Early warning method and system for faults of Internet of Things

Abstract

The invention relates to the technical field of computer science, in particular to an Internet of Things fault early warning method and system, and the method comprises the following steps: recognizing the relation between the state of Internet of Things equipment and a fault mode, collecting the data of the Internet of Things equipment during normal operation and fault, recording the change rule of equipment parameters through the data, and obtaining an initial state data set. According to the method, data in normal and fault states are deeply analyzed, the state change rule is extracted, and the state network model is constructed, so that the deviation between real-time data and model prediction can be dynamically simulated, the abnormal fault probability can be marked, and the accuracy and efficiency of fault early warning are improved; according to the risk assessment and early warning strategy optimization based on the deviation analysis result, the response measures are more timely and effective, the automatic processing capability and the fault prevention capability are greatly improved, the service interruption and economic loss are reduced, and the method has a remarkable effect on guaranteeing the operation safety and improving the efficiency.

Description

Technical Field

[0001] The present invention relates to the field of computer science and technology, and in particular to an early warning method and system for Internet of Things failures. Background Art

[0002] The Internet of Things (IoT) is a major branch of computer science encompassing technologies that embed electronic devices, software, sensors, and network connectivity in various physical devices, including vehicles, household appliances, and objects. Connectivity and sensors collect data and interact with cloud computing services, enabling remote monitoring, management, and control of devices. The IoT enables real-world objects to automatically exchange and process data over the internet, significantly improving resource utilization, system efficiency, and economic benefits. This technology field continues to expand and is now being applied to a wide range of areas, including smart homes, smart cities, the Industrial Internet, and healthcare.

[0003] IoT fault early warning methods involve collecting data from sensors and devices within IoT systems, using algorithms to monitor device status and performance in real time, and predicting and identifying faults or issues. This approach aims to detect problems before they occur, preventing service interruptions and further economic losses caused by equipment failures. This is particularly important in critical infrastructure such as industrial manufacturing, transportation systems, and smart buildings. Real-time data analysis and early warnings enable maintenance personnel to promptly carry out repairs or replacements, optimizing system efficiency and safety.

[0004] Existing technologies rely primarily on basic data collection and processing, lacking in-depth dynamic analysis of the complex relationships between IoT device states and failure modes. This limitation prevents fault warning systems from accurately capturing all potential device anomalies and responding to rapidly changing device states in real time, increasing system vulnerability, especially in complex industrial environments. The lack of effective state change modeling and risk assessment makes it impossible to effectively predict and respond to sudden failures, leading to production stagnation or major safety incidents. For example, in traffic control systems, minor equipment failures, if not detected and addressed promptly, can evolve into more serious system failures, causing traffic congestion and even accidents, impacting urban operational efficiency and public safety. These technical limitations significantly increase maintenance costs and operational risks, especially in environments with a wide variety of equipment and stringent operational requirements. Summary of the Invention

[0005] In order to solve the problem of the lack of in-depth dynamic analysis of the complex relationship between the status of IoT devices and failure modes in the existing technology. This limitation causes the fault warning system to be unable to accurately capture all potential equipment anomalies, and it is difficult to respond to rapidly changing equipment status in real time, increasing the vulnerability of the system, especially in complex industrial environments. The lack of effective state change modeling and risk assessment makes it impossible to effectively predict and respond to sudden failures, resulting in production stagnation or major safety accidents. For example, in a traffic control system, if a minor equipment failure is not discovered and handled in time, it will evolve into a more serious system failure, causing traffic congestion or even accidents, affecting urban operation efficiency and public safety. The limitations of this technology significantly increase maintenance costs and operational risks, especially in situations where there are many types of equipment and strict operating requirements. The embodiment of the present invention provides an IoT failure warning method and system. The technical solution is as follows:

[0006] In one aspect, a method for early warning of Internet of Things failures is provided, the method comprising:

[0007] S1: Identify the relationship between IoT device status and failure mode, collect data on IoT devices during normal operation and failure, and obtain the initial state data set by recording the change patterns of device parameters;

[0008] S2: Based on the initial state dataset, a network is designed to reflect device state changes, the dependencies between parameters are identified, nodes are used to represent differentiated IoT device states, edges are used to represent the probability of state transitions, and a device state network model is constructed.

[0009] S3: Perform dynamic simulation using the device state network model, input real-time data using the Internet of Things, simulate state changes, record deviations between model predictions and real-time data, mark states with abnormal fault probabilities, and obtain deviation analysis results;

[0010] S4: Based on the deviation analysis results, the risk level of the IoT fault state is evaluated, the warning threshold is adjusted, the warning strategy is optimized, and the differentiated risk states are matched to obtain a fault warning plan;

[0011] S5: Implement the fault warning solution, monitor the operation status of the Internet of Things in real time, adjust the operation parameters of the Internet of Things devices to match the requirements of the fault warning, evaluate the speed and efficiency of the fault warning, and obtain the warning effect evaluation results.

[0012] As a further solution of the present invention, the initial state data set includes temperature change data, pressure level records, and current fluctuation data; the equipment status network model includes normal operation, inefficient operation, potential failures, and state transition probability of the node; the deviation analysis results include the degree of state deviation, failure probability state mark, and key fault indication data; the fault warning scheme includes the adjusted warning threshold, optimized monitoring strategy, and response measures for risk status; the warning effect evaluation results include fault warning response speed, warning efficiency, and real-time equipment status monitoring data.

[0013] As a further solution of the present invention, the steps of identifying the relationship between the state of IoT devices and failure modes, collecting data of IoT devices during normal operation and failure, and obtaining an initial state data set by recording the change patterns of device parameters are as follows:

[0014] S101: Identify the relationship between IoT device status and failure mode, monitor the operating parameters of IoT devices in normal and failure states, including temperature, voltage, and flow, and obtain the original parameter data set;

[0015] S102: Filtering data points showing abnormalities based on the original parameter data set, including temperature increases and voltage decreases, and marking the differences compared with normal data to obtain an abnormal data marking set;

[0016] S103: Analyze data change patterns using the abnormal data tag set, associate abnormal data with device failure modes, analyze the relationship between the initial state and the failure state of the IoT device, and obtain an initial state data set.

[0017] As a further solution of the present invention, based on the initial state data set, a network that reflects device state changes is designed, the dependencies between parameters are identified, nodes are used to represent differentiated IoT device states, and edges represent the probability of state transitions. The steps of constructing a device state network model are specifically as follows:

[0018] S201: Based on the initial state data set, record data of the IoT device in normal operation and fault state, record numerical changes of temperature, voltage and flow, and monitor parameter fluctuations in real time to obtain a device state data set;

[0019] S202: Using the device status data set, screening key performance indicators, recording temperature exceeding limits and voltage anomalies, adjusting data collection frequency and key parameters, optimizing data entry time intervals and monitoring point layout, and obtaining an optimized data set;

[0020] S203: Using the optimized data set, compare parameter changes under normal and fault conditions, mark abnormal data points before the fault occurs, formulate association rules between parameter anomalies and IoT device faults, and build a device status network model.

[0021] As a further solution of the present invention, the steps of performing dynamic simulation through the device state network model, inputting real-time data using the Internet of Things, simulating state changes, recording the deviation between the model prediction and the real-time data, marking the state with abnormal fault probability, and obtaining the deviation analysis results are specifically as follows:

[0022] S301: Using the device status network model, inputting real-time data from the Internet of Things, including temperature readings and voltage levels, to perform dynamic simulation of the IoT device status and obtain dynamic simulation data;

[0023] S302: Comparing the deviation between the dynamic simulation data and the real-time data, performing real-time analysis on the data points in the time series, calculating weighted deviation values ​​of multiple windows, and obtaining deviation values ​​of key parameters;

[0024] S303: Based on the deviation values ​​of the key parameters, mark the IoT device states with abnormal deviations and failure probabilities, record voltage anomalies when the temperature rises, identify potential failure states, and obtain deviation analysis results.

[0025] As a further solution of the present invention, the formula for calculating the weighted deviation values ​​of the multiple windows is:

[0026]

[0027] Among them, Z is the deviation evaluation value, x i Represents the value of real-time data at the i-th data point, μ i represents the value of the dynamic simulation data at the i-th data point, n represents the total number of data points in the sliding window, and w1 and w2 are weight coefficients.

[0028] As a further solution of the present invention, based on the deviation analysis results, the risk level of the IoT fault state is evaluated, the warning threshold is adjusted, the warning strategy is optimized, and the differentiated risk state is matched to obtain the fault warning solution. Specifically, the steps are as follows:

[0029] S401: Using the deviation analysis results, perform feature analysis and classification on the data points of abnormal temperature increase and voltage mutation, calculate the risk level, and obtain a risk level assessment record;

[0030] S402: Adjusting the IoT fault warning threshold based on the risk level assessment record, lowering the fault warning threshold for temperature anomalies, identifying potential faults, and obtaining an optimized warning strategy;

[0031] S403: Based on the optimized early warning strategy, formulate response measures for differentiated risk states, implement immediate inspection and maintenance for the risk states, and obtain a fault early warning plan.

[0032] As a further solution of the present invention, the formula for calculating the risk level is:

[0033]

[0034] Where R is the risk level value, T represents the real-time temperature of the device, V represents the real-time voltage of the device, ΔT represents the rate of change of temperature, ΔV represents the rate of change of voltage, and a, b, c, and d are weight coefficients.

[0035] As a further solution of the present invention, the steps of implementing the fault warning solution, monitoring the operating status of the Internet of Things in real time, adjusting the operating parameters of the Internet of Things devices, matching the requirements of the fault warning, evaluating the speed and efficiency of the fault warning, and obtaining the warning effect evaluation results are as follows:

[0036] S501: Implement the fault warning solution, monitor the operating status of the equipment in real time through the Internet of Things, including temperature, voltage, and flow, verify the consistency of the monitoring activities with the fault warning solution, and obtain real-time monitoring records;

[0037] S502: Adjusting operating parameters of IoT devices based on the real-time monitoring records to match fault warning requirements, responding to fault risks by adjusting temperature thresholds of running devices, and obtaining parameter adjustment records;

[0038] S503: Based on the parameter adjustment record, the reaction speed and processing efficiency of the IoT fault warning are evaluated, the time efficiency and success rate of the fault handling process are monitored, and the warning effect evaluation result is obtained.

[0039] On the other hand, an electric vehicle state monitoring system is provided, wherein the electric vehicle state monitoring system is used to execute the above electric vehicle state monitoring method, and the system includes:

[0040] The state recognition module collects data on IoT devices during normal operation and failure, records the changing patterns of device parameters, and obtains a device parameter data set;

[0041] The model building module designs a device state change network based on the device parameter data set, identifies the dependencies between parameters, uses nodes to represent differentiated IoT device states, and edges to represent the probability of state transitions, thereby building a device state network model.

[0042] The dynamic simulation module dynamically simulates the device state network model, inputs real-time data, simulates state changes, records the deviation between the model prediction and the real-time data, marks abnormal states, and obtains deviation analysis results;

[0043] The risk assessment module assesses the risk level of the IoT fault status based on the deviation analysis results, adjusts the warning threshold, optimizes the warning strategy, and obtains a fault warning plan;

[0044] The parameter optimization module implements the fault warning solution, monitors the operating status of IoT devices in real time, adjusts the device operating parameters to match the fault warning requirements, and obtains the adjusted operating parameters;

[0045] The effect evaluation module evaluates the speed and efficiency of the fault warning based on the adjusted operating parameters to obtain a warning effect evaluation result.

[0046] The beneficial effects brought about by the technical solution provided by the embodiment of the present invention include at least:

[0047] By establishing a deep relationship between IoT device states and failure modes, more accurate device monitoring and fault prediction can be achieved. By conducting in-depth analysis of data in both normal and faulty states, extracting patterns of state change, and constructing a state network model, the deviation between real-time data and model predictions can be dynamically simulated. Abnormal failure probabilities are flagged, improving the accuracy and efficiency of fault warnings. Risk assessment and early warning strategy optimization based on deviation analysis results make response measures more timely and effective, significantly enhancing automated processing capabilities and fault prevention capabilities, reducing service interruptions and economic losses. This has a significant impact on ensuring operational safety and improving efficiency, particularly in critical infrastructure applications. BRIEF DESCRIPTION OF THE DRAWINGS

[0048] Figure 1 It is a schematic diagram of the workflow of the present invention;

[0049] Figure 2 This is a detailed flow chart of S1 of the present invention;

[0050] Figure 3 This is a detailed flow chart of S2 of the present invention;

[0051] Figure 4 This is a detailed flow chart of S3 of the present invention;

[0052] Figure 5 This is a detailed flow chart of S4 of the present invention;

[0053] Figure 6 This is a detailed flow chart of S5 of the present invention;

[0054] Figure 7 It is a system flow chart of the present invention. DETAILED DESCRIPTION

[0055] The technical solution of the present invention is described below in conjunction with the accompanying drawings.

[0056] In the embodiments of the present invention, words such as "exemplarily" and "for example" are used to indicate examples, illustrations, or explanations. Any embodiment or design described as an "exemplary" in the present invention should not be interpreted as being preferred or advantageous over other embodiments or designs. Rather, the use of the word "exemplary" is intended to present concepts in a concrete manner. Furthermore, in the embodiments of the present invention, "and / or" can mean both or either of the two.

[0057] In order to make the technical problems, technical solutions and advantages to be solved by the present invention clearer, a detailed description will be given below with reference to the accompanying drawings and specific embodiments.

[0058] See also Figure 1 The embodiment of the present invention provides an early warning method for Internet of Things failures. The processing flow of the method may include the following steps:

[0059] S1: Identify the relationship between IoT device status and failure mode, collect data on IoT devices during normal operation and failure, and obtain the initial state data set by recording the change patterns of device parameters;

[0060] S2: Based on the initial state dataset, a network is designed to reflect device state changes. The dependencies between parameters are identified, and nodes are used to represent differentiated IoT device states. Edges represent the probability of state transitions, thus constructing a device state network model.

[0061] S3: Dynamically simulate the device status network model, using the Internet of Things to input real-time data, simulate state changes, record the deviation between the model prediction and real-time data, mark the state with abnormal failure probability, and obtain deviation analysis results;

[0062] S4: Based on the deviation analysis results, the risk level of the IoT fault status is assessed, the warning threshold is adjusted, the warning strategy is optimized, and the differentiated risk status is matched to obtain a fault warning plan;

[0063] S5: Implement the fault warning plan, monitor the operation status of the Internet of Things in real time, adjust the operating parameters of the Internet of Things equipment to match the requirements of fault warning, evaluate the speed and efficiency of fault warning, and obtain the warning effect evaluation results.

[0064] The initial state data set includes temperature change data, pressure level records, and current fluctuation data. The equipment status network model includes the normal operation, inefficient operation, potential failures, and state transition probability of the node. The deviation analysis results include the degree of state deviation, failure probability state mark, and key fault indication data. The fault warning plan includes the adjusted warning threshold, optimized monitoring strategy, and response measures for risk status. The warning effect evaluation results include fault warning response speed, warning efficiency, and real-time equipment status monitoring data.

[0065] See also Figure 2 ,Identify the relationship between the state and failure mode of IoT devices, collect data of IoT devices during normal operation and failure, and obtain the initial state data set by recording the change pattern of device parameters.,Specific steps are as follows:

[0066] S101: Identify the relationship between IoT device status and failure mode, monitor the operating parameters of IoT devices in normal and failure states, including temperature, voltage, and flow, and obtain the original parameter data set. The execution process is as follows;

[0067] In the process of identifying the relationship between the state of IoT devices and failure modes, it is necessary to collect various parameters about the operation of the equipment, such as temperature, voltage and flow, through the IoT device monitoring system. The data is collected in real time by sensors and transmitted to the central processing system through the data interface. The central processing system records the data in time series. The temperature data is measured by thermistors, the voltage data is measured by voltmeters, and the flow data is obtained by flow meters. The accuracy of the data directly affects the efficiency and accuracy of subsequent fault diagnosis, and the original parameter data set is obtained.

[0068] S102: Based on the original parameter data set, screen out data points that exhibit abnormalities, including temperature increases and voltage decreases, and mark the differences compared to normal data. The execution process for obtaining the abnormal data mark set is as follows;

[0069] Filter out abnormal data points according to the formula:

[0070] T n =T o +ΔA

[0071] and

[0072] V n =V o -ΔV

[0073] Calculate the temperature rise and voltage drop, where T o Represents the temperature during normal operation, ΔA represents the temperature change, V o Represents the voltage during normal operation, and ΔV represents the voltage change;

[0074] Set the normal operating temperature T o =25℃ and normal voltage V o =220V, if the temperature rises by 5 degrees, the voltage drops by 10V, according to the formula:

[0075] T n =25+5=30

[0076] The voltage value is according to the formula:

[0077] V n =220-10=210

[0078] Anomalous data points in operating parameters can be quickly identified.

[0079] S103: Analyze the data change pattern using the abnormal data tag set, associate the abnormal data with the device failure mode, analyze the relationship between the initial state and the failure state of the IoT device, and obtain the initial state data set. The execution process is as follows;

[0080] Analyze the data change patterns and classify the abnormal data obtained by screening. Different data change patterns correspond to different equipment failure modes. For example, an abnormal temperature increase indicates a cooling system failure, and a voltage drop indicates a power supply problem. Use statistical analysis methods such as standard deviation analysis and trend analysis on the data to determine the severity and failure mode of the abnormal data. It is also necessary to perform time series analysis on the data to identify the specific time point when the failure occurred. Through comprehensive analysis, we can form an in-depth understanding of the relationship between the initial state and the failure state of the IoT device, provide a scientific basis for equipment maintenance and failure prevention, and obtain the initial state data set.

[0081] See also Figure 3 Based on the initial state data set, a network that reflects the device state changes is designed, the dependencies between parameters are identified, nodes are used to represent differentiated IoT device states, and edges represent the probability of state transitions. The specific steps for building a device state network model are as follows:

[0082] S201: Based on the initial state data set, record the data of the IoT device in normal operation and fault state, record the value changes of temperature, voltage and flow, and monitor parameter fluctuations in real time to obtain the device state data set. The execution process is as follows;

[0083] To record the data of IoT devices in normal operation and fault conditions, it is necessary to monitor the operating parameters of the devices in real time, including temperature, voltage, and flow. The parameters are collected from the devices in real time through sensors and sent to the central monitoring system through wireless or wired networks. The system records and stores the data in real time. The temperature sensor is responsible for measuring the operating temperature of the device, the voltage sensor monitors the power supply voltage, and the flow sensor records the flow rate of the flowing medium. The data can be used to understand the performance of the device in normal operation and fault conditions in detail, and obtain the device status data set.

[0084] S202: Using the equipment status data set, screen key performance indicators, record temperature excursions and voltage anomalies, adjust the data collection frequency and key parameters, optimize the data entry time interval and monitoring point layout, and obtain the optimized data set. The execution process is as follows;

[0085] Use the equipment status data set to filter key performance indicators according to the formula:

[0086] ΔT=TT avg

[0087] and

[0088] ΔV=VV avg

[0089] Calculate the temperature exceeding the limit and the voltage abnormality, where ΔT represents the temperature difference that needs to be adjusted, T represents the current temperature, and T avg represents the average value of normal temperature, V represents the current voltage, V avg Represents the normal voltage average value;

[0090] Set the normal temperature average value T avg =25℃ and the average value of normal voltage V avg =220V, the currently monitored temperature is 30 degrees, the voltage is 210V, according to the formula:

[0091] ΔT=30-25=5

[0092] Voltage abnormality is calculated according to the formula:

[0093] ΔV=210-220=10

[0094] Such calculations help identify data points that exhibit critical anomalies in performance.

[0095] S203: Using the optimized data set, compare parameter changes under normal and fault conditions, mark abnormal data points before the fault occurs, formulate association rules between parameter anomalies and IoT device faults, and construct a device status network model. The execution process is as follows;

[0096] It is necessary to mark abnormal data points before a fault occurs. The data points show parameter values ​​that are significantly different from the normal state, such as sharp changes in temperature and voltage. By marking, we can further formulate association rules between parameter anomalies and IoT device failures. By analyzing parameter changes, we can predict potential equipment failures and achieve early warning of equipment failures. This is extremely critical for maintenance and operation management, and it is necessary to build a network model of equipment status.

[0097] See also Figure 4 ,dynamic simulation is performed through the device status network model, real-time data is inputted using the Internet of Things, state changes are simulated, the deviation between the model prediction and real-time data is recorded, and the state with abnormal fault probability is marked. The specific steps to obtain the deviation analysis results are as follows:

[0098] S301: Using the device status network model, real-time data from the IoT, including temperature readings and voltage levels, is input to perform a dynamic simulation of the IoT device status. The execution process for obtaining dynamic simulation data is as follows;

[0099] Input real-time data from the Internet of Things, including temperature readings and voltage levels, to ensure that the network model can accurately receive and process input data from various sensors. The data includes real-time readings from temperature sensors and voltage sensors. The model dynamically simulates the current status of the Internet of Things device through algorithms. During the simulation process, the model will consider various factors that affect the operation of the device, such as ambient temperature changes, voltage fluctuations, etc. Through comprehensive information, the model can predict the operation problems of the device, provide a basis for further fault diagnosis, and obtain dynamic simulation data.

[0100] S302: Compare the deviation between the dynamic simulation data and the real-time data, perform real-time analysis on the data points in the time series, calculate the weighted deviation values ​​of multiple windows, and obtain the deviation values ​​of key parameters. The execution process is as follows;

[0101] The formula for calculating the weighted deviation value of multiple windows is:

[0102]

[0103] Among them, Z is the deviation evaluation value, x i Represents the value of real-time data at the i-th data point, μ i represents the value of the dynamic simulation data at the i-th data point, n represents the total number of data points in the sliding window, and w1 and w2 are weight coefficients;

[0104] Parameter meaning and setting value:

[0105] x i is the real-time value of the i-th data point obtained from the real-time monitoring device, μ iis the simulated value of the ith data point obtained from the simulation, and the sliding window is set to include n = 5 data points. The real-time data x is obtained by monitoring the equipment. i and the simulated data μ at the corresponding time point i , in this example, x i With μ i The specific values ​​of come from the records of the last five sampling points. The actual values ​​are: i =[10, 12, 11, 13, 14], the simulation value is μ i =[9, 11, 11, 12, 15];

[0106] n is the total number of data points in the sliding window, which is set to 5;

[0107] w1 and w2 are weight coefficients, which are set based on historical data analysis. w1 = 0.3 adjusts the impact of the first-order absolute deviation, and w2 = 0.7 adjusts the impact of the second-order variance.

[0108] Substitute the parameters into the formula for calculation:

[0109]

[0110] The result of 0.74 shows that the weighted calculation of the first-order and second-order deviations reflects the degree of deviation between the real-time data and the simulation data in the current sliding window. Through this value, the accuracy of the simulation system can be evaluated and the parameters can be adjusted in time to reduce future deviations and optimize the system's response and prediction capabilities.

[0111] S303: Based on the deviation values ​​of key parameters, mark the IoT device states with abnormal deviations and failure probabilities, record voltage anomalies when the temperature rises, identify potential failure states, and obtain the deviation analysis results. The execution process is as follows;

[0112] Mark IoT device states with abnormal deviations and failure probabilities. During this process, the temperature and voltage data collected from the sensors are analyzed, with particular attention paid to whether the voltage fluctuates abnormally when the temperature rises. The deviation value is calculated using data analysis software, and data points where the deviation value exceeds the preset threshold are marked. The marked data points indicate that the device is entering a potential failure state. In this way, device failures can be identified and handled early to avoid more serious equipment damage or downtime, and obtain deviation analysis results.

[0113] See also Figure 5 ,According to the deviation analysis results, the risk level of IoT fault status is evaluated, the warning threshold is adjusted, the warning strategy is optimized, and the differentiated risk status is matched.,The specific steps for obtaining the fault warning solution are as follows:

[0114] S401: Using the deviation analysis results, perform feature analysis and classification on the data points of abnormal temperature increase and voltage mutation, and calculate the risk level. The execution process of obtaining the risk level assessment record is as follows;

[0115] The formula for calculating risk level is:

[0116]

[0117] Where R is the risk level value, T represents the real-time temperature of the device, V represents the real-time voltage of the device, ΔT represents the rate of change of temperature, ΔV represents the rate of change of voltage, and a, b, c, and d are weight coefficients;

[0118] Parameter meaning and setting value:

[0119] T represents the real-time temperature of the device, which is obtained through real-time monitoring of the temperature sensor. The set value is 45°C. Based on the real-time monitoring data, it indicates that the device is experiencing overheating;

[0120] V represents the real-time voltage of the device, which is obtained through real-time monitoring by the voltage sensor. The set value is 220V, indicating that the device operating voltage is normal;

[0121] ΔT represents the rate of change of temperature, which is determined by analyzing the temperature data over the past hour. The set value is 0.5°C / min, indicating that the temperature is slowly rising;

[0122] ΔV represents the rate of change of voltage. By analyzing the voltage data over the past hour, we found that the set value is 2V / min, indicating that the voltage fluctuates slightly.

[0123] The weights a, b, c, and d are set based on equipment failure data and expert opinions. The weights reflect the relative importance of each feature in the risk assessment, among which the temperature change rate is considered to be the most critical factor;

[0124] Substitute the parameters into the formula for calculation:

[0125]

[0126] The results show that the equipment is at a medium risk level. This low value indicates that although the temperature is gradually increasing, the current risk level has not yet reached a high-risk state. The equipment needs to be continuously monitored. Once the risk level approaches or exceeds the preset high-risk threshold (such as 50), maintenance or emergency measures should be taken immediately.

[0127] S402: Based on the risk level assessment record, adjust the threshold for IoT fault warnings, lower the threshold for temperature anomaly fault warnings, identify potential faults, and obtain the optimized warning strategy. The execution process is as follows:

[0128] By analyzing the abnormal temperature events and related failure conditions recorded in historical data and lowering the temperature thresholds indicating higher risk levels based on the risk level assessment records, potential failures can be identified earlier and preventive measures can be taken. The adjusted thresholds are verified for effectiveness through simulation tests to ensure that the new thresholds are neither too sensitive to frequently issue false alarms, nor sensitive enough to capture all key risk points, resulting in an optimized early warning strategy.

[0129] S403: Based on the optimized early warning strategy, response measures are formulated for differentiated risk states, and immediate inspection and maintenance are performed for risk states. The execution process of the fault early warning plan is as follows;

[0130] Develop response measures for differentiated risk states, including implementing different inspection and maintenance plans for IoT devices with different risk levels. For example, for high-risk devices, implement more frequent inspections and immediate maintenance to prevent failures. For medium- and low-risk devices, adjust the inspection frequency based on actual conditions. This differentiated measure can effectively allocate resources, ensure that focus is placed on equipment with problems, maximize maintenance efficiency and equipment operation stability, provide effective early warning and response for different risk levels, and obtain fault early warning solutions.

[0131] See also Figure 6 , implement the fault warning plan, monitor the operation status of the Internet of Things in real time, adjust the operating parameters of the Internet of Things equipment, match the requirements of fault warning, evaluate the speed and efficiency of fault warning, and obtain the warning effect evaluation results. The specific steps are:

[0132] S501: Implement the fault warning plan. Use the Internet of Things to monitor the operating status of the equipment in real time, including temperature, voltage, and flow. Verify the consistency of monitoring activities with the fault warning plan and obtain real-time monitoring records. The execution process is as follows;

[0133] Through the Internet of Things, the operating status of the equipment can be monitored in real time. The sensors installed on the equipment continuously collect key operating parameters such as temperature, voltage and flow. The data is transmitted to the monitoring center in real time through the Internet of Things. The monitoring center software compares the data with the preset fault indicators to verify whether the monitoring activities are consistent with the fault warning plan, ensuring that all monitoring data can be updated in real time. In this process, the data update frequency and timeliness of each sensor are crucial to the accuracy of the fault warning, and real-time monitoring records are obtained.

[0134] S502: Based on the real-time monitoring records, the operating parameters of the IoT devices are adjusted to match the fault warning requirements. The temperature thresholds of the running devices are adjusted to respond to the fault risk. The execution process of obtaining the parameter adjustment records is as follows;

[0135] Adjust the operating parameters of the IoT device according to the formula:

[0136] T new =T current +ΔT

[0137] Adjust the temperature threshold of the running device, where T current Represents the current temperature of the device, and ΔT represents the temperature difference that needs to be adjusted;

[0138] Set the current temperature of the device T current =50℃. According to the fault warning requirements, the temperature needs to be lowered by 5 degrees to respond to potential fault risks. Substitute into the formula:

[0139] T new =50-5=45

[0140] Through this calculation, device parameters can be adjusted in real time to ensure that the device operates in a safe state.

[0141] S503: Based on the parameter adjustment records, the response speed and processing efficiency of the IoT fault warning are evaluated, the time efficiency and success rate of the fault handling process are monitored, and the execution process of obtaining the warning effect evaluation result is as follows;

[0142] Evaluate the response speed and processing efficiency of IoT fault warnings, analyze parameter adjustment records, record the start and end time of each fault handling, calculate the total time consumed in the processing, and record the success rate of each handling, that is, the ratio of the number of successful fault resolutions to the total number of faults. The data can be used to evaluate the overall efficiency of the fault warning system, which is an important basis for optimizing the fault warning system, can help improve system design, improve the efficiency and success rate of fault handling, and obtain warning effect evaluation results.

[0143] See also Figure 7 On the other hand, an electric vehicle state monitoring system is provided, which is used to execute the above electric vehicle state monitoring method, and the system includes:

[0144] The state recognition module collects data on IoT devices during normal operation and failure, records the changing patterns of device parameters, and obtains a device parameter data set;

[0145] The model building module designs a device state change network based on the device parameter dataset, identifies the dependencies between parameters, uses nodes to represent differentiated IoT device states, and edges to represent the probability of state transitions, thus building a device state network model.

[0146] The dynamic simulation module dynamically simulates the device status network model, inputs real-time data, simulates status changes, records the deviation between the model prediction and real-time data, marks abnormal status, and obtains deviation analysis results;

[0147] The risk assessment module assesses the risk level of IoT fault status based on the deviation analysis results, adjusts the warning threshold, optimizes the warning strategy, and obtains a fault warning plan;

[0148] The parameter optimization module implements the fault warning plan, monitors the operating status of IoT devices in real time, adjusts the device operating parameters to match the fault warning requirements, and obtains the adjusted operating parameters;

[0149] The effect evaluation module evaluates the speed and efficiency of fault warning based on the adjusted operating parameters and obtains the warning effect evaluation result.

[0150] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any modifications or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in the present invention should be included in the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be based on the scope of protection of the claims.

Claims

1. A method for early warning of Internet of Things failures, characterized in that: The following steps are involved: Identify the relationship between IoT device status and failure modes, collect data on IoT devices during normal operation and failure, and obtain the initial state data set by recording the changing patterns of device parameters; Based on the initial state data set, a network that reflects device state changes is designed, the dependencies between parameters are identified, nodes are used to represent differentiated IoT device states, edges represent the probability of state transitions, and a device state network model is constructed; Perform dynamic simulation through the device state network model, use the Internet of Things to input real-time data, simulate state changes, record the deviation between the model prediction and real-time data, mark the state with abnormal fault probability, and obtain deviation analysis results; Based on the deviation analysis results, the risk level of the IoT fault state is evaluated, the warning threshold is adjusted, the warning strategy is optimized, and the differentiated risk states are matched to obtain a fault warning plan; Implement the fault warning solution, monitor the operation status of the Internet of Things in real time, adjust the operating parameters of the Internet of Things equipment, match the requirements of fault warning, evaluate the speed and efficiency of fault warning, and obtain the warning effect evaluation results.

2. The early warning method for Internet of Things failure according to claim 1, characterized in that: The initial state data set includes temperature change data, pressure level records, and current fluctuation data; the equipment status network model includes the normal operation, inefficient operation, potential failure, and state transition probability of the node; the deviation analysis results include the state deviation degree, failure probability state mark, and key fault indication data; the fault warning plan includes the adjusted warning threshold, the optimized monitoring strategy, and the response measures for the risk state; the warning effect evaluation results include the fault warning response speed, warning efficiency, and real-time equipment status monitoring data.

3. The early warning method for Internet of Things failure according to claim 1, characterized in that: Identify the relationship between IoT device status and failure mode, collect data on IoT devices during normal operation and failure, and record the changing patterns of device parameters to obtain the initial state data set. The specific steps are as follows: Identify the relationship between IoT device status and failure modes, monitor the operating parameters of IoT devices in normal and faulty states, including temperature, voltage, and flow, and obtain the original parameter data set; According to the original parameter data set, data points showing abnormal performance are screened, including temperature increase and voltage decrease, and differences compared with normal data are marked to obtain an abnormal data marked set; By using the abnormal data tag set, the data change pattern is analyzed, the abnormal data is associated with the device failure mode, and the relationship between the initial state and the failure state of the Internet of Things device is analyzed to obtain the initial state data set.

4. The early warning method for Internet of Things failure according to claim 1, characterized in that: Based on the initial state dataset, a network that reflects device state changes is designed. The dependencies between parameters are identified. Nodes are used to represent differentiated IoT device states, and edges represent the probability of state transitions. The specific steps for constructing a device state network model are as follows: Based on the initial state data set, record the data of the IoT device in normal operation and fault state, record the numerical changes of temperature, voltage and flow, and monitor parameter fluctuations in real time to obtain a device state data set; Using the equipment status data set, key performance indicators are screened, temperature exceeding the standard and voltage anomalies are recorded, data collection frequency and key parameters are adjusted, and data entry time intervals and monitoring point layout are optimized to obtain an optimized data set; By using the optimized data set, the parameter changes under normal and fault conditions are compared, abnormal data points before the fault occurs are marked, association rules between parameter anomalies and IoT device faults are formulated, and a device status network model is constructed.

5. The early warning method for Internet of Things failure according to claim 1, characterized in that: The steps for performing dynamic simulation through the device state network model, inputting real-time data using the Internet of Things, simulating state changes, recording the deviation between the model prediction and the real-time data, marking the abnormal state of fault probability, and obtaining the deviation analysis results are as follows: Using the device status network model, real-time data from the Internet of Things, including temperature readings and voltage levels, is input to perform dynamic simulation of the status of the IoT device to obtain dynamic simulation data; Comparing the deviation between the dynamic simulation data and the real-time data, performing real-time analysis on the data points in the time series, calculating weighted deviation values ​​of multiple windows, and obtaining deviation values ​​of key parameters; According to the deviation values ​​of the key parameters, the IoT device states with abnormal deviations and failure probabilities are marked, the voltage anomalies when the temperature rises are recorded, the potential fault states are identified, and the deviation analysis results are obtained.

6. The method for early warning of Internet of Things failure according to claim 5, characterized in that: The formula for calculating the weighted deviation values ​​of the multiple windows is: Among them, Z is the deviation evaluation value, x i Represents the value of real-time data at the i-th data point, μ i represents the value of the dynamic simulation data at the i-th data point, n represents the total number of data points in the sliding window, and w1 and w2 are weight coefficients.

7. The early warning method for Internet of Things failure according to claim 1, characterized in that: Based on the deviation analysis results, the risk level of the IoT fault state is assessed, the warning threshold is adjusted, the warning strategy is optimized, and the differentiated risk states are matched to obtain the fault warning solution. The specific steps are as follows: Using the deviation analysis results, feature analysis and classification are performed on the data points of abnormal temperature increase and voltage mutation, and risk levels are calculated to obtain a risk level assessment record; Based on the risk level assessment records, the threshold for IoT fault warning is adjusted, the fault warning threshold for temperature anomalies is lowered, potential faults are identified, and an optimized warning strategy is obtained; Based on the optimized early warning strategy, response measures are formulated for differentiated risk states, and immediate inspection and maintenance are carried out for risk states to obtain a fault early warning plan.

8. The method for early warning of Internet of Things failure according to claim 7, characterized in that: The formula for calculating the risk level is: Where R is the risk level value, T represents the real-time temperature of the device, V represents the real-time voltage of the device, ΔT represents the rate of change of temperature, ΔV represents the rate of change of voltage, and a, b, c, and d are weight coefficients.

9. The early warning method for Internet of Things failure according to claim 1, characterized in that: The specific steps for implementing the fault warning solution, monitoring the IoT operating status in real time, adjusting the IoT device operating parameters to match the fault warning requirements, evaluating the speed and efficiency of fault warning, and obtaining the warning effect evaluation results are as follows: Implement the fault warning plan and monitor the operating status of the equipment in real time, including temperature, voltage, and flow, through the Internet of Things, verify the consistency of monitoring activities with the fault warning plan, and obtain real-time monitoring records; Adjust the operating parameters of IoT devices based on the real-time monitoring records to match fault warning requirements, and respond to fault risks by adjusting the temperature thresholds of running devices to obtain parameter adjustment records; Based on the parameter adjustment records, the reaction speed and processing efficiency of the Internet of Things fault warning are evaluated, the time efficiency and success rate of the fault handling process are monitored, and the warning effect evaluation results are obtained.

10. An early warning system for Internet of Things failures, characterized in that: The method for early warning of Internet of Things failure according to any one of claims 1 to 9, wherein the system comprises: The state recognition module collects data on IoT devices during normal operation and failure, records the changing patterns of device parameters, and obtains a device parameter data set; The model building module designs a device state change network based on the device parameter data set, identifies the dependencies between parameters, uses nodes to represent differentiated IoT device states, and edges to represent the probability of state transitions, thereby building a device state network model. The dynamic simulation module dynamically simulates the device state network model, inputs real-time data, simulates state changes, records the deviation between the model prediction and the real-time data, marks abnormal states, and obtains deviation analysis results; The risk assessment module assesses the risk level of the IoT fault status based on the deviation analysis results, adjusts the warning threshold, optimizes the warning strategy, and obtains a fault warning plan; The parameter optimization module implements the fault warning solution, monitors the operating status of IoT devices in real time, adjusts the device operating parameters to match the fault warning requirements, and obtains the adjusted operating parameters; The effect evaluation module evaluates the speed and efficiency of the fault warning based on the adjusted operating parameters to obtain a warning effect evaluation result.

Images