A Coal Bunker Intelligent Unblocking System and Method

Abstract

This invention belongs to the field of coal bunker unblocking technology, specifically involving an intelligent coal bunker unblocking system and method. By synchronously collecting current timing signals and coal flow timing signals, load deviation duration and flow suppression duration are constructed respectively. By analyzing the synergistic coupling relationship between the two in time, a multi-dimensional joint criterion is established to achieve collaborative verification and judgment of coal blockage status. This can minimize the misjudgment caused by relying solely on flow rate judgment and significantly improve the reliability of coal blockage identification. After determining that there is a risk of coal blockage, the unblocking head is first triggered to perform an unblocking action. During the unblocking head's extension, the pressure timing curve of the hydraulic main oil circuit is simultaneously collected to construct real-time status feedback of the unblocking execution process. Based on this status feedback, the unblocking effect is evaluated and subsequent control strategies are dynamically decided. This realizes a leap from open-loop execution to closed-loop control of perception-evaluation-regulation, which to a certain extent avoids blind repeated impacts and insufficient impacts.

Description

Technical Field

[0001] This invention belongs to the field of coal bunker unblocking technology, specifically relating to an intelligent coal bunker unblocking system and method. Background Technology

[0002] In thermal power plants, the raw coal bunker is a crucial buffer and storage facility connecting the coal conveying system and the pulverizing system. However, due to factors such as the high moisture content of the coal, uneven particle size distribution, and the bunker's geometric structure, blockages such as coal bridging, arching, and rodent holes can easily occur within the raw coal bunker, causing feed interruptions to the coal feeder and subsequently disrupting combustion. Traditional methods of clearing blockages, such as manual tapping and using bunker wall vibrators, are inefficient, have limited effectiveness, and pose safety risks.

[0003] To improve the automation level of unblocking operations, some intelligent unblocking solutions have been proposed in the existing technology. For example, Chinese invention patent with publication number CN115321029A discloses an intelligent unblocking device and equipment for raw coal bunkers in power plants. This solution determines whether coal blockage has occurred by setting up a coal flow detection unit for the coal feeder. When coal blockage is determined, an unblocking device with an adjustable angle and reciprocating impact is used to clear the blockage.

[0004] This solution represents an advancement from manual to automatic unblocking, but it has the following limitations in its implementation: 1. The existing solution mainly relies on downstream coal flow as the basis for judging coal blockage. This single criterion has limitations. When the coal flow fluctuates or decreases due to non-blocking factors, it is easy to cause misjudgment, which will affect the reliability of the diagnostic system. At the same time, it may frequently trigger ineffective unblocking, leading to increased fatigue damage to the hydraulic system and mechanical components.

[0005] 2. The existing solution activates the unblocking device for mechanical impact when coal blockage is detected. The entire unblocking process is carried out at a fixed frequency and number of times. The system cannot sense or evaluate the magnitude of the resistance encountered by the unblocking device during impact, whether the impact actually acts on the dense coal body, and the effect of a single impact. This results in a lack of status feedback in the entire unblocking process, which may lead to over-action on slightly brittle coal or under-action on hard arches, resulting in low overall unblocking efficiency and resource utilization efficiency. Summary of the Invention

[0006] In view of the shortcomings of the prior art described above, the purpose of this invention is to provide an intelligent coal bunker unblocking system and method to solve the problems existing in the prior art.

[0007] The objective of this invention can be achieved through the following technical solutions: The first aspect of this invention provides an intelligent coal bunker unblocking system, comprising the following modules: a coal blockage diagnosis module: based on the current timing signal of the coal feeding equipment and the downstream coal flow timing signal, generating a load deviation duration characterizing the continuous state of motor overload and a flow suppression duration characterizing the continuous state of poor material feeding, and determining coal blockage based on the synergistic relationship between the two.

[0008] Unblocking Trigger Module: When a risk of coal blockage is detected, an unblocking command is generated, which controls the hydraulic station to start and drives the mechanical hydraulic actuator to make the unblocking head perform an initial unblocking action that includes the stages of pushing out, holding and retracting.

[0009] Effect evaluation module: During the process of pushing out the blockage clearing head, hydraulic system pressure data is collected simultaneously to form a pressure time series curve, and pressure characteristics including peak pressure, pressure rise rate and high pressure platform duration are extracted from it.

[0010] The unblocking control module first verifies the unblocking effect by extracting pressure features. After successful verification, it performs template similarity matching to determine the unblocking effect scenario. Based on the determined unblocking restricted scenario or unblocking smooth scenario, it controls subsequent unblocking actions by dynamically changing the number of unblocking attempts.

[0011] The second aspect of the present invention provides a method for intelligent unblocking of coal bunkers, comprising the following steps: S1, generating a load deviation duration characterizing the continuous state of motor overload and a flow suppression duration characterizing the continuous state of poor material feeding based on the current timing signal of the coal feeding equipment and the downstream coal flow timing signal.

[0012] S2. Coal blockage is determined based on the synergistic relationship between load deviation duration and flow suppression duration.

[0013] S3. When a risk of coal blockage is detected, a blockage clearing command is generated, controlling the hydraulic station to start and drive the mechanical hydraulic actuator, so that the blockage clearing head performs an initial blockage clearing action including the push-out, holding and retraction phases.

[0014] S4. During the process of pushing out the blockage clearing head, the hydraulic system pressure data is collected simultaneously to form a pressure time series curve, and pressure characteristics including peak pressure, pressure rise rate and high pressure platform duration are extracted from it.

[0015] S5. First, verify the blockage-clearing effect of the extracted pressure features. After the verification is successful, perform template similarity matching to determine the blockage-clearing effect scenario.

[0016] S6. Based on the determined congestion clearing scenarios (limited or unimpeded), control subsequent congestion clearing actions by dynamically changing the number of clearing attempts.

[0017] Combining all the above technical solutions, the positive effects of this invention are as follows: 1. This invention simultaneously collects current timing signals and coal flow timing signals in the drive motor circuit of the coal feeding equipment and the downstream material metering point of the coal bunker, respectively constructs load deviation duration and flow suppression duration, and establishes a multi-dimensional joint criterion by analyzing the synergistic coupling relationship between the two in time, thereby realizing the synergistic verification and judgment of coal blockage status. This can minimize the misjudgment caused by relying solely on flow determination and significantly improve the reliability of coal blockage identification.

[0018] 2. Upon determining the risk of coal blockage, this invention first triggers the unblocking head to perform an unblocking action. During the unblocking head's extension, the pressure timing curve of the main hydraulic circuit is simultaneously collected, and pressure characteristics are extracted to construct real-time status feedback for the unblocking execution process. Based on this status feedback, the unblocking effect is evaluated, and subsequent control strategies are dynamically decided. This achieves a leap from open-loop execution to closed-loop control of perception-evaluation-regulation, which to a certain extent avoids blind repeated impacts and insufficient impacts, and helps to improve the overall unblocking efficiency and resource utilization efficiency. Attached Figure Description

[0019] The present invention will be further described with reference to the accompanying drawings, but the embodiments in the drawings do not constitute any limitation on the present invention. For those skilled in the art, other drawings can be obtained based on the following drawings without creative effort.

[0020] Figure 1 This is a schematic diagram of the module connection of an intelligent coal bunker unblocking system according to the present invention.

[0021] Figure 2 This is a schematic diagram of the unblocking device in this invention.

[0022] Figure 3 This diagram illustrates the implementation steps of an intelligent coal bunker unblocking method according to the present invention. Detailed Implementation

[0023] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0024] Example 1

[0025] This invention provides an intelligent coal bunker unblocking system, including a coal blockage diagnosis module, an unblocking trigger module, an effect evaluation module, and an unblocking control module.

[0026] Please see Figure 1As shown, through the orderly collaboration of perception, execution, and feedback control, the various functional modules construct a closed-loop congestion clearing data chain with multi-source state perception as input, effect evaluation as the core, and adaptive control as the output.

[0027] The coal blockage diagnosis module is used to generate a load deviation duration characterizing the continuous state of motor overload and a flow suppression duration characterizing the continuous state of poor material feeding based on the current timing signal of the coal feeding equipment and the downstream coal flow timing signal, and to determine coal blockage based on the synergistic relationship between the two.

[0028] In a coal bunker system, raw coal is discharged from the bottom of the bunker by gravity and then quantitatively conveyed to the downstream pulverizing or combustion system via feeding equipment such as vibrating feeders, belt feeders, or screw feeders. This feeding equipment is driven by an electric motor, and its operating current directly reflects the mechanical load on the equipment. Simultaneously, downstream conveyor belts are typically equipped with coal flow metering devices such as electronic belt scales, nuclear scales, or microwave mass flow meters to monitor the material conveying rate in real time.

[0029] Under normal operating conditions, there is a stable mapping relationship between the motor load current and the output coal flow rate of the coal feeding equipment: the greater the amount of coal conveyed per unit time, the greater the resistance that the motor needs to overcome, and the higher the current; conversely, the current decreases. This electromechanical-material coupling characteristic constitutes the inherent consistency basis of the system's operating state.

[0030] When a blockage occurs inside the coal bunker, although the coal feeding equipment is still powered on, its actual operating conditions deviate significantly: For vibrating coal feeders: the interruption of coal flow causes the equipment to be in a near-no-load state, and the motor current is lower than the normal value.

[0031] For screw or belt feeders: if the outlet is blocked by dense coal blocks, the motor may enter an overload state due to stall, and the current will rise abnormally.

[0032] At the same time, the blockage directly caused a significant decrease or even zero in downstream coal flow.

[0033] Therefore, the abnormal deviation of the current signal and the unexpected decay of the coal flow are synergistic in time, and the two constitute an early and quantifiable sign of coal blockage.

[0034] Based on the above explanation, coal blockage can be determined in the coal bunker by coordinating the current timing signal of the coal feeding equipment and the downstream coal flow timing signal.

[0035] In the specific implementation of the above modules, the acquisition of current timing signals and coal flow timing signals includes the following: the motor operating current signal is acquired in real time by a current sensor deployed on the drive motor of the coal feeding equipment to form a current timing signal.

[0036] A coal flow metering device installed downstream of the coal bunker outlet is used to collect instantaneous coal flow signals in real time, forming a coal flow time sequence signal.

[0037] The acquired current time-series signal and coal flow time-series signal are timestamped to ensure that the current signal and coal flow signal are aligned under a unified time reference.

[0038] Furthermore, given that coal blockage is a gradual development process rather than an instantaneous event, when identifying load deviation based on current time-series signals and flow suppression based on coal flow time-series signals, coal blockage cannot be directly determined based solely on amplitude anomalies at a single moment. This invention introduces a time dimension as a criterion, where the duration of the anomaly reflects the stability of the deviation state. Brief anomalies may originate from operating condition disturbances or measurement noise, while persistent anomalies indicate that the material flow has entered a stage of substantial obstruction, possessing the physical characteristics of coal blockage.

[0039] Based on the above considerations, the specific generation process of load deviation duration and flow suppression duration is as follows: the real-time acquired current timing signal is filtered to suppress high-frequency noise and transient interference.

[0040] The processed current timing signal is compared with the allowable fluctuation range of the reference current, which is the rated current of the equipment.

[0041] If the current exceeds the allowable fluctuation range of the reference current at a certain moment, a load deviation is identified at that moment, and the duration from the occurrence of the deviation is recorded as the load deviation duration.

[0042] It should be noted that the rated current is the design current value for the motor to operate safely for a long period of time under rated load, rated voltage, and rated speed, and is clearly given on the equipment nameplate or technical manual.

[0043] When the coal feeding equipment is delivering coal at a stable rate, its actual operating current should fluctuate slightly around the rated current. This allowable fluctuation range is provided by the equipment manufacturer's technical specifications and is usually ±10%, reflecting the reasonable disturbance boundary of the current under normal operating conditions.

[0044] The real-time coal flow time-series signal is filtered.

[0045] The filtered coal flow signal is compared with a preset steady-state flow range, which is the statistical distribution range of coal flow during the steady-state operation period when the equipment is confirmed to have no coal blockage under the same operating conditions.

[0046] Since coal flow rate is affected by various factors such as feeder opening, coal quality, and silo pressure, there is no fixed rated flow rate. Therefore, a steady-state flow range must be constructed based on historical steady-state data under the same operating conditions where no coal blockage has occurred in order to truly reflect the current expected flow rate.

[0047] The same operating conditions mentioned above refer to the same coal feeder control commands, the same unit load range, and similar environmental conditions such as season and coal type / batch.

[0048] In the example implementation, the steady-state flow range is obtained as follows: In the historical coal flow time series data that matches the current operating conditions, after removing all known coal blockage, clearing actions and transition processes, only the coal flow sequence of the fault-free steady-state operation period is retained, and the mean and standard deviation are calculated based on the sequence. The interval formed by the mean ± 2 times the standard deviation is taken as the steady-state flow range.

[0049] If the coal flow rate at a certain moment is lower than the lower limit of the steady-state flow rate range, then flow suppression is identified at that moment, and the duration from the start of flow suppression is recorded as the flow suppression duration.

[0050] In a preferred embodiment of the present invention, the determination of coal blockage is made by applying the load deviation duration and the flow suppression duration, as described below: when only load deviation is detected, the coal flow suppression signal is detected synchronously within the time window of load deviation. When flow suppression is detected, it is determined that there is a risk of coal blockage; otherwise, the monitoring status is maintained.

[0051] If only flow suppression is detected, load deviation signals will be detected synchronously within the time window of flow suppression. If load deviation is detected, it is determined that there is a risk of coal blockage; otherwise, the monitoring status will be maintained.

[0052] Understandably, when only one of load deviation and flow suppression is identified, the possibility of misjudgment due to a single signal should be considered. Given that coal blockage is a gradual and cumulative process, if blockage does occur, it can trigger observable anomalous responses in both. Therefore, after either anomalous signal is identified, the other signal must be simultaneously verified within its duration window.

[0053] If another type of abnormal signal is detected within this window, the two constitute a spatiotemporal collaborative evidence chain of multi-source heterogeneous signals, which has high causal consistency and process stability, and can be determined with high confidence to be a real coal blockage event.

[0054] If no other type of abnormal signal is detected, it indicates that the reliability of the current single signal is low. In this case, the monitoring status should be maintained and the evolution trend of subsequent signals should be continuously tracked in order to avoid misjudgment caused by isolated anomalies.

[0055] Of course, if only a single signal anomaly is detected in multiple diagnoses, and no coordinated response of another type of signal is observed within its duration window, it may be due to a sensor malfunction. In this case, an alarm should be triggered and manual intervention required.

[0056] When both load deviation and flow suppression are detected simultaneously, the duration of both load deviation and flow suppression is compared with the judgment time limit. If both reach the judgment time limit, it is determined that there is a risk of coal blockage. If the duration of either abnormal state does not reach the judgment time limit, the monitoring status is maintained.

[0057] The aforementioned judgment time limit should be set based on the typical time scale of coal blockage failure development. Specifically, according to on-site operation observation and historical event statistics, when blockages such as coal bridging or arching occur in the coal bunker, the resulting load deviation and flow suppression anomalies usually have continuous characteristics, generally lasting from several seconds to tens of seconds. In order to effectively distinguish between real blockages and short-term process fluctuations or measurement noise, the judgment time limit is set to 10 seconds.

[0058] It is understandable that when load deviation and flow suppression are detected simultaneously, it indicates that the two heterogeneous signals exhibit a co-occurrence characteristic in time. In order to eliminate instantaneous interference, the duration of the abnormal state of the two is jointly verified. Only when both reach or exceed the corresponding time limit can the co-occurrence phenomenon be confirmed to have process stability, and thus be judged as a real coal blockage event with high confidence.

[0059] This invention fully considers that coal blockage is not an isolated event caused by a single abnormal sensor signal, but rather a comprehensive manifestation of a physical imbalance in the coupling between the electromechanical system and the material conveying system. Therefore, by fusing current timing signals and coal flow timing signals, a logical diagnostic mechanism is constructed that mutually verifies coal flow suppression and load deviation, achieving high-precision, low-false-alarm identification of real coal blockage events.

[0060] When the unblocking trigger module determines that there is a risk of coal blockage, it generates an unblocking command. The electrical control cabinet controls the hydraulic station to start and drive the mechanical hydraulic actuator, so that the unblocking head performs an initial unblocking action that includes the stages of pushing out, holding and retracting.

[0061] See Figure 2 As shown, the unblocking device provided by the present invention consists of three main parts: a mechanical hydraulic actuator, a hydraulic station, and an electrical control cabinet.

[0062] As the end-efficiency actuator for unblocking operations, the mechanical hydraulic actuator acts directly on the blocked area of ​​the coal bunker to complete the physical impact and pushing operations. Its core components include the unblocking head and the hydraulic cylinder.

[0063] The unblocking head is a special impact pushing component installed at the front end of the actuator. Its function is to extend into the coal bunker under the drive of the hydraulic cylinder and apply directional mechanical force to blockages such as coal bridging and arching, so as to break, disturb or push them through.

[0064] The hydraulic cylinder, as a power conversion element, converts the high-pressure hydraulic energy provided by the hydraulic station into linear reciprocating motion, driving the unblocking head to complete the complete unblocking cycle of pushing out, holding, and retracting.

[0065] As the power source and energy control unit of the system, the main function of the hydraulic station is to provide a stable and controllable high-pressure hydraulic oil flow to drive the cylinder.

[0066] The electrical control cabinet, as the control center of the system, is responsible for controlling the on / off state of the solenoid valves of the hydraulic station and precisely scheduling the timing of the blockage clearing action.

[0067] The effect evaluation module simultaneously collects hydraulic system pressure data during the unblocking head ejection process, forms a pressure time series curve, and extracts pressure characteristics including peak pressure, pressure rise rate, and high-pressure platform duration.

[0068] It should be noted that the pushing action of the unblocking head is essentially a work process in which the hydraulic cylinder, driven by high pressure, overcomes the external resistance exerted by the blockage in the coal bunker. During this process, the hydraulic station provides the necessary hydraulic fluid to the cylinder, and the working pressure in its main oil circuit changes dynamically with the load resistance: when the unblocking head contacts and pushes against the dense coal body, the resistance increases, causing the pressure in the main oil circuit to rise; conversely, if no effective resistance is encountered, such as empty pushing or smooth coal flow, the pressure remains at a lower level.

[0069] To monitor this mechanical interaction process in real time, a pressure sensor can be integrated into the main oil supply line of the hydraulic station to continuously collect the pressure response signal of the hydraulic circuit and generate a pressure time-series curve. This curve fully records the dynamic evolution characteristics of resistance during the unblocking process, constituting direct physical feedback on the unblocking execution status.

[0070] Given that the mechanical properties of the blockage, such as strength, density, and structural stability, act on the hydraulic actuator in the form of resistance and leave characteristics in the pressure response, the pressure characteristics in the pressure time-series curve can be used to evaluate the unblocking effect.

[0071] As one way to achieve the above scheme, extracting pressure features including peak pressure, pressure rise rate, and high-pressure plateau duration from the pressure time series curve specifically includes the following: extracting the global maximum pressure value of the pressure time series curve as the peak pressure.

[0072] Numerical differentiation is performed on the pressure time series curve to obtain the first derivative sequence. The starting time of pressure rise is determined by the first derivative being greater than zero and continuously rising. The time when the first derivative first reaches a local maximum is defined as the ending time of pressure rise.

[0073] The pressure rise phase is defined by the start and end of the pressure rise. The ratio of the pressure increment to the time interval during this phase is used as the pressure rise rate.

[0074] The duration for which the pressure value remains at the peak level after reaching the peak pressure in the statistical pressure time series curve is taken as the high-pressure plateau duration.

[0075] It should be noted that the pressure rise rate, peak pressure, and high-pressure platform duration extracted from the pressure time series curve correspond to the mechanical response of different physical stages in the unblocking process. Specifically, the physical stages are as follows: Initial resistance breakthrough stage: The pressure rises rapidly, corresponding to the initial rupture of the coal arch bridge or the wall-adhering structure. The pressure rise rate in this stage reflects the dynamic loading capacity of the hydraulic actuator to break through the initial blockage resistance. The higher the rate, the faster the system response.

[0076] Peak load stage: The pressure reaches its maximum value, i.e., the peak pressure, which corresponds to the maximum instantaneous load borne by the clearing head during the interaction with the blockage. This value, to a certain extent, characterizes the ultimate compressive strength of the blockage.

[0077] Stable pushing phase: If the blockage is not broken through in one go, the clearing head will continue to apply pushing force, and the system pressure will remain relatively stable at a high level for a period of time, forming a high-pressure plateau. The duration of the high-pressure plateau reflects the length of time the clearing head has a continuous and effective effect on the blockage. A long duration indicates that the blockage is dense and requires continuous force, while a very short duration or no duration may be a momentary disturbance or ineffective clearing. It is a key criterion for determining whether the clearing action is truly and effectively acting on the blockage.

[0078] These three parameters together constitute a blockage mechanics characteristic vector, which can characterize the material resistance characteristics encountered in a single unblocking operation.

[0079] After extracting pressure characteristics from the pressure time-series curve during the blockage clearing process, the blockage clearing effect scenario can be determined based on these characteristics, and a closed-loop feedback control strategy can be constructed to achieve adaptive regulation of the blockage clearing action.

[0080] The blockage clearing control module is used to first verify the blockage clearing effect by extracting pressure features. After the verification is successful, template similarity matching is performed to determine the blockage clearing effect scenario. Based on the determined blockage clearing restricted scenario or blockage clearing unobstructed scenario, the subsequent blockage clearing actions are controlled by dynamically changing the number of blockage clearing operations.

[0081] Considering that the clearing action may not actually act on the blockage due to reasons such as insufficient travel or failure of the actuator to be effectively triggered, if the clearing effect scenario is directly judged based on pressure characteristics, it is easy to misjudge invalid actions as specific clearing scenarios, resulting in inappropriate control strategies.

[0082] Therefore, the present invention adopts a two-stage evaluation mechanism. First, the duration of the high-pressure platform is used to verify whether the clearing action actually acts on the blockage. Only when the effective effect is confirmed will the clearing effect scenario be further determined.

[0083] Applying the above description, the unblocking effect verification process is as follows: the extracted high-pressure platform duration is compared with the set minimum action duration threshold. If the minimum action duration threshold is reached, the unblocking effect scenario is directly determined. Otherwise, it is considered that the unblocking action has not effectively acted on the blockage. Then, the mechanical hydraulic actuator is re-driven to control the unblocking head to perform a new round of unblocking action, and a new pressure time curve is collected simultaneously.

[0084] Extract the high-pressure platform duration from the newly acquired pressure time-series curve again, and repeat the above discrimination logic until any of the following termination conditions are met: a) The high-pressure platform duration reaches the minimum action time threshold, and enter the blockage clearing effect scenario judgment.

[0085] b) If the number of retries reaches the limit, such as 2 or 3 times, it is determined that the blockage clearing is abnormal, triggering an alarm, and manual intervention can be initiated.

[0086] In a specific embodiment of the above scheme, the minimum action duration threshold represents the lower limit of the time required for the clearing head and the blockage body to form effective mechanical contact and continuously apply force.

[0087] When setting this threshold, a theoretical lower limit can be calculated based on the clearing stroke and speed. Specifically, the operation is as follows: Let the total stroke of the clearing head be... The launch speed is The entire launch time is approximately Based on engineering practice, effective unblocking typically requires the high-pressure platform to operate for at least 30% of the deployment time to ensure the unblocking head applies sufficient and continuous force to the blocked area. The minimum application time threshold at this point can be [value missing]. .

[0088] Furthermore, the scenario for determining the effectiveness of unblocking is as follows: under the same operating conditions, historical unblocking event samples that have been verified by downstream coal flow recovery and marked as unblocked are collected, and the statistical mean of pressure characteristics is calculated as the benchmark value for each historical unblocking event sample.

[0089] The peak pressure, pressure rise rate, and high-pressure plateau duration extracted from a single effective blockage clearing action are normalized to their corresponding baseline values, forming a three-dimensional normalized feature vector, specifically represented as follows: ,in , , These represent the peak pressure, pressure rise rate, and high-pressure plateau duration extracted from a single effective unblocking action, respectively. , , These represent the baseline values ​​for peak pressure, pressure rise rate, and high-pressure plateau duration, respectively.

[0090] The mean of the normalized feature vectors of the original pressure characteristics of all historical congestion clearing events are calculated in the same way to construct the congestion clearing template vector.

[0091] Since the samples used to construct the congestion clearing and smooth flow template vector are derived from the smooth flow condition samples, the direction of the vector represents the typical pressure characteristic ratio relationship during congestion clearing and smooth flow.

[0092] Calculate the cosine similarity between the 3D normalized feature vector and the congestion clearing template vector.

[0093] It is important to clarify that cosine similarity, as a directional metric, depends only on the relative proportions between the dimensions of the feature vector and is insensitive to overall amplitude scaling. Therefore, even if fluctuations in the hydraulic system's operating conditions cause a systematic shift in the absolute amplitude of the pressure characteristics during the unblocking operation, as long as the intrinsic proportional relationship between peak pressure, pressure rise rate, and high-pressure platform duration remains consistent with historical unblocked samples, cosine similarity can still provide a high similarity score. This demonstrates that cosine similarity aligns with the physical essence of unblocking effectiveness assessment.

[0094] The calculated cosine similarity is compared with the threshold value for determining unobstructed flow. When the threshold value is reached or exceeded, it is determined to be a scenario where the blockage is cleared and the flow is unobstructed. Otherwise, it is determined to be a scenario where the blockage is restricted, meaning that the blockage has been acted upon but has not been effectively cleared.

[0095] The aforementioned threshold for determining smooth flow is the decision boundary that distinguishes between two scenarios: smooth flow after clearing blockages and restricted flow after clearing blockages. It can be set using the statistical distribution of historical labeled samples, as follows: First, construct a validation dataset: collect a sufficient number of historical clearing events under the same working conditions. Each case is labeled as follows after downstream coal flow recovery: Smooth flow: coal flow recovers to the steady-state flow range after clearing blockages.

[0096] Restricted category: Insufficient traffic recovery or multiple congestion clearing operations required.

[0097] Subsequently, for each labeled sample, the cosine similarity between its three-dimensional normalized pressure feature vector and the template vector of the unblocking and smooth flow scenario is calculated, thereby obtaining the empirical distribution of cosine similarity between the unblocking and restricted samples respectively. It can usually be seen that the unblocking class is concentrated in the high value range, while the restricted class is distributed in the low value range.

[0098] Based on this, the lowest valley position between the two types of distributions is selected as the critical value for determining smooth flow, which can maximize the separation of the two types of scenarios.

[0099] It is necessary to add to the above scheme that the reason for choosing to construct a clearing unobstructed template vector instead of a clearing restricted template vector when determining the clearing scenario is that: under the same working conditions, once the blockage is effectively cleared and the coal flow is restored to smooth flow, the resistance of the clearing head is small and the process can be repeated. Its hydraulic response, such as the pressure rise rate, peak pressure, and high pressure platform duration, shows a stable, concentrated, and reproducible characteristic pattern, which is easy to form a highly cohesive template vector through historical sample clustering.

[0100] The constraints encompass varying degrees of physical states, resulting in pressure characteristics that are diffusely distributed across a multidimensional space, with blurred boundaries and a lack of typicality. Forcibly constructing a constrained template would lead to poor template representativeness, making it unsuitable as a matching target.

[0101] Furthermore, depending on whether the blockage is restricted or unobstructed, the subsequent blockage-clearing actions are controlled by dynamically changing the number of clearing cycles as follows: When the blockage is restricted, a repeated impact command is generated, and the blockage-clearing head is controlled to automatically perform an additional push-and-retract impact cycle a preset number of times after completing the initial blockage-clearing action, so as to enhance the breaking and disturbance effect on dense blockages.

[0102] After each additional impact action is completed, the pressure time-series curve is collected synchronously, the pressure features are re-extracted, a three-dimensional normalized feature vector is constructed, and its cosine similarity with the template vector of the blockage clearing scenario is calculated again until the cosine similarity reaches the threshold for determining unobstructed flow or the number of additional impacts reaches the upper limit, at which point the impact is terminated.

[0103] The maximum number of additional impacts under the above conditions is to prevent equipment overload.

[0104] When the system is determined to be in a clearing and unblocking scenario, a return-to-standby command is generated, which controls the clearing head to directly enter the safe return-to-standby process after completing the initial clearing action.

[0105] In the innovative implementation of this invention, the initial unblocking action of the unblocking head also includes safety monitoring, including the following safety monitoring: the two ends of the cylinder of the mechanical hydraulic actuator are equipped with stroke detection sensors, including a forward positioning sensor and a retraction positioning sensor.

[0106] The forward positioning sensor is used to detect whether the clearing head has moved to the preset maximum push-out position.

[0107] The return-to-position sensor is used to detect whether the unblocking head has fully returned to its initial safe position.

[0108] During the process of the unblocking head executing the push command, the forward extension sensor monitors the end point of the stroke limit. When the sensor reports the forward extension status, the hydraulic output is terminated to prevent mechanical impact or equipment damage caused by overtravel.

[0109] After the clearing head executes the retraction command, the retraction return sensor confirms the return status. When the sensor reports the return status, it is considered that the single action cycle has been safely completed, and the system is allowed to enter the safe return or standby state.

[0110] This invention utilizes a dual position feedback mechanism, constructed by configuring forward exit and retraction stroke detection sensors at both ends of the hydraulic cylinder, to not only actively protect the movement limits of the unclogging head but also ensure the integrity of the unclogging process and the controllability of the system state.

[0111] Example 2

[0112] See Figure 3 As shown, the present invention proposes an intelligent unblocking method for coal bunkers, including the following steps: S1, generating a load deviation duration characterizing the continuous state of motor overload and a flow suppression duration characterizing the continuous state of poor material feeding based on the current timing signal of the coal feeding equipment and the downstream coal flow timing signal.

[0113] S2. Coal blockage is determined based on the synergistic relationship between load deviation duration and flow suppression duration.

[0114] S3. When a risk of coal blockage is detected, a blockage clearing command is generated, controlling the hydraulic station to start and drive the mechanical hydraulic actuator, so that the blockage clearing head performs an initial blockage clearing action including the push-out, holding and retraction phases.

[0115] S4. During the process of pushing out the blockage clearing head, the hydraulic system pressure data is collected simultaneously to form a pressure time series curve, and pressure characteristics including peak pressure, pressure rise rate and high pressure platform duration are extracted from it.

[0116] S5. First, verify the blockage-clearing effect of the extracted pressure features. After the verification is successful, perform template similarity matching to determine the blockage-clearing effect scenario.

[0117] S6. Based on the determined congestion clearing scenarios (limited or unimpeded), control subsequent congestion clearing actions by dynamically changing the number of clearing attempts.

[0118] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented using software, the above embodiments can be implemented, in whole or in part, in the form of a computer program product.

[0119] Those skilled in the art will recognize that the modules and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0120] In addition, the functional modules in the various embodiments of this application can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0121] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any variations or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application. Therefore, the scope of protection of this application should be determined by the scope of the claims.

[0122] Finally, the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A coal bunker intelligent unblocking system, characterized in that, Includes the following modules: Coal blockage diagnosis module: Based on the current timing signal of the coal feeding equipment and the downstream coal flow timing signal, it generates the load deviation duration characterizing the continuous state of motor overload and the flow suppression duration characterizing the continuous state of poor material feeding, and determines coal blockage based on the synergistic relationship between the two. Unblocking Trigger Module: When a risk of coal blockage is detected, an unblocking command is generated, which controls the hydraulic station to start and drives the mechanical hydraulic actuator to make the unblocking head perform an initial unblocking action that includes the stages of pushing out, holding and retracting. Effect evaluation module: During the process of pushing out the blockage clearing head, hydraulic system pressure data is collected simultaneously to form a pressure time series curve, and pressure characteristics including peak pressure, pressure rise rate and high pressure platform duration are extracted from it; The unblocking control module first verifies the unblocking effect of the extracted pressure features. After the verification is successful, template similarity matching is performed to determine the unblocking effect scenario. Based on the determined unblocking restricted scenario or unblocking smooth scenario, the module adaptively controls the subsequent unblocking actions. The acquisition of current timing signals and coal flow timing signals includes the following: real-time acquisition of motor operating current signals by current sensors deployed on the drive motor of the coal feeding equipment to form current timing signals; real-time acquisition of instantaneous coal flow signals by coal flow metering devices set downstream of the coal bunker outlet to form coal flow timing signals; and timestamp alignment processing of the acquired current timing signals and coal flow timing signals. The specific implementation process for generating the load deviation duration characterizing the continuous state of motor overload and the flow suppression duration characterizing the continuous state of poor material feeding is as follows: The real-time acquired current time-series signal is filtered; the processed current time-series signal is compared with the allowable fluctuation range of the reference current, which is the rated current of the equipment; if the current exceeds the allowable fluctuation range of the reference current at a certain moment, a load deviation is identified at that moment, and the duration from the occurrence of the deviation is recorded as the load deviation duration; the real-time acquired coal flow time-series signal is filtered; the filtered coal flow signal is compared with a preset steady-state flow range, which is the statistical distribution range of coal flow during the steady-state operation period when the equipment is confirmed not to have experienced coal blockage under the same operating conditions; when the coal flow at a certain moment is lower than the lower limit of the steady-state flow range, flow suppression is identified at that moment, and the duration from the occurrence of the flow suppression is recorded as the flow suppression duration; The pressure characteristics are extracted as follows: the global maximum pressure value of the pressure time series curve is extracted as the peak pressure; the pressure time series curve is numerically differentiated to obtain the first derivative sequence, and the pressure rise start time is determined by the first derivative being greater than zero and continuously rising; the time when the first derivative first reaches a local maximum is defined as the pressure rise termination time; the pressure rise start time and pressure rise termination time constitute the pressure rise stage, and the ratio of pressure increment to time interval is calculated as the pressure rise rate during this stage; the duration for which the pressure value remains at the peak level after the pressure time series curve reaches the peak pressure is statistically analyzed as the high-pressure plateau duration.

2. The intelligent coal bunker unblocking system as described in claim 1, characterized in that: The determination of coal blockage is described below: When only load deviation is detected, the coal flow inhibition signal is detected synchronously within the time window of load deviation. When flow inhibition is detected, it is determined that there is a risk of coal blockage; otherwise, the monitoring status is maintained. When only flow suppression is detected, load deviation signal is detected synchronously within the time window of flow suppression. When load deviation is detected, it is determined that there is a risk of coal blockage; otherwise, the monitoring status is maintained. When both load deviation and flow suppression are detected simultaneously, the duration of both load deviation and flow suppression are compared with the judgment time limit. If both reach the judgment time limit, it is determined that there is a risk of coal blockage; otherwise, the monitoring status is maintained.

3. The intelligent coal bunker unblocking system as described in claim 1, characterized in that: The unblocking effect is verified through the following execution process: The extracted high-pressure platform duration is compared with the set minimum action time threshold. If the minimum action time threshold is reached, the blockage clearing effect scenario is directly judged. Otherwise, it is considered that the blockage clearing action has not been effectively applied to the blockage. In this case, the mechanical hydraulic actuator is re-driven to control the blockage clearing head to perform a new round of blockage clearing action, and a new pressure time curve is collected simultaneously. Extract the high-pressure platform duration again from the newly acquired pressure time-series curve, and repeat the above scenario determination until any of the following termination conditions are met: a) When the duration of the high-pressure platform reaches the minimum action duration threshold, the blockage clearing effect scenario is determined. b) If the maximum number of retries is reached, the blockage clearing operation is deemed abnormal, and an alarm is triggered.

4. The intelligent coal bunker unblocking system as described in claim 1, characterized in that: The criteria for determining the effectiveness of the congestion clearing effect are as follows: Under the same operating conditions, historical blockage clearing events were collected and marked as unobstructed after downstream coal flow recovery verification. For this sample set, the statistical mean of pressure characteristics was calculated as the benchmark value. The peak pressure, pressure rise rate, and high-pressure platform duration extracted from a single effective unblocking action are normalized with the baseline values ​​of the corresponding features to form a three-dimensional normalized feature vector. The mean of the normalized feature vectors of the original pressure characteristics of all historical congestion clearing events are calculated in the same way to construct the congestion clearing template vector; Calculate the cosine similarity between the 3D normalized feature vector and the congestion clearing template vector; The calculated cosine similarity is compared with the smooth flow threshold. When the smooth flow threshold is reached or exceeded, it is determined to be a clearing and smooth flow scenario; otherwise, it is determined to be a clearing and restricted scenario.

5. The intelligent coal bunker unblocking system as described in claim 1, characterized in that: The process of controlling subsequent congestion clearing actions based on whether the congestion clearing scenario is restricted or unimpeded is as follows: When the scenario is determined to be a restricted blockage clearing scenario, a repeated impact command is generated to control the blockage clearing head to automatically perform an additional push-retract impact cycle a preset number of times after performing the initial blockage clearing action; After each additional impact action is completed, the pressure time-series curve is collected synchronously, the pressure features are re-extracted, a three-dimensional normalized feature vector is constructed, and the cosine similarity is recalculated until the cosine similarity reaches the unobstructed critical value or the number of additional impacts reaches the upper limit, at which point the impact is terminated. When the system is determined to be in a clearing and unblocking scenario, a return-to-standby command is generated, which controls the clearing head to enter a safe return-to-standby process after completing the initial clearing action.

6. The intelligent coal bunker unblocking system as described in claim 5, characterized in that: The initial unblocking action performed by the unblocking head also includes safety monitoring: The mechanical hydraulic actuator is equipped with stroke detection sensors at both ends of the cylinder, including a forward extension sensor and a retraction sensor, which are used to detect the forward extension state and the retraction state of the clearing head, respectively. During the process of the unblocking head executing the push command, the forward extension sensor monitors the end point limit of the stroke. When the sensor reports the forward extension status, the hydraulic output is terminated. After the clearing head executes the retraction command, the retraction and positioning sensor confirms the retraction status. When the sensor reports the retraction and positioning status, it is considered that the single action cycle has been safely completed, and the system is allowed to enter the safe retraction or standby state.

7. A method for intelligent unblocking of coal bunkers, characterized in that: Includes the following steps: S1. Based on the current timing signal of the coal feeding equipment and the downstream coal flow timing signal, generate the load deviation duration characterizing the continuous state of motor overload and the flow suppression duration characterizing the continuous state of poor material feeding. S2. Determine coal blockage based on the synergistic relationship between load deviation duration and flow suppression duration; S3. When a risk of coal blockage is detected, a blockage clearing command is generated, which controls the hydraulic station to start and drives the mechanical hydraulic actuator, so that the blockage clearing head performs an initial blockage clearing action that includes the stages of pushing out, holding and retracting. S4. During the process of pushing out the blockage clearing head, the hydraulic system pressure data is collected simultaneously to form a pressure time series curve, and pressure characteristics including peak pressure, pressure rise rate and high pressure platform duration are extracted from it. S5. First, the extracted pressure features are used to verify the blockage-clearing effect. After the verification is passed, template similarity matching is performed to determine the blockage-clearing effect scenario. S6. Adaptively control subsequent congestion clearing actions based on whether the congestion clearing scenario is restricted or unimpeded.

Images