Intelligent monitoring method and device based on drying bed blockage and bed collapse risk prompt

Abstract

The invention provides an intelligent monitoring method and device based on drying bed blockage and collapse risk prompt, and belongs to the technical field of industrial process intelligent early warning and fault diagnosis. Aiming at the problems that the traditional DCS has many fixed value alarm false alarms, a static light indicator cannot reflect a fault evolution process, and the coping experience of an operator is insufficient, the system combines multiple regression and a state estimation model to carry out real-time monitoring and trend analysis on key parameters of a drying bed. And when the parameters deviate from a normal model or meet a preset logic condition, the system automatically triggers graded early warning and pushes expert processing suggestions.

Description

Technical Field

[0001] This invention relates to the field of intelligent early warning and fault diagnosis technology for industrial processes, and in particular to a smart monitoring method and device based on the risk warning of drying bed blockage and collapse. Background Technology

[0002] Drying beds, as key equipment in wastewater treatment systems of thermal power plants, are widely used in the evaporation and crystallization process of high-salt wastewater. With the development of industrial automation and intelligent diagnostic technologies, wastewater treatment systems based on DCS platforms have established a complete technical system covering data acquisition, process control, and status monitoring. Specifically, this system achieves real-time monitoring of core parameters such as drying fan frequency, current, and bed temperature through the collaborative operation of distributed processing units and I / O modules. Traditional monitoring methods mainly rely on static setpoint alarm strategies, using preset electrical parameter thresholds and equipment status logic to determine faults, but their technological evolution has consistently failed to overcome the limitations of "post-event response."

[0003] However, existing alarm mechanisms that directly use fixed thresholds have systemic flaws. Specifically, static setpoints cannot adapt to the dynamic changes in the drying bed under different loads and operating conditions, leading to misjudgments or omissions of key abnormal features such as sudden changes in bed temperature. Therefore, in typical fault scenarios such as air distribution plate blockage (fan frequency and current divergence), nozzle blockage, and bed collapse risk, traditional methods can only trigger alarms through a single parameter exceeding its limit, failing to establish a dynamic early warning model based on the nonlinear correlation between parameters. The problem of delayed early warning for bed collapse risk is particularly prominent; the resulting cascading failures in the electrostatic precipitator system can cause unplanned unit shutdowns, and existing technologies lack a quantitative description of the fault development process, making it difficult for operators to take effective intervention measures in the early stages of a fault. Summary of the Invention

[0004] The present invention aims to at least partially solve one of the technical problems in the related art.

[0005] Therefore, the first objective of this invention is to propose a smart monitoring method based on the risk warning of dry bed blockage and collapse.

[0006] Another objective of this invention is to provide a smart monitoring device for alerting users of the risk of blockage and collapse in a drying bed.

[0007] The third objective of this invention is to provide a computer device.

[0008] A fourth objective of this invention is to provide a non-transitory computer-readable storage medium.

[0009] To achieve the above objectives, a first aspect of the present invention proposes a smart monitoring method based on the risk warning of dry bed blockage and collapse, comprising: S1, Collect multi-dimensional operational data of the drying bed in the wastewater treatment system of a thermal power plant. The operational data includes electrical parameters, equipment status parameters, and process parameters. S2, perform nonlinear correlation analysis on the operating data based on the multivariate state estimation model, calculate the residual between the actual parameters and the estimated parameters, and determine whether there is an abnormal state based on the dynamic statistical confidence interval; S3. When an abnormal state is detected, the expert rule base is called to diagnose the abnormality type and generate corresponding early warning information and handling opinions. The handling opinions include backflushing operation guidance, standby equipment switching suggestions or fluidization parameter adjustment schemes. S4. The warning information and handling suggestions are displayed in a hierarchical and visual manner through a human-computer interaction interface, and a closed-loop decision-making link of warning-diagnosis-handling suggestions is established.

[0010] In one embodiment of the present invention, S1 includes: S11, collect real-time temperature data of bed temperature 1 and bed temperature 2 of drying bed A / B, for subsequent judgment of uneven bed temperature distribution; S12 collects frequency and current data of drying fans A / B to analyze the correlation between ventilation volume and load anomalies.

[0011] In one embodiment of the present invention, S2 includes: S21 uses three consecutive months of historical data on normal operation of the drying bed as the training set, and removes interference from abnormal operating conditions through data cleaning; S22, Based on the dynamic calculation formula of statistical confidence interval ,in The mean of the parameters, Standard deviation The confidence coefficient is used to determine whether the residuals exceed the normal range.

[0012] In one embodiment of the present invention, S3 includes: S31, when the temperature difference between bed temperature 1 and bed temperature 2 is detected to be greater than... When this happens, the nozzle blockage diagnosis rule is triggered and instructions to check whether the nozzle and pipeline are unobstructed are sent. S32, when the bed temperature fluctuation is detected to exceed... When this occurs, the bed collapse risk diagnosis rule is triggered, and a handling suggestion to adjust the air volume and air pressure to the normal fluidization range is pushed.

[0013] In one embodiment of the present invention, S4 includes: S41 uses a yellow warning sign to indicate an abnormal state and a red alarm sign to indicate a risk of bed collapse. S42 displays a bed temperature trend graph via a pop-up window, and overlays a threshold line indicating the temperature difference between bed temperature 1 and bed temperature 2. and .

[0014] To achieve the above objectives, a second aspect of the present invention provides a smart monitoring device for detecting the risk of blockage and collapse in a drying bed, comprising: The data acquisition module is used to collect multi-dimensional operational data of the drying bed in the wastewater treatment system of a thermal power plant. The operational data includes electrical parameters, equipment status parameters, and process parameters. The state estimation and analysis module performs nonlinear correlation analysis on the operating data based on a multivariate state estimation model, calculates the residuals between the actual parameters and the estimated parameters, and determines whether there is an abnormal state based on the dynamic statistical confidence interval. The expert rule diagnosis module is used to call the expert rule library to diagnose the abnormality type when an abnormal state is detected, and generate corresponding early warning information and handling opinions. The handling opinions include backflush operation guidance, standby equipment switching suggestions or fluidization parameter adjustment schemes. The closed-loop decision display module is used to display the early warning information and handling opinions in a hierarchical and visual manner through a human-computer interaction interface, and to establish a closed-loop decision link of early warning-diagnosis-handling suggestions.

[0015] This invention discloses a smart monitoring method and device based on the risk warning of dry bed blockage and collapse, which enables early and accurate warning and intelligent diagnosis of the risk of dry bed blockage and collapse, improves the timeliness and accuracy of fault identification, and effectively reduces the risk of unplanned downtime and maintenance costs.

[0016] To achieve the above objectives, a third aspect of this application provides a computer device, including a processor and a memory; wherein the processor runs a program corresponding to the executable program code stored in the memory, in order to implement a smart monitoring method based on the risk warning of dry bed blockage and collapse as described in the first aspect embodiment.

[0017] To achieve the above objectives, a fourth aspect of this application provides a non-transitory computer-readable storage medium storing a computer program that, when executed by a processor, implements a smart monitoring method for alerting the risk of blockage and collapse of a dry bed as described in the first aspect embodiment.

[0018] Additional aspects and advantages of the invention will be set forth in part in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. Attached Figure Description

[0019] The above and / or additional aspects and advantages of the present invention will become apparent and readily understood from the following description of the embodiments taken in conjunction with the accompanying drawings, wherein: Figure 1 This is a flowchart of a smart monitoring method based on the risk warning of dry bed blockage and collapse according to an embodiment of the present invention; Figure 2 This is a schematic diagram of the drying bed status monitoring according to an embodiment of the present invention; Figure 3 This is a structural diagram of a smart monitoring device based on the risk warning of dry bed blockage and collapse according to an embodiment of the present invention; Figure 4 It is a computer device according to an embodiment of the present invention. Detailed Implementation

[0020] It should be noted that, unless otherwise specified, the embodiments and features described in the present invention can be combined with each other. The present invention will now be described in detail with reference to the accompanying drawings and embodiments.

[0021] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0022] The following description, with reference to the accompanying drawings, describes a smart monitoring method and apparatus based on the risk warning of dry bed blockage and collapse according to an embodiment of the present invention.

[0023] Example 1 Figure 1 This is a flowchart of a smart monitoring method based on the risk warning of dry bed blockage and collapse according to an embodiment of the present invention, such as... Figure 1 As shown, it includes: S1, collect multi-dimensional operational data of the drying bed in the wastewater treatment system of a thermal power plant. The operational data includes electrical parameters, equipment status parameters, and process parameters.

[0024] Specifically, this step involves collecting multi-dimensional operational data of the drying bed in the wastewater treatment system of a thermal power plant, and is a fundamental link in the entire intelligent early warning and diagnosis system. In some implementations, this step uses the distributed processing unit (DPU) and I / O modules of the DCS platform to achieve comprehensive perception of the drying bed's operating status, providing high-quality, real-time data support for subsequent anomaly detection and fault diagnosis.

[0025] Furthermore, the operational data collected by the system mainly includes three categories: electrical parameters, equipment status parameters, and process parameters. Electrical parameters include the frequency and current of the drying fans A and B, such as the frequency of the drying fans. and current These parameters reflect the operating load and ventilation capacity of the blower. Equipment status parameters cover the on / off status and control mode of various electric doors, regulating doors, blowers, and pumps, such as the "open" status of the main electric isolation door for hot air and the main electric isolation door for cold air in the #2 furnace drying zone, and whether the cold air regulating door of drying blower A is in "automatic mode." This status information is used to determine whether the drying bed is in sequential control operation. Process parameters include bed temperature 1, bed temperature 2, air pressure, flow rate, etc., such as the temperature difference between bed temperature 1 and bed temperature 2. It is used to identify abnormal phenomena such as nozzle blockage or uneven material distribution.

[0026] Furthermore, the system has strict requirements for acquisition frequency, accuracy, and data integrity. The acquisition frequency for electrical parameters is typically 100ms, and the accuracy must meet the ±0.5% error requirement for analog signals in the IEC60870-5-104 standard. Equipment status parameters are acquired digitally, and the response time should be less than 50ms to ensure real-time status changes. Process parameters such as temperature and pressure must conform to the sampling period and engineering unit (EU) settings of process variables (PV) in the DCS system. For example, the sampling period for bed temperature measuring points is 1s, the range is 0–200°C, and the accuracy is ±1°C.

[0027] Furthermore, this step is deployed in the DCS monitoring server of the thermal power plant's wastewater treatment system, acquiring data in real time via the OPC protocol or the DCS's native interface. The collected data will be used to construct a training set for a multivariate state estimation (MSET) model and as input variables for the expert rule base. For example, when determining if the air distribution plate is blocked, the system must simultaneously meet the generator power requirement. Multiple sub-conditions related to the sequential control of the drying bed's operating status.

[0028] Furthermore, this step provides the system with comprehensive and accurate operational data support, which is a key prerequisite for achieving a closed loop from anomaly detection to fault diagnosis. Through the fusion analysis of multi-dimensional data, the system can capture early abnormal signals that are difficult to identify with traditional setpoint alarms, thereby significantly improving the safety and stability of the drying bed operation.

[0029] Furthermore, S1 includes: S11: Collect real-time temperature data of bed temperature 1 and bed temperature 2 of drying beds A / B for subsequent judgment of uneven bed temperature distribution.

[0030] Specifically, in the intelligent early warning and diagnosis system for the drying bed of the wastewater treatment system of the present invention, the real-time temperature data of drying bed A / B bed temperature 1 and bed temperature 2 are collected to determine the uneven distribution of bed temperature. This step uses the I / O module of the DCS platform to perform high-frequency sampling and real-time transmission of temperature signals from multiple measuring points inside the drying bed, providing basic data support for subsequent anomaly detection and fault diagnosis.

[0031] Furthermore, this step employs a distributed data acquisition architecture, periodically sampling temperature points such as temperature 1 and temperature 2 of drying beds A and B via the analog input (AI) module of the DCS. The sampling frequency is typically set to 1 second per sampling to ensure high sensitivity in capturing temperature change trends. The acquired temperature data is expressed in engineering units (e.g., °C) and uploaded in real-time to the monitoring server via the DCS network using the OPC protocol or a dedicated interface for use by the background analysis module. In some implementations, the system can also filter the raw data, such as using a first-order low-pass filter with a cutoff frequency set to 0.1 Hz, to eliminate transient noise interference and improve data stability.

[0032] Furthermore, the temperature range of bed temperature 1 and bed temperature 2 is typically between 100℃ and 300℃, depending on the operating conditions of the drying bed and the characteristics of the material. The system determines whether the bed temperature distribution is uniform by comparing the real-time difference between bed temperature 1 and bed temperature 2. For example, in rule three, when the temperature difference exceeds... When this occurs, the system will trigger a nozzle blockage alarm, indicating that the material or moisture is severely unevenly distributed on the bed surface, which may be caused by localized nozzle blockage.

[0033] Furthermore, this step is widely applied in the operation monitoring of the drying bed in thermal power plant wastewater treatment systems. During the sequential operation of the drying bed, the system continuously collects temperature data and makes a comprehensive judgment based on the current spraying / water spraying status. This data acquisition process needs to be synchronized with other measuring points in the DCS (such as fan frequency, current, etc.) to achieve multi-parameter joint analysis and improve diagnostic accuracy.

[0034] Furthermore, it provides high-precision and timely input data for subsequent anomaly detection based on MSET and fault diagnosis based on expert rule base. Through the acquisition and analysis of real-time temperature data, the system can identify abnormal bed temperature distribution, thereby issuing early warnings before faults occur, effectively improving the safety and stability of the drying bed operation.

[0035] S12 collects frequency and current data of drying fans A / B to analyze the correlation between ventilation volume and load anomalies.

[0036] Specifically, in the intelligent early warning and diagnosis system for the drying bed of the wastewater treatment system of the present invention, the frequency and current data of the drying fans A / B are collected. This step provides data support for subsequent fault diagnosis based on multivariate state estimation (MSET) and expert rule base by acquiring the operating parameters of the fans in real time.

[0037] Furthermore, this step relies on the I / O module of the DCS platform to collect real-time data from the frequency converters (VFDs) and current transformers (CTs) of the dryer fans A and B via standard Modbus TCP or OPCUA protocols. Frequency data reflects the fan speed control signal, typically in Hz, with a sampling period of 1 to 5 seconds; current data is measured by the CT to determine the actual operating current of the motor, in amperes (A), with the sampling frequency consistent with the frequency data. The collected data is encapsulated into standard DCS measurement point signals and transmitted to the monitoring server via the DCS network for real-time analysis and historical data modeling.

[0038] Furthermore, the system focuses on the DCS dynamic coupling relationship between the frequency and current of the drying fan. For example, when the fan frequency is higher than 40Hz ( However, the current is less than 120A. When the value is 0, it indicates that although the fan is increasing the ventilation volume, the actual load has not increased accordingly, which may suggest that the air distribution panel is blocked or the air duct resistance is abnormal. This parameter combination is one of the core bases in the expert rule base for judging the risk of air distribution panel blockage.

[0039] Furthermore, this step is applicable to the continuous operation monitoring of drying beds in wastewater treatment systems of thermal power plants. During the operation of drying beds A / B in boiler #2, the system continuously collects fan frequency and current data, and performs comprehensive analysis in conjunction with other process parameters (such as bed temperature, air pressure, etc.).

[0040] Furthermore, this step provides the system with crucial anomaly detection input signals, helping to identify potential ventilation anomalies by analyzing the divergence between frequency and current before the fan triggers a traditional setpoint alarm. Combining residual analysis of the MSET model with logical judgments based on expert rules, the system can provide early warnings for faults such as drying bed blockage and collapse, thereby improving the safety and economy of equipment operation and reducing the risk of unplanned downtime.

[0041] S2, based on the multivariate state estimation model, perform nonlinear correlation analysis on the operating data, calculate the residual between the actual parameters and the estimated parameters, and determine whether there is an abnormal state based on the dynamic statistical confidence interval.

[0042] Specifically, in the intelligent analysis and diagnosis layer of the present invention, nonlinear correlation analysis is performed on the drying bed operation data based on a multivariate state estimation model. This step identifies abnormal deviations in the operating state by constructing a dynamic statistical model between parameters, thereby achieving early warning of risks before the fault triggers a traditional setpoint alarm.

[0043] Furthermore, the MSET model is first trained based on multi-dimensional operational data collected from a historical database. This data includes, but is not limited to, key parameters such as temperature, air pressure, flow rate, fan frequency, and current at various measuring points on the drying bed. During model training, it is necessary to ensure that the data covers a variety of typical healthy operating conditions to improve the model's generalization ability to normal states. The training data typically needs to meet a sampling frequency of no less than 1 second / time, and the recommended time window length is 30 minutes to 2 hours to capture the dynamic coupling relationship between parameters.

[0044] Furthermore, the MSET model learns the intrinsic correlations between parameters through nonlinear regression or state-space modeling methods. During the runtime phase, the system inputs parameters collected in real time into the model, and the model outputs the "estimated parameter values" that should be present under healthy operating conditions. Subsequently, the system calculates the residuals between the actual parameters and the estimated parameters, i.e. ,in For the first The actual measured values ​​of each parameter These are model estimates. The residuals reflect the degree of deviation between the system's operating state and the normal pattern.

[0045] Furthermore, the system employs dynamic statistical confidence intervals (DSCI) to determine anomalies in the residuals. DSCI is based on the statistical characteristics of historical residuals (such as mean). Standard deviation The confidence interval is calculated dynamically and is usually set as follows: ,in This is the confidence coefficient, with a recommended value of 2.5 to 3.0, corresponding to a confidence level of 95% to 99%. When any residual exceeds this range, the system determines that the parameter is in an abnormal state and triggers an alert.

[0046] Furthermore, the monitoring server deployed on the DCS platform binds real-time data with the DCS measurement point addresses. Its application scenarios cover the early identification of typical faults such as drying bed air distribution plate blockage, nozzle blockage, and bed collapse risk, and it is especially suitable for complex working conditions with slow parameter changes and significant nonlinear coupling.

[0047] Furthermore, this step effectively enhances the system's sensitivity and robustness to abnormal states, avoiding the false alarms and missed alarms problems of traditional set-value alarms. Through nonlinear correlation analysis, the system can capture the implicit relationships between parameters, thus issuing early warnings before the fault manifests as a significant numerical deviation, providing operators with an intervention window and significantly enhancing the safety and stability of the drying bed operation.

[0048] Furthermore, S2 includes: S21 uses three consecutive months of historical data on normal operation of the drying bed as the training set, and removes interference from abnormal operating conditions through data cleaning.

[0049] Specifically, in the implementation of this invention, three consecutive months of normal operating history data of the drying bed are used as the training set. Abnormal operating conditions are removed through data cleaning, aiming to provide the MSET model with high-quality and representative health status data, thereby improving the model's ability to identify abnormal behaviors and the accuracy of early warning.

[0050] Further, this step first extracts three consecutive months of drying bed operation data from the DCS's historical database, covering multiple key measurement points such as drying fan frequency, current, bed temperature 1 and bed temperature 2, air pressure, and flow rate. The data acquisition period should cover different load conditions (e.g., generator power ranging from 0MW to full load) to ensure the model has good generalization ability. Subsequently, the system cleans the raw data, removing abnormal operating condition data caused by equipment failure, misoperation, DCS signal drift, or communication interruption. The data cleaning process includes, but is not limited to: removing values ​​exceeding reasonable physical ranges (e.g., bed temperature above 1200°C or below 50°C), identifying and filtering continuous abnormal fluctuations, and eliminating outliers caused by DCS sampling anomalies through time series consistency analysis.

[0051] Furthermore, data cleaning requires setting reasonable thresholds based on actual operational experience. For example, the normal fluctuation range of the drying fan frequency is typically 30Hz to 50Hz, and the normal current range is 120A to 180A. The temperature difference between bed temperature 1 and bed temperature 2 should generally not exceed 20°C under healthy conditions; exceeding this value may be a precursor to nozzle blockage. The cleaned dataset should meet the following standards: temporal continuity (sampling interval of 1 minute), data integrity (missing rate less than 5%), and representativeness of operating conditions (covering at least 80% of typical operating modes).

[0052] Furthermore, this step is applicable to long-term health status modeling of dryer beds in thermal power plant wastewater treatment systems. Using a high-quality dataset after cleaning, the MSET model can learn the normal parameter coupling relationships of the dryer bed under different loads, seasons, and material properties, thereby identifying potential anomalies through residual analysis during real-time operation. For example, when the fan frequency increases but the current does not increase synchronously, the model can identify a possible trend of air distributor blockage.

[0053] Furthermore, this step provides a solid data foundation for subsequent anomaly detection and intelligent diagnosis. By removing interfering data, the accuracy of model training is significantly improved, which helps to enhance the sensitivity and specificity of early warnings, reduce false alarms and missed alarms, and thus realize the transformation from "post-event alarm" to "pre-event early warning," providing operators with more reliable decision support.

[0054] S22, Based on the dynamic calculation formula of statistical confidence interval ,in The mean of the parameters, Standard deviation The confidence coefficient is used to determine whether the residuals exceed the normal range.

[0055] Furthermore, this step first extracts operational data of the drying bed under various healthy operating conditions from the historical database as training samples. The system then performs sliding window statistical analysis on each key parameter (such as bed temperature, air pressure, and flow rate) to calculate its current mean. with standard deviation Confidence coefficient It is usually set based on actual operating experience, for example, at a 95% confidence level. Or, in industrial settings, to improve sensitivity, settings can be configured... Or higher. Confidence interval Used to define the reasonable fluctuation range of parameters under normal operating conditions.

[0056] Furthermore, the system collects current parameter values ​​in real time. And calculate its estimate compared with the output of the MSET model. The residuals between If the residual If the parameter deviates from the normal pattern, an anomaly warning is triggered. This method considers not only the absolute value of the parameter but also its degree of deviation from the overall system state, thus effectively identifying slowly varying and nonlinear anomalies.

[0057] Furthermore, this step is deployed on the DCS monitoring server and tightly integrated with the real-time data acquisition module of the DCS platform. Its technical specifications include residual threshold sensitivity, confidence interval width, and early warning response time, typically requiring a residual detection delay of no more than 10 seconds to ensure timely early warning. This step acts as the "first line of defense" in the system, providing data support for subsequent fault diagnosis using the expert rule base, significantly improving the accuracy of identifying the drying bed's operating status and its early warning capabilities.

[0058] S3. When an abnormal state is detected, the expert rule base is called to diagnose the abnormality type and generate corresponding early warning information and handling opinions. The handling opinions include backflushing operation guidance, standby equipment switching suggestions or fluidization parameter adjustment schemes.

[0059] Furthermore, the expert rule base consists of multiple predefined diagnostic rules, each corresponding to a specific anomaly type and its processing logic. The system matches real-time collected DCS measurement data (such as dryer fan frequency, current, bed temperature difference, etc.) with the conditions in the rules to determine whether the triggering conditions for a certain type of fault are met. For example, when the dryer fan frequency is detected... And current When this occurs, the system will trigger a "air distributor blockage" alarm and send a response: "Check the air distributor ventilation and prepare for backflushing or shutdown for cleaning." This type of rule is based on modeling the typical electrical behavior of the fan in a blocked state, i.e., the frequency increases but the current decreases, indicating that the fan is not effectively driving the air volume under high load, thus judging that the air distributor may be blocked.

[0060] Furthermore, each rule in the system has a clearly defined trigger threshold and logical combination method. For example, the determination of nozzle blockage depends on the temperature difference between bed temperature 1 and bed temperature 2. Furthermore, the system is in either shotcreting or water spraying mode. This temperature difference threshold is set based on statistical analysis of uneven bed temperature distribution caused by nozzle blockage in historical fault data, and has high engineering applicability. In addition, the trigger condition for the bed collapse risk warning is a short-term fluctuation in bed temperature. This indicator reflects drastic changes in the fluidization state of the bed material and is an early sign of bed collapse accidents.

[0061] Furthermore, this step is widely applied to the operation monitoring of dryer beds in wastewater treatment systems of thermal power plants. When the system detects blockage of the air distribution plate, nozzle blockage, or risk of bed collapse, operators can receive tiered early warning information (such as yellow warnings and red alarms) on the intelligent monitoring HMI interface of the DCS operator station, and simultaneously obtain expert opinions. These opinions include backflushing operation guidance, suggestions for switching to standby equipment, or fluidization parameter adjustment schemes. Specific operations can be performed manually or automatically in conjunction with the DCS sequential control system, thereby achieving rapid response and precise handling.

[0062] Furthermore, this step significantly improves the system's fault diagnosis accuracy and response efficiency. By transforming expert experience into executable rule logic, the system can not only identify anomalies but also provide targeted handling suggestions, effectively reducing the operator's misjudgment rate and operational risks. Simultaneously, this mechanism achieves knowledge standardization and automation, providing solid support for the intelligent operation and maintenance of thermal power plant wastewater treatment systems.

[0063] Furthermore, S3 includes: S31, when the temperature difference between bed temperature 1 and bed temperature 2 is detected to be greater than... When this happens, the nozzle blockage diagnosis rule is triggered and instructions to check whether the nozzle and pipeline are unobstructed are sent.

[0064] Specifically, when the temperature difference between bed temperature 1 and bed temperature 2 is detected to be greater than... When this occurs, the system will trigger the nozzle blockage diagnosis rule and push the instruction to "check whether the nozzle and pipeline are unobstructed." Its technical implementation principle and operation method are as follows: Furthermore, this step involves real-time monitoring of the temperature difference between two key temperature measuring points in drying bed A (or B)—bed temperature 1 and bed temperature 2—to determine if localized nozzle blockage has occurred. Bed temperature 1 and bed temperature 2 are typically located in different areas of the drying bed to reflect the uniformity of the temperature distribution on the bed material surface. Under normal operating conditions, due to the uniform distribution of the sprayed slurry or water, the temperature difference between bed temperature 1 and bed temperature 2 should be maintained within a small range, typically not exceeding [a certain value]. When the temperature difference exceeds This indicates that the material or moisture is severely unevenly distributed on the bed surface, which may be due to partial blockage of the nozzles or obstruction of the pipeline.

[0065] Furthermore, the triggering conditions for this rule consist of three parts: First, the generator power must be greater than [a certain value]. First, to ensure the unit is in operation; second, the drying bed sequential control operation is in place, meaning the relevant electric doors, regulating doors, and fans meet the preset logic conditions; finally, the system is in spraying or water spraying mode, and the temperature difference between bed temperature 1 and bed temperature 2 exceeds [a certain threshold]. These conditions together form the diagnostic criteria for nozzle blockage, ensuring that the system only makes judgments under relevant operating conditions and avoiding false alarms.

[0066] Furthermore, this rule is deployed in the expert rule engine of the DCS platform. It acquires the values ​​of bed temperature 1 and bed temperature 2 through the real-time data acquisition module and compares them within the rule execution cycle. If the temperature difference exceeds the set threshold, the system immediately pops up a "nozzle blockage" alarm on the HMI interface and pushes specific handling suggestions, such as "check whether the relevant nozzles and pipelines are unobstructed, adjust the grouting / water pressure or switch to a standby bed".

[0067] Furthermore, by quantifying indicators of abnormal temperature distribution, early identification of nozzle blockage, a typical fault, is achieved. Compared to traditional fixed-value alarms, this rule, based on multi-parameter logical combinations, has higher diagnostic accuracy and specificity, helping operators to take timely measures to prevent serious consequences such as uneven bed fluidization, localized overheating, or even bed collapse caused by nozzle blockage.

[0068] S32, when the bed temperature fluctuation is detected to exceed... When this occurs, the bed collapse risk diagnosis rule is triggered and the processing suggestion of "adjusting air volume and air pressure to the normal fluidization range" is pushed.

[0069] Specifically, in some implementations, when the temperature fluctuation of the drying bed is detected to exceed a certain threshold... When this occurs, the system will trigger the bed collapse risk diagnosis rule and push the processing suggestion of "adjusting the air volume and air pressure to the normal fluidization range". This step is based on rule four defined in the expert rule base, and its technical implementation principle involves real-time monitoring and anomaly judgment of the bed temperature change rate.

[0070] Furthermore, the system continuously acquires temperature data from a measurement point on bed temperature 1 of the drying bed A in furnace #2 via the real-time data acquisition function of the DCS platform. This measurement point is typically provided by thermocouples or infrared temperature measuring devices installed on the bed surface, with a sampling frequency generally set to 1 second / time to ensure high sensitivity in capturing temperature changes. After entering the spraying state, the system begins dynamic monitoring of bed temperature 1, calculating its maximum and minimum temperature difference within a set time window (e.g., 30 seconds to 2 minutes). If this difference exceeds... This will trigger a bed collapse risk warning.

[0071] Furthermore, the key threshold for this rule is the fluctuation range of bed temperature 1. This value is set based on statistical analysis of temperature anomalies preceding bed collapse events in historical operational data. Furthermore, the system needs to determine whether it is currently in a shotcreting state, which is confirmed by the output signal (such as `SprayOn`) of the shotcreting control logic in the DCS. In practical applications, this rule is typically linked with the anomaly detection results of the MSET model, forming a dual early warning mechanism of "data-driven + rule-based reasoning".

[0072] Furthermore, this step is applicable to the operation monitoring of drying beds in wastewater treatment systems of thermal power plants, especially under conditions such as high load, high humidity, or abnormal bed material particle size, where the risk of bed collapse increases significantly. The system pushes early warning information to operators in the HMI interface through red pop-ups or trend change prompts, along with clear operational suggestions, such as adjusting the drying fan frequency, regulating air pressure, and checking the bed material particle size, to restore the normal fluidization state of the bed.

[0073] Furthermore, by capturing abnormal fluctuations in bed temperature, the system can provide early warnings before bed collapse occurs, thus offering operators a crucial window for intervention. Once bed collapse occurs, it can lead to bed material subsidence, fluidization failure, and even a chain reaction of malfunctions in the electrostatic precipitator system. Therefore, this rule effectively enhances the system's fault prediction capabilities, reduces the risk of equipment damage, and strengthens the operational stability and safety of the wastewater treatment system.

[0074] S4. The warning information and handling suggestions are displayed in a hierarchical and visual manner through a human-computer interaction interface, and a closed-loop decision-making link of warning-diagnosis-handling suggestions is established.

[0075] Specifically, in the third part of this invention—the Human-Machine Interaction and Execution Layer—the step of "displaying the early warning information and handling opinions in a hierarchical and visual manner through a human-machine interface, and establishing a closed-loop decision-making link of early warning-diagnosis-handling suggestions" is a key step in realizing the system's intelligence, visualization, and operability. This step, by constructing a multi-level, multi-dimensional HMI interface, presents the early warning and diagnostic results from the intelligent analysis and diagnosis layer to the operators in an intuitive and hierarchical manner, and simultaneously pushes expert handling opinions, thereby forming a closed-loop control process from anomaly detection to fault diagnosis and then to handling suggestions.

[0076] Furthermore, this step begins with custom interface development based on the OVATIONDMI or HMI framework. The system subscribes in real-time to warning signals from the MSET model and expert rule engine via OPCUA or the DCS native interface, including warning level (e.g., yellow warning, red alarm), occurrence time, parameter name, anomaly type, and corresponding handling suggestions. Warning information is displayed on the HMI interface using dynamic graphical elements (e.g., flashing icons, trend curves, status bars) and pop-up prompts, ensuring that operators can identify potential risks immediately. Simultaneously, the system supports historical trend retrospective functionality, facilitating the analysis of anomaly evolution.

[0077] Furthermore, the grading of early warning information is based on preset confidence intervals and rule-based triggering conditions. For example, when the MSET model detects that the bed temperature residual exceeds 3σ (standard deviation), the system triggers a yellow warning; if the fan frequency is also met... And current If the alarm is triggered, it will be upgraded to a red alert. The appropriate action will be pushed out based on predefined rules in the rule base, such as "Check the ventilation of the air distribution panel and prepare for backflushing or shutdown for cleaning."

[0078] Furthermore, this step is widely applied to the operation monitoring of dryer beds in wastewater treatment systems of thermal power plants. Operators can monitor the health status of the dryer beds in real time via a dedicated HMI interface on the DCS operator station, and promptly respond to system-generated treatment suggestions, thereby effectively preventing major accidents such as bed collapse and blockages.

[0079] Furthermore, by visually integrating early warning information with handling suggestions, the response efficiency and decision-making accuracy of operators have been significantly improved, realizing the transformation from "passive response" to "proactive intervention," and providing key support for building an intelligent, efficient, and safe wastewater treatment system for thermal power plants.

[0080] Furthermore, S4 includes: S41 uses a yellow warning sign to indicate an abnormal state and a red alarm sign to indicate a risk of bed collapse.

[0081] Specifically, in the "human-machine interaction and execution layer" of this invention, an abnormal state is indicated by a yellow warning sign and a red alarm sign is indicated by a risk of bed collapse. This step uses a hierarchical color coding mechanism on the HMI (human-machine interface) of the DCS operator station to provide real-time and intuitive alarm display of the drying bed's operating status, thereby improving the efficiency and accuracy of operators' response to potential faults.

[0082] Furthermore, this step involves embedding a dynamic status indicator module into the existing drying bed monitoring screen using the graphical development tools of the DCS platform. The system performs residual analysis by comparing real-time collected measurement data with the estimated values ​​output by the MSET model, or dynamically triggers early warning states based on logical judgments from the expert rule base. Early warning states are divided into two levels: a yellow warning indicates that parameter behavior deviates from the normal pattern but has not yet reached the fault threshold, while a red alarm indicates that serious risks such as bed collapse have been triggered, requiring immediate intervention.

[0083] Furthermore, the trigger for a yellow alert is when the residuals of key parameters in the MSET model exceed a set statistical confidence interval (such as a 95% confidence interval), while a red alert is based on the bed temperature fluctuation threshold defined in Rule 4, i.e., after entering the shotcreting state, if the bed temperature 1 (or bed temperature 2) fluctuates (temperature difference at the same point) within a short period of time, it is greater than [a certain threshold]. This threshold, validated using historical fault data, exhibits high sensitivity and specificity, effectively identifying precursors to bed collapse.

[0084] Furthermore, this step is deployed at the DCS operator station of the thermal power plant's wastewater treatment system, allowing operators to visually observe changes in the drying bed's status during monitoring. Figure 2As shown. For example, when the system detects abnormal bed temperature fluctuations, the bed collapse risk indicator on the HMI screen changes from green to red, and an alarm window containing handling suggestions pops up, prompting operators to immediately adjust the air volume and air pressure to the normal fluidization range, check the bed material particle size and thickness, and stop feeding if necessary.

[0085] Furthermore, by combining graded color-coding with real-time trend tracking, dynamic and visual monitoring of the drying bed's operating status is achieved, effectively avoiding the lag and false alarm problems of traditional fixed-value alarms. Simultaneously, the synchronized delivery of early warning information and expert opinions improves the decision-making efficiency and accuracy of operators, thereby significantly reducing the probability of major accidents such as bed collapse and ensuring the safe and stable operation of the thermal power plant's wastewater treatment system.

[0086] S42 displays a bed temperature trend graph via a pop-up window, and overlays a threshold line indicating the temperature difference between bed temperature 1 and bed temperature 2. and .

[0087] Specifically, in the intelligent monitoring system of this invention, a bed temperature trend graph is displayed via a pop-up window, and a temperature difference threshold line between bed temperature 1 and bed temperature 2 is overlaid. and This step employs HMI graphical development technology based on the DCS platform at the technical implementation level, combined with a real-time data acquisition and processing mechanism, to ensure that operators can intuitively and promptly identify abnormal bed temperature trends.

[0088] Furthermore, the temperature difference threshold in this step is set based on statistical analysis of historical operating data and verification by expert experience. As an early warning threshold for nozzle clogging, it reflects the typical characteristics of uneven material or moisture distribution on the bed surface; and This serves as an emergency early warning threshold for bed collapse risk, used to identify early signals of disrupted bed fluidization. In practical applications, the system calculates in real time... And compare it with the above threshold to trigger the corresponding warning logic.

[0089] Furthermore, this pop-up function is deployed in the dedicated HMI interface of the DCS operator station. Operators can view bed temperature trends at any time during monitoring, especially during shotcreting or water spraying operations, where the system automatically activates this function to help determine if there is a risk of nozzle blockage or bed collapse. This step forms a closed loop with Rules 3 and 4 in the expert rule base, providing operators with intuitive decision-making support.

[0090] Furthermore, this step transforms abstract temperature difference data into perceptible visual signals through graphical means, significantly improving the response speed and accuracy of operators in judging abnormal conditions of the drying bed. Simultaneously, it complements the anomaly detection mechanism of the MSET model, enhancing the system's ability to identify unknown anomalies, thereby achieving a shift from "post-event alarm" to "pre-event warning," effectively reducing equipment failure rates and maintenance costs.

[0091] An embodiment of the present invention provides a smart monitoring method based on the risk warning of dry bed blockage and collapse, which can realize early and accurate warning and intelligent diagnosis of the risk of dry bed blockage and collapse, effectively improving the operational safety and maintenance efficiency of the wastewater treatment system of thermal power plants.

[0092] Example 2 This invention proposes an intelligent monitoring system based on the risk warning of dry bed blockage and collapse. The system aims to: This invention enables early and accurate warnings of drying bed air distribution plate blockage, nozzle blockage, and the risk of bed collapse. A dual diagnostic model based on multivariate state estimation and an expert rule base is constructed, forming a closed loop from anomaly detection to fault diagnosis. While issuing warnings, specific and actionable expert advice is provided to operators to assist in rapid and correct decision-making. This effectively reduces unplanned downtime of the drying bed and serious accidents (such as cascading failures of the electrostatic precipitator system caused by bed collapse), lowering equipment maintenance workload and costs. The core of the system described in this invention lies in constructing a hierarchical, intelligent analysis and decision-making architecture, whose system structure and working principle consist of the following three core parts: Furthermore, the data perception and acquisition layer includes: Foundation: Based on the existing OVATION DCS and its distributed processing unit (DPU). Data Source: The system collects multi-dimensional measurement data related to the operation of the drying bed in real time through the DCS's I / O modules, including but not limited to: Electrical parameters: generator power, frequency of drying fans A / B, and current of drying fans A / B. Equipment status: open, closed, and fault status of various electric doors / regulating doors; on / off status of fans and pumps. Process parameters: temperature at various points on drying beds A / B (bed temperature 1, bed temperature 2, etc.), air pressure, flow rate, etc. Data transmission: The above real-time data is transmitted to a monitoring server deployed with the system of this invention via a DCS network.

[0093] Furthermore, the intelligent analysis and diagnosis layer includes a hybrid intelligent diagnosis model that combines "Multivariate State Estimation (MSET)" with "Expert Rule Base".

[0094] In one embodiment of the present invention, the anomaly warning module based on multivariate state estimation (MSET) includes: Model Construction: Normal operating data of the drying bed under various typical health conditions is selected from historical databases to form a training dataset. Working Principle: The MSET model learns the inherent (non-linear) correlations between multiple parameters in these health data. During system operation, real-time data is input into the MSET model, which outputs a set of "estimated values" that should be present under healthy conditions. The real-time measurements are compared with the estimated values, and the residuals are calculated. Anomaly Detection: When the residual of any key parameter exceeds the set statistical confidence interval, even if the absolute value of the parameter does not reach the traditional alarm threshold, the system will issue an "anomaly warning," indicating to operators that the parameter's behavior has deviated from the normal pattern and may pose a potential risk.

[0095] In one embodiment of the present invention, the fault diagnosis module based on an expert rule base includes: Specifically, this module encapsulates the experience of senior operations and maintenance personnel, as well as the conclusions of historical accident analysis, forming the following executable diagnostic logic: Rule 1: Risk Warning Regarding Clogged Air Vents on Drying Sheets Triggering logic: Condition 1 AND Condition 2 Condition 1: Generator power > 0 MW (indicating that the unit is in operation).

[0096] Condition 2: The drying bed sequential control operation is in place. This status is determined by the comprehensive satisfaction of all the following sub-conditions: #2 Furnace Drying Zone Hot Air Main Electric Isolation Door Open #2 Furnace Drying Zone Cold Air Main Electric Isolation Door Open #2 Drying Fan A Imported Electric Door Opening #2 Dryer Fan A Cold Air Damper is in automatic mode. #2 Furnace Drying Fan A Closed Output result: When the above conditions are met, the HMI screen will display a "Pressure Distributor Blockage Risk Warning" to remind operators to strengthen monitoring.

[0097] Rule 2: Drying bed sheet air valve blockage alarm Triggering logic: Condition 1 AND Condition 2 Condition 1: Dryer fan frequency > 40 Hz (indicating that the system is attempting to increase ventilation).

[0098] Condition 2: Dryer fan current <120 A (indicating that the fan is hindered from working, the load is light, and it is a typical blockage characteristic).

[0099] Output result: Triggered "Air distribution panel blockage" alarm, and pushed handling opinion: "Check the ventilation of the air distribution panel, prepare for backflushing or shutdown for cleaning".

[0100] Rule 3: Clogged nozzles in the drying bed Triggering logic: Condition 1 AND Condition 2 AND Condition 3 Condition 1: Generator power > 0 MW.

[0101] Condition 2: The drying bed sequential control operation has been put into operation (judgment logic is the same as rule 1).

[0102] Condition 3: The system is in the spraying or water spraying state, and the temperature difference between bed temperature 1 and bed temperature 2 of drying bed A of furnace #2 is greater than 20℃ (indicating that the material or moisture is severely unevenly distributed on the bed surface and the nozzles are partially blocked).

[0103] Output result: Triggers "Nozzle blockage" alarm and pushes handling suggestions: "Check whether the relevant nozzles and pipelines are unobstructed, adjust the spraying / water pressure or switch to the standby bed."

[0104] Rule 4: Risk Warning of Drying Bed Collapse Triggering logic: Condition 1 Condition 1: After entering the spraying state, the temperature 1 (or temperature 2) of bed A in the drying bed of furnace #2 fluctuates by more than 40℃ within a short period of time (temperature difference at the same point) (indicating that the fluidization state of the bed material is destroyed, and local dead zones or channels appear, which is a precursor to bed collapse).

[0105] Output result: Triggered an emergency warning of "high risk of bed collapse" and pushed the following handling suggestions: "Immediately adjust the air volume and air pressure to the normal fluidization range, check the particle size and thickness of the bed material, and stop feeding if necessary."

[0106] In one embodiment of the present invention, the human-computer interaction and execution layer includes: Early Warning Display: On the DCS operator station, visual alarms are displayed through a dedicated intelligent monitoring HMI interface, using tiered color schemes (e.g., yellow alert, red alarm), pop-up windows, and trend tracking. Expert Opinion Push: Alarm information not only describes the problem but also directly includes "expert handling opinions" as defined in the above rules. Intervention Execution: Operators can manually adjust operations based on system guidance, or, after deep integration with the DCS sequential control system in the future, the system can recommend or automatically execute a series of adjustment commands.

[0107] The embodiments of the present invention also have the following technical effects: Safety Benefits: The system has successfully transitioned from "post-event alarms" to "pre-event warnings." For example, in 2025, the system provided an early warning of the blockage trend in #2 dryer bed A, allowing operators to intervene in time and successfully preventing a chain reaction of shutdowns caused by bed collapse, which could have resulted in blockage of the electrostatic precipitator inlet distribution plate and dust accumulation on the electrode plates, thus ensuring the inherent safety of the unit. Economic Benefits: Cost Reduction: Significantly reduced losses from unplanned downtime due to dryer bed malfunctions and expensive emergency repair costs. Efficiency Improvement: Through accurate early warnings, maintenance work was transformed from "emergency repairs" to "planned maintenance," reducing the workload and difficulty of cleaning and maintaining the dryer bed, and lowering labor costs.

[0108] Example 3 To achieve the above embodiments, such as Figure 4 As shown, this embodiment also provides a smart monitoring device 10 based on the risk warning of dry bed blockage and collapse, including: The data acquisition module 100 is used to collect multi-dimensional operating data of the drying bed in the wastewater treatment system of a thermal power plant. The operating data includes electrical parameters, equipment status parameters, and process parameters. The state estimation and analysis module 200 performs nonlinear correlation analysis on the operating data based on a multivariate state estimation model, calculates the residuals between the actual parameters and the estimated parameters, and determines whether there is an abnormal state based on the dynamic statistical confidence interval. The expert rule diagnosis module 300 is used to call the expert rule library to diagnose the abnormality type when an abnormal state is detected, and generate corresponding early warning information and handling opinions. The handling opinions include backflush operation guidance, standby equipment switching suggestions or fluidization parameter adjustment schemes. The closed-loop decision display module 400 is used to display the early warning information and processing opinions in a hierarchical and visual manner through a human-computer interaction interface, and to establish a closed-loop decision link of early warning-diagnosis-processing suggestions.

[0109] Furthermore, the data acquisition module 100 is also used for: Real-time temperature data of drying beds A / B, temperature 1 and temperature 2 are collected for subsequent judgment of uneven bed temperature distribution; Collect frequency and current data of drying fans A and B to analyze the correlation between ventilation volume and load anomalies.

[0110] Furthermore, the state estimation analysis module 200 is also used for: Three consecutive months of normal operating data of the drying bed were used as the training set, and abnormal operating conditions were removed through data cleaning. Based on the dynamic calculation formula of statistical confidence interval ,in The mean of the parameters, Standard deviation The confidence coefficient is used to determine whether the residuals exceed the normal range.

[0111] This invention provides an intelligent monitoring device based on the risk warning of drying bed blockage and collapse, which enables early and accurate warning and intelligent diagnosis of drying bed blockage and collapse risks, improves the timeliness and accuracy of fault identification, and effectively reduces the risk of unplanned downtime and maintenance costs.

[0112] Example 4 The present invention also provides an electronic device such as Figure 4 As shown, it includes a processor and a memory. The memory stores executable instructions. When the processor executes the instructions, it implements the above-mentioned intelligent monitoring method based on the risk warning of dry bed blockage and collapse.

[0113] Example 5 The present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the above-mentioned intelligent monitoring method based on the risk warning of blockage and collapse of a drying bed.

[0114] In the description of this specification, the references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., refer to specific features, structures, materials, or characteristics described in connection with that embodiment or example, which are included in at least one embodiment or example of the present invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples. Moreover, without contradiction, those skilled in the art can combine and integrate the different embodiments or examples described in this specification, as well as the features of different embodiments or examples.

[0115] Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include at least one of that feature. In the description of this invention, "a plurality of" means at least two, such as two, three, etc., unless otherwise explicitly specified.

Claims

1. A smart monitoring method based on the risk warning of dry bed blockage and collapse, characterized in that, include: S1, Collect multi-dimensional operational data of the drying bed in the wastewater treatment system of a thermal power plant. The operational data includes electrical parameters, equipment status parameters, and process parameters. S2, perform nonlinear correlation analysis on the operating data based on the multivariate state estimation model, calculate the residual between the actual parameters and the estimated parameters, and determine whether there is an abnormal state based on the dynamic statistical confidence interval; S3. When an abnormal state is detected, the expert rule base is called to diagnose the abnormality type and generate corresponding early warning information and handling opinions. The handling opinions include backflushing operation guidance, standby equipment switching suggestions or fluidization parameter adjustment schemes. S4. The warning information and handling suggestions are displayed in a hierarchical and visual manner through a human-computer interaction interface, and a closed-loop decision-making link of warning-diagnosis-handling suggestions is established.

2. The method as described in claim 1, characterized in that, S1 includes: S11, collect real-time temperature data of bed temperature 1 and bed temperature 2 of drying bed A / B, for subsequent judgment of uneven bed temperature distribution; S12 collects frequency and current data of drying fans A / B to analyze the correlation between ventilation volume and load anomalies.

3. The method as described in claim 1, characterized in that, The S2 includes: S21 uses three consecutive months of historical data on normal operation of the drying bed as the training set, and removes interference from abnormal operating conditions through data cleaning; S22, Based on the dynamic calculation formula of statistical confidence interval ,in The mean of the parameters, Standard deviation, The confidence coefficient is used to determine whether the residuals exceed the normal range.

4. The method as described in claim 1, characterized in that, The S3 includes: S31, when the temperature difference between bed temperature 1 and bed temperature 2 is detected to be greater than... When this happens, the nozzle blockage diagnosis rule is triggered and instructions to check whether the nozzle and pipeline are unobstructed are sent. S32, when the bed temperature fluctuation is detected to exceed... When this occurs, the bed collapse risk diagnosis rule is triggered, and a handling suggestion to adjust the air volume and air pressure to the normal fluidization range is pushed.

5. The method as described in claim 1, characterized in that, The S4 includes: S41 uses a yellow warning sign to indicate an abnormal state and a red alarm sign to indicate a risk of bed collapse. S42 displays a bed temperature trend graph via a pop-up window, and overlays a threshold line indicating the temperature difference between bed temperature 1 and bed temperature 2. and .

6. A smart monitoring device for detecting and alerting the risk of blockage and collapse in a drying bed, characterized in that, include: The data acquisition module is used to collect multi-dimensional operational data of the drying bed in the wastewater treatment system of a thermal power plant. The operational data includes electrical parameters, equipment status parameters, and process parameters. The state estimation and analysis module performs nonlinear correlation analysis on the operating data based on a multivariate state estimation model, calculates the residuals between the actual parameters and the estimated parameters, and determines whether there is an abnormal state based on the dynamic statistical confidence interval. The expert rule diagnosis module is used to call the expert rule library to diagnose the abnormality type when an abnormal state is detected, and generate corresponding early warning information and handling opinions. The handling opinions include backflush operation guidance, standby equipment switching suggestions or fluidization parameter adjustment schemes. The closed-loop decision display module is used to display the early warning information and handling opinions in a hierarchical and visual manner through a human-computer interaction interface, and to establish a closed-loop decision link of early warning-diagnosis-handling suggestions.

7. The apparatus as claimed in claim 6, characterized in that, The data acquisition module is also used for: Real-time temperature data of drying beds A / B, temperature 1 and temperature 2 are collected for subsequent judgment of uneven bed temperature distribution; Collect frequency and current data of drying fans A and B to analyze the correlation between ventilation volume and load anomalies.

8. The apparatus as claimed in claim 6, characterized in that, The state estimation and analysis module is also used for: Three consecutive months of normal operating data of the drying bed were used as the training set, and abnormal operating conditions were removed through data cleaning. Based on the dynamic calculation formula of statistical confidence interval ,in The mean of the parameters, Standard deviation, The confidence coefficient is used to determine whether the residuals exceed the normal range.

9. A computer device, characterized in that, Including processor and memory; The processor reads executable program code stored in the memory to run a program corresponding to the executable program code, so as to implement the intelligent monitoring method based on the risk warning of dry bed blockage and collapse as described in any one of claims 1-5.

10. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When executed by the processor, the program implements a smart monitoring method based on the risk warning of dry bed blockage and collapse as described in any one of claims 1-5.

Images