A Method and System for Self-Sensing the Status of High-Risk Equipment in Thermal Power Plants Based on the Internet of Things

Abstract

This invention discloses a method and system for self-sensing the status of high-risk equipment in thermal power plants based on the Internet of Things (IoT), specifically relating to the field of industrial process monitoring and fault diagnosis technology. It addresses the problem that existing single-equipment independent monitoring technologies cannot effectively detect anomalies in the coupling relationship between equipment and accurately locate the root cause of faults. This invention collects operating parameters of high-risk equipment through IoT sensing terminals, divides coupling units based on the process chain, verifies the temporal correlation of parameter responses, and constructs a dataset of coupled and correlated equipment parameters. It then identifies and distinguishes between independent anomalies of single equipment and anomalies in the coupling relationship between equipment. By extracting the slope of temporal abrupt changes, analyzing coupling phase differences and peak matching, it traces the root cause of the anomaly back along the process chain. Finally, by combining historical data comparison and supplementary parameter verification, it integrates and generates a self-sensing result of the coupled operating status of high-risk equipment in thermal power plants, including anomaly characteristics, equipment location, and root cause indication.

Description

Technical Field

[0001] This invention relates to the field of industrial process monitoring and fault diagnosis technology, and more specifically, to a method and system for self-sensing the status of high-risk equipment in thermal power plants based on the Internet of Things. Background Technology

[0002] The safe and stable operation of thermal power plants is crucial for ensuring energy supply. Boilers, steam turbines, coal mills, and other high-risk equipment, as the core of the production system, require effective sensing and control of their operational status for preventing malfunctions and mitigating safety risks. In the process of industrial intelligentization, the Internet of Things (IoT) technology has been gradually applied to the status monitoring of high-risk equipment in thermal power plants. By deploying sensing terminals to collect equipment operating parameters and transmitting them to the control platform via the network, abnormal early warnings are achieved, becoming an important technical means for equipment operation control and maintenance management. In actual production, high-risk equipment in thermal power plants does not operate in isolation but forms an interconnected and coupled operating system through material flow and energy transfer. The operating status of each piece of equipment influences the others, constituting the core operating mode of thermal power plant production control.

[0003] Existing IoT-based status awareness technologies for high-risk equipment in thermal power plants primarily employ a single-device, single-point independent monitoring and analysis approach. This approach fails to adapt to the core characteristics of coupled operation of high-risk equipment, resulting in insufficient ability to perceive anomalies associated with different equipment and trace the root cause of faults. Consequently, existing technologies struggle to accurately distinguish between equipment faults and anomalies caused by coupled relationships, leading to misjudgments and an inability to precisely locate the root cause of faults. This results in a lack of targeted operation and maintenance management, affecting the reliability of equipment operation control and failing to meet the core requirements of status awareness technology for the safe and stable operation of high-risk equipment in thermal power plants. Summary of the Invention

[0004] In order to overcome the above-mentioned defects of the prior art, the present invention provides a method and system for self-sensing the status of high-risk equipment in thermal power plants based on the Internet of Things to solve the problems mentioned in the background art.

[0005] To achieve the above objectives, the present invention provides the following technical solution:

[0006] The IoT-based method for self-sensing the status of high-risk equipment in thermal power plants includes the following steps:

[0007] S1. Collect the operating parameters of each device in the coupled operation system of high-risk equipment in thermal power plants through IoT sensing terminals;

[0008] S2. Based on the material flow and energy transfer process link of high-risk equipment in thermal power plants, couple units are divided, and the change time sequence of each operating parameter in the same couple unit is collected. The correlation of parameter response time sequence is verified and the correlation mapping is performed to obtain the equipment parameter coupling correlation dataset.

[0009] S3. Identify abnormal features in the coupled and correlated dataset of equipment parameters, and distinguish between independent abnormal features of a single device and coupled and correlated abnormal features between devices;

[0010] S4. Extract the temporal abrupt change slope of the abnormal coupling characteristics between equipment, combine it with the process link analysis of the coupling phase difference, and determine the root equipment pointing information of the coupling anomaly by peak matching and tracing back upstream along the process link.

[0011] S5. Based on the root source device pointing information, retrieve the historical abnormal operation data of the corresponding root source device for comparison, collect supplementary operation parameters for verification, and obtain the verified root source device pointing information.

[0012] S6. Integrate the independent anomaly features of a single device, the coupled and associated anomaly features, and the verified root cause device pointing information to generate the self-sensing results of the coupled operation status of high-risk equipment in thermal power plants.

[0013] Furthermore, S1 includes:

[0014] Identify the types of high-risk equipment included in the coupled operation system of high-risk equipment in thermal power plants;

[0015] Based on the core operating mechanism of each type of high-risk equipment, determine the categories of operating parameters that need to be monitored for each type of high-risk equipment;

[0016] For each type of high-risk equipment, deploy IoT sensing terminals corresponding to the categories of operating parameters that need to be monitored;

[0017] The deployed IoT sensing terminals synchronously collect the operating parameters of each device in the coupled operation system of high-risk equipment in thermal power plants according to the preset collection frequency.

[0018] Furthermore, based on the core operating mechanism of each type of high-risk equipment, the categories of operating parameters that need to be monitored for each type of high-risk equipment are determined, including:

[0019] For boiler equipment, the categories of operating parameters that need to be monitored include drum pressure, main steam temperature, and furnace negative pressure.

[0020] For steam turbine equipment, the categories of operating parameters that need to be monitored include shaft vibration, bearing temperature, and cylinder expansion.

[0021] For coal mill equipment, the categories of operating parameters that need to be monitored include bearing vibration, gearbox oil temperature, and outlet pulverized coal temperature.

[0022] Furthermore, S2 includes:

[0023] Analyze the material flow path and energy transfer path of high-risk equipment in thermal power plants, and divide the process into coupled units including upstream and downstream equipment based on the continuity of the process flow.

[0024] For each coupling unit, the operating parameters of each device in the high-risk equipment coupling operation system of the thermal power plant are extracted from the operating parameters of each device collected through the Internet of Things sensing terminal.

[0025] Align the timestamps of the operating parameters of each device within the same coupling unit to form a change time series with a unified time base;

[0026] Based on the change time series, we analyze the chronological order and delay relationship between the changes in the operating parameters of upstream equipment and the changes in the operating parameters of downstream equipment to verify the correlation of parameter response time series.

[0027] The upstream and downstream device operating parameters, whose parameter response timing correlation has been verified, are associated and mapped to generate and store a device parameter coupled association dataset.

[0028] Furthermore, S3 includes:

[0029] Based on the coupling and association of device parameters with the operating parameters of each device in the dataset, it is determined whether the operating parameters deviate from their corresponding normal operating threshold range.

[0030] Mark operating parameters that deviate from the normal operating threshold range as abnormal data points;

[0031] For each coupled unit, analyze the marked abnormal data points in its device parameter coupling correlation dataset;

[0032] If an abnormal data point appears only on a single device in a coupling unit, and the operating parameters of other devices in the coupling unit that have parameter response timing correlation with it are normal, then the feature corresponding to the abnormal data point is identified as a single device independent abnormal feature.

[0033] If abnormal data points appear on multiple devices in the coupling unit, and there is a verified correlation between the abnormal data points of these devices in terms of parameter response timing, then the common features corresponding to these abnormal data points are identified as abnormal features of inter-device coupling correlation.

[0034] Furthermore, operating parameters that deviate from the normal operating threshold range are marked as abnormal data points, including:

[0035] The real-time values ​​of the operating parameters are continuously compared with the upper and lower boundary values ​​of their corresponding normal operating threshold range.

[0036] When the real-time value of an operating parameter exceeds the boundary of its corresponding normal operating threshold range more than a preset number of times, the operating parameter is determined to deviate from the normal operating threshold range.

[0037] In the time series of the coupled dataset of equipment parameters, anomaly markers are added to the time points and their parameter values ​​that are determined to be deviated, forming abnormal data points.

[0038] Furthermore, S4 includes:

[0039] From the time series of abnormal data points corresponding to the abnormal characteristics of coupling between devices, the instantaneous change rate of the change time sequence of each abnormal data point is calculated as the time sequence abrupt change slope.

[0040] Based on the material flow and energy transfer process link direction on which the coupling unit is divided, the upstream and downstream order of each device that has abnormal characteristics of inter-device coupling correlation in the process link is determined;

[0041] The time difference between the abnormal start time of the upstream equipment and the abnormal start time of the downstream equipment in the calculation process link is used as the coupling phase difference.

[0042] Match the peak value of the timing mutation slope of the upstream equipment in the process link with the peak value of the timing mutation slope of the downstream equipment to verify the consistency of the transmission of timing patterns.

[0043] Along the verified process link, starting from the downstream device exhibiting abnormal inter-device coupling characteristics, trace back to the upstream source device. The source device that first appears with the peak slope of the time-series abrupt change in the tracing path and matches the downstream peak shape is identified as the root cause device of the coupling abnormality.

[0044] Furthermore, S5 includes:

[0045] Based on the identified root cause device information of the coupling-related anomaly, retrieve the historical operational anomaly data corresponding to the root cause device when it experienced operational anomalies in previous periods from the storage history of the device parameter coupling-related dataset;

[0046] The currently identified abnormal features of inter-device coupling are compared with the abnormal features in the historical operational abnormal data of the root cause device.

[0047] For the root cause devices that are being compared for consistency, supplementary operating parameters in addition to the already monitored operating parameters are collected through their IoT sensing terminals.

[0048] Based on the collected supplementary operating parameters, verify whether the operating status of the root cause device is abnormal;

[0049] When both the consistency comparison results of historical abnormal operation data and the verification results of supplementary operation parameters support that the root source device is the source of the abnormality, the verified root source device pointing information is confirmed.

[0050] Furthermore, S6 includes:

[0051] It collects and identifies individual device-specific anomaly features, inter-device coupled and correlated anomaly features, and verified root cause device pointing information;

[0052] Associate and bind the abnormal characteristics of inter-device coupling with corresponding relationships with the verified root source device pointing information;

[0053] Based on the independent anomaly characteristics of a single device, the coupled anomaly characteristics between devices that have been associated and bound, and the verified root source device pointing information, a structured state description containing anomaly characteristic categories, anomaly device location, and coupled anomaly root source pointing information is generated.

[0054] Based on the generated structured state description, the final self-sensing results of the coupled operating state of high-risk equipment in thermal power plants are formed.

[0055] On the other hand, the present invention provides an Internet of Things-based self-sensing system for the status of high-risk equipment in thermal power plants, comprising the following modules:

[0056] The parameter acquisition module is used to collect the operating parameters of each device in the coupled operation system of high-risk equipment in thermal power plants through IoT sensing terminals;

[0057] The dataset generation module is used to divide the material flow and energy transfer process link of high-risk equipment in thermal power plants into coupling units, collect the change time series of each operating parameter in the same coupling unit, verify the correlation of parameter response time series and associate mapping, and obtain the equipment parameter coupling correlation dataset.

[0058] The feature differentiation module is used to identify abnormal features in the coupled and correlated dataset of device parameters, and to distinguish between independent abnormal features of a single device and coupled and correlated abnormal features between devices.

[0059] The information determination module is used to extract the temporal abrupt change slope of the abnormal features of coupling association between equipment, combine it with the coupling phase difference analysis of the process link, and determine the root equipment pointing information of the coupling association anomaly by peak matching and tracing back along the upstream of the process link.

[0060] The information verification module is used to retrieve historical abnormal operation data of the corresponding root source device based on the root source device pointing information, compare them, collect supplementary operation parameters for verification, and obtain the verified root source device pointing information.

[0061] The result generation module is used to integrate the independent anomaly features of a single device, the coupled and associated anomaly features, and the verified root cause device pointing information to generate the self-sensing results of the coupled operating status of high-risk equipment in thermal power plants.

[0062] Compared with the prior art, the present invention has the following beneficial effects:

[0063] 1. By constructing a dataset of coupled and correlated equipment parameters based on the material and energy process links, a digital representation and online analysis of the dynamic coupled operation relationship between high-risk equipment in thermal power plants was achieved. This elevates the perspective of state awareness from traditional isolated monitoring of single equipment to a holistic understanding of the coupled operation system of multiple equipment. By verifying the temporal correlation of parameter responses and dividing the coupling units, the chain reaction of downstream equipment caused by changes in the state of upstream equipment can be effectively captured. This clearly separates and presents the abnormal correlation between equipment that was originally hidden in complex interactions, overcoming the defect of existing technologies that misjudge correlation anomalies as multiple independent faults due to ignoring coupling characteristics. This significantly improves the completeness of state awareness and the depth of situational understanding.

[0064] 2. By extracting the slope of time-series abrupt changes, analyzing the coupled phase difference, and performing peak matching and reverse tracing along the process chain, an automatic diagnostic logic chain from abnormal phenomena to the root cause of faults was established. This enables the system not only to alert where the abnormality is, but also to accurately determine its origin, achieving a leap in fault diagnosis from symptom description to root cause location. Combined with a cross-validation mechanism of comparing historical abnormal data and verifying supplementary parameters, the reliability and confidence of the root cause equipment pointing information are greatly improved. The final generated structured state self-perception results integrate multi-dimensional information on independent anomalies, coupled anomalies, and their root cause pointing, providing operators with direct, clear, and actionable decision support. This qualitatively improves the pertinence, preventativeness, and reliability of operation and maintenance management, meeting the core requirements of accurate perception and intelligent diagnosis for the safe and stable operation of high-risk equipment. Attached Figure Description

[0065] Figure 1 This is a flowchart of the Internet of Things-based self-sensing method for the status of high-risk equipment in thermal power plants according to the present invention.

[0066] Figure 2 This is a schematic diagram of the structure of the Internet of Things-based self-sensing system for the status of high-risk equipment in thermal power plants according to the present invention. Detailed Implementation

[0067] 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 of ordinary skill in the art without creative effort are within the scope of protection of the present invention.

[0068] Example 1: Figure 1 This invention presents a self-sensing method for the status of high-risk equipment in thermal power plants based on the Internet of Things, which includes the following steps:

[0069] S1. Collect the operating parameters of each device in the coupled operation system of high-risk equipment in thermal power plants through IoT sensing terminals;

[0070] S2. Based on the material flow and energy transfer process link of high-risk equipment in thermal power plants, couple units are divided, and the change time sequence of each operating parameter in the same couple unit is collected. The correlation of parameter response time sequence is verified and the correlation mapping is performed to obtain the equipment parameter coupling correlation dataset.

[0071] S3. Identify abnormal features in the coupled and correlated dataset of equipment parameters, and distinguish between independent abnormal features of a single device and coupled and correlated abnormal features between devices;

[0072] S4. Extract the temporal abrupt change slope of the abnormal coupling characteristics between equipment, combine it with the process link analysis of the coupling phase difference, and determine the root equipment pointing information of the coupling anomaly by peak matching and tracing back upstream along the process link.

[0073] S5. Based on the root source device pointing information, retrieve the historical abnormal operation data of the corresponding root source device for comparison, collect supplementary operation parameters for verification, and obtain the verified root source device pointing information.

[0074] S6. Integrate the independent anomaly features of a single device, the coupled and associated anomaly features, and the verified root cause device pointing information to generate the self-sensing results of the coupled operation status of high-risk equipment in thermal power plants.

[0075] When implementing an IoT-based self-sensing method for the status of high-risk equipment in thermal power plants, the first step is data acquisition. This step aims to comprehensively acquire the operating parameters of each device in the coupled operation system of high-risk equipment in the thermal power plant through IoT sensing terminals. The specific implementation process is as follows: First, the scope and composition of the coupled operation system of high-risk equipment in the thermal power plant are defined, and all types of high-risk equipment included in the system are acquired. This acquisition process is based on the actual production process and equipment safety management requirements of the thermal power plant. By analyzing the system diagram and operating procedures of the entire plant, equipment that is at a critical node in the material flow and energy transfer chain and whose failure may trigger a chain reaction or significant safety risks is identified. For example, in a typical coal-fired power plant, the identified high-risk equipment types usually include boiler equipment, steam turbine equipment, and coal mill equipment. These devices are closely coupled through the transfer of media such as steam and pulverized coal, forming the core operation system. Acquiring the equipment types is the foundation for all subsequent monitoring and analysis activities, ensuring the completeness of the monitoring scope.

[0076] After identifying the types of high-risk equipment, the categories of operating parameters to be monitored for each type are scientifically determined based on the core operating mechanisms of each identified high-risk equipment type. Operating mechanisms refer to the inherent physical and chemical laws governing the energy conversion, media handling, or power transmission processes of the equipment. The principle for determining the categories of operating parameters is to select key physical quantities that can directly and sensitively reflect the core functional status, health level, and fault symptoms of the equipment. For boiler equipment, its core operating mechanism involves fuel combustion, steam-water heat absorption, and phase change; therefore, the categories of operating parameters to be monitored include drum pressure (reflecting the pressure of the steam-water system), main steam temperature (reflecting the degree of steam superheat), and furnace negative pressure (reflecting the combustion stability of the furnace). For steam turbine equipment, its core operating mechanism involves high-temperature, high-pressure steam expansion driving rotor rotation; therefore, the categories of operating parameters to be monitored include shaft vibration (reflecting the smoothness of rotor operation), bearing temperature (reflecting the lubrication and heat dissipation status of bearings), and cylinder expansion (reflecting the thermal expansion of the cylinder). For coal mill equipment, its core operating mechanism involves the grinding and conveying of raw coal. Therefore, the categories of operating parameters to be monitored include bearing vibration, which reflects the mechanical state of rotating parts; gearbox oil temperature, which reflects the lubrication and load state of the transmission system; and outlet pulverized coal temperature, which reflects the quality of the pulverized coal and the system temperature. Determining these categories of operating parameters provides clear targets for the targeted deployment of sensing terminals.

[0077] For each type of high-risk equipment and its identified categories of operating parameters requiring monitoring, corresponding IoT sensing terminals are deployed. An IoT sensing terminal is a device integrating specific sensors, signal conditioning circuits, analog-to-digital converters, and communication interfaces. During deployment, based on the specific physical characteristics of the operating parameters and the equipment structure, the IoT sensing terminals are installed in locations capable of accurately sensing the target parameters. For example, for the operating parameter category of steam drum pressure, a pressure transmitter is deployed as an IoT sensing terminal, with its pressure tap connected to the pressure measurement point interface of the boiler drum. For the operating parameter category of shaft vibration, a vibration sensor is deployed as an IoT sensing terminal, with its probe installed on a measuring plane perpendicular to the axis near the turbine bearing housing or journal. For the operating parameter category of gearbox oil temperature, a temperature sensor is deployed as an IoT sensing terminal, with its temperature probe inserted into the lubricating oil temperature measuring sleeve of the coal mill gearbox or closely attached to the temperature measuring point on the outer wall of the gearbox. The deployment process ensures that each IoT sensing terminal can stably convert the physical quantity corresponding to the operating parameter category it monitors into a standard electrical signal.

[0078] After deploying all IoT sensing terminals, the deployed terminals synchronously collect operating parameters of each device in the high-risk equipment coupling operation system of the thermal power plant according to a preset collection frequency. The preset collection frequency is set based on the dynamic change characteristics of the monitored operating parameters and the needs of state analysis. The setting method is as follows: for relatively slow-changing parameters such as temperature and pressure, the collection frequency can be set to 1 Hz, that is, 1 data point is collected per second; for rapidly changing vibration parameters, the collection frequency needs to be set to 1000 Hz or higher, that is, 1000 or more data points are collected per second, to meet the needs of spectrum analysis. Synchronous collection refers to synchronizing the time of each distributed IoT sensing terminal through a unified clock source of the plant-level monitoring information system or based on a network time protocol, so that the timestamps attached to the data collected by all terminals have a consistent and accurate time reference. This is a prerequisite for subsequent cross-device time-series correlation analysis. The IoT sensing terminals continuously transmit the collected operating parameter values, along with their precise timestamps, to the central data processing node for aggregation and storage via network transmission methods such as industrial Ethernet, wireless LAN, or fieldbus. This completes the comprehensive and synchronous data collection of the operating status of the coupled operation system of high-risk equipment in thermal power plants, providing the raw data foundation for subsequent coupled analysis and status self-sensing. The entire data collection process is continuous and automatic.

[0079] After collecting operating parameters, a coupled analysis based on the material flow and energy transfer process links is implemented to construct a coupled correlation dataset of equipment parameters that reflects the mutual influence between equipment. The specific implementation process is as follows: First, the material flow paths and energy transfer paths of high-risk equipment in thermal power plants are analyzed. The material flow path refers to the physical route of working fluids such as coal, water, steam, and flue gas transported between equipment. For example, raw coal enters the coal mill from the coal bunker via the coal feeder, and the ground coal powder is carried by the primary air into the boiler furnace for combustion. Feedwater circulates through the economizer, steam drum, downcomer, and water-cooled wall to generate steam, which then enters the steam turbine to perform work. The energy transfer path refers to the path of conversion and transfer of chemical energy, thermal energy, mechanical energy, and electrical energy between equipment. For example, the chemical energy of coal is converted into the thermal energy of steam in the boiler, the thermal energy of steam is converted into the mechanical energy of the rotor in the steam turbine, and the mechanical energy is converted into electrical energy in the generator. Based on the interpretation of the plant's overall process system diagram and the experience of the operating personnel, continuous and direct process connections in these paths are identified.

[0080] Based on the continuity of the identified process flow, coupled units comprising upstream and downstream equipment are defined. The principle of this division is to group adjacent high-risk equipment with direct material input / output or energy transfer relationships into a single coupled unit. Equipment providing material or energy is called upstream equipment, and equipment receiving material or energy is called downstream equipment. For example, boiler equipment and steam turbine equipment are directly connected via main steam pipelines and reheat steam pipelines. The high-temperature, high-pressure steam generated by the boiler drives the steam turbine; therefore, boiler equipment and steam turbine equipment can be classified as one coupled unit, with the boiler equipment as upstream and the steam turbine equipment as downstream. As another example, coal mill equipment and boiler equipment are connected via pulverized coal pipelines. The pulverized coal ground by the coal mill is sent to the boiler for combustion; therefore, coal mill equipment and boiler equipment can be classified as another coupled unit, with the coal mill equipment as upstream and the boiler equipment as downstream. A single piece of equipment may belong to multiple coupled units simultaneously, reflecting the networked coupling characteristics of the operating system. Dividing the system into coupled units is the process of decomposing the entire complex operating system into several more easily analyzable basic relational groups.

[0081] For each defined coupling unit, operating parameters belonging to each device within that unit are extracted from the massive real-time data obtained by collecting operating parameters of each device in the high-risk equipment coupling operation system of a thermal power plant through IoT sensing terminals. Specifically, based on the equipment composition of the coupling unit, for example, a boiler-turbine coupling unit includes boiler equipment and turbine equipment, operating parameters such as drum pressure, main steam temperature, and furnace negative pressure collected by IoT sensing terminals deployed on the boiler equipment, and operating parameters such as shaft vibration, bearing temperature, and cylinder expansion collected by IoT sensing terminals deployed on the turbine equipment, are retrieved from the storage of the central data processing node within the same time period. The extraction operation ensures that the data time range covers a complete operating condition stage or a preset analysis time window.

[0082] Because the acquisition cycles and clock fine-tuning of different IoT sensing terminals may have slight differences, it is necessary to align the timestamps of the operating parameters extracted from each device within the same coupling unit to form a change time sequence with a unified time base. Alignment methods can employ time series interpolation or nearest neighbor matching. For example, using a unified, high-precision timeline as a reference, such as one reference point per second, for each device's operating parameter sequence, its values ​​near each reference time are calculated using a linear interpolation algorithm to obtain a value corresponding to that reference time, thus ensuring that the operating parameter sequences of all devices within the coupling unit have the same set of timestamps. Another method is to use the acquisition time of a key parameter as a reference and match the values ​​of other parameters to the closest time to that reference time. After alignment, each operating parameter within the coupling unit forms a change time sequence with strictly synchronized timestamps and equally spaced or comparable data points.

[0083] Based on a unified timeline of changes, this study analyzes the chronological order and delay relationship between changes in operating parameters of upstream and downstream equipment to verify the correlation of parameter response timing. The analysis involves selecting the change sequences of one or more key operating parameters from both upstream and downstream equipment. Through observation or calculation, it is determined whether the time point of significant change in upstream operating parameters always precedes the time point of corresponding change in downstream operating parameters. For example, in a boiler-turbine coupling unit, the study analyzes the change sequence of boiler main steam temperature and turbine bearing temperature. When the main steam temperature rises due to combustion adjustments, it observes whether the turbine bearing temperature also shows an upward trend after a reasonable delay. The delay relationship refers to the time lag between downstream parameter changes and upstream parameter changes. This lag should conform to the physical laws determined by the working fluid transfer rate or heat transfer rate. Verification can be performed by calculating the average time difference and consistency ratio of upstream changes leading downstream changes across multiple events. If a statistically stable chronological order and a reasonable delay range exist, a correlation of parameter response timing is determined between the upstream and downstream parameters. This is key evidence to confirm the existence of a causal or strong coupling relationship between the devices.

[0084] The upstream and downstream operating parameters, whose parameter response time-series correlations have been verified, are mapped together to generate and store a device parameter coupling correlation dataset. This correlation mapping involves creating a data structure or database table where each record contains not only the value of a certain operating parameter of the upstream device at a specific time, but also the value of the downstream operating parameter with the verified parameter response time-series correlation at the corresponding delay time. It also records the attributes of this correlation, such as the coupling unit, upstream and downstream device identifiers, specific operating parameter category, and the statistically obtained typical delay time. For example, for the verified correlation between boiler main steam temperature and turbine bearing temperature, a record in the device parameter coupling correlation dataset might contain a timestamp T, a boiler main steam temperature value A, and a turbine bearing temperature value B at timestamp T+Δt, where Δt is the verified typical delay time. This dataset is continuously updated, incorporating new synchronously collected data and verified correlations. The generated and stored equipment parameter coupling and correlation dataset is a structured data set that depicts the dynamic coupling relationship between high-risk equipment in thermal power plants, providing core analytical basis for subsequent anomaly feature identification and root cause tracing.

[0085] After obtaining the coupled dataset of equipment parameters, anomaly feature identification and classification steps are implemented to distinguish abnormal features characterizing equipment failure or deterioration from normal operational fluctuations, and to differentiate between independent abnormal features of a single device and coupled abnormal features between devices. The specific implementation process is as follows: First, based on the operating parameters of each device in the coupled dataset of equipment parameters, it is determined whether each operating parameter deviates from its corresponding normal operating threshold range. The normal operating threshold range is a numerical interval set for each specific operating parameter category, used to define the normal fluctuation range of the parameter under healthy operating conditions. This range is obtained by selecting a sufficiently long period of data for the parameter in the equipment's historical operating database, during which the equipment has been confirmed to be in normal operating condition, and calculating its statistical characteristics. For example, for the boiler drum pressure parameter, all drum pressure data from the boiler's stable load operation within the last three months are selected, and their average value and standard deviation are calculated. The lower boundary of the normal operating threshold range is set as the average value minus three times the standard deviation, and the upper boundary is set as the average value plus three times the standard deviation. Another approach is to combine equipment design specifications, operating procedures, and expert experience to directly define a fixed safe operating range. For example, the normal operating threshold range for the steam drum pressure can be set to ±5% of the design working pressure. Each category of operating parameters has its own independently defined normal operating threshold range.

[0086] The real-time values ​​of operating parameters are continuously compared with the upper and lower boundary values ​​of their corresponding normal operating threshold ranges. This comparison process is performed in real-time or near real-time. For each newly arrived operating parameter data point in the device parameter coupled association dataset, its value is compared with the upper and lower boundary values ​​of the pre-stored normal operating threshold range for that parameter. The judgment condition is whether the value is greater than the upper boundary value or less than the lower boundary value.

[0087] When the real-time value of an operating parameter continuously exceeds its corresponding normal operating threshold range boundary more than a preset number of times, the operating parameter is determined to have deviated from the normal operating threshold range. The preset number of times is an integer used to filter out transient interference or measurement noise, and its setting is based on the understanding of the normal fluctuation characteristics of the equipment parameters and the accuracy of the sensor. For example, the preset number of times can be set to 5. This means that in a continuous data acquisition cycle, if the value of an operating parameter is higher than its upper boundary value for 5 consecutive times, or lower than its lower boundary value for 5 consecutive times, then the operating parameter is finally determined to have a real deviation, rather than a momentary disturbance. This continuous judgment mechanism improves the reliability of anomaly detection and reduces false alarms.

[0088] In the time series of the coupled dataset of equipment parameters, anomaly markers are added to the time points and parameter values ​​identified as deviations, forming anomalous data points. Anomaly markers can be special label fields or status bits; for example, adding a flag bit to the data record to mark normal data points as 0 and anomalous data points as 1. Simultaneously, the specific operating parameter category, equipment identifier, and direction of deviation (whether it is excessively high or excessively low) corresponding to the anomalous data point are recorded. These marked anomalous data points constitute the basic raw material for subsequent feature analysis.

[0089] For each coupled unit identified in the previous steps, analyze all marked abnormal data points in its corresponding device parameter coupling association dataset. The analysis includes: identifying which devices within the coupled unit exhibited abnormal data points, how many abnormal data points occurred in each device, which operational parameter categories these abnormal data points correspond to, and the temporal distribution of these abnormal data points.

[0090] If analysis reveals that anomalous data points appear only on a single device within the analyzed coupled unit, and the operating parameters of other devices within that coupled unit that have a verified time-series correlation with the anomalous data point are normal, then the feature corresponding to that anomalous data point is identified as a single-device independent anomalous feature. For example, in a coupled unit containing a coal mill and a boiler, only the bearing vibration parameters of the coal mill show continuous anomalous data points, while the operating parameters of the boiler, which have a verified time-series correlation with the coal mill, such as the furnace negative pressure, are within their normal operating threshold range and no anomalous data points appear within the corresponding time period. In this case, the abnormal bearing vibration of the coal mill is identified as a single-device independent anomalous feature. This usually suggests that the anomaly may originate from a mechanical failure of the coal mill itself, such as bearing wear, without significantly affecting the downstream boiler equipment.

[0091] If analysis reveals that anomalous data points appear on multiple devices within the analyzed coupled unit, and these anomalous data points exhibit a verified temporal correlation of parameter responses, then the common feature corresponding to these anomalous data points is identified as an inter-device coupling correlation anomaly. For example, in a coupled unit containing a boiler and a turbine, the boiler's main steam temperature parameter first shows an anomalous increase, followed by an anomalous increase in the turbine's bearing temperature parameter after a verified delay. The anomalous data points from these two devices exhibit a clear chronological order, consistent with the previously verified temporal correlation of boiler main steam temperature change preceding turbine bearing temperature change. In this case, the associated phenomena of the boiler main steam temperature anomaly and the turbine bearing temperature anomaly are jointly identified as an inter-device coupling correlation anomaly. This indicates that the anomaly may originate from upstream equipment and propagate through the process chain to downstream equipment, or that multiple devices may be simultaneously affected by a common root cause. After identification and differentiation are completed, the independent anomaly features of a single device and the coupled anomaly features between devices will be output as two distinct types of status indication information for use in subsequent root cause tracing and result integration steps.

[0092] After identifying and classifying abnormal features, a root cause tracing step is implemented for the identified inter-device coupling and correlation anomaly features. This aims to locate the initial source of the chain of anomalies by analyzing the dynamic characteristics of anomaly propagation, i.e., to determine the root device information of the coupling and correlation anomaly. The specific implementation process is as follows: First, from the time series of the abnormal data points corresponding to the inter-device coupling and correlation anomaly features, the instantaneous rate of change of each abnormal data point's change sequence is calculated as the temporal abrupt change slope. The inter-device coupling and correlation anomaly features correspond to a set of time-related abnormal data points, which originate from the operating parameters of multiple devices within the coupled unit. For each abnormal data point, the instantaneous rate of change of its operating parameter change sequence at that point needs to be calculated. The calculation method uses numerical differentiation. Specifically, for a time series of changes in an operating parameter, the time corresponding to the abnormal data point to be calculated is taken, and then data points from the previous and next fixed time intervals are taken. For example, data points from 0.5 seconds before and 0.5 seconds after that time are taken. The difference in parameter values ​​between these two data points is divided by the time difference to obtain an approximate instantaneous rate of change at the time of the abnormal data point. This value is defined as the temporal abrupt change slope at that point. The time interval is selected based on the acquisition frequency and physical inertia of the operating parameter being analyzed. For rapidly changing vibration parameters, the interval may be 0.01 seconds; for slowly changing temperature parameters, the interval may be 2 seconds. The calculated temporal abrupt change slope is a signed numerical value. Its absolute value represents the severity of the abnormal change, and the sign represents whether the change is increasing or decreasing.

[0093] Based on the material flow and energy transfer process link direction used when dividing the coupling units, the upstream and downstream order of each device exhibiting abnormal inter-device coupling characteristics in the process link is determined. This process link direction is already defined when dividing the coupling units. For example, in a coupling unit consisting of a boiler and a steam turbine, the direction of material and energy transfer is from the boiler to the steam turbine; therefore, the boiler is the upstream device, and the steam turbine is the downstream device. When abnormal inter-device coupling characteristics involve multiple devices within the unit, the order of these devices from the upstream to the downstream can be listed based on this pre-determined transfer direction. This order forms the logical path basis for subsequent time difference calculations and reverse tracing.

[0094] The time difference between the anomaly initiation time of upstream and downstream equipment in the calculation process link is used as the coupling phase difference. The anomaly initiation time refers to the precise timestamp corresponding to the first marked abnormal data point of a device in the current coupling correlation anomaly feature. For upstream equipment, its anomaly initiation time is determined and denoted as time Tup; for the immediately adjacent downstream equipment, its anomaly initiation time is determined and denoted as time Tdown. The coupling phase difference ΔT is obtained by subtracting Tup from Tdown, and its value should be positive, indicating that there is a time lag between the anomaly response of the downstream equipment and the anomaly initiation of the upstream equipment. This time difference needs to conform to the physical laws of the time required for the working fluid to be transported in the pipeline between equipment or the time required for heat transfer. For example, in the steam transmission from the boiler to the turbine, this phase difference is usually on the order of several seconds to tens of seconds, depending on the pipeline length and steam flow rate. Calculating the coupling phase difference is a key step in quantifying the anomaly propagation delay.

[0095] Matching the peak values ​​of the time-series abrupt change slopes of upstream and downstream equipment in the process chain verifies the consistency of the transmission of their time-series patterns. The peak value of the time-series abrupt change slope refers to the maximum absolute value of the calculated time-series abrupt change slope reached during the initial stage of an anomaly. Matching verification includes two aspects: First, peak polarity matching, checking whether the signs of the peak values ​​of the time-series abrupt change slopes of upstream and downstream equipment are consistent. For example, if the peak value of the time-series abrupt change slope for the upstream boiler main steam temperature is positive, indicating a sudden temperature rise, then the peak value of the time-series abrupt change slope for the downstream turbine bearing temperature should also be positive, indicating a subsequent sudden temperature rise. This shows that the physical effects of the anomaly propagation are consistent. Second, peak intensity correlation assessment. Under the premise of consistent polarity, there should be a reasonable proportional relationship between the downstream and upstream peak intensities. This proportional relationship can be obtained through the statistical relationship of the rates of change between the two in historical normal coupling data. For example, the downstream peak intensity is usually between 0.3 and 0.8 times the upstream peak intensity, with the specific ratio determined by factors such as equipment thermal inertia. If the upstream and downstream peak values ​​meet expectations in terms of polarity and intensity, the verification is successful, and the timing pattern is considered to have consistent transmission, supporting the hypothesis that the anomaly propagates along the process link.

[0096] Following the verified process flow, starting from the downstream device exhibiting abnormal inter-device coupling characteristics, a reverse tracing process is initiated towards the upstream source device. This reverse tracing is a logical iterative process. The starting point is the downstream device, examining the directly upstream devices supplying it with matter or energy. Operating parameter data for this upstream device is obtained, and it is checked whether it was also identified as having abnormal data points in this event. Its time-series abrupt change slope peak is calculated. If the upstream device exhibits an anomaly, and its anomaly start time is earlier than that of the downstream device, and its time-series abrupt change slope peak matches the downstream device's peak value using the aforementioned method, then this upstream device is marked as a possible source in the current tracing stage. Then, using this device as the new current point, the same checks are performed on even more upstream devices, examining the earlier the anomaly start time and the matching of peak values.

[0097] The source device that first exhibits the peak slope of the temporal abrupt change in the tracing path and matches the peak shape of the downstream device is identified as the root cause of the coupled anomaly. During reverse tracing, when an upstream device is checked, if it is found to have no abnormal data points, or its anomaly start time is not earlier than its downstream device, or its peak shape cannot match the downstream device, the tracing terminates at the downstream device of that upstream device. At this point, the last device that simultaneously meets the conditions of earliest anomaly start and peak shape matching is determined as the root cause of the entire chain of anomalies. For example, in a transmission chain consisting of a coal mill, boiler, and steam turbine, if tracing starts from the downstream steam turbine and finds a matching anomaly in the boiler earlier than the steam turbine, and tracing upwards from the boiler to the coal mill reveals a matching anomaly in the coal mill earlier than the boiler, and there are no other devices above the coal mill or the coal mill's anomaly cannot be traced back to a previous cause, then the coal mill is identified as the root cause of this coupled anomaly. The root cause device information clarifies the priority direction and core objects of fault diagnosis, providing direct target input for subsequent verification and operation and maintenance decisions.

[0098] After identifying the root cause equipment of the coupled anomaly through time-series analysis, a cross-validation step is implemented. This aims to further enhance the confidence of the root cause location by comparing historical data and supplementing monitoring, thereby obtaining the verified root cause equipment information. The specific implementation process is as follows: First, based on the identified root cause equipment information of the coupled anomaly, which indicates the high-risk equipment traced as a suspected source, historical operational anomaly data corresponding to the root cause equipment's operational anomalies in past periods is retrieved from the storage history of the equipment parameter coupled association dataset. The storage history of the equipment parameter coupled association dataset is a time-series database that not only stores current real-time data but also archives past historical data, including records of each event marked as an anomaly by the equipment in its history. During retrieval, the identifier of the current root cause equipment is used as the query key to retrieve all historical records associated with the equipment and marked as anomaly in the database. For example, if the root cause equipment points to a coal mill, the database is queried for all historical events and their complete time-series data of abnormal data points in parameters such as bearing vibration and gearbox oil temperature of the coal mill equipment in the past period. These historical operational anomaly data contain detailed information about the anomaly characteristics, such as the start and end times of the anomaly, the categories of the anomaly parameters, and the numerical sequences of the anomaly data points.

[0099] The process involves comparing the currently identified inter-device coupling and correlation anomaly features with the anomaly features in the historical operational anomaly data of the root cause device. The currently identified inter-device coupling and correlation anomaly features refer to the set of anomaly features identified in this event involving multiple devices, but the comparison here focuses on the portion of the anomaly performance belonging to the current root cause device. The consistency comparison is performed in three dimensions. The first dimension is the consistency of anomaly parameter categories, checking whether the operational parameter categories of the current root cause device exhibiting anomalies have appeared in historical anomaly records. For example, if the coal mill exhibits abnormal bearing vibration in this instance, the historical records are checked to see if the coal mill has ever experienced abnormal bearing vibration. The second dimension is the consistency of temporal morphology, comparing the morphological characteristics of the current root cause device's operational sequence curve for this anomaly with the operational sequence curves of the same parameter category in historical anomaly events. Morphological comparison is achieved by calculating the quantified difference value between the two curves. Specifically, under the premise of time axis alignment, the absolute value of the difference between the values ​​of the two curves at multiple corresponding time points is calculated, and then these absolute values ​​are summed or averaged to obtain the final quantified difference value. The smaller the quantified difference value, the more similar the shapes of the two curves. The calculated quantified difference value is compared with a preset shape difference threshold. The method for setting the shape difference threshold is to collect multiple sets of time-series curves of the same parameter category under the historical normal operating conditions of the device, calculate the quantified difference values ​​between them, and take the higher percentile of these values ​​as the shape difference threshold, such as the 95th percentile. If the currently calculated quantified difference value is less than the preset shape difference threshold, it is determined that the shapes are consistent. The third dimension is the consistency of the anomaly intensity range, which compares the degree to which the value of the current anomaly data point deviates from the normal operating threshold range, and whether the degree of deviation of the same parameter category in historical anomalies is within a similar numerical range. Only when the current anomaly feature shows significant consistency with the historical anomaly feature in the main dimensions is the comparison result considered valid.

[0100] For the root cause equipment undergoing consistency comparison, supplementary operating parameters, in addition to the monitored operating parameters, are collected through its deployed IoT sensing terminals. Monitored operating parameters refer to those categories of operating parameters that are routinely and continuously monitored, as determined in step S1 based on the equipment's core operating mechanism. Supplementary operating parameters are other parameters temporarily collected to further verify specific equipment states. The determination of supplementary operating parameters is based on the type of the current suspected anomaly and the equipment mechanism. For example, if the root cause equipment is a coal mill and the anomaly is concentrated in bearing vibration, the supplementary operating parameters can be determined as the online monitoring values ​​of the coal mill motor current and lubricating oil particle size. Collection is achieved by sending specific commands to the IoT sensing terminal network controlling the coal mill to activate or adjust relevant sensors. Supplementary operating parameters provide evidence reflecting other aspects of the equipment's state beyond the monitored parameters.

[0101] Based on the collected supplementary operating parameters, the verification process checks whether the root cause equipment's operating status is abnormal. The verification process first sets temporary evaluation thresholds for these supplementary operating parameters. For example, for the temporarily collected coal mill motor current, the evaluation threshold is dynamically set based on the motor's rated current and the current load rate. The setting method is to take the typical operating current value of the motor under the current load as a benchmark, and set 110% of this benchmark value as the temporary alarm threshold. For lubricating oil particle size, the evaluation threshold is set according to equipment maintenance standards; for example, setting a temporary abnormality threshold as more than 100,000 particles larger than a certain size per milliliter of lubricating oil. Then, the real-time collected supplementary operating parameter values ​​are compared with these corresponding temporary evaluation thresholds. If the supplementary operating parameters also show clear signs of abnormality, such as the motor current value exceeding its temporary alarm threshold and the lubricating oil particle size value exceeding its temporary abnormality threshold, the verification result supports that the root cause equipment's operating status is abnormal. If all supplementary operating parameters are within the normal range defined by their corresponding temporary evaluation thresholds, the verification result does not support this.

[0102] When both the consistency comparison results of historical operational anomaly data and the verification results of supplementary operational parameters support that the root cause device is the source of the anomaly, the verified root cause device identification information is confirmed. "Both support" means that, under the preset judgment logic, the independent conclusions of both verification paths are affirmative. For example, the consistency comparison shows that the current vibration anomaly pattern is highly similar to the vibration pattern of a bearing in the early stage of wear in the past, and the calculated difference quantification value is lower than the preset pattern difference threshold. Simultaneously, the supplementary collected lubricating oil particle size value repeatedly exceeds the temporary anomaly threshold of 100,000 particles per milliliter. If both conditions are met, the coal mill is determined to be the verified source of the anomaly. If only one path supports this, such as historical comparison being consistent but supplementary parameters being normal, or supplementary parameters being abnormal but historical patterns not matching, the verification result may be uncertain. The finally confirmed verified root cause device identification information is a location conclusion that has undergone cross-validation of multi-source evidence, and it will be output as a core component of the final state perception result.

[0103] After completing anomaly identification, root cause tracing, and cross-validation, the final result integration and output step is implemented to generate a comprehensive, clear, and actionable self-awareness result of the coupled operating status of high-risk equipment in thermal power plants. The specific implementation process is as follows: First, all independent anomaly features of individual equipment, all coupled anomaly features between equipment, and all verified root cause equipment pointing information obtained after the verification step are collected from the previous steps. This collection operation is performed at the central data processing node, which establishes a temporary result workspace. All records of independent anomaly features of individual equipment, coupled anomaly features between equipment, and verified root cause equipment pointing information from the output list of the anomaly identification step are imported into this workspace. This collection ensures that all scattered, different types of status judgment fragments are centralized, preparing for subsequent unified integration.

[0104] The system associates and binds inter-device coupling-related anomaly features with corresponding relationships to verified root source device information. A corresponding relationship means that an inter-device coupling-related anomaly feature and a verified root source device information logically point to the same set of chained anomaly events; that is, the root source device is verified as having triggered this set of inter-device coupling-related anomalies. The specific implementation method for this association binding is to assign a unique event identifier to each inter-device coupling-related anomaly feature record and each verified root source device information record. When the system logic determines that the tracing result of an inter-device coupling-related anomaly feature and a verified root source device information point to the same device and their time windows overlap, the event identifiers of these two records are linked, or a new integrated record is created. This record contains all attribute fields of both the inter-device coupling-related anomaly feature and the verified root source device information. For example, for an inter-device coupling-related anomaly feature concerning boiler main steam temperature anomaly and turbine bearing temperature anomaly, if its verified root source device information is confirmed to be the boiler, then these two information entities are bound together, forming an anomaly-root source pairing.

[0105] Based on the collected single-device independent anomaly features, the associated and bound inter-device coupled anomaly features, and the verified root source device information, a structured state description is generated, including anomaly feature categories, anomaly device locations, and the root source of coupled anomalies. The generation process follows a predefined structured template. This template defines the fields of the description document or data object. For each single-device independent anomaly feature, its key attributes are extracted and filled into the corresponding fields of the template. These attributes include the anomaly feature category as single-device independent anomaly, the anomaly device location information (i.e., the unique identifier of the device where the anomaly occurred), the anomaly's operating parameter category, the anomaly start time, and a brief description of the anomaly. The coupled anomaly root source field is marked as none, as this is an independent anomaly. For each associated and bound inter-device coupled anomaly feature paired with the verified root source device information, its key attributes are extracted and filled into the template. These attributes include the anomaly feature category as inter-device coupled anomaly, the anomaly device location information (a list of unique identifiers of all devices involved in the anomaly), the associated anomaly operating parameter category list, a timeline summary of the anomaly propagation, and the coupled anomaly root source field, which explicitly contains the unique identifier of the verified root source device. The structured state description is ultimately represented as a list or a set of data objects, where each entry clearly contains all the above field information, enabling any reader to quickly understand what anomalies exist, which devices they occur on, and where the root cause of a chain of anomalies lies.

[0106] Based on the generated structured status descriptions, the final self-perception results of the coupled operating status of high-risk equipment in thermal power plants are generated. Generating the final result means converting the structured status descriptions into a standard output format that can be directly used by upper-level application systems or operators. This format can be an Extensible Markup Language (EXPLAIN) document following a specific pattern, one or more records in a Structured Query Language (SCL) database table, or a predefined graphical user interface (GUI) data interface protocol. For example, all entries in the structured status description list are sorted by timestamp and encapsulated into a data package following JavaScript object notation. This data package serves as the final output of this analysis cycle. Simultaneously, corresponding result publishing mechanisms can be triggered, such as pushing the data package to the data interface of the thermal power plant's monitoring screen for display in a specific visualization format; storing it in a historical status record library for trend analysis; or integrating it with a work order management system to automatically generate preliminary maintenance suggestion work orders. The final self-perception results of the coupled operating status of high-risk equipment in thermal power plants are a comprehensive status report that integrates the results of the entire chain from data acquisition to intelligent analysis, is machine-readable and human-understandable, and marks the completion of a complete self-perception closed loop.

[0107] Example 2: Figure 2A schematic diagram of the IoT-based self-sensing system for the status of high-risk equipment in thermal power plants is provided. The IoT-based self-sensing system for the status of high-risk equipment in thermal power plants includes the following modules:

[0108] The parameter acquisition module is used to collect the operating parameters of each device in the coupled operation system of high-risk equipment in thermal power plants through IoT sensing terminals;

[0109] The dataset generation module is used to divide the material flow and energy transfer process link of high-risk equipment in thermal power plants into coupling units, collect the change time series of each operating parameter in the same coupling unit, verify the correlation of parameter response time series and associate mapping, and obtain the equipment parameter coupling correlation dataset.

[0110] The feature differentiation module is used to identify abnormal features in the coupled and correlated dataset of device parameters, and to distinguish between independent abnormal features of a single device and coupled and correlated abnormal features between devices.

[0111] The information determination module is used to extract the temporal abrupt change slope of the abnormal features of coupling association between equipment, combine it with the coupling phase difference analysis of the process link, and determine the root equipment pointing information of the coupling association anomaly by peak matching and tracing back along the upstream of the process link.

[0112] The information verification module is used to retrieve historical abnormal operation data of the corresponding root source device based on the root source device pointing information, compare them, collect supplementary operation parameters for verification, and obtain the verified root source device pointing information.

[0113] The result generation module is used to integrate the independent anomaly features of a single device, the coupled and associated anomaly features, and the verified root cause device pointing information to generate the self-sensing results of the coupled operating status of high-risk equipment in thermal power plants.

[0114] All calculations involved in the embodiments are dimensionless numerical calculations, and the preset parameters and thresholds in the calculations are set by those skilled in the art according to the actual situation.

[0115] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0116] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and inventive constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0117] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0118] In the several embodiments provided in this application, it should be understood that the disclosed systems, apparatuses, and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0119] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0120] In conclusion, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for self-sensing the status of high-risk equipment in thermal power plants based on the Internet of Things, characterized in that, Includes the following steps: S1. Collect the operating parameters of each device in the coupled operation system of high-risk equipment in thermal power plants through IoT sensing terminals; S2. Based on the material flow and energy transfer process link of high-risk equipment in thermal power plants, couple units are divided, and the change time sequence of each operating parameter in the same couple unit is collected. The correlation of parameter response time sequence is verified and the correlation mapping is performed to obtain the equipment parameter coupling correlation dataset. S3. Identify abnormal features in the coupled and correlated dataset of equipment parameters, and distinguish between independent abnormal features of a single device and coupled and correlated abnormal features between devices; S4. Extract the temporal abrupt change slope of the abnormal coupling characteristics between equipment, combine it with the coupling phase difference analysis of the process link, and determine the root equipment pointing to the coupling anomaly by peak matching and tracing back upstream along the process link, including: From the time series of abnormal data points corresponding to the abnormal characteristics of coupling between devices, the instantaneous change rate of the change time sequence of each abnormal data point is calculated as the time sequence abrupt change slope. Based on the material flow and energy transfer process link direction on which the coupling unit is divided, the upstream and downstream order of each device that has abnormal characteristics of inter-device coupling correlation in the process link is determined; The time difference between the abnormal start time of the upstream equipment and the abnormal start time of the downstream equipment in the calculation process link is used as the coupling phase difference. Match the peak value of the timing mutation slope of the upstream equipment in the process link with the peak value of the timing mutation slope of the downstream equipment to verify the consistency of the transmission of timing patterns. Along the verified process link, starting from the downstream device exhibiting abnormal inter-device coupling characteristics, trace back to the upstream source device. The source device that first appears with the peak slope of the time-series abrupt change in the tracing path and matches the downstream peak shape is identified as the root cause device pointing to the abnormal coupling. S5. Based on the root source device pointing information, retrieve the historical abnormal operation data of the corresponding root source device for comparison, collect supplementary operation parameters for verification, and obtain the verified root source device pointing information. S6. Integrate the independent anomaly features of a single device, the coupled and associated anomaly features, and the verified root cause device pointing information to generate the self-sensing results of the coupled operation status of high-risk equipment in thermal power plants.

2. The method for self-sensing the status of high-risk equipment in thermal power plants based on the Internet of Things as described in claim 1, characterized in that, S1 includes: Identify the types of high-risk equipment included in the coupled operation system of high-risk equipment in thermal power plants; Based on the core operating mechanism of each type of high-risk equipment, determine the categories of operating parameters that need to be monitored for each type of high-risk equipment; For each type of high-risk equipment, deploy IoT sensing terminals corresponding to the categories of operating parameters that need to be monitored; The deployed IoT sensing terminals synchronously collect the operating parameters of each device in the coupled operation system of high-risk equipment in thermal power plants according to the preset collection frequency.

3. The method for self-sensing the status of high-risk equipment in thermal power plants based on the Internet of Things according to claim 2, characterized in that, Based on the core operating mechanism of each type of high-risk equipment, the categories of operating parameters that need to be monitored for each type of high-risk equipment are determined, including: For boiler equipment, the categories of operating parameters that need to be monitored include drum pressure, main steam temperature, and furnace negative pressure. For steam turbine equipment, the categories of operating parameters that need to be monitored include shaft vibration, bearing temperature, and cylinder expansion. For coal mill equipment, the categories of operating parameters that need to be monitored include bearing vibration, gearbox oil temperature, and outlet pulverized coal temperature.

4. The method for self-sensing the status of high-risk equipment in thermal power plants based on the Internet of Things according to claim 1, characterized in that, S2 include: Analyze the material flow path and energy transfer path of high-risk equipment in thermal power plants, and divide the process into coupled units including upstream and downstream equipment based on the continuity of the process flow. For each coupling unit, the operating parameters of each device in the high-risk equipment coupling operation system of the thermal power plant are extracted from the operating parameters of each device collected through the Internet of Things sensing terminal. Align the timestamps of the operating parameters of each device within the same coupling unit to form a change time series with a unified time base; Based on the change time series, we analyze the chronological order and delay relationship between the changes in the operating parameters of upstream equipment and the changes in the operating parameters of downstream equipment to verify the correlation of parameter response time series. The upstream and downstream device operating parameters, whose parameter response timing correlation has been verified, are associated and mapped to generate and store a device parameter coupled association dataset.

5. The method for self-sensing the status of high-risk equipment in thermal power plants based on the Internet of Things according to claim 1, characterized in that, S3 include: Based on the coupling and association of device parameters with the operating parameters of each device in the dataset, it is determined whether the operating parameters deviate from their corresponding normal operating threshold range. Mark operating parameters that deviate from the normal operating threshold range as abnormal data points; For each coupled unit, analyze the marked abnormal data points in its device parameter coupling correlation dataset; If an abnormal data point appears only on a single device in a coupling unit, and the operating parameters of other devices in the coupling unit that have parameter response timing correlation with it are normal, then the feature corresponding to the abnormal data point is identified as a single device independent abnormal feature. If abnormal data points appear on multiple devices in the coupling unit, and there is a verified correlation between the abnormal data points of these devices in terms of parameter response timing, then the common features corresponding to these abnormal data points are identified as abnormal features of inter-device coupling correlation.

6. The method for self-sensing the status of high-risk equipment in thermal power plants based on the Internet of Things as described in claim 5, characterized in that, Operating parameters that deviate from the normal operating threshold range are marked as abnormal data points, including: The real-time values ​​of the operating parameters are continuously compared with the upper and lower boundary values ​​of their corresponding normal operating threshold range. When the real-time value of an operating parameter exceeds the boundary of its corresponding normal operating threshold range more than a preset number of times, the operating parameter is determined to deviate from the normal operating threshold range. In the time series of the coupled dataset of equipment parameters, anomaly markers are added to the time points and their parameter values ​​that are determined to be deviated, forming abnormal data points.

7. The method for self-sensing the status of high-risk equipment in thermal power plants based on the Internet of Things according to claim 1, characterized in that, S5 include: Based on the identified root cause device information of the coupling-related anomaly, retrieve the historical operational anomaly data corresponding to the root cause device when it experienced operational anomalies in previous periods from the storage history of the device parameter coupling-related dataset; The currently identified abnormal features of inter-device coupling are compared with the abnormal features in the historical operational abnormal data of the root cause device. For the root cause devices that are being compared for consistency, supplementary operating parameters in addition to the already monitored operating parameters are collected through their IoT sensing terminals. Based on the collected supplementary operating parameters, verify whether the operating status of the root cause device is abnormal; When both the consistency comparison results of historical abnormal operation data and the verification results of supplementary operation parameters support that the root source device is the source of the abnormality, the verified root source device pointing information is confirmed.

8. The method for self-sensing the status of high-risk equipment in thermal power plants based on the Internet of Things according to claim 1, characterized in that, S6 include: It collects and identifies individual device-specific anomaly features, inter-device coupled and correlated anomaly features, and verified root cause device pointing information; Associate and bind the abnormal characteristics of inter-device coupling with corresponding relationships with the verified root source device pointing information; Based on the independent anomaly characteristics of a single device, the coupled anomaly characteristics between devices after association and binding, and the verified root cause device pointing information, a structured state description containing anomaly characteristic categories, anomaly device location, and coupled anomaly root cause pointing is generated; Based on the generated structured state description, the final self-sensing results of the coupled operating state of high-risk equipment in thermal power plants are formed.

9. A self-sensing system for the status of high-risk equipment in thermal power plants based on the Internet of Things (IoT), used to implement the self-sensing method for the status of high-risk equipment in thermal power plants based on the IoT as described in any one of claims 1-8, characterized in that, Includes the following modules: The parameter acquisition module is used to collect the operating parameters of each device in the coupled operation system of high-risk equipment in thermal power plants through IoT sensing terminals; The dataset generation module is used to divide the material flow and energy transfer process link of high-risk equipment in thermal power plants into coupling units, collect the change time series of each operating parameter in the same coupling unit, verify the correlation of parameter response time series and associate mapping, and obtain the equipment parameter coupling association dataset. The feature differentiation module is used to identify abnormal features in the coupled and correlated dataset of device parameters, and to distinguish between independent abnormal features of a single device and coupled and correlated abnormal features between devices. The information determination module is used to extract the temporal abrupt change slope of the abnormal features of coupling association between equipment, combine it with the coupling phase difference analysis of the process link, and determine the root equipment pointing information of the coupling association anomaly by peak matching and tracing back upstream along the process link. The information verification module is used to retrieve historical abnormal operation data of the corresponding root source device based on the root source device pointing information, compare them, collect supplementary operation parameters for verification, and obtain the verified root source device pointing information. The result generation module is used to integrate the independent anomaly features of a single device, the coupled and associated anomaly features, and the verified root cause device pointing information to generate the self-sensing results of the coupled operating status of high-risk equipment in thermal power plants.

Images