Efficient cutting machining method and system for titanium alloy parts for automobiles

Abstract

The invention relates to the technical field of metal cutting machining, and discloses an efficient cutting machining method and system for a titanium alloy part for an automobile. Signals are collected through multi-physics field sensing to estimate the state of a cutting interface, feedback instructions are calculated based on state deviation, electric field parameters are adjusted in real time, and closed-loop control is formed; an efficient cutting machining system for titanium alloy parts for automobiles comprises a high-frequency pulse energy source, a cutting fluid supply module, a multi-physics field sensing module, a state estimation module and a feedback control module. The cutting environment is optimized through the in-situ plasma layer. The multi-physical field signals are fused for depth state perception, adaptive model predictive control is adopted, predictive optimal adjustment of the cutting process is achieved, and the problems that in the prior art, perception is shallow, and control is lagged are solved.

Description

Technical Field

[0001] The present invention relates to the technical field of metal cutting, and in particular to a high-efficiency cutting method and system for titanium alloy parts for automobiles. Background Art

[0002] Due to its excellent high specific strength, corrosion resistance, and high-temperature resistance, titanium alloys are increasingly used in the automotive industry, especially in key engine components, exhaust systems, and lightweight structural parts. However, the physical properties of titanium alloys themselves also make them a typical difficult-to-machine material. Their low thermal conductivity makes it difficult to quickly conduct the large amount of heat generated during cutting, resulting in a sharp increase in the temperature of the cutting area. At the same time, they are extremely chemically active at high temperatures and easily bond with the tool material, exacerbating the tool's adhesive wear and diffusion wear. These factors work together to result in short tool life, poor machined surface quality, and low machining efficiency, seriously restricting the further promotion and application of titanium alloy parts in the automotive field.

[0003] Existing technologies usually use high-pressure cooling or low-temperature cooling to improve the cutting environment, but the cutting fluid has limited permeability in the high-pressure and high-temperature cutting zone, making it difficult to achieve the ideal cooling and lubrication effect. At the same time, in terms of cutting process monitoring and control, most existing methods rely on monitoring single or limited external physical quantities such as cutting force and vibration. This method is difficult to deeply reveal the inherent evolution laws of complex physical processes such as friction, heat transfer and tool micro-wear at the cutting interface. Therefore, control strategies based on such limited information are often relatively passive and cannot accurately and proactively regulate the dynamically changing cutting process, resulting in a large room for improvement in the stability and optimization of the machining process.

[0004] Therefore, the present invention provides a high-efficiency cutting method and system for titanium alloy parts for automobiles to solve the shortcomings of the prior art. Summary of the Invention

[0005] In response to the shortcomings of the existing technology, the present invention provides a high-efficiency cutting processing method and system for titanium alloy parts for automobiles, which solves the problems of low processing efficiency of titanium alloy parts, severe tool wear and unstable processing process caused by harsh cutting environment and imprecise process control.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solutions: a high-efficiency cutting method for automotive titanium alloy parts, the method comprising the following steps:

[0007] S1, applying a high-frequency pulse electric field between the tool and the titanium alloy workpiece, and supplying cutting fluid to the cutting area to perform cutting processing;

[0008] S2. During the cutting process, a multi-physics field sensing module is used to collect dynamic physical signals between the tool and the titanium alloy workpiece in real time to generate an observation vector;

[0009] S3, inputting the observation vector into a state space model, estimating the state of the cutting interface between the tool and the titanium alloy workpiece, and obtaining a state estimation vector;

[0010] S4. Calculating a feedback control instruction for adjusting the high-frequency pulse electric field based on a deviation between the state estimation vector and a preset reference state vector;

[0011] S5. According to the calculated feedback control instruction, the parameters of the high-frequency pulse electric field are adjusted in real time by a high-frequency pulse energy source.

[0012] Preferably, in step S1, the step of applying a high-frequency pulse electric field between the tool and the titanium alloy workpiece and supplying cutting fluid to the cutting area to perform cutting processing includes:

[0013] Through the synergistic effect of the high-frequency pulse electric field and the cutting fluid, the cutting fluid is induced to undergo a phase change at the cutting interface between the tool and the titanium alloy workpiece to generate a plasma layer with both lubrication and heat insulation properties.

[0014] Preferably, in step S2, the step of collecting dynamic physical signals between the tool and the titanium alloy workpiece in real time by a multi-physics field sensing module includes:

[0015] The signals representing the mechanical properties of the cutting process, the signals representing the microscopic fracture properties of the material, and the signals representing the thermal properties are collected to form the dynamic physical signals.

[0016] Preferably, in step S3, the step of inputting the observation vector into a state space model to estimate the state of the cutting interface between the tool and the titanium alloy workpiece includes:

[0017] The state estimation vector is determined to characterize the physical state of the cutting interface that cannot be directly measured, where the physical state consists of an interface equivalent friction coefficient, a heat flux distribution coefficient, and a tool wear rate.

[0018] Preferably, in step S3, the step of estimating the state of the cutting interface between the tool and the titanium alloy workpiece includes:

[0019] The state estimation vector is calculated by using a Kalman filter with the observation vector as input and combined with the state space model.

[0020] Preferably, in step S4, the step of calculating a feedback control instruction for adjusting the high-frequency pulse electric field based on a deviation between the state estimation vector and a preset reference state vector includes:

[0021] An optimization problem is solved by a model predictive control algorithm, wherein the optimization problem aims to minimize the prediction deviation between the state estimation vector and the preset reference state vector in the future prediction time domain, thereby obtaining an optimal control sequence, and using the first control instruction of the optimal control sequence as the feedback control instruction.

[0022] Preferably, before executing step S1, feedforward control instructions are pre-calculated according to a preset cutting process path and using the state space model;

[0023] In step S5, the step of adjusting the parameters of the high-frequency pulse electric field in real time by a high-frequency pulse energy source according to the calculated feedback control instruction includes:

[0024] The feedforward control instruction and the feedback control instruction are superimposed to generate a composite control instruction, and the parameters of the high-frequency pulse electric field are adjusted in real time according to the composite control instruction.

[0025] Preferably, in step S3, the model parameters of the state-space model are updated online, and the online updating step includes comparing the observation output predicted by the state-space model with the observation vector to generate a residual, and adjusting the model parameters of the state-space model according to the generated residual.

[0026] Preferably, the step of adjusting the model parameters of the state-space model includes:

[0027] The updated value of the model parameter at the current moment is calculated by combining the residual and the estimated value of the model parameter at the previous moment through a recursive least squares algorithm.

[0028] The present invention also provides a high-efficiency cutting system for titanium alloy parts for automobiles, the system comprising the following modules:

[0029] A high-frequency pulse energy source is used to apply a high-frequency pulse electric field between the tool and the titanium alloy workpiece, and to adjust the parameters of the high-frequency pulse electric field in real time according to feedback control instructions;

[0030] A cutting fluid supply module, used for supplying cutting fluid to the cutting area;

[0031] A multi-physics field sensing module, used for collecting dynamic physical signals between the tool and the titanium alloy workpiece in real time during the cutting process to generate an observation vector;

[0032] The multi-physics field sensing module includes a force sensor for collecting mechanical signals, an acoustic emission sensor for collecting microscopic fracture signals, and an infrared thermal imager for collecting thermal signals;

[0033] a state estimation module, configured to receive the observation vector and estimate the state of the cutting interface between the tool and the titanium alloy workpiece based on a state space model to generate a state estimation vector;

[0034] A feedback control module is used to calculate the feedback control instruction based on the deviation between the state estimation vector and a preset reference state vector, and send the feedback control instruction to the high-frequency pulse energy source.

[0035] The present invention provides a method and system for efficiently cutting titanium alloy parts for automobiles. It has the following beneficial effects:

[0036] 1. This invention synergizes a high-frequency pulsed electric field with cutting fluid to generate an in-situ plasma layer at the cutting interface. This enables active regulation of the friction and heat transfer characteristics of the titanium alloy cutting interface. Compared to conventional methods that rely on external cooling and lubrication, this invention addresses the difficulties of traditional cutting fluids in effectively entering the cutting zone and their limited lubrication and thermal insulation effects, providing a new physical approach for optimizing titanium alloy machining.

[0037] 2. This invention estimates the internal state of the cutting interface by integrating multi-physics signals such as mechanics, acoustic emission, and thermal analysis, and combining them with a state-space model. Key internal parameters, including interface friction, heat flux distribution, and tool wear rate, are derived. This contrasts with existing technologies that rely solely on a single, externally measurable physical quantity for process monitoring. This solves the problem of ambiguous understanding of cutting mechanisms and inaccurate state judgments due to incomplete information, significantly improving the depth and accuracy of process perception.

[0038] 3. This invention utilizes a model predictive control algorithm to establish a future-oriented closed-loop feedback control system. This control strategy can predict the future evolution of cutting conditions and determine the optimal control sequence. Compared to passive error-correcting control methods such as PID commonly used in existing technologies, this invention addresses the shortcomings of delayed response and prone to overshoot when addressing time lag and nonlinearities in the cutting process. It also achieves predictable and optimal adjustment of high-frequency pulsed electric field parameters.

[0039] 4. This invention designs an online adaptive parameter update mechanism for the state-space model. The system utilizes a recursive algorithm to continuously modify model parameters based on the deviation between model predictions and actual observations. This provides the control system with learning and adaptive capabilities. This overcomes the shortcomings of existing fixed-parameter models, which are unable to adapt to dynamic processes such as tool wear and changing operating conditions. This ensures the long-term effectiveness and robustness of the state estimation and control strategy throughout the entire machining cycle. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1 is a flow chart of the method of the present invention;

[0041] Figure 2 This is a system architecture diagram of the present invention. DETAILED DESCRIPTION

[0042] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the drawings in the present specification. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0043] See also Figure 1 The embodiment of the present invention provides a method and system for efficiently cutting titanium alloy parts for automobiles, the method comprising the following steps:

[0044] S1, applying a high-frequency pulse electric field between the tool and the titanium alloy workpiece, and supplying cutting fluid to the cutting area to perform cutting processing;

[0045] In this embodiment, step S1 is intended to apply a high-frequency pulse electric field between the tool and the titanium alloy workpiece, and simultaneously supply cutting fluid to the cutting area to perform efficient cutting processing.

[0046] Specifically, the high-frequency pulsed electric field is provided by a high-frequency pulsed energy source. This energy source outputs an electrical signal of a specific frequency and pulse width. This signal is connected to electrodes between the tool and the workpiece, creating an electric field at the cutting interface. The cutting fluid supply module is responsible for delivering cutting fluid to the cutting area, ensuring that the cutting fluid adequately covers the cutting interface.

[0047] The high-frequency pulse electric field and the cutting fluid cooperate in the following manner:

[0048] When a high-frequency pulsed electric field is applied to the cutting interface between the tool and the titanium alloy workpiece, the dielectric molecules in the cutting fluid are ionized by the strong electric field, inducing a phase change in the cutting fluid at the cutting interface, thereby generating a plasma layer with both lubricating and thermal insulation properties. This plasma layer significantly reduces friction between the tool and the workpiece, reducing cutting forces and effectively isolating the conduction of cutting heat to the tool and workpiece, thereby improving cutting performance.

[0049] In one possible implementation, the formation of the plasma layer involves microscopic physical and chemical changes in the cutting fluid. When a high-frequency pulsed electric field acts on the cutting fluid, the oscillating effect of the electric field can weaken or even break the molecular bonds in the cutting fluid, forming free ions and electrons. Furthermore, the energy of the high-frequency pulse can excite these free particles, causing them to reach an excited state. These particles then transfer energy to the surrounding cutting fluid molecules through collisions and other means, causing a local temperature increase and a phase transition, ultimately forming a plasma. This plasma layer is dynamically formed and maintained during the cutting process, and its thickness, density, and stability are influenced by the parameters of the high-frequency pulsed electric field and the properties of the cutting fluid.

[0050] Generally, the electric field parameters output by a high-frequency pulse energy source, including but not limited to voltage, frequency, pulse width, and duty cycle, can be adjusted based on the specific cutting conditions and desired machining results. For example, adjusting the electric field frequency and pulse width can optimize the efficiency and stability of the plasma layer, thereby regulating its lubrication and thermal insulation properties.

[0051] Preferably, the cutting fluid may be a liquid having specific dielectric constant and thermophysical properties, so as to better respond to the action of the high-frequency pulse electric field and promote the effective generation of the plasma layer.

[0052] In one possible implementation, the generation and maintenance of the plasma layer can be described by the following process:

[0053] When a high-frequency pulsed electric field is applied, the electric field strength in the local area of ​​the cutting fluid is sufficient to cause dielectric breakdown, resulting in ionization of the cutting fluid molecules and generation of plasma. This process is expressed as:

[0054] Liquid+ElectricalEnergy→Plasma;

[0055] In the formula: Liquid refers to cutting fluid; Electrical Energy refers to the electrical energy provided by the high-frequency pulse electric field; Plasma refers to the generated plasma.

[0056] Plasma exhibits both lubricating and insulating properties at the cutting interface. The lubrication mechanism may involve the ions and free electrons in the plasma forming an "electric blanket" effect at the interface, reducing direct contact. Furthermore, the high temperature and pressure of the plasma may induce the cutting fluid to produce lubricating chemicals. The insulating mechanism stems from the inherently low thermal conductivity of the plasma, which hinders the transfer of cutting heat into the tool and workpiece, effectively reducing the temperature in the cutting zone.

[0057] S2. During the cutting process, a multi-physics field sensing module is used to collect dynamic physical signals between the tool and the titanium alloy workpiece in real time to generate an observation vector;

[0058] In this embodiment, step S2 involves using a multi-physics sensing module to collect a set of dynamic physical signals between the tool and the titanium alloy workpiece in real time during the cutting process to generate an observation vector. The acquisition of dynamic physical signals aims to fully reflect the physical state of the cutting interface under transient conditions, providing sufficient input information for subsequent state estimation.

[0059] Specifically, the acquisition of the dynamic physical signal includes the following aspects:

[0060] Acquire signals characterizing the mechanical properties of the cutting process. This signal directly reflects the mechanical behavior of the interaction between the tool and the workpiece during the cutting process. Typically, this signal is acquired using a force sensor, such as a piezoelectric force sensor or a strain gauge sensor. These sensors are typically mounted on the tool holding system or workpiece fixture to measure the components of the cutting force in different directions, such as the main cutting force, feed force, and back force. These mechanical signals can reflect information such as the tool's force state, cutting load, and chip formation mechanism.

[0061] Acquire signals that characterize the material's microscopic fracture characteristics. This signal is crucial for monitoring microscopic damage and internal changes in the material during the cutting process. Preferably, this signal can be acquired using an acoustic emission sensor. Acoustic emission sensors can capture in real time the transient elastic waves generated by the material during processes such as microcrack initiation, propagation, phase transformation, or plastic deformation. The amplitude, frequency, and energy of these acoustic emission signals are closely related to the material's microscopic fracture behavior and serve as an important basis for determining tool wear, chip formation patterns, and material damage.

[0062] Acquire signals representing thermal properties. The cutting process is accompanied by the generation of a large amount of heat. Monitoring the temperature distribution in the cutting area is crucial for assessing tool life, workpiece surface integrity, and the thermal insulation effect of the plasma layer. Preferably, the signals can be acquired using an infrared thermal imager. Infrared thermal imagers can non-contactly measure the temperature field in the cutting area and generate temperature distribution images. These thermal signals reflect the generation, transfer, and dissipation of cutting heat and are crucial for understanding the thermodynamic state of the cutting interface and the thermal insulation properties of the plasma layer.

[0063] In one possible implementation, the multi-physics field sensing module integrates the aforementioned sensors (e.g., force sensors, acoustic emission sensors, infrared thermal imagers, etc.) and is equipped with corresponding data acquisition and signal processing units. The data acquisition unit is responsible for synchronously acquiring raw data from different sensors and performing analog-to-digital conversion. The signal processing unit performs pre-processing on the collected raw signals, such as filtering, denoising, and feature extraction, to eliminate noise interference and extract useful information features.

[0064] The preprocessed signals, representing mechanical, microscopic fracture, and thermal properties, collectively constitute the dynamic physical signal. These dynamic physical signals are then integrated and encoded to form a unified observation vector. This observation vector is a multi-dimensional, real-time data set, each component of which carries important information about a specific aspect of the cutting interface.

[0065] In general, the observation vector is expressed as:

[0066] y(t)=[F x (t),F y (t),F z (t),AE(t),T(t),…] T ;

[0067] Where: y(t) refers to the observation vector generated at time t; F x (t),F y (t),F z (t) refers to the components of the cutting force in the x, y, and z directions collected at time t; AE(t) refers to a characteristic value of the acoustic emission signal collected at time t, such as its energy or effective value; T(t) refers to the representative temperature value of the cutting area collected at time t, such as the tool tip temperature or chip temperature; “…” indicates that the observation vector can contain other signal features that can reflect the physical state of the cutting process, such as vibration signals, tool wear image features, etc., to further enrich the dimension of the observation information.

[0068] S3, inputting the observation vector into a state space model, estimating the state of the cutting interface between the tool and the titanium alloy workpiece, and obtaining a state estimation vector;

[0069] In this embodiment, step S3 inputs the observation vector into the state-space model to estimate the state of the cutting interface between the tool and the titanium alloy workpiece, ultimately generating a state estimation vector. This step aims to extract key physical state information within the cutting interface that is difficult to directly measure from multi-dimensional real-time sensor data.

[0070] Specifically, the state estimation vector is determined to characterize the non-directly measurable physical states of the cutting interface. These physical states are key factors affecting cutting process efficiency, quality, and tool life. Typically, these physical states are comprised of the interface equivalent friction coefficient, heat flux distribution coefficient, and tool wear rate.

[0071] The equivalent interface friction coefficient reflects the frictional resistance between the tool and workpiece during cutting. Changes in its value directly affect the cutting force and heat generation. Because the cutting interface is exposed to high temperatures, high pressures, and a constantly changing dynamic environment, direct measurement of the friction coefficient is extremely difficult.

[0072] Heat flux distribution coefficient: This characterizes the distribution of the total heat generated during cutting between the tool, workpiece, and chip. Understanding heat flux distribution is crucial for controlling cutting temperature, avoiding thermal damage, and optimizing cutting fluid cooling. Similar to the coefficient of friction, the complex transfer mechanism of cutting heat makes its direct measurement challenging.

[0073] Tool wear rate: This indicates the degree of tool wear per unit time and is a key indicator for evaluating tool life and guiding tool replacement strategies. Tool wear is a gradual and complex physical process. Direct, real-time measurement often relies on optical or contact detection, which is difficult to perform continuously during the cutting process.

[0074] In order to achieve accurate estimation of these physical states that cannot be directly measured, in this embodiment, the step of estimating the state of the cutting interface between the tool and the titanium alloy workpiece is implemented by a Kalman filter. The Kalman filter is a recursive state estimation algorithm that can optimally estimate the state of a dynamic system under noisy measurement data. Specifically, the observation vector is used as the input of the Kalman filter and combined with a pre-established state space model. The state space model describes the dynamic behavior of the cutting interface, and its state space model formula is:

[0075] Formula (1): x k+1 =A k x k +B k u k +w k ;

[0076] Formula (2): y k =C k x k +D k u k +v k ;

[0077] Where: x krefers to the system state vector at time k, i.e., the state estimation vector, whose components include the interface equivalent friction coefficient, heat flux distribution coefficient and tool wear rate; u k Refers to the system input vector at time k, which may include high-frequency pulse electric field parameters, cutting parameters (such as cutting speed, feed rate, cutting depth), etc.; y k Refers to the system observation vector at time k, that is, the dynamic physical signal observation vector; A k Refers to the state transfer matrix at time k, which describes the evolution of the state vector from time k to time k+1; B k Refers to the input control matrix at time k, which describes the influence of the input vector on the state vector; C k Refers to the observation matrix at time k, which describes how the state vector is mapped to the observation vector; D k Refers to the input direct matrix at time k, which describes the direct impact of the input vector on the observation vector; w k refers to the process noise at time k, reflecting the uncertainty of the model and the unmodeled disturbance; v k refers to the measurement noise at time k, reflecting the error of sensor measurement.

[0078] The state estimate vector can be calculated accurately and in real time through the prediction and update steps of the Kalman filter. The prediction step uses Equation (1) to predict the current state based on the state estimate at the previous moment and the current input. The update step uses Equation (2) to compare the actual observation data with the predicted observation data and correct the predicted state based on the difference between the two, thereby obtaining a more accurate estimate of the current state.

[0079] In this embodiment, the model parameters of the state-space model are updated online. The online update mechanism is crucial to ensure that the model maintains high accuracy in the face of changes in actual cutting conditions, accumulated tool wear, and environmental disturbances. The steps of online update specifically include:

[0080] The observed output predicted by the state-space model is compared with the generated observation vector to generate a residual. The residual represents the difference between the model predicted value and the actual measured value, which is a direct reflection of the model's inaccuracy.

[0081] Then, the model parameters of the state space model are adjusted according to the generated residual. The model parameters include but are not limited to the state transfer matrix A k , input control matrix B k , observation matrix C k and input pass-through matrix D kSpecifically, the step of adjusting the model parameters of the state-space model is implemented using a recursive least squares algorithm. The recursive least squares algorithm is an online parameter estimation method that can incrementally correct model parameters using new measurement data without reprocessing all historical data. The algorithm combines the residuals with the previous estimate of the model parameters to calculate the current updated values ​​of the model parameters.

[0082] Preferably, the update equation of the recursive least squares algorithm is:

[0083]

[0084] P k =λ -1 (P k-1 -K k H k P k-1 );

[0085] Where: Refers to the model parameter estimation vector at time k; K k Refers to the gain matrix at time k; y k Refers to the observation vector at time k; H k Refers to the regression matrix at time k, which is determined by the model structure and input and output data; P k Refers to the parameter estimation error covariance matrix at time k; λ refers to the forgetting factor, which is used to adjust the weight of the influence of historical data on the current parameter estimation. It usually takes a value between 0 and 1, such as 0.95 to 0.99, and is used for online adaptation of time-varying systems.

[0086] S4. Calculating a feedback control instruction for adjusting the high-frequency pulse electric field based on a deviation between the state estimation vector and a preset reference state vector;

[0087] In this embodiment, step S4 calculates feedback control instructions for adjusting the high-frequency pulsed electric field based on the deviation between the state estimation vector obtained in S3 and a preset reference state vector. This is the core step in achieving closed-loop control of the cutting process. Its purpose is to drive the actual cutting interface state to the desired ideal state by actively adjusting the electric field parameters.

[0088] Specifically, the preset reference state vector represents the optimal physical state of the cutting interface under specific processing requirements, for example, an ideal combination of interface equivalent friction coefficient, heat flux distribution coefficient and tool wear rate that can take into account both processing efficiency and tool life.

[0089] To achieve precise and robust control of the complex dynamic system of the cutting process, in this embodiment, the feedback control command calculation step is implemented using a model predictive control (MPC) algorithm. MPC is an advanced control strategy whose core concept is to use a dynamic process model to predict the system's behavior over a period of time and, based on this prediction, to solve the optimal control problem.

[0090] In one possible implementation, the model predictive control algorithm determines the current control action within each control cycle by solving an online, finite-horizon open-loop optimization problem. The optimization problem aims to minimize the predicted deviation between the state estimate vector and a preset reference state vector over the future prediction horizon, while also taking into account the energy consumption or rate of change of the control input to achieve smooth control.

[0091] In general, the objective function of the optimization problem is as follows:

[0092]

[0093] Where: J(k) refers to the objective function value at the current time k; N p Refers to the length of the prediction horizon, that is, the number of steps the controller predicts ahead; N c Refers to the length of the control time domain, that is, the length of the future control input sequence calculated by the controller, generally N c ≤N p ; x(k+i|k) refers to the predicted value of the state vector at the current time k for the future time k+i, which is based on the state space model established and updated online in S3; ref (k+i) refers to the preset reference state vector at the future time k+i; Δu(k+i|k) refers to the control input increment at the future time k+i calculated at the current time k, that is, u(k+i)-u(k+i-1), and the control input u corresponds to the parameters of the high-frequency pulse electric field, such as voltage or frequency; Q and R are positive semidefinite and positive definite weight matrices, respectively, which are used to adjust the degree of penalty for state tracking error and control input change, and are adjustable parameters in controller design.

[0094] The solution to the optimization problem satisfies constraints that reflect the practical limitations of the physical system. These constraints may optionally include:

[0095] System dynamic model constraints:

[0096] x(k+i+1|k)=A k x(k+i|k)+B k u(k+i|k);

[0097] Control input constraints:

[0098] u min ≤u(k+i|k)≤u max ;

[0099] Control input increment constraint:

[0100] Δu min ≤Δu(k+i|k)≤Δu max ;

[0101] State variable constraints:

[0102] x min ≤x(k+i|k)≤x max ;

[0103] By solving the above constrained optimization problem, we can obtain a control domain N in the future. c The optimal control sequence ΔU within * ={Δu(k|k),Δu(k+1|k),…,Δu(k+N c -1|k)} T .

[0104] Based on the model predictive control's rolling horizon optimization strategy, only the first control instruction of the optimal control sequence, Δu(k|k), is used to calculate the actual control action at the current moment. The current feedback control instruction, u(k), is the sum of the previous control instruction, u(k-1), and the currently calculated optimal control increment, Δu(k|k). This instruction is output to step S5 for adjusting the high-frequency pulse energy source.

[0105] In the next control cycle, at time k+1, the system collects a new observation vector, updates the state estimate, and re-solves the optimization problem based on this new state information, obtaining a new optimal control sequence and applying its first element. This cycle repeats, achieving continuous closed-loop control of the cutting interface state.

[0106] S5. According to the calculated feedback control instruction, the parameters of the high-frequency pulse electric field are adjusted in real time by a high-frequency pulse energy source;

[0107] In this embodiment, step S5 involves adjusting the parameters of the high-frequency pulsed electric field in real time via the high-frequency pulse energy source based on the calculated control instructions. This is the execution end of the entire closed-loop control circuit, converting the decision instructions generated in the previous steps into physical interventions in the cutting process.

[0108] In order to achieve both rapid response to foreseeable disturbances and accurate correction of unforeseeable deviations, this embodiment adopts a composite control strategy combining feedforward and feedback.

[0109] Specifically, before executing the cutting process (i.e., step S1), the system will pre-calculate the feedforward control instructions. The calculation is based on a preset cutting process path and uses the state space model established in S3. The preset cutting process path is usually derived from a CNC program (such as G code), which contains known and upcoming change information such as cutting depth, feed rate, tool trajectory, etc. By taking these known path information as input and using the state space model for simulation prediction, it is possible to estimate how the cutting interface state (such as the interface equivalent friction coefficient, heat flux distribution coefficient, etc.) will evolve in the absence of active intervention. The feedforward control instruction is a compensatory control quantity calculated in advance to offset these foreseeable deviations from the ideal state caused by changes in the processing path.

[0110] In one possible implementation, the calculation of the feedforward control command aims to make the predicted state trajectory as close as possible to the reference state trajectory.

[0111] During the cutting process, step S5 superimposes the above-mentioned pre-calculated feedforward control instructions with the feedback control instructions calculated in real time by the model predictive control algorithm in S4 to generate a composite control instruction.

[0112] In general, the generation process of the compound control instruction can be expressed as follows:

[0113] u comp (k)=u ff (k)+u fb (k);

[0114] Where: u comp (k) refers to the composite control instruction generated at time k; u ff (k) refers to the pre-calculated feedforward control command corresponding to time k; u fb (k) refers to the feedback control instruction calculated by step S4 at time k, that is, the first control instruction of the optimal control sequence output by the model predictive control algorithm.

[0115] In this way, the feedforward control part can quickly respond to known disturbances such as changes in the cutting path, providing the basic control quantity for the system, while the feedback control part accurately compensates and corrects the deviations caused by model uncertainty, unmodeled dynamics, and external unknown disturbances based on the real-time state estimation results.

[0116] Finally, the system adjusts the parameters of the high-frequency pulse electric field in real time through the high-frequency pulse energy source according to the generated composite control instruction. Each component of the composite control instruction vector will be correspondingly parsed into specific adjustment values ​​for the output parameters of the high-frequency pulse energy source, such as adjusting the amplitude of the output voltage, the frequency of the pulse, the pulse width or the duty cycle, etc. By adjusting these electric field parameters in real time and accurately, the characteristics of the plasma layer generated at the cutting interface (such as density, thickness, temperature, etc.) can be directly affected, thereby actively regulating the friction and heat transfer behavior of the cutting interface, and ultimately achieving the goal of maintaining the cutting interface state near the desired reference value.

[0117] See also Figure 2 The present invention also provides a high-efficiency cutting system for automotive titanium alloy parts, which includes the following modules:

[0118] A high-frequency pulse energy source is used to apply a high-frequency pulse electric field between the tool and the titanium alloy workpiece, and to adjust the parameters of the high-frequency pulse electric field in real time according to feedback control instructions;

[0119] The energy source is electrically connected to the cutting tool and the titanium alloy workpiece, applying a high-frequency pulsed electric field between them. The energy source receives control instructions from a feedback control module and adjusts its output electric field parameters, such as voltage, frequency, and pulse width, in real time based on these instructions, thereby actively regulating the physical properties of the plasma layer generated at the cutting interface.

[0120] A cutting fluid supply module, used for supplying cutting fluid to the cutting area;

[0121] It is used to stably supply cutting fluid to the cutting area. The cutting fluid not only serves as a cooling and lubricating medium, but more importantly, it serves as a working medium that undergoes phase change to generate plasma under the action of the high-frequency pulse electric field, and works in synergy with the high-frequency pulse energy source.

[0122] A multi-physics field sensing module, used for collecting dynamic physical signals between the tool and the titanium alloy workpiece in real time during the cutting process to generate an observation vector;

[0123] The multi-physics field sensing module includes a force sensor for collecting mechanical signals, an acoustic emission sensor for collecting microscopic fracture signals, and an infrared thermal imager for collecting thermal signals;

[0124] During the cutting process, the module collects dynamic physical signals between the tool and the titanium alloy workpiece in real time to generate an observation vector. The multi-physics sensing module integrates a set of sensors, including a force sensor for collecting cutting force signals that characterize the mechanical properties of the cutting process; an acoustic emission sensor for collecting acoustic emission signals that characterize the microscopic fracture characteristics of the material; and an infrared thermal imager for collecting temperature field signals that characterize the thermal properties of the cutting area. This module processes and fuses these various synchronously collected signals to form an observation vector that comprehensively reflects the state of the cutting interface.

[0125] a state estimation module, configured to receive the observation vector and estimate the state of the cutting interface between the tool and the titanium alloy workpiece based on a state space model to generate a state estimation vector;

[0126] The module is configured to receive observation vectors from the multi-physics sensing module and estimate the state of the cutting interface between the tool and the titanium alloy workpiece based on a state-space model to generate a state estimation vector. The state estimation vector characterizes physical states that cannot be directly measured, such as the interface equivalent friction coefficient, heat flux distribution coefficient, and tool wear rate. The module preferably performs state estimation using a Kalman filter and integrates a recursive least squares algorithm for online updating of the parameters of the state-space model to adapt to changes in cutting conditions.

[0127] a feedback control module, configured to calculate the feedback control instruction based on a deviation between the state estimation vector and a preset reference state vector, and send the feedback control instruction to the high-frequency pulse energy source;

[0128] The module is configured to calculate the feedback control instructions based on the deviation between the state estimation vector generated by the state estimation module and a preset reference state vector. This module preferably uses a model predictive control algorithm to solve an optimization problem online to obtain an optimal control instruction sequence. The feedback control module transmits the calculated feedback control instructions to the high-frequency pulse energy source, thereby forming a closed-loop control system to accurately maintain the actual physical state of the cutting interface at the desired reference state.

[0129] While embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions, and variations may be made to these embodiments without departing from the principles and spirit of the invention, and that the scope of the invention is defined by the appended claims and their equivalents.

Claims

1. A high-efficiency cutting method for titanium alloy parts for automobiles, characterized in that: The method comprises the following steps: S1, applying a high-frequency pulse electric field between the tool and the titanium alloy workpiece, and supplying cutting fluid to the cutting area to perform cutting processing; S2. During the cutting process, a multi-physics field sensing module is used to collect dynamic physical signals between the tool and the titanium alloy workpiece in real time to generate an observation vector; S3, inputting the observation vector into a state space model, estimating the state of the cutting interface between the tool and the titanium alloy workpiece, and obtaining a state estimation vector; S4. Calculating a feedback control instruction for adjusting the high-frequency pulse electric field based on a deviation between the state estimation vector and a preset reference state vector; S5. According to the calculated feedback control instruction, the parameters of the high-frequency pulse electric field are adjusted in real time by a high-frequency pulse energy source.

2. The high-efficiency cutting method for automotive titanium alloy parts according to claim 1, characterized in that: In step S1, the steps of applying a high-frequency pulse electric field between the tool and the titanium alloy workpiece and supplying cutting fluid to the cutting area for cutting processing include: Through the synergistic effect of the high-frequency pulse electric field and the cutting fluid, the cutting fluid is induced to undergo a phase change at the cutting interface between the tool and the titanium alloy workpiece to generate a plasma layer with both lubrication and heat insulation properties.

3. The high-efficiency cutting method for automotive titanium alloy parts according to claim 1, characterized in that: In step S2, the step of collecting dynamic physical signals between the tool and the titanium alloy workpiece in real time through a multi-physics field sensing module includes: The signals representing the mechanical properties of the cutting process, the signals representing the microscopic fracture properties of the material, and the signals representing the thermal properties are collected to form the dynamic physical signals.

4. The high-efficiency cutting method for automotive titanium alloy parts according to claim 1, characterized in that: In step S3, the step of inputting the observation vector into the state space model to estimate the state of the cutting interface between the tool and the titanium alloy workpiece includes: The state estimation vector is determined to characterize the physical state of the cutting interface that cannot be directly measured, where the physical state consists of an interface equivalent friction coefficient, a heat flux distribution coefficient, and a tool wear rate.

5. The high-efficiency cutting method for automotive titanium alloy parts according to claim 4, characterized in that: In step S3, the step of estimating the state of the cutting interface between the tool and the titanium alloy workpiece includes: The state estimation vector is calculated by using a Kalman filter with the observation vector as input and combined with the state space model.

6. The high-efficiency cutting method for automotive titanium alloy parts according to claim 1, characterized in that: In step S4, the step of calculating and obtaining a feedback control instruction for adjusting the high-frequency pulse electric field based on the deviation between the state estimation vector and the preset reference state vector includes: An optimization problem is solved by a model predictive control algorithm, wherein the optimization problem aims to minimize the prediction deviation between the state estimation vector and the preset reference state vector in the future prediction time domain, thereby obtaining an optimal control sequence, and using the first control instruction of the optimal control sequence as the feedback control instruction.

7. The high-efficiency cutting method for automotive titanium alloy parts according to claim 1, characterized in that: Before executing step S1, pre-calculating feedforward control instructions according to a preset cutting process path and using the state space model; In step S5, the step of adjusting the parameters of the high-frequency pulse electric field in real time by a high-frequency pulse energy source according to the calculated feedback control instruction includes: The feedforward control instruction and the feedback control instruction are superimposed to generate a composite control instruction, and the parameters of the high-frequency pulse electric field are adjusted in real time according to the composite control instruction.

8. The high-efficiency cutting method for automotive titanium alloy parts according to claim 1, characterized in that: In step S3, the model parameters of the state-space model are updated online, and the online updating step includes comparing the observation output predicted by the state-space model with the observation vector to generate a residual, and adjusting the model parameters of the state-space model according to the generated residual.

9. The high-efficiency cutting method for automotive titanium alloy parts according to claim 8, characterized in that: The step of adjusting the model parameters of the state-space model includes: The updated value of the model parameter at the current moment is calculated by combining the residual and the estimated value of the model parameter at the previous moment through a recursive least squares algorithm.

10. An efficient cutting system for titanium alloy parts for automobiles, applied to the method according to any one of claims 1 to 9, characterized in that: The system includes the following modules: A high-frequency pulse energy source is used to apply a high-frequency pulse electric field between the tool and the titanium alloy workpiece, and to adjust the parameters of the high-frequency pulse electric field in real time according to feedback control instructions; A cutting fluid supply module, used for supplying cutting fluid to the cutting area; A multi-physics field sensing module, used for collecting dynamic physical signals between the tool and the titanium alloy workpiece in real time during the cutting process to generate an observation vector; The multi-physics field sensing module includes a force sensor for collecting mechanical signals, an acoustic emission sensor for collecting microscopic fracture signals, and an infrared thermal imager for collecting thermal signals; a state estimation module, configured to receive the observation vector and estimate the state of the cutting interface between the tool and the titanium alloy workpiece based on a state space model to generate a state estimation vector; A feedback control module is used to calculate the feedback control instruction based on the deviation between the state estimation vector and a preset reference state vector, and send the feedback control instruction to the high-frequency pulse energy source.

Images