Overpressure graded interlocking control method and system for thermal equipment

Abstract

This invention relates to the field of equipment control technology and discloses a method and system for graded overpressure interlocking control of thermal equipment. The method includes: calculating the pressure change rate, time margin, and material disturbance index; triggering a first-level interlock if the cavity pressure does not reach the first-level pressure relief threshold but meets the pre-trigger condition or if the cavity pressure reaches the first-level pressure relief threshold; if the cavity pressure continues to rise, entering a pressure relief stage constrained by material disturbance; executing a second-level interlocking pre-protection if the cavity pressure reaches a preset advance protection pressure or the time margin is less than or equal to a preset critical time margin threshold; triggering a hard shutdown if the cavity pressure is greater than or equal to a preset second-level shutdown threshold; exiting the interlocking control state if the cavity pressure drops below a preset recovery threshold and the pressure change rate is not greater than zero for a duration that reaches a preset recovery judgment duration. This invention balances the pressure relief rate and material stability through graded overpressure interlocking control, reducing the probability of shutdown.

Description

Technical Field

[0001] This invention relates to the field of equipment control technology, and more specifically, to a method and system for overpressure graded interlocking control of thermal equipment. Background Technology

[0002] In thermal equipment, pressure control within sealed chambers is typically required during high-temperature processing, sintering, heat treatment, or reaction heating to maintain stable process atmosphere and conditions. To this end, pressure sensors and controllable gas paths for inlet, outlet, and exhaust are installed, and constant pressure operation is achieved by adjusting valve openings, flow rates, or heating power through a control system. When affected by factors such as thermal expansion, process venting, obstructed exhaust, or actuator lag, the chamber pressure may rise abnormally. Therefore, existing thermal equipment typically employs a graded protection system.

[0003] Chinese Patent CN120406377B discloses a pressure balance control method for the coordinated operation of a quench boiler and a pyrolysis furnace. The method includes: real-time data acquisition using pressure sensors deployed at the inlet and outlet of the quench boiler and temperature sensors in the reaction section of the pyrolysis furnace; constructing a pressure state matrix; and predicting pressure distribution characteristics by combining pressure gradient changes and flow path information. A coordinated control model is used to analyze pressure intensity distribution, flow delay characteristics, and trends; sub-units are dynamically divided; and pressure intensity gradients are calculated to generate the spatiotemporal distribution characteristics of pressure balance. Based on these characteristics, the system automatically adjusts valve opening, flow rate, and coolant distribution to achieve optimal resource allocation between the quench boiler and the pyrolysis furnace.

[0004] However, in thermal equipment involving powders, granules, flakes, fibrous materials, or unfixed loads, excessive venting intensity can easily create strong local pressure differentials and airflow impacts, leading to material carry-out, displacement, stratification, or contamination, and consequently, process failure. If venting is restricted to reduce material disturbance, the pressure may not drop sufficiently. Existing technologies lack targeted interlocking control measures for the conflict between pressure safety and material stability in overpressure scenarios of thermal equipment. They also lack constraints and controls on the pressure relief stage based on the pressure evolution trend and the risk of material disturbance, making it difficult to balance rapid pressure reduction and process integrity. Summary of the Invention

[0005] The purpose of this invention is to address the problems existing in the overpressure control process of thermal equipment, such as response lag, difficulty in balancing pressure relief and material stability, and insufficient protection reliability. This invention provides a graded interlocking control method and system for overpressure control of thermal equipment, so as to achieve early warning, graded interlocking, constrained pressure relief, and independent hardware protection during the overpressure evolution process.

[0006] This invention provides a method for graded interlocking control of overpressure in thermal equipment, comprising the following steps:

[0007] Collect the cavity pressure of the thermal equipment, and calculate the pressure change rate, time margin, and material disturbance index based on the cavity pressure;

[0008] When the chamber pressure does not reach the preset first-level pressure relief threshold but meets the pre-triggering condition, or when the chamber pressure reaches the preset first-level pressure relief threshold, the first-level interlock is triggered to suppress the source term of the intake air path, heating power and exhaust buffer.

[0009] If the chamber pressure continues to rise after the first-level interlock, it enters the pressure relief stage constrained by material disturbance. Based on the time margin and material disturbance index, the exhaust gas path is controlled to relieve pressure while meeting safety and material constraints, and an upgrade strategy is triggered.

[0010] If the chamber pressure reaches the preset advance protection pressure, or the time margin is less than or equal to the preset critical time margin threshold, the second-level interlocking pre-protection is executed; if the chamber pressure is greater than or equal to the preset second-level shutdown threshold, the second-level hardware overpressure switch triggers a hard shutdown.

[0011] When the chamber pressure drops below the preset recovery threshold and the pressure change rate is not greater than zero for a duration that reaches the preset recovery judgment duration, the interlock control state is exited and the system gradually returns to the normal constant pressure control mode.

[0012] This invention provides a thermal equipment overpressure graded interlocking control system, which stores computer-readable instructions and, when read, can execute the aforementioned thermal equipment overpressure graded interlocking control method; the system includes:

[0013] The pressure derivation module collects the cavity pressure of the thermal equipment and calculates the pressure change rate, time margin, and material disturbance index based on the cavity pressure.

[0014] The first-level interlock module triggers the first-level interlock when the cavity pressure does not reach the preset first-level pressure relief threshold but meets the pre-triggering condition, or when the cavity pressure reaches the preset first-level pressure relief threshold, thereby performing source term suppression on the intake air path, heating power, and exhaust buffer.

[0015] If the chamber pressure continues to rise after the first-level interlock, the pressure relief control module enters the pressure relief stage constrained by material disturbance. Based on the time margin and material disturbance index, it controls the exhaust gas path to relieve pressure while meeting safety and material constraints, and triggers the upgrade strategy.

[0016] The secondary protection module will execute secondary interlocking pre-protection if the chamber pressure reaches the preset pre-protection pressure or the time margin is less than or equal to the preset critical time margin threshold. If the chamber pressure is greater than or equal to the preset secondary shutdown threshold, the secondary hardware overpressure switch will trigger a hard shutdown.

[0017] The recovery control module exits the interlock control state and gradually returns to the normal constant pressure control mode when the cavity pressure drops below the preset recovery threshold and the pressure change rate is not greater than zero for a period of time that reaches the preset recovery judgment duration.

[0018] The beneficial effects of this invention are as follows: By preprocessing the cavity pressure and calculating derived quantities such as pressure change rate, pressure acceleration, predicted pressure, time margin, and material disturbance index, this invention can pre-trigger judgment before the pressure reaches the first-level pressure relief threshold, thereby identifying the overpressure evolution trend earlier and entering interlocking measures in advance, reducing the risk of pressure exceeding the second-level shutdown threshold due to response lag; by simultaneously implementing intake suppression, heat source unloading, and exhaust buffering in the first-level interlocking, the source of pressure rise is suppressed first, and then pressure relief control is entered. Compared with simply relying on the exhaust valve for rapid release, this can reduce the local pressure difference and jet intensity in the exhaust channel, which is beneficial to reducing material carry-out, displacement, and process disturbance.

[0019] By introducing the material disturbance index into the pressure relief stage, a hierarchical interlocking mechanism is established that combines safety constraints and material constraints. Furthermore, the extraction gas path can be activated for coordinated pressure reduction when necessary. This transforms pressure relief control from solely pursuing rapid pressure reduction to a constraint control that balances pressure safety and material stability. Therefore, it better addresses the conflict between pressure relief rate and material stability in thermal equipment, reduces the probability of secondary hard shutdowns, and improves safety and process consistency under complex operating conditions. By employing an independent secondary hardware overpressure switch to trigger hard shutdowns, a final safety barrier independent of the software control system is formed, thereby improving the reliability of overpressure protection. Attached Figure Description

[0020] Figure 1 This is a flowchart illustrating the overpressure graded interlocking control method for thermal equipment according to the present invention.

[0021] Figure 2 This is an example diagram of triggering the first-level interlock of the overpressure graded interlock control method for thermal equipment of the present invention;

[0022] Figure 3 This is an example diagram of the triggering and upgrading strategy of the overpressure graded interlocking control method for thermal equipment of the present invention;

[0023] Figure 4 This is a module example diagram of the thermal equipment overpressure graded interlocking control system of the present invention. Detailed Implementation

[0024] The subject matter described herein will now be discussed with reference to exemplary embodiments. It should be understood that these embodiments are discussed only to enable those skilled in the art to better understand and implement the subject matter described herein, and changes may be made to the function and arrangement of the elements discussed without departing from the scope of this specification. Various processes or components may be omitted, substituted, or added as needed in the examples. Furthermore, features described in some examples may be combined in other examples.

[0025] A method and system for graded interlocking control of overpressure in thermal equipment, including the following embodiments:

[0026] Example 1

[0027] A method for graded interlocking control of overpressure in thermal equipment is applied to thermal equipment with a sealed cavity, including sintering furnaces, heat treatment furnaces, or reaction heating devices. The thermal equipment is equipped with a pressure sensor and at least two controllable gas paths, including at least two of an inlet gas path, an exhaust gas path, and an extraction gas path. The exhaust gas path includes at least an exhaust pipe connected to the cavity of the thermal equipment and an exhaust valve assembly installed on the exhaust pipe; optionally, it also includes a balancing valve, an exhaust filter, and at least one pressure relief branch, which includes a small-diameter pressure relief valve and a quick-opening valve. The control system of the thermal equipment periodically samples the cavity pressure at a preset sampling period and outputs commands for valve opening, pump speed, and heating power. The default value of the preset sampling period is 100 milliseconds, determined by the minimum executable cycle of the control system hardware and the response time of the pressure sensor. The specific value is determined based on the sampling response bandwidth of the pressure sensor and must meet the requirements of the Nyquist sampling theorem for the dynamic characteristic frequency of the cavity pressure. The reason for using periodic sampling instead of event-triggered sampling is that pressure changes within the cavity of a thermal equipment are a continuous physical process, with dynamic characteristic frequencies typically ranging from several hertz to tens of hertz. Periodic sampling can discretize the pressure signal at fixed time intervals, providing a time-consistent data sequence for subsequent differential calculations and trend extrapolation, thus avoiding additional numerical errors introduced by uneven sampling intervals. The thermal equipment is also equipped with a two-stage hardware overpressure switch, independent of the software control system, as a final safety barrier. The secondary hardware overpressure switch is independent of the software control system in terms of sensor, logic solver, power supply, and execution link. Specifically, the secondary hardware overpressure switch is equipped with an independent pressure sensor. The pressure tap of this independent pressure sensor is physically separated from the pressure tap of the pressure sensor used by the software control system on the cavity wall. The logic solver of the secondary hardware overpressure switch is implemented by an independent hard-wired relay circuit or an independent safety instrument system, without relying on the processor and communication bus of the software control system. The power supply of the secondary hardware overpressure switch is independent of the power supply of the software control system. The emergency pressure relief actuator driven by the secondary hardware overpressure switch is an independently set safety relief valve, physically independent of the exhaust valve group controlled by the software control system. This ensures that even if the software control system experiences a common cause failure, the secondary hardware overpressure switch can still independently perform the overpressure protection function, meeting the design requirements of an independent protection layer. The secondary hardware overpressure switch is used to trigger a hard stop when the cavity pressure is greater than or equal to a preset secondary stop threshold. The execution logic of the hard stop is implemented by an independent hard-wired relay circuit or an independent safety instrument system, without relying on the processor and communication bus of the software control system. The hard shutdown execution actions include at least cutting off the heating power output and cutting off the air intake execution link, and locking the equipment into a shutdown protection state; optionally, it also includes driving an independently set safety pressure relief valve to open in order to implement emergency pressure relief of the cavity.The safety relief valve and the exhaust valve group controlled by the software control system are physically independent of each other, thereby ensuring that the secondary hardware overpressure switch can still independently perform the overpressure protection function when the software control system fails due to a common cause.

[0028] This method is as follows Figure 1 As shown, it includes the following steps:

[0029] Step 100: Collect the cavity pressure of the thermal equipment, and calculate the pressure change rate, time margin, and material disturbance index based on the cavity pressure.

[0030] The system sets a preset process set pressure, a preset first-level pressure relief threshold, a preset second-level shutdown threshold, and a preset recovery threshold. The preset recovery threshold is less than or equal to the preset process set pressure, the preset process set pressure is less than the preset first-level pressure relief threshold, and the preset first-level pressure relief threshold is less than the preset second-level shutdown threshold. The reason for using a four-level progressive threshold setting is that the overpressure process in thermal equipment exhibits a gradual evolution from normal fluctuations to dangerous overpressure. A single threshold cannot distinguish between overpressure states of different severity levels, nor can it provide a decision-making basis for graded responses in the control system. By setting four thresholds from low to high, the control system can initiate mild source term suppression measures in the early stages when the pressure deviates from the normal range, gradually escalate the pressure relief as the pressure further increases, execute the strongest protective action when the pressure approaches the equipment's tolerance limit, and finally orderly exit the interlock state after the pressure drops back to a safe level. This hierarchical structure allows the control system to adopt a control strategy that matches the degree of risk in each pressure range, avoiding unnecessary interference to the process due to over-response in the case of slight overpressure, while ensuring timely and sufficiently strong protective measures in the case of severe overpressure. The four thresholds mentioned above are all determined based on the rated operating pressure and safety margin requirements of the thermal equipment. The default value for the preset process setting pressure is 80% of the rated operating pressure of the thermal equipment, determined by the target atmosphere pressure in the process formula. The default value for the preset first-level pressure relief threshold is 1.1 times the preset process setting pressure, i.e., 10% higher than the process setting pressure. This value is determined based on the statistical upper limit of pressure fluctuation during normal operation and must be greater than the maximum pressure fluctuation under normal operating conditions to avoid false triggering. The default value for the preset second-level shutdown threshold is 1.3 times the preset process setting pressure, i.e., 30% higher than the process setting pressure. This value is determined based on the maximum allowable operating pressure of the equipment casing and seals, and must have sufficient safety margin to ensure that the structural integrity of the equipment is not damaged at this pressure. The default value for the preset recovery threshold is 0.95 times the preset process setting pressure, i.e., slightly lower than the process setting pressure. It is used for hysteresis judgment when exiting the interlock state to avoid frequent switching of the control system near the threshold. This value is determined based on the steady-state accuracy of the constant pressure control loop and must ensure that the constant pressure control loop can stably take over after exiting the interlock.

[0031] Within each preset sampling period, the control system acquires the raw value of the cavity pressure at the current moment and performs signal preprocessing on the raw value before calculating the derived values. The signal preprocessing employs an exponentially weighted moving average filtering method. This method is chosen because the raw signal acquired by the pressure sensor inevitably contains electrical noise, electromagnetic interference, and the sensor's own quantization noise. If differential operations are directly performed on the noisy raw signal to calculate the pressure change rate and pressure acceleration, the noise components will be significantly amplified by the differential operation, leading to large spurious fluctuations in the derived values, which in turn can cause false or missed triggering of the interlocking decision logic. Compared to simple moving average filtering and median filtering, the exponentially weighted moving average filtering method has the advantages of a simple recursive calculation structure, requiring only the storage of the previous moment's filtered value without maintaining a sliding window data buffer. This makes it suitable for achieving effective noise suppression with extremely low computational overhead in real-time control systems with short preset sampling periods. Furthermore, this method provides an adjustable trade-off between noise suppression capability and signal tracking speed through a preset filtering smoothing factor, allowing for targeted configuration based on the sensor characteristics and pressure dynamics of different thermal equipment. Specifically, the filtered cavity pressure at the current moment is equal to the preset smoothing factor multiplied by the original cavity pressure value at the current moment, plus 1, minus the preset smoothing factor, and then multiplied by the filtered cavity pressure at the previous moment. The preset smoothing factor ranges from 0 to 1, excluding 0 and 1, with a default value of 0.3. The specific value is determined based on the signal-to-noise ratio of the pressure sensor and the dynamic characteristic frequency of the cavity pressure. The higher the sensor noise, the smaller the preset smoothing factor should be to enhance the smoothing effect; the higher the dynamic characteristic frequency of the cavity pressure, the larger the preset smoothing factor should be to reduce the phase delay introduced by filtering. It is necessary to ensure that the cutoff frequency of the filtered signal is not less than twice the dynamic characteristic frequency of the cavity pressure. All subsequent calculations of derived quantities and the cavity pressure determination involved in steps 200-500 are based on the filtered cavity pressure, and the original cavity pressure value is no longer used directly.

[0032] Based on the filtered chamber pressure, the control system calculates derived quantities, including the pressure change rate, pressure acceleration, predicted pressure, time margin, and material disturbance index. These derived quantities are selected as inputs representing the pressure evolution trend because a single instantaneous chamber pressure value only reflects the current pressure state and cannot provide crucial decision-making information such as whether the pressure is rising or falling, whether the rate of rise is accelerating, what pressure level will be reached in the near future, how much time margin remains before triggering shutdown protection, and the extent of the disturbance risk to the material within the chamber caused by pressure relief operations. The pressure change rate provides first-order trend information on pressure evolution, enabling the control system to distinguish between different operating conditions such as stable pressure, slow rise, and rapid rise. Pressure acceleration provides second-order trend information on pressure evolution, enabling the control system to determine whether the pressure rise is accelerating or decelerating, thereby predicting whether current control measures are effective. Predicted pressure integrates first-order and second-order trend information into a quantitative estimate of future pressure, allowing the control system to take preventative measures before the pressure reaches a dangerous threshold. The time margin translates the predicted pressure into the remaining time before triggering a secondary shutdown, providing the control system with an intuitive measure of urgency. This facilitates comparison with the actuator's response delay to determine whether immediate escalation of control measures is necessary. The material disturbance index quantifies the risk of material being carried out or displaced by the airflow during depressurization into a single value, enabling the control system to balance rapid depressurization with material safety constraints.

[0033] The pressure change rate is calculated by subtracting the filtered cavity pressure value from the previous sampling time from the current sampling time, and then dividing the difference by the preset sampling period. In other words, the ratio of the filtered pressure difference between two adjacent sampling times to the preset sampling period characterizes how quickly the cavity pressure changes over time. Since the pressure change rate is calculated based on the filtered signal, the amplification effect of high-frequency noise from the sensor on the first-order differential signal is effectively suppressed.

[0034] Pressure acceleration is calculated by subtracting the pressure change rate from the previous sampling time from the current sampling time's pressure change rate, and then dividing the difference by a preset sampling period. In other words, the ratio of the pressure change rate difference between two adjacent sampling times to the preset sampling period characterizes the pressure change rate's trend over time. Pressure acceleration is introduced because relying solely on the pressure change rate for trend judgment has a lag. When abnormal conditions such as a sudden increase in gas supply or blockage of the exhaust channel occur within the cavity, the pressure change rate will rapidly rise from a low value. Pressure acceleration can detect this accelerating upward trend before the pressure change rate reaches a dangerous level, providing an earlier warning signal to the control system. Because the second-order difference amplifies noise significantly more strongly than the first-order difference, a secondary smoothing process is required after calculating the pressure acceleration. This secondary smoothing also employs an exponentially weighted moving average filtering method. Applying secondary smoothing to the pressure acceleration separately, rather than relying solely on the smoothing effect of the cavity pressure after front-end filtering, is because the second-order difference operation has a secondary amplification characteristic for residual noise in the signal. Even if the noise in the cavity pressure is significantly suppressed after filtering, considerable high-frequency fluctuations may still occur after two differences. If no additional smoothing is applied to the pressure acceleration, these fluctuations will be directly transmitted to the calculation of predicted pressure and time margin, causing the interlocking decision logic to frequently jitter near the threshold. Specifically, the smoothed pressure acceleration at the current moment equals the preset acceleration smoothing factor multiplied by the original calculated pressure acceleration value at the current moment, plus 1, minus the preset acceleration smoothing factor, and then multiplied by the smoothed pressure acceleration from the previous moment. The default value of the preset acceleration smoothing factor is 0.2, and its range is from 0.1 to 0.4. The specific value is determined based on the ratio of the noise amplitude to the effective signal amplitude of the pressure acceleration signal. The larger this ratio is, the smaller the preset acceleration smoothing factor should be selected. All subsequent calculations and judgments involving pressure acceleration are based on the smoothed pressure acceleration.

[0035] Based on the aforementioned derivatives, the predicted pressure uses a second-order Taylor expansion to extrapolate the cavity pressure at a future moment. The reason for choosing a second-order Taylor expansion instead of a first-order linear extrapolation or higher-order extrapolation methods is that first-order linear extrapolation only utilizes information about the rate of pressure change, assuming a constant pressure change. When conditions such as accelerated thermal expansion of gas within the cavity or gradual blockage of the exhaust passage cause the pressure to rise rapidly, the first-order extrapolation will systematically underestimate the future pressure, leading to insufficient prediction of overpressure risk by the control system. The second-order Taylor expansion introduces a pressure acceleration term on top of the first-order extrapolation, capturing the trend of the pressure change rate itself. This provides a more accurate prediction when the pressure is rising rapidly and avoids over-warning when the pressure is rising rapidly. While higher-order Taylor expansions theoretically offer higher fitting accuracy, the amplification effect of third-order and higher-order difference operations on noise increases dramatically, making it difficult to obtain reliable high-order derivative estimates in engineering practice, and thus reducing prediction stability. Therefore, the second-order Taylor expansion achieves a suitable balance between prediction accuracy and numerical stability for overpressure control scenarios in thermal equipment. Specifically, the predicted pressure equals the filtered cavity pressure at the current moment, plus the pressure change rate at the current moment multiplied by a preset prediction lead, plus the smoothed pressure acceleration at the current moment multiplied by the square of the preset prediction lead, and then taking half of that. The second-order Taylor expansion is applicable only if the state of each actuator within the cavity remains unchanged within the time window covered by the preset prediction lead; that is, the intake valve opening, the equivalent opening of the exhaust valve group, the suction valve speed command, and the heating power command do not change within this time window. When the control system issues an actuator state change command at a certain sampling moment, it should start from the changed actuator state again in the next preset sampling period after the command takes effect, collect the new filtered cavity pressure, and recalculate the pressure change rate and smoothed pressure acceleration. This serves as the new benchmark for subsequent predicted pressure calculations, i.e., resetting the prediction model, thereby avoiding extrapolation distortion caused by sudden changes in pressure dynamic characteristics during actuator transitions. The preset prediction lead is the time interval between the current moment and the predicted future moment. Its default value is 3 seconds, and the value ranges from 1 second to 10 seconds. The specific value is determined according to the cavity volume and the maximum possible air intake flow. The larger the cavity volume, the slower the pressure changes, and a larger preset prediction lead can be selected to obtain a longer prediction window. The smaller the cavity volume, the faster the pressure changes, and a smaller preset prediction lead should be selected to ensure prediction accuracy.

[0036] Based on the predicted pressure, the time margin represents the shortest time required for the predicted pressure to first reach the preset secondary shutdown threshold from the current moment. The reason for introducing the time margin as a derived variable is that although the predicted pressure provides a pressure estimate for a fixed future moment, the control system needs to know how much time remains before triggering the secondary shutdown when making interlocking decisions, rather than the pressure value at a specific moment. The time margin transforms the pressure prediction problem into a time prediction problem, allowing the control system to directly compare the remaining time with the response delay of the actuator, thereby determining whether there is a sufficient time window to execute the pressure relief action. The time margin is solved analytically rather than numerically iteratively because the predicted pressure model based on second-order Taylor expansion is a quadratic function of time. Its intersection with the preset secondary shutdown threshold can be directly obtained by solving a quadratic equation without iterative calculation. This ensures that the time margin solution can be completed within a defined computation time in each preset sampling period, eliminating the risk of iteration non-convergence and meeting the deterministic requirements of real-time control systems. Specifically, the difference between the preset secondary shutdown threshold and the filtered cavity pressure at the current moment is defined as the pressure margin. When the smoothed pressure acceleration is greater than zero, the predicted pressure is a quadratic function with respect to the preset prediction lead, opening upwards. In this case, expanding the condition that the predicted pressure equals the preset secondary shutdown threshold yields a quadratic equation with respect to the preset prediction lead. The discriminant of this equation equals the square of the pressure change rate plus twice the smoothed pressure acceleration multiplied by the pressure margin. If the discriminant is less than zero, it indicates that the predicted pressure will not reach the preset secondary shutdown threshold within the preset prediction window, and the time margin is set to positive infinity. If the discriminant is greater than or equal to zero, the smallest root greater than zero of the two roots of the quadratic equation is taken as the time margin. If neither root is greater than zero, the time margin is set to positive infinity. When the smoothed pressure acceleration is zero, the predicted pressure degenerates into a linear function with respect to the preset prediction lead. In this case, if the pressure change rate is greater than zero, the time margin equals the pressure margin divided by the pressure change rate; if the pressure change rate is less than or equal to zero, the cavity pressure will not rise to the preset secondary shutdown threshold, and the time margin is set to positive infinity. When the smoothed pressure acceleration is less than zero, the predicted pressure is a downward-opening quadratic function of the preset prediction lead. At this point, the predicted pressure has a maximum value. If this maximum value is less than the preset secondary shutdown threshold, the time margin is set to positive infinity. If the maximum value is greater than or equal to the preset secondary shutdown threshold, the aforementioned quadratic equation is solved, and the smallest root greater than zero is taken as the time margin. Furthermore, if the time margin obtained through the above analytical method is greater than the preset prediction window, the time margin is also set to positive infinity, indicating that there is currently no risk of triggering a secondary shutdown within the preset prediction window.The default value of the preset prediction window is 10 seconds, and the value ranges from 5 seconds to 30 seconds. The specific value is determined based on the thermal inertia of the cavity and the gas path response delay, and it needs to cover the time span from the current moment to the worst case situation when the cavity pressure reaches the preset secondary shutdown threshold.

[0037] The material disturbance index is used to quantify the risk of material being carried out or displaced by airflow during pressure relief. The reason for introducing the material disturbance index is that during pressure relief in thermal equipment, the high-speed airflow within the cavity can impact and drag the material. If the pressure relief rate is too fast or the opening of the exhaust valve assembly changes too drastically, the material may be carried out of the cavity by the airflow into the exhaust pipe, causing irrecoverable losses, or it may displace within the cavity, disrupting the loading geometry and affecting the uniformity of subsequent processes. Traditional overpressure protection methods typically only focus on the safety constraints of the pressure itself, aiming solely at reducing pressure as quickly as possible during pressure relief, ignoring the disturbance effect on the material during the pressure relief process. The material disturbance index integrates multiple observable physical quantities related to the risk of material carry-out into a single dimensionless index, enabling the control system to simultaneously consider both pressure safety and material safety objectives in pressure relief decisions, seeking an optimal balance between the two. The material disturbance index is a weighted normalized linear combination of multiple observable physical quantities. The reason for using this weighted normalized linear combination method is that the dimensions, ranges, and physical meanings of each observable physical quantity are different, making direct numerical comparison or summation impossible. Normalization maps each physical quantity to a unified dimensionless interval, making them comparable. Weighted summation allows assigning different importance weights to each physical quantity based on its contribution to the risk of material embezzlement, enabling the material disturbance index to reflect the relative influence of each factor. Before calculating each component, each observable physical quantity is first normalized and saturated. The specific calculation method is as follows: Divide the pressure difference across the exhaust filter by the preset normalized upper limit of the pressure difference across the exhaust filter. If the quotient is greater than 1, it is counted as 1; if it is less than 0, it is counted as 0. Multiply the saturated truncated value by the preset weighting coefficient corresponding to the pressure difference across the exhaust filter as the first component; Divide the particulate concentration signal in the exhaust pipe by the preset normalized upper limit of the particulate concentration signal. If the quotient is greater than 1, it is counted as 1; if it is less than 0, it is counted as 0. Multiply the saturated truncated value by the preset weighting coefficient corresponding to the particulate concentration signal as the second component; The absolute value of the vibration characteristic signal near the valve is divided by a preset normalized upper limit for the vibration characteristic signal. If the quotient is greater than 1, it is counted as 1; if it is less than 0, it is counted as 0. The truncated value is then multiplied by a preset weighting coefficient corresponding to the vibration characteristic signal, and this is taken as the third component. The absolute value of the equivalent opening change rate of the exhaust valve group is divided by a preset normalized upper limit for the equivalent opening change rate of the exhaust valve group. If the quotient is greater than 1, it is counted as 1; if it is less than 0, it is counted as 0. The truncated value is then multiplied by a preset weighting coefficient corresponding to the equivalent opening change rate of the exhaust valve group, and this is taken as the fourth component. The material disturbance index is equal to the sum of these four components.

[0038] By applying saturation truncation to each normalized component, the values ​​of each component are ensured to remain between 0 and 1 even when the sensor exceeds its range or experiences abnormal noise spikes. Combined with the constraint that the sum of four preset weighting coefficients equals 1, the material disturbance index is guaranteed to strictly range between 0 and 1, inclusive. The effect of saturation truncation is that when a sensor outputs an abnormally large value due to transient interference or range overflow, the contribution of that component is cut off at its maximum weighting value. This prevents the overall material disturbance index from being distorted due to the abnormality of a single sensor, thus ensuring the numerical stability and consistency of the physical meaning of the material disturbance index under various operating conditions. The pressure difference before and after the exhaust filter, particle concentration signal, vibration characteristic signal, and the equivalent opening change rate of the exhaust valve assembly are selected as the four input components of the material disturbance index because these four physical quantities characterize the degree of material disturbance during the pressure relief process from four different dimensions: airflow load, particle carryover, mechanical vibration, and valve action speed. The pressure difference across the exhaust filter directly reflects the airflow rate through the filter; a larger pressure difference indicates a stronger airflow and a higher risk of material being dragged by the airflow. Particle concentration is the most direct evidence that material has begun to be carried out by the airflow; an increase in its value means that material carry-out is occurring. Vibration characteristic signals reflect the pulsation and turbulence intensity of the airflow in the exhaust pipeline. Severe vibration is usually accompanied by unstable jets, which can easily cause intermittent strong impacts on the material. The equivalent opening change rate of the exhaust valve assembly reflects the swiftness of the valve action. Rapid valve opening generates transient pressure gradients and localized high-speed airflow within the cavity, resulting in a much greater impact on the material than slow opening.

[0039] Furthermore, the control system verifies the validity of each sensor signal within each preset sampling period. When a sensor signal exhibits abnormal conditions such as exceeding its range, communication interruption, or signal jump amplitude exceeding a preset abnormal jump threshold, the normalized component corresponding to that sensor is locked to its most recent valid value, and the preset weight coefficient corresponding to that sensor is temporarily set to zero. Simultaneously, this weight is redistributed to the remaining valid components according to the original weight ratio of the other valid sensors. This ensures that the material disturbance index can still provide meaningful risk assessment even when a sensor experiences a partial failure. The reason for using dynamic weight redistribution instead of simply setting the faulty sensor component to zero or keeping it unchanged is that if the faulty sensor component is directly set to zero, the material disturbance index will be systematically low, potentially causing the control system to allow a large pressure relief even when the actual material disturbance risk is high, resulting in material loss. If the faulty sensor component is kept at its most recent valid value without adjusting its weight, the contribution of this component will be frozen to a historical value that may no longer reflect the current operating conditions during the duration of the fault, reducing the material disturbance index's ability to track changes in actual risk. By proportionally redistributing the weights of faulty sensors to the remaining effective sensors, the material disturbance index can still fully utilize the remaining effective information even with a reduced number of sensors, maintaining a reasonable assessment of material disturbance risk. The default value of the preset abnormal jump threshold is 50% of the preset normalized upper limit value of the corresponding sensor divided by the preset sampling period. That is, when the signal change exceeds 50% of the normalized upper limit value within a single sampling period, it is determined to be an abnormal jump.

[0040] Among them, the pressure difference before and after the exhaust filter reflects the airflow load state of the filter component; the particle concentration signal is obtained by a light scattering dust sensor or a charge particle sensor installed in the exhaust air path; the vibration characteristic signal is used to identify the strong injection or pulsation stage; the equivalent opening change rate of the exhaust valve group is set to the equivalent opening command value of the exhaust valve group at the current sampling time minus the equivalent opening command value of the exhaust valve group at the previous sampling time, and the difference is then divided by the preset sampling period.

[0041] The sum of the preset weighting coefficients for the above four components equals 1, and the value of each preset weighting coefficient ranges from 0 to 1, inclusive. The default preset weighting coefficient for the pressure difference across the exhaust filter corresponding to the first component is 0.3, the default preset weighting coefficient for the particle concentration signal corresponding to the second component is 0.3, the default preset weighting coefficient for the vibration characteristic signal corresponding to the third component is 0.2, and the default preset weighting coefficient for the equivalent opening change rate of the exhaust valve group corresponding to the fourth component is 0.2. The method for determining the above default weights is as follows: the pressure difference across the exhaust filter and the particle concentration signal are the most direct indicators of the risk of material carryover, and therefore are given higher weights; the vibration characteristic signal and the equivalent opening change rate of the exhaust valve group are given lower weights as auxiliary criteria. In practical applications, the preset weighting coefficients can be adjusted according to the material characteristics of the specific equipment and the exhaust pipeline structure through offline calibration tests or statistical analysis of historical operating data. During adjustment, it is necessary to ensure that the sum of the four preset weighting coefficients always equals 1.

[0042] The preset normalized upper limit values ​​for each component correspond to the preset normalized upper limit values ​​for the pressure difference across the exhaust filter, particle concentration signal, vibration characteristic signal, and the equivalent opening change rate of the exhaust valve assembly, respectively. The preset normalized upper limit value for the pressure difference across the exhaust filter is determined based on the maximum allowable operating pressure difference of the filter under clean conditions; the default value is obtained from the rated pressure difference parameters provided by the filter manufacturer. The preset normalized upper limit value for the particle concentration signal is determined based on the upper limit of the range of the particle concentration sensor in the exhaust pipeline; the default value is the sensor's full-scale value. The preset normalized upper limit value for the vibration characteristic signal is determined based on the vibration amplitude of the vibration sensor near the exhaust valve under the maximum allowable pressure relief flow rate; the default value is obtained from the maximum vibration amplitude measured during the equipment's full-load pressure relief test. The preset normalized upper limit value for the equivalent opening change rate of the exhaust valve assembly is determined based on the maximum operating rate of the valve actuator; the default value is the reciprocal of the shortest time required for the actuator to go from fully closed to fully open multiplied by 100% of the opening.

[0043] A preset acceptable upper limit for the material disturbance index is set as a constraint boundary for material stability during the depressurization process. The default value of the preset acceptable upper limit is 0.7, and the value ranges from 0.5 to 0.9. The specific value is determined based on the particle size distribution, bulk density, and adhesion characteristics of the material. Materials with smaller particle size and lower bulk density are more easily carried out by the airflow, and a lower preset acceptable upper limit should be selected; materials with larger particle size and higher bulk density can select a higher preset acceptable upper limit.

[0044] Step 200: When the cavity pressure does not reach the preset first-level pressure relief threshold but meets the pre-triggering condition, or when the cavity pressure reaches the preset first-level pressure relief threshold, the first-level interlock is triggered, and source term suppression is performed on the intake air path, heating power, and exhaust buffer, specifically as follows: Figure 2 As shown.

[0045] During the routine constant pressure control operation of the thermal equipment according to the preset process pressure, the control system continuously calculates the pressure change rate, smoothed pressure acceleration, time margin and material disturbance index set in step 100 in each preset sampling cycle, forming a real-time monitoring layer of the dynamic trend of the cavity pressure. This monitoring layer is superimposed on the original constant pressure control loop without changing the control logic of the constant pressure control loop itself.

[0046] Based on the time margin and pressure change rate obtained in step 100, a dual criterion of threshold and trend is used for pre-trigger determination. The reason for using dual criteria instead of a single threshold criterion is that a single pressure threshold criterion can only trigger interlocking when the cavity pressure has already reached the preset first-level pressure relief threshold. At this point, the pressure has deviated from the normal range, and the control system's response has an inherent time delay. Under rapid pressure rise conditions, the interlocking action may not have enough time to produce sufficient pressure reduction before the pressure reaches the preset second-level shutdown threshold. By introducing time margin and pressure change rate as trend criteria, the control system can trigger interlocking in advance when the cavity pressure has not yet reached the preset first-level pressure relief threshold but has already shown a dangerous upward trend, thus gaining more execution time for subsequent source suppression and pressure relief actions. The time margin criterion focuses on assessing whether there is sufficient remaining time before triggering the second-level shutdown under the current pressure trend, while the pressure change rate criterion focuses on identifying the rapid pressure rise. The two criteria complement each other from different perspectives, covering operating scenarios requiring early intervention. The condition for the pre-trigger determination is that the current cavity pressure has not yet reached the preset first-level pressure relief threshold, that is, the current cavity pressure is less than the preset first-level pressure relief threshold, but condition one or condition two is met.

[0047] Condition one is that the time margin is less than or equal to the preset pre-trigger time margin threshold, indicating that according to the current pressure rise trend, the cavity pressure will reach the preset secondary shutdown threshold in a short period of time. The default value of the preset pre-trigger time margin threshold is 5 seconds, and the value ranges from 2 seconds to 10 seconds. The specific value is determined based on the longest response time required for the control system to execute all actions of the first-level interlock, and must be greater than this longest response time to ensure that the interlock actions have a sufficient execution window.

[0048] Condition two is that the pressure change rate is greater than or equal to the preset pre-trigger pressure change rate threshold, indicating that the cavity pressure is rising rapidly. The default value of the preset pre-trigger pressure change rate threshold is 5% of the preset process set pressure divided by 1 second, that is, when the pressure increase per second reaches 5% of the preset process set pressure, an early warning is triggered. The specific value is determined based on the statistical distribution of the pressure change rate during normal equipment operation, and must be greater than 3 times the standard deviation of the pressure change rate under normal operating conditions to avoid false triggering.

[0049] When the aforementioned pre-triggering conditions are met or the cavity pressure is greater than or equal to the preset first-level pressure relief threshold, the control system triggers the first-level interlock, executing actions A (intake suppression), B (heat source unloading), and C (balance buffering) in sequence according to low disturbance priority. The reason for prioritizing these three actions according to low disturbance is that in the early stages of an overpressure event, although the cavity pressure is high, it usually has not yet reached a level that endangers equipment safety. At this time, the primary goal of the control strategy is to curb the pressure from continuing to rise while minimizing disturbance to the process and materials within the cavity. Intake suppression reduces the amount of gas input into the cavity by cutting off or reducing the pressure boosting source. This action does not involve opening the exhaust channel, therefore it produces almost no additional airflow impact on the materials within the cavity, making it the least disturbing pressure reduction method. Heat source unloading slows down the rate of gas thermal expansion by reducing the heating power. This action also does not involve changes in the airflow channel, resulting in minimal direct disturbance to the materials, but a trade-off needs to be struck between temperature uniformity and pressure reduction rate. Balancing buffering disperses potential subsequent exhaust airflow by pre-opening balancing valves. Although it involves gas path operation, the opening is small and the purpose is buffering, resulting in far less disturbance to the material than directly opening the exhaust valve assembly for pressure relief. By executing these measures sequentially according to the above priorities, the control system prioritizes the less disruptive approach at each step, only activating subsequent, more disruptive approaches when the preceding measures are insufficient to curb the pressure rise, thereby achieving minimal control over material disturbance.

[0050] Action A, intake suppression, includes setting the intake valve opening to 0 or reducing it to the opening value required to maintain the minimum airflow, and locking this state for a preset intake suppression duration. During this preset intake suppression duration, intake is not allowed to resume, to prevent intake and exhaust from creating a counter-current and exacerbating pressure fluctuations. The default value for the preset intake suppression duration is 10 seconds, with a range of 5 to 30 seconds. The specific value is determined based on the ratio of the cavity volume to the maximum exhaust capacity of the exhaust channel, ensuring that the exhaust channel has sufficient time to expel excess gas accumulated during intake within this duration.

[0051] Action B, heat source reduction, includes simultaneously determining whether the pressure change rate is greater than zero and whether the time margin is continuously shortening during the execution of Action A. The condition for continuously shortening the time margin is that, within a consecutive preset time margin shortening judgment window of sampling periods, the time margin of each sampling period is less than the time margin of the previous sampling period. The default value of the preset time margin shortening judgment window is 3 sampling periods, with a range of 2 to 5 sampling periods. The specific value is determined based on the short-term fluctuation amplitude of the time margin signal and must be greater than the typical non-monotonic fluctuation period caused by the small fluctuations in the filtered pressure signal, to avoid misjudging a non-shortening state due to occasional single-cycle rebounds. If the above conditions are met, the heating power command is reduced. Specifically, the current heating power command is subtracted from the preset single-time reduction amplitude. If the result is lower than the preset minimum allowable heating power, the heating power command is set to the preset minimum allowable heating power; otherwise, the heating power command is set to the value after subtracting the preset single-time reduction amplitude. This action suppresses the continuous pressure rise caused by gas thermal expansion by reducing the heat source input. The default value for the preset single-load reduction range is 20% of the current heating power command, with a range of 10% to 50%. The specific value is determined based on the process's sensitivity to temperature uniformity. Temperature-sensitive processes should select a smaller preset single-load reduction range to avoid the impact of sudden temperature drops on product quality. Non-temperature-sensitive processes can select a larger preset single-load reduction range to accelerate the voltage reduction rate. The default value for the preset minimum allowable heating power is 10% of the rated heating power, determined by the heating power required to maintain the minimum process temperature in the process formula.

[0052] Action C, balancing and buffering, includes opening the balancing valve to a preset equivalent opening command value to reduce the local pressure difference and jet intensity in the exhaust channel, preventing the material disturbance index set in step 100 from suddenly increasing due to subsequent pressure relief actions. The default value of the preset equivalent opening command value is 30% of the maximum opening of the balancing valve, with a range of 10% to 50%. The specific value is determined based on the ratio of the balancing valve diameter to the exhaust valve diameter and the sensitivity of the material in the cavity to airflow disturbance. When the material sensitivity is high, a larger preset equivalent opening command value should be selected to fully divert the flow; when the material sensitivity is low, a smaller value can be selected.

[0053] By executing actions A, B, and C, the intake valve is closed, the heating power is reduced, and the balancing valve is opened, thereby collectively reducing the pressure rise source term within the cavity. The synergistic effect of these three actions is that intake suppression cuts off the continuous input of external gas into the cavity, eliminating the main external source term for pressure rise; heat source unloading reduces the internal pressure rise rate caused by the thermal expansion of gas within the cavity, weakening the internal source term for pressure rise; and balancing buffering establishes a buffer channel for possible subsequent pressure relief operations by pre-opening the balancing valve, reducing the local pressure gradient at the exhaust channel inlet, so that strong transient jets will not be generated due to sudden pressure difference when subsequent pressure relief operations are initiated. The synergistic effect of these three actions significantly reduces the net pressure rise rate within the cavity. In many cases of slight overpressure, the cavity pressure can naturally fall back simply through source term suppression, without needing to enter the subsequent active pressure relief stage, thus handling overpressure events with minimal material disturbance. After executing the above actions, the control system continues to calculate the pressure change rate, time margin, and material disturbance index in each preset sampling period. If the cavity pressure still does not show a downward trend, it proceeds to step 300. The condition for determining that the cavity pressure has not yet shown a downward trend is that after all three automatic actions (A, B, and C) have been completed, and after a preset source term suppression observation time has elapsed, the pressure change rate is still greater than zero, meaning that the filtered cavity pressure is still increasing cycle by cycle. The default value of the preset source term suppression observation time is 5 seconds, with a range of 3 to 10 seconds. The specific value is determined based on the pressure response delay caused by the thermal inertia of the gas in the cavity after the intake valve is closed and the heating power is reduced. It must be greater than this response delay to ensure that the effect of the source term suppression action is fully reflected in the cavity pressure change.

[0054] Step 300: If the cavity pressure continues to rise after the first-level interlock, it enters the pressure relief stage constrained by material disturbance. Based on the time margin and material disturbance index, the exhaust gas path is controlled to relieve pressure while satisfying safety and material constraints, and an upgrade strategy is triggered, specifically as follows: Figure 3 As shown.

[0055] After the source term suppression action in step 200 is executed, if the cavity pressure continues to rise, the control system enters the pressure relief stage constrained by material disturbance. The criterion for the cavity pressure continuing to rise is the same as the criterion for the cavity pressure not showing a downward trend in step 200, that is, after all automatic actions A, B, and C are completed and a preset source term suppression observation time has elapsed, the pressure change rate is still greater than zero. The reason for entering the pressure relief stage constrained by material disturbance is that when the source term suppression measures have been implemented but the cavity pressure is still rising, it indicates that the residual pressure-increasing factors in the cavity exceed the capacity of the source term suppression, and the cavity pressure must be reduced by actively venting. However, although directly opening the vent valve assembly at maximum capacity can reduce the pressure most quickly, the resulting high-speed airflow will cause serious disturbance to the material in the cavity, which may lead to irreversible losses such as material carry-out, displacement, or breakage. Therefore, this step applies safety constraints and material constraints simultaneously during the pressure relief process, so that the control system can complete the pressure reduction task with the lowest possible material disturbance cost while ensuring that the cavity pressure does not reach the preset secondary shutdown threshold. When calculating the equivalent opening command of the exhaust valve assembly, both safety constraints and material constraints must be satisfied.

[0056] Safety constraints include ensuring that the time margin returns to above the preset lower limit of the safety time margin, i.e., the time margin is greater than or equal to the preset lower limit of the safety time margin. The default value of the preset lower limit of the safety time margin is 3 seconds, and the range is from 1 second to 8 seconds. The specific value is determined based on the total delay time from the triggering of pressure relief in the control system to the exhaust valve group reaching the target opening degree. It must be greater than this total delay time to ensure that the pressure relief action takes effect before the cavity pressure reaches the preset secondary shutdown threshold.

[0057] Material constraints include ensuring that the material disturbance index in step 100 does not exceed a preset acceptable upper limit, i.e., the material disturbance index is less than or equal to the preset acceptable upper limit, and the equivalent opening change rate of the exhaust valve assembly does not exceed a preset opening change rate upper limit. The default value of the preset opening change rate upper limit is 10% of the full-scale opening per second, and the value range is 5% to 20% of the full-scale opening per second. The specific value is determined based on the volume of the exhaust pipeline and the resistance characteristics of the pipeline after the valve. The smaller the pipeline volume and the lower the resistance, the more sensitive the airflow response to the opening change. A smaller preset opening change rate upper limit should be selected to avoid excessively strong transient jets.

[0058] Based on the above dual constraints, the following segmented control law is used to control the exhaust gas path depressurization:

[0059] When the time margin exceeds the preset lower limit of the safety time margin and the material disturbance index is greater than or equal to the preset acceptable upper limit multiplied by the preset proximity scaling factor, a pulse pressure relief mode is adopted. The reason for using a pulse pressure relief mode instead of continuous constant opening pressure relief is that when the material disturbance index is already greater than or equal to the preset acceptable upper limit multiplied by the preset proximity scaling factor, continuously venting at a large opening will cause the material disturbance index to continuously exceed the limit, while continuously venting at a small opening may not provide a sufficient pressure reduction rate to meet the safety constraints. The pulse pressure relief mode, by periodically switching between large and small openings, achieves a high instantaneous pressure reduction rate during the large opening phase and allows the material disturbance index to fall back during the small opening phase. This simultaneously satisfies the safety constraint's requirement for pressure reduction rate and the material constraint's limitation on the disturbance index in a time-averaged sense. The effect of this intermittent pressure relief method is similar to applying time-dimensional modulation to the exhaust airflow. Under the premise of not exceeding the instantaneous peak constraint of the material disturbance index, the average pressure relief amount is flexibly controlled by adjusting the preset duty cycle, achieving a dynamic balance between safety constraints and material constraints. The default value of the preset proximity scaling factor is 0.8, and its range is from 0.7 to 0.95. The specific value is determined based on the fluctuation range of the material disturbance index during the pressure relief process. The larger the fluctuation range, the smaller the preset proximity scaling factor should be selected to reserve sufficient margin. The smaller the fluctuation range, the larger the preset proximity scaling factor can be selected to fully utilize the pressure relief capacity. When using the pulse pressure relief mode, the equivalent opening of the exhaust valve group periodically switches between the preset low-level pressure relief opening and the preset medium-level pressure relief opening according to a preset duty cycle. The preset low-level pressure relief opening is smaller than the preset medium-level pressure relief opening, so that the average pressure relief amount meets the safety constraints, while the transient jet intensity is controlled to avoid the material disturbance index exceeding the limit. The default value of the preset low-level pressure relief opening is 10% of the equivalent maximum opening of the exhaust valve group, which is determined by the valve opening corresponding to the minimum exhaust flow rate required to maintain the cavity pressure from continuing to rise. The default value of the preset medium-range pressure relief opening is 30% of the equivalent maximum opening of the exhaust valve assembly, determined by the valve opening required for the maximum exhaust flow rate when the material disturbance index just does not exceed the preset acceptable upper limit. The default value of the preset duty cycle is 0.5, with a range of 0.2 to 0.8. The specific value is determined based on the requirement that the average pressure relief amount within the pulse pressure relief cycle must meet the safety constraints. When the preset safety time margin lower limit is small, a larger preset duty cycle should be selected to increase the average pressure relief amount; conversely, a smaller preset duty cycle can be selected.

[0060] When the time margin is less than or equal to the preset lower limit of the safety time margin, it indicates an increased risk that the cavity pressure is approaching the preset secondary shutdown threshold. The pressure relief level is then increased, and the equivalent opening of the exhaust valve assembly is updated recursively as follows: the current equivalent opening of the exhaust valve assembly is added to the product of the preset time margin error gain and the difference between the preset lower limit of the safety time margin and the current time margin. The result, after saturation limiting processing, is used as the equivalent opening of the exhaust valve assembly at the next sampling time. The reason for adopting the recursive update method based on the time margin error is that when the time margin is already below the preset lower limit of the safety time margin, the core challenge facing the control system is to reverse the upward trend of the cavity pressure within the limited remaining time. At this time, the adjustment range of the equivalent opening of the exhaust valve assembly should be proportional to the urgency. The difference between the preset lower limit of the safety time margin and the current time margin is the time margin error. The larger this error, the more urgent the current operating condition, requiring a larger increment in the equivalent opening of the exhaust valve assembly to accelerate pressure reduction. This error-based recursive control structure has adaptive characteristics. When the time margin error is large, it automatically increases the pressure relief intensity; when the time margin error decreases due to the pressure relief effect, it automatically decreases the increment of the pressure relief intensity, avoiding the problems of insufficient response or over-adjustment that may be caused by fixed step size adjustment. Saturation limiting processing ensures that the equivalent opening of the exhaust valve group is always within the physically feasible range, preventing the numerical calculation result from exceeding the actual stroke range of the actuator. The default value of the preset time margin error gain is 0.1, and the value range is 0.05 to 0.3. The specific value is determined according to the slope of the flow characteristic curve and the cavity volume of the exhaust valve group. The larger the slope of the valve flow characteristic curve and the smaller the cavity volume, the greater the pressure change caused by the unit opening change. A smaller preset time margin error gain should be selected to avoid over-adjustment, and vice versa to speed up the response. The saturation limiting process means that if the calculated result is less than 0, it is set to 0; if the calculated result is greater than the preset equivalent maximum opening of the exhaust valve group, the preset equivalent maximum opening of the exhaust valve group is used; otherwise, the calculated result itself is used. In other words, the output is limited to the range from 0 to the preset equivalent maximum opening of the exhaust valve group. The default value of the preset equivalent maximum opening of the exhaust valve group is 100%, i.e., the valve group is fully open, determined by the upper limit of the physical stroke of each valve in the exhaust valve group and the equivalent maximum flow capacity after parallel combination. Simultaneously, within each preset sampling period, the absolute value of the rate of change of the equivalent opening of the exhaust valve group is limited to not exceeding the preset upper limit of the rate of change of opening.

[0061] During the depressurization process, the control system monitors the particle concentration signal in the exhaust pipeline and the pressure difference before and after the exhaust filter in real time. If signs of material carryover are detected, i.e., the rate of change of the particle concentration signal is greater than the preset rapid rise threshold of particle concentration and the current value of the particle concentration signal exceeds the preset particle concentration alarm threshold, the adjustment step of the equivalent opening of the exhaust valve group is immediately reduced, the depressurization mode is switched to a higher frequency small-amplitude pulse depressurization, and the depressurization task is shared by increasing the preset single load reduction amplitude of the heat source load reduction in step 200 and activating the extraction channel. The rate of change of the particle concentration signal is the particle concentration signal value at the current sampling time minus the particle concentration signal value at the previous sampling time, and the difference is divided by the preset sampling period. The default value of the preset rapid rise threshold of particle concentration is 10% of the preset normalized upper limit of particle concentration signal divided by 1 second, that is, when the increase in particle concentration signal per second reaches 10% of the preset normalized upper limit, it is judged as a rapid rise. The specific value is determined according to the statistical upper limit of the rate of change of particle concentration signal during normal depressurization, and must be greater than the statistical upper limit to avoid misjudgment. The default value of the preset particle concentration alarm threshold is 60% of the preset normalized upper limit of the particle concentration signal, and the value range is 40% to 80% of the preset normalized upper limit. The specific value is determined according to the economic value of the material and the process's tolerance for material loss. For high-value materials, a lower preset particle concentration alarm threshold should be selected to detect signs of material carry-out as early as possible.

[0062] If the above-mentioned control of the exhaust gas path pressure relief is still insufficient to curb the pressure rise trend, an upgrade strategy is triggered. The upgrade trigger condition is to meet any of the following conditions: the cavity pressure continues to rise and the time margin is still decreasing. The determination condition for the continuous rise of the cavity pressure is that the pressure change rate is greater than zero within a preset upgrade pressure rise determination window sampling period. The default value of the preset upgrade pressure rise determination window is 3 sampling periods, and the value range is 2 to 5 sampling periods. The determination condition for the time margin still decreasing is the same as the determination condition for the continuous decrease of the time margin in step 200 action B; or after the cavity pressure reaches the preset first-level pressure relief threshold, there is still no reversal of the pressure change rate from positive to negative after a preset upgrade waiting time; or the material disturbance index has approached the preset acceptable upper limit, which restricts the further opening of the exhaust channel. The determination condition for the material disturbance index approaching the preset acceptable upper limit is the same as the determination condition for the material disturbance index approaching the preset acceptable upper limit in the pulse pressure relief mode in step 300, that is, the material disturbance index is greater than or equal to the preset acceptable upper limit multiplied by the preset proximity ratio factor. The default value of the preset upgrade waiting time is 15 seconds, and the range is from 5 seconds to 30 seconds. The specific value is determined based on the typical response time required from the execution of the first-level pressure relief action to the production of an observable pressure reduction effect. It must be greater than this typical response time to avoid unnecessary upgrades due to premature judgment.

[0063] Once the upgrade trigger conditions are met, the actions are executed sequentially in order of lower disturbance but stronger pressure reduction capability: Action D (activating the extraction channel), Action E (valve group sequence switching), and Action F (adaptive valve effect gain update). This sequential execution is because the upgrade strategy aims to enhance the overall pressure reduction capability by introducing additional pressure reduction methods and optimizing the efficiency of existing methods, given that the pressure relief capacity of the exhaust channel is limited by material constraints. Action D, activating the extraction channel, introduces a pressure reduction path physically independent of the exhaust channel. Since the airflow direction and velocity distribution characteristics of the extraction channel differ from those of the exhaust channel, its impact mode on the material is also different. Therefore, it can share some of the pressure reduction task without significantly increasing the material disturbance index, making it the upgrade method with the best disturbance efficiency. Action E, valve group sequence switching, achieves a significant pressure reduction effect in a short time by arranging the timing of valves of different diameters in the exhaust valve group, while minimizing the duration of continuous material impact. This is a method of optimizing the pressure relief method within the exhaust channel. Action F adaptively updates the valve effect gain by estimating in real time the actual impact of changes in the equivalent opening of the exhaust valve assembly on the pressure change rate. This enables the control system to adjust the control strategy based on the actual pressure reduction efficiency under the current operating conditions, avoiding ineffective material disturbance caused by blindly increasing the equivalent opening of the exhaust valve assembly when the pressure reduction efficiency decreases.

[0064] Action D, activating the evacuation channel, involves opening the evacuation valve and increasing the valve speed command to a preset evacuation operating value, transferring a portion of the pressure reduction task from the exhaust channel to the evacuation channel. The default value of the preset evacuation operating value is 70% of the vacuum pump's rated speed, with a range of 50% to 100% of the rated speed. The specific value is determined based on the pipeline resistance characteristics of the evacuation channel and the flow and pressure characteristic curves of the vacuum pump. It is necessary to ensure that the evacuation channel can provide effective exhaust flow under the current cavity pressure, while avoiding excessively high pump speeds that could lead to excessively strong local negative pressure near the evacuation port and disturb the material. Because the airflow velocity and direction characteristics of the evacuation channel differ from direct exhaust, the impact on the material within the cavity is generally smaller, which is beneficial for enhancing pressure reduction capabilities without significantly increasing the material disturbance index.

[0065] The action E valve group sequence switching includes peak-shaving and stabilization sequence control of the exhaust valve group. Specifically, the needle valve or small-diameter pressure relief valve is first opened to release the chamber pressure until the absolute value of the pressure change rate is less than a preset pre-triggered pressure change rate threshold. Then, the quick-opening valve is briefly opened for rapid peak shaving. The duration of this brief opening is the preset single-opening duration of the quick-opening valve, with a default value of 1 second and a range of 0.5 to 3 seconds. The specific value is determined based on the diameter of the quick-opening valve and the chamber volume. It is necessary to ensure that the amount of gas discharged by the quick-opening valve within this duration is sufficient to produce an observable pressure reduction effect, while not exceeding the maximum exhaust volume allowed when the material disturbance index rises from its current value to the preset acceptable upper limit. Subsequently, the quick-opening valve is immediately closed, and the system returns to the needle valve or small-diameter pressure relief valve to maintain pressure relief. This sequence, by limiting the opening time of the large-diameter valve to the preset single-opening duration of the quick-opening valve, achieves a rapid pressure reduction effect while reducing the continuous impact on the material.

[0066] Action F, adaptively updating the valve effect gain, involves the control system estimating in real time the impact of changes in the equivalent opening of the exhaust valve assembly on the pressure change rate under the current operating conditions during the execution of the upgrade strategy, and setting this as the valve effect gain. The reason for introducing the valve effect gain and adaptively updating it is that the impact of changes in the equivalent opening of the exhaust valve assembly on the chamber pressure change rate is not a fixed constant, but dynamically changes with factors such as the gas state, material accumulation state, exhaust filter resistance, and pipeline temperature within the chamber. During overpressure event handling, as pressure relief proceeds, the exhaust filter may experience increased resistance due to particle trapping, and the material accumulation state within the chamber may change due to airflow impact. These factors can lead to different pressure reduction effects from the same change in the equivalent opening of the exhaust valve assembly. If the control system always uses a fixed valve effect gain parameter, it may continuously increase the equivalent opening of the exhaust valve assembly when the actual pressure reduction efficiency decreases, without achieving the expected pressure reduction effect, and instead causing unnecessary material disturbance. By estimating the valve effect gain in real time, the control system can perceive the actual pressure reduction efficiency of the exhaust channel under the current operating conditions and adjust the control strategy accordingly, promptly switching to other pressure reduction methods when the pressure reduction efficiency decreases. Since the pressure change rate is affected not only by the equivalent opening of the exhaust valve assembly, but also by the combined influence of multiple factors such as the release of residual pressure after the intake has been cut off, changes in heating power, the operation of the extraction channel, the material accumulation state, and pipeline resistance, the calculation of the valve effect gain must be performed under conditions that isolate interference from other factors as much as possible. Specifically, the valve effect gain is updated only when the following conditions are met simultaneously: the intake valve opening remains unchanged in adjacent sampling windows, the heating power command remains unchanged in adjacent sampling windows, the extraction valve speed command remains unchanged in adjacent sampling windows, and the absolute value of the change in the equivalent opening of the exhaust valve assembly is greater than the preset minimum effective opening change. The default value of the preset minimum effective opening change is 1% of the full-range opening, used to exclude invalid small changes caused by actuator dead zones or quantization errors. When the above conditions are met simultaneously, the valve effect gain is calculated by dividing the change in the pressure change rate in adjacent sampling windows by the sum of the change in the equivalent opening of the exhaust valve assembly in adjacent sampling windows and a preset minimum positive number, where the preset minimum positive number is used to avoid numerical overflow caused by a zero denominator. The default value of the preset minimum positive number is 0.000001, determined by the floating-point precision of the control system's numerical calculations. It must be much smaller than the minimum effective value of the equivalent opening change of the exhaust valve assembly under normal operating conditions, while being greater than the machine precision of the floating-point calculations to ensure numerical stability. When the above conditions are not simultaneously met, the control system keeps the valve effect gain unchanged at the most recently updated effective value and does not perform update calculations to avoid incorrectly attributing multi-factor coupling effects to changes in valve opening.Furthermore, to further suppress random deviations in single calculations, the control system employs an exponentially weighted moving average to smooth the valve effect gain. Specifically, the smoothed valve effect gain at the current moment equals a preset valve effect gain smoothing factor multiplied by the original calculated valve effect gain at the current moment, plus 1, minus the preset valve effect gain smoothing factor, and then multiplied by the smoothed valve effect gain from the previous moment. The default value of the preset valve effect gain smoothing factor is 0.3, with a range of 0.1 to 0.5. All subsequent determinations and uses of the valve effect gain are based on the smoothed valve effect gain.

[0067] If the absolute value of the smoothed valve effect gain continuously decreases within a consecutive preset gain attenuation judgment window sampling period, the judgment condition for continuous decrease is that in each sampling period within the consecutive preset gain attenuation judgment window, the absolute value of the smoothed valve effect gain is less than the absolute value of the smoothed valve effect gain in the previous sampling period. This indicates that the pressure reduction efficiency of the exhaust channel is decreasing. Possible reasons include increased exhaust filter resistance or changes in the cavity loading state. In this case, the control system prioritizes increasing the suction valve speed command or increasing the preset single-time load reduction amplitude of the heat source in step 200, rather than continuing to increase the equivalent opening of the exhaust valve group, to avoid unnecessary material disturbance caused by excessive opening of the exhaust valve group when the pressure reduction efficiency is low. The default value of the preset gain attenuation judgment window is 5 sampling periods, and the value range is 3 to 10 sampling periods. The specific value is determined according to the fluctuation characteristics of the smoothed valve effect gain signal, and must be greater than the typical fluctuation period of the smoothed signal to avoid misjudging the decrease in pressure reduction efficiency due to short-term fluctuations.

[0068] Step 400: If the cavity pressure reaches the preset advance protection pressure, or the time margin is less than or equal to the preset critical time margin threshold, the secondary interlocking pre-protection is executed. If the cavity pressure is greater than or equal to the preset secondary shutdown threshold, the hard shutdown is triggered by the independent secondary hardware overpressure switch.

[0069] When the chamber pressure reaches the preset advance protection pressure, which is greater than the preset first-level pressure relief threshold and less than the preset second-level shutdown threshold, or when the time margin set in step 100 is less than or equal to the preset critical time margin threshold, and the preset critical time margin threshold is less than the preset safety time margin lower limit, the control system enters the second-level interlock pre-protection stage. The default value of the preset advance protection pressure is the arithmetic mean of the preset first-level pressure relief threshold and the preset second-level shutdown threshold, i.e., the midpoint of the two thresholds. The specific value is determined based on the size of the software intervention window that needs to be retained between the first-level pressure relief and the second-level shutdown. This window must be large enough for the control system to complete the coordinated pressure relief action of the entire channel. The default value of the preset critical time margin threshold is 1.5 seconds, with a range of 0.5 seconds to 3 seconds. The specific value is determined based on the maximum delay time from the control system issuing the full-channel pressure relief command to all valves and pumps reaching the target state, and must be greater than this maximum delay time. The goal of this stage is to pull the chamber pressure back to the safe range through software control before the second-level hardware overpressure switch triggers a hard shutdown.

[0070] After entering the secondary interlock pre-protection stage, the control system executes the following mandatory actions:

[0071] The intake valve opening is forcibly set to 0, completely cutting off the intake; the heating power command is forcibly reduced to the minimum heating power allowed by the preset process.

[0072] Simultaneously, an emergency relaxation is triggered, temporarily relaxing the preset acceptable upper limit of the material disturbance index set in step 100 from the preset normal value to the preset emergency value, where the preset emergency value is greater than the preset normal value. The reason for triggering the emergency relaxation is that when the cavity pressure has reached the preset early protection pressure or the time margin has fallen below the preset critical time margin threshold, the core contradiction faced by the control system is that, under the constraint of material limitations, the pressure reduction capacity of the exhaust and extraction channels may be insufficient to pull the cavity pressure back to the safe range before the secondary hardware overpressure switch is triggered. If the normal preset acceptable upper limit of the material disturbance index is maintained, the adjustment space of the equivalent opening of the exhaust valve group will be strictly constrained, which may lead to insufficient pressure reduction speed and ultimately trigger the hard shutdown of the secondary hardware overpressure switch. A hard shutdown means that all heating and atmosphere control functions of the thermal equipment are interrupted instantly, which will not only cause the scrapping of the entire batch of process products, but may also cause thermal stress damage to the equipment lining and seals due to the sudden drop in temperature. In contrast, while temporarily relaxing the preset acceptable upper limit of the material disturbance index may result in slight material displacement, as long as it does not exceed the technical boundary of irreversible process loss, this slight displacement can usually be compensated for in subsequent processes by adjusting process parameters. The loss is far less than the overall loss caused by a hard shutdown. Therefore, emergency relaxation is essentially a risk trade-off between slight material disturbance and a hard shutdown of the equipment. It allows for greater depressurization operation space by accepting material disturbance within a controllable range, avoiding more severe consequences of a hard shutdown. The preset normal value is the preset acceptable upper limit set in step 100, with a default value of 0.7. The default value of the preset emergency value is 0.9, with a range of 0.8 to 1.0, determined based on the maximum tolerable disturbance level of the material under short-term high-intensity airflow impact. The technical boundary of irreversible process loss is set as follows: irreversible material displacement leading to damage to the charging geometry, thus affecting the uniformity of subsequent processes; material being carried out of the cavity by airflow into the exhaust pipe, resulting in non-recoverable material loss; or material breaking, stratifying, or chemically contaminating due to severe disturbance. The selection of the preset emergency value must ensure that, under this disturbance level, although the material may be slightly displaced, it will not exceed the technical boundary of the aforementioned irreversible process loss.

[0073] This relaxation only lasts for a preset emergency relaxation period. After this time window ends, the preset acceptable upper limit of the material disturbance index gradually recovers from the preset emergency value to the preset normal value in a linear, gradual manner over a preset gradual recovery time, rather than switching instantaneously. This is to avoid abrupt changes in the constraint boundary during pressure relief that could cause a step adjustment in the equivalent opening of the exhaust valve assembly. The reason for using linear gradual recovery instead of instantaneous switching is that if the preset acceptable upper limit of the material disturbance index were instantly restored from the preset emergency value to the preset normal value at the end of the emergency relaxation period, the actual value of the material disturbance index might still be between the preset normal value and the preset emergency value. The sudden tightening of the constraint boundary would force the control system to significantly reduce the equivalent opening of the exhaust valve assembly within a preset sampling period to meet the tightened constraint. This step adjustment would cause a sudden change in airflow within the cavity, which could potentially cause secondary disturbances in the material. At the same time, the sudden decrease in the pressure reduction rate could also cause the cavity pressure to rise again. Linear gradual recovery allows for a smooth transition of the constraint boundary, and the equivalent opening of the exhaust valve assembly can gradually decrease as the constraint boundary slowly tightens, avoiding abrupt changes in control action. The default value for the preset gradual recovery time is 5 seconds, with a range of 2 to 10 seconds. The specific value is determined based on the response sensitivity of the equivalent opening of the exhaust valve assembly to changes in the constraint boundary. The default value for the preset emergency relaxation duration is 10 seconds, with a range of 5 to 20 seconds. The specific value is determined based on the typical time required for the chamber pressure to drop from the preset advance protection pressure to the preset recovery threshold through coordinated depressurization across the entire channel.

[0074] To prevent the control system from repeatedly triggering emergency relaxation near the entry condition in step 400, a preset emergency relaxation cooling time is set. This means that after the end of an emergency relaxation duration window and the gradual recovery is complete, the control system is not allowed to trigger emergency relaxation again within the preset emergency relaxation cooling time. Even if the chamber pressure meets the entry condition in step 400 again, only the forced actions of cutting off the air intake and reducing the heating power are executed, without further relaxing the preset acceptable upper limit of the material disturbance index. The reason for setting a cooling time is that without a cooling time limit, when the chamber pressure fluctuates repeatedly near the preset pre-protection pressure, the control system may trigger emergency relaxation multiple times in a short period. Each relaxation would subject the material to a high-intensity airflow impact, and the cumulative disturbance effect might exceed the material's tolerance, causing irreversible displacement or carry-out of the material, even if each individual relaxation does not exceed the disturbance level corresponding to the preset emergency value. The cooling time mechanism ensures that after an emergency relaxation is completed, the control system has sufficient time to observe the pressure relief effect and allow the material state to stabilize, avoiding excessively frequent repeated impacts on the material. The default value for the preset emergency relaxation cooling time is 30 seconds, with a range of 15 to 60 seconds. The specific value is determined based on the shortest possible time for the chamber pressure to rise from the preset early protection pressure to the preset recovery threshold and then back to the preset early protection pressure. This time must be greater than the shortest possible time to avoid repeated relaxation of material constraints during the same overpressure event. If the chamber pressure continues to rise to the preset secondary shutdown threshold within the cooling time, a hard shutdown is triggered by an independent secondary hardware overpressure switch. By temporarily relaxing material constraints, a larger pressure relief operation space is gained to avoid triggering a hard shutdown caused by the secondary hardware overpressure switch.

[0075] Within the preset emergency relaxation duration window, the exhaust channel and the extraction channel work simultaneously, that is, the equivalent opening of the exhaust valve group and the speed command of the extraction valve are increased at the same time, but the preset upper limit of the opening change rate is still used for limitation, that is, the absolute value of the equivalent opening change rate of the exhaust valve group does not exceed the preset upper limit of the opening change rate, so as to avoid unacceptable impact on the material by a one-time strong injection.

[0076] Step 500: When the chamber pressure drops below the preset recovery threshold and the pressure change rate is not greater than zero for a duration that reaches the preset recovery judgment duration, exit the interlock control state and gradually return to the normal constant pressure control mode.

[0077] The conventional constant pressure control mode refers to the closed-loop pressure regulation method adopted by thermal equipment during normal process operation. Its control objective is to maintain the cavity pressure near the preset process set pressure. In the conventional constant pressure control mode, the control system uses the preset process set pressure as the target value and adjusts the outputs of four actuators—the inlet valve opening, the equivalent opening of the exhaust valve group, the suction valve speed command, and the heating power command—in real time through the constant pressure control loop to keep the cavity pressure stable within the allowable fluctuation range near the preset process set pressure. The constant pressure control loop works as follows: in each preset sampling period, the control system compares the filtered cavity pressure with the preset process set pressure. Based on the deviation and its trend, it calculates the output commands of each actuator according to pre-tuned control parameters. When the cavity pressure is lower than the preset process set pressure, the inlet valve opening or heating power command is appropriately increased to raise the cavity pressure; when the cavity pressure is higher than the preset process set pressure, the equivalent opening of the exhaust valve group or the suction valve speed command is appropriately increased to lower the cavity pressure, thereby dynamically stabilizing the cavity pressure near the preset process set pressure. In the normal constant pressure control mode, each actuator is at its normal constant pressure control operating point. This normal constant pressure control operating point refers to the steady-state output value of each actuator when the cavity pressure stabilizes at the preset process set pressure. Specifically, the intake valve opening is at the normal intake opening determined by the process formula; this normal intake opening is the steady-state opening value of the intake valve required to maintain the balance between the target atmosphere flow rate and the cavity pressure. The equivalent opening of the exhaust valve assembly is at the normal exhaust opening calculated by the constant pressure control loop based on the intake flow rate and cavity leakage. To stabilize the chamber pressure at the preset process set pressure under the current inlet flow rate and heating power conditions, the exhaust valve assembly's steady-state equivalent opening value is required. The evacuation valve speed command is set to the conventional evacuation valve speed determined by the process formula. This conventional evacuation valve speed is the steady-state speed command value of the vacuum pump required to maintain the chamber atmosphere replacement rate and pressure balance. If evacuation assistance is not required in the current process stage, the conventional evacuation valve speed is zero. The heating power command is set to the conventional heating power determined by the temperature control loop. This conventional heating power is the steady-state output value of the heating power required to maintain the target temperature in the current process stage. The specific values ​​of the above four conventional constant pressure control operating points vary depending on the process formula, process stage, and material loading in the chamber. They are automatically adjusted and maintained by the constant pressure control loop during steady-state operation.

[0078] When the chamber pressure drops below the preset recovery threshold and the pressure change rate remains less than or equal to zero for a preset recovery judgment duration, the control system determines that the overpressure event has been eliminated, exits the interlock control state, and gradually returns to the normal constant pressure control mode. The reason for using both the pressure threshold and the pressure change rate duration as conditions for recovery judgment is that if only the chamber pressure being below the preset recovery threshold is used as the condition for exiting the interlock, in situations where the chamber pressure briefly drops due to the pressure relief action and then rises again due to residual pressure rise factors, the control system may prematurely exit the interlock state and resume air intake and heating, causing the chamber pressure to exceed the preset first-level pressure relief threshold again and re-trigger the interlock, resulting in repeated entry and exit of the interlock state. By adding the condition that the pressure change rate remains less than or equal to zero for a preset recovery judgment duration, the control system ensures that before exiting the interlock, it has confirmed that the chamber pressure has not only dropped to a safe level, but that the pressure drop or stabilization trend has lasted for a sufficiently long time, eliminating the possibility of the pressure rising again after a brief drop. The default value for the preset recovery determination duration is 5 seconds, and the range is from 3 seconds to 15 seconds. The specific value is determined based on the fluctuation period of the cavity pressure near the recovery threshold. It must be greater than this fluctuation period to avoid misjudging that the overpressure event has been eliminated due to a brief drop in pressure.

[0079] During the process of exiting the interlock control state, the recovery of the intake valve opening and heating power command is limited by the rising slope. That is, within each preset sampling period, the increment of the intake valve opening is less than or equal to the product of the preset maximum recovery slope value of the intake valve and the preset sampling period, and the increment of the heating power command is less than or equal to the product of the preset maximum recovery slope value of the heating power and the preset sampling period. The reason for imposing the rising slope limitation on the recovery process is that the intake valve has been closed and the heating power has been reduced during the overpressure event handling. If the intake valve opening and heating power command are instantly restored to the normal intake opening and normal heating power corresponding to the normal constant pressure control operating point after exiting the interlock, the sudden influx of a large amount of gas and the sudden increase in heating power will generate a new pressure boost in the cavity. Since the equivalent opening of the exhaust valve group and the suction valve speed command are also synchronously reverting at this time, the pressure reduction capacity of the exhaust channel is weakening. The rapid recovery of intake and heating is very likely to trigger a secondary pressure spike, causing the cavity pressure to exceed the preset first-level pressure relief threshold again and re-triggering the interlock. By limiting the opening of the intake valve and the rising slope of the heating power command, the recovery process proceeds slowly and gradually. The pressure boosting term within the cavity increases progressively, giving the constant pressure control loop sufficient time to adjust and stabilize each increment, thus avoiding pressure overshoot during the recovery process. The default value for the preset maximum recovery slope of the intake valve is 2% of the full-scale opening per second, with a range of 1% to 5% of the full-scale opening per second. The specific value is determined based on the sensitivity of the intake flow rate to the cavity pressure; higher sensitivity requires a smaller recovery slope. The default value for the preset maximum recovery slope of the heating power is 2% of the rated power per second, with a range of 1% to 5% of the rated power per second. The specific value is determined based on the degree of influence of heating power changes on the thermal expansion rate of the gas within the cavity.

[0080] During the exit from the interlock control state, the equivalent opening of the exhaust valve group and the speed command of the extraction valve synchronously and gradually return to the normal constant pressure control operating point. Specifically, the equivalent opening of the exhaust valve group gradually recovers to the normal exhaust opening at a rate not exceeding the upper limit of the preset opening change rate, and the speed command of the extraction valve gradually recovers to the normal extraction valve speed according to the same recovery ratio as the equivalent opening of the exhaust valve group. The recovery ratio refers to the degree of completion of the recovery of the equivalent opening of the exhaust valve group from the current value to the normal exhaust opening, and this degree of completion is used as the conversion basis for the recovery of the extraction valve speed command from the current value to the normal extraction valve speed. That is, the extraction valve speed command will synchronously complete the same percentage of recovery as the equivalent opening of the exhaust valve group, so that the constant pressure control loop resumes closed-loop regulation of the chamber pressure after all actuators return to the steady-state output value.

[0081] After the recovery is complete, the control system returns to the normal constant pressure control bypass monitoring state in step 200, and continues to monitor the dynamic trend of the cavity pressure in real time.

[0082] Throughout the overpressure event handling process from step 200 to step 400, the control system continuously records the following key process trajectory data: filtered chamber pressure, pressure change rate, smoothed pressure acceleration, inlet valve opening, equivalent opening of exhaust valve group, suction valve speed command, heating power command, material disturbance index, particle concentration signal, pressure difference before and after exhaust filter, and execution timestamps of each action.

[0083] After an overpressure event ends, i.e., after exiting the interlock state, the control system outputs the minimum disturbance feasible strategy parameters for this overpressure event based on the recorded data of this event through the following quantitative analysis method. The reason for using a post-event quantitative analysis method to correct the strategy parameters is that the various parameters of the control system during the overpressure event handling process are initial values ​​set based on equipment commissioning or historical experience. These initial values ​​may not fully adapt to changes in operating conditions caused by factors such as material batch differences, equipment aging, and changes in the degree of clogging of exhaust filters during actual operation. By systematically analyzing the process trajectory data of each overpressure event, the control system can extract key characteristic quantities reflecting the current equipment state and material properties from the actual operating data, and accordingly make targeted corrections to the strategy parameters, enabling the control strategy to better adapt to actual operating conditions in subsequent batches. This parameter correction method based on historical event data does not require the establishment of an accurate cavity pressure dynamics model; it only relies on directly observable process data to achieve stepwise parameter optimization, making it suitable for complex industrial systems such as thermal equipment where it is difficult to establish accurate mechanistic models. First, the control system extracts the following characteristic quantities from the process trajectory data of this event: the response delay time from the triggering of the first-level interlock to the start of pressure reduction in the chamber; the peak value of the material disturbance index during the entire event and its corresponding equivalent opening of the exhaust valve group; the actual average pressure reduction rate within each pulse cycle in the pulse pressure relief mode; and the changing trend of the smoothed valve effect gain during the event. These four characteristic quantities are selected as inputs for parameter correction because the response delay time reflects the overall response speed of the control system from triggering the interlock to producing the actual pressure reduction effect, and this characteristic quantity is directly related to the rationality of the preset safety time margin lower limit. The peak value of the material disturbance index reflects the maximum degree of disturbance experienced by the material in this event, and this characteristic quantity is directly related to the conservatism of the preset duty cycle and the upper limit of the preset opening change rate. The actual average pressure reduction rate reflects the actual pressure reduction efficiency of the pulse pressure relief mode, and this characteristic quantity is directly related to the rationality of the preset duty cycle setting.

[0084] The change trend of the smoothed valve effect gain reflects the evolution of the exhaust channel pressure reduction efficiency during the event process. This characteristic is directly related to whether the preset safety time margin lower limit needs to reserve additional margin for the decrease in pressure reduction efficiency. Based on the above characteristics, the control system modifies the strategy parameters according to the following rules: If the peak value of the material disturbance index in this event is lower than 80% of the preset acceptable upper limit and the response delay time is greater than twice the preset safety time margin lower limit, the preset duty cycle of the next batch will be reduced by the preset duty cycle adjustment step size based on the current usage value to further reduce material disturbance when the safety margin is sufficient; If the peak value of the material disturbance index in this event exceeds 95% of the preset acceptable upper limit or triggers the emergency relaxation in step 400, the preset opening change rate upper limit of the next batch will be reduced by the preset opening change rate adjustment step size based on the current usage value to constrain the transient jet intensity during the depressurization process; If the smoothed valve effect gain decreases by more than 50% relative to the initial value of the event in this event, the preset safety time margin lower limit of the next batch will be increased by the preset time margin adjustment step size based on the current usage value to reserve more response time for the condition of reduced depressurization efficiency. The default value for the preset duty cycle adjustment step size is 0.05, the default value for the preset opening change rate adjustment step size is 1% of the full-scale opening per second, and the default value for the preset time margin adjustment step size is 0.5 seconds. Each of the above adjustment step sizes has a cumulative adjustment upper limit, meaning that the absolute offset of the parameter value after multiple batches of correction relative to the corresponding baseline initial configuration parameter must not exceed the preset maximum cumulative offset. The baseline initial configuration parameter refers to the initial parameter value determined and archived after equipment commissioning, serving as a unified comparison benchmark for subsequent batches of parameter corrections. It remains unchanged during the determination of the cumulative adjustment upper limit and is not updated due to the correction results of a particular batch. The default value for the preset maximum cumulative offset is 30% of the corresponding baseline initial configuration parameter value to prevent the parameter from deviating from the reasonable working range after multiple unidirectional corrections. Each of the above adjustment step sizes has a cumulative adjustment upper limit, meaning that the absolute offset of the parameter value after multiple batches of correction relative to the corresponding baseline initial configuration parameter must not exceed the preset maximum cumulative offset. The baseline initial configuration parameters refer to the initial parameter values ​​determined and archived after the equipment is debugged. They serve as a unified comparison benchmark for subsequent batch parameter corrections and remain unchanged during the determination of the cumulative adjustment upper limit. They are not updated due to the correction results of a particular batch. The default value of the preset maximum cumulative offset is 30% of the initial configuration parameter value. It is determined based on the degree of influence of the corresponding parameter on safety constraints, material constraints, and control stability. It is used to limit the maximum allowable correction range of the parameter and prevent the parameter from deviating from the reasonable working range after multiple batch corrections.The reason for setting a cumulative adjustment upper limit is that parameter correction based on single event data may be affected by the special operating conditions of that event, resulting in deviations. If the cumulative correction amount is not limited, when encountering deviations in the same direction in multiple consecutive batches, the parameters may be gradually pushed to extreme values, deviating from the reasonable operating range verified by the system during the equipment commissioning phase, which would reduce the performance of the control strategy under normal operating conditions. The cumulative adjustment upper limit ensures that the adaptive correction of parameters always takes place within a limited range near the initial configuration parameters, allowing parameters to be moderately optimized based on actual operating data while preventing parameters from becoming uncontrollable due to cumulative deviations. The output minimum disturbance feasible strategy parameters include the corrected preset duty cycle, the upper limit of the preset opening change rate, the lower limit of the preset safety time margin, and the smoothed valve effect gain estimated in step 300.

[0085] The aforementioned minimum disturbance feasible strategy parameters are stored and used for initialization in step 100 of the next similar process stage, serving as the initial parameters for the next batch. This achieves cross-batch adaptive parameter tuning, improving the consistency and effectiveness of the control strategy across different batches. The initial parameters for the next batch are used for control initialization, while the baseline initial configuration parameters are only used for comparison and constraint of the cumulative adjustment upper limit; their meanings differ. The effect of cross-batch adaptive parameter tuning lies in the fact that during long-term operation of thermal equipment, the resistance of the exhaust filter gradually increases due to particle retention, the response characteristics of the valve actuator gradually change due to wear, and the particle size distribution and bulk density of the material fluctuate due to differences in raw material batches. These slowly changing factors make it difficult for fixed control parameters to maintain optimal performance across all batches. By transferring the experience of handling each overpressure event to subsequent batches in the form of quantitative parameter corrections, the control system can track the slow drift of equipment status and material characteristics, continuously optimizing the control effect of material disturbances while ensuring safety constraints. This allows the control strategy to gradually approach the optimal configuration under the current equipment state as the number of operating batches accumulates.

[0086] Example 2

[0087] See Figure 4 As shown, a thermal equipment overpressure graded interlocking control system is provided, which stores computer-readable instructions and, when the computer-readable instructions are read, can execute the aforementioned thermal equipment overpressure graded interlocking control method. The system includes:

[0088] The pressure derivation module 101 collects the cavity pressure of the thermal equipment and calculates the pressure change rate, time margin, and material disturbance index based on the cavity pressure.

[0089] The first-level interlock module 102 triggers the first-level interlock when the cavity pressure does not reach the preset first-level pressure relief threshold but meets the pre-triggering condition, or when the cavity pressure reaches the preset first-level pressure relief threshold, and performs source term suppression on the intake air path, heating power and exhaust buffer.

[0090] If the cavity pressure continues to rise after the first-level interlock, the pressure relief control module 103 enters the pressure relief stage constrained by material disturbance. Based on the time margin and material disturbance index, it controls the pressure relief of the exhaust gas path under the conditions of meeting safety constraints and material constraints, and triggers the upgrade strategy.

[0091] The secondary protection module 104 performs secondary interlocking pre-protection if the cavity pressure reaches the preset advance protection pressure or the time margin is less than or equal to the preset critical time margin threshold. If the cavity pressure is greater than or equal to the preset secondary shutdown threshold, a hard shutdown is triggered by an independent secondary hardware overpressure switch.

[0092] When the chamber pressure drops below the preset recovery threshold and the pressure change rate is not greater than zero for a period of time that reaches the preset recovery judgment duration, the recovery control module 105 exits the interlock control state and gradually recovers to the normal constant pressure control mode.

[0093] The embodiments of the present invention have been described above. However, the embodiments are not limited to the specific implementation methods described above. The specific implementation methods described above are merely illustrative and not restrictive. Those skilled in the art can make more equivalent embodiments under the guidance of the present embodiments, and all of them are within the protection scope of the present embodiments.

Claims

1. A method for graded interlocking control of overpressure in thermal equipment, characterized in that, Includes the following steps: Collect the cavity pressure of the thermal equipment, and calculate the pressure change rate, time margin, and material disturbance index based on the cavity pressure; When the cavity pressure does not reach the preset first-level pressure relief threshold but meets the pre-triggering condition, or when the cavity pressure reaches the preset first-level pressure relief threshold, the first-level interlock is triggered to suppress the source term of the intake air passage, heating power and exhaust buffer. If the chamber pressure continues to rise after the first-level interlock, it enters the pressure relief stage constrained by material disturbance. Based on the time margin and material disturbance index, the exhaust gas path is controlled to relieve pressure while meeting safety and material constraints, and an upgrade strategy is triggered. If the chamber pressure reaches the preset advance protection pressure, or the time margin is less than or equal to the preset critical time margin threshold, the second-level interlocking pre-protection is executed; if the chamber pressure is greater than or equal to the preset second-level shutdown threshold, the second-level hardware overpressure switch triggers a hard shutdown. When the chamber pressure drops below the preset recovery threshold and the pressure change rate is not greater than zero for a duration that reaches the preset recovery judgment duration, the interlock control state is exited and the system gradually returns to the normal constant pressure control mode.

2. The overpressure graded interlocking control method for thermal equipment according to claim 1, characterized in that, Methods for calculating pressure change rate, time margin, and material disturbance index include: The original cavity pressure value is smoothed and filtered to obtain the smoothed cavity pressure; The rate of pressure change is calculated based on the filtered cavity pressure. The pressure acceleration is calculated based on the pressure change rate and then smoothed. Based on the current cavity pressure, pressure change rate, and smoothed pressure acceleration, a short-term prediction of future pressure is made to obtain the predicted pressure. The time margin is obtained by calculating the shortest time required for the predicted pressure to reach the preset level 2 shutdown threshold for the first time based on the predicted pressure. The material disturbance index is obtained by a weighted combination of multiple observable physical quantities after normalization.

3. The overpressure graded interlocking control method for thermal equipment according to claim 1, characterized in that, Pre-triggering conditions include at least one of the following: The time margin is less than or equal to the preset pre-trigger time margin threshold; The pressure change rate is greater than or equal to the preset pre-trigger pressure change rate threshold.

4. The overpressure graded interlocking control method for thermal equipment according to claim 1, characterized in that, The actions executed by the first-level interlock include: Intake suppression sets the intake valve opening to zero or reduces it to the minimum airflow required, and locks the preset intake suppression duration. When the pressure change rate is greater than zero and the time margin continues to shorten, the heating power command will be reduced by a preset single reduction range, and will not be lower than the preset minimum heating power allowed by the process. Balance buffer: Open the balance valve to the preset equivalent opening command value of the balance valve to reduce the local pressure difference and jet intensity in the exhaust channel.

5. The overpressure graded interlocking control method for thermal equipment according to claim 1, characterized in that, Methods for controlling exhaust gas pressure relief include: When the time margin is greater than the preset safety time margin lower limit and the material disturbance index is greater than or equal to the preset acceptable upper limit multiplied by the preset proximity proportional factor, the pulse pressure relief mode is adopted, and the equivalent opening of the exhaust valve group in the exhaust gas path switches periodically between the preset low pressure relief opening and the preset medium pressure relief opening according to the preset duty cycle. When the time margin is less than or equal to the preset safety time margin lower limit, the equivalent opening of the exhaust valve group in the exhaust gas path at the current moment is added to the product of the preset time margin error gain and the difference between the preset safety time margin lower limit and the current time margin, and saturation limiting processing is performed.

6. The overpressure graded interlocking control method for thermal equipment according to claim 5, characterized in that, During the depressurization process, if the particle concentration signal exceeds the preset particle concentration alarm threshold, the adjustment step of the equivalent opening of the exhaust valve group is reduced, and a higher frequency small-amplitude pulse depressurization is switched. The depressurization task is shared by increasing the heat source deload amplitude in the first-level interlock and activating the air extraction channel. The particle concentration signal is obtained by a light scattering dust sensor or a charge particle sensor installed in the exhaust gas path of the thermal equipment.

7. The overpressure graded interlocking control method for thermal equipment according to claim 1, characterized in that, Methods for triggering upgrade strategies include: The upgrade strategy is triggered when any upgrade trigger condition is met. The upgrade triggering conditions include: after the cavity pressure reaches the preset first-level pressure relief threshold, there is still no reversal from positive to negative pressure change rate after a preset upgrade waiting time; the material disturbance index is greater than or equal to the preset acceptable upper limit multiplied by the preset proximity proportional factor. The upgrade strategy includes sequentially performing actions to enable the extraction channel and switch the exhaust valve group sequence.

8. The overpressure graded interlocking control method for thermal equipment according to claim 7, characterized in that, Methods for enabling the sequence switching of the extraction channel and exhaust valve group include: The activation of the extraction channel involves opening the extraction valve and increasing the extraction valve speed command to the preset extraction working value, thereby transferring the pressure reduction task from the exhaust channel to the extraction channel. The sequence of the exhaust valve group is switched as follows: first, the small-diameter pressure relief valve is opened to relieve pressure until the absolute value of the pressure change rate is less than the preset pre-triggered pressure change rate threshold. Then, the quick-opening valve is opened briefly to quickly reduce the pressure peak. After that, the quick-opening valve is closed and the valve returns to the small-diameter valve to maintain pressure relief.

9. The overpressure graded interlocking control method for thermal equipment according to claim 1, characterized in that, The methods for implementing secondary interlocking pre-protection include: Force the intake valve opening to zero and reduce the heating power command to the preset minimum allowable heating power; Trigger emergency relaxation, temporarily relax the preset acceptable upper limit of the material disturbance index from the preset normal value to the preset emergency value; During the execution of the secondary interlocking pre-protection, the emergency relaxation only lasts for the preset emergency relaxation duration. After the expiration, the preset acceptable upper limit of the material disturbance index is restored from the preset emergency value to the preset normal value in a linear gradual manner within the preset gradual recovery time. Set a preset emergency relaxation cooldown time, during which emergency relaxation cannot be triggered again.

10. The overpressure graded interlocking control method for thermal equipment according to claim 1, characterized in that, Methods for gradually reverting to the conventional constant pressure control mode include: During the process of exiting the interlock control state, within each preset sampling period, the increment of the intake valve opening is less than or equal to the product of the preset intake valve maximum recovery slope value and the preset sampling period, and the increment of the heating power command is less than or equal to the product of the preset heating power maximum recovery slope value and the preset sampling period. During the process of exiting the interlock control state, the equivalent opening degree of the exhaust valve group and the speed command of the extraction valve synchronously return to the normal constant pressure control operating point of the normal constant pressure control mode.

11. A graded interlocking control system for overpressure in thermal equipment, characterized in that, It is used to store computer-readable instructions, which, when read, can execute the overpressure graded interlocking control method for thermal equipment as described in any one of claims 1-10; the system includes: The pressure derivation module collects the cavity pressure of the thermal equipment and calculates the pressure change rate, time margin, and material disturbance index based on the cavity pressure. The first-level interlock module triggers the first-level interlock when the cavity pressure does not reach the preset first-level pressure relief threshold but meets the pre-triggering condition, or when the cavity pressure reaches the preset first-level pressure relief threshold, thereby performing source term suppression on the intake air path, heating power, and exhaust buffer. If the chamber pressure continues to rise after the first-level interlock, the pressure relief control module enters the pressure relief stage constrained by material disturbance. Based on the time margin and material disturbance index, it controls the exhaust gas path to relieve pressure while meeting safety and material constraints, and triggers the upgrade strategy. The secondary protection module will execute secondary interlocking pre-protection if the chamber pressure reaches the preset early protection pressure or the time margin is less than or equal to the preset critical time margin threshold. If the chamber pressure is greater than or equal to the preset secondary shutdown threshold, the secondary hardware overpressure switch will trigger a hard shutdown. The recovery control module exits the interlock control state and gradually returns to the normal constant pressure control mode when the cavity pressure drops below the preset recovery threshold and the pressure change rate is not greater than zero for a period of time that reaches the preset recovery judgment duration.

Images