Roller eccentricity compensation method and system based on preloading calibration and dual-mode cooperation

Abstract

The application provides a kind of based on preloading calibration and double mode collaborative roll eccentricity compensation method and system, belong to metallurgical automation control technical field.The method includes the preloading force is applied after zero and position closed loop is kept, and stiffness coupling error is eliminated;Low-speed multi-circle acquisition and with encoder Z pulse as reference phase alignment, superimposed average generates initial eccentric profile;Second system executes harmonic resolution, and output look-up table+harmonic parameter double mode compensation data;Primary system based on rolling force / thickness signal fundamental frequency phase real-time tracking eccentric phase drift, dynamic rotation look-up table index;Through thickness variance / cumulative rolling amount double threshold event trigger adaptive resynchronization.The application solves the problem that domestic non-ferrous rolling mill has no second level direct connection, manual input deviation is large, and phase misalignment caused by work roll slip, measured thickness fluctuation fundamental frequency decay rate is greater than or equal to 70%, calibration-application whole process is less than or equal to 15s, without stopping harmonic correction, significantly improve plate and strip thickness precision and production stability.

Description

Technical Field

[0001] This invention relates to the field of automation control technology in the metallurgical industry, and more specifically, to a method and system for roll eccentricity compensation based on preload calibration and dual-mode coordination. Background Technology

[0002] Roll eccentricity is one of the core factors causing periodic fluctuations in strip thickness, especially in high-speed cold rolling and foil rolling processes, where eccentricity-induced thickness fluctuations can account for 30% to 60% of the total tolerance. Existing eccentricity compensation technologies are mainly divided into three categories: (1) Offline grinding machine data lookup table method: A compensation table is generated using the geometric profile measured by the roll grinding machine, and the encoder angle is used to... The table is used to overlay the values ​​onto the roll gap setting. Representative patents: CN103447412A, CN108856382B.

[0003] (2) Online adaptive filtering method: Based on the rolling force / thickness signal, recursive least squares (RLS), Kalman filtering, or adaptive notch filter is used to identify the eccentric harmonic parameters in real time. Representative patents: CN109201789A, CN111451234B.

[0004] (3) Phase domain signal processing method: The time domain signal is converted into a phase domain signal with the roll phase as a parameter, and the eccentricity component is extracted by comb filtering or linearized phase domain averaging. Representative literature: "Application of Phase Domain Filtering in Roll Eccentricity Compensation", Metallurgical Automation, 2021.

[0005] For typical production line configurations in China's non-ferrous metal processing industry (such as a high proportion of low- and mid-end rolling mills, low penetration rate of secondary systems, and multi-variety, small-batch production mode), the above three types of technologies all have significant defects, specifically manifested as follows: Reliance on grinding machine data: More than 80% of non-ferrous rolling mill grinding machines in China are not connected to the Level 2 (L2) control system. Offline grinding machine data needs to be manually entered into the compensation system, which not only results in amplitude and phase deviations, but also the static geometric profile measured by the grinding machine cannot reflect the stiffness coupling deformation under rolling conditions, leading to a significant decrease in compensation accuracy.

[0006] The phase misalignment problem is prominent: there is a relative slippage of 3% to 8% between the work roll and the support roll during the cold rolling process. Traditional technology uses a fixed encoder angle lookup table method, which cannot dynamically correct the phase deviation caused by slippage. Instead, it may amplify the thickness fluctuation due to the compensation phase misalignment.

[0007] Lack of dynamic adaptability: Offline calibration technology cannot track dynamic changes such as roll thermal expansion, wear, and hydraulic system parameter drift, requiring frequent shutdowns for harmonic correction, which seriously affects the OEE of the production line; Although some online algorithms have dynamic identification capabilities, they are not adaptable to changes in working conditions and are prone to failure in scenarios such as changing specifications or rolls.

[0008] Communication architecture limitations: Traditional L1 (basic automation) and L2 (process automation) systems mostly use CSV file exchange or Modbus polling to transmit data, with communication delays of up to 2 seconds or more, which is far from meeting the real-time requirements of eccentricity compensation (≤100ms), causing the compensation signal to lag behind the actual eccentricity interference.

[0009] Algorithm-execution disconnect: Most algorithm research focuses only on the core compensation logic, without binding it to the PLC execution link. It lacks industrial-grade design such as safety limiting, confidence degradation, and anomaly handling. In practical applications, it is easy to cause equipment damage or product scrap due to signal abnormalities or hardware failures.

[0010] The aforementioned shortcomings have resulted in poor application of existing technologies on domestic non-ferrous rolling mill production lines, necessitating the development of an integrated compensation scheme that combines online calibration, dynamic phase correction, high-speed communication, and industrial-grade safety design. Therefore, this paper proposes a roll eccentricity compensation method and system based on preload calibration and dual-mode coordination. Summary of the Invention

[0011] The purpose of this invention is to address the problems identified in the existing background technology. To achieve the above-mentioned objective, this invention provides the following technical solution: a roll eccentricity compensation method based on preload calibration and dual-mode coordination, comprising the following steps: S1. Zero-point preload stage: The mill support rolls are brought close to the mechanical zero position, and a preset preload force is applied. It also switches to the position closed-loop holding mode to eliminate bearing clearance and elastic deformation of the hydraulic system; S2. Multi-turn phase alignment acquisition: Control the roll to rotate at low speed N times, use the encoder zero-position pulse as the phase reference, acquire displacement signals at high speed, align the phases of each turn through a cross-correlation algorithm, and then superimpose and average them to generate the initial eccentric profile. ; S3. Secondary System Harmonic Decomposition and Dual-Mode Output: The initial eccentric profile is transmitted to the secondary control system to extract the 1st to 3rd order harmonic parameters. and 0.1◦ resolution compensation table The data is then transmitted back to the primary PLC system. S4. Online Phase Correction for Primary System: During the rolling process, the eccentric phase drift is estimated in real time based on the fundamental frequency phase of the rolling force or exit thickness signal. The lookup index of the compensation table is dynamically rotated to output the real-time compensation amount. ; S5. Event-triggered adaptive resynchronization: When the export thickness deviation variance exceeds a preset threshold. Or the cumulative rolling amount reaches the set value At this time, a new round of S1 to S3 processes is triggered to compensate for the cumulative phase error and the effects of thermal expansion.

[0012] As a preferred technical solution of the present invention, the preload force ,in For rated rolling force, coefficient The value range is 0.1 to 0.15.

[0013] As a preferred technical solution of the present invention, in step S2, the number of sampling cycles N=5, and the sampling resolution is... sampling frequency And calculate the coherence coefficient of the 5-lap data. ,like This will trigger a re-sampling alarm.

[0014] As a preferred technical solution of the present invention, in step S4, the phase tracking gain With the linear speed of the rolls Adaptive adjustment, the adjustment formula is: ,in As the reference gain, This is the speed sensitivity coefficient.

[0015] As a preferred technical solution of the present invention, the Snap7 communication library is used between the primary system and the secondary system to achieve state synchronization through the event bit handshake protocol. The original data and the compensation table adopt a block read and write strategy, and the length of a single read and write is ≤2000 bytes.

[0016] As a preferred technical solution of the present invention, the triggering condition of step S5 adopts dual threshold logic: thickness fluctuation variance Or cumulative rolling kilometers And the minimum trigger interval is ≥15 minutes.

[0017] As a preferred technical solution of the present invention, a hard limit is set before the compensation output. ,in And when the confidence level issued by the secondary system When the time is right, it will automatically switch to safe mode (compensation amount = 0).

[0018] As a preferred technical solution of the present invention, the cross-correlation phase alignment in step S2 is implemented using FFT-accelerated convolution, which reduces the computational complexity by Down to ,in This represents the number of sampling points per lap.

[0019] As a preferred technical solution of the present invention, step S4 includes extracting the fundamental frequency phase of the rolling force / thickness signal in real time during the rolling process. And calculate the phase drift: Dynamic rotating lookup index: Output compensation amount: Gain coefficient; Safety limit: Switch to safe mode when the confidence level Conf < 0.3.

[0020] A roll eccentricity compensation system based on preload calibration and dual-mode collaboration includes a sensor acquisition module, a first-level PLC execution module, a communication middleware module, a second-level edge computing module, and a human-machine interaction module. The sensor acquisition module is used to acquire the support roll encoder angle, hydraulic cylinder magnetic scale displacement, and rolling force signals; The first-level PLC execution module is based on the Siemens S7-1500 series, which realizes preload control, low-speed rotation drive, online phase correction algorithm execution and compensation superposition to the AGC feedforward channel; The communication middleware module is developed based on the Snap7 library and adopts a 4-bit event handshake protocol and a block read / write strategy to achieve reliable 72KB-level data transmission between the primary system and the secondary system. The secondary edge computing module is deployed on an industrial control computer, running phase alignment, harmonic extraction, and confidence assessment algorithms to generate... Resolution compensation table and 1st to 3rd order harmonic parameters; The human-computer interaction module is used to display the compensation status, confidence level, and phase drift, and supports manual triggering of calibration and parameter tuning.

[0021] Compared with the prior art, the beneficial effects of the present invention are as follows: This invention utilizes preload and pressure holding to eliminate stiffness coupling errors, employs 5-turn phase alignment and superimposed averaging to suppress random noise, and achieves an eccentric profile repeatability accuracy of ≤±0.005mm. This invention uses L1 online phase correction to track the slippage of the work roll in real time, with a phase tracking error ≤ ±1°, avoiding the need for compensation-induced reverse amplification fluctuations. The L2 calculation time of this invention is ≤2s (3600 points × 5 cycles), and the L1 lookup table + phase correction period is ≤1ms, meeting the requirements of high-speed rolling (≥15m / s). This invention's event-triggered mechanism replaces fixed-cycle calibration, automatically adapting to different materials (aluminum / copper), specifications, and rolling processes, reducing invalid calibrations by more than 30%. This invention features triple protection: confidence assessment, safety limiting, and anomaly degradation, along with automatic retrying after communication timeout, ensuring the safe operation of the rolling mill. This invention does not require modification of the grinding machine or the addition of a high-precision sensor. It is based on the existing magnetic scale / encoder + edge industrial control computer, and the cost of modifying a single production line is ≤80,000 yuan. Attached Figure Description

[0022] Figure 1This is a schematic diagram of the system architecture provided by the present invention. Detailed Implementation

[0023] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are specific implementations of the present invention and are not limited to all embodiments.

[0024] Therefore, the following detailed description of embodiments of the present invention is not intended to limit the scope of the claimed invention, but merely illustrates some embodiments of the invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0025] It should be noted that, in the absence of conflict, the embodiments and features and technical solutions in the embodiments of the present invention can be combined with each other. It should be noted that similar reference numerals and letters in the following figures indicate similar items. Therefore, once an item is defined in one figure, it does not need to be further defined and explained in subsequent figures.

[0026] Example 1: A roll eccentricity compensation method based on preload calibration and dual-mode collaboration. This invention proposes a closed-loop compensation architecture of "preload calibration to eliminate stiffness coupling error + L2 offline LUT generation + L1 online phase correction + event-triggered adaptive resynchronization", as follows: Figure 1 As shown. The method of the present invention includes the following core steps: S1. Zero-based preload stage S1.1 The support roller is brought close to the mechanical zero position to establish an absolute angle reference; S1.2 Apply preload Switch to position closed-loop hold mode; S1.3 Waiting for the hydraulic system to stabilize This eliminates bearing clearance and cylinder elastic deformation. S2. Multi-turn phase alignment acquisition.

[0027] S2.1 controls the rolls to rotate at low speed N revolutions. ; S2.2 uses the encoder Z-pulse as the phase reference to acquire the magnetic scale displacement signal at high speed. ; S2.3 performs cross-correlation phase alignment on N-cycle data:

[0028] S2.4 Phase alignment followed by averaging generates an initial eccentric profile. ; S2.5 Calculation of coherence coefficient of 5-cycle data ,like Triggered resampling alarm. S3.L2 harmonic decomposition and dual-mode output.

[0029] S3.1 will Transmitted to L2 edge computing nodes; S3.2 Perform FFT to extract 1st to 3rd order harmonic parameters:

[0030] S3.3 generation Resolution Compensation Table And calculate the data confidence level:

[0031] S3.4 via Snap7 back S4.L1 Online Phase Correction

[0032] During the S4.1 rolling process, the fundamental frequency phase of the rolling force / thickness signal is extracted in real time.

[0033] S4.2 Calculate the phase drift: ; S4.3 Dynamic Rotation Lookup Index: ; S4.4 Output Compensation Amount: This is the gain coefficient; S4.5 Safety Limit: When the confidence level (Conf) < 0.3, switch to safe mode. S5. Event-triggered adaptive resynchronization.

[0034] S5.1 Real-time monitoring of outlet thickness deviation variance With cumulative rolling volume ; S5.2 Triggering conditions (execution occurs if any one of them is met): ;

[0035] S5.3 Utilize the belt threading gap or the roll changing preparation period to automatically execute the S1~S3 process; S5.4 updates LUT and harmonic parameters to achieve adaptive calibration without downtime.

[0036] System Composition The system of this invention includes the following modules: M1: Sensor acquisition module: support roll encoder (with Z pulse), hydraulic cylinder sealing magnetic scale, rolling force sensor, sampling rate ≥1kHz; M2: L1 PLC execution module: based on Siemens S7-1500, realizing preload control, low-speed rotation, online phase correction, and compensation amount superimposed to the AGC feedforward channel; M3: Communication Middleware Module: Developed based on the Snap7 library, it adopts an event bit handshake protocol (4-bit state machine) + block read / write strategy (single ≤2000 bytes) to ensure reliable transmission of 72KB raw data; M4: L2 edge computing module: Deployed on an industrial control computer (i5 / 8G / Win10 IoT), running C++ / Rust algorithm programs to perform phase alignment, harmonic extraction, and confidence assessment; M5: Human-Computer Interaction Module: The HMI displays the compensation status, confidence level, and phase drift, and supports manual triggering of calibration and parameter tuning.

[0037] Example 2: The components and parameters of this example are as follows: PLC: Uses Siemens S7-1516-3PN / DP model, with a scan cycle of up to 1ms, supports PUT / GET access methods, and has efficient data processing and interaction capabilities.

[0038] Encoder: Heidenhain ERN1387 model, 2048 lines per revolution, with built-in Z pulse, IP67 protection rating, can accurately provide position information under complex working conditions.

[0039] Magnetic ruler: Equipped with Novotechnik TRS-0200 model, with a measuring range of 200mm, a resolution of up to 0.001mm, and an IP67 protection rating, suitable for high-precision length measurement scenarios.

[0040] Rolling force sensor: The Kistler 9321B model is used, with a range of up to 2000kN and a frequency response range of 0 to 500Hz, which can accurately capture the force changes during the rolling process.

[0041] Edge industrial control computer: The Advantech UNO-2484G model is selected, equipped with an i5-8365UE processor, 8GB of memory and a 256GB solid-state drive, and pre-installed with the Win10 IoT system to meet the data processing and storage needs of the edge.

[0042] Communication scheme: Snap7 v1.4.2 combined with Profinet protocol is used to realize data interaction through event bit handshake mechanism, support block read and write mode, and set 5-second timeout threshold to ensure communication stability and reliability.

[0043] DB Block Communication Protocol (Key Part) Listing 1: DB100 Handshake Signal Area Definition (ST Language) / / DB_EccComm (DB100) - Handshake signal area (Byte 0-15) STRUCT L1_DataReady:Bool; / / %DB100.DBX0.0 - L1 set: Sampling data is ready L1_AckReceived:Bool; / / %DB100.DBX0.1 - L2 set: Read confirmed L2_CalcDone:Bool; / / %DB100.DBX0.2 - L2 set: Calculation complete L1_TableTaken: Bool; / / %DB100.DBX0.3 - L1 set: Result taken CommError: Word; / / %DB100.DBD2 - Error codes: 0=OK, 1=ReadFail, 2=Timeout ProcessStatus:Int; / / %DB100.DBD4-0=Idle,1=Processing,2=Done,3=Error / / ...Reserved bytes END_STRUCT

[0044] Listing 2: Core Logic of Phase Alignment + Harmonic Extraction / / Cross-correlation phase alignment (optimization: FFT to accelerate convolution) fn phase_align(ref_sig: &[f32], target: &[f32]) -> usize { let n = ref_sig.len(); / / Calculate the cross-correlation frequency domain expression using rustfft let mut planner=FftPlanner:: <f32>::new(); let fft = planner.plan_fft_forward(n); let ref_fft = fft_process(ref_sig, &fft); let tgt_fft = fft_process(target, &fft); / / Cross-correlation = IFFT(FFT(ref) * conj(FFT(tgt))) let cross:Vec <complex32>=ref_fft.iter() .zip(tgt_fft.iter()) .map(l(a,b)la*b.conj()) .collect(); let corr=ifft_process(&cross,&fft); / / Finding the peak position is the phase shift. corr.iter() .enumerate() .max_by(la,bl a.1.re.partial_cmp(&b.1.re).unwrap()) .map(l(idx,_)lidx) .unwrap_or(0) / / Harmonic parameter extraction (1st to 3rd order) fn extract_harmonics(signal:&[f32], orders:&[usize])->Vec<(f32, f32)>{ let n = signal.len(); let fft=planner.plan_fft_forward(n); let spectrum=fft_process(signal,&fft); orders.iter().map(l&kl{ if k>=n / 2{return(o.0, 0.o);} let c = Spectrum[k]; let amp = 2.0 * C.norm() / nasf32; / / Amplitude normalization let phase = c.arg() * 180.0 / PI; / / Phase rotation angle (amp, phase) }).collect()

[0045] Listing 3: Core Logic of the FB_PhaseTracker Function Block FUNCTION_BLOCK FB_PhaseTracker VAR_INPUT ForceSignal:Real; / / Real-time value of rolling force RollSpeed:Real; / / Roll linear speed (m / s) RollDiameter:Real; / / Support roller diameter (mm) Harm1_Phase_Init:Real; / / First-order initial phase (°) issued by L2. Enable:Bool; END_VAR VAR_OUTPUT PhaseDrift: Real; / / Estimated phase drift (°) Valid:Bool; END_VAR VAR EccFreq ForceBand EstPhase Gain :Real;:Real;:Real; :Real / / Eccentric theoretical frequency (Hz) / / Signal after bandpass filtering / / Real-time estimated phase:=0.02; / / Tracking gain END_VAR / / 1. Calculate the eccentric fundamental frequency EccFreq:=RollSpeed*1000.0 / (PI*RollDiameter); / / 2. Bandpass filtering to extract off-center frequency bands (±20% bandwidth) ForceBand := BandPass_Butterworth(ForceSignal, EccFreq * 0.8,EccFreq)*1.2); / / 3. Zero crossing detection + phase calculation IF Enable AND ABS(ForceBand)>0.01 THEN EstPhase:=PhaseFromZeroCrossing(ForceBand,EccFreq); / / 4. First-order inertial filter updates drift amount PhaseDrift:=PhaseDrift+Gain*(EstPhase-Harm1_Phase_Init- PhaseDrift); / / 5. Limiting and Validity Judgment Valid:=ABS(PhaseDrift)<=10.0; / / Valid within ±10° IF NOT Valid THEN PhaseDrift:=0.0;END_IF;END_IF END_FUNCTION_BLOCK

[0046] Triggering condition: thickness deviation variance

[0047] Recommended threshold:

[0048] Adjustment basis: The thickness standard deviation during normal rolling is calculated using a 1.5x coefficient corresponding to the 3σ principle.

[0049] Triggering condition: Cumulative rolling mileage

[0050] Recommended thresholds: 80km for aluminum, 50km for copper Adjustment basis: The threshold value is determined based on the statistical analysis of measured data on roll wear rate.

[0051] Triggering condition: Minimum trigger interval Recommended threshold: no less than 15 minutes Adjustment criteria: Limit the shortest interval to avoid frequent calibrations disrupting the normal production rhythm.

[0052] Triggering condition: calibrated execution window Recommended thresholds: Belt threading stage, roll change preparation period, low-speed bite section Adjustment basis: The calibration operation is carried out using the inherent idle gap of the production process to achieve zero downtime operation.

[0053] Example 3: A method for roll eccentricity compensation based on preload calibration and dual-mode coordination, comprising the following steps: S1. Zero-position preload stage: The mill support roll is brought close to the mechanical zero position, and a preset preload force is applied. It also switches to the position closed-loop holding mode to eliminate bearing clearance and elastic deformation of the hydraulic system; S2. Multi-turn phase alignment acquisition: Control the roll to rotate at low speed N times, use the encoder zero-position pulse as the phase reference, acquire displacement signals at high speed, align the phases of each turn through a cross-correlation algorithm, and then superimpose and average them to generate the initial eccentric profile. ; S3. Second-level system harmonic decomposition and dual-mode output: The initial eccentric profile is transmitted to the second-level control system to extract the 1st to 3rd order harmonic parameters. and 0.1◦ resolution compensation table The data is then transmitted back to the primary PLC system. S4. Online Phase Correction for Primary System: During the rolling process, the eccentric phase drift is estimated in real time based on the fundamental frequency phase of the rolling force or exit thickness signal. The lookup index of the dynamic rotation compensation table outputs the real-time compensation amount. ; S5. Event-triggered adaptive resynchronization: When the export thickness deviation variance exceeds a preset threshold. Or the cumulative rolling amount reaches the set value At this time, a new round of S1 to S3 processes is triggered to compensate for the cumulative phase error and the effects of thermal expansion.

[0054] Preload ,in For rated rolling force, coefficient The value range is 0.1 to 0.15.

[0055] In step S2, the number of sampling cycles N=5, and the sampling resolution is... sampling frequency And calculate the coherence coefficient of the 5-lap data. ,like This will trigger a re-sampling alarm.

[0056] In step S4, the phase tracking gain With the linear speed of the rolls Adaptive adjustment, the adjustment formula is: ,in As the reference gain, This is the speed sensitivity coefficient.

[0057] The Level 1 and Level 2 systems use the Snap7 communication library and achieve state synchronization through an event bit handshake protocol. The raw data and compensation table adopt a block read / write strategy, with a single read / write length of ≤2000 bytes.

[0058] The triggering condition for step S5 uses a dual-threshold logic: thickness fluctuation variance. Or cumulative rolling kilometers And the minimum trigger interval is ≥15 minutes.

[0059] Set hard limit before compensation output ,in And when the confidence level issued by the secondary system When the time is right, it will automatically switch to safe mode (compensation amount = 0).

[0060] The cross-correlation phase alignment in step S2 is implemented using FFT-accelerated convolution, reducing the computational complexity from... Down to ,in This represents the number of sampling points per lap.

[0061] Step S4 includes extracting the fundamental frequency phase of the rolling force / thickness signal in real time during the rolling process. And calculate the phase drift: Dynamic rotating lookup index: Output compensation amount: Gain coefficient; Safety limit: Switch to safe mode when the confidence level Conf < 0.3.

[0062] A roll eccentricity compensation system based on preload calibration and dual-mode collaboration includes a sensor acquisition module, a first-level PLC execution module, a communication middleware module, a second-level edge computing module, and a human-machine interaction module. The sensor acquisition module is used to acquire the support roll encoder angle, hydraulic cylinder magnetic scale displacement, and rolling force signals; The first-level PLC execution module is based on the Siemens S7-1500 series, which realizes preload control, low-speed rotation drive, online phase correction algorithm execution and compensation superposition to the AGC feedforward channel; The communication middleware module is developed based on the Snap7 library and adopts a 4-bit event handshake protocol and a block read / write strategy to achieve reliable 72KB-level data transmission between the primary system and the secondary system. The secondary edge computing module is deployed on an industrial control computer, running phase alignment, harmonic extraction, and confidence assessment algorithms to generate... Resolution compensation table and 1st to 3rd order harmonic parameters; The human-computer interaction module is used to display the compensation status, confidence level, and phase drift, and supports manual triggering of calibration and parameter tuning.

[0063] The above embodiments are only used to illustrate the present invention and are not intended to limit the technical solutions described herein. Although the present invention has been described in detail with reference to the above embodiments, the present invention is not limited to the specific embodiments described above. Therefore, any modifications or equivalent substitutions to the present invention, as well as all technical solutions and improvements that do not depart from the spirit and scope of the invention, are covered within the scope of the claims of the present invention.

Claims

1. A method for compensating roll eccentricity based on preload calibration and dual-mode coordination, characterized in that, Includes the following steps: S1. Zero-point preload stage: The mill support rolls are brought close to the mechanical zero position, and a preset preload force is applied. It also switches to the position closed-loop holding mode to eliminate bearing clearance and elastic deformation of the hydraulic system; S2. Multi-turn phase alignment acquisition: Control the roll to rotate at low speed N times, use the encoder zero-position pulse as the phase reference, acquire displacement signals at high speed, align the phases of each turn through a cross-correlation algorithm, and then superimpose and average them to generate the initial eccentric profile. ; S3. Secondary System Harmonic Decomposition and Dual-Mode Output: The initial eccentric profile is transmitted to the secondary control system to extract the 1st to 3rd order harmonic parameters. and 0.1◦ resolution compensation table The data is then transmitted back to the primary PLC system. S4. Online Phase Correction for Primary System: During the rolling process, the eccentric phase drift is estimated in real time based on the fundamental frequency phase of the rolling force or exit thickness signal. The lookup index of the compensation table is dynamically rotated to output the real-time compensation amount. ; S5. Event-triggered adaptive resynchronization: When the export thickness deviation variance exceeds a preset threshold. Or the cumulative rolling amount reaches the set value At this time, a new round of S1 to S3 processes is triggered to compensate for the cumulative phase error and the effects of thermal expansion.

2. The roll eccentricity compensation method based on preload calibration and dual-mode coordination according to claim 1, characterized in that: The preload ,in For rated rolling force, coefficient The value range is 0.1 to 0.

15.

3. The roll eccentricity compensation method based on preload calibration and dual-mode coordination according to claim 1, characterized in that: In step S2, the number of sampling cycles N=5, and the sampling resolution is... sampling frequency And calculate the coherence coefficient of the 5-lap data. ,like This will trigger a re-sampling alarm.

4. The roll eccentricity compensation method based on preload calibration and dual-mode coordination according to claim 1, characterized in that: In step S4, the phase tracking gain With the linear speed of the rolls Adaptive adjustment, the adjustment formula is: ,in As the reference gain, This is the speed sensitivity coefficient.

5. The roll eccentricity compensation method based on preload calibration and dual-mode coordination according to claim 1, characterized in that: The Level 1 and Level 2 systems use the Snap7 communication library and achieve state synchronization through an event bit handshake protocol. The raw data and compensation table adopt a block read / write strategy, with a single read / write length of ≤2000 bytes.

6. The roll eccentricity compensation method based on preload calibration and dual-mode coordination according to claim 1, characterized in that: The triggering condition for step S5 uses a dual-threshold logic: thickness fluctuation variance. Or cumulative rolling kilometers And the minimum trigger interval is ≥15 minutes.

7. The roll eccentricity compensation method based on preload calibration and dual-mode coordination according to claim 1, characterized in that: Set hard limit before compensation output ,in And when the confidence level issued by the secondary system When the time is right, it will automatically switch to safe mode (compensation amount = 0).

8. The roll eccentricity compensation method based on preload calibration and dual-mode coordination according to claim 1, characterized in that: The cross-correlation phase alignment in step S2 is implemented using FFT-accelerated convolution, reducing the computational complexity from... Down to ,in This represents the number of sampling points per lap.

9. The roll eccentricity compensation method based on preload calibration and dual-mode coordination according to claim 1, characterized in that: Step S4 includes extracting the fundamental frequency phase of the rolling force / thickness signal in real time during the rolling process. And calculate the phase drift: Dynamic rotating lookup index: Output compensation amount: Gain coefficient; Safety limit: Switch to safe mode when the confidence level Conf < 0.

3.

10. A roll eccentricity compensation system for implementing the method of any one of claims 1 to 9, characterized in that, It includes a sensor acquisition module, a primary PLC execution module, a communication middleware module, a secondary edge computing module, and a human-machine interaction module; The sensor acquisition module is used to acquire the support roll encoder angle, hydraulic cylinder magnetic scale displacement, and rolling force signals; The first-level PLC execution module is based on the Siemens S7-1500 series, which realizes preload control, low-speed rotation drive, online phase correction algorithm execution and compensation superposition to the AGC feedforward channel; The communication middleware module is developed based on the Snap7 library and adopts a 4-bit event handshake protocol and a block read / write strategy to achieve reliable 72KB-level data transmission between the primary system and the secondary system. The secondary edge computing module is deployed on an industrial control computer, running phase alignment, harmonic extraction, and confidence assessment algorithms to generate... Resolution compensation table and 1st to 3rd order harmonic parameters; The human-computer interaction module is used to display the compensation status, confidence level, and phase drift, and supports manual triggering of calibration and parameter tuning.

Images