Humidity coupling pressure relief method of pneumatic pipeline conveying system for production materials of flexible freight bags

Abstract

The invention relates to the technical field of material conveying and industrial automation, in particular to a humidity coupling pressure relief method of a container bag production material pneumatic pipeline conveying system which comprises a central control unit, a humidity sensor, a pressure sensor, a dust concentration sensor, an air door executing mechanism and a pulse buffer valve. By constructing a closed-loop control logic which takes the humidity change rate as a core predictive variable and is deeply coupled with the air door opening, the pulse valve frequency and the pressure fluctuation, conversion from passive response to predictive pre-regulation is realized, and the method comprises the steps of risk predictive modeling, pre-regulation instruction generation, dynamic compensation, failure interception, model self-evolution and the like. The humidity coupling pressure relief method solves or at least relieves the problem that in the prior art, passive response is conducted only depending on a pressure threshold value, and the sudden dust explosion risk caused by sudden change of environment humidity is difficult to deal with, and the humidity coupling pressure relief method of the pneumatic pipeline conveying system for the container bag production materials is provided.

Description

Technical Field

[0001] This invention belongs to the field of material conveying and industrial automation technology, and particularly relates to a humidity-coupled pressure relief method for a pneumatic pipeline conveying system for bulk bag production materials. Background Technology

[0002] In modern industrial production, especially in fields involving the handling of large quantities of powder materials such as FIBCs (Flexible Intermediate Bulk Containers), pneumatic pipeline conveying systems have become an indispensable core technology in material flow processes due to their high efficiency, good sealing properties, and ease of automation. This technology utilizes the energy of compressed gas to achieve long-distance, high-throughput material transport within sealed pipelines, significantly improving production continuity and cleanliness. However, the stable and safe operation of the system remains the cornerstone of technological development in this field. Precise control of the internal pipeline pressure and the design of abnormal pressure relief mechanisms are crucial to ensuring that the entire production process is not interrupted or even catastrophic.

[0003] To address the risk of sudden pressure surges in pipelines caused by poor material flow or fluctuations in operating conditions, existing technologies commonly employ pressure threshold-responsive pressure relief schemes based on mechanical structures or simple electrical controls. For example, a pneumatic pipeline conveyor disclosed in Chinese patent CN116101790B constructs a typical passive pressure relief mechanism by incorporating a spring and a pressure relief hole structure. Specifically, this scheme presets a pressure threshold; when the air pressure inside the pipeline exceeds this threshold, the thrust of the airflow overcomes the preload of the spring, opening the pressure relief channel and releasing the excess pressure, effectively preventing material blockage or pipeline damage caused by excessive pressure. This design concept is clear and the structure is reliable, playing a crucial role in solving the problem of gradual pressure overshoot caused by changes in conventional flow resistance. It represents the mainstream technical paradigm for pressure safety control in this field during a specific historical period. Its core lies in using pipeline pressure as the sole monitoring and triggering variable, constructing a linear control logic that responds after exceeding the limit.

[0004] However, the specific production processes in these industries place more stringent demands on system reliability (higher precision, more intelligent control, and greater resistance to special operating conditions). The inherent limitations of the passive pressure relief defense strategy based on static pressure thresholds begin to reveal their deep-seated flaws when facing these new safety requirements. Taking the high-temperature, low-humidity, and high-speed unloading of materials in large-diameter bulk bags as an example, the primary safety threat is no longer the single pressure threshold, but rather the highly sudden dust explosion risk induced by drastic changes in environmental parameters. Under these conditions, a sharp decrease in ambient humidity is a prerequisite parameter that causes a rapid accumulation of static charge on the material surface, increasing the flammability and explosion sensitivity of the dust cloud. Traditional pressure relief defense systems are "blind" to changes in these environmental parameters. Their logic control does not include consideration of risk parameters such as humidity. Therefore, their defense measures can only passively wait for other physical effects caused by humidity changes (such as material adhesion and local blockage) to eventually be transmitted and manifest as a sudden increase in pressure. However, a sudden increase in pressure is a process with a significant physical delay. By the time the pressure sensor finally detects an over-limit signal and triggers pressure relief measures, the dust concentration inside the pipe may have already crossed the danger zone, and the system is actually in a high-risk state. The fundamental decoupling between the control variable (pressure) and the risk factor (humidity change) causes the system to completely lose its ability to predict and warn, let alone intervene in advance or provide proactive defense, forming a delayed defense that only addresses the symptoms and not the root cause. The deeper contradiction lies in the fact that one is facing dynamic, multi-parameter coupled risk factors, but using a static, single-parameter driven control strategy. This mismatch in strategy leads to delayed defense or even failure at critical moments, making it difficult to meet the new requirements of modern industrial safety standards for predictive defense and proactive interception.

[0005] Therefore, how to break through the existing framework of passive response based solely on pressure thresholds and establish a predictive control model that can deeply couple key environmental precursor information such as humidity change rate with pressure relief actions, thereby shifting from post-event pressure relief to pre-event pre-regulation and enabling precise intervention in the early stages of risk formation, has become a key challenge and an urgent technical problem for those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to overcome the shortcomings of the prior art, solve or at least alleviate the defects in the existing river flood evolution simulation and flood control decision support system, which are caused by the inability to provide real-time and accurate feedback of the physical execution status of flood control dispatching instructions to the digital simulation model, and the lack of a unified high-precision time reference, resulting in a disconnect between simulation prediction and reality and reduced decision reliability. The invention provides a humidity-coupled pressure relief method for a pneumatic pipeline conveying system for bulk bag production materials.

[0007] To achieve the above objectives, the present invention provides the following technical solution: a humidity-coupled pressure relief method for a pneumatic pipeline conveying system for bulk bag production materials. The method is deployed in a system comprising a central control unit, a humidity sensor, a pressure sensor, a dust concentration sensor located at the unloading end of the pipeline outlet, and a damper actuator and a pulse buffer valve electrically connected to the central control unit. The method includes the following steps: Risk prediction modeling and parameter initialization steps: Within the first preset time period before the material unloading task starts, the central control unit drives the humidity sensor to continuously collect the humidity value at the pipeline outlet and calculates the humidity change rate based on the collected time series humidity data; when the central control unit determines that there is a potential dust explosion risk based on the calculation result of the humidity change rate, it activates the risk prediction logic. Pre-adjustment instruction generation and scenario adaptation steps: Based on the activation state of the risk prediction logic, the central control unit generates and issues pre-adjustment instructions to the damper actuator and the pulse buffer valve in advance before the material enters the pipeline on a large scale. The pre-adjustment instructions are used to adjust the damper opening driven by the damper actuator and the working frequency of the pulse buffer valve to a preset non-standard working state, so as to actively intervene in the airflow conditions in the pipeline and suppress the risk of dust explosion caused by drastic changes in humidity.

[0008] To further realize the present invention, the following technical solutions may be preferred: Preferably, the risk prediction modeling and parameter initialization steps specifically include: Dynamic environment evolution data acquisition sub-step: Within the first preset time period, the central control unit drives the humidity sensor to continuously acquire the humidity value at the outlet of the pipe at the first preset sampling frequency, forming a time series humidity data stream; Risk prediction model construction sub-step: The central control unit runs a risk prediction algorithm model. The model receives the time series humidity data stream as input and calculates the humidity decrease rate in real time. When the absolute value of the humidity decrease rate calculated by the central control unit is greater than the preset humidity change rate threshold, the high temperature and low humidity risk prediction logic is activated, and a pre-adjustment intensity parameter for subsequent instruction generation is generated.

[0009] Preferably, in the dynamic environment evolution data acquisition sub-step, the central control unit also synchronously drives the pressure sensor to record instantaneous pressure fluctuations in the pipeline at a preset sampling interval, and calculates the slope of the pressure-time curve in real time; and The risk prediction model construction sub-step includes an enhanced monitoring mechanism: if the absolute value of the slope of the pressure-time curve calculated by the central control unit is greater than a preset pressure threshold, an execution command is triggered. The execution command forcibly increases the sampling frequency of the humidity sensor from the first preset sampling frequency to the second preset sampling frequency to ensure that the data input accuracy of the risk prediction algorithm model is improved during periods of drastic pressure changes.

[0010] Preferably, the risk prediction modeling and parameter initialization step further includes: Model parameter dynamic freezing sub-step: When the temperature inside the pipeline is higher than the preset temperature threshold and the output confidence of the risk prediction algorithm model is greater than or equal to the preset confidence threshold, the central control unit will lock the currently calculated damper opening benchmark value and forcibly block the system from executing the conventional, pressure feedback-based adjustment logic. Furthermore, in response to the execution command, if the pressure fluctuation has triggered an increase in the sampling frequency of the humidity sensor, the central control unit automatically extends the duration of the risk prediction window used to calculate the humidity change rate from the first preset duration to the second preset duration to compensate for data disturbances that may be introduced due to sudden pressure changes.

[0011] Preferably, the pre-adjustment instruction generation and scene adaptation steps specifically include: Emergency scenario pre-adjustment start-up sub-step: When the central control unit detects that the absolute value of the humidity change rate is continuously greater than the preset humidity change rate threshold, the central control unit generates and issues the pre-adjustment command in advance. The pre-adjustment command drives the damper actuator to adjust the opening of the damper to a preset multiple of the standard opening. Opening frequency mapping generation sub-step: The central control unit dynamically calculates the initial operating frequency of the matching pulse buffer valve by querying the internally stored nonlinear mapping function or lookup table based on the target opening value of the damper determined in the emergency scenario pre-adjustment start-up sub-step.

[0012] Preferably, the emergency scenario pre-adjustment initiation sub-step includes a compensatory adjustment mechanism: if the duration of the risk prediction window is extended due to the absolute value of the slope of the pressure-time curve being greater than the preset pressure threshold, then when the central control unit executes the emergency scenario pre-adjustment initiation sub-step, it immediately increases the target multiple of the damper opening from the preset multiple to a higher preset multiple; and The opening frequency mapping generation sub-step includes a nested correction mechanism: if the compensatory adjustment mechanism is triggered, causing the damper opening to increase to a higher preset multiple, the central control unit, when executing the opening frequency mapping generation sub-step, nests and corrects the mapping relationship to further reduce the operating frequency of the corresponding pulse buffer valve.

[0013] Preferably, the method further includes a dynamic compensation and failure interception step, wherein the dynamic compensation and failure interception step dynamically corrects the execution deviation by real-time monitoring of system status feedback during the execution of the pre-adjustment instruction, and specifically includes: Pre-adjustment execution and deviation monitoring sub-step: The central control unit drives the damper actuator and the pulse buffer valve to start working according to the determined parameters, and simultaneously starts trend monitoring of the dust concentration sensor to calculate the dust concentration change rate; if the dust concentration increase rate calculated by the central control unit is greater than the first preset concentration change threshold, it indicates that the pre-adjustment intensity is insufficient, and the system immediately nests the emergency scenario pre-adjustment start sub-step to further increase the damper opening value.

[0014] Preferably, the dynamic compensation and failure interception step further includes: The three-level linkage sub-step for failure interception: When the dust concentration increase rate continues to be greater than the second preset concentration change threshold and the duration exceeds the first preset duration, it indicates that the aforementioned adjustment measures have failed to effectively control the risk. The central control unit will simultaneously perform three mandatory operations: multiply the damper opening driven by the damper actuator by the first gain coefficient; multiply the operating frequency of the pulse buffer valve by the second gain coefficient; and trigger and open a backup airflow channel controlled by the central control unit to introduce additional clean gas for dilution and purging.

[0015] Preferably, the method further includes an effect verification and model self-evolution step, wherein the effect verification and model self-evolution step quantifies and evaluates the actual effect of pre-adjustment, and uses the evaluation results to drive the continuous optimization of the control model, specifically including: Pre-adjustment effect real-time evaluation sub-step: After the material unloading task is started, the central control unit compares the theoretical value of dust concentration predicted by the internal model with the concentration value actually measured by the dust concentration sensor, and calculates the deviation between the two. Model parameter dynamic calibration sub-step: When the absolute value of the deviation is greater than the first preset deviation threshold, the central control unit resets the core parameters of the risk prediction algorithm model. Specifically, it multiplies the weight coefficient of the humidity change rate in the model by a calibration factor and forcibly binds and updates the pre-adjustment trigger threshold in the risk prediction logic.

[0016] Preferably, the effect verification and model self-evolution steps further include: Successful experience solidification and storage sub-step: When the absolute value of the deviation is less than or equal to the second preset deviation threshold, it indicates that the current model parameters and mapping relationship can accurately cope with the current working conditions. The central control unit locks the currently used prediction coefficients and mapping functions and stores them in the database of non-volatile memory. Furthermore, the method includes a scene feature learning mechanism: if the failure interception three-level linkage sub-step has been activated in the current task, the system automatically extracts and marks the scene features that caused the failure, and associates and stores the scene features with the successful response strategy to provide priority adaptation rules for the next similar scenario uninstallation task.

[0017] The beneficial effects of this invention are: This invention represents a fundamental shift from passive response to predictive intervention. By using the rate of change in humidity as a core precursor variable, this invention enables accurate prediction and early intervention at the initial stage of dust explosion risk formation, significantly earlier than traditional hysteresis response mechanisms that rely solely on pressure over-limit signals, thus greatly enhancing the system's safety margin. This invention dynamically correlates multiple key parameters such as humidity, pressure, and dust concentration, employing a "prediction-pre-adjustment-compensation-model iteration" logic. This gives the system powerful self-adaptive, self-calibrating, and self-optimizing capabilities when facing complex and changing operating conditions, ensuring the accuracy and robustness of adjustment actions.

[0018] Meanwhile, this invention provides a tiered and interconnected failure interception and compensation mechanism. By setting up multiple defense layers from deviation monitoring and dynamic compensation to three-level interconnected failure interception, this invention ensures that even in extreme cases where the prediction model deviates or the initial adjustment measures are ineffective, the system can still activate stronger intervention measures at each level, effectively preventing safety incidents. Attached Figure Description

[0019] Figure 1 This is a block diagram of the overall architecture of the system of the present invention; Figure 2 This is a schematic diagram of the overall process of the method of the present invention; Figure 3 This is a detailed flowchart illustrating the risk prediction modeling and parameter initialization steps of the present invention. Figure 4 This is a detailed flowchart illustrating the pre-adjustment instruction generation and scene adaptation steps of the present invention. Figure 5 This is a detailed flowchart illustrating the dynamic compensation and failure interception steps of the present invention; Figure 6This is a detailed flowchart illustrating the effect verification and model self-evolution steps of the present invention; Figure 7 This is a schematic diagram illustrating the relationship between the humidity change rate and the pre-adjustment of the damper opening according to the present invention. Figure 8 This is a schematic diagram of the three-level linkage mechanism for failure interception of the present invention. Detailed Implementation

[0020] In the description of this invention, it should also be noted that, unless otherwise explicitly specified and limited, the terms "set," "install," "connect," and "link" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0021] 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. Example 1

[0022] This embodiment discloses a pneumatic pipeline conveying system for producing materials for container bags, referring to... Figure 1 The system constructs a closed-loop pneumatic conveying and status monitoring circuit at the physical level. Its core components include a humidity sensor for collecting ambient humidity information at the pipeline outlet unloading end, a pressure sensor for real-time monitoring of the internal pressure dynamics of the pipeline, a dust concentration sensor for measuring the dust concentration inside the pipeline, a damper actuator for precisely regulating the airflow in the pipeline, a pulse buffer valve for generating pulsed airflow to buffer and guide materials, and a backup airflow channel as an emergency safety measure. All of the above sensors and actuators are electrically connected and communicate with a central control unit via an industrial fieldbus.

[0023] The central control unit is the core of the entire system, and its hardware can be a Siemens SIMATIC S7-1500 series programmable logic controller (PLC). The humidity sensor preferably uses the Vaisala HMP110 probe, which provides high-fidelity input data for the risk prediction model. The pressure sensor uses an Endershaus Cerabar PMP51 absolute pressure sensor, which can detect instantaneous pressure fluctuations caused by material conveying obstructions. The dust concentration sensor uses a SICK DUSTHUNTER SP100 device based on the forward scattering principle, which can output the dust mass concentration per unit volume in real time, providing quantitative evidence for verifying the pre-adjustment effect. The damper actuator is driven by a stepper motor or servo motor with a high-precision encoder, ensuring that fine-tuning commands for the damper opening are accurately executed. The pulse buffer valve is a Goyen RCAC series diaphragm valve, which can generate a stable and rapid pulsed airflow according to the frequency commands output by the central control unit. The backup airflow channel is opened and closed by a solenoid valve controlled by a central control unit, which can introduce dry and clean compressed air in an emergency to quickly dilute and purge the environment inside the pipeline. Example 2

[0024] Based on the system architecture of Embodiment 1, this embodiment further discloses a humidity-coupled pressure relief method for a pneumatic pipeline conveying system for bulk bag production materials. The overall process of this method is as follows: Figure 2 As shown, its core lies in upgrading the traditional passive safety response based on pressure thresholds into a predictive proactive safety intervention paradigm based on humidity change rate prediction, multivariate coupled regulation, and possessing self-learning and self-evolution capabilities. (Refer to...) Figures 2 to 6 This method starts with risk prediction, goes through pre-adjustment, dynamic compensation, effect verification, and finally ends with a safe reset.

[0025] This method is activated when a material unloading task is initiated. First, the system performs a risk prediction modeling and parameter initialization step, the detailed process of which is as follows: Figure 3As shown. Its purpose is to establish a quantitative correlation between potential risks and control parameters in advance by analyzing the evolution trend of the environmental state before materials enter the pipeline on a large scale. After the unloading task command is issued, the system enters a first preset duration, specifically a 5-second standby observation period. During this period, a specific function block of the central control unit (e.g., FB100 "RiskModel_Init" created in S7-1500) is activated and begins to execute the dynamic environmental evolution data acquisition sub-step. The central control unit continuously acquires the 4-20mA analog signal output by the humidity sensor through its high-speed analog input module at a first preset sampling frequency of 10Hz. After analog-to-digital conversion and linear calibration, the acquired data is encapsulated into a data frame containing a high-precision timestamp and humidity measurement value, and sent to a first-in-first-out (FIFO) circular buffer. At the same time, the central control unit synchronously records the pressure sensor readings at a sampling interval of 50 milliseconds and uses the differential method to calculate the slope of the pressure-time curve in real time, namely the instantaneous pressure change rate dP / dt. This slope value is used as the dynamic basis for subsequent adjustment of the data acquisition strategy.

[0026] Next, the system performs a risk prediction model construction sub-step. The central control unit embeds a pre-set risk prediction algorithm model, whose main logic is to calculate the derivative of the humidity data time series within the annular buffer, i.e., the humidity decrease rate dH / dt. First, a moving average filter is applied to the humidity data of the most recent 10 sampling points to eliminate sensor noise; then, the linear regression slope is calculated based on the smoothed humidity data. As long as the absolute value of this slope, i.e., the absolute value of the humidity decrease rate, is greater than a first humidity change rate threshold, such as 2% / second, the system executes a high-temperature, low-humidity risk prediction logic. Extensive practical experiments have shown that when humidity decreases at a rate exceeding 2% / second, the accumulation rate of static charge generated by friction on the material surface increases significantly, leading to a sharp increase in the risk of dust explosion; therefore, this is set as the first humidity change rate threshold. Once the high-temperature, low-humidity risk prediction logic is activated, a pre-adjustment intensity coefficient is immediately generated. This coefficient is set as the product of the absolute value of the current humidity change rate and a first preset coefficient (set to 0.8), and is directly used to generate subsequent pre-adjustment instructions, realizing a quantitative correspondence between predicted risk and intervention intensity. As a nested enhanced monitoring method, during data acquisition, if the absolute value of the pressure fluctuation slope dP / dt calculated by the central control unit is greater than a first pressure threshold, such as 5 kPa / s, it indicates that there may be slight material blockage or uneven flow in the pipeline, and the system immediately invokes an interrupt service routine. This routine forcibly increases the sampling frequency of the humidity sensor to 20 Hz and achieves this by changing the sampling period of the analog input module. In this way, it can be ensured that in the early stage of unstable pressure conditions, the risk prediction model can obtain data input with higher time resolution, thereby improving the accuracy of prediction.

[0027] Next, the system executes a dynamic model parameter freezing sub-step. When the temperature measured by the temperature sensor inside the pipeline is higher than a second temperature threshold (e.g., 80°C), and the risk confidence level output by the risk prediction model is greater than or equal to a third confidence threshold (e.g., 85%), the central control unit executes a locking command. This locking command is used to lock the currently determined damper opening reference value and the pulse frequency in the data block, and to set a global flag. This global flag is configured to forcibly shield one or more PID control loops based on the pressure feedback to avoid conflicts between different control logics. While the central control unit is performing the dynamic freezing of the model parameters, if the humidity sampling frequency has increased from a low frequency to a high frequency due to pressure fluctuations, the risk prediction window length is adjusted from the first preset time length to the second preset time length. The first preset time length corresponds to a data length of 5 seconds (i.e., 50 data points @ 10Hz) in the annular buffer, and the second preset time length corresponds to a data length of 8 seconds (i.e., 160 data points @ 20Hz) in the annular buffer. By increasing the sample size for data smoothing, the signal mutations that may be caused by pressure changes are compensated for, ensuring the stability of the risk prediction model's output under complex operating conditions.

[0028] After completing risk prediction and parameter initialization, the system enters the pre-adjustment instruction generation and scenario adaptation steps, the detailed process of which is as follows: Figure 4 As shown. The core task of this step is to transform the abstract risk prediction results into specific and quantifiable control commands for the damper actuator and pulse buffer valve. The system executes the emergency scenario pre-adjustment initiation sub-step. Based on the output of the risk prediction model, if the central control unit detects that the absolute value of the humidity change rate is continuously greater than the first humidity change rate threshold (2% / second), a pre-adjustment command is generated and issued 3 seconds in advance (relative to the time point when materials enter the pipeline on a large scale). Figure 7As shown, this drives the damper actuator to adjust the damper opening to 1.5 times the standard opening (e.g., a 45-degree opening angle corresponding to normal flow rate), i.e., a 67.5-degree opening angle (the preset multiple is set to 1.5 times, based on the experimentally calibrated mapping relationship between humidity change rate and safety margin: for every 1% / second increase in humidity change rate, the opening multiple increases by 0.1 times). The corresponding physical fine-tuning displacement is 0.12 mm, and this instruction is implemented by sending a precise number of pulses to the stepper motor driver. This pre-adjustment instruction is given the highest execution priority in the PLC's program organization block and can be used to forcibly interrupt and override any initial setting instruction generated based on normal operating conditions. As a compensatory adjustment mechanism, if the risk prediction window is extended due to pressure mutation in the aforementioned steps, the central control unit will immediately increase the target multiple of the damper opening from 1.5 times to 1.7 times, i.e., adjust it to a 76.5-degree opening angle, when executing this sub-step. This opening gain adjustment is designed to proactively compensate for the slight response delay that may be caused by the extension of the prediction window by further increasing the gas flow rate, thus ensuring a safety margin.

[0029] Correspondingly, the system also includes a dynamic suppression sub-step for normal scenarios. When the output confidence of the risk prediction model remains below a second preset confidence threshold, such as 70%, it indicates that the current environment is stable and the system is in a low-risk state. The central control unit will freeze the damper opening at the standard operating value and forcibly disable the enhancement logic in the subsequent dynamic compensation mechanism to prevent energy waste and interference with the stability of material conveying caused by performing dynamic compensation under normal operating conditions. As a nested backtracking mechanism to prevent scenario misjudgment, if the absolute value of the pressure fluctuation slope dP / dt monitored by the pressure sensor suddenly drops during a period judged as a normal scenario, and the change is greater than the second pressure threshold (e.g., 3 kPa / s), this indicates a sudden collapse of material or a drastic change in the flow state within the pipeline. The central control unit will immediately interrupt the current normal scenario suppression logic and forcibly return to the risk prediction model construction sub-step to reassess the risk level of the current operating condition using the latest sensor data.

[0030] Subsequently, the system performs an opening frequency mapping generation sub-step. Based on the target damper opening value calculated in the emergency scenario pre-adjustment start-up sub-step, the central control unit dynamically calculates the initial operating frequency of the matching pulse buffer valve by querying a nonlinear mapping function or lookup table stored in the non-volatile memory area. This lookup table is pre-calibrated based on fluid dynamics simulations and extensive experimental data, aiming to ensure that the increase in airflow caused by the increase in damper opening matches the pressure relief strength of the pulse valve, and that the two are coordinated. For example, when the damper opening is set to 1.5 times the standard opening (67.5 degrees), the corresponding pulse buffer valve frequency is set to 0.7 times the standard frequency (e.g., 5Hz), i.e., 3.5Hz. The purpose of reducing the frequency is to increase airflow while extending the pulse interval to avoid excessive disturbance to the material flow field. This calculated frequency value will be written as a core execution parameter into the register of the pulse valve controller. As a nested correction mechanism to ensure precise matching of adjustment intensity, if the aforementioned compensatory adjustment mechanism is triggered, causing the damper opening to increase to 1.7 times (76.5 degrees), the central control unit will, during this sub-step, nestedly correct the mapping relationship, further reducing the corresponding pulse buffer valve frequency to 0.5 times the standard frequency, i.e., 2.5Hz. In this way, a precise and coordinated inverse relationship is formed between the pre-adjusted opening increase and the enhanced pressure relief intensity, achieving more refined control. After the opening ratio is increased, the system automatically lowers the pulse frequency according to the inverse coordinated rule of opening increase - frequency decrease (for every 0.1 times increase in opening, the frequency decreases by 0.1Hz).

[0031] Next, the system enters the dynamic compensation and failure interception step, the detailed process of which is as follows: Figure 5 As shown. This step, after the pre-adjustment command is executed, dynamically corrects any possible execution deviations by monitoring the system status feedback in real time, and sets up a multi-level failure interception mechanism to deal with extreme situations. Specifically, the system executes the pre-adjustment execution and deviation monitoring sub-steps. The central control unit generates the parameters determined in the sub-step according to the opening frequency mapping, and drives the damper actuator and pulse buffer valve to start working through pulse width modulation (PWM) output or fieldbus commands. At the same time, the system synchronously starts trend monitoring of the dust concentration sensor, records the dust concentration value at 200-millisecond intervals, and calculates its rate of change dC / dt. If, during this monitoring process, the dust concentration rise rate calculated by the central control unit is greater than the first concentration change threshold, for example, 0.5 mg / m³ / s, it indicates that the current pre-adjustment intensity is insufficient to suppress dust rising. The system will immediately nest and execute the aforementioned emergency scenario pre-adjustment start sub-step, further increasing the damper opening value by a fixed step (for example, increasing it by 5%), thereby taking proactive and escalating intervention before the dust concentration reaches the dangerous threshold.

[0032] Next, if deviation monitoring triggers an increase in the damper opening, the system will activate the dynamic compensation mechanism sub-step. The central control unit automatically expands the sampling range of the humidity sensors. Specifically, in addition to the existing single-point humidity sensor at the pipe outlet, the system will activate two additional humidity sensors arranged along the length of the pipe (e.g., located in the middle and beginning of the pipe), expanding humidity data acquisition from a single point to multiple points. Through multi-point sampling, the system can obtain humidity distribution gradient information within the pipe. The central control unit calculates the humidity gradient dH / dx. If the absolute value of this gradient is greater than a preset gradient threshold, such as 1% / m, it indicates the existence of a significant humidity non-uniformity area within the pipe, which may be a significant factor leading to localized static electricity accumulation and increased risk. At this point, the system will nest the sub-step of generating the opening frequency mapping, but instead of using the preset lookup table, it will reset the mapping relationship between opening and frequency based on the current humidity gradient information through a correction algorithm (for example, superimposing a compensation term proportional to the gradient value on the original mapping relationship; specifically, the correction algorithm is: new frequency = frequency - K * dH / dx, where K is the gradient compensation coefficient, set to 0.3 Hz / (% / m)) to dynamically adapt to this non-uniform working condition.

[0033] To cope with extreme situations, the system is equipped with a three-level linkage sub-step for failure interception, the mechanism of which is as follows: Figure 8 As shown, when the dust concentration rise rate dC / dt continuously exceeds the second concentration change threshold (e.g., 0.8 mg / m³ / s), and this state lasts for more than 1 second, it indicates that all the aforementioned adjustment measures have failed to effectively control the risk, and the system will determine an emergency failure state. At this time, the central control unit will simultaneously execute three mandatory, parallel operations: First, multiply the opening of the damper actuator by a gain factor of 1.8 based on the current value to achieve maximum ventilation; second, multiply the operating frequency of the pulse buffer valve by a gain factor of 1.3 based on the current value to forcefully clear materials through high-frequency pulses; third, trigger and open the solenoid valve of the backup airflow channel to introduce additional clean gas for emergency dilution and purging. As a final insurance mechanism, if the dust concentration increase rate does not show a downward trend within the preset 2-second time after the above three-level linkage operation is executed, the central control unit will forcibly interrupt all execution and compensation logic and jump to the model parameter dynamic freezing sub-step to forcibly reset and refresh the internal state variables (such as integral terms, historical data buffers, etc.) of the entire prediction model. This is equivalent to performing a "hot start" on the control system to achieve absolute interception of the risk of system runaway.

[0034] Furthermore, the system execution effect verification and model self-evolution steps are detailed as follows: Figure 6As shown. This step aims to quantitatively evaluate the actual effect of pre-conditioning and use the evaluation results to drive continuous optimization of the control model, forming a closed loop of learning and evolution. Specifically, 2 seconds after the material unloading task starts, the system executes a real-time evaluation sub-step of the pre-conditioning effect. A simplified material conveying model runs inside the central control unit, which predicts a theoretical value of dust concentration based on the current material characteristics (pre-input), damper opening, and airflow velocity. Then, the system compares this theoretical value with the concentration value actually measured by the dust concentration sensor and calculates the relative deviation between the two. This deviation value will serve as the decision-making basis, forcibly determining whether the system enters the dynamic calibration sub-step of model parameters or the successful experience solidification and storage sub-step. If the calculated absolute value of the deviation is greater than the preset deviation threshold, such as 15%, it indicates that there is a large gap between the prediction model and the actual operating conditions. The system will immediately nest and execute the enhanced monitoring mechanism in the risk prediction model construction sub-step, that is, forcibly increase the sampling frequency of the humidity sensor to 20Hz to provide higher precision data input for subsequent model parameter iterations.

[0035] Subsequently, based on the evaluation results, the system may execute a dynamic calibration sub-step for model parameters. When the absolute value of the deviation exceeds 15%, it indicates that the model's failure has reached a point where its predictive ability needs to be corrected. The central control unit will reset the core parameters of the risk prediction model online. The humidity change rate weighting coefficient used to calculate risk confidence in the model will be multiplied by a calibration factor, for example, 1.2 (the calibration factor is dynamically calculated based on historical deviation data: when the deviation exceeds 15%, the calibration factor = 1 + (absolute deviation value - 15%) / 50), increasing the model's sensitivity to humidity changes. This forcibly binds and updates the pre-adjustment trigger threshold in the risk prediction logic, making the next prediction more sensitive. This is a deep self-optimization mechanism. If model parameter calibration is triggered twice consecutively within a complete material unloading task cycle, the system will determine that not only is there a problem in the prediction layer, but there may also be a mismatch in the execution mapping layer. At this point, the system will nest the sub-step of generating the opening frequency mapping, but instead of querying, it will modify the points already saved in the lookup table. For example, it will slightly modify the points around the current operating point (damper opening, pulse frequency) and move from the prediction layer to the execution mapping layer for linkage optimization.

[0036] Conversely, if the evaluation result is ideal, the system proceeds to execute the successful experience solidification and storage sub-step. When the absolute value of the deviation is less than or equal to 5%, it indicates that the current model parameters and mapping relationship are suitable for the working condition. The central control unit locks the currently used prediction coefficients, weight parameters, and opening-frequency mapping relationship, and stores them in the PLC's non-volatile memory database as a successful "scenario-strategy" pair. Locking this strategy will forcibly prevent the dynamic adjustment function of the risk prediction model until the end of the current task cycle to maintain the stability of the strategy. As a scenario feature learning mechanism, if the failure interception three-level linkage sub-step was activated in this task and the risk was successfully controlled, the system will automatically extract and mark the scenario features that caused the failure (e.g., humidity gradient greater than 1% / m and pressure fluctuation greater than 5kPa / s), and forcibly associate and store this feature with the final successful response strategy (e.g., damper opening 2.1 times, pulse frequency 1.0 times), so that when facing similar extreme scenarios again, the system can directly call the optimized historical successful experience and prioritize the adaptation rules.

[0037] Finally, as the material unloading task nears completion, the system executes a final-state safety zeroing step. This step ensures that after each task, all nested logic triggered during system operation and all modified parameters are safely and thoroughly stopped, released, and cleared, preventing any residual state from causing unpredictable impacts on subsequent operations. The system then executes a final-state risk confirmation sub-step. The central control unit confirms the unloading process is safely completed and issues an unloading completion signal only when the dust concentration sensor's measurement value is less than or equal to the safety threshold (e.g., 5 mg / m³) for 5 consecutive seconds, and the absolute value of its rate of decrease is greater than 0.2 mg / m³ / s. The issuance of the unloading completion signal has a mandatory dependency on confirming the verification results of the effect verification step; that is, the system can only terminate under controlled conditions. If the model parameter dynamic calibration step was performed in this task, the system, while issuing the unloading completion signal, must also store the calibrated model parameters in a scenario library corresponding to a specific material type or environmental condition (e.g., "high static PP material - winter drying"), thereby enabling continuous evolution of the control method.

[0038] Subsequently, the system executes a system state safety reset sub-step. Upon receiving the unloading completion signal, the central control unit immediately drives the damper actuator to precisely reset it to the initial fully closed position, ensuring that its position error is controlled within ±0.05mm via encoder feedback. Simultaneously, the control signal of the pulse buffer valve is shut off, and the solenoid valve of the backup airflow channel is also de-energized and closed. This reset action forcibly clears all dynamic parameters, flags, and temporary variables generated in RAM during the pre-adjustment and dynamic compensation steps. As a nested operation to ensure the accuracy of the sensor reference, if an abnormal event was marked due to a sudden pressure change or abnormal humidity gradient during task execution, a zero-point calibration procedure for the humidity sensor will be nested and executed during the system reset. This procedure controls the system to introduce a reference gas with known humidity flowing through the sensor and corrects the sensor's zero-point offset based on the deviation between the reading and the known value, thereby eliminating the potential interference of historical data drift on the next prediction. At this point, the entire closed-loop control process is complete. Example 3

[0039] This embodiment aims to verify the effectiveness of the method of the present invention in simulating the risks caused by sudden environmental changes in actual production.

[0040] Experimental conditions: The conveyed material was polypropylene (PP) powder with an average particle size of 50 μm, which is prone to static electricity. The conveying pipeline was 20 meters long and DN200 in diameter. The initial ambient temperature was 25℃ and relative humidity was 55%RH. Before the material conveying task started, a malfunction of the workshop's air conditioning system was simulated, introducing a dry airflow to rapidly reduce the humidity at the pipeline outlet.

[0041] System configuration: The pneumatic pipeline conveying system and humidity-coupled pressure relief method for FIBC production materials described in this invention are adopted. The threshold parameters are set as described above: the first humidity change rate threshold is 2% / second, the first pressure threshold is 5kPa / s, and the first concentration change threshold is 0.5mg / m³ / s, etc.

[0042] Experimental process and results: 1. At t=0s, the material conveying task is initiated. The system then enters a 5-second risk prediction modeling and parameter initialization phase.

[0043] 2. From t=0s to t=2s, the ambient humidity remains at 55%RH, and the humidity change rate is close to 0. The system determines this to be a normal scenario.

[0044] 3. At t=2.1s, the dry airflow begins to affect the pipe outlet, and the humidity sensor reading begins to drop.

[0045] 4. At t=3.5s, the central control unit calculates that the humidity decreased from 54%RH to 51.5%RH in the past second, with an absolute rate of change of 2.5% / second, which is greater than the threshold of 2% / second. The risk prediction model is activated.

[0046] 5. t = 3.6s, the system generates a pre-adjustment command in advance, driving the damper actuator to adjust the opening from the standard 45 degrees to 1.5 times 67.5 degrees. At the same time, the frequency of the pulse buffer valve is reduced from the standard 5Hz to 3.5Hz through a lookup table.

[0047] 6. At t=5s, the initialization phase ends, and materials begin to enter the conveying pipeline on a large scale. Because the damper has been opened in advance, a stronger negative pressure and a higher airflow velocity are formed inside the pipeline, effectively suppressing the initial adhesion of PP powder to the pipe wall due to static electricity.

[0048] 7. From t=5s to t=15s, the material was conveyed stably. The dust concentration sensor reading stabilized at around 45mg / m³ after a brief rise, with the peak value not exceeding 50mg / m³. The pressure sensor reading fluctuated smoothly without any sudden pressure changes.

[0049] 8. At t=7s, the system execution effect was verified. The theoretical dust concentration predicted by the model was 42mg / m³, which was less than 15% compared with the actual measured 45mg / m³. The deviation was approximately 7.1% (45-42) / 42. The system was judged to have successfully predicted and adjusted the dust concentration, and the current parameter combination was stored in the successful experience database.

[0050] By introducing a predictive adjustment mechanism based on humidity change rate, precise and gradual proactive intervention can be carried out at the nascent stage of risk formation. Compared with existing passive response technologies that rely solely on pressure thresholds, this greatly advances the response time, effectively avoids the dangerous accumulation of dust concentration and pressure in pipelines, and significantly improves the system's operational stability and intrinsic safety level.

[0051] The above are merely preferred embodiments of the present invention, but the scope of protection of the present invention is not limited thereto. Any equivalent substitutions or modifications made by those skilled in the art within the scope of the technology disclosed in the present invention, based on the technical solution and inventive concept of the present invention, should be covered within the scope of protection of the present invention.

Claims

1. A humidity coupling pressure relief method for a pneumatic pipe conveying system of a bag production line, the method is deployed in a system comprising a central control unit, a humidity sensor arranged at an unloading end of a pipe outlet, a pressure sensor, a dust concentration sensor, and a damper actuator and a pulse buffer valve electrically connected to the central control unit, characterized in that, The method comprises the following steps: Risk prediction modeling and parameter initialization step: within a first preset time length before the start of the material unloading task, the central control unit drives the humidity sensor to continuously collect the humidity value at the outlet end of the pipeline, and calculates the humidity change rate based on the collected time series humidity data; when the central control unit judges that there is a potential dust explosion risk based on the calculation result of the humidity change rate, the risk prediction logic is activated; Pre-adjustment instruction generation and scene adaptation step: based on the activation state of the risk prediction logic, the central control unit generates and issues a pre-adjustment instruction to the damper actuator and the pulse buffer valve in advance before the material enters the pipeline in large quantities, the pre-adjustment instruction is used to drive the damper actuator to adjust the damper opening degree, and drive the pulse buffer valve to adjust the working frequency to a preset non-standard working state, so as to actively intervene in the airflow condition in the pipeline, thereby inhibiting the dust explosion risk caused by the sharp change of humidity.

2. The method of claim 1, wherein, The risk prediction modeling and parameter initialization step specifically comprises: Dynamic environment evolution data collection sub-step: within the first preset time length, the central control unit drives the humidity sensor to continuously collect the humidity value at the outlet end of the pipeline at a first preset sampling frequency, forming a time series humidity data stream; Risk prediction model construction sub-step: the central control unit internally runs a risk prediction algorithm model, which receives the time series humidity data stream as input and calculates the humidity drop rate in real time; when the absolute value of the humidity drop rate calculated by the central control unit is greater than the preset humidity change rate threshold, the risk prediction logic is activated, and a pre-adjustment intensity parameter for subsequent instruction generation is generated.

3. The method of claim 2, wherein, In the dynamic environment evolution data collection sub-step, the central control unit also synchronously drives the pressure sensor to record the instantaneous pressure fluctuation in the pipeline at a preset sampling interval, and calculates the slope of the pressure-time curve in real time; and In the risk prediction model construction sub-step, an enhanced monitoring mechanism is included: if the absolute value of the slope of the pressure-time curve calculated by the central control unit is greater than the preset pressure threshold, an execution instruction is triggered, which forcibly increases the sampling frequency of the humidity sensor from the first preset sampling frequency to a second preset sampling frequency, so as to ensure that the data input accuracy of the risk prediction algorithm model is improved during the sharp change of pressure.

4. The method of claim 3, wherein, The risk prediction modeling and parameter initialization step further comprises: Model parameter dynamic freezing sub-step: when the temperature in the pipeline is higher than the preset temperature threshold and the output confidence of the risk prediction algorithm model is greater than or equal to the preset confidence threshold, the central control unit locks the currently calculated pre-adjustment damper opening degree reference value, and forcibly blocks the system from executing the regular pressure feedback-based adjustment logic; And, as a response to the execution instruction, if the pressure fluctuation has triggered the humidity sensor sampling frequency to be raised, the central control unit automatically extends the length of the risk prediction window for calculating the humidity change rate from the first preset length to a second preset length to compensate for the data disturbance that may be introduced by the pressure mutation.

5. The method of claim 4, wherein, The pre-conditioning instruction generation and scene adaptation step specifically includes: An emergency scene pre-conditioning sub-step: when the central control unit monitors that the absolute value of the humidity change rate is continuously greater than the preset humidity change rate threshold, the central control unit generates and issues the pre-conditioning instruction in advance, which drives the damper actuator to adjust the opening degree of the damper to a preset multiple of the standard opening degree; An opening degree-frequency mapping generation sub-step: according to the target opening degree value of the damper determined in the emergency scene pre-conditioning sub-step, the central control unit dynamically calculates the initial working frequency of the pulse buffer valve that matches the target opening degree value by querying the internally stored nonlinear mapping function or lookup table.

6. The method of claim 5, wherein, In the emergency scene pre-conditioning sub-step, a compensatory adjustment mechanism is included: if the length of the risk prediction window is extended due to the absolute value of the pressure-time curve slope being greater than the preset pressure threshold, the central control unit immediately raises the target multiple of the damper opening degree from the preset multiple to a higher preset multiple when executing the emergency scene pre-conditioning sub-step; In the opening degree-frequency mapping generation sub-step, a nested correction mechanism is included: if the compensatory adjustment mechanism is triggered and the damper opening degree is raised to a higher preset multiple, the central control unit nestedly corrects the mapping relationship to further reduce the working frequency of the pulse buffer valve when executing the opening degree-frequency mapping generation sub-step.

7. The method of claim 5, wherein, The method further includes a dynamic compensation and failure interception step, which dynamically corrects the execution deviation by monitoring the system state feedback in real time during the execution of the pre-conditioning instruction, and specifically includes: A pre-conditioning execution and deviation monitoring sub-step: the central control unit drives the damper actuator and the pulse buffer valve to start working according to the determined parameters, and simultaneously starts monitoring the trend of the dust concentration sensor to calculate the dust concentration change rate; if the dust concentration rise rate calculated by the central control unit is greater than a first preset concentration change threshold, it indicates that the pre-conditioning intensity is insufficient, and the system immediately nestedly executes the emergency scene pre-conditioning sub-step to further raise the damper opening degree value by a preset step length based on the current value.

8. The method of claim 7, wherein, The dynamic compensation and failure interception step further includes: The failure interception three-level linkage sub-step: when the dust concentration rising rate continues to be greater than the second preset concentration change threshold and the duration exceeds the first preset time length, indicating that the risk cannot be effectively controlled, the central control unit will simultaneously perform three mandatory operations: multiplying the damper opening degree driven by the damper actuator by a first gain coefficient on the current basis; multiplying the working frequency of the pulse buffer valve by a second gain coefficient on the current basis; and triggering and opening a standby air flow channel controlled by the central control unit to introduce additional clean gas for dilution and purging.

9. The method of claim 8, wherein, The method further comprises an effect verification and model self-evolution step, which quantitatively evaluates the actual effect of pre-conditioning and continuously optimizes the control model using the evaluation results to drive the control model, and specifically comprises: A pre-conditioning effect real-time evaluation sub-step: after the material unloading task is started, the central control unit compares the theoretical value of the dust concentration predicted by the internal model with the actual measured concentration value of the dust concentration sensor, and calculates the deviation between the two; A model parameter dynamic calibration sub-step: when the absolute value of the deviation is greater than a first preset deviation threshold, the central control unit dynamically calibrates the core parameters of the risk prediction algorithm model, specifically by multiplying the weight coefficient of the humidity change rate in the model by a calibration factor, and forcibly binding and updating the pre-conditioning trigger threshold in the risk prediction logic.

10. The method of claim 9, wherein, The effect verification and model self-evolution step further comprises: A successful experience solidification storage sub-step: when the absolute value of the deviation is less than or equal to a second preset deviation threshold, indicating that the current model parameters and mapping relationship can accurately respond to the current working condition, the central control unit locks the current prediction coefficient and mapping function used and stores them in the database of the non-volatile memory; Furthermore, the method comprises a scene feature learning mechanism: if the failure interception three-level linkage sub-step has been activated in this task, the system automatically extracts and marks the scene features that led to the failure, and stores the scene features in association with the successful coping strategies, providing priority adaptation rules for the next unloading task in a similar scenario.

Images