Coal gasification oxygen-coal ratio control method based on DCS built-in APC

Abstract

The invention belongs to the technical field of industrial control, discloses a coal gasification oxygen-coal ratio control method based on a DCS built-in APC, and aims to solve the problems of frequent manual intervention, slow coal quality change response and low control precision in traditional control. A robust multivariable pre-estimation model is constructed based on steady-state historical data and is adaptively updated, a card edge control and hierarchical adaptive adjustment strategy is adopted, the model is integrated into a control system to realize full-automatic control, and a full-process configuration debugging and multi-fault tolerance mechanism is matched, so that the core advantage of the method is that manual operation is thoroughly eliminated; model adjustment is rapidly completed when the coal ash content changes, so that the CH4 content of the product is stable, the effective gas yield is greatly increased, the annual failure rate of the system is reduced, and the safety, stability and economic benefits of the coal gasification process are remarkably improved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial control, and particularly relates to a coal gasification oxygen-to-coal ratio control method based on a DCS built-in APC. BACKGROUND

[0002] Coal gasification technology occupies a core position in coal chemical production such as synthetic ammonia, methanol, and ethylene glycol, and the oxygen-to-coal ratio of a gasification furnace is a key process parameter that affects the gasification reaction efficiency, synthesis gas quality, and furnace safety.

[0003] In the prior art, an operator usually needs to frequently adjust manually according to a small amount of real-time collected process parameters, such as furnace temperature, pressure, and synthesis gas components. This experience-based adjustment method, on the one hand, leads to high labor intensity and low efficiency of the operator, and on the other hand, due to the experience difference of different operators, the control accuracy and consistency are difficult to guarantee, and frequent manual adjustment easily causes working condition fluctuations, resulting in high standard deviation of CH4 content in synthesis gas, affecting the yield and quality of downstream products, and even possibly causing safety problems such as over-temperature and over-pressure due to improper adjustment. According to industry statistics, under the traditional control mode, the operator may manually adjust the oxygen-to-coal ratio dozens of times a day, and 80% of unplanned shutdowns of coal chemical gasification furnaces are directly related to manual adjustment errors.

[0004] In addition, the coal gasification process is a complex system with multiple variables and strong coupling, which is easily affected by coal quality fluctuations, changes in coal powder feed quantity, and other external disturbances. The existing PID and other classic control algorithms are difficult to effectively respond to these nonlinear and time-varying disturbances, leading to control lag and overshoot, and failing to accurately control the oxygen-to-coal ratio in the optimal range, resulting in insufficient coal gasification or waste of oxygen. In particular, when the coal quality changes greatly, the traditional control method responds slowly and cannot be adjusted in time, leading to abnormal synthesis gas components and affecting production efficiency and product quality. SUMMARY

[0005] The technical problem to be solved by the present application is that the coal gasification oxygen-to-coal ratio control method needs to be frequently adjusted manually, and when the coal quality changes greatly, the traditional control method responds slowly and cannot be adjusted in time, leading to abnormal synthesis gas components. To solve the above problems, the present application provides a coal gasification oxygen-to-coal ratio control method based on a DCS built-in APC.

[0006] To achieve the above purpose, the present application adopts the following technical scheme: a coal gasification oxygen-to-coal ratio control method based on a DCS built-in APC, comprising the following steps: S1. Multi-dimensional data acquisition and preprocessing: Oxygen purity and gasifier operating parameters are acquired in real time through the hardware layer of the DCS system. The gasifier operating parameters include temperature, pressure, pulverized coal feed rate, and syngas composition. The core hardware includes a mass flow meter, a K-type thermocouple, a pressure transmitter, an online gas chromatograph, and an electric regulating valve. The accuracy of oxygen purity measurement is ±0.1%, gasifier temperature measurement is ±1℃, pressure measurement is ±0.075%, pulverized coal feed rate measurement is ±0.1%, and CH4 detection accuracy in syngas is ±1ppm. S2. Dynamic Model Construction and Adaptive Update: A dynamic model is constructed based on at least 3 months of steady-state historical data. The steady-state criteria are temperature fluctuation ≤ ±20℃ and pressure fluctuation ≤ ±0.1MPa. A robust multivariate predictor controller is trained using the least squares method with regularization. The regularization coefficient is 0.001 to 0.01. The prediction time domain length is 10 to 20 sampling periods, and the control time domain length is 5 to 10 sampling periods. The model prediction error is ≤5%. A rolling time domain optimization strategy is adopted, with an optimization window length of 15 to 20 minutes. When the coal ash content changes by ±3%, the model is adaptively adjusted within 10 minutes. Under steady-state conditions, the deviation between the predicted and actual CH4 content values ​​is ≤10ppm. S3. Edge control and adaptive adjustment strategy execution: Edge control sets a safety upper limit based on coal type, with a limit of 0.01 Nm. 3 The oxygen-coal ratio is gradually increased in increments of / kg; when the heat load demand exceeds 90% of the design load, edge optimization is initiated, pausing for 3 minutes after each increase. If any of the following conditions are met: CH4 content > 195ppm, temperature > 1500℃, or pressure > 4.0MPa, the ratio is immediately reduced by 0.02Nm. 3 / kg; Adaptive adjustment corrects the set value according to the set trigger conditions and rate, with the adjustment priority being safety constraints > quality constraints > economic constraints, to ensure that the CH4 content is ≤195ppm; S4. Fully automated control implementation: The Honeywell EPKSDCS system integrates the C300 controller as the control core, and adopts a hierarchical network architecture to realize data transmission and control command issuance; the APC operation interface is integrated into the operator station, supporting three-level permission management and dual authentication. S5. System full-process configuration, deployment and debugging: including initialization settings, parameter configuration, permission and calibration settings, deployment and debugging, completing device communication configuration, parameter standardization settings, calibration mechanism establishment and multi-condition debugging and verification; S6. Fault Tolerance Guarantee: Fault tolerance mechanisms are set up for scenarios such as sensor failure, actuator jamming, and model failure to ensure that the annual system failure rate is ≤0.5%, ultimately achieving an effective gas production increase of ≥10% and reducing the frequency of manual operation to 0 times / day.

[0007] Further, the data collection in S1 also includes coal quality parameters and steam production; the coal quality parameters include ash content of 15-25% and moisture content of 5-10%, which are obtained by combining real-time collection by an online coal quality analyzer and daily updates in the laboratory; the steam production measurement range is 0-50 t / h; the sampling frequency of all collected parameters is ≤1 Hz, and a timestamp synchronization technology is used; the oxygen flow is subjected to 5-point moving average filtering processing, and abnormal values within ±3σ are eliminated, wherein σ is the standard deviation of the collected data within the last 1 hour; the coal powder feed quantity is cross-verified with the data of a weighing sensor; when the oxygen purity is lower than 99.8%, the oxygen-coal ratio calculation result is automatically corrected.

[0008] Further, the modeling data quantity of the dynamic model in S2 is ≥100,000 groups, and the update period is ≤5 minutes; when the coal quality ash content changes by ±3%, the model completes self-adaptive adjustment within 10 minutes, and the deviation between the predicted value and the actual value of CH4 content under a steady state is ≤10 ppm.

[0009] Further, the safety upper limit of the card edge control in S3 is set according to the coal type: 0.85 Nm 3 / kg for lignite, 0.83 Nm 3 / kg for bituminous coal, and 0.87 Nm 3 / kg for anthracite; the trigger conditions for self-adaptive adjustment include a change of ±3% in the coal quality ash content, a change of ±2% in the moisture content, a change of ±300 kg / h in the coal powder feed quantity, a change of ±500 Nm 3 / h in the oxygen flow, a fluctuation of ±20 ppm in the CH4 content, and a change of ±500 Nm 3 / h in the effective gas production, and the adjustment rate is 0.005 Nm 3 / kg / second.

[0010] Further, the hierarchical network architecture in S4 includes an L1 layer field network, an L2 layer control network, and an L3 layer operation network, wherein the L1 layer field network adopts a PROFIBUS-DP protocol; the L2 layer control network is a redundant Ethernet architecture, the communication rate is 1000 Mbps, the redundant switching time is ≤50 ms; the control response time is ≤2 seconds, and the signal transmission delay is ≤100 ms; the L3 layer operation network is used by operating personnel to realize monitoring and control operations through a terminal.

[0011] Further, the APC operation interface in S4 includes a real-time monitoring panel, a parameter setting module, a fault alarm module, and a historical data tracing module; the trend chart refresh frequency of the real-time monitoring panel is 1 time / second, and the historical data storage time length is ≥1 year; the three-level permissions are as follows: an administrator can modify all parameters and control algorithms, a process engineer can modify control strategy related parameters, and an operator can only view real-time / historical data and confirm alarm information; the permission verification adopts a dual verification mode of an account password and an operation permission code.

[0012] Further, the fault-tolerant mechanism in S6 is as follows: when the sensor fails, the redundant sensor is automatically switched within 1 second and an alarm is triggered, the maintenance response time is less than or equal to 4 hours; when the actuator is stuck, the manual control mode is switched within 2 seconds, and the maintenance response time is less than or equal to 8 hours; when the model fails, the standby model is loaded within 5 seconds, and the automatic control mode is restored within 24 hours.

[0013] Further, the initialization setting and parameter configuration in S5 include: Hardware initialization: power-on self-test is performed on the C300 controller, communication link connectivity is confirmed, and the bit error rate is required to be less than or equal to 0.001%; Software initialization: load the APC control core program, and configure the initial algorithm parameters of the robust multivariable predictive controller; Reference parameters: set the initial value of the oxygen-coal ratio according to the coal type, wherein the initial value of the oxygen-coal ratio for lignite is 0.8 Nm 3 / kg, the initial value of the oxygen-coal ratio for bituminous coal is 0.78 Nm 3 / kg, and the initial value of the oxygen-coal ratio for anthracite is 0.82 Nm 3 / kg; at the same time, set the steady-state judgment reference value, the temperature is 1200-1500℃, the pressure is 2.5-4.0 MPa, and the coal powder feed rate is 5000-8000 kg / h; Acquisition parameters: the sampling frequency of oxygen flow and coal powder feed rate is 0.5 Hz, and the sampling frequency of temperature and pressure is 1 Hz; configure the 5-point sliding average filter window parameters corresponding to the oxygen flow and the ±3σ abnormal value rejection threshold; Model and control parameters: set the modeling data threshold to be greater than or equal to 100,000 groups, the update period to be 3-5 minutes, the optimization window length to be 15-20 minutes, and the step size, residence time, and callback amplitude of the edge control parameters.

[0014] Further, the permission and calibration setting in S5 includes: Permission configuration: set the operation permission according to a three-level permission and a double verification mechanism; Sensor calibration: the Rosemount 3051S mass flowmeter, the Siemens SITRAN SP300 pressure transmitter, the K-type thermocouple, and the Agilent 7890B online gas chromatograph are calibrated once a month, once a quarter, once a week, and once a day respectively; the calibration error is required to be less than or equal to 50% of the corresponding device measurement accuracy, and the calibration data is automatically synchronized to the dynamic model; Coal quality calibration: the online coal quality analyzer is compared and calibrated with the laboratory data every 2 hours, and the deviation is required to be less than or equal to 1 percentage point, and the coal quality parameter library is updated daily.

[0015] Further, the deployment and debugging in S5 include: Network deployment: allocate a unique communication address for all on-site hardware devices, and configure the communication rate and switching mechanism of the L2 layer redundant Ethernet; System commissioning: 72 hours of continuous commissioning, simulating working condition fluctuation to verify the effectiveness of control; during commissioning, the CH4 content is required to be ≤195ppm, and the standard deviation is ≤20ppm, and the effective gas production fluctuation is ≤±5%; Fault commissioning: simulate sensor failure, actuator jamming, model failure scenarios, verify the response speed of the fault handling mechanism, and ensure that the switching time meets the set requirements.

[0016] Technical effects and advantages of the present application: 1. The present application embeds the APC function in the DCS system, realizes the full-automatic closed-loop control of oxygen-coal ratio from data acquisition, model prediction to control instruction issuing through highly integrated software and hardware, completely eliminates manual adjustment, reduces the manual operation frequency to 0 times / day, greatly reduces the labor intensity of the operator and the risk of misoperation due to experience difference or fatigue, and improves the safety and stability of system operation.

[0017] 2. The present application can accurately push the oxygen-coal ratio to the upper limit of the optimal value under the premise of meeting various safety constraints, thereby maximizing the production of effective gas, and combining with model self-adaptive adjustment, the coal quality change can be quickly responded, and always maintained in the optimal control interval, finally realizing the improvement of effective gas production and bringing significant economic benefit growth. BRIEF DESCRIPTION OF DRAWINGS

[0018] The disclosure of the present application will be described with reference to the accompanying drawings. It should be understood that the drawings are for illustrative purposes only and are not intended to limit the scope of protection of the present application. In the drawings, the same reference numerals are used to refer to the same parts: Figure 1 is the oxygen-coal ratio dynamic control flowchart of the present application; Figure 2 is the DCS embedded APC system architecture diagram of the present application; Figure 3 is the CH4 content and effective gas flow trend comparison diagram of the present application. DETAILED DESCRIPTION

[0019] It is easy to understand that according to the technical scheme of the present application, those skilled in the art can propose a plurality of structure modes and implementation modes which can be replaced with each other without changing the essential spirit of the present application. Therefore, the following specific embodiments and drawings are only exemplary description of the technical scheme of the present application, and should not be regarded as the whole or regarded as the limitation or restriction of the technical scheme of the present application.

[0020] The application provides a coal gasification oxygen-to-coal ratio control method based on a DCS built-in APC, aiming at solving the problems of low control precision, poor stability, frequent manual intervention, limited system reliability and insufficient economic benefits in the prior art. The application realizes full automation, high precision and high reliability control of the oxygen-to-coal ratio by deeply integrating advanced process control functions into a distributed control system, combining high-precision data acquisition, dynamic model adaptive updating and intelligent optimization control strategies, and maximally improving the effective gas production.

[0021] Embodiment one: system overall architecture and data acquisition and preprocessing This embodiment describes the overall architecture, data acquisition range and key preprocessing methods of the control method of the application, in cooperation with the DCS built-in APC system architecture diagram shown in Figure 2 .

[0022] The application adopts the Honeywell EPKSR601 DCS system as a core platform, integrates the APC function in the DCS through the C300 controller, and constitutes an integrated control system. The system architecture is divided into an L1 layer field network, an L2 layer control network and an L3 layer operation network, and realizes hierarchical management of data transmission and control instruction issuing.

[0023] S1. Multi-dimensional data acquisition and preprocessing In the L1 layer field network, a series of high-precision sensors and analyzers are used to acquire the key operation parameters of the gasification furnace and the related feeding system in real time. These parameters include: Oxygen purity: the oxygen purity is acquired in real time by an outlet purity analyzer of an air separation device, for example, a zirconia analyzer, and the measurement accuracy can reach ±0.1%. When the oxygen purity is lower than 99.8%, the system will automatically correct the calculation result of the oxygen-to-coal ratio to compensate for the influence of the change of the oxygen purity on the actual oxygen supply.

[0024] Gasification furnace operation parameters: Temperature: the temperature of multiple temperature measuring points in the furnace hearth of the gasification furnace is monitored in real time by a K-type thermocouple, for example, three temperature measuring points of upper, middle and lower. The measurement range of the thermocouple is 0-1800℃, and the accuracy is ±1℃. The system takes the average value of multiple points as the furnace hearth temperature, and has the function of automatically switching the standby sensor of the abnormal point, to ensure the reliability of the temperature data.

[0025] Pressure: the pressure in the furnace is acquired in real time by a pressure transmitter, for example, a SITRAN SP300 pressure transmitter. The measurement range of the pressure transmitter is 0-6.0MPa, and the accuracy is ±0.075%. The data will be subjected to pressure compensation calculation to correct the influence of temperature on pressure measurement.

[0026] Coal feed rate: Real-time acquisition through mass flow meters, such as 3051S mass flow meters, with a measurement range of 0-10000 Nm 3 / h, with an accuracy of ±0.1%, and to improve accuracy, the coal feed rate data will be cross-verified with independent weighing sensor data.

[0027] Synthesis gas components: Real-time analysis of key components in synthesis gas, including CH4, CO, CO2, etc., through online gas chromatographs, such as 7890B online gas chromatographs, with a response time ≤30s, and the detection accuracy of CH4 can reach ±1ppm. The CH4 content data is smoothed to remove the interference caused by chromatographic peaks.

[0028] Other extended data collection: Coal quality parameters: including ash content and moisture content, with ash content ranging from 15% to 25% and moisture content ranging from 5% to 10%. These data are collected in real time through online coal quality analyzers and supplemented by laboratory daily coal quality test data to ensure the accuracy and timeliness of coal quality data.

[0029] Steam production: Real-time acquisition through steam outlet flow meters, with a measurement range of 0-50 t / h, for stable working condition judgment.

[0030] The sampling frequency of all collected parameters is set to ≤1Hz, and time stamp synchronization technology is used to ensure strict matching between different parameter data at the same time, providing high-quality time series data for subsequent dynamic model construction.

[0031] In the data preprocessing stage, 5-point moving average filtering is performed on the oxygen flow data to eliminate random noise, and based on the standard deviation (σ) of the data collected in the last 1 hour, abnormal values deviating from the average value ±3σ are removed to ensure the cleanliness of data input.

[0032] Core hardware devices include: mass flow meters, K-type thermocouples, pressure transmitters, online gas chromatographs, and electric regulating valves as actuators. These high-precision and high-reliability hardware devices are the basis for precise control.

[0033] Hierarchical network architecture: L1 layer field network adopts PROFIBUS-DP protocol, connecting each sensor, analyzer and actuator with C300 controller; L2 layer control network is a redundant Ethernet architecture, with a communication rate of 1000 Mbps and a redundant switching time of ≤50 ms, this layer network bears the high-speed data exchange between C300 controller and EXPERION SERVER, ConsoleStation, ensures the fast response of the control loop, the control response time is ≤2 seconds, the signal transmission delay is ≤100 ms, greatly improves the reliability and real-time performance of the system; L3 layer operation network connects EXPERION SERVER and FLEXStation operation station, for the operator to realize monitoring and control operation through the terminal, and is physically isolated from L2 layer through a network lock, to ensure data security.

[0034] Example two: dynamic model construction and adaptive updating and control strategy execution This embodiment details the core of the control method of the application, the construction of the dynamic model, the adaptive updating mechanism and the execution of the card edge control and adaptive adjustment strategy, which cooperates Figure 1 The oxygen-coal ratio dynamic control flowchart shown in the figure.

[0035] S2. Dynamic model construction and adaptive updating In the C300 controller inside the DCS system, an advanced APC algorithm module is built in, which is responsible for constructing and maintaining the dynamic model of the oxygen-coal ratio of the gasification furnace.

[0036] Model construction: based on at least 3 months of steady-state historical data to construct a dynamic model. The steady-state determination standard is that the gas furnace temperature fluctuation is ≤±20℃, and the furnace pressure fluctuation is ≤±0.1MPa. In order to ensure the comprehensiveness and robustness of the model, the modeling data amount is not less than 100,000 groups. Least square method+regularization processing is used to train the robust multivariable predictive controller, i.e. RMPC, wherein the regularization coefficient is 0.001-0.01. This RMPC model can handle complex coupled systems with multiple inputs and multiple outputs, such as the dynamic correlation of oxygen-coal ratio with oxygen flow, CH4 content, temperature, pressure, coal quality, effective gas production and other key parameters. The prediction time domain length is set to 10-20 sampling periods, and the control time domain length is 5-10 sampling periods. After training, the prediction error of the model is strictly controlled to ≤5%.

[0037] Model adaptive updating: in order to cope with internal and external disturbances such as coal quality change and catalyst activity decay, the application establishes an efficient model adaptive updating mechanism.

[0038] Rolling horizon optimization strategy is adopted, and the latest running data is real-time incorporated for model parameter correction, and the optimization window length is 15-20 minutes.

[0039] The model update cycle is set to ≤5 minutes, ensuring that the model always maintains the latest system dynamic characteristics.

[0040] When the coal ash changes by ±3%, such as from 15% to 18% or 12%, the system can complete the adaptive adjustment of the model within 10 minutes. After adjustment, the deviation of the CH4 content prediction value from the actual value under steady-state conditions is ≤10 ppm, ensuring the rapid response and accurate prediction of the model to changes in working conditions.

[0041] S3. Card edge control and adaptive adjustment strategy execution The present application combines Figure 1 control logic flow, based on dynamic models, through card edge control and adaptive adjustment strategy, to achieve optimal control of oxygen-coal ratio.

[0042] Card edge control: The goal of card edge control is to maximize effective gas production while ensuring the safe operation of the gasifier.

[0043] Safety upper limit setting: The safety upper limit of the oxygen-coal ratio will be set according to different coal types, such as 0.85 Nm 3 / kg for lignite, 0.83 Nm 3 / kg for bituminous coal, and 0.87 Nm 3 / kg for anthracite. These upper limit values are determined based on the equipment tolerance limit of the gasifier, reaction thermodynamics, and historical operation experience after risk assessment.

[0044] Card edge optimization start condition and process: When the heat load demand exceeds 90% of the design load, the system starts card edge optimization. The system will gradually increase the oxygen-coal ratio by 0.01 Nm 3 / kg, and after each increase, the system will stop for 3 minutes for observation.

[0045] Safety callback mechanism: During the process of increasing the oxygen-coal ratio, the system will monitor key safety parameters in real time. If any of the following conditions is met: CH4 content in synthesis gas > 195 ppm, gasifier temperature > 1500℃, or furnace pressure > 4.0 MPa, the system will immediately adjust the oxygen-coal ratio by 0.02 Nm 3 / kg, adjusting the oxygen-coal ratio back to the safety interval, and stabilizing before attempting optimization again. The CH4 content in synthesis gas > 195 ppm is a hard constraint that cannot be broken, the gasifier temperature > 1500℃ is a soft constraint that can be temporarily overheated but needs to be quickly adjusted, and the furnace pressure > 4.0 MPa is a hard constraint that cannot be broken. This mechanism effectively prevents safety accidents caused by excessive optimization.

[0046] Adaptive adjustment: Adaptive adjustment ensures that the oxygen-coal ratio set value can be quickly and accurately corrected when the working conditions fluctuate.

[0047] Trigger condition: the trigger condition of adaptive adjustment is multifaceted, including: coal ash variation ± 3%, moisture variation ± 2%, coal powder feed variation ± 300 kg / h, oxygen flow variation ± 500 Nm 3 / h, CH4 content fluctuation ± 20 ppm, effective gas production variation ± 500 Nm 3 / h.

[0048] Adjustment rate: when any trigger condition is met, the system will correct the oxygen-coal ratio set value at an adjustment rate of 0.005 Nm 3 / kg / sec to avoid system oscillation caused by too fast adjustment, as shown in the rate limiting unit in Figure 1 , with a rate limit of 0.0017 / sec.

[0049] Adjustment priority: the adjustment strategy follows strict priority: safety constraint > quality constraint > economic constraint. This means that the system first ensures the safe operation of the gasifier, then guarantees the quality of the synthesis gas, and finally considers maximizing economic benefits, which embodies the concept of safety production as the core.

[0050] Through the above dynamic model construction and adaptive update, combined with the edge control and adaptive adjustment strategy, the present application can realize precise, stable, safe and efficient control of the oxygen-coal ratio of coal gasification. Figure 1 The input parameter module (such as total oxygen of coal burner, heat load setting, CH4 content, etc.), correction and constraint module (heat load correction, rate limitation, boundary limitation), and execution output module (coal powder adjustment and oxygen flow adjustment of 1#~4# coal burners) in

[0051] Example three: system full process configuration, deployment debugging and fault tolerance protection This example combines Figure 2 and Figure 3 , and details the system full process configuration, deployment debugging process, and key fault tolerance protection mechanism of the control method of the present application, and further illustrates the advantages of the present application through effect verification.

[0052] S4. Full automation control The present application integrates C300 controller as the control core in Honeywell EPKS DCS system, realizes APC function through DCS internal bus, rather than traditional external hanging mode, which fundamentally solves the signal delay and communication interruption problem existing in traditional APC system, and realizes real full automation control.

[0053] Layered network architecture: the system adopts layered network architecture of L1 layer field network, L2 layer control network and L3 layer operation network.

[0054] L1 layer field network adopts PROFIBUS-DP protocol, connects all field devices, and realizes high real-time data acquisition and instruction issuing.

[0055] L2 layer control network is a redundant Ethernet architecture, the communication rate is 1000Mbps, and the redundant switching time is less than or equal to 50ms. This layer network bears high-speed data exchange between C300 controller and server, ensures the fast response of the control loop, the control response time is less than or equal to 2 seconds, the signal transmission delay is less than or equal to 100ms, and the reliability and real-time performance of the system are greatly improved.

[0056] L3 layer operation network is used for operators to realize monitoring and control operation through terminals, and is physically isolated from L2 layer through a network lock, to ensure network security.

[0057] APC operation interface: the APC operation interface is integrated in the FLEXStation operation station, and realizes intelligentization and convenience of man-machine interaction. The interface includes: Real-time monitoring panel: real-time data and trend chart of core parameters such as oxygen-coal ratio set value / actual value, CH4 content, temperature, pressure and effective gas production are presented, and the trend chart refresh frequency is 1 time / second.

[0058] Parameter setting module: key parameters such as card edge upper limit, CH4 threshold value and adjustment step can be set.

[0059] Fault alarm module: when the device fails or the parameter exceeds the standard, the sound and light alarm is automatically triggered, and the fault positioning and processing suggestion are provided.

[0060] Historical data tracing module: more than 1 year of operation data is stored, flexible query is supported, and process optimization and fault analysis are facilitated.

[0061] Permission management and verification: the APC operation interface supports three-level permission management and double verification. Specifically, the administrator can modify all parameters and control algorithms; the process engineer can modify the control strategy related parameters; the operator can only view real-time / historical data and confirm alarm information. The permission verification adopts double verification mode of account password combined with operation permission code, to ensure system security and prevent unauthorized modification and operation.

[0062] S5. System whole-process configuration and deployment debugging The present application needs detailed configuration before system deployment, and each function is verified through strict debugging.

[0063] Initialization setting and parameter configuration: Hardware initialization: power-on self-test is performed on the C300 controller, communication link connectivity is confirmed, and the error code rate is required to be less than or equal to 0.001%.

[0064] Software initialization: Load APC control core program and configure initial algorithm parameters of robust multivariable predictive controller, such as regularization coefficient, prediction horizon length, control horizon length, etc.

[0065] Reference parameters: Set the initial value of oxygen-coal ratio according to coal type, where lignite is 0.8 Nm 3 / kg, bituminous coal is 0.78 Nm 3 / kg, and anthracite is 0.82 Nm 3 / kg. At the same time, set the steady-state judgment reference value, such as temperature 1200-1500℃, pressure 2.5-4.0 MPa, and coal powder feed rate 5000-8000 kg / h.

[0066] Acquisition parameters: Configure oxygen flow and coal powder feed rate sampling frequency as 0.5 Hz, temperature and pressure sampling frequency as 1 Hz. Configure 5-point sliding average filter window parameters corresponding to oxygen flow and ±3σ outlier rejection threshold.

[0067] Model and control parameters: Set modeling data threshold ≥100,000 groups, update period 3-5 minutes, optimization window length 15-20 minutes, and parameters such as step size, residence time, and callback amplitude of edge control.

[0068] Permissions and calibration settings: Permission configuration: According to three-level permissions (administrator, process engineer, operator) and double verification mechanism, configure permissions for operation groups and individuals.

[0069] Sensor calibration: Regularly calibrate key sensors to ensure data accuracy. For example, Rosemount 3051S mass flowmeter, Siemens SITRAN SP300 pressure transmitter are calibrated once a month; K-type thermocouple is calibrated once a quarter; Agilent 7890B online gas chromatograph is calibrated once a week. The calibration error is ≤50% of the corresponding device measurement accuracy. Calibration data is automatically synchronized to the dynamic model to ensure the accuracy of model input data.

[0070] Coal quality calibration: Online coal quality analyzer is compared and calibrated with laboratory data every 2 hours, with a deviation of ≤1 percentage point, and coal quality parameter library is updated daily.

[0071] Deployment and debugging: Network deployment: Assign a unique communication address to all on-site hardware devices, configure L2 layer redundant Ethernet communication rate and switching mechanism to ensure normal communication of all devices.

[0072] System commissioning: 72 hours of continuous commissioning, simulating various operating condition fluctuations to verify the effectiveness of control. During commissioning, CH4 content is required to be ≤195 ppm, with a standard deviation of ≤20 ppm, and effective gas production fluctuation of ≤±5%, to ensure the stability and control effect of the system in actual operation.

[0073] Fault commissioning: simulate sensor failure, actuator sticking, model failure scenarios, verify the response speed and effectiveness of the fault handling mechanism in S6, ensure that the switching time meets the set requirements.

[0074] S6. Fault tolerance protection The present application designs multiple fault tolerance mechanisms for possible key fault scenarios to ensure safe operation and continuity of production in abnormal situations.

[0075] Sensor failure: when a sensor fails, showing abnormal data, no signal, etc., the system can automatically switch to redundant sensors to obtain data within 1 second, and immediately trigger an alarm to notify the operator. The response time of maintenance personnel is required to be ≤4 hours to replace or repair the faulty sensor in time.

[0076] Actuator sticking: when the electric regulating valve and other actuators stick, the system can detect the deviation between the actual output and the command output within 2 seconds, and automatically switch to manual control mode, while sending an alarm. The operator can perform emergency intervention in manual mode to ensure stable production. The response time of maintenance personnel is required to be ≤8 hours.

[0077] Model failure: when the dynamic model's prediction ability decreases or fails due to external interference or abnormal calculation, such as a large deviation between the predicted CH4 content and the actual value, the system can automatically load the verified backup model to take over control within 5 seconds. At the same time, the system records the model failure reason and starts online retraining or manual intervention to ensure the automatic control mode is restored within 24 hours.

[0078] Through the configuration, deployment, commissioning and perfect fault tolerance protection of the above system, the present application can ensure that the annual failure rate is ≤0.5%, and finally achieve an effective gas production increase of ≥10% and a manual operation frequency reduction of 0 times / day, thereby achieving the expected technical and economic benefits.

[0079] Effect verification The present application has achieved significant effect improvement in practical application, as shown in Figure 3

[0080] CH4 content stability improvement: Figure 3 ​The left curve of figure 1 shows that before the APC system is put into operation, the CH4 content in the synthesis gas fluctuates violently, and the standard deviation is as high as 238.9 ppm, while after the APC system is put into operation, Figure 3 The right curve of figure 1 shows that the CH4 content curve tends to be smooth, and the standard deviation is significantly reduced to below 194.61 ppm, and is stably controlled within the safety threshold of 195 ppm, verifying the effectiveness of the dynamic model and the adaptive adjustment strategy, which shows that the present application can significantly improve the stability of the synthesis gas quality.

[0081] Effective gas production is improved: Figure 3 The change trend of the effective gas flow is also compared. Before the APC is put into operation, the effective gas flow is maintained at the benchmark value level, and after the APC is put into operation, the effective gas flow curve shows a steady upward trend, and finally stabilizes at about 70753 Nm 3 / h, which is increased by 82.59 Nm 3 / h compared with the benchmark value, with an increase of 11.27%, which directly reflects the core advantage of the corner optimization strategy in maximizing the effective gas production within the safety boundary, and achieves significant economic benefits.

[0082] Automation and reliability: In the 180-day continuous operation test in a certain coal chemical enterprise, the system has no downtime, all controls are automatically completed by the APC system, the manual operation frequency is reduced to 0 times / day, the CH4 content is always stable below 195 ppm, the heat load fluctuation is ≤±4%, and the system annual failure rate is far lower than 0.5%.

[0083] Finally, it should be noted that: the above examples are only used to illustrate the technical solutions of the present application, and not to limit it; although the present application has been described in detail with reference to the foregoing examples, those skilled in the art should understand that: it can still modify the technical solutions recorded in the foregoing examples, or make equivalent replacement for part of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the spirit and scope of the technical solutions of the embodiments of the present application.

Claims

1. A coal gasification oxygen-to-coal ratio control method based on DCS built-in APC, characterized by, Comprising the following steps: S1. Multidimensional data acquisition and preprocessing: real-time acquisition of oxygen purity, gasifier operating parameters, including temperature, pressure, coal powder feed rate, and syngas composition through DCS system hardware layer devices; core hardware includes mass flow meter, K-type thermocouple, pressure transmitter, online gas chromatograph, electric regulating valve; wherein the oxygen purity measurement accuracy is ±0.1%, the gasifier temperature measurement accuracy is ±1℃, the pressure measurement accuracy is ±0.075%, the coal powder feed rate measurement accuracy is ±0.1%, and the CH4 detection accuracy in syngas is ±1ppm; S2. Dynamic model construction and adaptive update: based on at least 3 months of steady-state historical data, a dynamic model is constructed, and the steady-state determination standard is temperature fluctuation ≤±20℃ and pressure fluctuation ≤±0.1MPa; a robust multivariate predictive controller is trained using least squares method + regularization processing, the regularization coefficient is 0.001-0.01, the prediction time domain length is 10-20 sampling periods, the control time domain length is 5-10 sampling periods, and the model prediction error is ≤5%; a rolling time domain optimization strategy is adopted, and the optimization window length is 15-20 minutes; when the coal ash content changes by ±3%, the model adaptive adjustment is completed within 10 minutes, and the deviation between the predicted value and the actual value of CH4 content under steady-state conditions is ≤10ppm; S3. Card edge control and adaptive adjustment strategy execution: The card edge control sets a safety upper limit according to the coal type, and gradually increases the oxygen-coal ratio by 0.01 Nm 3 / kg as a step size; when the heat load demand exceeds 90% of the design load, the card edge optimization is started, and after each step increase, it stays for 3 minutes; if any of the following conditions is met: CH4 content > 195 ppm, temperature > 1500℃, pressure > 4.0MPa, immediately adjust back by 0.02 Nm 3 / kg; adaptive adjustment adjusts the set value according to the set trigger condition and rate correction, and the adjustment priority is safety constraint > quality constraint > economic constraint, to ensure that the CH4 content is ≤195 ppm; S4. Full automatic control implementation: C300 controller is integrated as the control core in the system, and a hierarchical network architecture is adopted to realize data transmission and control instruction issuance; the APC operation interface is integrated in the operation station, supporting three-level permission management and double verification; S5. System full-process configuration and deployment debugging: including initialization setting, parameter configuration, permission and calibration setting, deployment debugging, completing device communication configuration, parameter standardization setting, calibration mechanism establishment, and multi-condition debugging verification; S6. Fault tolerance guarantee: fault tolerance mechanisms are set for sensor failure, actuator jamming, and model failure scenarios, ensuring that the system annual failure rate is ≤0.5%, and finally achieving ≥10% improvement in effective gas production and reducing manual operation frequency to 0 times / day.

2. The coal gasification oxygen-to-coal ratio control method based on DCS built-in APC according to claim 1, characterized in that, The data acquisition in S1 also includes coal quality parameters and steam production; the coal quality parameters include ash content of 15%-25% and moisture content of 5%-10%, which are obtained through online coal quality analyzer in combination with daily updates from the laboratory; the steam production measurement range is 0-50t / h; the sampling frequency of all collected parameters is ≤1Hz, and timestamp synchronization technology is adopted; oxygen flow is subjected to 5-point moving average filtering processing and ±3σ abnormal values are removed, wherein σ is the standard deviation of the collected data in the last 1 hour; coal powder feed rate is cross-verified with weighing sensor data; when the oxygen purity is lower than 99.8%, the oxygen-coal ratio calculation result is automatically corrected.

3. The method for controlling the oxygen-coal ratio in a coal gasification process based on a DCS embedded APC according to claim 1, characterized in that, The modeling data volume of the dynamic model in S2 is ≥100,000 groups, and the update period is ≤5 minutes; when the coal ash content changes by ±3%, the model completes adaptive adjustment within 10 minutes, and the deviation between the predicted value and the actual value of CH4 content under steady-state conditions is ≤10ppm.

4. The method for controlling the oxygen-coal ratio in a coal gasification process based on a DCS embedded APC according to claim 1, characterized in that, The safety upper limit of the S3 card edge control is set according to coal types: lignite 0.85 Nm 3 / kg, bituminous coal 0.83 Nm 3 / kg, anthracite 0.87 Nm 3 / kg; the trigger conditions of adaptive adjustment include coal ash variation ±3%, moisture variation ±2%, coal powder feed quantity variation ±300 kg / h, oxygen flow variation ±500 Nm 3 / h, CH4 content fluctuation ±20 ppm, effective gas production variation ±500 Nm 3 / h, and the adjustment rate is 0.005 Nm 3 / kg / s.

5. The method for controlling the oxygen-coal ratio in a coal gasification process based on a DCS embedded APC according to claim 1, characterized in that, The S4 layered network architecture comprises an L1 layer field network, an L2 layer control network and an L3 layer operation network, wherein the L1 layer field network adopts a PROFIBUS-DP protocol; the L2 layer control network is a redundant Ethernet architecture, the communication rate is 1000 Mbps, the redundant switching time is less than or equal to 50 ms; the control response time is less than or equal to 2 seconds, and the signal transmission delay is less than or equal to 100 ms; and the L3 layer operation network is used for realizing monitoring and control operation by an operator through a terminal.

6. The method for controlling the oxygen-coal ratio in a coal gasification process based on a DCS embedded APC according to claim 1, characterized in that, The APC operation interface in the S4 comprises a real-time monitoring panel, a parameter setting module, a fault alarm module and a historical data tracing module; the trend chart refresh frequency of the real-time monitoring panel is 1 time per second, and the historical data storage time length is greater than or equal to 1 year; the three-level permissions are as follows: an administrator can modify all parameters and control algorithms, a process engineer can modify control strategy related parameters, and an operator can only view real-time / historical data and confirm alarm information; and the permission verification adopts a double verification mode of account password and operation permission code.

7. The method for controlling the oxygen-coal ratio in a coal gasification process based on a DCS embedded APC according to claim 1, characterized in that, The fault tolerance mechanism in the S6 is as follows: when a sensor fails, the redundant sensor is automatically switched to within 1 second and an alarm information is triggered, and the maintenance response time is less than or equal to 4 hours; when an actuator is stuck, the manual control mode is switched to within 2 seconds, and the maintenance response time is less than or equal to 8 hours; and when a model fails, a standby model is loaded within 5 seconds, and the automatic control mode is restored within 24 hours.

8. The method for controlling the oxygen-coal ratio in a coal gasification process based on a DCS embedded APC according to claim 1, characterized in that, The initialization setting and parameter configuration in the S5 comprise: hardware initialization: performing power-on self-test on the C300 controller, confirming the communication link connectivity, and requiring that the bit error rate be less than or equal to 0.001%; software initialization: loading the APC control core program, and configuring the initial algorithm parameters of the robust multivariable predictive controller; Reference parameters: set initial value of oxygen-coal ratio according to coal type, wherein lignite is 0.8 Nm 3 / kg, bituminous coal is 0.78 Nm 3 / kg, and anthracite is 0.82 Nm 3 / kg; and set steady state determination reference value, wherein temperature is 1200-1500℃, pressure is 2.5-4.0 MPa, and coal powder feeding amount is 5000-8000 kg / h; acquisition parameters: the sampling frequency of oxygen flow and coal powder feed quantity is 0.5 Hz, the sampling frequency of temperature and pressure is 1 Hz, 5-point sliding average filter window parameters corresponding to oxygen flow are configured, and a ±3σ abnormal value elimination threshold is configured; model and control parameters: setting the modeling data quantity threshold to be greater than or equal to 100,000 groups, the update period to be 3-5 minutes, the optimization window length to be 15-20 minutes, and the step length, residence time and callback amplitude of the card edge control and the like parameters.

9. The method for controlling the oxygen-coal ratio in a coal gasification process based on a DCS embedded APC according to claim 1, characterized in that, The permission and calibration setting in the S5 comprises: permission configuration: setting operation permissions according to three-level permissions and a double verification mechanism; sensor calibration: the Rosemount 3051S mass flowmeter, the Siemens SITRAN SP300 pressure transmitter, the K-type thermocouple and the Agilent 7890B online gas chromatograph are calibrated once a month, once a quarter and once a week respectively; the calibration allowable error is less than or equal to 50% of the corresponding device measurement accuracy, and the calibration data is automatically synchronized to the dynamic model; coal quality calibration: the online coal quality analyzer is compared and calibrated with the laboratory data every 2 hours, the deviation is required to be less than or equal to 1 percentage point, and the coal quality parameter library is updated daily.

10. The method for controlling the oxygen-coal ratio in a coal gasification process based on a DCS embedded APC according to claim 1, characterized in that, The deployment and debugging in the S5 comprise: network deployment: allocating a unique communication address for all field hardware devices, and configuring the communication rate and switching mechanism of the L2 layer redundant Ethernet; System commissioning: 72 hours of continuous commissioning, simulating working condition fluctuations to verify the effectiveness of control; during commissioning, the CH4 content is required to be ≤195 ppm, and the standard deviation is ≤20 ppm, and the effective gas production fluctuation is ≤±5%; Fault commissioning: Simulate sensor failure, actuator jamming, and model failure scenarios to verify the response speed of the fault handling mechanism and ensure that the switching time meets the set requirements.

Images