An industrial real-time control method and process control system for lumber processing equipment
By integrating heterogeneous signal processing of the spindle servo motor q-axis current and acoustic emission signal, and combining digital twin model with disturbance observer, feedforward-feedback coordinated control of sawing equipment is realized, solving the problem of lag response to sudden changes in cutting force in existing technologies, and improving processing quality and equipment safety.
Patent Information
- Application Number
- CN202611457799.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2026-09-18
- Publication Date
- 2026-11-17
AI Technical Summary
Existing industrial real-time control software and advanced process control solutions for sawing equipment rely on the hysteresis feedback and post-event monitoring mechanism of mechanical sensors. They cannot complete the perception and active suppression within the microsecond to millisecond time window when a sudden change in cutting force occurs, resulting in a decline in processing quality, increased equipment wear and tear, and forced production line to slow down and stop.
By integrating the q-axis current signal of the spindle servo motor and the acoustic emission sensor signal, and combining a digital twin model with a disturbance observer, feedforward-feedback coordinated control is achieved. This allows for real-time estimation and active suppression of sudden changes in cutting force. An active electromagnetic guide is used for dynamic compensation, thereby improving control precision and response speed.
It achieves refined active compensation during the initial stage of sudden cutting force changes, avoids saw jamming and washboard effects, improves the quality of machined surfaces and the safety of equipment operation, and reduces the impact on overall production capacity.
Smart Images

Figure CN123113582A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of automatic control technology in the wood processing industry, specifically to an industrial real-time control method and process control system for sawn timber processing equipment. Background Technology
[0002] In large-scale industrial lumber processing production lines, the sawing system is usually driven by a spindle servo motor to rotate the saw blade at high speed, and the wood is conveyed along the processing direction by a feed servo system.
[0003] The distribution of knots, grain, and density within the wood is significantly uneven, and the cutting resistance fluctuates drastically depending on the local characteristics of the material, which can easily lead to industrial failures such as lateral vibration of the saw blade and saw jamming.
[0004] Currently, the control schemes widely used in industrial settings consist of a spindle servo drive system, a feed servo drive system, mechanical sensors, a control unit, and a human-machine interface terminal. The control logic is as follows: after the mechanical sensors collect the spindle vibration or feed resistance signals, they are uploaded to the control unit. The control unit performs threshold judgment on the signals based on a steady-state model or a proportional-integral-derivative feedback algorithm. When an abnormality is detected, it triggers overall speed reduction or shutdown protection.
[0005] Because the signals collected by the mechanical vibration and torque sensors lag behind the moment of the sudden change in cutting force, this type of control scheme can only start to respond after the saw blade has already experienced macroscopic vibration or the motor has been overloaded. Moreover, the response method is to reduce the speed or stop the machine as a whole. It lacks a refined active compensation method for local transient disturbances, which makes it difficult to balance the processing quality and production efficiency of the production line.
[0006] Some high-end production lines are equipped with offline optical or X-ray scanning equipment to pre-scan wood defects. However, due to the influence of sawdust and resin contamination, the detection confidence decreases as the continuous operation time of the production line increases, making it difficult to achieve hard real-time spatial alignment with online cutting control.
[0007] Therefore, existing industrial real-time control software and advanced process control systems are significantly inadequate in dealing with sudden changes in cutting force during sawing, and there is an urgent need for a new type of industrial automatic control system device. Summary of the Invention
[0008] The technical problem this invention aims to solve is that existing industrial real-time control software and advanced process control schemes for sawing equipment rely on the hysteresis feedback and post-event monitoring mechanism of mechanical sensors. They cannot complete the perception and active suppression within the microsecond to millisecond time window when a sudden change in cutting force occurs. Moreover, the response after triggering protection is to reduce the overall speed or stop the machine. There is a lack of refined active compensation methods for local transient disturbances, which leads to a decline in processing quality, increased equipment wear and tear, and forced production line to reduce speed and stop.
[0009] To solve the above-mentioned technical problems, the present invention provides an industrial real-time control method for sawing equipment, comprising: acquiring the q-axis current signal of the spindle servo motor and the acoustic emission signal output by the acoustic emission sensor installed at the spindle bearing seat;
[0010] The q-axis current signal is bandpass filtered to obtain the current ripple component, and the estimated value of the cutting load torque is obtained by solving the spindle rotor motion equation.
[0011] After envelope demodulation of the acoustic emission signal, energy integration is performed within a preset sliding time window to obtain the acoustic emission energy characteristic value;
[0012] The estimated value of the cutting load torque and the characteristic value of the acoustic emission energy are fused together according to a preset weighting coefficient to obtain the fused transient cutting force estimate.
[0013] Determine whether the absolute value of the rate of change of the estimated transient cutting force is greater than a preset abrupt change detection threshold. If it is greater, determine that a transient cutting force abrupt change event has occurred.
[0014] Based on a pre-established nominal state-space model of the wood processing dynamics, synchronous simulation and deduction are performed within a real-time controller to obtain the predicted values of the system state.
[0015] Based on the equivalent disturbance excitation obtained by converting the transient cutting force estimate and the system state prediction, the disturbance estimate is obtained in real time through a disturbance observer.
[0016] The sliding surface function is constructed based on the tracking error between the reference trajectory and the actual state of the system, and the feedback control quantity is calculated according to the sliding mode control law.
[0017] The disturbance estimate is superimposed with the feedback control quantity to obtain the feedforward-feedback coordinated control quantity;
[0018] The feedforward-feedback coordinated control quantity is sent to the feed servo driver and the spindle servo driver to perform closed-loop control of the feed servo system.
[0019] The present invention provides a process control system for sawing equipment that implements the above method, including a mechanical execution body layer, a sensing and perception layer, an edge real-time control layer and an execution drive layer, wherein each layer interacts with the other through an industrial real-time bus for hard real-time data exchange.
[0020] Compared with the prior art, the beneficial effects of the present invention by adopting the above technical solution are as follows:
[0021] 1. This invention utilizes a heterogeneous signal soft measurement method that integrates the q-axis current ripple of the spindle servo motor and the acoustic emission signal. It does not rely on optical detection methods that are susceptible to wood chips and resin contamination. Instead of relying on the post-event sensing method triggered by vibration in the existing technology, the perception delay of sudden changes in cutting force is compressed to the decoupled identification stage when the sudden change in cutting force occurs, thus fundamentally alleviating the problem of feedback lag.
[0022] 2. This invention constructs a feedforward compensation mechanism by embedding a digital twin model in the edge real-time controller and a disturbance observer. This mechanism intercepts disturbances before the saw blade undergoes macroscopic lateral vibration or plastic deformation. It also outputs control quantities in coordination with the sliding mode feedback control law, which helps to avoid saw clamping and washboard effects, and improves the quality of the processed surface and the safety of equipment operation.
[0023] 3. This invention refines the coarse response of overall deceleration or shutdown in the prior art into local and instantaneous active compensation actions by using a nonlinear dynamic trajectory compensation method that executes deceleration, deviation adjustment and recovery in sequence, and linkage stiffness compensation of active electromagnetic guides. This helps to reduce the impact of local transient disturbances on the overall production line capacity.
[0024] 4. This invention improves the reliability of the solution in harsh industrial environments by cross-checking the signals of the current sensor and the acoustic emission sensor and setting a fault degradation mechanism, so that the system can still maintain basic real-time anti-interference capability based on the other sensor signal even when some sensor signals are abnormal. Attached Figure Description
[0025] Figure 1 This is an overall architecture diagram of the system in an embodiment of the present invention;
[0026] Figure 2 This is a flowchart illustrating the execution of the transient cutting force soft measurement algorithm in an embodiment of the present invention.
[0027] Figure 3 This is a flowchart illustrating the execution of multi-axis dynamic trajectory compensation and fault-tolerant control in an embodiment of the present invention. Detailed Implementation
[0028] The following is in conjunction with the appendix Figure 1-3 The specific embodiments of the present invention will be further described below. It should be noted that the description of these embodiments is for the purpose of helping to understand the present invention, but does not constitute a limitation of the present invention.
[0029] The specific embodiments of the present invention will be described in detail below with reference to the technical solution of the present invention, so that those skilled in the art can implement the present invention.
[0030] I. System Overall Architecture
[0031] The process control system provided by this invention consists of four parts at the physical level: a mechanical execution body layer, a sensing and perception layer, an edge real-time control layer, and an execution drive layer.
[0032] The mechanical actuator layer includes a spindle, saw blade, feed servo system, and active electromagnetic guides installed on both sides of the saw blade. The active electromagnetic guides are used to dynamically compensate for the stiffness of the saw blade by applying controllable electromagnetic force.
[0033] The sensing layer includes a current sensor and an acoustic emission sensor. The current sensor is installed inside or at the output end of the spindle servo motor to collect the q-axis current signal of the spindle servo motor. The acoustic emission sensor is installed in a safe position in the non-contact cutting area of the spindle bearing housing to collect broadband acoustic emission signals. Both types of sensors are indirect measurement elements installed in the non-contact cutting area, eliminating the need to place optical detection devices that are susceptible to contamination by sawdust and resin in the cutting area.
[0034] The edge real-time control layer is an industrial control computer platform that integrates a field-programmable gate array or runs a real-time operating system. Internally, it integrates a transient cutting force soft measurement module, a wood processing process dynamics digital twin model module, a disturbance observer module, and a sliding mode feedback controller module in the form of functional modules. The modules are interconnected through internal high-speed data channels and are equipped with control law superposition points to superimpose and synthesize feedforward compensation and feedback control quantities.
[0035] The execution drive layer includes a spindle servo driver, a feed servo driver, and an electromagnetic guide driver, which are used to drive the spindle servo motor, the feed servo system, and the active electromagnetic guide, respectively.
[0036] The sensing layer and the edge real-time control layer, as well as the edge real-time control layer and the execution drive layer, are all connected via an industrial real-time bus. The communication cycle of the industrial real-time bus is no higher than 125 microseconds to ensure that signal acquisition and control command issuance meet the real-time requirements of microsecond to millisecond level. The control signal flow is as follows: the sensing signal is uploaded from the sensing layer to the transient cutting force soft measurement module. The cutting force estimate output by the transient cutting force soft measurement module is simultaneously sent to the digital twin model module and the disturbance observer module. The disturbance estimate output by the disturbance observer module and the feedback control quantity output by the sliding mode feedback controller module are combined at the control law superposition point to form a feedforward-feedback coordinated control quantity, which is then applied to the feed servo system and the active electromagnetic guide through the execution drive layer to form a complete closed loop.
[0037] II. Microsecond-level soft measurement method for transient cutting force
[0038] The core idea of the soft measurement method described in this invention is not to directly measure the cutting force itself, but to fuse the time-frequency domain characteristics of two heterogeneous signals, namely the spindle servo motor current ripple and the acoustic emission signal, to decouple the transient estimate of the equivalent cutting force in real time.
[0039] The electromagnetic torque equation of the spindle servo motor is:
[0040] ;
[0041] In the formula, Indicates the spindle servo motor at time Output electromagnetic torque;
[0042] This is the motor torque constant, determined by the motor's inherent parameters;
[0043] The spindle servo motor at time The q-axis current.
[0044] The equation of motion for the main shaft rotor is:
[0045] ;
[0046] In the formula, The equivalent rotational inertia of the main spindle system;
[0047] Main axis angular velocity;
[0048] The equivalent torque of the cutting load is the load torque corresponding to the transient cutting force to be estimated.
[0049] It is the viscous friction coefficient of the main spindle system.
[0050] The inverse solution of the above equation yields the current-based estimation formula for the cutting load torque:
[0051] ;
[0052] In the formula, This is an estimated value for the cutting load torque;
[0053] This is the estimated angular velocity value after filtering by the velocity observation process; the meanings of the other symbols are the same as in the previous formula.
[0054] Acoustic emission signal energy feature extraction employs a sliding window energy integration method:
[0055] ;
[0056] In the formula, For a moment The acoustic emission signal energy within the corresponding sliding window;
[0057] For acoustic emission sensors at time The instantaneous amplitude of the acquired signal;
[0058] The sliding window duration is denoted by , and its value ranges from tens to hundreds of microseconds.
[0059] The two types of features are weighted and fused to obtain the fused transient cutting force estimate:
[0060] ;
[0061] In the formula, This is the estimated value of the transient cutting force after fusion;
[0062] , These are the weighting coefficients for the current characteristic component and the acoustic emission characteristic component, respectively, and are fixed coefficients predetermined based on historical calibration data;
[0063] This is a compensation constant term used to correct steady-state baseline offset.
[0064] The mutation detection criterion is:
[0065] ;
[0066] In the formula, The threshold for detecting sudden changes in the rate of change of cutting force is set. When this criterion is met, a transient cutting force change event is determined to have occurred.
[0067] The specific implementation process of the soft measurement method is as follows: the edge real-time control layer synchronously acquires the q-axis current signal through the industrial real-time bus with a sampling period of no more than 125 microseconds. and acoustic emission signals The two signals are time-stamped and aligned to eliminate phase deviations caused by differences in the acquisition channels.
[0068] The current signal is preprocessed by bandpass filtering to extract the current ripple component and filter out the power frequency fundamental wave and high-frequency switching noise.
[0069] Substituting the purified current signal into the spindle rotor motion equation, and combining the estimated angular velocity and its differential component output from the velocity observation stage, the estimated cutting load torque is calculated. ;
[0070] The raw acoustic emission signal is demodulated using Hilbert envelope and bandpass filtered to obtain the envelope characteristics of the broadband acoustic emission signal, and then processed within a preset sliding window. Energy integration is performed internally to obtain the acoustic emission energy characteristics. ;
[0071] The two feature values are fused using a weighted fusion formula, and the fused transient cutting force estimate is output. ;
[0072] right The rate of change is obtained by performing a difference operation and compared with a preset mutation detection threshold. The comparison is performed. When the absolute value of the rate of change is greater than the threshold, it is determined that a transient cutting force change has occurred at the current moment. The change trigger flag and the corresponding cutting force estimate are output. The result is then sent to the digital twin model module and the disturbance observer module simultaneously to enter the feedforward compensation control process.
[0073] When the absolute value of the rate of change is not greater than the threshold, return to continue signal acquisition for the next sampling cycle.
[0074] III. Digital Twin and Disturbance Observer Feedforward-Feedback Cooperative Interception Control Method
[0075] After obtaining the transient cutting force estimate, the present invention further embeds a digital twin model module of wood processing dynamics into the edge real-time control layer, combines it with the disturbance observer module to achieve feedforward compensation, and coordinates with the sliding mode feedback controller module to output control quantity.
[0076] The dynamic characteristics of the feed servo system and the spindle system are established as the following nominal state-space model:
[0077] ;
[0078] In the formula, This is the system state vector, which includes the spindle angular velocity, feed displacement, and feed rate components.
[0079] The control input is the torque or speed command received by the servo driver.
[0080] The equivalent concentrated disturbance is used to uniformly characterize the equivalent effect of the sudden change in cutting force caused by the sudden change in wood density on the system;
[0081] , These are the system state matrix and the input matrix, respectively, which are obtained from the mechanical parameters of the spindle and the feed system.
[0082] The digital twin model module performs synchronous real-time simulation and deduction within the edge real-time control layer using the aforementioned nominal model, and outputs predicted values of the system state. , serving as a reference benchmark for the disturbance observer module.
[0083] The disturbance observer module uses the following equation to reconstruct concentrated disturbances in real time:
[0084] ;
[0085] In the formula, This is the estimated value of the disturbance;
[0086] To perturb the observer gain, the pole placement method is used for tuning to ensure that the observation error converges to zero at an exponential rate.
[0087] The meanings of the remaining symbols are the same as in the previous formula. In specific implementation, Output from the transient cutting force soft measurement module Through equivalent transformation and numerical differentiation, the information on sudden changes in cutting force is directly introduced into the disturbance reconstruction process.
[0088] The sliding mode feedback controller module defines the sliding surface as follows:
[0089] ;
[0090] In the formula, For sliding surface functions;
[0091] For system state tracking error, For reference trajectory;
[0092] These are integral coefficients used to eliminate steady-state errors.
[0093] The sliding mode feedback control law is:
[0094] ;
[0095] In the formula, For feedback control;
[0096] The gain is switched to ensure that the system state reaches and remains near the sliding surface;
[0097] It is a symbolic function;
[0098] This is the proportional gain.
[0099] The feedforward-feedback coordinated control quantity is:
[0100]
[0101] The meanings of the symbols in the formula are the same as those in the previous formulas. The control quantity is synthesized by the control law superposition point and then sent to the execution driving layer.
[0102] The specific implementation process of the control method is as follows: the edge real-time control layer receives the estimated cutting force value output by the transient cutting force soft measurement module. And the mutation trigger flag, which is converted into the perturbation input of the state space model input channel, and used as the input excitation of the perturbation observer module;
[0103] The digital twin model module is based on the nominal state-space model and combines the current actual control commands with the state measurement values from the previous cycle to synchronously and recursively calculate the predicted system state values. ;
[0104] The disturbance observer module calculates the disturbance estimate in real time by recursively using the observer equation, combined with the equivalent disturbance excitation and the model prediction output. ;
[0105] The sliding mode feedback controller module is based on the current tracking error. And its integral term to calculate the sliding surface function And calculate the feedback control quantity according to the sliding mode control law. ;
[0106] The control law superposition point will affect the disturbance estimate. With feedback control quantity By superimposing and synthesizing the feedforward-feedback coordinated control quantity, the feedforward-feedback coordinated control quantity is obtained. And when the amplitude of the disturbance estimate is greater than the preset flutter warning threshold, a flutter warning flag is generated simultaneously, triggering the subsequent multi-axis dynamic trajectory compensation process;
[0107] The edge real-time control layer will coordinate feedforward and feedback control quantities. In the next bus communication cycle, the data is sent to the feed servo driver and the spindle servo driver to complete one control loop, and then continues to enter the cyclic control of the next sampling cycle.
[0108] IV. Multi-axis microsecond-level dynamic trajectory compensation and fault-tolerant control methods
[0109] When the disturbance estimate exceeds the flutter warning threshold, the edge real-time control layer further generates a nonlinear dynamic trajectory compensation command and links the active electromagnetic guide to implement stiffness compensation, while simultaneously performing multi-sensor fault-tolerant judgment.
[0110] The feed rate compensation command is:
[0111] ;
[0112] In the formula, The feed rate command after compensation;
[0113] This is the original feed rate reference value;
[0114] The speed reduction factor is determined by the calibration of the production line process parameters;
[0115] This is the estimated value of the aforementioned disturbance.
[0116] The electromagnetic guide compensation force is:
[0117] ;
[0118] In the formula, The reverse compensating electromagnetic force applied to the saw blade by the active electromagnetic guide;
[0119] This is the electromagnetic force gain coefficient.
[0120] The specific implementation process of the compensation method is as follows: the edge real-time control layer receives the flutter early warning flag and the corresponding disturbance estimate. ,in accordance with The amplitude is used to classify the compensation level; the larger the amplitude, the stronger the corresponding compensation action.
[0121] Based on the compensation level, a three-stage nonlinear trajectory generation rule is implemented sequentially, consisting of a deceleration phase, a deviation adjustment phase, and a recovery phase. First, the instantaneous deceleration command is calculated using the feed rate compensation formula. Then, combined with the feed lateral fine-tuning bias, a complete dynamic trajectory compensation command sequence is generated;
[0122] The reverse compensation electromagnetic force command of the active electromagnetic guide is calculated synchronously based on the electromagnetic force compensation formula. And within the same bus communication cycle, the instruction and the trajectory compensation instruction are sent together;
[0123] The feed servo driver and the electromagnetic guide driver receive and execute the above instructions respectively. After the feed system completes the instantaneous deceleration and fine-tuning actions according to the compensation trajectory, it automatically executes the recovery segment trajectory when the disturbance estimate falls below the chatter warning threshold, gradually restoring the original reference feed speed. ;
[0124] While performing the above compensation actions, the edge real-time control layer cross-verifies the signals fed back by the current sensor and the acoustic emission sensor. If it is determined that the signal of one of the sensor channels is abnormal, such as the signal being constant for a long time or exceeding the physical reasonable range, the channel is set to a fault degradation state, switched to a single-source soft measurement mode that only relies on the other sensor signal, and an alarm prompt is triggered on the human-machine interface to ensure that the system can still maintain basic real-time anti-interference capability in the event of partial sensor failure.
[0125] V. Parameter Description of the Implementation Examples
[0126] In one specific embodiment, the industrial real-time bus adopts a fieldbus with a communication cycle of 125 microseconds, and the sampling frequency of the q-axis current of the spindle servo motor is synchronized with the bus cycle;
[0127] Sliding window duration The value is one hundred microseconds;
[0128] mutation detection threshold The vibration warning threshold and compensation level division range are all calibrated by the production line for different batches of wood materials during the offline calibration stage and written into the parameter configuration file of the edge real-time control layer. After the above parameter configuration, the present invention can complete soft measurement sensing within a time window of several microseconds to tens of microseconds after the transient cutting force change occurs, and complete the generation and superposition of feedforward compensation before the cutting force change actually causes macroscopic vibration of the saw blade, so as to realize the active resolution of transient cutting force change.
[0129] VI. Alternative Implementation Methods
[0130] In other embodiments, the acoustic emission sensor mounted on the spindle bearing housing can be replaced with a high-frequency response vibration acceleration sensor to collect high-frequency vibration acceleration signals from the spindle or saw blade support structure, and the envelope demodulation of this signal can then replace the acoustic emission energy characteristics. While participating in the aforementioned weighted fusion calculation, the remaining soft measurement, feedforward compensation, and trajectory compensation processes remain unchanged, thus enabling early detection and proactive suppression of transient cutting force changes.
[0131] In other embodiments, the combination of the disturbance observer module and the sliding mode feedback controller module can be replaced with a combination of a disturbance-state joint estimator based on extended Kalman filtering and a model predictive controller. That is, the extended Kalman filter is used to jointly estimate the system state and equivalent disturbance in each sampling period, and the estimation results are input into the model predictive controller to perform online rolling optimization to solve the control sequence, replacing the original sliding mode feedback control law to output feedback control quantity. The remaining soft measurement sensing and multi-axis trajectory compensation execution links remain unchanged, which can also achieve the feedforward-feedback cooperative active suppression effect for sudden changes in cutting force.
[0132] The above embodiments are merely illustrative examples of the technical solutions of the present invention and are not intended to limit the scope of protection of the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.
[0133] Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make possible changes and modifications without departing from the spirit and scope of the present invention. Therefore, any modifications, equivalent changes and alterations made to the above embodiments based on the technical essence of the present invention without departing from the content of the technical solution of the present invention shall fall within the protection scope defined by the claims of the present invention.
Claims
1. An industrial real-time control method of a sawn timber processing plant, characterized in that, include: Acquire the q-axis current signal of the spindle servo motor and the acoustic emission signal output by the acoustic emission sensor installed at the spindle bearing housing; The q-axis current signal is bandpass filtered to obtain the current ripple component, and the estimated value of the cutting load torque is obtained by solving the spindle rotor motion equation. After envelope demodulation of the acoustic emission signal, energy integration is performed within a preset sliding time window to obtain the acoustic emission energy characteristic value; The estimated value of the cutting load torque and the characteristic value of the acoustic emission energy are fused together according to a preset weighting coefficient to obtain the fused transient cutting force estimate. Determine whether the absolute value of the rate of change of the estimated transient cutting force is greater than a preset abrupt change detection threshold. If it is greater, determine that a transient cutting force abrupt change event has occurred. Based on a pre-established nominal state-space model of the wood processing dynamics, synchronous simulation and deduction are performed within a real-time controller to obtain the predicted values of the system state. Based on the equivalent disturbance excitation obtained by converting the transient cutting force estimate and the system state prediction, the disturbance estimate is obtained in real time through a disturbance observer. The sliding surface function is constructed based on the tracking error between the reference trajectory and the actual state of the system, and the feedback control quantity is calculated according to the sliding mode control law. The disturbance estimate is superimposed with the feedback control quantity to obtain the feedforward-feedback coordinated control quantity; The feedforward-feedback coordinated control quantity is sent to the feed servo driver and the spindle servo driver to perform closed-loop control of the feed servo system.
2. The method according to claim 1, characterized in that, The preset weighting coefficients are fixed coefficients predetermined based on historical calibration data.
3. The method according to claim 1, characterized in that, When the amplitude of the disturbance estimate is greater than the preset flutter warning threshold, a flutter warning flag is generated, and a nonlinear dynamic trajectory compensation command is generated based on the flutter warning flag. The nonlinear dynamic trajectory compensation command includes a deceleration stage, a skew adjustment stage, and a recovery stage executed sequentially. In the deceleration stage, a compensated feed rate command is calculated based on the disturbance estimate. In the skew adjustment stage, a lateral fine-tuning skew is applied to the feed trajectory. In the recovery stage, when the disturbance estimate falls back below the flutter warning threshold, the feed rate is gradually restored to the feed rate reference value before compensation.
4. The method according to claim 3, characterized in that, While generating the nonlinear dynamic trajectory compensation command, the electromagnetic guide compensation force command is calculated based on the disturbance estimate, and the electromagnetic guide compensation force command and the nonlinear dynamic trajectory compensation command are sent to the electromagnetic guide driver corresponding to the active electromagnetic guide installed on both sides of the saw blade in the same bus communication cycle.
5. The method according to claim 1, characterized in that, While acquiring the q-axis current signal and the acoustic emission signal, signal anomaly detection is performed on the q-axis current signal and the acoustic emission signal. When an anomaly is detected in one of the signals, the corresponding sensor channel is set to a fault degradation state, and the calculation of the transient cutting force estimate is switched to be based solely on the other signal.
6. The method according to any one of claims 1 to 5, characterized in that, The spindle servo motor and the acoustic emission sensor, as well as the real-time controller and the feed servo driver and the spindle servo driver, are all connected via an industrial real-time bus with a communication cycle of no more than 125 microseconds.
7. A process control system for sawn timber processing equipment, characterized in that, It includes a mechanical execution body layer, a sensing and perception layer, an edge real-time control layer, and an execution drive layer, wherein: The mechanical actuator body layer includes a spindle, a saw blade, a feed servo system, and active electromagnetic guides installed on both sides of the saw blade. The sensing layer includes a current sensor mounted on the spindle servo motor and an acoustic emission sensor mounted on the spindle bearing housing. The current sensor is used to collect the q-axis current signal of the spindle servo motor, and the acoustic emission sensor is used to collect the acoustic emission signal. The edge real-time control layer includes a transient cutting force soft measurement module, a digital twin model module, a disturbance observer module, and a sliding mode feedback controller module. The transient cutting force soft measurement module is used to fuse the q-axis current signal and the acoustic emission signal to obtain a transient cutting force estimate. The digital twin model module is used to synchronously simulate and deduce the system state prediction based on the nominal state space model of the wood processing dynamics. The disturbance observer module is used to reconstruct the disturbance estimate based on the transient cutting force estimate and the system state prediction. The sliding mode feedback controller module is used to calculate the feedback control quantity based on the tracking error. The edge real-time control layer also includes a control law superposition point, which is used to superimpose the disturbance estimate and the feedback control quantity to synthesize a feedforward-feedback coordinated control quantity. The execution drive layer includes a spindle servo driver, a feed servo driver, and an electromagnetic guide driver, which are used to receive the feedforward-feedback coordinated control quantity and drive the corresponding execution components. The sensing layer and the edge real-time control layer, as well as the edge real-time control layer and the execution drive layer, are all connected via an industrial real-time bus.
8. The process control system according to claim 7, characterized in that, The edge real-time control layer also includes a fault-tolerant judgment module, which is used to cross-verify the signals fed back by the current sensor and the acoustic emission sensor, and set the corresponding channel to a fault-degraded state when it is determined that one of the signals is abnormal.
9. The process control system according to claim 7, characterized in that, The edge real-time control layer is also used to generate a nonlinear dynamic trajectory compensation command when the amplitude of the disturbance estimate is greater than the preset flutter warning threshold, and to link the active electromagnetic guide to execute the reverse compensation electromagnetic force.
