Method and device for detecting coal blockage of coal feeder

Abstract

The invention discloses a method and a device for detecting coal blockage of a coal feeder. The detection method comprises the following steps: acquiring a rotating speed real-time value, a current real-time value and a material density detection value of the coal feeder in a current detection period; in combination with a coal feeding rate instruction value received by a coal feeder, calculating normal operation intervals corresponding to the rotating speed real-time value, the current real-time value and the material density detection value respectively based on a dynamic mapping function; and when the rotating speed real-time value, the current real-time value and the material density detection value deviate from the normal operation intervals corresponding to the coal feeding rate instruction value and exceed preset threshold values, it is judged that coal blockage happens to the coal feeder, and a vibration assembly in a raw coal bunker for supplying coal to the coal feeder is started. The rotating speed, the current and the material density of the coal feeder are monitored in real time, the normal operation interval of each parameter under the given coal feeding rate instruction is dynamically calculated in combination with the preset transfer function model, coal blockage is judged and the raw coal bunker rapping device is triggered once the detected value deviation interval exceeds the threshold value, the response time difference is shortened, and the hidden danger of overload tripping of the coal feeder is avoided.

Description

Technical Field

[0001] The present invention relates to the technical field of coal feeder equipment detection, and in particular to a method and device for detecting coal blockage in a coal feeder. Background Art

[0002] As a key component of the pulverizing system in coal-fired power plants, weighing coal feeders ensure continuous and uniform coal delivery to the pulverizers. Widely used electronic weighing and metering coal feeders (such as the CS2024) utilize microcomputer control for precise weighing and automatic speed regulation, dynamically adjusting the coal feed rate based on boiler load demand. Raw coal enters the feeder from the coal bunker through a coal gate. A belt conveyor transports the raw coal to the coal outlet, where it falls through a coal drop pipe into the pulverizer. Any residual coal dust on the belt surface is removed by a cleaning scraper.

[0003] In actual operation, when high-moisture coal is mixed and burned, the humidity and viscosity of the coal increase significantly, resulting in increased agglomeration of coal particles. Due to the cone-shaped structure of the lower part of the raw coal bin, the fluidity of the coal flow in the bin decreases, especially in the area 1.5 meters above the coal gate and the straight pipe section at the bottom of the gate, where coal blockages frequently occur. The existing technology only installs a coal-break monitoring switch at the outlet of the coal feeder, while coal blockages mostly occur at the entrance, and there is a serious lag from the occurrence of coal blockage to the triggering of the alarm. The more prominent contradiction is that although the straight pipe at the bottom of the coal gate is equipped with an electric vibrating device, there is a lack of mechanical clearing means in the high-incidence area of ​​coal blockage above the gate. Relying on manual knocking is not only time-consuming and labor-intensive, but also difficult to guarantee the treatment effect.

[0004] The current coal blockage monitoring mechanism suffers from fundamental flaws. The initial signs of coal blockage often manifest as a "bridging" phenomenon within the coal bunker. At this point, the material density gradually decreases due to the thinning of the coal seam. To maintain the required coal feed rate, the feeder speed is passively and continuously increased. This process creates two risks: First, if the speed reaches full load and still fails to meet the required speed, the system lacks a warning mechanism, which can easily cause the feeder to overload and trip. Second, the coal outage alarm is only triggered when the coal seam at the tail of the belt is too thin, causing the detection baffle to drop. At this point, the coal blockage has accumulated to a critical level, making conventional vibration ineffective in quickly restoring a stable coal flow. Summary of the Invention

[0005] The purpose of the embodiment of the present invention is to provide a method and device for detecting coal blockage in a coal feeder. By real-time monitoring of the speed, current and material density of the coal feeder, combined with a preset transfer function model, the normal operating range of each parameter under a given coal feeding rate instruction is dynamically calculated. Once the detection value deviates from the range and exceeds the threshold, coal blockage is determined and the raw coal bin vibration device is triggered, which significantly shortens the response time difference (such as the risk can be identified before the speed reaches full load), avoids the hidden danger of coal feeder overload tripping, and at the same time, the vibration device is linked to clear the blockage in time, greatly reducing the need for manual intervention, and ensuring the continuity of coal feeding and the stability of unit operation.

[0006] To solve the above technical problems, a first aspect of an embodiment of the present invention provides a method for detecting coal feeder blockage, comprising the following steps: Obtain the real-time speed value, current value and material density detection value of the coal feeder in the current detection cycle; Obtaining a time series data set of coal feed rate command values, speed measured values, current measured values, and material density measured values ​​under multiple operating conditions; generating a dynamic mapping function from the coal feed rate command value to the speed, current, and material density through machine learning algorithm training based on the time series data set; calculating the dynamic normal operating range of the speed, current, and material density using the dynamic mapping function in combination with the coal feed rate command value received by the coal feeder; and adaptively adjusting the normal operating range as the coal feed rate command value changes; When the real-time value of the rotational speed, the real-time value of the current and the material density detection value deviate from the normal operating range corresponding to the coal feeding rate instruction value by more than a preset threshold, it is determined that the coal feeder is blocked by coal, and the vibration component in the raw coal bin that supplies coal to the coal feeder is started.

[0007] Furthermore, based on the time series data set, a dynamic mapping function is generated from the coal feed rate command value to the speed, current and material density through machine learning algorithm training, wherein the mapping function output is a parameter prediction interval, including: Preprocessing the time series data set to extract the dynamic coupling relationship between the rotational speed, current, and material density under different coal feeding rate command values, wherein the dynamic coupling relationship reflects the coordinated fluctuation characteristics of the parameters as the coal feeding rate command value changes; The dynamic coupling relationship is input into a machine learning training framework, and a functional mapping relationship between the coal feed rate command value and the speed, current and material density is constructed through multivariate regression analysis, wherein each functional mapping relationship outputs a predicted reference value and a confidence interval of the corresponding parameter; The function mapping relationship is integrated to generate a dynamic mapping function. The dynamic mapping function takes the coal feeding rate instruction value as input and outputs the parameter prediction interval of the rotational speed, current and material density. The parameter prediction interval is composed of the prediction reference value plus or minus the real-time floating deviation.

[0008] Furthermore, before integrating the function mapping relationship to generate a dynamic mapping function, the method further includes: Analyze the historical fluctuation ranges of the speed, current, and material density corresponding to each of the coal feed rate command values ​​in the time series data set, where the historical fluctuation ranges reflect the natural deviation range of the parameters under stable operating conditions; Calculating the parameter floating boundary under the confidence probability based on the historical fluctuation range, wherein the confidence probability represents the statistical reliability of the parameter measured value falling within the prediction interval; Combined with the working condition characteristics corresponding to the current coal feeding rate instruction value, the parameter floating boundary is dynamically scaled to generate a real-time floating deviation. The real-time floating deviation changes with the coal feeding rate instruction value, and the width of the parameter prediction interval is adaptively adjusted.

[0009] Furthermore, the raw coal bunker includes a cylindrical coal bunker, a conical bucket coal bunker and a coal gate arranged in sequence from top to bottom, the conical bucket coal bunker is equipped with a first coal vibration assembly, and a plurality of second coal vibration assemblies are further arranged on the upper part of the first coal vibration assembly in the vertical direction; The vibrating assembly in the raw coal bin that starts supplying coal to the coal feeder comprises: When coal blockage is determined, a first start signal is sent to a first coal vibration component of the raw coal bin, the first coal vibration component being located outside the conical bucket-shaped coal bin; After the first coal vibrating assembly is started, a second start signal is sent to the second coal vibrating assembly after a preset delay, and the second coal vibrating assembly is vertically distributed on the upper part of the conical bucket-shaped coal bunker; The first coal vibration assembly and the second coal vibration assembly are controlled to continuously vibrate in coordination until the real-time value of the rotational speed, the real-time value of the current and the material density detection value return to the normal operating range.

[0010] Furthermore, the controlling the first coal vibration assembly and the second coal vibration assembly to continuously coordinate vibration until the real-time speed value, the real-time current value, and the material density detection value return to a normal operating range includes: Setting a layered vibration mode, causing the first coal vibration component to act on the coal flow channel at the lower part of the conical bucket-shaped coal bunker at a first vibration frequency, while causing at least one second coal vibration component to act on the coal shelf area at the upper part of the conical bucket-shaped coal bunker at a second vibration frequency, wherein the second vibration frequency is higher than the first vibration frequency to match the crushing characteristics of coal agglomerates; Real-time monitoring of the recovery rate of the real-time value of the rotational speed and the material density detection value relative to the normal operating range, and dynamically adjusting the amplitudes of the first vibration frequency and the second vibration frequency according to the recovery rate; When the real-time value of the rotational speed and the detected value of the material density remain within the normal operating range for a stable period of time, the rapping intensity of the first coal vibration assembly and the second coal vibration assembly is gradually reduced until it stops.

[0011] Furthermore, the real-time monitoring of the recovery rate of the real-time value of the rotational speed and the material density detection value relative to the normal operating range, and the dynamic adjustment of the amplitudes of the first vibration frequency and the second vibration frequency according to the recovery rate, includes: Calculating a weighted sum of the absolute value of the difference between the real-time speed value and the lower limit of the corresponding normal operating range and the absolute value of the difference between the material density detection value and the upper limit of the normal operating range, and defining the rate of change of the weighted sum as the comprehensive recovery rate; When the comprehensive recovery rate is lower than a preset recovery threshold, the amplitude of the second vibration frequency is increased proportionally while maintaining the amplitude of the first vibration frequency unchanged to preferentially break up the agglomerates in the upper coal shed area; When the comprehensive recovery rate is higher than the preset recovery threshold but the material density detection value does not reach the corresponding normal operating range, the amplitudes of the first vibration frequency and the second vibration frequency are simultaneously increased to enhance the unblocking of the entire coal flow; When the real-time value of the rotational speed and the detected value of the material density both enter the normal operating range, the current vibration frequency amplitude is maintained until a stable duration is reached.

[0012] Furthermore, the vibration intensity reduction process is divided into several progressive attenuation stages, each attenuation stage is maintained for a fixed duration and the intensity reduction decreases step by step, and the number of the progressive attenuation stages is positively correlated with the cumulative duration during which the parameters remain normal within the stable duration; After each level of attenuation is completed, the real-time value of the rotational speed and the material density detection value are detected to see whether they deviate from the normal operating range. If deviation occurs, the previous level of rapping intensity is immediately restored and the stabilization time is accumulated again; When the speed is attenuated to the lowest intensity gear and the real-time value of the rotational speed and the material density detection value remain normal, the second coal vibration component is turned off and the first coal vibration component is kept running at the basic intensity for the final recovery time; After the final recovery time is over and there is no abnormal fluctuation in the parameters, all coal vibration components are stopped.

[0013] Accordingly, a second aspect of an embodiment of the present invention provides a coal feeder blockage detection device, which performs coal feeder blockage detection based on the above-mentioned coal feeder blockage detection method, comprising: A data acquisition module is used to obtain the real-time speed value, current real-time value and material density detection value of the coal feeder in the current detection cycle; a data calculation module, which is used to obtain a time series data set of coal feed rate command values, speed measured values, current measured values, and material density measured values ​​under multiple operating conditions; based on the time series data set, generate a dynamic mapping function from the coal feed rate command value to the speed, current, and material density through machine learning algorithm training; and calculate the dynamic normal operating range of the speed, current, and material density through the dynamic mapping function in combination with the coal feed rate command value received by the coal feeder. The normal operating range is adaptively adjusted as the coal feed rate command value changes; The coal blockage judgment module is used to determine that the coal feeder is blocked by coal when the real-time value of the rotational speed, the real-time value of the current and the material density detection value deviate from the normal operating range corresponding to the coal feeding rate instruction value by more than a preset threshold value, and to start the vibration component in the raw coal bin that supplies coal to the coal feeder.

[0014] Accordingly, the third aspect of an embodiment of the present invention provides an electronic device comprising: at least one processor; and a memory connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the above-mentioned coal feeder blockage detection method.

[0015] Accordingly, a fourth aspect of an embodiment of the present invention provides a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the above-mentioned method for detecting coal blockage in a coal feeder.

[0016] The above technical solutions of the embodiments of the present invention have the following beneficial technical effects: 1. A dynamic mapping function trained based on historical time series data calculates in real time the adaptive normal operating ranges for speed, current, and material density under different coal feed rate commands, and dynamically adjusts the width of these ranges by introducing confidence probabilities. When a parameter deviates from this range by more than a threshold (e.g., a sudden drop in material density or an abnormal increase in speed), the system triggers an early warning at the initial stage of coal blockage accumulation. Compared to traditional passive monitoring that relies solely on outlet coal shut-off switches, this system accelerates response times by several minutes (e.g., identifying risks before the speed reaches full load), avoiding overload tripping accidents while also reducing false alarm rates. 2. Add multiple layers of vibration components to the upper part of the conical bunker. When a coal blockage is detected, the lower rapping device is activated first, followed by a delayed triggering of the upper high-frequency rapping, forming a blockage-clearing network covering the entire coal bunker. A layered rapping mode is used: the lower low-frequency rapping clears the coal flow channel, while the upper high-frequency rapping breaks up coal aggregates. The rapping amplitude is dynamically adjusted based on the parameter recovery rate. For example, when the recovery rate is low, the upper high-frequency rapping is prioritized to achieve targeted blockage clearing. This completely solves the blind spot above the coal gate in the original system, which lacked a rapping device. The blockage clearing efficiency is increased by over 50%, and there is no need to shut down the coal feeder. 3. Real-time monitoring of the rate at which the rotational speed and material density return to the normal range, and dynamic adjustment of the vibration intensity: When recovery is good, the vibration intensity is gradually reduced and eventually turned off; if the parameters rebound, the intensity is immediately restored; through a progressive attenuation strategy, the number of attenuation stages and the intensity reduction are intelligently matched according to the stabilization time to avoid energy waste and equipment wear caused by excessive vibration; at the same time, the basic vibration intensity is retained until the parameters are completely stable, reducing the risk of secondary coal blockage; this closed-loop control reduces vibration energy consumption by 30%, maintains the continuity of coal feeding, and guarantees the unit output. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1 This is a schematic diagram of the vibration of the raw coal bin provided by an embodiment of the present invention; Figure 2 This is a flow chart of a method for detecting coal blockage in a coal feeder provided by an embodiment of the present invention; Figure 3 This is a module block diagram of a coal feeder coal blockage detection device provided by an embodiment of the present invention.

[0018] Reference numerals: 01. Second coal vibration component, 02. Coal gate, 03. First coal vibration component, 1. Data acquisition module, 2. Data calculation module, 3. Coal blockage judgment module. DETAILED DESCRIPTION

[0019] To make the objectives, technical solutions, and advantages of the present invention more clearly understood, the present invention will be further described in detail below in conjunction with specific embodiments and with reference to the accompanying drawings. It should be understood that these descriptions are merely exemplary and are not intended to limit the scope of the present invention. In addition, in the following description, descriptions of well-known structures and technologies are omitted to avoid unnecessary confusion of the concepts of the present invention.

[0020] Please refer to Figure 1 and Figure 2 A first aspect of an embodiment of the present invention provides a method for detecting coal feeder blockage, comprising the following steps: Step S100, obtaining the real-time value of the rotation speed, the real-time value of the current and the material density detection value of the coal feeder in the current detection cycle.

[0021] A sensor system integrated into the coal feeder acquires key operating parameters in real time. The real-time speed value reflects the actual operating speed of the belt drive motor; the real-time current value indicates the load status of the coal feeder motor and indirectly correlates to the coal flow resistance; and the material density value, calculated by a weighing sensor and a volume measurement unit (coal feed rate = material density × belt speed × cross-sectional area), directly indicates the thickness uniformity of the coal layer on the belt. These parameters are synchronously collected at a fixed detection interval (e.g., once per second) and form the raw data foundation for coal blockage diagnosis.

[0022] Step S200, obtain a time series data set of coal feeding rate command values, speed measured values, current measured values ​​and material density measured values ​​under multiple working conditions, and generate a dynamic mapping function from the coal feeding rate command value to the speed, current and material density through machine learning algorithm training based on the time series data set. Combined with the coal feeding rate command value received by the coal feeder, the dynamic normal operating range of the speed, current and material density is calculated through the dynamic mapping function, and the normal operating range is adaptively adjusted as the coal feeding rate command value changes.

[0023] This model is generated through training using historical operating data: First, coal feed rate command values ​​and corresponding measured values ​​for speed, current, and material density under multiple operating conditions are collected. Machine learning algorithms (such as multivariate regression analysis) are then used to establish a dynamic mapping relationship between the command value and each parameter. The model output is not a fixed threshold, but rather a normal operating range that adaptively changes with the command value. For example, when the coal feed rate command is 40 t / h, the model calculates the normal speed range to be [600, 700] r / min and the normal material density range to be [9500, 11000] kg / m³ based on historical fluctuations. The width of these ranges also takes into account confidence levels (e.g., 95%) and is dynamically scaled based on operating condition characteristics to ensure that the ranges cover natural fluctuations under stable operating conditions.

[0024] Step S300: When the real-time speed value, the real-time current value and the material density detection value deviate from the normal operating range corresponding to the coal feeding rate instruction value and exceed the preset threshold, it is determined that the coal feeder is blocked by coal, and the vibration component in the raw coal bin that supplies coal to the coal feeder is started.

[0025] When any parameter (speed, current or material density) continuously deviates from its corresponding normal operating range and exceeds the preset threshold (such as the speed exceeds the upper limit by 10% or the density is lower than the lower limit by 15%), the system determines that coal blockage has occurred. The judgment logic prioritizes early abnormal characteristics: for example, a sudden drop in material density (thinning of the coal seam) is accompanied by an abnormal increase in speed (forced to increase the speed to maintain the coal feed). Although the traditional coal shortage alarm is not triggered at this time, it indicates that there is a risk of "bridging" at the coal bunker entrance. The vibration component is started immediately after the judgment: the layered vibration device is activated in sequence, first triggering the original vibration device at the bottom of the cone bucket, and then starting the newly added three-layer vibration device above the coal gate after a delay of 10 seconds, forming a coordinated vibration from bottom to top. The vibration continues until the parameters return to the normal range, and the vibration intensity is dynamically optimized through a graded adjustment strategy to avoid excessive vibration.

[0026] Furthermore, in step S200, based on the time series data set, a dynamic mapping function is generated from the coal feed rate command value to the speed, current and material density through machine learning algorithm training, wherein the mapping function output is a parameter prediction interval, including: Step S210 , preprocessing the time series data set to extract the dynamic coupling relationship among the rotational speed, current and material density under different coal feeding rate instruction values. The dynamic coupling relationship reflects the coordinated fluctuation characteristics of the parameters as the coal feeding rate instruction value changes.

[0027] Deep preprocessing of the time series dataset focused on exploring the coordinated response mechanisms between parameters. First, the data was grouped by coal feed rate command value, analyzing the linkage characteristics of speed, current, and material density when the command changed. Under normal operating conditions, an increase in the coal feed rate command causes a simultaneous increase in material density and speed (reflecting the coordination between coal seam thickening and conveying speed increases). However, under abnormal operating conditions (such as the initial stages of coal blockage), an anomalous coupling of a continuous decrease in material density and an abnormal increase in speed is observed (due to "bridging" causing the coal seam to thin, forcing the system to increase speed to compensate for the coal loss). The preprocessing process quantifies the intensity of the synchronization of parameter fluctuations within different command segments through sliding window statistics and correlation calculations, eliminating occasional noise interference, and ultimately extracting the dynamic coupling relationship matrix under stable operating conditions.

[0028] In step S220, the dynamic coupling relationship is input into the machine learning training framework, and a functional mapping relationship between the coal feed rate instruction value and the speed, current and material density is constructed through multivariate regression analysis, where each functional mapping relationship is output as a predicted reference value and confidence interval of the corresponding parameter.

[0029] The dynamic coupling relationship is input into a machine learning training framework, and a multivariate regression model is used to construct independent mappings between the coal feed rate command value and each parameter. Taking the speed as an example, the model takes the command value as input and, based on the co-fluctuation characteristics of the coupling relationship (e.g., the theoretical increase in speed as the command increases), outputs a predicted speed baseline value for that command (e.g., 40 T / H corresponds to 650 r / min). The model also calculates the dispersion of historical data for the same command (e.g., ±50 r / min) and generates a confidence interval (e.g., [600, 700] r / min) based on a confidence probability (e.g., 95%). A mapping relationship is generated simultaneously between current and material density. The confidence interval for material density specifically addresses the historical fluctuation range under high-humidity coal conditions (e.g., the interval width is expanded by 15% when the humidity exceeds 20%). The function mapping for each parameter maintains adaptive operating conditions, ensuring that the baseline value and interval are dynamically calibrated with coal quality characteristics.

[0030] Step S230: integrating the function mapping relationship to generate a dynamic mapping function. The dynamic mapping function takes the coal feed rate instruction value as input and outputs the parameter prediction interval of the rotation speed, current and material density. The parameter prediction interval is composed of the prediction reference value plus or minus the real-time floating deviation.

[0031] The independent function mappings for speed, current, and material density are integrated to construct a unified dynamic mapping function. This function uses the real-time coal feed rate command value as its sole input and simultaneously outputs prediction ranges for the three parameters. This range consists of two components: a predicted baseline value corresponding to the current command value (e.g., a speed of 650 r / min); and a real-time floating deviation (e.g., ±50 r / min). This deviation is not a fixed value but is dynamically calculated based on the current operating conditions. For example, under high-humidity coal conditions, the floating deviation of material density automatically extends to ±20% of the baseline value (compared to ±10% under dry coal conditions). The resulting parameter prediction range (e.g., [8500, 10500] kg / m³ for material density) reflects both the theoretical expected value of the parameter and a reasonable range of deviations under fluctuating operating conditions, forming an adaptive monitoring threshold.

[0032] Furthermore, before generating the dynamic mapping function from the integrated function mapping relationship in step S230, the following steps are further included: Step S231 , analyzing the historical fluctuation ranges of the speed, current and material density corresponding to each coal feed rate instruction value in the time series data set, where the historical fluctuation ranges reflect the natural deviation range of the parameters under stable working conditions.

[0033] For stable operating condition segments within the time series data set (e.g., parameter fluctuations with a standard deviation below a threshold for 10 consecutive minutes), historical fluctuation characteristics of speed, current, and material density are statistically analyzed by grouping by coal feed rate command value. This analysis includes calculating the dispersion of the measured parameter values ​​relative to their mean at each command value (e.g., material density fluctuates within ±850 kg / m³ at 40 T / H), associating them with operating condition labels, and specifically noting the fluctuation amplitude for high-moisture coal (moisture content >15%) (e.g., material density fluctuations expand to ±1200 kg / m³ under the same command). By fitting fluctuation distribution curves for different coal qualities and load ranges, the natural deviation range of the parameters in the absence of coal blockage is quantified, providing a statistical basis for setting floating boundaries.

[0034] Step S232 , calculating the parameter floating boundary under the confidence probability based on the historical fluctuation range, where the confidence probability represents the statistical reliability of the parameter measured value falling within the prediction interval.

[0035] Based on historical fluctuation data, probabilistic statistical methods are used to calculate parameter fluctuation limits at a confidence level (e.g., 95%). For example, for material density, historical data for high-humidity coal operating conditions with a 40 tonnes / hour (T / H) command follows a normal distribution N(9500,400²). Therefore, the fluctuation limit corresponding to a 95% confidence level is ±784 kg / m³ (or 1.96 standard deviations). This limit represents the statistical reliability of the measured parameter value falling within the prediction interval, and its width directly reflects the stability of the operating conditions. The wetter the coal and the more frequent the load changes, the wider the fluctuation limit. Current and speed are calculated simultaneously, with the speed fluctuation limit requiring additional consideration of baseline offset corrections caused by equipment aging.

[0036] In step S233, the parameter floating boundary is dynamically scaled to generate a real-time floating deviation in combination with the operating condition characteristics corresponding to the current coal feeding rate instruction value. The real-time floating deviation changes with the coal feeding rate instruction value, and the width of the parameter prediction interval is adaptively adjusted.

[0037] In actual operation, the system dynamically scales the floating margins calculated in step S223b based on the real-time operating conditions corresponding to the current coal feed rate command value (such as coal moisture content and equipment operating hours). When the coal moisture sensor detects a moisture content greater than 15%, the floating margin for material density automatically expands to 1.3 times the historical baseline value (for example, from ±784 kg / m³ to ±1019 kg / m³). If the equipment operates continuously for more than 100 hours, the floating margin for speed is narrowed to 0.8 times to offset abnormal fluctuations caused by mechanical wear. This scaled floating margin is the real-time floating deviation, which is adaptively adjusted based on the command value and operating conditions. Ultimately, it is used to construct the parameter prediction interval (e.g., predicted baseline ± real-time floating deviation).

[0038] The raw coal bin for feeding coal to the coal feeder in the present invention includes a cylindrical coal bin, a conical coal bin and a coal gate 02 arranged in sequence from top to bottom. The conical coal bin is equipped with a first coal vibration assembly, and a plurality of second coal vibration assemblies 01 are arranged vertically on the upper part of the first coal vibration assembly.

[0039] The raw coal bunker adopts a three-stage vertical layout: the top is a cylindrical bunker, which stores raw coal. Its vertical wall structure facilitates uniform distribution of coal particles. The middle section transitions to a conical bunker, where a tapered slope reduces the cross-sectional area of ​​the coal flow, guiding coal particles toward the coal gate 02. The bottom section connects to the coal gate 02 and serves as a channel for the coal flow to enter the coal feeder. To address areas prone to coal blockage, a first coal vibration assembly is installed on the outer wall of the conical bunker. Its vibration force acts on the coal flow channel below the cone, clearing any localized blockages. To further address the problem of shed coal within 1.5 meters above the coal gate 02, where coal particles easily agglomerate and form "bridges," several second coal vibration assemblies 01 are added vertically above the conical bunker. These assemblies are installed at the top, middle, and bottom sections of the cone (e.g., a three-tiered layout). The high-frequency vibration force directly impacts the coal agglomerates. The two sets of vibration components form spatial coordination: the lower vibration ensures the smooth flow of coal channels, and the upper vibration prevents the accumulation of coal in the shed, thus achieving clearing and coverage of the entire warehouse section.

[0040] The first coal vibrator assembly 03 is installed below the coal gate. Its mounting position, optimized through structural mechanics, ensures efficient transmission of the rapping force to the inner wall of the cone bucket and the coal mass within, directly impacting the core blockage area, where wet coal adhesion easily forms "bridging" or "shedding" formations. The first coal vibrator assembly 03 corresponds to the existing straight-pipe electric vibrator installed below the coal gate. However, in this solution, its functionality is further integrated and enhanced. It not only mechanically disrupts the liquid bridge and van der Waals forces between high-moisture coal particles, breaking up the agglomerated blockage structure, but also serves as the bottom-level actuator of the stratified rapping system, synergizing spatially and temporally with the second coal vibrator assembly installed above. Its vibration parameters (such as frequency and amplitude) are calibrated based on the cone bucket's structural characteristics and coal viscosity, ensuring that while clearing localized blockages, excessive vibration avoids coal flow collapse or structural resonance. As the primary vibrator actuator after coal blockage is detected, it performs the critical task of rapidly alleviating inlet flow resistance and creating conditions for coordinated vibration operations in the upper layers.

[0041] The segmented silo structure and three-dimensional layout of vibration components completely eliminate the blind spot problem of traditional raw coal silos. The lower vibrating assembly in the conical bucket maintains a smooth base coal flow, while the upper multi-layered vibration assembly precisely targets areas prone to coal blockage (particularly within 1.5 meters above the coal gate). High-frequency, targeted vibration quickly breaks up coal bridging. The spatial coordination of the layered vibration force significantly improves blockage clearing efficiency, avoids the lag and uncertainty of manual vibration, and ensures continuous and stable operation of the coal feeder.

[0042] The step S300 of starting the vibration assembly in the raw coal bin for supplying coal to the coal feeder includes: Step S310: When coal blockage is determined, a first start signal is sent to a first coal vibration component of the raw coal bin, where the first coal vibration component is located outside the cone-shaped coal bin.

[0043] When the system detects a coal blockage, it first sends a start command to the first coal vibrating assembly located outside the conical bunker. This assembly corresponds to the existing electric vibrating device installed on the straight pipe below the coal gate in the technical solution. It mechanically intervenes to address the initial coal blockage at the feeder inlet. Vibrating the plate directly acts on the key blockage points in the coal flow channel, forcibly breaking down the coal particle agglomeration caused by high humidity and alleviating flow resistance in the inlet area. This primary response action is intended to quickly eliminate the direct impact of local blockages on the conveyor coal feeder.

[0044] Step S320: After the first coal vibration component is started, a second start signal is sent to the second coal vibration component 01 after a preset delay. The second coal vibration component 01 is vertically distributed on the upper part of the cone-shaped coal bunker.

[0045] After ensuring the activation of the first coal vibrating assembly, a preset delay (for example, 10 seconds) is introduced before sending a start signal to the second coal vibrating assembly 01, located vertically above the conical bunker. This first coal vibrating assembly corresponds to the newly added three-layer rapping device. This delayed activation is based on the dynamic characteristics of coal flow: the initial relief of inlet blockage allows the "shelf coal" accumulated in the bunker's upper portion to sink. Re-activating the upper layer rapping at this point prevents coal flow shock and overload. This layered rapping strategy precisely matches the bunker's structural characteristics. The upper 1.5m area of ​​the conical bunker is a high-risk area for wet coal adhesion. By vertically covering different heights, the three-layer device systematically breaks down the arched blockage structure in the bunker's conical body.

[0046] Step S330, controlling the first coal vibration assembly and the second coal vibration assembly 01 to continuously coordinate vibration until the real-time value of the rotational speed, the real-time value of the current and the material density detection value return to the normal operating range.

[0047] The first and second vibrating components enter a continuous coordinated vibration phase, during which the DCS system monitors the feeder speed, current, and material density in real time. The system terminates vibration only when these three parameters return to the normal operating range established by the machine learning model (for example, when feeding 40 t / h, the speed should approach 650 rpm and the density should be approximately 10,256 kg / m³). This continuous coordinated mechanism overcomes the limitations of single vibration: the lower device maintains an unobstructed inlet, while the upper device continuously pulverizes the vertebral coal, forming a coordinated mechanical wave transmission process in both time and space until the coal flow parameters stabilize within the preset thresholds.

[0048] The collaborative vibration solution significantly improves the efficiency of coal blockage treatment through hierarchical vibration timing and spatial coverage. While retaining the original inlet vibration, a new vertebral vibration device is added to solve key blockage points. Combined with the intelligent start-stop control based on the data model, the coal blockage intervention node is advanced from the traditional coal outage alarm to the parameter abnormality stage. The risk of the coal feeder tripping due to speed overload is avoided, the need for manual vibration is reduced, and the coal blockage is quickly eliminated while maintaining the continuous operation of the unit. The entire mechanism relies on the transformation of the existing DCS architecture, ensuring the reliability of the coal-fired system while having the economical implementation of the project.

[0049] Furthermore, in step S330, controlling the first coal vibration assembly and the second coal vibration assembly 01 to continuously cooperate in vibration until the real-time value of the rotational speed, the real-time value of the current, and the material density detection value return to the normal operating range includes: Step S331, set the layered vibration mode, so that the first coal vibration component acts on the coal flow channel at the lower part of the conical bucket-shaped coal bin at a first vibration frequency, and at the same time, at least one second coal vibration component 01 acts on the upper shelf coal area of ​​the conical bucket-shaped coal bin at a second vibration frequency, wherein the second vibration frequency is higher than the first vibration frequency to match the crushing characteristics of the coal agglomerates.

[0050] During the coordinated vibration stage, the layered vibration strategy is initiated: the first coal vibration component acts on the coal flow channel at the bottom of the conical bucket-shaped coal bunker (corresponding to the position of the original electric vibration device) at a lower first vibration frequency. This frequency fully considers the continuity requirements of coal flow transportation to avoid coal flow collapse caused by excessive vibration; at the same time, at least one second coal vibration component 01 acts on the shed coal area at the top of the conical bucket-shaped coal bunker (corresponding to the newly added three-layer device) at a higher second vibration frequency. The high-frequency vibration is specifically aimed at the agglomerate structure formed by wet coal within 1.5m of the coal bunker cone. The high-frequency mechanical wave can effectively destroy the liquid bridge force and van der Waals force between coal particles. The frequency value is calibrated through coal viscosity experimental data to ensure accurate matching of the crushing threshold of coal agglomerates.

[0051] Step S332: monitor in real time the recovery rate of the real-time value of the rotational speed and the material density detection value relative to the normal operating range, and dynamically adjust the amplitudes of the first vibration frequency and the second vibration frequency according to the recovery rate.

[0052] Real-time monitoring of the recovery rate of the rotational speed and material density relative to the normal operating range (for example, the speed of the rotational speed approaching the target value of 650R / MIN from the abnormal value of 1450R / MIN) directly reflects the effectiveness of vibration in breaking the coal blockage structure. When the recovery rate is lower than expected, the amplitude of the second vibration frequency is automatically increased to strengthen the upper agglomerate crushing, while the amplitude of the first vibration frequency is moderately increased to promote the dredging of the coal flow channel; conversely, when the recovery rate is too fast, the vibration amplitude is reduced to avoid coal flow impact overload. This dynamic adjustment relies on the rate-vibration intensity mapping relationship constructed by the machine learning model to ensure that the vibration energy accurately matches the real-time coal blockage severity.

[0053] Step S333: When the real-time value of the rotational speed and the material density detection value remain within the normal operating range for a stable period of time, the vibration intensity of the first coal vibration assembly and the second coal vibration assembly 01 is gradually reduced until it stops.

[0054] When the speed and material density remain stable within the normal operating range (e.g., at 40 t / h, the speed remains at 650 ± 50 rpm and the density remains at 10,256 ± 500 kg / m³) for a preset stability period (usually longer than the mill system response period), the vibration intensity is gradually attenuated. First, the high-frequency vibration of the second vibration component is reduced to the maintenance frequency to eliminate the risk of recurrence of upper shed coal. Once the material density remains stable, the amplitude of the first vibration component is gradually reduced, and finally, the vibration is completely stopped. This staged exit prevents sudden changes in coal flow and ensures continuous coal feeding.

[0055] The hierarchical collaborative control mechanism accurately matches the coal blockage characteristics of different areas in the coal bunker through frequency differentiation, and achieves optimal distribution of vibration energy through dynamic adjustment based on parameter recovery rate. The segmented exit strategy minimizes mechanical loss and energy consumption while ensuring the complete elimination of coal blockage. It upgrades the vibration process from the traditional fixed intensity mode to an adaptive system that is linked to the coal flow status in real time, avoiding processing delays caused by insufficient vibration and preventing coal flow fluctuations caused by excessive vibration. The entire control logic is deeply integrated into the advanced DCS control algorithm, achieving a dual improvement in coal blockage control efficiency and equipment safety with zero human intervention.

[0056] Furthermore, in step S332, the recovery rate of the real-time value of the rotational speed and the material density detection value relative to the normal operating range is monitored in real time, and the amplitudes of the first vibration frequency and the second vibration frequency are dynamically adjusted according to the recovery rate, including: Step S3321, calculate the weighted sum of the absolute value of the difference between the real-time speed value and the lower limit of the normal operating range and the absolute value of the difference between the material density detection value and the upper limit of the normal operating range, and define the change rate of the weighted sum as the comprehensive recovery rate.

[0057] The system calculates in real time the absolute difference between the speed deviation from the lower limit of the normal operating range (for example, if the command is 40 t / h and the lower limit is 600 rpm, the actual value is 1450 rpm, resulting in a difference of 850) and the absolute difference between the material density deviation from the upper limit of the range (the upper limit is 10756 kg / m³, while the actual value is 4566 kg / m³, resulting in a difference of 6190). Using preset weighting factors (such as a speed weight of 0.6 and a density weight of 0.4), the weighted sum is calculated. The rate of change represents the comprehensive recovery rate. This accurately quantifies the coal blockage status: an abnormal increase in speed indicates insufficient belt load, while a sudden drop in density indicates a thinning coal seam. The weighted sum captures the dynamics of coal flow recovery.

[0058] Step S3322: When the comprehensive recovery rate is lower than the preset recovery threshold, the amplitude of the second vibration frequency is increased proportionally, while the amplitude of the first vibration frequency is maintained unchanged to preferentially break up the agglomerates in the upper coal shed area.

[0059] When the comprehensive recovery rate falls below a preset threshold (e.g., weighted sum reduction <100 units per minute), it indicates insufficient agglomerate removal in the upper shed area. At this point, the high-frequency amplitude of the second vibration component is proportionally increased (for example, from 50Hz to 65Hz), while maintaining the low-frequency amplitude of the first vibration component unchanged. This targeted increase in vibration energy is applied to the upper 1.5m of the cone bucket, increasing the mechanical impact force to break the arched structure formed by the wet coal while maintaining vibration intensity in the lower layer to prevent overload and disturbance in the coal flow channel.

[0060] Step S3323: When the comprehensive recovery rate is higher than the preset recovery threshold but the material density detection value does not reach the corresponding normal operating range, the amplitudes of the first vibration frequency and the second vibration frequency are simultaneously increased to enhance the overall coal flow.

[0061] If the overall recovery rate exceeds the threshold but the material density is below the normal range (e.g., rate > 100 units / minute and density < 9756 kg / m³), this indicates insufficient coal flow throughout the entire system. Simultaneously increasing the amplitude of the first and second vibration components (e.g., from 30 Hz to 40 Hz for the first component and from 50 Hz to 60 Hz for the second component) will accelerate the removal of coal blocks stuck to the cone and promote overall coal flow subsidence by enhancing the synergistic effect of the lower coal flow's thrust and the upper agglomeration-breaking force.

[0062] Step S3324: When the real-time value of the rotational speed and the material density detection value both enter the normal operating range, the current vibration frequency amplitude is maintained until a stable duration is reached.

[0063] When both the speed and material density enter the normal operating range (e.g., a stable speed of 620-680 rpm and a density of 9756-10756 kg / m³), maintain the current vibration amplitude until the preset stabilization time is reached (e.g., three mill cycles). This continuous vibration phase serves to consolidate the dredging effect, prevent coal particles from reaggregating due to residual moisture, and provide a buffer for subsequent rapping intensity reduction.

[0064] The dynamic adjustment mechanism quantitatively defines the coal flow recovery state, enabling precise, on-demand distribution of vibration energy. The top-priority strategy focuses on addressing areas with high coal blockage risk, the global enhancement mode addresses systemic flow obstructions, and steady-state maintenance ensures thorough treatment. The complete control logic upgrades the rapping process from being experience-driven to data-driven, significantly shortening the coal blockage treatment time while eliminating manual intervention. It also avoids equipment overload and secondary coal flow blockage caused by traditional fixed-intensity rapping, fundamentally improving the reliability of the pulverizing system's continuous operation.

[0065] Furthermore, in step S333, when the real-time value of the rotational speed and the material density detection value remain within the normal operating range for a stable period of time, the rapping intensity of the first coal vibration assembly and the second coal vibration assembly 01 is reduced in stages until it stops, including: Step S3331, divide the vibration intensity reduction process into several gradual attenuation stages, each attenuation stage maintains a fixed duration and the intensity reduction decreases step by step, and the number of gradual attenuation stages is positively correlated with the cumulative duration during which the parameters remain normal within the stable duration.

[0066] The vibration intensity attenuation is divided into multiple progressive stages (e.g., four-stage attenuation). Each stage lasts a fixed duration (e.g., 30 seconds), with the intensity reduction decreasing gradually (e.g., a 40% reduction in the first stage of the second component, followed by a 10% reduction in the final stage). The number of stages is dynamically correlated with the stability of the parameter: if the rotational speed and density remain normal for an extended period (e.g., over 5 minutes), indicating that the coal flow has stabilized, additional attenuation stages can be added to achieve a smooth exit. Conversely, if short-term stability is observed, the number of stages can be reduced to prevent premature cessation and the subsequent recurrence of coal blockage. Matching the reaggregation characteristics of coal particles can prevent a rebound in coal flow caused by a sudden drop in intensity.

[0067] Step S3332: After each level of attenuation phase ends, check whether the real-time value of the rotational speed and the material density detection value deviate from the normal operating range. If deviation occurs, immediately restore the previous level of vibration intensity and re-accumulate the stable time.

[0068] After each level of attenuation is completed, the system monitors the rotational speed and material density in real time to see if they deviate from their normal ranges (e.g., if the rotational speed suddenly increases to 700 rpm or the density drops below 9,000 kg / m³). If so, the vibration intensity is immediately restored to the previous level and the accumulated stabilization time is reset. This mechanism addresses the "unblocking-reblocking" risk unique to wet coal environments, ensuring rapid response and maintaining vibration energy above the critical elimination value until coal flow parameters stabilize.

[0069] Step S3333: When the intensity is attenuated to the lowest gear and the real-time value of the rotational speed and the material density detection value remain normal, the second coal vibration component 01 is turned off and the first coal vibration component is kept running at the basic intensity for the final recovery time.

[0070] When the intensity decays to the lowest level (e.g., 10% of the base intensity of the second component) and parameters remain normal, first shut down the upper second coal vibrating component 01. At this point, maintain the lower first component at base intensity (e.g., 30% of rated intensity) for the final recovery time (e.g., 2 minutes), using its continuous vibration to eliminate residual adhesion at the bottom of the coal bunker. This step is based on the principles of coal flow dynamics: after the upper layer of coal is broken, the bottom coal flow channel must maintain micro-vibration to prevent particle deposition and avoid high-frequency vibration that interferes with belt weighing accuracy.

[0071] Step S3334: After the final recovery time is over and there is no abnormal fluctuation in the parameters, stop the operation of all coal vibration components.

[0072] After the final recovery period has elapsed and there are no abnormal parameter fluctuations (e.g., speed fluctuations <±3%, density fluctuations <±5%), the system stops all rapping components. Before termination, a final data check is performed to ensure that the coal flow at the mill inlet has reached a "self-sustaining flow state," meaning that the material density and speed match the current coal feed command and the current value is below the full-load threshold, indicating that the risk of coal blockage has been completely eliminated.

[0073] The segmented attenuation mechanism completely solves the problem of coal flow fluctuation caused by the "sudden start and stop" of traditional vibration through progressive intensity adjustment and real-time feedback closed loop. The abnormal retreat strategy effectively addresses the risk of re-blocking in a wet coal environment, and the step-by-step exit prioritizes the dredging effect in key areas. The entire process minimizes mechanical wear and energy consumption while ensuring the complete elimination of coal blockage, transforming the vibration termination process into an intelligent decision-making process that is deeply coupled with the coal flow state. This prevents secondary blockage caused by premature exit and also avoids equipment life reduction caused by excessive vibration, ultimately achieving a seamless transition from fault intervention to smooth operation of the pulverizing system.

[0074] Accordingly, please refer to Figure 3 A second aspect of an embodiment of the present invention provides a coal feeder coal blockage detection device, which performs coal feeder coal blockage detection based on the above-mentioned coal feeder coal blockage detection method, comprising: Data acquisition module 1, which is used to obtain the real-time value of the speed, real-time value of the current and material density detection value of the coal feeder in the current detection cycle; Data calculation module 2 is used to obtain a time series data set of coal feed rate command values, speed measured values, current measured values, and material density measured values ​​under multiple working conditions; based on the time series data set, a dynamic mapping function is generated from the coal feed rate command value to the speed, current, and material density through machine learning algorithm training; and in combination with the coal feed rate command value received by the coal feeder, a dynamic normal operating range of the speed, current, and material density is calculated through the dynamic mapping function, and the normal operating range is adaptively adjusted as the coal feed rate command value changes; The coal blockage judgment module 3 is used to determine that the coal feeder is blocked by coal when the real-time value of the speed, the real-time value of the current and the material density detection value deviate from the normal operating range corresponding to the coal feeding rate instruction value by more than the preset threshold value, and to start the vibration component in the raw coal bin that supplies coal to the coal feeder.

[0075] Accordingly, the third aspect of an embodiment of the present invention provides an electronic device comprising: at least one processor; and a memory connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor so that the at least one processor executes the above-mentioned coal feeder blockage detection method.

[0076] Accordingly, a fourth aspect of an embodiment of the present invention provides a computer-readable storage medium having computer instructions stored thereon, which, when executed by a processor, implement the above-mentioned method for detecting coal blockage in a coal feeder.

[0077] The embodiments of the present invention are intended to provide a method and device for detecting coal blockage in a coal feeder, which has the following effects: 1. A dynamic mapping function trained based on historical time series data calculates in real time the adaptive normal operating ranges for speed, current, and material density under different coal feed rate commands, and dynamically adjusts the width of these ranges by introducing confidence probabilities. When a parameter deviates from this range by more than a threshold (e.g., a sudden drop in material density or an abnormal increase in speed), the system triggers an early warning at the initial stage of coal blockage accumulation. Compared to traditional passive monitoring that relies solely on outlet coal shut-off switches, this system accelerates response times by several minutes (e.g., identifying risks before the speed reaches full load), avoiding overload tripping accidents while also reducing false alarm rates. 2. Add multiple layers of vibration components to the upper part of the conical bunker. When a coal blockage is detected, the lower rapping device is activated first, followed by a delayed triggering of the upper high-frequency rapping, forming a blockage-clearing network covering the entire coal bunker. A layered rapping mode is used: the lower low-frequency rapping clears the coal flow channel, while the upper high-frequency rapping breaks up coal aggregates. The rapping amplitude is dynamically adjusted based on the parameter recovery rate. For example, when the recovery rate is low, the upper high-frequency rapping is prioritized to achieve targeted blockage clearing. This completely solves the blind spot above the coal gate in the original system, which lacked a rapping device. The blockage clearing efficiency is increased by over 50%, and there is no need to shut down the coal feeder. 3. Real-time monitoring of the rate at which the rotational speed and material density return to the normal range, and dynamic adjustment of the vibration intensity: When recovery is good, the vibration intensity is gradually reduced and eventually turned off; if the parameters rebound, the intensity is immediately restored; through a progressive attenuation strategy, the number of attenuation stages and the intensity reduction are intelligently matched according to the stabilization time to avoid energy waste and equipment wear caused by excessive vibration; at the same time, the basic vibration intensity is retained until the parameters are completely stable, reducing the risk of secondary coal blockage; this closed-loop control reduces vibration energy consumption by 30%, maintains the continuity of coal feeding, and guarantees the unit output.

[0078] Those skilled in the art will appreciate that the embodiments of the present application may be provided as methods, systems, or computer program products. Therefore, the present application may take the form of an entirely hardware embodiment, an entirely software embodiment, or an embodiment combining software and hardware. Furthermore, the present application may take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to magnetic disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0079] The present application is described with reference to the flowcharts and / or block diagrams of the methods, devices (systems), and computer program products according to the embodiments of the present application. It should be understood that each process and / or block in the flowchart and / or block diagram, as well as the combination of processes and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing device to produce a machine, so that the instructions executed by the processor of the computer or other programmable data processing device generate instructions for implementing the processes in the flowchart and / or block diagram. Figure 1 a process or multiple processes and / or boxes Figure 1 A device that provides the functions specified in a block or multiple blocks.

[0080] These computer program instructions may also be stored in a computer readable memory that can direct a computer or other programmable data processing device to work in a specific manner, so that the instructions stored in the computer readable memory produce an article of manufacture comprising an instruction device, which implements the process Figure 1 a process or multiple processes and / or boxes Figure 1 The function specified in one or more boxes.

[0081] These computer program instructions can also be loaded onto a computer or other programmable data processing device so that a series of operational steps are executed on the computer or other programmable device to produce a computer-implemented process, thereby providing the instructions executed on the computer or other programmable device for implementing the process. Figure 1 a process or multiple processes and / or boxes Figure 1 A step that specifies a function in one or more boxes.

[0082] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to the above embodiments, ordinary technicians in the field should understand that the specific implementation methods of the present invention can still be modified or replaced by equivalents. Any modification or equivalent replacement that does not depart from the spirit and scope of the present invention should be covered by the scope of protection of the claims of the present invention.

Claims

1. A method for detecting coal feeder blockage, characterized in that: The steps include: Obtain the real-time speed value, current value and material density detection value of the coal feeder in the current detection cycle; Obtaining a time series data set of coal feed rate command values, speed measured values, current measured values, and material density measured values ​​under multiple operating conditions; generating a dynamic mapping function from the coal feed rate command value to the speed, current, and material density through machine learning algorithm training based on the time series data set; calculating the dynamic normal operating range of the speed, current, and material density using the dynamic mapping function in combination with the coal feed rate command value received by the coal feeder; and adaptively adjusting the normal operating range as the coal feed rate command value changes; When the real-time value of the rotational speed, the real-time value of the current and the material density detection value deviate from the normal operating range corresponding to the coal feeding rate instruction value by more than a preset threshold, it is determined that the coal feeder is blocked by coal, and the vibration component in the raw coal bin that supplies coal to the coal feeder is started.

2. The method for detecting coal feeder blockage according to claim 1, characterized in that: The method of generating a dynamic mapping function from the coal feed rate command value to the speed, current, and material density through machine learning algorithm training based on the time series data set, wherein the mapping function output is a parameter prediction interval, including: Preprocessing the time series data set to extract the dynamic coupling relationship between the rotational speed, current, and material density under different coal feeding rate command values, wherein the dynamic coupling relationship reflects the coordinated fluctuation characteristics of the parameters as the coal feeding rate command value changes; The dynamic coupling relationship is input into a machine learning training framework, and a functional mapping relationship between the coal feed rate command value and the speed, current and material density is constructed through multivariate regression analysis, wherein each functional mapping relationship outputs a predicted reference value and a confidence interval of the corresponding parameter; The function mapping relationship is integrated to generate a dynamic mapping function. The dynamic mapping function takes the coal feeding rate instruction value as input and outputs the parameter prediction interval of the rotational speed, current and material density. The parameter prediction interval is composed of the prediction reference value plus or minus the real-time floating deviation.

3. The method for detecting coal feeder blockage according to claim 2, characterized in that: Before integrating the function mapping relationship to generate a dynamic mapping function, the method further includes: Analyze the historical fluctuation ranges of the speed, current, and material density corresponding to each of the coal feed rate command values ​​in the time series data set, where the historical fluctuation ranges reflect the natural deviation range of the parameters under stable operating conditions; Calculating the parameter floating boundary under the confidence probability based on the historical fluctuation range, wherein the confidence probability represents the statistical reliability of the parameter measured value falling within the prediction interval; Combined with the working condition characteristics corresponding to the current coal feeding rate instruction value, the parameter floating boundary is dynamically scaled to generate a real-time floating deviation. The real-time floating deviation changes with the coal feeding rate instruction value, and the width of the parameter prediction interval is adaptively adjusted.

4. The method for detecting coal blockage in a coal feeder according to any one of claims 1 to 3, characterized in that: The raw coal bunker includes a cylindrical coal bunker, a conical bucket coal bunker and a coal gate arranged in sequence from top to bottom. The conical bucket coal bunker is equipped with a first coal vibration assembly. A plurality of second coal vibration assemblies are further arranged on the upper part of the first coal vibration assembly in the vertical direction. The vibrating assembly in the raw coal bin that starts supplying coal to the coal feeder comprises: When coal blockage is determined, a first start signal is sent to a first coal vibration component of the raw coal bin, the first coal vibration component being located outside the conical bucket-shaped coal bin; After the first coal vibrating assembly is started, a second start signal is sent to the second coal vibrating assembly after a preset delay, and the second coal vibrating assembly is vertically distributed on the upper part of the conical bucket-shaped coal bunker; The first coal vibration assembly and the second coal vibration assembly are controlled to continuously vibrate in coordination until the real-time value of the rotational speed, the real-time value of the current and the material density detection value return to the normal operating range.

5. The method for detecting coal feeder blockage according to claim 4, characterized in that: The controlling the first coal vibration assembly and the second coal vibration assembly to continuously coordinate vibration until the real-time speed value, the real-time current value, and the material density detection value return to the normal operating range includes: Setting a layered vibration mode, causing the first coal vibration component to act on the coal flow channel at the lower part of the conical bucket-shaped coal bunker at a first vibration frequency, while causing at least one second coal vibration component to act on the coal shelf area at the upper part of the conical bucket-shaped coal bunker at a second vibration frequency, wherein the second vibration frequency is higher than the first vibration frequency to match the crushing characteristics of coal agglomerates; Real-time monitoring of the recovery rate of the real-time value of the rotational speed and the material density detection value relative to the normal operating range, and dynamically adjusting the amplitudes of the first vibration frequency and the second vibration frequency according to the recovery rate; When the real-time value of the rotational speed and the detected value of the material density remain within the normal operating range for a stable period of time, the rapping intensity of the first coal vibration assembly and the second coal vibration assembly is gradually reduced until it stops.

6. The method for detecting coal feeder blockage according to claim 5, characterized in that: The real-time monitoring of the recovery rate of the real-time value of the rotational speed and the material density detection value relative to the normal operating range, and dynamically adjusting the amplitudes of the first vibration frequency and the second vibration frequency according to the recovery rate, includes: Calculating a weighted sum of the absolute value of the difference between the real-time speed value and the lower limit of the corresponding normal operating range and the absolute value of the difference between the material density detection value and the upper limit of the normal operating range, and defining the rate of change of the weighted sum as the comprehensive recovery rate; When the comprehensive recovery rate is lower than a preset recovery threshold, the amplitude of the second vibration frequency is increased proportionally while maintaining the amplitude of the first vibration frequency unchanged to preferentially break up the agglomerates in the upper coal shed area; When the comprehensive recovery rate is higher than the preset recovery threshold but the material density detection value does not reach the corresponding normal operating range, the amplitudes of the first vibration frequency and the second vibration frequency are simultaneously increased to enhance the unblocking of the entire coal flow; When the real-time value of the rotational speed and the detected value of the material density both enter the normal operating range, the current vibration frequency amplitude is maintained until a stable duration is reached.

7. The method for detecting coal feeder blockage according to claim 5, characterized in that: When the real-time speed value and the material density detection value remain within the normal operating range for a stable period of time, the rapping intensity of the first coal vibration assembly and the second coal vibration assembly is gradually reduced until the rapping stops, including: The vibration intensity reduction process is divided into a number of gradual attenuation stages, each attenuation stage is maintained for a fixed time and the intensity reduction decreases step by step, and the number of the gradual attenuation stages is positively correlated with the cumulative time during which the parameters remain normal within the stable time period; After each level of attenuation is completed, the real-time value of the rotational speed and the material density detection value are detected to see whether they deviate from the normal operating range. If deviation occurs, the previous level of rapping intensity is immediately restored and the stabilization time is accumulated again; When the speed is attenuated to the lowest intensity gear and the real-time value of the rotational speed and the material density detection value remain normal, the second coal vibration component is turned off and the first coal vibration component is kept running at the basic intensity for the final recovery time; After the final recovery time is over and there is no abnormal fluctuation in the parameters, all coal vibration components are stopped.

8. A coal feeder blockage detection device, characterized in that: The method for detecting coal blockage of a coal feeder according to any one of claims 1 to 7 is used to detect coal blockage of a coal feeder, comprising: A data acquisition module is used to obtain the real-time speed value, current real-time value and material density detection value of the coal feeder in the current detection cycle; a data calculation module, which is used to obtain a time series data set of coal feed rate command values, speed measured values, current measured values, and material density measured values ​​under multiple operating conditions; based on the time series data set, generate a dynamic mapping function from the coal feed rate command value to the speed, current, and material density through machine learning algorithm training; and calculate the dynamic normal operating range of the speed, current, and material density through the dynamic mapping function in combination with the coal feed rate command value received by the coal feeder. The normal operating range is adaptively adjusted as the coal feed rate command value changes; The coal blockage judgment module is used to determine that the coal feeder is blocked by coal when the real-time value of the rotational speed, the real-time value of the current and the material density detection value deviate from the normal operating range corresponding to the coal feeding rate instruction value by more than a preset threshold value, and to start the vibration component in the raw coal bin that supplies coal to the coal feeder.

9. An electronic device, characterized in that: include: at least one processor; And a memory connected to the at least one processor; wherein the memory stores instructions that can be executed by the at least one processor, and the instructions are executed by the at least one processor to enable the at least one processor to execute the coal feeder blockage detection method as described in any one of claims 1-7.

10. A computer-readable storage medium, characterized in that Computer instructions are stored thereon, and when the instructions are executed by a processor, a method for detecting coal blockage in a coal feeder as described in any one of claims 1-7 is implemented.

Images