Dual overpressure safety control system and method for a gas-water dual-purpose stove

Abstract

The application relates to a double anti-overpressure safety control system and method of a steam-water dual-purpose furnace, which comprises an enhanced signal acquisition unit, an electronic control unit, an executing mechanism, a physical protection unit, a man-machine interaction and data management unit; the physical protection unit comprises three-stage relief devices independently installed on a steam pocket; the enhanced signal acquisition unit comprises a pressure monitoring module, a flow monitoring module, a temperature monitoring module and a safety component state monitoring module; the man-machine interaction and data management unit comprises an HMI and a data storage module; the executing mechanism comprises a frequency conversion burner, an electric feed water pump and an electric regulating valve V1 receiving PLC analog output; the PLC programmable logic controller controls the frequency conversion burner, the electric feed water pump and the electric regulating valve V1 based on the data collected in each module of the enhanced signal acquisition unit. The application greatly reduces the probability of serious overpressure accidents caused by single element or system failure.

Description

Technical Field

[0001] This invention relates to the field of industrial boiler safety control technology, and in particular to a dual overpressure prevention safety control system and method for a steam-water dual-purpose boiler. Background Technology

[0002] Currently, the operating pressure of dual-fuel boilers used in small and medium-sized processing enterprises (such as bamboo shoot steaming) is usually no higher than 0.1 MPa. Although they are not subject to the regulatory scope of the "Special Equipment Catalogue," their operational safety is equally crucial. The existing safety protection measures for such boilers mainly have the following problems: The protection methods are simplistic and lack reliability: most devices rely solely on mechanical safety valves or simple electrical contact pressure gauges for overpressure protection. Mechanical safety valves are prone to pressure drift or failure due to corrosion or jamming; electrical contact controls may completely lose their protective function due to sensor failure or circuit problems, and the system lacks redundancy design.

[0003] Poor control precision and large pressure fluctuations: Traditional start-stop or simple PID control is difficult to cope with rapid changes in steam load, which can easily cause large pressure fluctuations. This not only affects the stability of the production process, but also accelerates equipment wear and tear due to frequent load changes. Summary of the Invention

[0004] The purpose of this invention is to provide a dual overpressure protection safety control system and method for a steam-water dual-purpose boiler. It can provide ultimate protection for the PLC programmable logic controller to actively regulate the frequency converter, electric feed water pump and electric regulating valve V1 and the physical protection unit, which greatly reduces the probability of serious overpressure accidents caused by single component or system failure.

[0005] To achieve the above objectives, the present invention provides the following technical solution: a dual overpressure protection safety control system for a steam-water dual-purpose boiler, the system comprising an enhanced signal acquisition unit, an electronic control unit, an actuator, a physical protection unit, and a human-machine interaction and data management unit; The physical protection unit includes a three-stage relief device independently installed on the steam drum. The three-stage relief device includes a spring-loaded safety valve A, a rupture disc B, and a manual pressure relief valve C installed sequentially on the steam drum. The enhanced signal acquisition unit includes a pressure monitoring module, a flow monitoring module, a temperature monitoring module, and a safety component status monitoring module. The pressure monitoring module is a pressure sensor P1 installed on the top of the steam drum, used to acquire the steam drum pressure signal P(t) in real time. The flow monitoring module is a vortex flow meter F1 installed on the main steam pipeline, used to measure the instantaneous steam consumption D(t) in real time. The temperature monitoring module consists of a PT100 temperature sensor T1 installed on the steam outlet pipeline and a temperature sensor T2 installed on the feedwater pipeline, used to monitor the steam temperature and feedwater temperature respectively. The safety component status monitoring module consists of a microswitch S1 installed on the valve stem of the spring-loaded safety valve A and a high-sensitivity differential pressure switch ΔP1 or acoustic emission sensor AE1 connected upstream and downstream of the rupture disc B. The microswitch S1 is used to detect whether the safety valve has activated; the high-sensitivity differential pressure switch ΔP1 or acoustic emission sensor AE1 is used to detect minor leaks or rupture events of the rupture disc. The electronic control unit is a PLC programmable logic controller, and all data acquired by the enhanced signal acquisition unit is connected to the PLC programmable logic controller. The human-machine interaction and data management unit includes an HMI and a data storage module. The HMI is communicatively connected to the PLC programmable logic controller. The industrial touch screen is used to display real-time data, set parameters, query historical alarms and event reports, and the data management unit is used to continuously record operating data. The actuator includes a variable frequency burner, an electric water pump, and an electric regulating valve V1 that receive analog outputs from the PLC. The PLC programmable logic controller controls the variable frequency burner, the electric water pump, and the electric regulating valve V1 based on the data collected from each module in the enhanced signal acquisition unit.

[0006] Furthermore, the real-time data in the human-computer interaction and data management unit includes pressure P(t), risk coefficient R(t) curve, and health status of safety components.

[0007] Further steps include the following: Step S1: Using T as the sampling period, the steam drum pressure P(t), steam flow rate D(t), and feedwater temperature Tw signals are collected in real time through each module in the enhanced signal acquisition unit, and the pressure change rate dP / dt and dynamic risk prediction coefficient R(t) are obtained. Step S2: The PLC programmable logic controller enters different control zones based on the current pressure P(t) and the obtained R(t) value: If P(t) < 0.075MPa and R(t) < 0.4, it enters the normal adjustment zone and adopts "conventional PID control with feedforward compensation" to maintain pressure stability while gently intervening in the initial risk; if 0.075MPa ≤ P(t) < 0.085MPa or 0.4 ≤ R(t) < 0.8, it enters the early warning intervention zone and switches to "fuzzy PID adaptive control based on R(t) feedforward" for more proactive and compliant intervention; if P(t) ≥ 0.085MPa or R(t) ≥ 0.8, it enters the emergency intervention zone and executes the highest priority hard safety logic, immediately overriding all algorithm outputs. Step S3: If the pressure continues to rise and reaches the first physical safety threshold, the spring-loaded safety valve A will automatically open to release pressure. If the pressure continues to rise and reaches the final safety threshold, the rupture disc B will rupture to release pressure. The manual valve C is used for manual pressure release during maintenance. At the same time, when the system pressure exceeds 0.08 MPa, the PLC programmable logic controller will perform online testing to determine whether the safety valve core moves flexibly and whether there is any jamming, and record the status. The historical action data of the safety valve will be analyzed periodically to conduct health assessments and provide early warnings. Step S4: The system records the "pressure-time" curve, R(t) curve, all sensor data, control commands, valve status, alarm and event logs at a high frequency. Each data entry is accompanied by a precise timestamp, forming a traceable "digital black box".

[0008] Furthermore, the method includes the following steps: the first physical safety threshold is 0.09 MPa, and the final safety threshold is 0.12 MPa.

[0009] Furthermore, the formula for calculating the pressure change rate in step S1 is as follows: dP / dt = [P(t) - P(t-1)] / T; The dynamic risk prediction coefficient R(t) is calculated using an enhanced calculation formula that integrates multi-source information: P_pred = P(t) + dP / dt × Δt R(t) = α×[P(t) - Pset] / (Pmax - Pset) + β * / (dP / dt)max + γ* ; Where Pset is the pressure setpoint, Pmax is the design pressure, (dP / dt)max is the maximum allowable pressure change rate, D_avg is the sliding average of steam flow over a period of time, representing the current average load; D_max is the maximum range of the flow meter; α, β, γ, δ are weighting coefficients, satisfying α+β+γ+δ=1. F(P(t), t) is the short-term pressure trend prediction term; Δt is the prediction step size, and R(t) ranges from [0, 1].

[0010] Furthermore, in step 2, the "conventional PID control with feedforward compensation" is used to gently intervene in the initial risk while maintaining pressure stability. The specific steps are as follows: Step a1: Calculate the pressure deviation e(t): e(t) = Pset - P(t); Step a2: Calculate the standard PID output u_pid(t): u_pid(t) = Kp×e(t) + Ki×Σe(t) + Kd×[e(t) - e(t-1)] / T; Step a3: Calculate the dynamic feedforward quantity u_ff(t): u_ff(t) = Kf×R(t); Kf is the feedforward gain coefficient; Step a4: Synthesize the final control quantity u(t): u(t) = u_pid(t) + u_ff(t); After amplitude limiting, it is converted into a 4-20mA signal output to the variable frequency burner.

[0011] Furthermore, in step S2, switching to "fuzzy PID adaptive control based on R(t) feedforward" for more proactive and compliant intervention involves the following steps: Step b1, Fuzzification: Divide the pressure deviation e(t), deviation change rate de(t) / dt, dynamic risk coefficient R(t), and risk change rate dR(t) / dt into five fuzzy subsets: “Negative Large (NB)”, “Negative Small (NS)”, “Zero (Z)”, “Positive Small (PS)”, and “Positive Large (PB)”, and determine their membership functions. Step b2, Fuzzy Rule Base Design: Introduce R(t) and dR(t) / dt as key bases for adjusting PID parameters; Defuzzification and Parameter Self-Tuning: Based on all activated fuzzy rules, use the centroid method for defuzzification calculation to obtain the real-time adjustment amounts ΔKp, ΔKi, and ΔKd of the PID parameters, and update the PID parameters online. Kp = Kp0 + ΔKp, Ki = Ki0 + ΔKi, Kd = Kd0 + ΔKd; Wherein, Kp0, Ki0, and Kd0 are a set of pre-tuned basic PID parameters; Step b3: Calculate the control quantity u(t) using the updated Kp, Ki, Kd, ​​and output it to the actuator; Step b4: At the same time, the HMI interface displays a yellow warning and dynamically displays the real-time value and change curve of R(t).

[0012] Furthermore, the specific steps in step S2 to execute the highest priority hard safety logic, immediately overriding all algorithm outputs, are as follows: Step c1: Output a fixed signal to the variable frequency burner to maintain the minimum firepower. Step c2: Fully open the water supply regulating valve V1 to perform high-flow "quenching" cooling; Step c3: Trigger the red alarm of the on-site audible and visual alarm, the HMI will pop up an emergency alarm screen, and all data before and after the event will be forcibly recorded for one minute.

[0013] Furthermore, in step S3, the method for determining whether the safety valve core moves flexibly and is not stuck is that while the programmable logic controller is executing control intervention, it sends a short pulse signal to a parallel test solenoid valve, causing the valve core of safety valve A to make a micro-motion. Through the feedback of the micro-switch S1, it can be determined whether the valve core moves flexibly and is not stuck.

[0014] The beneficial effects of this invention: Compared with the prior art, this invention can achieve the following: Revolutionary improvements in safety and reliability: This is brought about by the technical feature of a "redundant architecture consisting of an electronic control unit and a physical protection unit." The electronic control unit enables proactive intervention, while the physical system provides ultimate protection. Both operate independently, greatly reducing the probability of severe overpressure accidents caused by the failure of a single component or system.

[0015] Significantly improved control accuracy and stability: This is due to the technical feature of "electronic active regulation based on fuzzy PID algorithm". This algorithm can adapt to the nonlinear and time-varying changes in steam load, and suppress pressure fluctuations faster and more smoothly than traditional PID, reducing the working pressure fluctuation range from the traditional ±0.01MPa to within ±0.005MPa, thereby improving the quality stability of supporting processes (such as bamboo shoot steaming).

[0016] It possesses intelligent early warning and operation and maintenance capabilities, brought about by the technical features of "calculation and real-time monitoring of dynamic risk assessment coefficient R(t)" and "data storage and remote communication module." The system can identify potential risks from pressure change trends, transforming passive alarms into proactive early warnings. Complete data traceability functions provide a solid data foundation for fault diagnosis, performance analysis, and preventative maintenance.

[0017] Wide applicability and low modification cost: This invention adopts a modular design, and the control method can be implemented through software programming. The main hardware additions are a PLC control cabinet and two-stage physical valves. This solution is easy to integrate into the modification of existing small and medium-sized steam-water dual-purpose boilers, is economical, and can be quickly promoted to various fields requiring low-pressure steam, such as food processing, small-scale chemical industry, and medical sterilization. Attached Figure Description

[0018] Figure 1 This is a system block diagram of the present invention; Figure 2 This is a flowchart of the method of the present invention; Figure 3 This is a schematic diagram of the fuzzy PID controller principle. Figure 4 A schematic diagram of the physical protection unit installation; Figure 5 This is a schematic diagram of the safety monitoring screen for a dual-purpose steam and water boiler. Detailed Implementation

[0019] The invention will now be further described with reference to the accompanying drawings.

[0020] Please see Figures 1 to 5 The present invention provides an embodiment: a dual overpressure protection safety control system for a dual-purpose steam and water boiler, the system comprising an enhanced signal acquisition unit, an electronic control unit, an actuator, a physical protection unit, and a human-machine interaction and data management unit; The physical protection unit includes a three-stage relief device independently installed on the steam drum. This three-stage relief device comprises a spring-loaded safety valve A, a rupture disc B, and a manual pressure relief valve C, sequentially installed on the steam drum. As the final safety barrier, the physical protection unit independently installs a three-stage relief device on the steam drum: spring-loaded safety valve A (set pressure 0.09 MPa), rupture disc B (design burst pressure 0.12 MPa), and manual pressure relief valve C (for maintenance). Crucially, the microswitch S1 of safety valve A and the rupture disc monitoring signal (ΔP1 or AE1) are both connected to the PLC's DI (digital input) module, allowing the status of the physical protection unit to be sensed.

[0021] The enhanced signal acquisition unit includes a pressure monitoring module, a flow monitoring module, a temperature monitoring module, and a safety component status monitoring module. The pressure monitoring module is a pressure sensor P1 installed on the top of the steam drum, used to acquire the steam drum pressure signal P(t) in real time. The flow monitoring module is a vortex flow meter F1 installed on the main steam pipeline, used to measure the instantaneous steam consumption D(t) in real time. The temperature monitoring module consists of a PT100 temperature sensor T1 installed on the steam outlet pipeline and a temperature sensor T2 installed on the feedwater pipeline, used to monitor the steam temperature and feedwater temperature respectively. The safety component status monitoring module consists of a microswitch S1 installed on the valve stem of the spring-loaded safety valve A and a high-sensitivity differential pressure switch ΔP1 or acoustic emission sensor AE1 connected upstream and downstream of the rupture disc B. The microswitch S1 is used to detect whether the safety valve has activated. The high-sensitivity differential pressure switch ΔP1... P1 or acoustic emission sensor AE1 is used to detect minute leaks or ruptures in the rupture disc. The pressure sensor P1 is a high-precision pressure sensor with a range of 0-0.15 MPa and an accuracy of 0.5 class. The high-precision sensor is chosen to more sensitively capture initial pressure fluctuations, providing a reliable data basis for risk prediction. The vortex flow meter F1 has a range of 0-1 t / h and is used to measure the instantaneous steam consumption D(t) in real time. This signal is a key feedforward quantity for assessing the balance between production and consumption steam and calculating dynamic risks. The microswitch S1 is used to detect whether the safety valve has activated, such as opening or reseating. A high-sensitivity differential pressure switch ΔP1 or an acoustic emission sensor AE1 is connected upstream and downstream of the rupture disc B to detect possible minute leaks or ruptures in the rupture disc. All sensor signals are conditioned and converted from analog to digital by an isolated transmitter before being transmitted to the PLC programmable logic controller.

[0022] The electronic control unit is a PLC (Programmable Logic Controller), and the data collected by the enhanced signal acquisition unit is all connected to the PLC. It is based on a high-performance PLC, such as the Siemens S7-1500 series. This PLC not only executes basic logic and control algorithms, but also has a built-in dynamic risk prediction calculation module and a safety component intelligent diagnostic algorithm module. The PLC communicates with each input / output module via a high-speed bus.

[0023] The human-machine interface and data management unit includes an HMI and a data storage module. The HMI is communicatively connected to the PLC programmable logic controller. The industrial touch screen is used to display real-time data, set parameters, and query historical alarms and event reports. The data management unit is used to continuously record operating data. The HMI is an industrial touch screen that communicates with the PLC via Ethernet. It is used to display real-time data (including pressure P(t), risk coefficient R(t) curve, and health status of safety components), set parameters, and query historical alarms and event reports.

[0024] The data storage module (which can use an industrial SD card or an embedded database) continuously records all operational data. Alarm information, diagnostic reports, and critical data can be pushed to the cloud platform or central monitoring center in real time via a 4G / 5G industrial gateway.

[0025] The actuator includes a variable frequency burner, an electric water pump, and an electric regulating valve V1 that receive analog outputs from the PLC. The PLC programmable logic controller controls the variable frequency burner, electric water pump, and electric regulating valve V1 based on data collected from various modules in the enhanced signal acquisition unit. The actuator includes a variable frequency burner (regulating gas flow), an electric water pump, and an electric regulating valve V1 (collectively controlling water flow) that receive 4-20mA analog outputs from the PLC programmable logic controller.

[0026] Please continue reading. Figure 1 As shown, in one embodiment of the present invention, the real-time data in the human-computer interaction and data management unit includes pressure P(t), risk coefficient R(t) curve, and health status of safety components.

[0027] Please continue reading. Figures 2 to 5 As shown, in one embodiment of the present invention, the following steps are included: Step S1: Using T as the sampling period, the steam drum pressure P(t), steam flow rate D(t), and feedwater temperature Tw signals are collected in real time through each module in the enhanced signal acquisition unit, and the pressure change rate dP / dt and dynamic risk prediction coefficient R(t) are obtained. Step S2: The PLC programmable logic controller enters different control zones based on the current pressure P(t) and the obtained R(t) value: If P(t) < 0.075MPa and R(t) < 0.4, it enters the normal adjustment zone and adopts "conventional PID control with feedforward compensation" to maintain pressure stability while gently intervening in the initial risk; if 0.075MPa ≤ P(t) < 0.085MPa or 0.4 ≤ R(t) < 0.8, it enters the early warning intervention zone and switches to "fuzzy PID adaptive control based on R(t) feedforward" for more proactive and compliant intervention; if P(t) ≥ 0.085MPa or R(t) ≥ 0.8, it enters the emergency intervention zone and executes the highest priority hard safety logic, immediately overriding all algorithm outputs. Step S3: If the pressure continues to rise and reaches the first physical safety threshold, the spring-loaded safety valve A will automatically open to release pressure. If the pressure continues to rise and reaches the final safety threshold, the rupture disc B will rupture to release pressure. The manual valve C is used for manual pressure release during maintenance. At the same time, when the system pressure exceeds 0.08 MPa, the PLC programmable logic controller will perform online testing to determine whether the safety valve core moves flexibly and whether there is any jamming, and record the status. The historical action data of the safety valve will be analyzed periodically to conduct health assessments and provide early warnings. Step S4: The system records the "pressure-time" curve, R(t) curve, all sensor data, control commands, valve status, alarm and event logs at a high frequency. Each data entry is accompanied by a precise timestamp, forming a traceable "digital black box".

[0028] Please continue reading. Figure 2 As shown, in one embodiment of the present invention, the method includes the following steps: the first physical security threshold is 0.09 MPa, and the final security threshold is 0.12 MPa.

[0029] Please continue reading. Figure 2 As shown, in one embodiment of the present invention, the formula for calculating the pressure change rate in step S1 is: dP / dt = [P(t) - P(t-1)] / T; The dynamic risk prediction coefficient R(t) is calculated using an enhanced calculation formula that integrates multi-source information: P_pred = P(t) + dP / dt × Δt R(t) = α×[P(t) - Pset] / (Pmax - Pset) + β * / (dP / dt)max + γ* ; Where Pset is the pressure setpoint, Pmax is the design pressure, (dP / dt)max is the maximum allowable pressure change rate, D_avg is the sliding average of steam flow over a period of time, representing the current average load; D_max is the maximum range of the flow meter; α, β, γ, δ are weighting coefficients, satisfying α+β+γ+δ=1. F(P(t), t) is the short-term pressure trend prediction term; Δt is the prediction step size, and R(t) ranges from [0, 1].

[0030] Please continue reading. Figure 2 As shown, in one embodiment of the present invention, step 2 employs "conventional PID control with feedforward compensation" to gently intervene in the initial risk while maintaining pressure stability. The specific steps are as follows: Step a1: Calculate the pressure deviation e(t): e(t) = Pset - P(t); Step a2: Calculate the standard PID output u_pid(t): u_pid(t) = Kp×e(t) + Ki×Σe(t) + Kd×[e(t) - e(t-1)] / T; Step a3: Calculate the dynamic feedforward quantity u_ff(t): u_ff(t) = Kf×R(t); Kf is the feedforward gain coefficient; Step a4: Synthesize the final control quantity u(t): u(t) = u_pid(t) + u_ff(t); After amplitude limiting, it is converted into a 4-20mA signal output to the variable frequency burner.

[0031] Please continue reading. Figure 2 , Figure 3 As shown, in one embodiment of the present invention, the specific steps of switching to "fuzzy PID adaptive control based on R(t) feedforward" in step S2 for more proactive and compliant intervention are as follows: Step b1, Fuzzification: Divide the pressure deviation e(t), deviation change rate de(t) / dt, dynamic risk coefficient R(t), and risk change rate dR(t) / dt into five fuzzy subsets: “Negative Large (NB)”, “Negative Small (NS)”, “Zero (Z)”, “Positive Small (PS)”, and “Positive Large (PB)”, and determine their membership functions. Step b2, Fuzzy Rule Base Design: Introduce R(t) and dR(t) / dt as key bases for adjusting PID parameters; Defuzzification and Parameter Self-Tuning: Based on all activated fuzzy rules, use the centroid method for defuzzification calculation to obtain the real-time adjustment amounts ΔKp, ΔKi, and ΔKd of the PID parameters, and update the PID parameters online. Kp = Kp0 + ΔKp, Ki = Ki0 + ΔKi, Kd = Kd0 + ΔKd; Wherein, Kp0, Ki0, and Kd0 are a set of pre-tuned basic PID parameters; Step b3: Calculate the control quantity u(t) using the updated Kp, Ki, Kd, ​​and output it to the actuator; Step b4: At the same time, the HMI interface displays a yellow warning and dynamically displays the real-time value and change curve of R(t).

[0032] Please continue reading. Figure 2As shown, in one embodiment of the present invention, the specific steps in step S2 of executing the highest priority hard safety logic to immediately cover all algorithm outputs are as follows: Step c1: Output a fixed signal to the variable frequency burner to maintain the minimum firepower. Step c2: Fully open the water supply regulating valve V1 to perform high-flow "quenching" cooling; Step c3: Trigger the red alarm of the on-site audible and visual alarm, the HMI will pop up an emergency alarm screen, and all data before and after the event will be forcibly recorded for one minute.

[0033] Please continue reading. Figure 2 As shown, in one embodiment of the present invention, the method for determining whether the safety valve core moves flexibly and whether it is stuck in step S3 is that the programmable logic controller sends a short pulse signal to a parallel test solenoid valve while performing control intervention, causing the valve core of safety valve A to make a micro-motion. Through the feedback of micro-switch S1, it can be determined whether the valve core moves flexibly and whether it is stuck. Specific Implementation Example 1 The system performs the following operations at fixed intervals (e.g., 100ms): Real-time acquisition of signals such as steam drum pressure P(t), steam flow rate D(t), and feedwater temperature Tw.

[0035] Calculate the pressure change rate dP / dt = (P(t) - P(t-1)) / T, where T is the sampling period.

[0036] Calculate the dynamic risk prediction coefficient R(t). This invention employs the following enhanced calculation formula that integrates multi-source information: R(t) = α×(P(t) - Pset) / (Pmax - Pset) + β * / (dP / dt)max + γ* Parameter description: Pset: Pressure setpoint (e.g., 0.07 MPa).

[0037] Pmax: Design pressure (0.1 MPa).

[0038] (dP / dt)max: The maximum allowable rate of pressure change (e.g., 0.01 MPa / s).

[0039] D_avg: The moving average of steam flow over a period of time (e.g., 60 seconds), representing the current average load.

[0040] D_max: Maximum flow rate of the flow meter (1t / h).

[0041] α, β, γ, δ: Weighting coefficients, satisfying α + β + γ + δ = 1. An example set of values ​​is: α = 0.4, β = 0.3, γ = 0.2, δ = 0.1. These coefficients can be optimized and tuned using actual operational data.

[0042] F(P(t), t): Short-term pressure trend prediction term. This invention uses a simplified prediction using a linear extrapolation model based on the current rate of pressure change: P_pred = P(t) + dP / dt×Δt, where Δt is the prediction step size (e.g., 3 seconds). Therefore, F(P(t), This gives the system a certain degree of foresight, enabling it to anticipate impending pressure deviation risks.

[0043] The value of R(t) ranges from [0, 1] and changes continuously. The larger the value, the higher the overpressure risk level assessed by the system.

[0044] S202: Intelligent feedforward-feedback composite control based on R(t) partitioning The controller enters different control zones based on the current pressure P(t) and the calculated value of R(t): Normal adjustment region (P(t) < 0.075 MPa and R(t) < 0.4): Strategy: Employ "conventional PID control with feedforward compensation" to maintain pressure stability while gently intervening in initial risks.

[0045] step: a. Calculate the pressure deviation: e(t) = Pset - P(t).

[0046] b. Calculate the standard PID output: u_pid(t) = Kp×e(t) + Ki×Σe(t) + Kd×[e(t) - e(t-1)] / T.

[0047] c. Calculate the dynamic feedforward: u_ff(t) = Kf×R(t). Kf is the feedforward gain coefficient (which can be set to a negative value, such as -5). Technical effect: When R(t) is greater than 0 due to factors such as flow fluctuations but has not reached the warning threshold, this additional feedforward term can make reverse fine adjustments to the burner power in advance, offsetting the potential pressure rise trend, achieving "early damping", and reducing the frequency of entering the warning zone.

[0048] d. Synthesized final control quantity: u(t) = u_pid(t) + u_ff(t). After amplitude limiting, it is converted into a 4-20mA signal output to the variable frequency burner.

[0049] Early warning intervention zone (0.075MPa ≤ P(t) < 0.085MPa or 0.4 ≤ R(t) < 0.8): Strategy: Switch to "Fuzzy PID Adaptive Control Based on R(t) Feedforward" for more proactive and compliant intervention.

[0050] step: a. Fuzzification: The input variables - pressure deviation e(t), deviation change rate de(t) / dt, dynamic risk coefficient R(t), and risk change rate dR(t) / dt - are divided into five fuzzy subsets: "negative large (NB)", "negative small (NS)", "zero (Z)", "positive small (PS)" and "positive large (PB)", and their membership functions are determined.

[0051] b. Fuzzy Rule Base Design: The core of the rule base lies in introducing R(t) and dR(t) / dt as key bases for adjusting PID parameters. An example is shown below: Rule 1: IF e is PS AND de / dt is PB AND R(t) is PS AND dR(t) / dt is PB, THEN ΔKp is PB, ΔKi is NS, ΔKd is PS.

[0052] Explanation: This rule describes a scenario where "pressure is slightly high, rising rapidly, and risk is accumulating at an accelerated pace." Systematic decision-making significantly enhances the proportional effect. To quickly suppress pressure and moderately weaken the role of points. To prevent over-integral saturation caused by rapid adjustment, a derivative action is added. To provide damping.

[0053] Rule 2: IF e is Z AND de / dt is NS AND R(t) is PM AND dR(t) / dt is Z, THEN ΔKp is PM, ΔKi is Z, ΔKd is PM.

[0054] Explanation: This rule describes a situation where "stress has stabilized, but the risk factor remains high and unchanged." System decision-making involves maintaining strong proportional and differential actions. This is to maintain the system's rigidity and rapid response capability under the current high-risk conditions and prevent pressure rebound.

[0055] c. Defuzzification and Parameter Self-Tuning: Based on all activated fuzzy rules, the centroid method is used for defuzzification calculation to obtain the real-time adjustment values ​​of the PID parameters ΔKp, ΔKi, and ΔKd. The PID parameters are updated online: Kp = Kp0 + ΔKp, Ki = Ki0 + ΔKi, Kd = Kd0 + ΔKd. Where Kp0, Ki0, and Kd0 are a pre-tuned set of basic PID parameters.

[0056] d. Calculate the control quantity u(t) using the updated Kp, Ki, Kd, ​​and output it to the actuator.

[0057] e. At the same time, the HMI interface displays a yellow warning and dynamically displays the real-time value and change curve of R(t).

[0058] Emergency intervention zone (P(t) ≥ 0.085MPa or R(t) ≥ 0.8): This area represents extremely high risk; the system executes the highest priority hard safety logic, immediately overriding all algorithm outputs. a. Output a fixed signal (e.g., 4mA) to the variable frequency burner to maintain the minimum firing power.

[0059] b. Fully open the water supply regulating valve V1 to perform high-flow "quenching" cooling.

[0060] c. Trigger the red alarm on the on-site audible and visual alarm, the HMI will display an emergency alarm screen, and all data before and after the event will be forcibly recorded for one minute.

[0061] S203: Intelligent hierarchical response and status diagnosis of physical protection units Graded response: If electronic control fails completely, the pressure continues to rise: When the first physical safety threshold (0.09 MPa) is reached, the spring-loaded safety valve A should automatically open to release pressure.

[0062] When the final safety threshold (0.12 MPa) is reached, rupture disc B ruptures to release pressure. Manual valve C is used for manual pressure release during maintenance.

[0063] Intelligent diagnostic function: Online testing and status recording: When the system pressure exceeds 0.08 MPa (close to the safety valve setpoint), the PLC, while executing control intervention, can send a brief pulse signal to a parallel test solenoid valve (not shown in the figure), causing the valve core of safety valve A to move slightly. Feedback from the microswitch S1 can determine whether the valve core moves smoothly and without jamming. The system records the pressure, time, and duration of each actual start-up or test action.

[0064] Health Assessment and Early Warning: The diagnostic algorithm module periodically analyzes historical operation data of safety valves. For example, if data analysis reveals that the actual recorded opening pressures of safety valve A in its last three operations were 0.091 MPa, 0.092 MPa, and 0.094 MPa, showing an upward drifting trend, the system determines that its spring may be fatigued. Even if the valve has not yet reached the point of failure to operate, the system will generate a predictive maintenance work order stating "Safety valve A's opening pressure shows a drifting trend; calibration is recommended" and push it to the management platform. Similarly, analyzing the historical background noise of the rupture disc monitoring signal can assess whether there is a chronic leak.

[0065] S204: Panoramic Data Recording, Intelligent Diagnosis, and Remote Early Warning The system records the pressure-time curve, R(t) curve, all sensor data, control commands, valve status, alarms and event logs at a high frequency. Each data entry is accompanied by a precise timestamp, forming a traceable "digital black box".

[0066] For any event that enters the "early warning intervention zone" or triggers physical protection actions, the system automatically correlates the preceding and following data to generate a structured event analysis report. The report not only records the event phenomena but also analyzes the causes (such as a sudden drop in steam flow of XX%), the system response process (such as how fuzzy PID parameters are adjusted), the control effect (whether the pressure was successfully suppressed below XX%), and the status of relevant safety components.

[0067] All data, analysis reports, and diagnostic warning information can be uploaded to the remote cloud platform via the communication module. The platform provides a visual dashboard that centrally displays the real-time R(t) values, safety component health indices, and warning information of each device, achieving a leap from single-machine intelligent overpressure prevention to group-based intelligent safety management.

[0068] Example Description Suppose that at a certain moment, the steam valve of the cooking line is suddenly shut off due to a malfunction, and the steam consumption D(t) drops sharply from 0.4t / h to nearly 0 within 2 seconds.

[0069] Initially: P(t) = 0.072 MPa, dP / dt ≈ 0, R(t) ≈ 0.2, the system is in the normal adjustment range and runs smoothly with feedforward compensated PID.

[0070] Three seconds later: Due to steam production exceeding steam consumption, P(t) rapidly rises to 0.079 MPa, dP / dt = 0.009 MPa / s, and the deviation of D(t) - D_avg is extremely large. Calculations show that R(t) jumps to 0.7. The system immediately enters the early warning intervention zone.

[0071] System Response: Based on the input states e(t) = negative (small), de / dt = positive (large), R(t) = positive (large), and dR(t) / dt = positive (large), the fuzzy PID controller matches the rule base and outputs adjustment commands that significantly increase Kp and Kd and decrease Ki. The controller rapidly reduces the burner power in an "aggressive but controlled" manner. Simultaneously, the HMI displays a yellow warning and the real-time R(t) curve.

[0072] Result: Before the pressure reached 0.085 MPa, the upward trend was effectively suppressed and began to decline, and the warning was lifted. The entire process was recorded by the system, and the event report showed that "the warning was triggered by a sudden drop in steam load, and the fuzzy PID control based on R(t) successfully intervened at a pressure of 0.082 MPa, with a peak pressure of 0.083 MPa."

[0073] In extreme cases: If the PLC output module malfunctions at this time, the electronic control will completely fail. The pressure will rise to 0.09 MPa, and safety valve A should open. The system will not only record the pressure relief event but also confirm the valve's normal operation via a microswitch signal and record the actual opening pressure (e.g., 0.091 MPa) for subsequent health analysis. If the safety valve also jams, the pressure will ultimately be maintained at 0.12 MPa by the rupture disc B. Afterward, the complete data chain can be retrieved to clearly trace the entire process from "electronic control failure" to "physical protection action," and the effectiveness of each safety barrier can be quantitatively assessed.

[0074] The above description is only a preferred embodiment of the present invention and should not be construed as a limitation of this application. All equivalent changes and modifications made in accordance with the scope of the patent application of the present invention should be covered by the present invention.

Claims

1. A dual overpressure protection safety control system for a steam-water dual-purpose boiler, characterized in that: The system includes an enhanced signal acquisition unit, an electronic control unit, an actuator, a physical protection unit, and a human-machine interaction and data management unit; The physical protection unit includes a three-stage relief device independently installed on the steam drum. The three-stage relief device includes a spring-loaded safety valve A, a rupture disc B, and a manual pressure relief valve C installed sequentially on the steam drum. The enhanced signal acquisition unit includes a pressure monitoring module, a flow monitoring module, a temperature monitoring module, and a safety component status monitoring module. The pressure monitoring module is a pressure sensor P1 installed on the top of the steam drum, used to acquire the steam drum pressure signal P(t) in real time. The flow monitoring module is a vortex flow meter F1 installed on the main steam pipeline, used to measure the instantaneous steam consumption D(t) in real time. The temperature monitoring module consists of a PT100 temperature sensor T1 installed on the steam outlet pipeline and a temperature sensor T2 installed on the feedwater pipeline, used to monitor the steam temperature and feedwater temperature respectively. The safety component status monitoring module consists of a microswitch S1 installed on the valve stem of the spring-loaded safety valve A and a high-sensitivity differential pressure switch ΔP1 or acoustic emission sensor AE1 connected upstream and downstream of the rupture disc B. The microswitch S1 is used to detect whether the safety valve has activated; the high-sensitivity differential pressure switch ΔP1 or acoustic emission sensor AE1 is used to detect minor leaks or rupture events of the rupture disc. The electronic control unit is a PLC programmable logic controller, and all data acquired by the enhanced signal acquisition unit is connected to the PLC programmable logic controller. The human-machine interaction and data management unit includes an HMI and a data storage module. The HMI is communicatively connected to the PLC programmable logic controller. The industrial touch screen is used to display real-time data, set parameters, query historical alarms and event reports, and the data management unit is used to continuously record operating data. The actuator includes a variable frequency burner, an electric water pump, and an electric regulating valve V1 that receive analog outputs from the PLC. The PLC programmable logic controller controls the variable frequency burner, the electric water pump, and the electric regulating valve V1 based on the data collected from each module in the enhanced signal acquisition unit.

2. The dual overpressure protection safety control system for a steam-water dual-purpose boiler according to claim 1, characterized in that: The real-time data in the human-computer interaction and data management unit includes pressure P(t), risk coefficient R(t) curve, and health status of safety components.

3. A method for a dual overpressure protection safety control system for a steam-water dual-purpose boiler as described in claim 1, characterized in that: Includes the following steps: Step S1: Using T as the sampling period, the steam drum pressure P(t), steam flow rate D(t), and feedwater temperature Tw signals are collected in real time through each module in the enhanced signal acquisition unit, and the pressure change rate dP / dt and dynamic risk prediction coefficient R(t) are obtained. Step S2: The PLC programmable logic controller enters different control zones based on the current pressure P(t) and the obtained R(t) value: If P(t) < 0.075MPa and R(t) < 0.4, it enters the normal adjustment zone and adopts "conventional PID control with feedforward compensation" to gently intervene in the initial risk while maintaining pressure stability; if 0.075MPa ≤ P(t) < 0.085MPa or 0.4 ≤ R(t) < 0.8, it enters the early warning intervention zone and switches to "fuzzy PID adaptive control based on R(t) feedforward" for more aggressive and compliant intervention; if P(t) ≥ 0.085MPa or R(t) ≥ 0.8, it enters the emergency intervention zone and executes the highest priority hard safety logic, immediately overriding all algorithm outputs. Step S3: If the pressure continues to rise and reaches the first physical safety threshold, the spring-loaded safety valve A will automatically open to release pressure. If the pressure continues to rise and reaches the final safety threshold, the rupture disc B will rupture to release pressure. The manual valve C is used for manual pressure release during maintenance. At the same time, when the system pressure exceeds 0.08 MPa, the PLC programmable logic controller will perform online testing to determine whether the safety valve core moves flexibly and whether there is any jamming, and record the status. The historical action data of the safety valve will be analyzed periodically to conduct health assessments and provide early warnings. Step S4: The system records the "pressure-time" curve, R(t) curve, all sensor data, control commands, valve status, alarm and event logs at a high frequency. Each data entry is accompanied by a precise timestamp, forming a traceable "digital black box".

4. The method for a dual overpressure protection safety control system for a steam-water dual-purpose boiler according to claim 3, characterized in that: The method includes the following steps: the first physical safety threshold is 0.09 MPa, and the final safety threshold is 0.12 MPa.

5. The method for dual overpressure protection safety control of a steam-water dual-purpose boiler according to claim 3, characterized in that: The formula for calculating the pressure change rate in step S1 is as follows: dP / dt = [P(t) - P(t-1)] / T; The dynamic risk prediction coefficient R(t) is calculated using an enhanced calculation formula that integrates multi-source information: P_pred = P(t) + dP / dt × Δt; ； R(t) = α×[P(t) - Pset] / (Pmax - Pset) + β * / (dP / dt)max + γ * ; Where Pset is the pressure setpoint, Pmax is the design pressure, (dP / dt)max is the maximum allowable pressure change rate, D_avg is the sliding average of steam flow over a period of time, representing the current average load; D_max is the maximum range of the flow meter; α, β, γ, δ are weighting coefficients, satisfying α+β+γ+δ=1. F(P(t), t) is the short-term pressure trend prediction term; Δt is the prediction step size, and R(t) ranges from [0, 1].

6. The method for dual overpressure protection safety control of a steam-water dual-purpose boiler according to claim 4, characterized in that: Step 2 employs "conventional PID control with feedforward compensation" to gently intervene in the initial risk while maintaining pressure stability. The specific steps are as follows: Step a1: Calculate the pressure deviation e(t): e(t) = Pset - P(t); Step a2: Calculate the standard PID output u_pid(t): u_pid(t) = Kp×e(t) + Ki×Σe(t) + Kd×[e(t) - e(t-1)] / T; Step a3: Calculate the dynamic feedforward quantity u_ff(t): u_ff(t) = Kf×R(t); Kf is the feedforward gain coefficient; Step a4: Synthesize the final control quantity u(t): u(t) = u_pid(t) + u_ff(t); After amplitude limiting, it is converted into a 4-20mA signal output to the variable frequency burner.

7. The method for dual overpressure protection safety control of a steam-water dual-purpose boiler according to claim 4, characterized in that: In step S2, the switch to "fuzzy PID adaptive control based on R(t) feedforward" for more proactive and compliant intervention is as follows: Step b1, Fuzzification: Divide the pressure deviation e(t), deviation change rate de(t) / dt, dynamic risk coefficient R(t), and risk change rate dR(t) / dt into five fuzzy subsets: "negative large (NB)", "negative small (NS)", "zero (Z)", "positive small (PS)", and "positive large (PB)", and determine their membership functions. Step b2, Fuzzy Rule Base Design: Introduce R(t) and dR(t) / dt as key bases for adjusting PID parameters; Defuzzification and Parameter Self-Tuning: Based on all activated fuzzy rules, use the centroid method for defuzzification calculation to obtain the real-time adjustment amounts ΔKp, ΔKi, and ΔKd of the PID parameters, and update the PID parameters online. Kp = Kp0 + ΔKp, Ki = Ki0 + ΔKi, Kd = Kd0 + ΔKd; Wherein, Kp0, Ki0, and Kd0 are a set of pre-tuned basic PID parameters; Step b3: Calculate the control quantity u(t) using the updated Kp, Ki, Kd, ​​and output it to the actuator; Step b4: At the same time, the HMI interface displays a yellow warning and dynamically displays the real-time value and change curve of R(t).

8. The method for dual overpressure protection safety control of a steam-water dual-purpose boiler according to claim 4, characterized in that: The specific steps in step S2 to execute the highest priority hard safety logic and immediately override all algorithm outputs are as follows: Step c1: Output a fixed signal to the variable frequency burner to maintain the minimum firepower. Step c2: Fully open the water supply regulating valve V1 to perform high-flow "quenching" cooling; Step c3: Trigger the red alarm of the on-site audible and visual alarm, the HMI will pop up an emergency alarm screen, and all data before and after the event will be forcibly recorded for one minute.

9. The method for dual overpressure protection safety control of a steam-water dual-purpose boiler according to claim 3, characterized in that: In step S3, the method for determining whether the safety valve core moves flexibly and is not stuck is that while the programmable logic controller is executing control intervention, it sends a short pulse signal to a parallel test solenoid valve, causing the valve core of safety valve A to make a micro-motion. Through the feedback of the micro-switch S1, it can be determined whether the valve core moves flexibly and is not stuck.

Images