A silo pump ash anti-blocking pipe system and an anti-blocking control method

Abstract

The application discloses a kind of warehouse pump ash conveying anti-blocking pipe systems and anti-blocking control methods, belong to pneumatic ash conveying intelligent anti-blocking control technical field.The application uses software and hardware collaborative prevention and control scheme, independently arranges ICS conveying actuator between every two adjacent warehouse pumps, and the timing data of pipeline pressure is collected in real time by its built-in high-precision pressure acquisition unit.Meanwhile, intelligent anti-blocking control software system is built, machine learning model is trained based on full-condition pressure sample, and fault pipeline section is accurately identified by initial ash plug, ash plug aggregation and pipe blocking fault.The system can issue differentiated instructions according to the degree of blockage, drive ICS conveying actuator to point and grade air supplement and blow, and break ash plug and dredge pipeline in advance.The application realizes early prediction of ash conveying pipeline blockage, unmanned automatic unblocking and self-adaptive energy-saving operation, effectively improves the operation stability and intelligent level of ash conveying system, and is suitable for various industrial enterprises dense-phase pneumatic ash conveying scenes.

Description

Technical Field

[0001] This invention relates to the field of intelligent control technology for pneumatic ash conveying anti-blockage, specifically to a silo pump ash conveying anti-blockage pipeline system and anti-blockage control method, which is mainly applicable to intelligent anti-blockage and automatic unblocking operation and maintenance scenarios of positive pressure dense phase silo pump pneumatic ash conveying pipelines in industries such as thermal power plants, chemical industry, and metallurgy. Background Technology

[0002] Dense-phase pneumatic conveying of fly ash is the mainstream technology for conveying dry ash materials in the industrial field. Using compressed air as the power carrier, it transports powdery materials such as fly ash from the silo pump along a closed pipeline to the ash storage equipment. It boasts advantages such as large conveying capacity, good sealing, low material loss, and a high degree of automation, and is widely used in various industrial production scenarios. However, in actual long-term operation, due to multiple factors such as uneven fly ash moisture content, fluctuations in the conveying air pressure, imbalance in the ash-air ratio, gradual changes in resistance over long distances, and changes in ambient temperature, powdery materials are prone to slow deposition, aggregation, and adhesion inside the conveying pipeline, gradually forming stable ash plugs. If not intervened in time, this can quickly develop into pipeline blockage.

[0003] Currently, traditional ash conveying anti-blockage technologies in the industry mainly adopt three modes: continuous purging under constant pressure throughout the entire area, manual purging at fixed points, and timed batch purging. These technologies have significant shortcomings. First, the monitoring methods are crude and singular, only able to identify overall pressure anomalies in the ash conveying pipeline, unable to accurately capture slight pressure fluctuations in local pipe sections, and difficult to pinpoint the specific location of initial ash plugs. Faults can only be detected after the blockage has fully formed and the system pressure is significantly abnormal, resulting in a severe delay in intervention. Second, the control logic is rigid and inflexible, mostly employing a uniform purging strategy of timed, quantitative, and pressure-controlled purging. It cannot adaptively adjust operating parameters according to the real-time blockage status of the pipeline, resulting in a large number of ineffective air replenishment operations, high system energy consumption, and long-term ineffective airflow scouring will aggravate wear on the inner wall of the pipeline. Third, the degree of automation in operation and maintenance is low. After severe blockage occurs, the system must be shut down for manual unblocking operations, which not only significantly reduces the continuous operating efficiency of the ash conveying system, but also accelerates the aging of equipment such as silo pumps, air compressors, and conveying pipelines due to frequent equipment start-ups and shutdowns, significantly increasing equipment operation and maintenance costs and downtime losses.

[0004] More critically, existing anti-clogging technologies lack deep hardware-software synergy. Most devices only have mechanical purging functions, lacking intelligent sensing and data analysis capabilities. The few systems equipped with pressure monitoring rely solely on fixed thresholds for simple judgment, lacking data training, feature extraction, and self-learning optimization capabilities. They cannot identify the weak abnormal signals of newly formed ash plugs and therefore cannot achieve predictive proactive anti-clogging. Addressing these shortcomings of existing technologies, developing a segmented, precise monitoring, intelligent identification and prediction, and targeted adaptive thrombus breaking intelligent anti-clogging system and control method for silo pumps has become a pressing technical problem for the industry. Summary of the Invention

[0005] The purpose of this invention is to overcome the shortcomings of existing technologies, such as weak anti-blockage capability, inaccurate fault location, delayed anti-blockage intervention, high energy consumption, low level of intelligence, and poor adaptability. It provides a silo pump ash conveying anti-blockage pipe system and anti-blockage control method. By deploying independent ICS conveying actuators between adjacent silo pumps and combining them with intelligent control software trained by big data machine learning, a hardware-software collaborative prevention and control system of "hardware segmented perception and execution + software intelligent analysis and control" is constructed. This system achieves accurate identification of ash plugs in the ash conveying pipeline, early prediction of blockage, and automatic plug breaking at fixed points, thereby reducing the pipeline blockage failure rate from the root and improving the operational stability and energy efficiency of the ash conveying system.

[0006] To achieve the above-mentioned objectives, the present invention adopts the following technical solution:

[0007] This invention discloses a silo pump ash conveying anti-blockage pipeline system, comprising multiple silo pumps arranged in series, ash conveying pipelines, several ICS conveying actuators, and an intelligent anti-blockage control software system. Adjacent silo pumps are connected through ash conveying pipelines to form a complete ash conveying channel. An ICS conveying actuator is installed independently on each independent pipeline section between every two sets of adjacent silo pumps, achieving full coverage and seamless monitoring and operation of the ash conveying pipeline. The ICS conveying actuator integrates a high-frequency pressure acquisition unit, an adjustable air injection and blowing unit, and a two-way communication control unit, enabling it to independently complete single-section pipeline pressure data acquisition, signal uploading, and command execution. The intelligent anti-blockage control software system generates an intelligent ash plug blockage identification model based on massive full-condition pressure data training. It can analyze the pressure fluctuation characteristics of each pipeline section in real time, accurately determine the pipeline operating conditions, locate the fault point, and issue differentiated control commands according to the degree of blockage, driving the corresponding ICS conveying actuator to complete adaptive air injection and thrombus breaking operations.

[0008] This invention also discloses a corresponding ash conveying anti-blockage control method for silo pumps, which corely encompasses six major processes: offline model training, real-time data acquisition across the entire domain, intelligent working condition analysis and positioning, graded adaptive thrombus breaking, closed-loop status monitoring, and data self-learning iteration. This invention, through labeled training of pressure data under multiple working conditions, enables the model to accurately capture the pressure characteristic differences of different fault states, abandoning the traditional coarse-grained mode of fixed threshold judgment. During the online operation phase, differentiated air replenishment strategies are adopted for different fault states such as initial ash plugs and formed blockages, achieving early elimination of minor hidden dangers and rapid unblocking of severe faults, all without manual intervention, forming a complete automated prevention and control closed loop. Simultaneously, relying on self-learning capabilities, the prevention and control effect is continuously optimized to adapt to complex and ever-changing ash conveying conditions.

[0009] Compared with existing traditional ash conveying anti-blockage technologies, the present invention has the following significant advantages:

[0010] 1. Segmented full-area monitoring and precise fault location: The segmented hardware architecture of deploying ICS conveying actuators one-to-one between adjacent silo pumps enables independent monitoring and control of each segment of the ash conveying pipeline. This completely solves the technical problem that traditional overall monitoring cannot identify local minor faults or locate the blockage point, and enables precise fault location and handling.

[0011] 2. Intelligent prediction and intervention, proactive prevention: Relying on big data machine learning models, it can capture the weak pressure abnormal signals of nascent gray emboli, realize early prediction of pipe blockage failure, transform the traditional passive blockage clearing mode after the failure occurs into an active blockage prevention mode at the bud stage of hidden danger, and significantly reduce the pipe blockage failure rate.

[0012] 3. Graded adaptive operation, energy saving and consumption reduction: The system adaptively adjusts the air supply pressure, duration and flow rate according to the degree of pipeline blockage. Differentiated operation strategies are adopted for different fault conditions to eliminate the ineffective energy consumption of traditional whole-area timed purging, effectively reduce the system's operating power consumption, and at the same time reduce the scouring and wear of the pipeline by the airflow, thus extending the service life of the equipment.

[0013] 4. Closed-loop control of hardware and software with a high degree of automation: It constructs a fully automatic prevention and control closed loop of hardware perception and execution + software intelligent judgment. No manual monitoring or intervention is required throughout the entire process. It automatically completes the entire process of monitoring, identification, thrombus breaking and resetting, which greatly reduces the workload of manual operation and maintenance and improves the level of automation of the ash conveying system.

[0014] 5. Self-iterative optimization and strong adaptability: The system has the ability to learn data on its own and can continuously optimize model parameters and control logic based on historical operating data. It can adapt to the dynamic changes of materials, working conditions and environment, and continuously optimize the prevention and control effect in long-term operation, adapting to various complex ash conveying scenarios. Attached Figure Description

[0015] Figure 1 This is a schematic diagram of the internal structure and principle of the ICS conveying actuator of the present invention;

[0016] Figure 2 This is a flowchart illustrating the overall process of the ash conveying anti-clogging control method for the silo pump of the present invention. Detailed Implementation

[0017] The present invention will be further described in detail and completely below with reference to specific embodiments. These embodiments are implemented based on the technical solution of the present invention, and provide detailed implementation methods and specific operation processes. However, the scope of protection of the present invention is not limited to the following embodiments.

[0018] 1. System hardware deployment and implementation

[0019] Multiple silo pumps are arranged in series along the material conveying direction. The discharge and inlet ends of adjacent silo pumps are connected by a straight ash conveying pipeline, forming a closed and continuous pneumatic ash conveying channel. An ICS conveying actuator is fixedly installed in the middle of the straight ash conveying pipeline section between every two sets of adjacent silo pumps. All ICS conveying actuators are independently numbered, communicate independently, and are independently controlled, without interfering with each other. The high-precision pressure sensor of the ICS conveying actuator is installed flush against the inner wall of the pipeline, without airflow obstruction, and can accurately collect dynamic pressure data of the gas-ash mixture in real time. The equipment has radially opened air supply holes that connect to the plant's compressed air source. The air supply is controlled by a built-in solenoid valve, and with a high-precision flow regulating valve, the air supply pressure, duration, and flow rate can be precisely adjusted. All ICS conveying actuators establish bidirectional real-time communication with the backend intelligent anti-blocking management software system via industrial Ethernet, ensuring the real-time performance and stability of data transmission and command execution.

[0020] 2. Implementation of offline training for intelligent models

[0021] The system collects offline operational data from the ash conveying system under all operating conditions. The data samples comprehensively cover low, medium, and high ash conveying loads, dry and humid fly ash states, and complex operating conditions such as gas source pressure fluctuations and ambient temperature changes. It includes pressure time-series data for five core operating conditions: normal conveying, sporadic initial formation of ash plugs, continuous accumulation of ash plugs, slight pipeline blockage, and severe pipeline blockage. The collected raw pressure data undergoes filtering, noise reduction, outlier removal, and data normalization preprocessing to eliminate data deviations caused by environmental interference and equipment errors. Subsequently, all data samples are accurately labeled with operating conditions to construct a standardized training dataset. Machine learning algorithms are used to iteratively train the dataset, focusing on extracting core features such as pressure fluctuation amplitude, fluctuation frequency, pressure steady-state offset, and pressure change rate under different operating conditions to construct an intelligent ash plug blockage identification model. The model weight parameters are continuously iterated and optimized until the model's operating condition identification accuracy reaches over 99%, at which point the model is solidified and the system is deployed.

[0022] 3. Implementation of online intelligent anti-blocking control

[0023] During normal operation of the ash conveying equipment, all ICS conveying actuators continuously collect pressure time-series data of the corresponding pipe sections at a high frequency of 100Hz and upload it to the intelligent anti-blocking and control software system in real time. The software system calls upon the fixed intelligent identification model to analyze the pipeline pressure characteristics segment by segment, realizing dynamic monitoring and analysis of the entire pipeline.

[0024] When the pipeline pressure is stable and the fluctuation range is within the preset normal threshold range, the system determines that the pipeline is delivering normally, maintains normal monitoring status, and does not trigger any gas replenishment operation.

[0025] When a small fluctuation and slow, continuous rise in pipeline pressure is detected, the model identifies it as an early warning state of initial ash plugs. The system outputs a short-term, low-flow, low-pressure air replenishment command to drive the ICS delivery actuator of the corresponding pipe section to provide micro-blowing, disturb the ash deposited in the pipe, and break up the initial ash plugs in advance, thus preventing blockage formation from the source.

[0026] When the system detects a continuous and significant increase in pipeline pressure, violent pressure fluctuations, or an abnormal rate of pressure change, the model determines that the pipeline is blocked by ash plugs or formed blockages. The system outputs a long-term, high-flow-rate, high-pressure air supply command to impact the blocked pipe section with high-intensity airflow, breaking up and solidifying the ash plugs, and quickly clearing the blocked pipeline.

[0027] During all gas replenishment and rupture operations, the system continuously monitors the dynamic changes in pipeline pressure in real time. Once the pressure data returns to the normal steady-state range, the fault is immediately determined to be eliminated, and a shutdown command is issued simultaneously. The corresponding ICS delivery actuator stops gas replenishment to avoid energy waste and wear on the inner wall of the pipeline caused by excessive gas replenishment.

[0028] 4. System self-learning iterative optimization implementation

[0029] The system features full-process data recording and autonomous iterative optimization capabilities. It can automatically store location information, raw pressure data, control parameters, operation duration, and results for every ash hydrant warning, hydrant rupture operation, and pipe blockage fault, building a dedicated operating condition database. The system periodically retrieves historical operating condition data to complete secondary iterative training of the model, adaptively optimizing fault identification thresholds and gas supply control parameters under different operating conditions. This continuously improves the model's identification accuracy and the system's adaptive control capabilities, adapting to subtle changes in material states, equipment parameters, and environmental conditions during long-term operation, ensuring the system's long-term stable and efficient operation.

[0030] The above description is merely a preferred embodiment of the present invention and is not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention. Structures, methods, and modules not described in detail in the present invention are all existing conventional technologies in the art and need not be elaborated upon.

Claims

1. A silo pump ash conveying anti-clogging pipe system, characterized in that, It includes multiple sets of cascaded pumps, ash conveying pipelines, several ICS conveying actuators, and an intelligent anti-blocking and control software system; the ash conveying pipelines connect adjacent cascade pumps to form a continuous pneumatic ash conveying channel; each independent ash conveying pipeline section between two adjacent cascade pumps is equipped with a separate ICS conveying actuator. The ICS delivery actuator integrates a pressure acquisition unit, a controllable air supply unit, and a communication control unit. The pressure acquisition unit is used to acquire real-time pressure timing data of the corresponding pipeline section at high frequency. The controllable air supply unit is used to receive control commands to complete the fixed-point air supply and blow-through thrombus breaking action. The communication control unit realizes two-way data interaction and command execution with the intelligent anti-blockage management software system. The intelligent anti-blockage control software system has a built-in trained ash embolism blockage identification model, which is used to receive pressure data uploaded by each ICS conveying actuator, analyze pipeline pressure fluctuation characteristics in real time, identify the ash embolism generation location, degree of aggregation and blockage fault status, and issue corresponding control commands to drive the matched ICS conveying actuator to operate precisely.

2. The ash conveying anti-clogging pipe system for a silo pump according to claim 1, characterized in that, The ICS conveying actuator is configured one-to-one with the pipeline section between the two adjacent pump compartments, enabling independent sensing, control, and unblocking of a single pipeline section, and completing the full-area fault monitoring and precise thrombus breaking operation of the ash conveying pipeline.

3. The ash conveying anti-clogging pipe system for a silo pump according to claim 1, characterized in that, The pressure acquisition unit uses a high-frequency, high-precision pressure sensor with a sampling frequency of no less than 100Hz, which can capture minute pressure fluctuation signals in the pipeline, accurately identify the weak pressure anomalies formed by the initial ash plugs, and realize early fault prediction.

4. The ash conveying anti-clogging pipe system for a silo pump according to claim 1, characterized in that, The controllable air supply unit connects to the plant's compressed air source and has a built-in controllable solenoid valve and flow regulating valve. It can adaptively adjust the air supply pressure, air supply duration, and air supply flow according to the judgment instructions of the software system, so as to meet the needs of puncture operations with different ash embolism thickness and different degrees of blockage.

5. The ash conveying anti-clogging pipe system for a silo pump according to claim 1, characterized in that, The intelligent anti-blocking and control software system includes a data acquisition module, a model training module, a real-time identification and analysis module, an instruction output module, a data storage module, and a human-computer interaction module. The data acquisition module receives pressure data and equipment operating status data uploaded by each ICS conveying actuator in real time. The model training module iteratively trains and optimizes the feature models for four operating conditions—normal conveying, initial ash plug formation, minor blockage, and severe blockage—based on the full-condition pressure sample data of the ash conveying system. The real-time identification and judgment module imports real-time pressure data into the trained model to accurately determine the pipeline operating conditions and fault locations. The command output module outputs differentiated control commands based on the judgment results. The data storage module stores historical operating condition data, model parameters, and operation logs. The human-machine interaction module enables real-time display of operating conditions, manual parameter setting, audible and visual fault alarms, and historical data query.

6. A method for preventing blockage in ash conveying in a silo pump, characterized in that, The system applied to the silo pump ash conveying anti-clogging pipe system according to any one of claims 1-5 includes the following steps: S1. System initialization self-test: Each ICS conveyor actuator is powered on and completes a hardware self-test, confirming that the pressure acquisition unit, controllable air supply unit and communication control unit are in normal standby mode, and the intelligent anti-blockage control software system loads the solidified model parameters and control thresholds. S2. Intelligent Model Offline Training: Collect pipeline pressure time-series data under all operating conditions of the silo pump ash conveying system, covering normal conveying, initial ash plug formation, ash plug accumulation, slight blockage, and severe blockage under different ash conveying loads, different gas source pressures, and different fly ash moisture conditions; perform noise reduction, outlier removal, normalization preprocessing on the raw data and complete the operating condition labeling; iteratively train the model through machine learning algorithms to optimize the model weight parameters and obtain an ash plug blockage identification model that can accurately distinguish various operating conditions; S3. Real-time monitoring of operating conditions across the entire area: During the operation of the ash conveying system, all ICS conveying actuators synchronously collect pressure data of the corresponding pipeline sections at high frequency and upload it to the intelligent anti-blockage control software system in real time. S4. Intelligent and precise judgment and location: The software system inputs real-time pressure data into the trained recognition model, and by extracting the core features of pressure fluctuation amplitude, fluctuation frequency, pressure steady-state offset, and pressure change rate, it determines the real-time operating status of the pipeline and accurately locates the specific pipeline segment where the ash plug or blockage fault occurs. S5, graded adaptive thrombus breaking operation: If a warning of initial ash thrombus is detected, a short-term, low-flow air replenishment command is output to drive the corresponding ICS conveyor actuator to provide micro-blowing assistance, breaking up the ash thrombus in advance and avoiding blockage formation; if an ash thrombus accumulation or pipe blockage fault is detected, a long-term, high-flow, high-pressure air replenishment command is output to forcefully impact and break up the ash thrombus and clear the blocked pipeline. S6. Closed-loop monitoring and reset: During the rupture operation, pipeline pressure data is continuously collected in real time. When the pressure data returns to the normal delivery threshold range, it is determined that the pipe blockage fault has been completely eliminated. The software system issues a shutdown command, the corresponding ICS delivery actuator stops the gas supply operation, the system resumes normal monitoring, and completes the single anti-blockage and clearing closed-loop control. S7. Data self-learning iteration: The system automatically stores the location, type, pressure data, rupture control parameters and operation results of each fault. It regularly iterates and optimizes the model identification threshold and control parameters based on historical operation data to continuously improve the system's adaptability and prevention accuracy.

7. The method for preventing blockage in silo pump ash conveying according to claim 6, characterized in that: In step S2, the training samples cover extreme working conditions such as alternating high and low loads, fluctuating gas source pressure, and changes in the dry and wet state of fly ash, ensuring that the trained model can be adapted to all scenarios and operate stably in all weather conditions.

8. The method for preventing blockage in silo pump ash conveying according to claim 6, characterized in that: In step S5, the system presets multiple pressure thresholds, corresponding to four working modes: normal operation, ash plug warning, minor blockage, and severe blockage. This enables graded, adaptive, and precise purging, eliminating ineffective gas supply across the entire system and reducing system energy consumption and equipment wear.

Images