reliable basis for subsequent adjustments to the processor's operating strategy.

[0151] In one embodiment of this application, the above-mentioned step S730 smooths the initial delay data according to the smoothing algorithm and corresponding parameters, including but not limited to steps S810 to S830, which are described in turn below.

[0152] S810, Monitor the residual fluctuation between the smoothed delay data and the initial delay data;

[0153] S820, monitors the rate of change of delayed data after smoothing;

[0154] S830. When the residual fluctuation or rate of change exceeds the preset threshold, adjust the smoothing algorithm or its parameters corresponding to the instruction according to the characteristics of the instruction, and monitor the residual fluctuation or rate of change according to the adjusted smoothing algorithm or its parameters until the residual fluctuation and rate of change return to the preset threshold range.

[0155] Specifically, monitoring the residual fluctuations between the smoothed delayed data and the initial delayed data refers to continuously calculating and analyzing the differences (i.e., residuals) between the smoothed delayed data and the original, unsmoothed initial delayed data, and evaluating the magnitude of these residual fluctuations. Residual fluctuations can be understood as a measure of how well the smoothing model fits the original data; larger fluctuations may indicate that the current smoothing algorithm or parameters are no longer applicable. For example, residual fluctuations can be quantified by calculating statistics such as the standard deviation, variance, or mean absolute deviation of the residuals.

[0156] The rate of change of the smoothed delayed data refers to the speed or trend of change of the smoothed delayed data over time in real time. The rate of change can reflect the dynamic characteristics of the delayed data, such as whether there are sudden increases or decreases in trend. For example, the rate of change can be evaluated by calculating the difference between the smoothed delayed data at adjacent time points, or by using methods such as moving averages or exponential smoothing.

[0157] In practical applications, the preset thresholds are limits pre-set based on the operational requirements of the integrated work terminal, system stability needs, and tolerance for the accuracy of delayed data. When residual fluctuations or rates of change exceed these thresholds, it indicates that the current smoothing effect may have deviated from expectations and intervention is required.

[0158] Furthermore, adjusting the smoothing algorithm or its parameters corresponding to the instruction based on the characteristics of the instruction means that when abnormal fluctuations or rates of change are detected, the system will intelligently select a more suitable smoothing algorithm (e.g., switch from simple moving average to exponentially weighted moving average, Kalman filter, etc.) or adjust the parameters of the existing algorithm (e.g., adjust the window size of the moving average, the exponential weighting coefficient, etc.) according to the specific type of the current key control instruction (e.g., whether it is a periodic instruction, an event-triggered instruction, or an emergency braking instruction) and its different requirements for real-time performance and accuracy.

[0159] Therefore, monitoring the residual fluctuation or the rate of change according to the adjusted smoothing algorithm or its parameters until the residual fluctuation and the rate of change return to the preset threshold range means that after adjusting the smoothing algorithm or parameters, the system will continue to monitor to verify the effectiveness of the adjustment. This process will continue until the residual fluctuation and the rate of change both fall back to an acceptable preset threshold range, ensuring the stability and accuracy of the smoothing process.

[0160] In some preferred embodiments, assuming that the latency data of critical control commands fluctuates little during the initial normal operation of the integrated terminal, the system uses a simple moving average algorithm for smoothing. However, at a certain moment, due to a sudden increase in network load, the initial latency data of critical control commands exhibits significant instantaneous spikes and persistent fluctuations. At this time, the system detects a sharp increase in the residual fluctuation between the smoothed latency data and the initial latency data, and the rate of change of the smoothed latency data also exceeds a preset threshold. Based on this, the system identifies the current command type and, according to its high real-time requirements, automatically adjusts the smoothing algorithm to an exponentially weighted moving average, and dynamically adjusts its smoothing coefficient (e.g., increasing the weighting factor to respond to the latest data faster). After adjustment, the system continues to monitor the residual fluctuation and rate of change until they return to the preset threshold range, thereby ensuring that the smoothed latency data can still accurately and quickly reflect the real latency situation when network conditions change, providing a reliable basis for timely adjustment of the processor's operating strategy.

[0161] This application's solution introduces a monitoring mechanism for residual fluctuations between smoothed and initial delay data, as well as the rate of change of the smoothed delay data, enabling real-time evaluation of the effectiveness and adaptability of the current smoothing process. When these monitoring indicators exceed preset thresholds, it indicates that the current smoothing algorithm or its parameters may no longer be suitable for the current delay data characteristics, such as overfitting, underfitting, or insufficient response to sudden changes. It is precisely this dynamic monitoring and feedback mechanism that allows the system to adaptively adjust the smoothing algorithm or its parameters according to the characteristics of critical control instructions. By continuously monitoring the effects of the adjustments and iteratively optimizing, it ensures that the smoothing process always accurately and robustly reflects the actual end-to-end delay of critical control instructions, thereby providing more reliable data support for subsequent adjustments to the processor's operating strategy.

[0162] Through the above technical solution, this application effectively addresses the problem of insufficient accuracy in dynamic environments caused by traditional fixed-parameter smoothing processing. This solution significantly improves the accuracy and robustness of smoothing delayed data for critical control commands by real-time monitoring of the smoothing effect and adaptively adjusting the smoothing algorithm and parameters. This enables the integrated terminal to more accurately perceive the system's real-time status, thereby allowing for more timely and reasonable adjustments to the processor's operating strategy. This effectively avoids misjudgments or delayed responses caused by inaccurate delayed data, further enhancing the system's security, stability, and reliability.

[0163] In one embodiment of this application, when multiple smoothed delay data are close to a preset threshold, the processor working strategy is adjusted according to the smoothed delay data in step S540, including but not limited to steps S910 to S950. Each step will be described in turn below.

[0164] S910, Identify the type of key control command corresponding to the smoothed delayed data;

[0165] S920. Based on the type, obtain the preset weight of the impact of the instruction on production safety.

[0166] S930. Calculate the comprehensive risk value of potential abnormal delays based on the instruction type corresponding to the delay data and its preset weight.

[0167] S940. Based on the comprehensive risk value, potential abnormal delays are classified into risk levels.

[0168] S950: Adjust processor operating strategy according to risk level.

[0169] Specifically, identifying the type of critical control command corresponding to the smoothed delayed data refers to the system classifying the received critical control commands into predefined command types based on their identifiers, function codes, or data content. For example, command types may include emergency stop commands, safety protection commands, production process control commands, and data acquisition commands. Each command type may correspond to different real-time requirements and varying degrees of impact on production safety.

[0170] The pre-set weights for the impact of commands on production safety, based on command type, can be understood as assigning a numerical value to each type of critical control command. This value reflects the potential harm to the production process, equipment safety, or personnel safety should the command be delayed or malfunctioning. For example, emergency stop commands may have the highest weight, while data acquisition commands may have a relatively lower weight. These weights can be preset during system design based on industry standards, safety specifications, and expert experience, and can be dynamically adjusted according to actual operating conditions.

[0171] In practical applications, based on the instruction type corresponding to the latency data and its preset weight, a comprehensive risk value for potential abnormal latency is calculated. Specifically, this involves combining each smoothed latency data point close to a preset threshold with its corresponding preset weight for the instruction type, and then using a specific algorithm (such as weighted average, product, or a more complex risk assessment model) to calculate a comprehensive risk index. This comprehensive risk value can quantify the impact of multiple potential latency anomalies on the overall system security. For example, the deviation of the smoothed latency data from the preset threshold can be multiplied by the instruction weight, and then the risk contribution of all relevant instructions can be summed or the maximum value can be taken.

[0172] Furthermore, based on the aforementioned comprehensive risk value, potential abnormal latency is classified into risk levels. This involves comparing the calculated comprehensive risk value with preset risk level thresholds to categorize the current potential abnormal latency into different risk levels, such as low risk, medium risk, high risk, or emergency risk. Each risk level corresponds to different processor work strategy adjustment priorities and response measures.

[0173] Therefore, adjusting the processor's operating strategy according to the aforementioned risk classification means that the system automatically or semi-automatically triggers corresponding adjustments to the processor's operating strategy based on the determined risk level. For example, for a low-risk level, only minor performance optimization or resource scheduling may be performed; for a medium-risk level, more aggressive performance monitoring and early warning may be initiated; and for a high-risk level, more stringent measures such as immediately reducing the frequency, limiting non-critical tasks, or even triggering partial safety shutdown procedures may be necessary to ensure production safety.

[0174] In some preferred embodiments, a specific example is given below. Assume the integrated control terminal is executing a complex industrial control task involving multiple critical control commands, such as an "emergency braking command," a "valve opening adjustment command," and a "temperature sensor data acquisition command." At a certain moment, the system detects that the smoothed delay of the "valve opening adjustment command" is 95ms, while its preset threshold is 100ms; simultaneously, the smoothed delay of the "temperature sensor data acquisition command" is 180ms, while its preset threshold is 200ms.

[0175] According to the scheme of this application, the types of these instructions are first identified. Assuming that the "emergency braking instruction" is identified as the highest safety level instruction, with a preset weight of 10; the "valve opening adjustment instruction" is a medium safety level instruction, with a preset weight of 7; and the "temperature sensor data acquisition instruction" is a lower safety level instruction, with a preset weight of 3.

[0176] Next, the system calculates the overall risk value of potential abnormal delays based on these instruction types and their preset weights. For example, the risk contribution can be defined as (delay data / threshold) * weight.

[0177] For the "valve opening adjustment command": Risk contribution = (95ms / 100ms) 7 = 0.95 7 = 6.65.

[0178] For the "temperature sensor data acquisition command": Risk contribution = (180ms / 200ms) 3 = 0.90 3 = 2.70.

[0179] Assuming the overall risk value is the sum of all risk contributions, then the overall risk value = 6.65 + 2.70 = 9.35.

[0180] Then, based on the aforementioned comprehensive risk value, potential abnormal delays are classified into risk levels. Assume the preset risk level thresholds are: 0-5 for low risk, 5-10 for medium risk, and above 10 for high risk. In this case, 9.35 falls into the medium risk level.

[0181] Finally, the processor's operating strategy is adjusted based on the aforementioned risk classification. Since the current risk level is medium, the processor's operating strategy may be adjusted as follows: prioritize the execution resources for the "valve opening adjustment instruction," moderately increasing its processing priority; simultaneously, continuously monitor the "temperature sensor data acquisition instruction," but refrain from immediate aggressive performance interventions to avoid unnecessary performance impact on the overall system. If the overall risk value reaches a high-risk level, a more stringent strategy may be triggered, such as temporarily suspending some non-critical tasks or activating backup processing units to ensure the real-time performance of the most critical instructions. In this way, the system can make targeted and efficient strategy adjustments based on the actual risk level, rather than simply handling each latency approaching the threshold independently.

[0182] Through the aforementioned technical solution, the integrated workstation can achieve more intelligent and proactive processor operation strategy adjustments when facing complex operating conditions, especially when the latency data of multiple critical control instructions simultaneously approaches preset thresholds. This solution avoids the limitations of simply responding to single latency events. Instead, by comprehensively assessing the combined risks of instruction type, security impact weight, and potential abnormal latency, the system can effectively classify risks, thereby prioritizing the handling of potential problems with the greatest impact on production safety. This significantly improves the safety, stability, and reliability of the workstation operation, reduces the risk of production accidents caused by accumulated latency or untimely response to critical instructions, and optimizes the allocation efficiency of processor resources.

[0183] See Figure 2 , Figure 2 This is a schematic diagram of an integrated work terminal operation safety control system provided in one embodiment of this application. The integrated work terminal operation safety control system 1000 includes:

[0184] The benchmark establishment module 1010 establishes a first power consumption benchmark and a first completion time benchmark for the processor when executing a preset computing task.

[0185] The real-time monitoring module 1020 is used to monitor the actual power consumption and actual completion time of the processor when executing the current computing task in real time.

[0186] The risk assessment module 1030 determines the risk of processor overheating based on the actual power consumption, the actual completion time, the first power consumption benchmark, and the first completion time benchmark.

[0187] The strategy adjustment module 1040 adjusts the processor's operating strategy based on the risk of processor overheating.

[0188] The benchmark establishment module aims to provide a stable reference benchmark for the processor's power consumption and completion time when executing preset computational tasks. In a preferred implementation, this benchmark establishment module can be a software component configured to drive the processor to execute a series of preset, representative computational tasks in a controlled environment after initial deployment or maintenance of the all-in-one terminal. During this process, the processor's instantaneous power consumption is continuously recorded via a high-precision power consumption measurement interface, and its average power consumption is calculated as a power consumption benchmark. Simultaneously, the start and end times of each preset task are precisely recorded via a system timer or task scheduler, thereby calculating the task completion time and using its average as the completion time benchmark. For example, a benchmark test program can be set up, containing various typical industrial control algorithms, recording the power consumption and completion time when the processor is running at its rated frequency and voltage while executing these algorithms. Alternatively, the benchmark establishment module can be a hardware unit, such as dedicated memory integrated on the motherboard, used to store power consumption and completion time data obtained by the manufacturer under standard test conditions before the processor leaves the factory, serving as the initial power consumption and completion time benchmarks. This data can be loaded and used during system startup.

[0189] The purpose of a real-time monitoring module is to continuously acquire the actual power consumption and actual completion time of the processor while executing the current computing task. Specifically, the real-time monitoring module can be a software agent running in the operating system or real-time operating system (RTOS) of an all-in-one job terminal, configured to interact with the processor's power management unit (PMU) to acquire instantaneous power consumption data of the processor at a high sampling rate (e.g., once per millisecond) and calculate the average actual power consumption within a specific time window. Simultaneously, this software agent also works with the task scheduler to record the start and end timestamps of the currently executing computing task, thereby calculating the actual completion time of the task. For example, when the processor executes a data acquisition task, the real-time monitoring module records the total time elapsed from task initiation to data processing completion and uses this as the actual completion time. As an alternative implementation, the real-time monitoring module can include an external sensor array, such as current and voltage sensors deployed on the processor's power supply path, to measure the processor's input power in real time and use this as the actual power consumption. The task completion time can then be determined by monitoring task-related output signals or status registers.

[0190] The comparison module aims to compare the real-time monitored actual power consumption with the actual task completion time and the power consumption and completion time benchmarks provided by the benchmark establishment module. Specifically, the comparison module can be a software routine configured to receive actual power consumption and actual completion time data from the real-time monitoring module, and power consumption and completion time benchmark data from the benchmark establishment module. This module then performs arithmetic comparison operations, such as calculating power consumption deviation rate and time deviation rate. For example, if the actual power consumption is a certain percentage higher than the power consumption benchmark, or the actual completion time is a certain percentage longer than the completion time benchmark, these deviations will be calculated and used as the basis for subsequent risk assessment. As a preferred implementation, the comparison module can be designed to perform efficient data processing to ensure real-time performance.

[0191] The risk assessment module aims to compare the real-time monitored power consumption with the actual task completion time and the power consumption and completion time benchmarks provided by the benchmark establishment module. Specifically, the comparison module can be a software routine configured to receive actual power consumption and actual completion time data from the real-time monitoring module, and power consumption and completion time benchmark data from the benchmark establishment module. This module then performs arithmetic comparison operations, such as calculating the power consumption deviation rate and time deviation rate. For example, if the actual power consumption is a certain percentage higher than the power consumption benchmark, or the actual completion time is a certain percentage longer than the completion time benchmark, these deviations will be calculated and used as the basis for subsequent risk assessment. Specifically, the risk assessment module can be a software algorithm configured to receive the power consumption deviation rate and time deviation rate calculated by the comparison module. When the power consumption deviation rate or time deviation rate exceeds a preset threshold, the module determines that the processor has an overheating risk. For example, if the actual power consumption is more than 15% higher than the power consumption benchmark, or the actual completion time is more than 10% longer than the completion time benchmark, the risk assessment module will determine that the processor may have an overheating risk. This is because when a processor experiences localized overheating, the leakage current of its internal transistors increases, leading to increased power consumption. Simultaneously, overheating also degrades processor performance, thus extending task completion time. As a further implementation method, the risk assessment module can perform a weighted average of the power consumption deviation rate and the time deviation rate to obtain a comprehensive risk index. When this risk index exceeds a preset threshold, it is determined that the processor is at risk of overheating.

[0192] The strategy adjustment module aims to take timely measures to prevent further damage once the risk assessment module determines that the processor is at risk of overheating. Specifically, the strategy adjustment module can be a software component configured to interact with the operating system's power management and task scheduling functions. When an overheating risk is detected, this module immediately triggers a frequency and voltage reduction strategy, adjusting the processor's clock frequency and core voltage to reduce its operating speed and power consumption, thereby reducing heat generation. For example, the processor frequency can be reduced from 2.0GHz to 1.5GHz, while the core voltage is reduced from 1.2V to 1.0V. Alternatively, the strategy adjustment module can dynamically adjust task scheduling priorities based on the level of overheating risk. For example, the priority of non-critical tasks can be reduced, or even suspended, to ensure that critical tasks receive sufficient computing resources and allow the processor more time to dissipate heat. Simultaneously, this module can also be configured to activate additional cooling fans or liquid cooling systems to enhance heat dissipation capabilities.

[0193] The integrated operational terminal safety control system proposed in this application, through the coordinated operation of a benchmark establishment module, a real-time monitoring module, a comparison module, a risk assessment module, and a strategy adjustment module, can more accurately capture changes in the processor's operating status. Traditional methods often rely on data from a single temperature sensor, which may not accurately reflect localized hot spots within the processor. This application's system, by combining two key indicators—power consumption and task completion time—can more sensitively detect the risk of localized overheating in the processor caused by factors such as aging thermal paste.

[0194] Specifically, when the thermal paste in an all-in-one terminal ages, the heat conduction efficiency inside the processor decreases, leading to a rise in localized temperature. This localized overheating increases the processor's leakage current, resulting in increased actual power consumption. Simultaneously, overheating also affects the processor's transistor performance, causing it to take longer to complete the same computational task, i.e., an extended actual completion time. The system in this application obtains actual power consumption and actual completion time through a real-time monitoring module and compares them with power consumption and completion time benchmarks provided by a benchmark establishment module, effectively identifying these anomalies. For example, if the actual power consumption is significantly higher than the power consumption benchmark, or the actual completion time is significantly longer than the completion time benchmark, the risk assessment module can determine that the processor is at risk of overheating.

[0195] Once the risk assessment module determines that the processor is at risk of overheating, the strategy adjustment module adjusts the processor's operating strategy according to the degree of risk. This strategy adjustment is dynamic and intelligent, and can include various methods such as frequency and voltage reduction, task scheduling optimization, and activation of auxiliary cooling. Compared with the single frequency reduction strategy in existing technologies that rely solely on temperature thresholds, the strategy adjustment implemented by the system in this application is more flexible and refined. For example, in the case of a minor overheating risk, the scheduling of non-critical tasks can be adjusted first, while in the case of a severe overheating risk, more aggressive frequency and voltage reduction measures can be taken. This risk assessment and strategy adjustment mechanism based on multi-dimensional data analysis can effectively prevent the processor from continuing to operate under localized overheating conditions, thereby extending the processor's lifespan, ensuring the long-term stable operation of the integrated workstation, and significantly improving the reliability and safety of equipment in industrial production environments.

Claims

1. A method for safe operation control of an integrated work terminal, characterized in that, include: Establish a first power consumption benchmark and a first completion time benchmark for the processor when executing a preset computing task; Real-time monitoring of the processor's actual power consumption and actual completion time while executing the current computing task; The processor overheating risk is determined based on the actual power consumption, the actual completion time, the first power consumption benchmark, and the first completion time benchmark. Adjust the processor's operating strategy based on the risk of processor overheating; The steps for adjusting the processor's operating strategy based on the risk of processor overheating include: When there is a risk of processor overheating, monitor real-time data from the production process to assess the criticality of the current production conditions to the real-time performance of computing. Based on the criticality of the production conditions, the processor's operating strategy is adjusted, wherein... When the production condition is of high real-time criticality, the priority of the processor's working strategy is reduced to the lowest level, and the execution of internal processor adjustment actions is restricted. When the production condition is of normal real-time criticality, the priority of the processor working strategy is restored to the normal level; The step of adjusting the processor's operating strategy based on the criticality of the production conditions further includes: Obtain key control commands and timestamp the key control commands; Calculate the actual end-to-end delay of the key control command based on the timestamp to obtain initial delay data; The initial delay data is smoothed to obtain smoothed delay data; Adjust the processor's operating strategy based on the smoothed delay data; The step of smoothing the initial delay data to obtain smoothed delay data includes: Obtain the type of the critical control command; Select the smoothing algorithm and corresponding parameters according to the type of the key control command; The initial delay data is smoothed according to the smoothing algorithm and the corresponding parameters to obtain smoothed delay data.

2. The integrated operation terminal safety control method according to claim 1, characterized in that, The step of determining the processor overheating risk based on the actual power consumption, the actual completion time, the first power consumption benchmark, and the first completion time benchmark includes: Within a preset time window, the processor is brought into a preset stable working state; The ambient temperature, actual power consumption, and actual time to complete the preset task are obtained by the processor under the stable operating state. Based on the ambient temperature under the stable operating state, the actual power consumption, and the actual time, the first power consumption reference and the first completion time reference are corrected to obtain the second power consumption reference and the second completion time reference. The processor overheating risk is determined based on the second power consumption benchmark and the second completion time benchmark.

3. The integrated operation terminal safety control method according to claim 2, characterized in that, The step of determining the processor overheating risk based on the second power consumption reference and the second completion time reference includes: Get the current ambient temperature of the processor; The criticality of obtaining the current production status of the processor; Based on the current ambient temperature, the criticality of the current production condition, the actual power consumption, and the actual completion time, the abnormal power consumption threshold and the abnormal completion time threshold are dynamically adjusted to obtain the corrected abnormal power consumption threshold and the corrected abnormal completion time threshold. The processor overheating risk is determined based on the second power consumption benchmark, the second completion time benchmark, the corrected power consumption anomaly threshold, and the corrected completion time anomaly threshold.

4. The integrated operation terminal safety control method according to claim 1, characterized in that, The step of calculating the actual end-to-end delay of the key control command based on the timestamp to obtain initial delay data includes: When sending critical control commands, a command sending timestamp is generated and embedded in the command data packet; When the instruction data packet passes through an intermediate network node, a local timestamp is recorded, and the local timestamp is synchronized with the instruction sending timestamp carried in the instruction data packet using a network time synchronization protocol to obtain the intermediate network node timestamp. Upon receiving the instruction data packet, record its local receiving timestamp; After the instruction is executed, record the local execution completion timestamp; Based on the local reception timestamp, the local execution completion timestamp, the instruction sending timestamp, and the intermediate network node timestamp, the actual end-to-end delay of the critical control instruction is calculated to obtain initial delay data.

5. The integrated operation terminal safety control method according to claim 1, characterized in that, After the step of smoothing the initial delay data according to the smoothing algorithm and corresponding parameters to obtain smoothed delay data, the method further includes: Monitor the residual fluctuation between the smoothed delay data and the initial delay data; Monitor the rate of change of delayed data after smoothing; When the residual fluctuation or the rate of change exceeds a preset threshold, the smoothing algorithm or its parameters corresponding to the instruction are adjusted according to the characteristics of the instruction. The residual fluctuation or the rate of change is monitored according to the adjusted smoothing algorithm or its parameters until the residual fluctuation and the rate of change return to the preset threshold range.

6. The integrated operation terminal safety control method according to claim 1, characterized in that, When multiple smoothed latency data are close to a preset threshold, the steps for adjusting the processor's operating strategy based on the smoothed latency data include: Identify the type of key control command corresponding to the smoothed delayed data; Based on the type, obtain a preset weight for the degree of impact of the instruction on production safety; Calculate the comprehensive risk value of potential abnormal delays based on the instruction type corresponding to the delay data and its preset weight; Based on the comprehensive risk value, potential abnormal delays are classified into risk levels. Adjust the processor's operating strategy based on the aforementioned risk classification.

7. An integrated operation terminal safety control system, characterized in that, The system includes: The benchmark establishment module establishes a first power consumption benchmark and a first completion time benchmark for the processor when executing a preset computing task; The real-time monitoring module is used to monitor the processor's actual power consumption and actual completion time when executing the current computing task. The risk assessment module determines the processor overheating risk based on the actual power consumption, the actual completion time, the first power consumption benchmark, and the first completion time benchmark. The strategy adjustment module adjusts the processor's operating strategy based on the risk of processor overheating. The strategy adjustment module includes: When there is a risk of processor overheating, monitor real-time data from the production process to assess the criticality of the current production conditions to the real-time performance of computing. Based on the criticality of the production conditions, the processor's operating strategy is adjusted, wherein... When the production condition is of high real-time criticality, the priority of the processor's working strategy is reduced to the lowest level, and the execution of internal processor adjustment actions is restricted. When the production condition is of normal real-time criticality, the priority of the processor working strategy is restored to the normal level; The adjustment of the processor's operating strategy based on the criticality of the production conditions also includes: Obtain key control commands and timestamp the key control commands; Calculate the actual end-to-end delay of the key control command based on the timestamp to obtain initial delay data; The initial delay data is smoothed to obtain smoothed delay data; Adjust the processor's operating strategy based on the smoothed delay data; The process of smoothing the initial delay data to obtain smoothed delay data includes: Obtain the type of the critical control command; Select the smoothing algorithm and corresponding parameters according to the type of the key control command; The initial delay data is smoothed according to the smoothing algorithm and the corresponding parameters to obtain smoothed delay data.

Images