Method and system for unmanned overfire air injection in a coal-fired boiler

Abstract

The application discloses a coal-fired boiler unmanned blowing method and system, and belongs to the technical field of automatic control. The method comprises the following steps: confirming a blowing starting instruction sequence and acquiring four types of multi-source operation data of a blowing device, including a stroke position, a steam working condition, electrical operation and a visual image; performing feature extraction, fusion and cross-validation classification on the multi-source operation data to obtain a transmission, steam inlet and steam leakage fault alarm signal; responding to the fault alarm signal, interrupting the blowing program control, locking the blowing device and reconfirming the blowing starting instruction sequence according to the latest dirtiness index. The application fills in the monitoring blind area through full-stroke continuous position detection, excludes single detection source misjudgment through mutual verification of multiple criteria, and realizes the whole-process unmanned blowing.

Description

Technical Field

[0001] This application relates to the field of automatic control technology, and in particular to a method and system for unmanned follow-up blowing of a coal-fired boiler. Background Technology

[0002] In coal-fired boilers, ash and slag commonly accumulate on the heating surfaces during operation, requiring periodic operation of numerous steam sootblowers for removal to maintain heat exchange efficiency and prevent overheating and corrosion. The sootblowers operate in a high-temperature, high-dust environment, with the transmission and steam inlet components being high-risk areas for failure. However, existing sootblower control systems generally rely on single-type detection signals to diagnose sootblower faults. For example, they may rely solely on motor overcurrent to determine transmission abnormalities, steam header pressure to determine steam inlet abnormalities, or limit switches to confirm gun positioning. This limited criterion leads to frequent false alarms and missed alarms. False alarms cause frequent interruptions to the sootblowing process, requiring manual verification. Missed alarms result in jammed gun barrels becoming deformed or even burned and broken by the high-temperature flue gas, forcing boiler shutdown. Consequently, fault handling and sootblower monitoring must be done manually, making unmanned operation of the unit difficult. Summary of the Invention

[0003] This application provides a method and system for unmanned soot blowing in coal-fired boilers, which at least solves the problems of existing soot blowing control systems relying on a single type of detection signal to determine soot blower faults, frequent false alarms and missed alarms, and difficulty in achieving unmanned operation due to reliance on manual on-site soot blowing for fault handling and soot blowing monitoring.

[0004] Firstly, this application provides a method for unmanned operation of a coal-fired boiler, the method comprising: The soot blowing start-up command sequence is confirmed and multi-source operating data of the soot blower is obtained. The soot blowing start-up command sequence is used to indicate the start-up order and start-up time of the soot blowers corresponding to each heating surface of the boiler. The multi-source operating data includes soot blower stroke position data, steam condition data, electrical operation data and visual image data. Based on the multi-source operating data, a fusion diagnostic process is performed to obtain fault alarm signals. The fusion diagnostic process includes multi-source operating feature extraction, feature fusion, and cross-validation classification. The fault alarm signals include transmission fault signals, steam inlet fault signals, and steam leakage fault signals of the sootblower. In response to the fault alarm signal, perform safety procedures on the faulty sootblower and reconfirm the sootblowing start command sequence.

[0005] The above technical solution identifies jamming and slippage faults by rate monitoring based on continuous position signals throughout the entire stroke; eliminates false alarms from a single detection source by multi-source cross-verification combining position deviation and time deviation with current and rate characteristics; achieves accurate early warning of steam inlet faults in a single sootblower by using pressure ratio and flow increment as dual criteria; and eliminates encoder cumulative errors by self-verification through the sliding average of position code values, thus overcoming the shortcomings of frequent false alarms and missed alarms caused by a single criterion. Furthermore, by automatically interrupting program control in response to fault alarms, retracting the gun to lock the faulty sootblower, and rearranging the sootblowing sequence according to the latest dirt index, it achieves a complete unmanned closed loop of automatic sootblowing strategy generation, automatic equipment execution, automatic process monitoring, and automatic fault handling, without the need for manual on-site follow-up blowing, effectively reducing the risk of gun barrel burnout and heated surface tube rupture.

[0006] Secondly, this application provides an unmanned follow-up blowing system for a coal-fired boiler, the system comprising: a field sensing unit, a DCS control unit, a fusion diagnostic unit, an unmanned follow-up blowing control unit, and a local execution unit; The field sensing unit is configured to correspond to the sootblower and includes an absolute position encoder installed on the sootblower transmission mechanism, a post-valve pressure transmitter installed after the lifting valve of a single sootblower, a main pipe pressure transmitter and a steam flow transmitter installed on the sootblower steam main pipe, a sootblower motor current acquisition module, and a field camera device deployed on the corresponding sootblower. The absolute position encoder is used to output the continuous position signal of the sootblower gun body throughout its entire stroke and calculate the real-time travel speed, thus forming the stroke position data; the valve downstream pressure transmitter, the main pipe pressure transmitter and the steam flow transmitter are used to collect the inlet steam pressure, main pipe pressure and total steam flow of a single sootblower, respectively, thus forming the steam operating condition data. The motor current acquisition module is used to acquire the operating current of the sootblower drive motor, which constitutes the electrical operation data. The on-site camera device is used to collect image data covering the soot blower protective cover, walking parts and valve stem sealing parts, which constitutes the visual image data. The unmanned soot blowing control unit is used to confirm the soot blowing start command sequence, which is used to indicate the start order and start time of the soot blowers corresponding to each heating surface of the boiler. The fusion diagnostic unit is used to perform fusion diagnostic processing based on the multi-source operating data to obtain fault alarm signals. The fusion diagnostic processing includes multi-source operating feature extraction, feature fusion, and cross-validation classification. The fault alarm signals include transmission fault signals, steam inlet fault signals, and steam leakage fault signals of the sootblower. The local execution unit includes a sootblower electrical control cabinet, which is used to respond to the fault alarm signal and perform safety procedures on the faulty sootblower; The unmanned soot blowing control unit is used to respond to the fault alarm signal and reconfirm the soot blowing start command sequence.

[0007] Thirdly, this application provides an electronic device comprising one or more processors and one or more memories, wherein at least one piece of program code is stored in the one or more memories, the program code being loaded and executed by the one or more processors to implement the operations performed by the unmanned follow-up blowing method for the coal-fired boiler.

[0008] Fourthly, this application also provides a computer-readable storage medium storing at least one piece of program code, which is loaded and executed by a processor to implement the operations performed by the unmanned follow-up blowing method for the coal-fired boiler.

[0009] Fifthly, this application also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the above-described unmanned blasting methods for coal-fired boilers. Attached Figure Description

[0010] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.

[0011] To more clearly illustrate the technical solutions in the embodiments of this disclosure or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, for those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0012] Figure 1 A schematic flowchart illustrating an unmanned follow-up blowing method for a coal-fired boiler, provided as an embodiment of this application; Figure 2-1 A flowchart illustrating the confirmation of the soot blowing start command sequence in an unmanned soot blowing method for a coal-fired boiler, provided in an embodiment of this application; Figure 2-2 This is a flowchart illustrating the process of acquiring multi-source operating data in an unmanned follow-up blowing method for a coal-fired boiler, as provided in an embodiment of this application. Figure 3 This is a schematic diagram of the integrated diagnostic processing in an unmanned follow-up blowing method for a coal-fired boiler provided in an embodiment of this application; Figure 4 A schematic diagram of the safety handling and instruction sequence rearrangement in a method for unmanned follow-up blowing of a coal-fired boiler provided in an embodiment of this application; Figure 5 This application provides a schematic diagram of the structure of an unmanned follow-up blowing system for a coal-fired boiler. Figure 6 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this application. Detailed Implementation

[0013] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the protection scope of this application.

[0014] It should be noted that, in the description of this application, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. The terms "first," "second," etc., in this application are used to distinguish similar objects and are not used to describe a specific order or sequence.

[0015] In order to illustrate the technical solutions provided in the embodiments of this application, some terms involved in the embodiments of this application will be introduced below.

[0016] Sootblower: A steam purging device used to remove ash and slag buildup on the heated surfaces of boiler water-cooled walls, superheaters, and reheaters. It includes long telescopic sootblowers and semi-telescopic sootblowers. During operation, the sootblower nozzle, driven by a motor, enters the furnace via a transmission mechanism, rotating and spraying steam to purge the heated surfaces as it travels. It then returns to its original position at the end of its stroke. In other words, each operation of the sootblower involves a complete stroke: "nozzle entry—purge—nozzle retraction." Any jamming at any point during this stroke can cause localized over-blowing damage to the heated surfaces.

[0017] DCS: Distributed Control System, which is responsible for controlling the boiler soot blowing procedure and interlocking the main boiler parameters.

[0018] Absolute position encoder: A sensor that directly outputs absolute position code values ​​and does not lose position information after power failure. That is, even if the system restarts, the encoder can still directly give the current full stroke position of the sootblower, without the need for zero-return calibration. The embodiments of this application use a multi-turn absolute encoder, which is installed on the drive shaft of the sootblower drive motor, and its output code value corresponds one-to-one with the 0-100% stroke position of the sootblower.

[0019] Lift valve: A valve installed on the steam inlet passage of a single sootblower to control the on / off of steam inlet to that sootblower. It can be understood that the pressure after the lift valve is a direct indicator of the actual steam inlet pressure obtained by that sootblower. If the lift valve does not operate properly, even if the main pipe pressure is normal, the sootblower will be in a "false commissioning" state.

[0020] Fouling Index: A quantitative indicator that characterizes the degree of ash and slag formation on a heated surface, calculated based on the heat transfer state parameters of the heated surface. The higher the fouling index, the more priority the heated surface needs to be given to soot blowing.

[0021] Flow characteristics: These are quantitative features characterizing the steam flow status of a single sootblower, including the pressure ratio between the pressure after the booster valve and the main pipe pressure, as well as the flow increment of the total steam flow in the main pipe relative to the rated flow of a single unit. In other words, flow characteristics jointly depict the steam health of a single sootblower from two dimensions: "whether the pressure is on" and "whether the flow is sufficient."

[0022] Active defense: After a fault alarm is triggered, the automatic execution of safety measures such as interrupting the soot blowing process, issuing a gun withdrawal command, and locking the faulty equipment will not require manual on-site intervention.

[0023] In related technologies, sootblowing control systems typically only collect limit switch signals at the sootblower's forward and backward positions. Faults are determined based on single-type detection signals such as motor overcurrent and main pipe pressure. Total pressure and flow measurement points are only configured on the sootblowing steam main pipe, and visual faults such as steam leakage rely on on-site monitoring and inspection by operators throughout the process. At the data perception level, traditional solutions cannot obtain the continuous position and operating rate of the entire stroke between the two limit switches; jamming and slippage in the middle of the stroke are completely invisible. Each sootblower lacks independent measurement points, leading to confusion between different operating conditions such as incomplete opening of the lifting valve, blockage of the steam inlet pipe, and fluctuations in main pipe pressure. At the diagnostic decision-making level, single criteria lack cross-validation mechanisms. Single sensor anomalies such as encoder drift and signal interference can trigger false alarms, causing frequent interruptions to the sootblowing process and requiring manual verification. Missed alarms, on the other hand, cause jammed nozzles to remain stuck in a localized area of ​​the furnace for extended periods, resulting in thinning of the heated surface tube wall and deformation or even burnt-out of the nozzle due to high-temperature flue gas. This approach means that fault handling and soot blowing monitoring can only be done manually, which not only increases the labor intensity of operators and poses safety hazards such as missed detections, delays, and untimely handling, but also makes it difficult to achieve unmanned operation of the unit.

[0024] To address the aforementioned technical issues, this application designs an unmanned sootblowing method for coal-fired boilers. This method confirms the sootblowing start-up command sequence and simultaneously acquires four types of multi-source operational data: stroke position, steam condition, electrical operation, and visual images. Continuous position detection throughout the entire stroke fills the monitoring blind spot between the limit switches at both ends. Through multi-source operational feature extraction, feature fusion, and cross-validation classification, position deviation, duration deviation, current fluctuation characteristics, and rate change characteristics are cross-verified to eliminate false positives and false negatives from a single detection source. Furthermore, pressure ratio and flow increment are used as dual criteria to accurately identify steam inlet faults in a single sootblower. In response to a fault alarm, the method automatically interrupts program control, retracts the nozzle, locks the faulty sootblower, and re-arranges the sootblowing start-up command sequence according to the latest dirt index, eliminating the need for manual on-site sootblowing. The application scenarios of the technical solution provided in this application are described below.

[0025] The technical solutions provided in this application are mainly applied to the field of automatic control and safety protection of coal-fired power plant boilers, especially in scenarios with extremely high requirements for safety monitoring, automatic fault handling, and unmanned operation of soot blowing equipment. Specifically, they mainly include the following application scenarios: Scenario 1: Unmanned transformation of the soot blowing system of a large coal-fired power unit.

[0026] Taking a 660MW coal-fired boiler as an example, this boiler is equipped with 44 long telescopic sootblowers and 12 semi-telescopic sootblowers, operating in a mode where two sootblowers work simultaneously. A complete sootblowing cycle lasts approximately 5 hours. Using the method provided in this application, four types of monitoring—encoder, downstream pressure transmitter, industrial camera, and motor current acquisition—are simultaneously activated, providing multi-dimensional real-time monitoring of the entire sootblower stroke. A 5-hour sootblowing cycle requires no on-site monitoring by operators; only remote confirmation of alarm information is needed.

[0027] Scenario 2: Soot blowing monitoring under deep peak shaving conditions.

[0028] When the unit participates in deep peak shaving, the load fluctuates frequently, and the fouling state of the heating surfaces changes rapidly. Fixed-sequence soot blowing is prone to local over-blowing or under-blowing. Applying the method of the embodiments of this application, the fouling index is dynamically calculated based on the latest inlet and outlet flue gas temperature, wall temperature and flue gas differential pressure of each heating surface, and the soot blowing priority is rearranged to achieve on-demand follow-up soot blowing. At the same time, the entire stroke is monitored to prevent the heating surface from being damaged by jamming and slippage under peak shaving conditions.

[0029] Scenario 3: Low-cost intelligent upgrade of old soot blowing systems.

[0030] There is no need to dismantle and rebuild the entire soot blowing system on a large scale. Only by installing an absolute position encoder, a pressure transmitter after the lifting valve and a field camera device on each soot blower, the method of this application can superimpose multi-source fusion diagnosis and active defense capabilities on the basis of existing DCS program control, effectively improving the safety monitoring level of soot blowing equipment in old production lines.

[0031] Scenario 4: Analysis and implementation of meeting the requirements for secondary power safety protection.

[0032] To meet the compliance requirement that the production control area and the production management area must be physically isolated, the method in this application embodiment can deploy the fusion diagnostic unit in the three security zones, and realize cross-zone linkage of alarm signals through forward unidirectional transparent transmission and reverse hard contact back transmission, so as to realize the closed loop of intelligent diagnosis and field control loop without crossing the network security boundary.

[0033] After introducing the implementation environment and application scenarios of the embodiments of this application, the technical solutions provided by the embodiments of this application are described below. (See also...) Figure 1 Taking a computer device (such as an industrial control computer or server) that is the execution subject of an unmanned follow-up blowing system for a coal-fired boiler as an example, the unmanned follow-up blowing method for a coal-fired boiler includes the following steps.

[0034] Step S101: Confirm the soot blowing start command sequence and obtain the multi-source operation data of the soot blower.

[0035] The soot blowing start-up command sequence is used to indicate the start-up order and start-up time of the soot blowers corresponding to each heating surface of the boiler. In other words, the soot blowing start-up command sequence not only specifies "who blows first and who blows last", but also specifies "when to start and what parameters to use" for each soot blower, serving as the execution blueprint for this round of soot blowing.

[0036] Multi-source operational data includes data on the sootblower's stroke position, steam conditions, electrical operation, and visual images. Understandably, these four types of data independently characterize the sootblower's operating status from four physical dimensions: "where the nozzle is moving," "whether steam is flowing," "how tired the motor is," and "whether any abnormalities are visible on-site." Failure in any one dimension is compensated for by the others.

[0037] For example, stroke position data may include continuous position signals of the gun body throughout its entire stroke and the real-time travel rate calculated accordingly; steam condition data may include the steam inlet pressure, main pipe pressure, and total steam flow of a single sootblower; electrical operation data may include the operating current of the drive motor; and visual image data may include image data covering the sootblower protective cover, traveling parts, and valve stem sealing parts.

[0038] Specifically, the heat exchange status parameters of each heating surface of the boiler are read, the fouling index of each heating surface is calculated and the soot blowing priority is sorted. Based on this, the operation group, start-up sequence and steam parameters of each soot blower are arranged to generate a soot blowing start command sequence. Data acquisition is started synchronously with the execution of the soot blowing start command sequence. Four types of multi-source operation data are acquired in real time through the encoder installed on the drive motor drive shaft, the pressure transmitter after the lifting valve, the pressure and flow transmitter on the main pipe, the motor current acquisition module and the on-site camera device.

[0039] Step S102: Based on multi-source operating data, perform fusion diagnostic processing to obtain fault alarm signals.

[0040] Fusion diagnostic processing includes multi-source operational feature extraction, feature fusion, and cross-validation classification. In other words, fusion diagnostic processing does not allow the four types of data to work independently and give separate fault conclusions. Instead, it first extracts features from the four types of data separately, then aligns and fuses them according to the blowing process nodes, and finally completes fault classification based on the fused features through mutual verification of multiple criteria. The judgment result of any one criterion must be verified by the other criteria before it can be confirmed.

[0041] The fault alarm signals include transmission fault signals, steam inlet fault signals, and steam leakage fault signals for the sootblower. These three types of fault signals correspond to the three major failure modes of the sootblower—mechanical abnormalities in the traveling mechanism, abnormalities in the steam inlet passage, and abnormalities in the seal leakage—covering the main risk sources in sootblower operation.

[0042] Specifically, differential calculations are performed on stroke position data to obtain rate change characteristics, flow state calculations are performed on steam operating condition data to obtain flow characteristics, current fluctuation statistics are performed on electrical operation data to obtain current fluctuation characteristics, and image recognition is performed on visual image data to obtain visual position characteristics and steam leakage image characteristics. The encoder position is compared with the visual position to obtain the position deviation, and the actual running time is compared with the standard stroke time to obtain the duration deviation. These are then aligned according to the soot blowing stroke nodes to form a fusion feature. Three types of fault alarm signals are output after cross-validation and classification, and self-checking correction is performed on continuous position signals.

[0043] Step S103: Respond to the fault alarm signal, perform safety procedures on the faulty sootblower, and reconfirm the sootblowing start command sequence.

[0044] The faulty sootblower is used to indicate the sootblower that needs to be taken out of operation for the current cycle, corresponding to the fault alarm signal. Understandably, the core of safe handling is "clean withdrawal, thorough locking, and orderly return to the original sequence"—that is, to safely remove the faulty equipment from operation, while ensuring that the on-demand sootblowing of the remaining heated surfaces is not halted by the faulty equipment.

[0045] Specifically, in response to a fault alarm signal, the current sootblowing process is interrupted, a retraction command is issued to the faulty sootblower to drive it back to its original position, and its restart permission is locked; full-dimensional operation data at the time of the fault is recorded for traceability; the faulty sootblower is removed from the sootblowing start command sequence, and the start order and start time of the remaining sootblowers are reconfirmed according to the latest dirt index of each heating surface of the boiler, and an updated sootblowing start command sequence is generated.

[0046] This embodiment replaces a single criterion with the fusion diagnosis of four types of data, and realizes fully unmanned operation of automatic strategy generation, automatic equipment execution, automatic process monitoring and automatic fault handling through automatic fault handling and dynamic rearrangement of instruction sequences. Its beneficial effects are: overcoming the defects of frequent false alarms and missed alarms due to single criteria, no manual on-site operation is required throughout the fault handling process, effectively reducing the risk of gun barrels being burned by heat in the furnace and tube bursting due to blown-out heating surfaces, and realizing unmanned operation of the entire soot blowing process.

[0047] It should be noted that the above steps S101-S103 are a simplified description of the unmanned follow-up blowing method for coal-fired boilers provided in the embodiments of this application. The method provided in the embodiments of this application will be described in more detail below with some examples.

[0048] In some embodiments, see Figure 2-1 The confirmed soot blowing start command sequence includes: Step S201: Based on the heat exchange status parameters of multiple heating surfaces of the boiler, confirm the fouling index of multiple heating surfaces.

[0049] The heat exchange status parameters include inlet and outlet flue gas temperatures, wall temperatures, and flue gas differential pressure data. These parameters reflect the degree of ash accumulation on the heated surface from the perspective of heat exchange effectiveness—the more severe the ash accumulation, the worse the heat absorption capacity of the heated surface, and the greater the deviation between the inlet and outlet flue gas temperature difference and the wall temperature. It is understandable that, compared to directly measuring the ash accumulation thickness, the heat exchange status parameters can be directly obtained from existing measuring points on the unit, without the need for additional sensors.

[0050] Specifically, for water-cooled walls, the fouling index is calculated based on the degree to which the outlet flue gas temperature deviates from the design value and the extent to which the wall temperature exceeds the set value. For superheaters and reheaters, the fouling index is calculated based on the deviations in their inlet and outlet flue gas temperatures and wall temperatures. For convective heating surfaces such as air preheaters, the fouling index is calculated based on the increase in their flue gas differential pressure. After normalizing each parameter, a weighted sum is obtained to obtain the fouling index for the corresponding heating surface. For example, if a superheater's outlet flue gas temperature is 18°C ​​higher than the design value and its wall temperature exceeds the limit by 6°C, its fouling index after normalization and weighting is 0.82, significantly higher than the adjacent water-cooled wall region with a fouling index of 0.35.

[0051] In addition, the enthalpy changes on the steam side of each heated surface can be combined with the normalized weighted calculation to further reflect the attenuation of the actual heat absorption capacity of the heated surface. The specific calculation formula for the fouling index is not limited in the embodiments of this application.

[0052] Step S202: Determine the priority order for soot blowing based on the dirt index.

[0053] The soot blowing priority ranking is used to characterize the order in which soot blowers corresponding to each heated surface receive soot blowing orders. In other words, the higher the dirtiness index of the heated surface, the earlier its corresponding soot blower will be put into operation, realizing on-demand soot blowing by "blowing where it's dirtiest first".

[0054] Specifically, each sootblower is sorted from highest to lowest according to the fouling index of its corresponding heated surface. For example, the sootblower corresponding to the superheater with a fouling index of 0.82 is placed at the top of the sequence, while the sootblower corresponding to the water-cooled wall with a fouling index of 0.35 is placed at the bottom of the sequence.

[0055] Step S203: Soot blowing priority sorting, arrange the operation group, start-up sequence and steam parameters of multiple soot blowers, and generate soot blowing start command sequence.

[0056] The operation grouping is used to determine the combination of sootblowers to be put into operation simultaneously; the start-up sequence is used to determine the start-up time of each group; and the steam parameters are used to determine the appropriate sootblowing steam pressure and flow rate for each group. Understandably, the supply capacity of the steam header must be taken into account during the arrangement to avoid a sudden drop in header pressure caused by the simultaneous operation of sootblowers within the group.

[0057] Specifically, in the operation mode of two soot blowers working simultaneously, adjacent soot blowers are grouped into the same group, the start time of each group is arranged in sequence, and the steam parameters of the corresponding group are determined according to the type of heated surface, thus generating a complete soot blowing start command sequence for this round.

[0058] In some embodiments, see Figure 2-2 The acquisition of multi-source operating data of the sootblower includes: Step S204: Perform continuous position detection on the gun body of the sootblower throughout its entire stroke to obtain stroke position data.

[0059] The travel position data includes continuous position signals throughout the gun's entire travel range and the real-time travel rate calculated accordingly. It is understandable that the essential difference between continuous position detection throughout the entire travel range and traditional two-end travel switch detection is that the former outputs a continuous position sequence within 0–100% of the travel range, reflecting the motion state at any point in the travel range, while the latter only has signals at the two endpoints of the forward and backward positions, making the intermediate process a complete black box for the system.

[0060] Specifically, a 16-bit multi-turn absolute encoder installed on the drive motor shaft outputs continuous position signals corresponding to the 0–100% stroke of the sootblower, and the real-time travel speed is calculated differentially according to the sampling period. For example, a long telescopic sootblower with a stroke of 10m and a rated speed of 2m / min has an encoder sampling period of 100ms, outputting 10 position code values ​​per second, and obtaining a continuous speed curve after differential processing.

[0061] Step S205: Detect pressure and flow rate on the downstream side of the steam inlet passage of a single sootblower to obtain steam operating condition data.

[0062] The steam operating condition data includes the inlet steam pressure, main pipe pressure, and total steam flow of a single sootblower. It is understandable that moving the pressure measurement point upstream to after the lift valve of each sootblower is key to achieving "single-unit" steam inlet diagnosis—the main pipe measurement point can only reflect the overall supply and cannot distinguish which specific sootblower has an abnormal steam inlet.

[0063] Specifically, a high-precision pressure transmitter with a range of 0–3 MPa is installed after the lifting valve of each sootblower to detect the single-unit steam inlet pressure; a main pipe pressure transmitter and a steam flow transmitter are installed on the sootblowing steam main pipe to collect the main pipe pressure and total steam flow, respectively.

[0064] Step S206: Detect the current of the drive motor of the sootblower to obtain electrical operation data.

[0065] The electrical operation data includes the operating current of the drive motor. The motor current directly reflects the transmission load—when jammed, the load increases and the current rises; when slipping, the load decreases sharply and the current drops abnormally. Specifically, the motor current acquisition module is connected to the drive motor circuit of the sootblower to collect the operating current in real time; the motor's rated current is 3A.

[0066] Step S207: Acquire images of the soot blower to obtain visual image data.

[0067] The visual image data includes image data covering the sootblower protective cover, moving parts, and valve stem sealing area. It is understood that visual images are the only "non-contact, appearance-based" data among the four types, capable of independently analyzing location and detecting appearance faults such as valve stem leakage that cannot be perceived by the other three types of data.

[0068] Specifically, industrial cameras are installed on the sootblower platforms on each floor of the boiler, with each camera covering one sootblower. The video signal is transmitted through an industrial optical transceiver, and the video image of the corresponding sootblower is automatically located and processed according to the current operating group.

[0069] This embodiment achieves the technical effect of on-demand follow-up soot blowing and full-stroke multi-source monitoring by calculating the dirt index, sorting priorities, arranging grouping timing to generate instruction sequences, and simultaneously constructing a four-dimensional perception system of position, steam, electricity, and vision. Its beneficial effects are: filling the monitoring blind spot of the entire stroke between the limit switches at both ends, realizing the accurate location and diagnosis of steam inlet faults of a single sootblower, providing a multi-dimensional and mutually verifiable basis for integrated diagnosis, and ensuring that the risk status changes of the sootblower can be captured in a timely manner.

[0070] In some embodiments, see Figure 3 The multi-source running feature extraction in step S102 includes: Step S301: Perform differential calculation processing on the travel position data of the multi-source operation data to obtain the rate change characteristics.

[0071] The rate change characteristics are used to quantify the evolution of the blowing gun's travel rate with the stroke. In other words, the rate change characteristics are the first window for judging the health of the transmission mechanism—the normal travel rate is smooth and stable, it continuously decreases when jammed, and drops sharply to zero when slipping.

[0072] Specifically, the continuous position signal of the encoder is differentially divided according to the sampling period to obtain the instantaneous travel rate during the forward and backward processes, generating a rate change curve of 0–100% of the full stroke.

[0073] Step S302: Perform steam flow state calculation processing on the steam operating condition data of multi-source operation data to obtain flow characteristics.

[0074] The flow characteristics include pressure ratio characteristics and flow increment characteristics. The pressure ratio characteristic is used to characterize whether the pressure in the steam inlet passage of the sootblower is "open"; the flow increment characteristic is used to characterize whether the flow is "sufficient". It can be understood that the two constitute a dual criterion - fluctuations in the main pipe pressure will cause synchronous fluctuations in the downstream pressure, but at this time the flow increment characteristic is normal, and the dual criterion can eliminate such interference.

[0075] Specifically, the ratio P1 / P0 of the pressure P1 after the booster valve to the pressure P0 of the main pipe is calculated as a pressure ratio characteristic; the change of the total steam flow rate Q of the main pipe relative to the start-up of the sootblower is monitored to obtain the flow increment characteristic of the corresponding single rated flow rate.

[0076] Step S303: Perform current fluctuation statistical processing on the electrical operation data of multi-source operation data to obtain current fluctuation characteristics.

[0077] Specifically, the deviation and frequency of the motor's operating current from the rated range of 3A are statistically analyzed. For example, when the motor is stuck, the current rises from 3A to 4.5A and remains there, while when it slips, the current drops abnormally.

[0078] Step S304: Perform image recognition processing on the visual image data of the multi-source operation data to obtain visual location features and steam leakage image features.

[0079] Visual position features are used to provide a second path of stroke detection results independent of the encoder; steam leakage image features are used to identify steam leaks in the valve stem and sealing areas. It is understandable that steam leakage is a visual fault, and since position, steam, and electrical data are all undetectable, the visual dimension is the sole criterion; therefore, the independent value of the visual channel is irreplaceable.

[0080] Specifically, the relative positions of the equally spaced reflective strips on the protective cover and the indicator arrows on the traveling parts are identified in the image. The visual position percentage corresponding to 0–100% of the full stroke is calculated by combining the total number of reflective strips and the spacing parameters. At the same time, the white steam leakage characteristics of the valve stem and sealing parts are identified based on the deep learning image recognition model, and the normal purging steam on the furnace side and the leakage steam on the valve body side are distinguished, and the leakage level is output.

[0081] In this embodiment, feature fusion includes: Step S305: Compare the continuous positions represented by the travel position data with the visual positions represented by the visual position features in real time to obtain the position deviation.

[0082] It is understandable that encoder position and visual position are two completely independent physical principles (electrical angle measurement and optical recognition) for travel detection. Under normal circumstances, the deviation between the two is minimal; when the deviation exceeds the limit, at least one of them is abnormal, and other features need to be combined to determine which one is abnormal.

[0083] Specifically, the position deviation is obtained by subtracting the encoder position from the visual position at the same moment. For example, when the encoder indicates 68% of the travel and the visual position indicates 53% of the travel, the position deviation is 15%.

[0084] Step S306: Compare the actual running time of a single forward and backward movement of the sootblower with the standard stroke time to obtain the time deviation.

[0085] Specifically, the standard travel time for a single forward movement is 5 minutes. If the actual running time exceeds 6 minutes, the time deviation exceeds the limit. The time deviation provides a holistic transmission verification method that does not rely on intermediate rate curves.

[0086] Step S307: Align the position deviation, duration deviation, current fluctuation characteristics, and rate change characteristics according to the soot blowing stroke nodes to obtain the joint analysis characteristics.

[0087] The blow-off stroke node is used to indicate the feature alignment reference based on the stroke percentage. It is understandable that the sampling frequencies and physical dimensions of the four types of features are different. Only by aligning to the same stroke node can we achieve mutual verification of multiple criteria at the same time and position. This is the data basis for the establishment of cross-validation classification.

[0088] Step S308: Based on the position deviation, duration deviation, and joint analysis features, a fusion feature is constructed.

[0089] Specifically, at the same travel node, the four types of features—position, duration, current, and rate—are aligned in time and space to form multi-dimensional joint analysis features, which in turn constitute fused features.

[0090] In this embodiment, cross-validation classification includes: Step S309: When the position deviation exceeds the set deviation threshold and continues for a set duration, accompanied by abnormal rate change characteristics or current fluctuation characteristics, determine the source of the abnormality and correct the determination result of a single detection source.

[0091] Specifically, when the deviation between the encoder position and the visual position exceeds 10% and lasts for more than 5 seconds, accompanied by abnormal speed or current, a jamming or slippage fault is confirmed. When the deviation exceeds the limit but the speed and current are normal, it is determined to be an encoder malfunction rather than a sootblower jamming. For example, if the encoder code value jumps due to interference, causing a sudden increase in position deviation, but the visual position, speed, and current are all normal, then only encoder maintenance is prompted without interrupting the sootblowing process, thus correcting the misjudgment from a single detection source.

[0092] Step S310: When the soot blower is in position or out position and the travel speed is zero, record the current continuous position code value and store it in the corresponding historical record. Take the average value of the records of the nearest set number of times as the reference code value and perform self-verification correction on the continuous position signal.

[0093] Specifically, when the sootblower is in its final position or out of position and its travel speed is zero, the current continuous position code value is recorded and stored in the corresponding historical record. The historical record queue has a length of 5, and the average of the last 5 records is used as the reference code value for 100% travel and 0% travel, respectively, to perform self-verification and correction on the continuous position signal. It can be understood that the position is independently verified by the limit switch, and the recorded code value at this time is the "true value anchor point". The cumulative error of the encoder over long-term operation is eliminated by the moving average, and no manual periodic calibration is required.

[0094] Step S311: When the speed change characteristics indicate that the travel speed is continuously lower than the set speed threshold or drops sharply to zero and continues for a set duration, or continuously higher than the set speed limit, output a transmission fault signal.

[0095] Specifically, when the travel speed is consistently below 0.5 m / min for more than 5 seconds, it is determined to be a jamming fault; when the motor operation feedback is normal but the speed suddenly drops to 0 for 3 seconds, it is determined to be a slippage fault; when the speed is consistently higher than the set speed limit, it is determined to be a slippage of the traveling wheels, and a transmission fault signal is output.

[0096] Step S312: When the pressure ratio characteristic is lower than the set pressure threshold and the flow increment characteristic does not reach the set flow increment of the corresponding single soot blower, output a steam inlet fault signal.

[0097] Specifically, under normal operating conditions, P1 / P0 should be no less than 0.85, and the flow rate in the main pipe should show an increase corresponding to the rated flow rate of a single unit. If P1 / P0 < 0.7 and the flow rate increase is insufficient, it is determined that the sootblower's lifting valve is not fully open or the steam inlet pipe is blocked, posing a risk of steam interruption and dry burning, and an inlet fault signal is output. Compared to a single pressure criterion, dual criteria can eliminate the interference of main pipe pressure fluctuations.

[0098] Step S313: When the steam leakage image features indicate that steam leakage has been identified in the valve stem and sealing parts, output a steam leakage fault signal.

[0099] Specifically, based on a deep learning image recognition model, the white vapor leakage characteristics of the valve stem and sealing parts are identified, and a steam leakage fault signal and fault level are output.

[0100] This embodiment achieves a multi-dimensional fusion diagnosis of rate, position, steam, current, duration, and vision by extracting rate, current, and visual features stepwise, aligning them according to stroke nodes to form fusion features, and outputting three types of fault signals through cross-validation classification and self-verification correction. Its beneficial effects are: effectively eliminating false alarms and missed alarms caused by single sensor failures such as encoder drift and signal interference; dual-criteria current diagnosis distinguishes between different operating conditions such as riser valve failure and main pipe pressure fluctuation, effectively improving the confidence of fault identification in unattended scenarios.

[0101] In some embodiments, see Figure 4 Step S103 specifically includes: Step S401: Determine the sootblower corresponding to the fault alarm signal to obtain the faulty sootblower.

[0102] Specifically, based on the equipment identifier carried by the fault alarm signal, the corresponding sootblower and its real-time grouping information are located.

[0103] Step S402: In response to the fault alarm signal, interrupt the current sootblowing program control process, issue a retraction command to the faulty sootblower to drive it back to its original position, and lock the restart permission of the faulty sootblower.

[0104] The "retract lance" command is used to drive the sootblowing lance out of the furnace and back to its original position; the "reactivation permission lockout" is used to prevent faulty equipment from being called back in subsequent sequences. Understandably, retracting the lance takes precedence over all programmable logic—every minute the lance barrel remains in the furnace increases the risk of it being scorched and deformed, and the heated surfaces being continuously damaged by blowing.

[0105] Specifically, the fault alarm signal triggers the active defense interlock, forcibly interrupting the current soot blowing program control process; the gun retraction command takes priority over the normal program control command and is responded to, directly driving the motor to move the soot blowing gun back to its original position. After retraction, the restart permission of the soot blower is automatically locked, and at the same time, an audible and visual alarm is popped up to notify the operators.

[0106] Step S403: Record the travel position data, steam condition data, electrical operation data, and visual image data at the time of the fault.

[0107] Understandably, comprehensive data at the moment of failure serves as both the basis for post-failure investigation and responsibility determination, and as material for iterative optimization of diagnostic criteria parameters. Specifically, comprehensive operational data at the moment of failure, including location, pressure, current, and video frames, is recorded and archived.

[0108] Step S404: Remove the faulty sootblower from the sootblowing start command sequence, and reconfirm the start order and start time of the remaining sootblowers based on the latest fouling index of multiple heating surfaces of the boiler, and generate an updated sootblowing start command sequence.

[0109] Specifically, after a faulty sootblower is removed and locked, the remaining sootblowers are dynamically rearranged based on the latest dirtiness index of the other heated surfaces. For example, if a superheater sootblower is removed due to transmission jamming, the high-dirtiness index water-cooled wall sootblower that was originally scheduled to follow it is started in advance, ensuring that the corresponding heated surface of that superheater area still receives the highest priority in the next round of sootblowing, and the faulty equipment does not affect the on-demand sootblowing of the other heated surfaces.

[0110] This embodiment achieves the technical effect of automatic and safe withdrawal of faulty equipment and unaffected soot blowing of other heated surfaces by means of interrupted program control, gun withdrawal lockout, full-dimensional data recording and dynamic rearrangement according to the latest dirt index after the faulty equipment is removed. Its beneficial effect is that no manual on-site operation is required throughout the fault handling process, and the handling delay is reduced from tens of minutes in the traditional manual on-site process to about 40 seconds, truly realizing unmanned blowing.

[0111] Figure 5 This is a schematic diagram of the structure of an unmanned follow-up blowing system for a coal-fired boiler provided in an embodiment of this application. See also... Figure 5 The unmanned follow-up blowing system 500 for coal-fired boilers includes: a field sensing unit 501, a DCS control unit 502, a fusion diagnostic unit 503, an unmanned follow-up blowing control unit 504, and a local execution unit 505.

[0112] The field sensing unit 501, configured for the sootblower, is used to acquire multi-source operating data of the sootblower. The field sensing unit 501 is the system's data input terminal, and its function is to fill the full-stroke blind zone of traditional two-end limit switch detection, providing multi-dimensional criteria for fusion diagnosis. Specifically, the field sensing unit 501 includes an absolute position encoder, a downstream pressure transmitter, a main pipe pressure transmitter, a steam flow transmitter, a motor current acquisition module, a field camera device, and a visual identification component installed on the sootblower body: the absolute position encoder is a 16-bit multi-turn absolute encoder, installed at the drive shaft of the sootblower drive motor, used to output continuous position signals of the gun body throughout its entire stroke and calculate the real-time travel speed, constituting stroke position data; the downstream pressure transmitter is installed downstream of the lifting valve of a single sootblower, with a range of 0–3MPa; the main pipe pressure transmitter and the steam flow transmitter are installed on the sootblowing steam main pipe, and the three collect the single-unit inlet steam pressure, main pipe pressure, and total steam flow respectively, constituting steam operating condition data; the motor current acquisition module is connected to the drive motor circuit to collect the operating current, constituting electrical operation data; the field camera device is an industrial camera, each covering one sootblower, and the video signal is transmitted through an industrial optical transceiver, collecting image data covering the protective cover, moving parts, and valve stem sealing parts, constituting visual image data. For example, when a soot blower needs to be diagnosed, the field sensing unit 501 automatically locates the corresponding camera image and measurement point channel according to its operation group, and outputs the full-dimensional operation data of the blower.

[0113] The DCS control unit 502 is used to implement safety interlocks between sootblowing program control and main parameters such as boiler MFT (Main Parameter Failure), and to execute safety interlocks in response to fault alarm signals. The DCS control unit 502 is the central hub for the system's safety interlocks, responsible for arbitrating commands between programmed control and active defense. Specifically, the DCS control unit 502 adds signal input channels for encoder position, downstream valve pressure, and cross-zone alarm; when responding to a fault alarm signal, it forcibly interrupts the current sootblowing program control process and issues a gun retraction command to the sootblower electrical control cabinet.

[0114] The fusion diagnostic unit 503 is used to perform fusion diagnostic processing based on multi-source operating data to obtain fault alarm signals. The fusion diagnostic unit 503 is the core of the system's diagnostic decision-making. Specifically, the fusion diagnostic unit 503 includes a full-stroke state analysis module, a steam condition diagnostic module, a visual intelligent recognition module, and a multi-source fusion diagnostic module: It executes the differential calculation in step S301 of the full-stroke state analysis module and the zero-position / full-position self-verification in step S310; the steam condition diagnostic module, when a single sootblower is in the operating command state, synchronously collects valve downstream pressure, main pipe pressure, and steam flow detection data, and executes the flow characteristic calculation in step S302; the visual intelligent recognition module executes the visual position analysis and steam leakage identification in step S304; the multi-source fusion diagnostic module executes the feature fusion and cross-validation classification in steps S305–S309 and S311–S313, and outputs transmission fault signals, steam inlet fault signals, and steam leakage fault signals.

[0115] The unmanned soot blowing control unit 504 is used to confirm the soot blowing start command sequence and respond to fault alarm signals to reconfirm the soot blowing start command sequence. Specifically, the unmanned soot blowing control unit 504 includes a heated surface cleanliness assessment subunit and a soot blowing strategy arrangement subunit: the heated surface cleanliness assessment subunit performs the dirt index calculation in step S201; the soot blowing strategy arrangement subunit performs priority sorting and grouping timing arrangement in steps S202–S203, and fault elimination and dynamic rearrangement in step S404, forming a dual closed-loop unmanned control for strategy optimization and safety protection. In one possible implementation, the unmanned sootblowing control unit 504 further includes an operation monitoring subunit. The operation monitoring subunit is connected to the DCS control unit 502 via an industrial Ethernet network. It is used to collect all operation data such as the operation grouping, stroke status, steam parameters, and motor current of all sootblowers, and to build a centralized monitoring interface for the sootblowing system. It displays the stroke percentage, real-time travel speed, steam parameters, current value, and fault information of each sootblower in real time, and supports historical data storage, fault tracing, and alarm push. The operation monitoring subunit also serves as a data outlet, transmitting unidirectional synchronous operation data to the three power production safety zones where the fusion diagnostic unit 503 is located through the cross-safety zone interaction unit.

[0116] The local execution unit 505 includes a sootblower electrical control cabinet, used to respond to fault alarm signals and perform safety procedures on the faulty sootblower. Specifically, the sootblower electrical control cabinet retains the original manual and programmable control loops, and is bidirectionally connected to the DCS control unit 502. It is used to receive control commands, drive the sootblower to complete forward, backward, and rotational movements, and provide feedback on the equipment operating status and motor current signals. The sootblower electrical control cabinet also includes an active defense priority control loop, which is used to execute the gun retraction command before the normal programmable command, directly driving the motor to move the sootblower gun back to its original position, and locking the restart permission of the faulty sootblower. For example, when the programmable command and the gun retraction command arrive simultaneously, the active defense priority control loop unconditionally executes the gun retraction first, ensuring that the gun barrel leaves the furnace in the shortest possible time.

[0117] In one possible implementation, the system further includes a visual identification component disposed on the body of the sootblower. The visual identification component provides a geometric reference for the visual intelligent recognition module, enabling the visual channel to resolve the 0–100% full-stroke position. Specifically, the visual identification component includes: multiple reflective strips disposed on the outer surface of the sootblower's protective cover, evenly spaced along the stroke direction, for example, 200 strips per unit with a spacing of 50mm; and indicator arrows disposed on the side of the sootblower's traveling carriage, with the arrow tips close to and pointing towards the reflective strip surface, moving synchronously with the sootblower. The visual intelligent recognition module identifies the relative position of the indicator arrows and reflective strips, and, combining the total number of reflective strips and spacing parameters, calculates the visual position percentage corresponding to the 0–100% full stroke, forming a second stroke detection independent of the encoder.

[0118] In one possible implementation, the integrated diagnostic unit 503 is deployed in the third safety zone of power production, while the field sensing unit 501 and the unmanned follow-blowing control unit 504 are deployed in the first safety zone of power production. The system also includes a cross-safety zone interaction unit. The cross-safety zone interaction unit is used to achieve the linkage between intelligent diagnosis and field control loops while meeting the safety protection requirements of the power secondary system. Specifically, the cross-safety zone interaction unit includes a forward unidirectional transparent transmission module and a reverse hard contact feedback module: the forward unidirectional transparent transmission module adopts a forward physical isolation device, which only allows the first safety zone to transmit the sootblower operating status, grouping information and analog data collected by the unmanned follow-blowing control unit 504 to the third safety zone in a unidirectional manner, and prohibits the third safety zone from transmitting network control commands to the first safety zone in the reverse direction, ensuring that the intelligent analysis algorithm does not directly issue control commands across the network; the reverse hard contact feedback module includes a relay output unit, which is used to convert the fault alarm signals such as transmission, steam inlet and steam leakage output by the third safety zone into passive dry contact switch signals, and physically connects them to the DI input channel of the DCS control unit 502 in the first safety zone through a control cable. For example, after a transmission fault signal of a sootblower is determined in the three safety zones, it is converted into a dry contact closing signal by a relay, which directly drives the DI channel of the DCS to trigger the active defense interlock without crossing the network security boundary.

[0119] It should be noted that the unmanned follow-up blowing system for coal-fired boilers provided in the above embodiments is only illustrated by the division of the aforementioned units and modules. In practical applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the equipment can be divided into different functional modules to complete all or part of the functions described above. Furthermore, the above system embodiments and the unmanned follow-up blowing method embodiments for coal-fired boilers belong to the same concept, and their specific implementation process is detailed in the method embodiments, which will not be repeated here.

[0120] To more intuitively illustrate the overall operation flow of the method and system provided in the embodiments of this application, the following is combined with... Figure 5 The system structure shown is illustrated using a complete unmanned follow-up blowing process of a 660MW unit as an example.

[0121] The boiler is equipped with 44 long telescopic sootblowers and 12 semi-telescopic sootblowers, with a stroke of 10m, a rated speed of 2m / min, and a standard forward travel time of 5 minutes. After the system is put into unattended sootblowing mode: the unattended sootblowing control unit 504 calculates and sorts the fouling index based on the flue gas temperature, wall temperature, differential pressure, and enthalpy of each heating surface, and generates a sequence of sootblowing start commands for two sootblowing units to work simultaneously; the DCS control unit 502 starts the sootblowers in sequence according to the time order, and four types of monitoring are started synchronously with the commands; the operation monitoring subunit of the unattended sootblowing control unit 504 synchronously displays the stroke percentage, travel speed, steam parameters, current value, and fault information of each sootblower in real time on the centralized monitoring interface.

[0122] At the 37-minute mark, a long telescopic sootblower reached 62% of its travel distance. The encoder differential calculation showed that the travel speed continuously decreased from 2 m / min to 0.3 m / min for more than 5 seconds, meeting the jamming criterion of a speed consistently below 0.5 m / min. Simultaneously, the encoder position deviated from the visual position by 15% for 6 seconds, and the motor current increased from 3A to 4.5A for 3 seconds. The position deviation, speed change, and current fluctuation corroborated each other at the 62% travel point, cross-validating the issue as transmission jamming rather than an encoder fault, and a transmission fault signal was output. The alarm signal is converted into a passive dry contact signal by the relay output unit of the reverse hard contact feedback module and fed back to the DI input channel of the DCS control unit 502 in safety zone 1. The DCS control unit 502 forcibly interrupts the sootblowing program and issues a retraction command. The active defense priority control loop of the local execution unit 505 drives the sootblowing gun to return to its original position and locks the equipment. The system records all-dimensional data at the moment of the fault, removes the sootblower from the sequence, rearranges the timing of the remaining sootblowers according to the latest dirt index, and continues to perform the current round of sootblowing. The entire process from fault identification to handling takes about 40 seconds and does not require the presence of operating personnel.

[0123] In contrast, if only the position deviation exceeds the limit while the speed and current are normal, it is determined to be an encoder malfunction. Only an encoder maintenance prompt is given without interrupting the sootblowing process. When the sootblower next reaches its designated position, the reference code value is updated using the average of the last five position code values ​​to complete self-verification, thus avoiding false alarms and interruptions caused by encoder drift. For example, if a sootblower is put into operation and P1 / P0 is 0.62 with insufficient increase in the main pipe flow, both criteria are met simultaneously. This indicates that the lifting valve is not fully open, resulting in an output steam inlet fault signal and execution of gun retraction, preventing the sootblowing gun from drying out and bending.

[0124] See Figure 6 This application also provides an electronic device. Figure 5 The unmanned follow-up control unit 504 and the fusion diagnostic unit 503 shown can be implemented by the electronic device 600. Typically, the electronic device 600 includes one or more processors 601 and one or more memories 602.

[0125] Processor 601 may include one or more processing cores, such as a quad-core processor or a hexa-core processor. Processor 601 may be implemented using at least one hardware form selected from DSP (Digital Signal Processing), FPGA (Field-Programmable Gate Array), and PLA (Programmable Logic Array). Processor 601 may also include a main processor and a coprocessor. The main processor, also known as a CPU (Central Processing Unit), is used to process data in the wake-up state; the coprocessor is a low-power processor used to process data in the standby state. In some embodiments, processor 601 may further include an AI (Artificial Intelligence) processor, which is used to handle neural network inference operations related to image recognition and fault classification.

[0126] The memory 602 may include one or more computer-readable storage media, which may be non-transitory. The memory 602 may also include high-speed random access memory and non-volatile memory, such as one or more disk storage devices or flash memory devices. In some embodiments, the non-transitory computer-readable storage media in the memory 602 are used to store at least one piece of program code, which is executed by the processor 601 to implement the unmanned blasting method for coal-fired boilers provided in the method embodiments of this application.

[0127] Those skilled in the art will understand that Figure 6 The structure shown does not constitute a limitation on the electronic device and may include more or fewer components than shown, or combine certain components, or use different component arrangements.

[0128] In addition, the device provided in the embodiments of this application may specifically be a chip, component or module. The chip may include a connected processor and a memory. The memory is used to store instructions. When the processor calls and executes the instructions, the chip can execute the unmanned blasting method for coal-fired boilers provided in the above embodiments.

[0129] This embodiment also provides a computer-readable storage medium storing computer program code. When the computer program code is run on a computer, the computer executes the above-mentioned related method steps to realize the unmanned blasting method for coal-fired boilers provided in the above embodiment.

[0130] This embodiment also provides a computer program product that, when run on a computer, causes the computer to perform the aforementioned related steps to realize the unmanned follow-up blowing method for coal-fired boilers provided in the above embodiment.

[0131] In this embodiment, the device, computer-readable storage medium, computer program product, or chip are all used to execute the corresponding methods provided above. Therefore, the beneficial effects they can achieve can be referred to the beneficial effects in the corresponding methods provided above, and will not be repeated here.

[0132] Through the above description of the embodiments, those skilled in the art will understand that, for the sake of convenience and brevity, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above.

[0133] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules or units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another apparatus, 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 units may be electrical, mechanical, or other forms.

[0134] 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.

Claims

1. A method for unmanned operation of a coal-fired boiler, characterized in that, The method includes: The soot blowing start-up command sequence is confirmed and multi-source operating data of the soot blower is obtained. The soot blowing start-up command sequence is used to indicate the start-up order and start-up time of the soot blowers corresponding to each heating surface of the boiler. The multi-source operating data includes soot blower stroke position data, steam condition data, electrical operation data and visual image data. Based on the multi-source operating data, a fusion diagnostic process is performed to obtain fault alarm signals. The fusion diagnostic process includes multi-source operating feature extraction, feature fusion, and cross-validation classification. The fault alarm signals include transmission fault signals, steam inlet fault signals, and steam leakage fault signals of the sootblower. In response to the fault alarm signal, perform safety procedures on the faulty sootblower and reconfirm the sootblowing start command sequence.

2. The method according to claim 1, characterized in that, The confirmation soot blowing start command sequence includes: Based on the heat transfer status parameters of multiple heating surfaces of the boiler, the fouling index of multiple heating surfaces is determined. The heat transfer status parameters include inlet and outlet flue gas temperature, wall temperature and flue gas differential pressure data. Based on the dirt index, the soot blowing priority ranking is determined. The soot blowing priority ranking is used to characterize the order in which the soot blowers corresponding to each heated surface are subjected to soot blowing. Based on the priority of soot blowing, the operation grouping, start-up sequence and steam parameters of multiple soot blowers are arranged to generate the soot blowing start-up command sequence; The acquisition of multi-source operating data of the sootblower includes: The gun body of the sootblower is continuously detected throughout its entire stroke to obtain stroke position data, which includes the continuous position signal of the gun body throughout its entire stroke and the real-time travel rate calculated accordingly. Pressure and flow rate are detected on the downstream side of the steam inlet passage of a single sootblower to obtain the steam operating condition data, which includes the steam inlet pressure, main pipe pressure and total steam flow rate of a single sootblower. The current of the drive motor of the sootblower is detected to obtain the electrical operation data, which includes the operating current of the drive motor. Image acquisition is performed on the soot blower to obtain the visual image data, which includes image data covering the soot blower protective cover, the traveling parts and the valve stem sealing part.

3. The method according to claim 1, characterized in that, The process involves fusing and diagnosing the multi-source operational data to obtain fault alarm signals. This fusing and diagnostic process includes multi-source operational feature extraction, feature fusion, and cross-validation classification. The fault alarm signals include transmission fault signals, steam inlet fault signals, and steam leakage fault signals for the sootblower. Differential calculation processing is performed on the travel position data of the multi-source operation data to obtain the rate change characteristics; The steam operating condition data of the multi-source operation data is processed by steam flow state calculation to obtain flow characteristics, which include pressure ratio characteristics and flow increment characteristics. The electrical operation data of the multi-source operation data are subjected to current fluctuation statistical processing to obtain current fluctuation characteristics; The visual image data of the multi-source operation data is processed by image recognition to obtain visual position features and steam leakage image features; The continuous positions represented by the travel position data are compared with the visual positions represented by the visual position features in real time to obtain the position deviation. The actual running time of a single forward and backward movement of the sootblower is compared with the standard travel time to obtain the time deviation. The position deviation, duration deviation, current fluctuation characteristics, and rate change characteristics are aligned according to the soot blowing stroke nodes to obtain joint analysis characteristics; Based on the positional deviation, the duration deviation, and the joint analysis features, a fusion feature is constructed; When the position deviation exceeds the set deviation threshold and continues for a set duration, and is accompanied by abnormal rate change characteristics or current fluctuation characteristics, the source of the abnormality is determined and the determination result of a single detection source is corrected. When the soot blower is in or out of position and its travel speed is zero, the current continuous position code value is recorded and stored in the corresponding historical record. The average value of the records from the nearest set number of times is used as the reference code value to perform self-checking and correction on the continuous position signal; When the rate change characteristics indicate that the travel rate is continuously lower than the set rate threshold or drops sharply to zero and remains so for a set duration, or remains higher than the set rate upper limit, the transmission fault signal is output. When the pressure ratio characteristic is lower than the set pressure threshold and the flow increment characteristic does not reach the set flow increment of the corresponding single soot blower, the steam inlet fault signal is output. When the steam leakage image features indicate that steam leakage has been identified in the valve stem and sealing area, the steam leakage fault signal is output.

4. The method according to claim 1, characterized in that, The response to the fault alarm signal includes performing safety procedures on the faulty sootblower and reconfirming the sootblowing start command sequence, including: Identify the sootblower corresponding to the fault alarm signal to obtain the faulty sootblower; In response to the fault alarm signal, the current soot blowing process is interrupted, a retraction command is issued to the soot blower to drive it back to its original position, and the restart permission of the faulty soot blower is locked. Record the travel location data, steam condition data, electrical operation data, and visual image data at the time of the fault. The faulty sootblower is removed from the sootblowing start-up command sequence, and the start-up order and start-up time of the remaining sootblowers are reconfirmed based on the latest dirt index of multiple heating surfaces of the boiler, thereby generating an updated sootblowing start-up command sequence.

5. An unmanned follow-up blowing system for a coal-fired boiler, characterized in that, The system includes: a field sensing unit, a DCS control unit, a fusion diagnostic unit, an unmanned follow-up blowing control unit, and a local execution unit; The field sensing unit is configured to correspond to the sootblower and includes an absolute position encoder installed on the sootblower transmission mechanism, a post-valve pressure transmitter installed after the lifting valve of a single sootblower, a main pipe pressure transmitter and a steam flow transmitter installed on the sootblower steam main pipe, a sootblower motor current acquisition module, and a field camera device deployed on the corresponding sootblower. The absolute position encoder is used to output the continuous position signal of the sootblower gun body throughout its entire stroke and calculate the real-time travel speed, thus forming the stroke position data; the valve downstream pressure transmitter, the main pipe pressure transmitter and the steam flow transmitter are used to collect the inlet steam pressure, main pipe pressure and total steam flow of a single sootblower, respectively, thus forming the steam operating condition data. The motor current acquisition module is used to acquire the operating current of the sootblower drive motor, which constitutes the electrical operation data. The on-site camera device is used to collect image data covering the soot blower protective cover, walking parts and valve stem sealing parts, which constitutes the visual image data. The unmanned soot blowing control unit is used to confirm the soot blowing start command sequence, which is used to indicate the start order and start time of the soot blowers corresponding to each heating surface of the boiler. The fusion diagnostic unit is used to perform fusion diagnostic processing based on the multi-source operating data to obtain fault alarm signals. The fusion diagnostic processing includes multi-source operating feature extraction, feature fusion, and cross-validation classification. The fault alarm signals include transmission fault signals, steam inlet fault signals, and steam leakage fault signals of the sootblower. The local execution unit includes a sootblower electrical control cabinet, which is used to respond to the fault alarm signal and perform safety procedures on the faulty sootblower; The unmanned soot blowing control unit is used to respond to the fault alarm signal and reconfirm the soot blowing start command sequence.

6. The system according to claim 5, characterized in that, The unmanned soot blowing control unit includes a heated surface cleanliness assessment subunit and a soot blowing strategy arrangement subunit; The heating surface cleanliness assessment subunit is used to calculate the dirt index of each heating surface based on the inlet and outlet flue gas temperature, wall temperature and flue gas differential pressure data of each heating surface of the boiler. The soot blowing strategy arrangement subunit is used to automatically sort the soot blowing priorities according to the dirt index, and dynamically arrange the operation group, start-up sequence and steam parameters of the long telescopic and semi-telescopic soot blowers according to the soot blowing priority sorting, and generate the soot blowing start command sequence. The absolute position encoder is a multi-turn absolute encoder, which is installed on the drive shaft of the sootblower drive motor; The on-site camera device is an industrial camera, and each industrial camera covers a soot blower; The motor current acquisition module is connected to the drive motor circuit of the soot blower.

7. The system according to claim 5, characterized in that, The fusion diagnostic unit includes a full-process status analysis module, a steam condition diagnostic module, a visual intelligent recognition module, and a multi-source fusion diagnostic module. The full-stroke state analysis module is used to perform differential calculation processing based on the continuous position signal of the absolute position encoder to obtain the rate change characteristics and perform stroke position self-check. The steam condition diagnostic module is used to simultaneously collect the detection data of the downstream pressure transmitter, the main pipe pressure transmitter and the steam flow transmitter when a single sootblower is in the running command state, perform steam flow state calculation and processing, and obtain flow characteristics including pressure ratio characteristics and flow increment characteristics. The visual intelligent recognition module is used to perform image recognition processing on the image data collected by the on-site camera device. By recognizing the relative position of the visual mark set on the sootblower protective cover and the indicator mark of the sootblower traveling part, the visual position feature is obtained, and the steam leakage image feature of the valve stem and sealing part is recognized. The multi-source fusion diagnostic module is used for feature fusion and cross-validation classification: it compares the position represented by the continuous position signal with the visual position represented by the visual position feature in real time to obtain the position deviation; it compares the actual running time of a single sootblower's single forward and backward movement with the standard stroke time to obtain the duration deviation; it combines the current fluctuation characteristics obtained from the working current collected by the motor current acquisition module with the rate change characteristics to determine jamming, slippage, lift valve failure, and steam leakage failure, corrects the judgment results of a single detection source, and outputs the transmission failure signal, the steam inlet failure signal, and the steam leakage failure signal.

8. The system according to claim 5, characterized in that, The DCS control unit is used to respond to the fault alarm signal, forcibly interrupt the current soot blowing process, and send a gun retraction command to the soot blower electrical control cabinet. The sootblower electrical control cabinet also includes an active defense priority control circuit. The active defense priority control circuit is used to execute the gun retraction command before the normal program control command. In response to the gun retraction command, the sootblower motor is driven to move the sootblower gun back to its original position, and the restart permission of the faulty sootblower is locked. The unmanned sootblowing control unit is used to record the full-dimensional operating data of the sootblower at the time of failure, remove the faulty sootblower from the sootblowing start command sequence, and reconfirm the start order and start time of the remaining sootblowers based on the latest dirt index of each heating surface of the boiler, and generate an updated sootblowing start command sequence.

9. The system according to claim 7, characterized in that, The system also includes: Visual identification components installed on the body of the soot blower; The visual identity component includes: Multiple reflective strips are provided on the protective cover of the soot blower and are distributed at equal intervals along the travel direction; An indicator arrow is provided on the traveling component of the soot blower, and the indicator arrow is close to and points towards the reflective strip; The image data acquired by the on-site camera device includes the relative positional relationship between the indicator arrow and the reflective strip; Based on the relative positional relationship, the visual positional features are obtained by performing image recognition through the visual intelligent recognition module.

10. The system according to claim 5, characterized in that, The fusion diagnostic unit is deployed in the third zone of power production safety, while the field sensing unit and the unmanned follow-up blowing control unit are deployed in the first zone of power production safety. The system also includes a cross-security zone interaction unit, which includes a forward unidirectional transparent transmission module and a reverse hard contact back transmission module. The forward unidirectional transparent transmission module adopts a forward physical isolation device to transmit the operating status, grouping information and analog data of the sootblower in one direction from the first safety zone to the third safety zone. The reverse hard contact feedback module includes a relay output unit, which is used to convert the fault alarm signal into a passive dry contact switch signal and connect it to the DI input channel of the DCS control unit in the safety zone through a control cable.

Images