Industrial networking device real-time state early warning method

Abstract

This invention relates to the field of industrial equipment monitoring technology and discloses a method for real-time status early warning of industrial networked equipment. The method includes: acquiring a set of real-time status data from multiple monitoring points in the industrial networked equipment, the set encompassing equipment operating parameter data, network communication parameter data, and environmental parameter data; comprehensively analyzing the real-time status data set to assess the risk values ​​of equipment operating status, network communication status, and environmental status; comparing these three types of risk values ​​with preset risk thresholds; and generating an early warning signal and executing an early warning operation based on the comparison results. This method, through multi-dimensional data acquisition and comprehensive analysis, achieves comprehensive monitoring and accurate early warning of the status of industrial networked equipment, effectively improving the reliability and timeliness of equipment status early warning, and is applicable to real-time monitoring scenarios for various types of industrial networked equipment.

Description

Technical Field

[0001] This invention relates to the field of industrial equipment monitoring technology, specifically a method for real-time status early warning of industrial networked equipment. Background Technology

[0002] With the continuous advancement of industrial intelligence, the number and types of industrial network devices are increasing daily. These devices are widely distributed in various scenarios such as production workshops and industrial parks, undertaking critical production tasks. The stable operation of these devices is directly related to the continuity of the production process and product quality. Once a malfunction or abnormality occurs, it may lead to a series of problems such as production interruption and increased costs. Currently, status monitoring of industrial networked devices often employs a single-parameter monitoring model, such as focusing only on mechanical parameters like operating temperature and speed, or solely on communication indicators like network transmission rate and latency. This single-dimensional monitoring approach has significant limitations and struggles to comprehensively reflect the true operating status of the equipment. For instance, when equipment operating parameters are normal but network communication fluctuates, data transmission may be interrupted, affecting remote control and scheduling. A single monitoring mechanism often fails to detect such problems in a timely manner. The impact of environmental factors on industrial networked equipment cannot be ignored. Environmental problems such as excessively high temperatures, abnormal humidity, and dust accumulation can accelerate equipment aging, reduce equipment performance, and even cause equipment failure. However, in existing monitoring systems, the integration of environmental parameters with equipment operating parameters and network communication parameters is low, making it difficult to achieve multi-factor collaborative analysis, resulting in insufficient accuracy and timeliness of early warnings. Traditional early warning methods often rely on preset fixed thresholds. When equipment operates in complex and variable environments, these fixed thresholds are prone to false alarms or missed alarms. For example, during the startup phase, equipment parameters may fluctuate beyond normal thresholds, but this does not necessarily indicate a fault. In such cases, fixed threshold warnings may generate false alarms. Conversely, as equipment parameters slowly deviate from normal ranges and gradually approach the fault threshold, fixed thresholds may fail to issue timely warnings, resulting in missed alarms. These issues hinder the improvement of operational efficiency for industrial networked equipment and fail to meet the demands of modern industrial production for real-time equipment status monitoring. Summary of the Invention

[0003] The purpose of this invention is to provide a real-time status early warning method for industrial networked devices to solve the problems mentioned in the background art.

[0004] To achieve the above objectives, the present invention provides a real-time status early warning method for industrial networked devices, the method comprising: Acquire a set of real-time status data from multiple monitoring points in an industrial network device, the set of real-time status data including device operating parameter data, network communication parameter data, and environmental parameter data; A comprehensive analysis is performed on the real-time status data set of the industrial network equipment to assess the risk values ​​of equipment operation status, network communication status, and environmental status. The risk values ​​of equipment operation status, network communication status, and environmental status obtained from the assessment are compared with the preset risk thresholds. Based on the comparison and judgment results, an early warning signal is generated and an early warning operation is performed.

[0005] Preferably, the equipment operating parameter data specifically includes the equipment internal temperature value, the equipment internal voltage value, and the equipment internal current value; the network communication parameter data specifically includes the network transmission delay value, the network data packet loss rate value, and the network bandwidth utilization value; and the environmental parameter data specifically includes the ambient temperature value, the ambient humidity value, and the ambient vibration intensity value.

[0006] Preferably, the specific steps for assessing the risk value of equipment operating status are as follows: Standardize and convert the internal temperature, internal voltage, and internal current values ​​of the equipment. Based on the standardized internal temperature, internal voltage, and internal current values ​​of the equipment, the risk value of the equipment's operating status is determined by a weighted average calculation method. The weighted average calculation method involves the weighting coefficients of the internal temperature value, the internal voltage value, and the internal current value of the equipment, and the sum of all weighting coefficients is a fixed value.

[0007] Preferably, the dynamic adjustment steps of the weighting coefficients in the weighted average calculation method are as follows: The weighting coefficient of the internal temperature value of the equipment is corrected based on the statistical value of the equipment runtime. Specifically, the statistical value of the equipment runtime is multiplied by the preset aging influence factor and then added to the basic weight. The weighting coefficient of the internal voltage value of the equipment is corrected based on the voltage fluctuation amplitude value during the equipment operation cycle. Specifically, the voltage fluctuation amplitude value is input into a preset sensitivity mapping function and the output is an adjustment amount. The weighting coefficient of the internal current value of the device is corrected based on the current load rate value of the device. Specifically, the difference between the current load rate value and the benchmark load threshold is multiplied by the load influence coefficient. The sum of all the corrected weighting coefficients remains a fixed value.

[0008] Preferably, the specific steps for assessing the risk value of network communication status are as follows: Perform data normalization transformation on network transmission latency, network packet loss rate, and network bandwidth utilization. Based on the normalized network transmission delay, network packet loss rate, and network bandwidth utilization, the network communication status risk value is determined through pattern recognition analysis. Pattern recognition analysis methods include extracting the changing trend characteristics of network transmission delay values, the fluctuation characteristics of network packet loss rate values, and the peak characteristics of network bandwidth utilization values, and integrating these characteristics into a comprehensive risk score.

[0009] Preferably, the specific steps for assessing the environmental status risk value are as follows: Perform quantization conversion on ambient temperature, ambient humidity, and ambient vibration intensity values; Based on the environmental temperature, humidity, and vibration intensity values ​​after scalar conversion, the environmental state risk value is determined through statistical analysis. The statistical analysis method involves calculating the standard deviation of ambient temperature, the coefficient of variation of ambient humidity, and the mean offset of ambient vibration intensity, and then integrating the calculation results into environmental risk assessment indicators.

[0010] Preferably, the specific steps for integrating equipment operation status risk values, network communication status risk values, and environmental status risk values ​​to generate a comprehensive risk index are as follows: Input the risk values ​​of equipment operation status, network communication status, and environmental status into the multi-level fusion model; The multi-level fusion model first scales the risk values ​​of equipment operation status, network communication status, and environmental status. Then, based on the scaled values, a decision tree classification method is used to output a comprehensive risk index; The decision tree classification method maps scaling values ​​to comprehensive risk indicators based on preset branching rules and leaf node thresholds.

[0011] Preferably, the specific steps for determining the risk threshold based on comprehensive risk indicators are as follows: Compare the comprehensive risk indicators with multiple preset risk threshold ranges; If the comprehensive risk indicators fall into the high-risk threshold range, an emergency warning operation will be triggered. If the comprehensive risk indicators fall into the medium risk threshold range, a normal early warning operation will be triggered; If the overall risk indicators fall within the low-risk threshold range, no warning action will be triggered, but a risk log will be recorded.

[0012] Preferably, the specific steps for performing the early warning operation are as follows: Select the warning output method according to the type of warning signal, including sound alarm, visual alarm or network notification; The sound alarm method outputs alarm sounds through audio devices, the visual alarm method displays warning icons through display devices, and the network notification method sends warning messages to the remote monitoring terminal through network protocols. Simultaneously, the execution time and detailed parameters of the early warning operation are recorded to the database.

[0013] The preferred steps for adjusting the early warning strategy in real time are as follows: Based on historical early warning records and real-time status data sets, update the risk threshold range and early warning output method; The update process includes analyzing the false alarm rate and false negative rate characteristics in historical early warning records, and dynamically adjusting the risk threshold based on the analysis results; At the same time, the priority order of early warning output methods is optimized based on the changing trend of equipment operation status risk values.

[0014] Compared with the prior art, the beneficial effects of the present invention are: By acquiring a real-time status data set consisting of equipment operating parameters, network communication parameters, and environmental parameters, multi-dimensional coverage of equipment status is achieved. This multi-parameter acquisition method breaks through the limitations of traditional single-parameter monitoring, enabling a more comprehensive capture of various status information during equipment operation. This allows for a comprehensive assessment of equipment status, moving beyond a single aspect to consider factors such as mechanical operation, network interaction, and environmental impact. In the comprehensive analysis phase, the risk values ​​of equipment operation, network communication, and environmental status are assessed separately, allowing for the individual identification of anomalies across different dimensions. This separate assessment approach clearly distinguishes the sources of various risks, avoids interference between different factors, and helps to accurately pinpoint potential problem areas. For example, when an early warning signal is generated, the assessment results of each risk value can quickly determine whether the problem stems from abnormal equipment operation, network transmission issues, or unsuitable environmental conditions, providing a clear direction for subsequent investigation and handling. By comparing the three types of risk values ​​obtained from the assessment with preset thresholds, a clear basis for early warning is provided. This multi-risk value comparison method, compared with traditional single-threshold judgment, can more meticulously distinguish the abnormal state of the equipment. Comparing different types of risk values ​​with their corresponding thresholds allows for accurate judgments on different aspects of anomalies, reducing misjudgments caused by single-threshold judgments. Based on the comparative judgment results, early warning signals are generated and early warning actions are executed, enabling the early warnings to directly impact the actual operation and maintenance process. Multi-dimensional risk assessment results make the early warning signals more targeted, allowing maintenance personnel to take appropriate measures based on the type of risk indicated by the early warning. For example, if the early warning stems from excessive network communication status risk values, network connectivity and communication equipment can be checked first; if it stems from abnormal environmental status risk values, environmental control equipment can be adjusted promptly. This targeted early warning action improves the efficiency of operation and maintenance response and avoids blind handling. This method's comprehensive analysis of three types of risk values ​​enables the consideration of correlations between different parameters. Equipment operating parameters, network communication parameters, and environmental parameters are not isolated; they influence each other. Comprehensive analysis can capture the potential correlations between parameters. For example, abnormal ambient humidity may cause fluctuations in equipment operating parameters, thereby affecting the stability of network communication. Comprehensive analysis can identify this chain reaction, thus taking into account the cumulative effects of multiple factors in early warning, making the warning more closely aligned with the actual operating conditions of the equipment. The pre-set risk threshold comparison mechanism provides flexibility for equipment early warning in different scenarios. Based on the equipment type, operating stage, and application scenario, corresponding risk thresholds can be set, allowing early warnings to adapt to diverse equipment needs. This highly adaptable early warning method can better address changes in equipment status under different operating conditions, reduce problems caused by incompatible fixed thresholds, and make early warnings more consistent with the complexities of actual production. Attached Figure Description

[0015] Figure 1 This is a schematic diagram illustrating the working principle of the real-time status early warning method for industrial networked devices according to the present invention. Figure 2 A flowchart for the dynamic adjustment of weighting coefficients; Figure 3 A flowchart for assessing the risk value of network communication status; Figure 4 A flowchart for generating comprehensive risk indicators; Figure 5 A flowchart for real-time adjustment of early warning strategies. Detailed Implementation

[0016] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0017] Please see Figure 1This invention provides a method for real-time status early warning of industrial networked devices, the method comprising: Step 1: Acquire a real-time status data set from multiple monitoring points in the industrial network device. This real-time status data set includes device operating parameter data, network communication parameter data, and environmental parameter data. Real-time data from each monitoring point is collected synchronously through the device's built-in sensors, network monitoring modules, and environmental monitoring devices, forming a complete real-time status data set to provide a foundational data source for subsequent analysis.

[0018] Step Two: Conduct a comprehensive analysis of the real-time status data set of the industrial networked devices to assess the risk values ​​of device operation status, network communication status, and environmental status. Using corresponding data analysis models and algorithms, process the parameter data for device operation, network communication, and the environment respectively to calculate their respective risk values, thus quantifying the status risks in different dimensions.

[0019] Step 3: Compare the assessed risk values ​​for equipment operation status, network communication status, and environmental status with preset risk thresholds. The preset risk thresholds are determined based on the equipment type, application scenario, and security requirements. The comparison clarifies the risk level of each risk value.

[0020] Step 4: Generate early warning signals and execute early warning actions based on the comparison and judgment results. Generate corresponding early warning signals for different risk levels, such as an emergency early warning signal for high risk and a normal early warning signal for medium risk. Execute the corresponding early warning actions according to preset rules to promptly remind relevant personnel to pay attention to the equipment status.

[0021] Example 1: The specific operating parameters of the equipment include internal temperature, internal voltage, and internal current. Internal temperature is collected by installing high-precision temperature sensors in key components such as the processor, motor, and capacitors. These sensors are connected to the main control module and acquire temperature data in real time at a preset sampling frequency, such as once every 0.5 seconds. The collected temperature values ​​directly reflect the heat dissipation and accumulation of the core components during operation. When the temperature exceeds the equipment's designed normal operating range, it may indicate overload, cooling system failure, or other problems.

[0022] The internal voltage values ​​of the device are acquired through a voltage monitoring circuit integrated into the device's power management module. This circuit can monitor the input voltage of each core circuit module in real time, including voltage data from different loops such as the motherboard power supply voltage and drive circuit voltage. During the acquisition process, the voltage monitoring circuit converts analog voltage signals into digital signals and transmits them to the data processing unit. The stability of the voltage value directly affects the normal operation of the device's electronic components. Excessive voltage may cause component breakdown, while insufficient voltage may lead to unstable operation or functional failure of the device.

[0023] The internal current value of the equipment is obtained using current transformers or Hall effect current sensors. These sensors are connected in series or coupled in the main power supply circuit of the equipment and can sense the current changes in the circuit in real time. The collected current data reflects the power consumption of the equipment under different load conditions. Abnormal current fluctuations may indicate problems such as short circuits in the internal circuits, component aging, or abnormal loads.

[0024] Network communication parameters specifically include network transmission delay, network packet loss rate, and network bandwidth utilization. Network transmission delay is measured by sending test data packets between the device and the communication node, recording the time difference between the sending and receiving of the data packet; this time difference is used as the transmission delay value. The measurement period can be dynamically adjusted according to network traffic levels; the measurement period can be shortened when network load is high to more promptly reflect delay changes. Transmission delay directly affects the timeliness of data interaction between industrial networked devices; excessive delay may lead to delayed control command execution or data synchronization failure.

[0025] The network packet loss rate is calculated by statistically analyzing packet transmission over a specific time window. In the device's network communication module, sent data packets are numbered and recorded, while the receiving end reports the numbers of received data packets. By comparing the total number of sent packets with the number of successfully received packets, the proportion of lost packets—the packet loss rate—is calculated. This value reflects the reliability of network communication; an excessively high loss rate leads to incomplete data transmission, affecting the collaborative operation of devices and the accuracy of status monitoring.

[0026] Network bandwidth utilization is calculated by monitoring the ratio of the actual data transmission volume of a device's network interface per unit time to the maximum supported bandwidth of that interface. The network communication module tracks upload and download traffic in real time and calculates the current bandwidth utilization based on the interface's rated bandwidth parameters. Excessive bandwidth utilization can lead to network congestion, further exacerbating transmission delays and packet loss.

[0027] The environmental parameters include ambient temperature, ambient humidity, and ambient vibration intensity. Ambient temperature and humidity values ​​are collected using integrated temperature and humidity sensors deployed around the equipment. The sensors should be installed as close to the equipment as possible while avoiding direct impact from the equipment's own heat dissipation; they are typically installed near the equipment casing or in a well-ventilated area of ​​the cabinet housing the equipment. The collected ambient temperature and humidity data reflect the thermal and humidity conditions of the external environment in which the equipment is located. Excessively high or low temperatures, or excessively high humidity, can affect the equipment's lifespan and operational stability. For example, high temperature and high humidity environments can easily lead to corrosion of internal components or a decline in insulation performance.

[0028] Environmental vibration intensity values ​​are obtained by installing vibration sensors on the equipment base or fixed bracket. The sensors can sense the vibration acceleration or amplitude transmitted from the external environment during equipment operation. Vibration intensity values ​​reflect the installation stability of the equipment and the vibration interference from the external environment. Continuous and strong vibrations may cause loosening of internal connectors and displacement of precision components, thereby leading to equipment failure.

[0029] All the above-mentioned parameter data are collected in real-time synchronously. The collected data is transmitted to the device's local data processing unit via the internal bus, forming a complete real-time status data set, providing comprehensive and accurate raw data for subsequent risk assessment. The data collection frequency is set according to the importance and rate of change of the parameters. For parameters that change rapidly, such as internal temperature, voltage, and current, the collection frequency is higher; for parameters that change relatively slowly, such as ambient temperature and humidity, the collection frequency can be appropriately reduced to ensure data timeliness while reducing the burden on data processing.

[0030] Example 2: See Figure 2 The system standardizes and transforms the internal temperature, voltage, and current values ​​of the equipment. During the transformation, the historical data range of each parameter under normal equipment operation must first be determined, including minimum and maximum values. By mapping the originally collected parameter values ​​to a fixed range, the influence of differences in units and magnitudes among different parameters is eliminated, ensuring that the transformed data has a unified comparison benchmark.

[0031] Based on the standardized internal temperature, voltage, and current values ​​of the equipment, a weighted average calculation method is used to determine the risk value of the equipment's operating status. During the calculation, each standardized parameter value is multiplied by its corresponding weighting coefficient, and the products are then summed to obtain the final risk value.

[0032] The weighted average calculation method involves weighting coefficients for the equipment's internal temperature, internal voltage, and internal current, with the sum of all weighting coefficients being a fixed value. This fixed value can be set according to the actual application scenario, for example, to 1, so as to intuitively reflect the proportion of each parameter in the risk assessment.

[0033] The steps for dynamically adjusting the weighting coefficients in the weighted average calculation method are as follows: The weighting coefficient of the internal temperature value of the equipment is adjusted based on the equipment runtime statistics. Specifically, the equipment runtime statistics are multiplied by a preset aging impact factor and then added to the base weight. The equipment runtime statistics are accumulated through the equipment's built-in timer, recording the total operating time from startup to the current moment. The preset aging impact factor is determined based on the equipment's material, structure, and long-term operating test data, reflecting the impact of the degree of aging that occurs with increasing operating time on temperature sensitivity. The base weight is the temperature weighting coefficient set during the initial operation phase of the equipment. Through the above adjustment method, as the equipment runtime increases, the weight of the temperature parameter in the risk assessment is correspondingly increased to adapt to the increased sensitivity to temperature changes after equipment aging.

[0034] The weighting coefficient of the internal voltage value of the equipment is adjusted based on the voltage fluctuation amplitude during the equipment's operating cycle. Specifically, the voltage fluctuation amplitude is input into a preset sensitivity mapping function, and the output is an adjustment amount. The equipment operating cycle can be set according to the equipment's working mode, such as a production batch or a work shift. Within each operating cycle, the voltage value is continuously monitored, and the difference between the maximum and minimum voltage values ​​is calculated, i.e., the voltage fluctuation amplitude. The preset sensitivity mapping function is generated by fitting a large amount of experimental data. The input of the function is the voltage fluctuation amplitude, and the output is the corresponding weighting adjustment amount. When the voltage fluctuation amplitude is large, the adjustment amount is positive, increasing the voltage weighting coefficient to highlight the impact of voltage instability on equipment operation; when the fluctuation amplitude is small, the adjustment amount is negative or zero, appropriately reducing the voltage weighting coefficient.

[0035] The weighting coefficient of the internal current value of the equipment is adjusted based on the current load rate value. Specifically, the difference between the current load rate value and the reference load threshold is multiplied by the load influence coefficient. The current load rate value is obtained through the equipment's load monitoring module and represents the percentage of the equipment's current actual load to its rated load. The reference load threshold is determined based on the equipment's design parameters and safety operation standards, and is typically a certain percentage of the rated load. The load influence coefficient reflects the degree of influence of load changes on the current parameter and is determined through operational tests of the equipment under different load conditions. When the current load rate value is higher than the reference load threshold, the difference is positive, and multiplying it by the load influence coefficient yields a positive adjustment, increasing the current weighting coefficient to emphasize the impact of current changes on equipment safety under high load conditions. When the load rate is lower than the reference threshold, the adjustment is negative, and the current weighting coefficient decreases accordingly.

[0036] The sum of all corrected weighting coefficients remains a fixed value. After individually correcting the weighting coefficients of each parameter, the corrected weighting coefficients need to be normalized to ensure that the sum of the three is still equal to the initially set fixed value. For example, if the sum of the corrected weighting coefficients for temperature, voltage, and current is greater than the fixed value, the coefficients are reduced proportionally; if it is less than the fixed value, it is increased proportionally to ensure the consistency and rationality of the weighted average calculation, so that the calculated risk value is always within a comparable range.

[0037] Example 3: See Figure 3 This process involves normalizing network transmission latency, network packet loss rate, and network bandwidth utilization. First, historical network parameter data is collected, and the mean and standard deviation of each parameter are calculated. By converting the original parameter values ​​into standard scores, the magnitude differences between different parameters are eliminated, ensuring that the transformed data exhibits the same statistical distribution characteristics.

[0038] Based on the normalized network transmission delay, network packet loss rate, and network bandwidth utilization, the network communication status risk value is determined through pattern recognition analysis.

[0039] The pattern recognition analysis method includes extracting the trend characteristics of network transmission delay, the fluctuation characteristics of network packet loss rate, and the peak characteristics of network bandwidth utilization, and integrating these characteristics into a comprehensive risk score. The weight coefficients for network transmission delay, network packet loss rate, and network bandwidth utilization are set using a combination of basic weights and dynamic adjustments, with the sum of the weight coefficients for the three parameters remaining a fixed value. Specifically, the basic weight for network transmission delay is determined based on the real-time requirements of industrial control commands, the basic weight for network packet loss rate is determined based on data transmission integrity requirements, and the basic weight for network bandwidth utilization is determined based on the network resource capacity limit. Simultaneously, the bandwidth utilization weight coefficient is adjusted based on the current network load level, the packet loss rate weight coefficient is adjusted based on the historical packet loss frequency, and the transmission delay weight coefficient is adjusted based on the command transmission delay limit. After adjustment, the weights are re-normalized to keep the sum of the three weight coefficients fixed, ensuring that the weight allocation aligns with the actual operating status of network communication and the needs of the industrial scenario.

[0040] The trend characteristics of network transmission latency are obtained by linearly fitting the latency values ​​of multiple consecutive sampling points. The slope of the fitted line reflects the direction and rate of latency change; a positive slope indicates an upward trend, and a larger absolute slope indicates a more pronounced trend. The fluctuation characteristics of network packet loss rate are obtained by calculating the dispersion of the loss rate within a certain time window; greater dispersion indicates poorer network communication stability. The peak characteristics of network bandwidth utilization are determined by detecting the maximum bandwidth utilization and its duration per unit time; a higher peak and a longer duration indicate a higher degree of network resource strain. These characteristics are converted into corresponding scores according to preset rules, and then weighted and summed to obtain the network communication status risk value. The weights are set according to the degree of influence of each characteristic on communication quality. The pre-defined rules are as follows: the slope of the network transmission delay value change trend is mapped to corresponding feature scores according to three states: increasing, stable, and decreasing. The higher the slope and the larger the absolute value, the higher the feature score. The dispersion of the network packet loss rate value is mapped to corresponding feature scores according to three ranges: low fluctuation, medium fluctuation, and high fluctuation. The higher the fluctuation, the higher the feature score. The peak value and duration of the network bandwidth utilization value are mapped to corresponding feature scores according to three states: low load, medium load, and high load. The higher the peak value and the longer the duration, the higher the feature score. The feature score mapping process is based on the historical feature range under normal communication conditions of the device, without fixed values ​​or proportional constraints, and only based on the degree to which the feature deviates from the normal range to complete the score conversion.

[0041] The specific steps for assessing the environmental status risk value are as follows: The ambient temperature, humidity, and vibration intensity values ​​are converted into quantifiable values. The ambient temperature value is converted to the deviation from the equipment's optimal operating temperature by subtracting the optimal operating temperature from the actual temperature; a positive result indicates the temperature is too high, and a negative result indicates the temperature is too low. The ambient humidity value is converted to a relative humidity percentage, directly reflecting the saturation level of water vapor in the air. The ambient vibration intensity value is converted to a ratio to the equipment's maximum permissible vibration intensity; the closer the ratio is to 1, the greater the potential impact of the vibration on the equipment.

[0042] Based on the environmental temperature, humidity, and vibration intensity values ​​after stoichiometric conversion, the environmental risk value is determined through statistical analysis.

[0043] The statistical analysis involves calculating the standard deviation of ambient temperature, the coefficient of variation of ambient humidity, and the mean offset of ambient vibration intensity, and then integrating the results into an environmental risk assessment index. The standard deviation of ambient temperature is calculated from temperature data over a specific time period, reflecting temperature fluctuations within that period; a large standard deviation indicates frequent and significant temperature changes. The coefficient of variation of ambient humidity is the ratio of the standard deviation to the mean of the humidity data, used to eliminate the influence of the mean on dispersion and more objectively reflect the fluctuation characteristics of humidity. The mean offset of ambient vibration intensity is the difference between the current mean vibration intensity and the mean vibration intensity during normal equipment operation; a positive difference indicates that the overall vibration intensity is relatively high.

[0044] When combining the above three calculation results into an environmental state risk value, the following formula is used:

[0045] in, Indicates the environmental status risk value. The standard deviation of ambient temperature values. The coefficient of variation represents the ambient humidity value. This represents the mean offset of environmental vibration intensity values. These are the fusion coefficients for temperature standard deviation, humidity coefficient of variation, and vibration mean offset, respectively. These are constants set based on the equipment's sensitivity to various environmental factors, and This formula integrates the statistical results from the three dimensions into a comprehensive environmental state risk value, thus fully reflecting the impact of environmental factors on equipment operation.

[0046] Example 4: See Figure 4The system inputs equipment operating status risk values, network communication status risk values, and environmental status risk values ​​into a multi-level fusion model. This model is stored in the equipment's data analysis module and can receive and process risk value data from different sources. The model's input interface is compatible with the three risk value data formats, ensuring that no information loss or distortion occurs during data transmission.

[0047] The multi-level fusion model first scales the risk values ​​for device operation status, network communication status, and environmental status. The scaling operation determines the conversion ratio based on the historical range of each risk value. For example, if the historical range for device operation status risk value is 0-100, network communication status risk value is 0-50, and environmental status risk value is 0-80, then each of the three is proportionally converted to the 0-1 range. This scaling provides a unified comparison benchmark for risk values ​​that were originally at different levels, facilitating subsequent fusion processing.

[0048] Then, based on the scaled values, a decision tree classification method is used to output a comprehensive risk index. The structure of the decision tree classification method is determined through pre-training and includes multiple branch nodes and leaf nodes. Branch nodes are configured with judgment conditions. For example, the first branch node uses the scaled equipment operating status risk value as the judgment criterion; if this value is greater than 0.6, it proceeds to the next level node A; otherwise, it proceeds to node B. The judgment condition for node A is whether the scaled network communication status risk value is greater than 0.5, the judgment condition for node B is whether the scaled environmental status risk value is greater than 0.4, and so on. Leaf nodes correspond to specific comprehensive risk index values. For example, after multiple levels of judgment, the final value at a certain leaf node is the comprehensive risk index.

[0049] The specific steps for determining the risk threshold based on comprehensive risk indicators are as follows: Compare the comprehensive risk index with multiple preset risk threshold ranges. The preset risk threshold ranges are determined by analyzing the equipment's historical failure data. For example, the comprehensive risk index ranges from 0 to 10, with the high-risk threshold range set to 7-10, the medium-risk threshold range set to 3-6, and the low-risk threshold range set to 0-2.

[0050] If the overall risk indicators fall into the high-risk threshold range, an emergency warning operation will be triggered. This emergency warning operation will activate the equipment's safety protection mechanisms, such as automatically reducing equipment load and activating the backup cooling system, while simultaneously sending the highest-level warning information to relevant systems.

[0051] If the overall risk indicators fall within the medium-risk threshold range, a standard warning operation will be triggered. This standard warning operation will display a notification on the device's local screen and record the relevant parameter trends for maintenance personnel to review periodically.

[0052] If the overall risk index falls within the low-risk threshold range, no warning action will be triggered, but a risk log will be recorded. The risk log contains the current overall risk index value, the values ​​of each sub-risk item, and the corresponding parameter data, stored in the device's local database, and can be retrieved through the data export function. Table 1 below shows examples of operations corresponding to different risk threshold ranges.

[0053] Table 1: Example of operation for different risk threshold ranges.

[0054]

[0055] Table 1 above clearly shows the corresponding handling methods for different comprehensive risk indicators, ensuring that early warning operations match the risk level, avoiding both excessive early warnings that lead to resource waste and insufficient early warnings that lead to equipment failure.

[0056] Example 5: See Figure 5 The system selects the appropriate warning output method based on the type of warning signal, including audible alarm, visual alarm, or network notification. Warning signals are categorized into emergency warning signals and general warning signals, each corresponding to a different combination of output methods. When the warning signal is an emergency warning, all three output methods are activated simultaneously; when it is a general warning, both visual alarm and network notification methods are activated.

[0057] The audible alarm system outputs alarm sounds via audio equipment, including speakers mounted on the equipment control panel and audible and visual alarms in the common areas of the workshop. Emergency alarms consist of high-frequency pulses lasting at least 30 seconds, repeating after a 5-second interval; regular alarms consist of low-frequency continuous sounds lasting 10 seconds before automatically stopping. The audio equipment volume automatically adjusts according to the ambient noise level, increasing to at least 1.5 times the ambient noise level in noisy workshop areas.

[0058] The visual alarm system displays warning icons via display devices, including the device's built-in LED indicators, the operating screen, and the large screen in the workshop monitoring center. During an emergency warning, the LED indicator flashes red rapidly, and the operating screen displays a full-screen red warning icon overlaid with the device number and risk type text. For a normal warning, the LED indicator flashes yellow slowly, and a yellow warning icon appears in the upper right corner of the operating screen. Upon receiving the warning signal, the large screen in the monitoring center highlights the corresponding device location on the factory floor plan and indicates the warning level.

[0059] The network notification method sends warning messages to remote monitoring terminals via network protocols, including MQTT and HTTP, ensuring that different types of monitoring terminals can receive the information. Remote monitoring terminals include mobile apps for administrators, computer clients, and industrial control platforms. The warning message includes the warning time, equipment identifier, risk level, specific risk parameter values, and recommended actions. If no read confirmation is received within 5 minutes of sending an emergency warning message, the system will automatically call the administrator for a voice reminder.

[0060] Simultaneously, the execution time and detailed parameters of the early warning operation are recorded to the database. The database adopts a distributed storage architecture to ensure that data is not lost due to single point of failure. The recorded detailed parameters include the early warning signal type, the execution status of each output method, the device operating parameters, network parameters, and environmental parameters at the time of early warning triggering. Each record automatically generates a unique identifier for easy subsequent querying and correlation analysis.

[0061] The specific steps for adjusting the early warning strategy in real time are as follows: Based on historical early warning records and real-time status data sets, the risk threshold range and early warning output method are updated. Historical early warning records are periodically extracted from the database, covering all early warning information from the past 6 months; the real-time status data set consists of parameter data within the current collection period, and the two are combined to form an analysis sample.

[0062] The update process involves analyzing the false alarm rate and missed alarm rate characteristics in historical warning records, and dynamically adjusting the risk thresholds based on the analysis results. The false alarm rate is determined by statistically analyzing the proportion and distribution of false warnings in the total warnings. For example, if three consecutive false alarms occur under the same parameter conditions, the risk threshold for the corresponding parameter will be increased by 10%. The missed alarm rate is determined by comparing actual fault cases with warning records. If the proportion of a certain type of fault that did not trigger a warning before its occurrence exceeds 20%, the corresponding risk threshold will be decreased by 8%. The adjusted risk thresholds will be automatically updated in the system, and the adjustment time and reason will be recorded.

[0063] Simultaneously, the priority order of early warning output methods is optimized based on the changing trends of equipment operating status risk values. By analyzing the change curves of equipment operating status risk values ​​over 10 consecutive data collection cycles, the development trend is determined. If the risk value shows a continuous upward trend, the priority of the audible alarm method is raised to the highest level; if the risk value fluctuates within the medium-risk range, the priority of the network notification method is adjusted to be higher than that of the visual alarm method; if the risk value decreases from high risk to medium risk, the emergency early warning output method combination is temporarily retained, and after continuous monitoring for two cycles, the normal settings are restored. After the priority adjustment, the system will execute early warning outputs according to the new order, ensuring that the most effective method takes precedence.

[0064] It should be noted that, in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such process, method, article, or apparatus.

[0065] Although embodiments of the invention have been shown and described, it will be understood by those skilled in the art that various changes, modifications, substitutions and alterations can be made to these embodiments without departing from the principles and spirit of the invention, the scope of which is defined by the appended claims and their equivalents.

Claims

1. A method for real-time status early warning of industrial networked devices, characterized in that, Includes the following steps: Acquire a set of real-time status data from multiple monitoring points in an industrial network device, the set of real-time status data including device operating parameter data, network communication parameter data, and environmental parameter data; A comprehensive analysis is performed on the real-time status data set of the industrial network equipment to assess the risk values ​​of equipment operation status, network communication status, and environmental status. The risk values ​​of equipment operation status, network communication status, and environmental status obtained from the assessment are compared with the preset risk thresholds. Based on the comparison and judgment results, an early warning signal is generated and an early warning operation is performed.

2. The real-time status early warning method for industrial networked devices according to claim 1, characterized in that, The specific equipment operating parameter data includes the internal temperature value, internal voltage value, and internal current value of the equipment; the specific network communication parameter data includes the network transmission delay value, network data packet loss rate value, and network bandwidth utilization value; and the specific environmental parameter data includes the ambient temperature value, ambient humidity value, and ambient vibration intensity value.

3. The real-time status early warning method for industrial networked equipment according to claim 2, characterized in that, The specific steps for assessing the risk value of equipment operating status are as follows: Standardize and convert the internal temperature, internal voltage, and internal current values ​​of the equipment. Based on the standardized internal temperature, internal voltage, and internal current values ​​of the equipment, the risk value of the equipment's operating status is determined by a weighted average calculation method. The weighted average calculation method involves the weighting coefficients of the internal temperature value, the internal voltage value, and the internal current value of the equipment, and the sum of all weighting coefficients is a fixed value.

4. The real-time status early warning method for industrial networked equipment according to claim 3, characterized in that, The dynamic adjustment steps for the weighting coefficients in the weighted average calculation method are as follows: The weighting coefficient of the internal temperature value of the equipment is corrected based on the statistical value of the equipment runtime. Specifically, the statistical value of the equipment runtime is multiplied by the preset aging influence factor and then added to the basic weight. The weighting coefficient of the internal voltage value of the equipment is corrected based on the voltage fluctuation amplitude value during the equipment operation cycle. Specifically, the voltage fluctuation amplitude value is input into a preset sensitivity mapping function and the output is an adjustment amount. The weighting coefficient of the internal current value of the device is corrected based on the current load rate value of the device. Specifically, the difference between the current load rate value and the benchmark load threshold is multiplied by the load influence coefficient. The sum of all the corrected weighting coefficients remains a fixed value.

5. The real-time status early warning method for industrial networked equipment according to claim 4, characterized in that, The specific steps for assessing the risk value of network communication status are as follows: Perform data normalization transformation on network transmission latency, network packet loss rate, and network bandwidth utilization. Based on the normalized network transmission delay, network packet loss rate, and network bandwidth utilization, the network communication status risk value is determined through pattern recognition analysis. Pattern recognition analysis methods include extracting the changing trend characteristics of network transmission delay values, the fluctuation characteristics of network packet loss rate values, and the peak characteristics of network bandwidth utilization values, and integrating these characteristics into a comprehensive risk score.

6. The real-time status early warning method for industrial networked equipment according to claim 5, characterized in that, The specific steps for assessing the environmental status risk value are as follows: Perform quantization conversion on ambient temperature, ambient humidity, and ambient vibration intensity values; Based on the environmental temperature, humidity, and vibration intensity values ​​after scalar conversion, the environmental state risk value is determined through statistical analysis. The statistical analysis method involves calculating the standard deviation of ambient temperature, the coefficient of variation of ambient humidity, and the mean offset of ambient vibration intensity, and then integrating the calculation results into environmental risk assessment indicators.

7. The real-time status early warning method for industrial networked equipment according to claim 6, characterized in that, The specific steps for integrating equipment operation status risk values, network communication status risk values, and environmental status risk values ​​to generate a comprehensive risk index are as follows: Input the risk values ​​of equipment operation status, network communication status, and environmental status into the multi-level fusion model; The multi-level fusion model first scales the risk values ​​of equipment operation status, network communication status, and environmental status. Then, based on the scaled values, a decision tree classification method is used to output a comprehensive risk index; The decision tree classification method maps scaling values ​​to comprehensive risk indicators based on preset branching rules and leaf node thresholds.

8. The method for real-time status early warning of industrial networked equipment according to claim 7, characterized in that, The specific steps for determining the risk threshold based on comprehensive risk indicators are as follows: Compare the comprehensive risk indicators with multiple preset risk threshold ranges; If the comprehensive risk indicators fall into the high-risk threshold range, an emergency warning operation will be triggered. If the comprehensive risk indicators fall into the medium risk threshold range, a normal early warning operation will be triggered; If the overall risk indicators fall within the low-risk threshold range, no warning action will be triggered, but a risk log will be recorded.

9. The real-time status early warning method for industrial networked equipment according to claim 8, characterized in that, The specific steps for performing the early warning operation are as follows: Select the warning output method according to the type of warning signal, including sound alarm, visual alarm or network notification; The sound alarm method outputs alarm sounds through audio devices, the visual alarm method displays warning icons through display devices, and the network notification method sends warning messages to the remote monitoring terminal through network protocols. Simultaneously, the execution time and detailed parameters of the early warning operation are recorded to the database.

10. The method for real-time status early warning of industrial networked equipment according to claim 8, characterized in that, The specific steps for adjusting the early warning strategy in real time are as follows: Based on historical early warning records and real-time status data sets, update the risk threshold range and early warning output method; The update process includes analyzing the false alarm rate and false negative rate characteristics in historical early warning records, and dynamically adjusting the risk threshold based on the analysis results; At the same time, the priority order of early warning output methods is optimized based on the changing trend of equipment operation status risk values.

Images