A hot-rolling ultra-fast cooling manifold flow control method based on online self-learning

Abstract

This invention proposes a flow control method for hot-rolled ultrafast cold manifolds based on online self-learning, implemented entirely on the existing L1 system. This method first queries the flow-valve position correction table to obtain the initial valve position; if the position is not found, a calibration curve is used. Actual flow rate is collected in real time; when it is below 5 m³ / h, PID control is disabled and an alarm is triggered. When the deviation is within the target flow rate ±5% dead zone, the PID remains silent to reduce valve action; if it exceeds this range, incremental PID closed-loop control is used. The valve position is locked after the deviation stabilizes for 2 seconds. After operator confirmation of normal operation, the current flow rate and valve position are stored in the correction table, overwriting the old value. This method significantly reduces valve wear and flow oscillation through dead zone silencing and anomaly isolation. The self-learning mechanism under manual supervision continuously optimizes the initial valve position with each production batch, constantly improving the system's silencing rate, while ensuring the purity of the correction table data and the long-term stability of the system.

Description

Technical Field

[0001] This invention belongs to the field of post-rolling cooling control technology, specifically relating to a hot-rolled ultra-fast cooling manifold flow control method based on online self-learning. Background Technology

[0002] The hot-rolled ultra-fast cooling system consists of dozens of branch manifolds. Each manifold requires precise control of the cooling water flow rate to achieve the required strip cooling rate and coiling temperature. The existing control scheme involves calibrating a separate offline "flow-valve position" curve for each manifold. During actual control, the initial valve position is obtained by looking up the respective curve based on the set flow rate, followed by closed-loop fine-tuning using a PID controller. The process is as follows: model sends out the set flow rate, the initial valve position is obtained by looking up the calibration curve of the corresponding manifold, valve execution, flow meter feedback, and PID adjustment.

[0003] This solution has the following problems: First, there are differences between offline calibration conditions and online conditions in terms of water supply pressure, valve nonlinear response, and pipeline resistance characteristics. Furthermore, factors such as water imbalance caused by switching between different manifolds, flow meter detection deviation, and valve response delay cause deviations between the calibrated flow rate and the actual flow rate. The PID controller continuously adjusts to eliminate these deviations, leading to frequent valve actions that exacerbate wear. The repeated flow oscillations also affect the uniformity of cooling.

[0004] Secondly, in order to eliminate the above-mentioned deviations and fluctuations, PID requires frequent adjustments. The system is always in a dynamic adjustment, the valves operate frequently, which aggravates wear and shortens lifespan. At the same time, the repeated oscillation of flow also affects the uniformity of cooling.

[0005] In addition, existing Chinese patent CN118788771A, "Cooling device and method for forming a stable temperature gradient in large-scale drum section differential temperature rolling," includes: S3. After rolling begins, the control unit continues to receive the temperature value detected by the temperature acquisition device and adjusts the opening of the flow regulating valve according to the temperature value to maintain a stable temperature gradient in the cylinder section.

[0006] The instruction manual states that "a flow meter is installed on the inlet pipe, and the flow meter is also connected to the control unit; the control unit can fine-tune the opening of the flow regulating valve according to the detected value of the flow meter." This solution corresponds to the control logic of the aforementioned prior art, and the problems of frequent valve operation and flow oscillation still exist. Summary of the Invention

[0007] This invention provides a hot-rolled ultrafast cold manifold flow control method based on online self-learning to solve at least one of the above-mentioned technical problems.

[0008] The technical solution adopted in this invention is as follows: A method for controlling the flow rate of hot-rolled ultrafast cooling manifolds based on online self-learning includes the following steps: S101: Receive the current set flow rate value of the manifold, query the flow-valve position correction table of the manifold, if a match is found, use the corrected valve position recorded in the match record as the initial valve position output, if no match is found, calculate the initial valve position output based on the calibration curve of the manifold. S102: After outputting the initial valve position, the actual flow rate is collected in real time; the minimum threshold of the flow meter is preset, and the flow meter reading is judged to be lower than the preset minimum threshold. If it is, it is judged to be abnormal, PID adjustment is prohibited and an alarm is triggered; if not, proceed to step S103. S103: Preset dead zone range, calculate the deviation between the actual flow rate and the set flow rate. If the deviation is within the preset dead zone range, the PID remains silent and returns to step S102; if the deviation exceeds the preset dead zone range, proceed to step S104. S104: PID intervention performs closed-loop regulation until the deviation returns to the dead zone range; S105: Timing starts after the deviation returns to the dead zone range. When the deviation remains stable within the dead zone range for a preset duration, the PID stops adjusting and the current valve position remains fixed. S106: After the operator confirms that the working condition is normal, the system triggers a confirmation signal and stores the current set flow rate value and the fixed valve position value in the flow-valve position correction table of the manifold. If there is an old record for the flow rate point, it will be overwritten directly.

[0009] Furthermore, the set flow rate value mentioned in step S101 is issued by the L2 system or manually entered by the operator through the operation interface.

[0010] For steel grades that have already been produced, the L2 system stores the corresponding set flow rate value, so in S101, the L2 system issues the set flow rate value; for new steel grades, the operator manually inputs it.

[0011] Furthermore, the preset minimum threshold mentioned in step S102 is 5 m³ / h. When the actual flow rate is lower than 5 m³ / h, it is determined that the valve is not open normally or the flow meter is faulty.

[0012] By setting a minimum flow threshold of 5 m³ / h, the system automatically disables PID regulation and triggers an alarm when the actual flow rate in the manifold falls below this value. This effectively prevents valve malfunctions caused by abnormal operating conditions such as valves not opening properly, flowmeter zero drift, or line interference, and eliminates water hammer impacts and equipment safety hazards caused by the continuous accumulation of valve position commands. At the same time, this forced isolation mechanism ensures that the system only performs control and data recording under real and effective media flow conditions, guaranteeing the data purity of the flow-valve position correction table from the source and preventing abnormal data from contaminating the self-learning database. This provides a solid safety and data guarantee for the long-term stable iteration of the entire control strategy.

[0013] Furthermore, the preset dead zone range mentioned in step S103 is ±5% of the target flow rate.

[0014] By setting a preset dead zone of ±5% of the target flow rate, the PID controller remains silent and does not adjust when the actual flow rate deviation is within this range. This effectively avoids high-frequency reciprocating valve position movements caused by minor fluctuations in the flow meter signal or process disturbances, significantly reducing mechanical wear of the regulating valve and energy consumption of the actuator. It also eliminates continuous oscillations in flow rate near the target value, resulting in a more stable and uniform cooling water output, which is beneficial to the stability of the strip cooling effect. Furthermore, PID closed-loop regulation is only triggered when the deviation exceeds the dead zone, making the system's control intervention more targeted and avoiding ineffective consumption of control resources within acceptable error ranges. This dead zone setting, combined with a minimum flow rate threshold of 5 m³ / h, forms a protective mechanism that ensures safety under extreme abnormal conditions and stability under normal operating conditions. This improves the reliability and service life of the equipment while guaranteeing the accuracy and stability of the ultra-fast cooling process control.

[0015] Furthermore, in step S104, the PLC calculates the opening correction amount according to the deviation between the actual flow rate and the set flow rate using the PID control algorithm, and gradually adjusts the valve opening until the deviation enters the preset dead zone range. The PID control algorithm is an incremental PID control algorithm.

[0016] An incremental PID control algorithm is used for closed-loop regulation. Each adjustment outputs only the correction increment relative to the current valve position. Since the algorithm does not require initial valve position value, the output is always based on the actual opening degree of the previous moment plus the fine adjustment, which can effectively eliminate the impact of water pressure fluctuations on the uniformity of strip cooling. This improves the fault tolerance and safety of the control system under rolling mill conditions. At the same time, an incremental limit can be used to impose a hard constraint on the single adjustment step size, ensuring rapid elimination of deviations while achieving smooth, controllable, and highly reliable valve action, providing core algorithmic support for the precise and stable control of ultra-fast cooling processes.

[0017] Furthermore, the preset duration in step S105 is 2 seconds.

[0018] Setting the stabilization timer to 2 seconds is intended to ensure that the deviation remains stable within the dead zone before locking the valve position and adjusting the flow-valve position correction table. This effectively filters out instantaneous fluctuations in the flow signal and prevents invalid data from being overwritten due to brief noise falling into the dead zone.

[0019] Furthermore, in step S106, the operator confirms that the main cooling water pressure is normal, with the normal range set at 0.82 MPa-0.87 MPa.

[0020] The manual verification of the main cooling water pressure aims to enhance human-machine collaboration. In the complex production process, intelligent control equipment has limited scope and cannot address other production issues. The manual verification process allows for monitoring of equipment status, assessment of the safety of the current operating conditions and the stability of equipment operation based on production experience, and timely implementation of corresponding measures.

[0021] Furthermore, the flow-valve position correction table is established independently for each manifold, and only one correction record is stored for each set flow value. The newly confirmed valve position value directly overwrites the old value.

[0022] By using a single-pipe, single-table data recording method, interference between data from multiple collection pipes is avoided, ensuring that each collection pipe can independently and accurately compensate. At the same time, the size of the data table is controlled to prevent redundancy and ensure that the system is always updated with the latest steady-state data, achieving simple and efficient self-learning. Beneficial effects

[0023] This invention provides a hot-rolled ultrafast cold manifold flow control method based on online self-learning, which is implemented entirely based on the existing L1 automation system without the need for additional hardware.

[0024] In terms of control strategy, PID only intervenes to adjust when the deviation exceeds the dead zone. When the flow is normal, the system is completely silent and the valve position remains fixed, which greatly reduces the number of valve actions and eliminates flow oscillation, ensuring uniform cooling.

[0025] Regarding the accuracy of the initial valve position, a self-learning mechanism combining manual verification and the flow-valve position correction table is used to record the accurate valve position after each successful adjustment. This information is then directly recalled for subsequent adjustments with the same flow rate setting, bypassing the offline calibration curve value. As production batches increase, the correction table can be continuously optimized, ensuring precise adjustment of the initial valve position and a sustained improvement in the system's quiet operation rate.

[0026] Regarding data security, this invention requires operators to confirm that the operating conditions are normal before storing the valve position in the flow-valve position correction table, and sets up flow meter malfunction protection to prevent erroneous learning caused by external factors such as flow meter failure or abnormal main pipe pressure, thus ensuring the reliability and validity of the correction table data. Attached Figure Description

[0027] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings: Figure 1 This is the control flowchart of the present invention.

[0028] Figure Labels S101: Receive the set flow rate, query the correction table and calibration curve, and determine the initial valve position.

[0029] S102: Output the initial valve position, collect the flow rate, and determine if the flow meter is malfunctioning.

[0030] S103: Calculate the deviation and determine whether it is within the dead zone.

[0031] S104: PID adjustment until the deviation returns to the dead zone.

[0032] S105: Stable timing, valve position locking.

[0033] S106: Manual confirmation, save to correction table. Detailed Implementation

[0034] To more clearly illustrate the overall concept of the present invention, a detailed description will be provided below with reference to the accompanying drawings and examples.

[0035] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and therefore the scope of protection of the invention is not limited to the specific embodiments disclosed below.

[0036] Furthermore, in the description of this invention, it should be understood that the terms "top," "bottom," "inner," "outer," "axial," "radial," "circumferential," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are only for the convenience of describing this invention and simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on this invention.

[0037] In this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," "linking," and "fixing," etc., should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral part; they can refer to a mechanical connection, an electrical connection, or a communication connection; they can refer to a direct connection or an indirect connection through an intermediate medium; they can refer to the internal communication of two components or the interaction between two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0038] In this invention, unless otherwise expressly specified and limited, the first feature "on" or "below" the second feature may be in direct contact with the first and second features, or indirect contact through an intermediate medium. In the description of this specification, references to terms such as "implementation," "example," "aspect," "specific example," or "specific example" indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0039] This embodiment takes a set of branch manifolds of an ultra-fast cooling system in a hot rolling production line as an example. Its hardware foundation is the existing L1 automated control system, including a PLC controller, an HMI human-machine interface, a manifold pneumatic shut-off valve, a pneumatic regulating valve, an electromagnetic flow meter, and a water supply main pressure gauge.

[0040] System prerequisites: Each manifold has been calibrated offline to obtain its own "flow-valve position" curve, which is stored in the PLC. An initially empty "flow-valve position correction table" is independently created in the PLC memory for each manifold, and each manifold correction table retains only one most recently corrected "flow-valve position" record.

[0041] HMI interface settings: Deviation dead zone range is ±5% of the set flow rate, stability judgment time is 2 seconds, minimum threshold for faulty flow meter is 5m³ / h, and a virtual confirmation button for manual HMI of a single manifold is set to trigger the correction table storage operation.

[0042] Example 1 The X70 pipeline steel is being produced, and the control system (L2) issues a set flow rate of 119 m³ / h to the manifold.

[0043] S101: The PLC queries the correction table for this manifold, but there is no record for "119" in the current table, so it is a no-match. The PLC calls the offline calibration curve for this manifold and finds that the corresponding initial valve position is 30%.

[0044] S102: The PLC outputs 30% as the initial opening command to the regulating valve, and the valve moves to the 30% opening. The PLC collects feedback from the flow meter and measures the actual flow rate as 96 m³ / h, which is greater than the minimum fault threshold of 5 m³ / h. The flow meter is then determined to be normal, and the process proceeds to S103.

[0045] S103: Calculation deviation = 19%, exceeding the ±5% dead zone, proceed to S104.

[0046] S104: The PLC initiates PID closed-loop regulation, continuously outputting regulation commands to drive the valve to gradually open. When the valve opening increases to 34%, the measured flow rate recovers to 116 m³ / h, with a deviation of approximately -2.5%, returning to the dead zone range.

[0047] S105: The system starts timing after detecting that the measured flow rate is within the target flow rate deviation range. If the flow rate remains within the deviation range for 2 consecutive seconds, the PID control stops adjusting, and the valve position is locked at 34%.

[0048] S106: After the operator confirms through the HMI that the main pipe pressure is about 0.85MPa, click the confirm learning button; the PLC will store the correspondence between the set flow rate of 119m³ / h and the locked valve position of 34% in the flow-valve position correction table of the manifold, overwriting the old record (if any).

[0049] Example 2 The same X70 pipeline steel of the same specification was produced again, and the same manifold received the set flow rate of 119 m³ / h again.

[0050] S101: The PLC queries the correction table and finds the record "flow rate 119m³ / h corresponds to valve opening of 34%", directly outputting 34% as the initial valve position, skipping the offline calibration curve.

[0051] S102: Valve actuated to 34% opening. Flow meter reading 117 m³ / h, exceeding the minimum fault threshold by 5 m³ / h.

[0052] S103: The deviation is approximately -1.6%. Within the ±5% dead zone, the PID remains silent, and the valve position remains unchanged.

[0053] The entire process required no adjustments, achieving long-term stability after a single learning phase.

[0054] Example 3 During the commissioning of the new steel grade process, no values ​​were issued from the secondary system. The operator manually entered the set flow rate of 80 m³ / h on the HMI. The system executed the same S101 to S107 procedures, and the data was saved to the correction table after successful learning.

[0055] Example 4 During production, the flow meter reading remained at 0 m³ / h (below the preset minimum threshold of 5 m³ / h). When the system executed S102, an anomaly was detected, PID regulation was immediately disabled, and an "alarm red value" was output on the HMI. On-site personnel then promptly repaired the issue before the system was put back into operation according to the established procedures.

[0056] For any parts not mentioned in this invention, existing technologies can be used or referenced.

[0057] The various embodiments in this specification are described in a progressive manner. The same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on describing the differences from other embodiments.

[0058] The above description is merely an embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of the present invention should be included within the scope of the claims of the present invention.

Claims

1. A flow control method for hot-rolled ultrafast cooling manifolds based on online self-learning, characterized in that, Includes the following steps: S101: Receive the current set flow rate value of the manifold, query the flow-valve position correction table of the manifold, if a match is found, use the corrected valve position recorded in the match record as the initial valve position output, if no match is found, calculate the initial valve position output based on the calibration curve of the manifold. S102: After outputting the initial valve position, the actual flow rate is collected in real time; the minimum threshold of the flow meter is preset, and the flow meter reading is judged to be lower than the preset minimum threshold. If it is, it is judged to be abnormal, PID adjustment is prohibited and an alarm is triggered; if not, proceed to step S103. S103: Preset dead zone range, calculate the deviation between the actual flow rate and the set flow rate. If the deviation is within the preset dead zone range, the PID remains silent and returns to step S102; if the deviation exceeds the preset dead zone range, proceed to step S104. S104: PID intervention performs closed-loop regulation until the deviation returns to the dead zone range; S105: Timing starts after the deviation returns to the dead zone range. When the deviation remains stable within the dead zone range for a preset duration, the PID stops adjusting and the current valve position remains fixed. S106: After the operator confirms that the working condition is normal, the system triggers a confirmation signal and stores the current set flow rate value and the fixed valve position value in the flow-valve position correction table of the manifold. If there is an old record for the flow rate point, it will be overwritten directly.

2. The method according to claim 1, characterized in that, The set flow rate value mentioned in step S101 is issued by the L2 system or manually entered by the operator through the operation interface.

3. The method according to claim 2, characterized in that, The preset minimum threshold mentioned in step S102 is 5 m³ / h. When the actual flow rate is lower than 5 m³ / h, it is determined that the valve is not open normally or the flow meter is faulty.

4. The method according to claim 3, characterized in that, The preset dead zone range mentioned in step S103 is ±5% of the target flow rate.

5. The method according to claim 4, characterized in that, In step S104, the PLC calculates the opening correction amount according to the deviation between the actual flow rate and the set flow rate using the PID control algorithm, and gradually adjusts the valve opening until the deviation enters the preset dead zone range. The PID control algorithm is an incremental PID control algorithm.

6. The method according to claim 5, characterized in that, The preset duration mentioned in step S105 is 2 seconds.

7. The method according to claim 6, characterized in that, In step S106, the operator confirms that the main cooling water pressure is normal, with the normal range set at 0.82 MPa-0.87 MPa.

8. The method according to claim 7, characterized in that, The flow-valve position correction table is established independently for each manifold. Only one correction record is stored for each set flow value, and the newly confirmed valve position value directly overwrites the old value.

Images