Detailed Description
The following description of the embodiments of the present invention will be made clearly and fully with reference to the accompanying drawings, in which it is evident that the embodiments described are some, but not all embodiments of the invention. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
The flow diagrams depicted in the figures are merely illustrative and not necessarily all of the elements and operations/steps are included or performed in the order described. For example, some operations/steps may also be split, combined, or partially combined, so that the order of actual execution may vary based on actual circumstances.
It is also to be understood that the terminology used in the description of the application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in this specification and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should be further understood that the term "and/or" as used in the present specification and the appended claims refers to any and all possible combinations of one or more of the associated listed items, and includes such combinations.
Some embodiments of the present application are described in detail below with reference to the accompanying drawings. The following embodiments and features of the embodiments may be combined with each other without conflict.
Referring to fig. 1, fig. 1 is a flowchart illustrating a method for controlling output power of a carbon dioxide laser therapeutic apparatus according to an embodiment of the present application, and as shown in fig. 1, the method for controlling output power of a carbon dioxide laser therapeutic apparatus according to an embodiment of the present application includes steps S100 to S600.
Step S100, finite element modeling is carried out on a light path, a laser tube and a cooling system of the carbon dioxide laser therapeutic instrument, and a three-dimensional temperature field model is obtained;
It is to be understood that the execution body of the present invention may be an output power control system of a carbon dioxide laser therapeutic apparatus, and may also be a terminal or a server, which is not limited herein. The embodiment of the invention is described by taking a server as an execution main body as an example.
Specifically, the geometric structure of the carbon dioxide laser therapeutic apparatus is subjected to parameterization modeling to obtain a complete three-dimensional geometric model comprising a light path, a laser tube and a cooling system, and each physical component and the interrelation thereof in the laser are reflected. According to the three-dimensional geometric model, the space inside the laser is divided into a plurality of finite element grid cells, and a finite element grid consisting of the plurality of grid cells is formed. Each grid cell represents a small portion of the geometric model, and the physical behavior of the overall system is approximately modeled by analyzing the collection of grid cells. The thermophysical parameters of the material, including thermal conductivity, specific heat capacity and density, are assigned to each grid cell to form a matrix containing these parameters reflecting the thermal conductivity characteristics of the material under different conditions so that the model simulates the thermal conductivity behavior of the laser during operation. And according to the working parameters of the laser and the parameters of the cooling system, applying a heat flow boundary condition to the boundary of the finite element grid, and simulating the heat flow and heat dissipation process of the laser in operation to obtain a complete heat conduction model. A system of partial differential equations of temperature over time is established for each grid cell in the thermal conduction model, by which equations the temperature change law of each cell is described. And performing time discretization according to the partial differential equation set to obtain temperature distribution data corresponding to each time step. And calculating thermal stress and thermal deformation based on the temperature distribution data to obtain a stress-strain matrix of the structural change of the laser, and describing the deformation and internal stress distribution condition of the laser under the action of heat. And (3) performing thermal-optical coupling analysis on the stress-strain matrix, combining thermodynamic and optical effects, and analyzing how the temperature change influences the optical performance of the laser to generate a three-dimensional temperature field model.
Step S200, continuous wave mode analysis is carried out on the three-dimensional temperature field model, and continuous wave power fluctuation influence data are obtained;
Specifically, the three-dimensional temperature field model is initialized to obtain initial temperature distribution in a continuous wave mode, so that the temperature distribution condition of the laser in a stable working state is established. And carrying out heat source loading calculation on the initial temperature distribution based on the working parameters of the continuous wave laser so as to simulate the heat input of the laser in the working process, obtain steady-state temperature distribution and reflect the temperature distribution state of the laser in a continuous wave mode. And carrying out Fourier transform on the steady-state temperature distribution to obtain the frequency spectrum characteristic of the temperature field. Through spectrum analysis, the change characteristics of the temperature field under different frequencies are revealed, and then the temperature field is subjected to modal decomposition to obtain a target temperature mode. Extraction of the temperature modes helps to identify the most critical temperature distribution modes for laser performance, which directly affect the power output and beam quality of the laser. And performing sensitivity analysis on the target temperature mode to obtain a response function of temperature to power fluctuation, and describing how the power output of the laser is affected by temperature change. Based on the response function, power fluctuation with different amplitudes and frequencies is simulated, and dynamic temperature distribution is obtained. The dynamic temperature profile reflects the transient response characteristics of the temperature field in the event of power fluctuations. And mapping the dynamic temperature distribution and the optical parameters to obtain the optical characteristic changes including the temperature of the gain medium and the parameters of the resonant cavity. And performing optical field calculation on the output of the laser through optical characteristic change to obtain change data of output power, wavelength and beam quality, and reflecting the optical performance of the laser in different working states. And carrying out statistical analysis and spectrum analysis on the change data to obtain the amplitude spectrum and the phase spectrum of the power fluctuation. And evaluating the influence of the power fluctuation on the performance of the laser through the amplitude spectrum and the phase spectrum, thereby comprehensively evaluating the sensitivity and the stability of the power fluctuation and obtaining continuous wave power fluctuation influence data containing fluctuation sensitivity and stability thresholds.
Step S300, carrying out pulse mode analysis on the three-dimensional temperature field model to obtain pulse power fluctuation influence data;
Specifically, the three-dimensional temperature field model is subjected to time discretization processing to obtain a high-time-resolution temperature field model suitable for a pulse mode. The transient nature of pulsed lasers requires that the model be able to accurately capture temperature changes on an extremely short time scale. And loading a pulse heat source to the model according to the working parameters of the pulse laser, and simulating the temperature change condition in a single pulse period to obtain the transient temperature distribution of the single pulse period. And performing fast Fourier transform on the transient temperature distribution to acquire the time-frequency characteristic of the temperature field. The time-frequency characteristic analysis can reveal the change rule of the temperature in the time and frequency dimensions, and is helpful for understanding the thermodynamic behavior in the pulse mode. Based on the time-frequency characteristic, the principal component analysis is carried out on the temperature field, and a target temperature change mode under the pulse mode is extracted from the principal component analysis. The target temperature change pattern represents the most representative and influential trend in the temperature field under pulsed operating conditions. And performing thermal-optical coupling analysis on the target temperature change mode to obtain an influence function of the temperature on pulse energy, peak power and pulse width. And according to the influence function, simulating different pulse parameters to obtain the temperature accumulation effect in the pulse sequence. The temperature accumulation effect describes the effect of a gradual increase in temperature under successive pulses and the effect on subsequent pulses, which helps to understand the thermal interactions between the pulses. And performing nonlinear fitting on the temperature accumulation effect, generating a mathematical model of heat accumulation among pulses, describing the heat effect accumulation among the pulses, and further performing pulse train simulation on laser output to simulate pulse energy stability and pulse shape change data under different operating conditions. And performing wavelet transformation on the pulse energy stability and pulse shape change data to obtain multi-scale fluctuation characteristics. The wavelet transform enables capturing detailed features of the pulse fluctuations on different time scales, which helps to fully evaluate the impact of the pulse power fluctuations. And obtaining pulse power fluctuation influence data comprising a pulse stability index and a heat accumulation threshold value through comprehensive evaluation of the multi-scale fluctuation characteristics.
Step S400, comprehensively processing continuous wave power fluctuation influence data and pulse power fluctuation influence data and performing dual-mode feedforward control analysis to obtain feedforward control parameters;
Specifically, data fusion is carried out on continuous wave power fluctuation influence data and pulse power fluctuation influence data, so that a power fluctuation feature matrix is obtained, and the influence of continuous wave and pulse fluctuation on laser output is reflected. Based on the power fluctuation feature matrix, cluster analysis is carried out on the continuous wave mode and the pulse mode, and feature boundaries under the two modes are identified. The feature boundary reveals the difference and commonality of the two modes under different conditions, which is the key for realizing mode switching. Blurring is performed on the mode feature boundaries to identify transition zones between the continuous wave mode and the pulse mode. The transition interval represents the overlap region that may occur when the laser is switched from one mode to another under different operating conditions. In order to ensure that the laser stably transits between different modes, a dual-mode switching function is constructed according to a transition interval, a smooth transition strategy of mode switching is generated, abrupt power fluctuation of the laser during mode switching is avoided, and continuity and stability of output are ensured. And carrying out state space modeling on the smooth transition strategy to obtain a state equation of the dual-mode system, and describing the dynamic behavior of the laser in continuous wave and pulse modes. Based on the state equation, a linear quadratic regulator is designed to find the optimal control law. The optimal control law is the core for ensuring the best performance of the laser in different modes, and the control effect is maximized by balancing the stability and the response speed of the system. In order to enable the optimal control law to be applied to an actual digital control system, discretization processing is carried out on the optimal control law, and a differential equation suitable for digital control is obtained. A feedforward compensator is constructed based on the differential equation. The main function of the feedforward compensator is to pre-calculate and compensate for the effects due to temperature and power fluctuations, thereby ensuring the stability of the output. The calculation of the compensation amount is realized by a compensation amount calculation function, and the compensation amount is dynamically adjusted according to the real-time power fluctuation condition. The compensation amount calculation function is adaptively adjusted to cope with different working conditions and environmental changes. By means of the adaptive adjustment, an adaptive compensator is generated, which can automatically optimize the control parameters according to the actual situation. The gain matrix and the compensation coefficient are combined to generate feedforward control parameters including these optimization parameters. The feedforward control parameters are applied to a control system of the laser, so that the accurate control and stability of the laser output power are ensured under different operation modes, and a more efficient treatment effect and higher safety are realized.
Step S500, performing multi-local control calculation on real-time laser output data of the carbon dioxide laser therapeutic instrument to obtain feedback control parameters;
Specifically, real-time laser output data of the carbon dioxide laser therapeutic apparatus is subjected to sliding window segmentation processing, and continuous time series data are divided into a plurality of local time series, wherein each local time series represents the change condition of laser output in a specific time period. Performing fast fourier transform on the plurality of local time sequences to extract spectral features of each of the local and reveal frequency components of the laser output over different time periods. And carrying out adaptive filtering processing on each local time sequence based on the local frequency spectrum characteristics. The self-adaptive filtering can effectively remove noise, and clearer and more accurate local output data are obtained. And carrying out nonlinear system identification on the denoised local data, and establishing a dynamic model in each local time period. The local dynamic model reflects the response characteristics of the laser output over a particular period of time. And designing a plurality of local PID controllers according to the local dynamic model, wherein each controller corresponds to one local dynamic model. By designing these controllers, a set of local PID parameter sets is obtained. And (3) carrying out fuzzy rule mapping on the local PID parameter set to obtain a global PID parameter adjustment strategy, optimizing the control effect in a global scope, and ensuring the cooperation and consistency among different local controllers. And constructing a nonlinear PID control model according to the global PID parameter adjustment strategy. The model can comprehensively control the laser output in a global range, and ensures that the output power of the laser can be kept stable under different operating conditions. Stability analysis is performed on the nonlinear PID control model to verify its reliability in practical applications. And obtaining Lyapunov stability conditions through stability analysis. According to the Lyapunov stability condition, an online parameter adjustment mechanism is generated to adjust control parameters in real time in the laser operation process. And performing numerical discretization on the online parameter adjustment mechanism to obtain feedback control parameters comprising a gain matrix, integral time and differential time. These feedback control parameters are used in the control system of the laser to ensure accurate control of the laser output power, achieving high efficiency, safety and stability during treatment.
And S600, carrying out multi-layer neural network analysis and laser energy distribution on the feedforward control parameters and the feedback control parameters to obtain a multi-stage laser energy distribution scheme.
Specifically, the feedforward control parameter and the feedback control parameter are spliced to form a 30-dimensional input vector X containing a gain matrix, a compensation coefficient and PID parameters. The 30-dimensional input vector X is input to the first hidden layer of the neural network. In this hidden layer, computation is performed by matrix multiplication w1x+b1, where W1 is a weight matrix of 64X30 and b1 is a 64-dimensional bias vector. Through calculation, a 64-dimensional intermediate result is obtained, then the result is processed through a ReLU activation function, the ReLU function sets a negative value part to be zero, and a positive value part is reserved, so that the 64-dimensional first-layer characteristic representation H1 is obtained. The 64-dimensional feature representation H1 is input to a second hidden layer, where it is also calculated by matrix multiplication w2h1+b2, where W2 is a weight matrix of 128x64 and b2 is a 128-dimensional bias vector. And obtaining a 128-dimensional intermediate result through calculation, and then processing through a ReLU activation function to generate a 128-dimensional second-layer characteristic representation H2. The 128-dimensional second layer feature representation H2 is input to the third hidden layer and again calculated by matrix multiplication w3h2+b3, where W3 is a weight matrix of 256x128 and b3 is a 256-dimensional bias vector. And obtaining a 256-dimensional intermediate result through calculation, and processing through a ReLU activation function to generate a 256-dimensional third-layer characteristic representation H3. In order to prevent overfitting, dropout processing is carried out on the third-layer characteristic representation H3, 50% of elements are randomly set to be zero, 256-dimensional noise reduction characteristic representation H3' is obtained, complexity of a neural network is effectively reduced, and generalization capability of a model is enhanced. The 256-dimensional noise reduction feature representation H3 'is input to the output layer and calculated by matrix multiplication W4H3' +b4, where W4 is a weight matrix of 10x256 and b4 is a 10-dimensional bias vector. And obtaining an initial laser energy distribution result through calculation. A Softmax function is applied to the initial laser energy distribution results to generate a probability distribution of the multi-stage energy distribution. The Softmax function converts the result into a probability distribution representing the corresponding energy distribution ratio for each phase. And generating multiple groups of candidate energy distribution schemes by using an inverse transformation sampling method according to the generated multi-stage energy distribution probability distribution, wherein each group of schemes comprises specific energy values of corresponding stages. The candidate schemes are further analyzed, wherein the safety evaluation and the calculation of the energy utilization efficiency are carried out on each group of candidate schemes, so that the selected schemes are safe and reliable, and the energy utilization efficiency can be maximized. And the scheme with the highest energy utilization efficiency is selected as a final multi-stage laser energy distribution scheme, so that the optimal distribution of laser energy is realized, and the maximization and safety of the treatment effect are ensured.
In the embodiment of the invention, the optical path, the laser tube and the cooling system of the carbon dioxide laser therapeutic instrument are subjected to finite element modeling to obtain a high-precision three-dimensional temperature field model, the continuous wave mode and the pulse mode are respectively subjected to deep analysis to obtain power fluctuation influence data under different working states, and the influence data of the continuous wave mode and the pulse mode are comprehensively processed to realize dual-mode feedforward control and solve the problem of power stability during switching among different working modes. The real-time laser output data is processed by utilizing the multi-local control calculation method, so that the feedback control parameters which are dynamically adjusted are obtained, and the resistance of the system to external interference is improved. The multi-layer neural network is adopted to analyze feedforward and feedback control parameters, so that intelligent distribution of laser energy is realized, and the energy utilization efficiency is optimized. By an online parameter adjustment mechanism, the self-adaptive optimization of control parameters is realized, and the adaptability of the system under different working conditions is enhanced. And the feedforward compensation and the nonlinear PID control are combined, so that the control accuracy and stability of the output power are obviously improved. And a safety evaluation mechanism is introduced in the energy distribution process, so that the safety and reliability of laser treatment are ensured. The control method has stronger flexibility through fuzzy rule mapping and multi-stage energy distribution.
In a specific embodiment, the process of executing step S100 may specifically include the following steps:
carrying out parameterized modeling on the geometric structure of the carbon dioxide laser therapeutic instrument to obtain a three-dimensional geometric model comprising a light path, a laser tube and a cooling system;
according to the three-dimensional geometric model, carrying out grid division on the internal space of the laser to obtain a finite element grid consisting of a plurality of grid units;
Distributing material thermophysical parameters to each grid unit to obtain a parameter matrix containing heat conductivity, specific heat capacity and density, and applying heat flow boundary conditions to the boundary of the finite element grid according to the working parameters of the laser and the parameters of a cooling system to obtain a heat conduction model;
Establishing a partial differential equation set of temperature change along with time for each grid unit in the heat conduction model, and performing time discretization according to the partial differential equation set to obtain temperature distribution data of each time step;
And calculating thermal stress and thermal deformation according to the temperature distribution data to obtain a stress-strain matrix of the laser structure change, and performing thermal-optical coupling analysis on the stress-strain matrix to generate a three-dimensional temperature field model.
Specifically, the geometric structure of the carbon dioxide laser therapeutic instrument is parameterized and modeled. The geometrical modeling is performed on each key part of the laser, such as a light path, a laser tube, a cooling system and the like, by using a parameterized design tool. The parameterized modeling has the advantage that various parameters in the model can be flexibly adjusted and optimized, so that the design can adapt to different requirements and conditions. For example, the length, diameter, and size of the fluid channels in the cooling system of the laser tube can be flexibly adjusted by parametric modeling. And meshing the internal space of the laser. By dividing the entire geometric model into a plurality of small grid cells, local details in the model can be better captured. Typically, critical areas of the laser, such as the laser tube inner wall, mirror areas in the optical path, etc., are more finely meshed to ensure that thermal effects and stresses in these areas can be accurately modeled. For example, in the area of the inner wall of the laser tube, a smaller mesh size is selected to capture small variations in heat transfer. After the grid division is completed, the thermophysical parameters of the material are distributed to each grid cell. These parameters include thermal conductivity (k), specific heat capacity (c) and density (ρ), and these parameter matrices reflect the response behavior of the material under different thermal environments. Thermal conductivity is a parameter describing the ability of a material to conduct heat, specific heat capacity is the amount of heat required to raise a unit temperature per unit mass of a material, and density is the ratio of the mass to the volume of the material. For example, for laser tube materials, quartz glass is selected, which has the characteristics of high thermal conductivity and low specific heat capacity, which is beneficial for rapid heat dissipation. The material of the cooling system may be copper metal, which has a high thermal conductivity capable of efficiently conducting heat to the cooling fluid. The boundary of the finite element mesh is subjected to a thermal flow boundary condition according to the operating parameters of the laser and parameters of the cooling system. The heat flow boundary conditions determine how heat is transferred across the different boundaries during operation of the laser. For example, on the inner wall of the laser tube, the heat generated by the laser will be transferred to the surrounding environment by thermal conduction, so that appropriate heat flow boundary conditions need to be applied in this region. Assuming that the heat flux density of the inner wall of the laser tube is q '', the heat flux boundary conditions are defined as follows:
Wherein T is temperature, n is direction perpendicular to boundary,The heat flux density at the boundary is expressed, and the formula describes the heat conduction amount in the normal direction. After the application of the heat flow boundary conditions is completed, a heat conduction model is constructed. The thermal conduction model describes the heat transfer process for each region inside the laser. In this model, a system of partial differential equations with temperature over time is built up for each grid cell to describe the distribution of heat over the different time nodes. This system of equations generally takes the form of a heat transfer equation:
where ρ is the density of the material, c is the specific heat capacity, T is the temperature, T is the time, k is the thermal conductivity, and Q is the internal heat source term. The equation describes how heat is transferred over different time nodes,And denotes a heat conduction term, and Q denotes a heat source term introduced due to laser irradiation or the like. And performing time discretization on the partial differential equation set, dividing a continuous time axis into a plurality of discrete time steps, and solving temperature distribution data on each time step by a numerical method. For example, the time derivative is discretized using an implicit differencing method, resulting in temperature distribution data at each time step. Thermal stress and thermal deformation are calculated from the temperature distribution data. In thermodynamic analysis, temperature changes typically cause stress and deformation of a material, which can be described by a stress-strain relationship assuming that the material has an elastic modulus E and a coefficient of thermal expansion α, then the thermal stress can be expressed as:
σ=E·α·ΔT;
Where σ is the thermal stress and Δt is the temperature change. The equation describes the magnitude of stress due to temperature changes. And constructing a stress-strain matrix according to the thermal stress data, and further simulating the deformation condition of the laser structure under the action of heat. The thermal-optical coupling analysis is performed on the stress-strain matrix, and the thermodynamic analysis and the optical performance analysis are combined to study how the temperature change affects the optical performance of the laser, such as the beam quality, the output power and the like. By analysis, a complete three-dimensional temperature field model is generated, the temperature distribution of the laser under different working conditions is described, and the influence of thermal stress and thermal deformation on the optical performance of the laser is reflected.
In a specific embodiment, the process of executing step S200 may specifically include the following steps:
Initializing the three-dimensional temperature field model to obtain initial temperature distribution in a continuous wave mode, and carrying out heat source loading calculation on the initial temperature distribution according to the working parameters of the continuous wave laser to obtain steady-state temperature distribution;
Performing Fourier transform on the steady-state temperature distribution to obtain the frequency spectrum characteristic of the temperature field, and performing modal decomposition on the temperature field according to the frequency spectrum characteristic to obtain a target temperature mode;
Performing sensitivity analysis on a target temperature mode to obtain a response function of temperature to power fluctuation, and simulating power fluctuation of different amplitudes and frequencies according to the response function to obtain dynamic temperature distribution;
Performing optical parameter mapping on the dynamic temperature distribution to obtain optical characteristic changes including the temperature of a gain medium and the parameters of a resonant cavity, and performing optical field calculation on the output of the laser according to the optical characteristic changes to obtain change data of output power, wavelength and beam quality;
And carrying out statistical analysis and spectrum analysis on the change data to obtain an amplitude spectrum and a phase spectrum of the power fluctuation, and comprehensively evaluating the influence of the power fluctuation according to the amplitude spectrum and the phase spectrum to obtain continuous wave power fluctuation influence data containing fluctuation sensitivity and stability threshold values.
Specifically, the three-dimensional temperature field model is initialized. The key to initializing the settings is to determine the initial conditions, including the initial temperature of the individual components and the setting of the ambient temperature. During operation of a continuous wave laser, the laser tube and other associated components gradually warm up, and thus the initial temperature distribution tends to be non-uniform, and it may generally be assumed that the laser is at ambient temperature prior to start-up, i.e., the global initial temperature is ambient, such as 20 ℃. According to the initial temperature distribution, heat source loading calculation is carried out on the initial temperature distribution by combining specific working parameters of the continuous wave laser, such as laser power, working frequency and running time, so as to simulate the temperature change process of the laser under continuous working. The core of the heat source loading calculation is to model the heat source distribution of the laser and solve the steady-state temperature distribution through a heat conduction equation. The heat conduction equation may describe the diffusion process of heat in a material, the equation:
Where ρ is the density of the material, c is the specific heat capacity, T is the temperature, T is the time, k is the thermal conductivity, and Q is the heat source term. The left term of the equation describes the temperature change over time, the first term on the right represents the diffusion of heat in space by conduction, and the second term is the direct heating contribution of the heat source to the system. In a continuous wave laser, the heat source term Q is typically generated by laser absorption, the distribution of which can be specifically modeled according to the optical design of the laser. And obtaining steady-state temperature distribution of the laser after a certain time by solving a partial differential equation, and displaying the temperature distribution condition of each part of the laser in a continuous working state. Fourier transforming the steady-state temperature profile to extract the spectral characteristics of the temperature field. Fourier transforms can transform the temperature distribution from the time domain to the frequency domain, revealing the dominant frequency components of the temperature change. The spectral characteristics of the temperature field are critical to understanding the thermal response of the laser at different frequencies and can help identify which frequency components contribute most to the temperature variation. Based on the spectral characteristics of the temperature field, the temperature field is subjected to modal decomposition, and the main modes in the temperature field, namely the modes with the greatest influence on the system behavior, are extracted. These modes can be considered as fundamental vibration modes in the temperature field, each mode corresponding to a typical form of temperature distribution. By modal decomposition, the complex temperature field is reduced to several main modes, thereby making it easier to analyze and understand the thermal behavior of the system. And carrying out sensitivity analysis on the target temperature mode, and determining the response condition of the temperature mode to the power fluctuation of the laser. The sensitivity of temperature to power fluctuations is quantified by a response function. Assuming that the fluctuation of the laser power is denoted as P (T) and the change of the temperature is denoted as T (T), the response function R (T) can be expressed as:
Here, δt (T) represents the temperature change amount, and δp (T) represents the power fluctuation amount. By this function the extent of the influence of different power fluctuations on the temperature field is quantified. Based on the response function, power fluctuation with different amplitudes and frequencies is simulated, and dynamic temperature distribution is obtained. The dynamic temperature profile shows how the laser temperature varies with power fluctuations during actual operation. Optical parameter mapping is performed. Temperature variations can directly affect the optical properties of the laser, such as the refractive index of the gain medium, the resonator length, etc. By mapping the dynamic temperature profile onto the optical parameters, variations in optical characteristics, including gain medium temperature and resonator parameters, are obtained. These optical characteristic variations will directly affect the output performance of the laser, such as output power, wavelength, and beam quality. The output of the laser is subjected to light field calculation. The propagation behavior of the laser in the cavity is simulated by an optical equation, and the changes of the output power, wavelength and beam quality of the laser are predicted. These calculations will reflect how temperature variations affect the output performance of the laser, especially the effect of power fluctuations on beam quality and wavelength stability under different operating conditions. Statistical analysis and spectral analysis are performed on the variation data of output power, wavelength and beam quality. Statistical analysis can help identify the characteristics of fluctuations in the laser output, while spectral analysis can reveal the dominant frequency components of these fluctuations. By analysis, an amplitude spectrum and a phase spectrum of the power fluctuation are obtained, and the spectrums describe the amplitude and phase change conditions of the power fluctuation at different frequencies. And comprehensively evaluating the influence of the power fluctuation according to the amplitude spectrum and the phase spectrum to obtain continuous wave power fluctuation influence data containing fluctuation sensitivity and stability threshold values.
In a specific embodiment, the process of executing step S300 may specifically include the following steps:
Performing time discretization on the three-dimensional temperature field model to obtain a high-time-resolution temperature field model suitable for a pulse mode, and loading a pulse heat source on the high-time-resolution temperature field model according to working parameters of a pulse laser to obtain transient temperature distribution of a single pulse period;
Performing fast Fourier transform on the transient temperature distribution to obtain time-frequency characteristics of a temperature field, and performing principal component analysis on the temperature field according to the time-frequency characteristics to obtain a target temperature change mode in a pulse mode;
Performing thermal-optical coupling analysis on the target temperature change mode to obtain an influence function of temperature on pulse energy, peak power and pulse width, and simulating different pulse parameters according to the influence function to obtain a temperature accumulation effect of a pulse sequence;
Performing nonlinear fitting on the temperature accumulation effect to obtain a mathematical model of heat accumulation among pulses, and performing pulse train simulation on laser output according to the mathematical model to obtain pulse energy stability and pulse shape change data;
And carrying out wavelet transformation on the pulse energy stability and pulse shape change data to obtain a multi-scale fluctuation characteristic, and comprehensively evaluating the pulse power fluctuation influence according to the multi-scale fluctuation characteristic to obtain pulse power fluctuation influence data containing a pulse stability index and a heat accumulation threshold value.
In particular, in the pulse mode, the laser emits high energy laser pulses at extremely short time intervals, resulting in transient temperature changes in the laser tube and surrounding medium. Such transient temperature variations have a significant impact on the thermal management and stability of the laser. In order to capture these rapidly changing temperature characteristics, a time discretization process is performed on the three-dimensional temperature field model, enabling it to simulate the operation of a pulsed laser with sufficiently high time resolution. The core of the time discretization process is to divide the continuous time domain into a plurality of discrete time steps. At each time step, the change in the temperature field is solved. This process is typically implemented using explicit or implicit numerical methods. For pulsed lasers, explicit methods are generally better able to capture rapid temperature changes due to their transient nature. Assuming a time step of Δt, at each time step the evolution of the temperature field can be described by a thermal conduction equation:
Wherein Tn is the temperature distribution at time step n, Tn+1 is the temperature distribution at time step n+1, deltat is the time step, k is the thermal conductivity of the material, ρ is the density, c is the specific heat capacity, and Qn is the heat source term at time step n. By this formula, the temperature distribution is updated stepwise at each time step, resulting in a high time resolution temperature field model suitable for the pulse mode. And loading a pulse heat source to the high-time-resolution temperature field model according to the working parameters of the pulse laser, and simulating transient heat effect generated by each laser pulse in the material. Pulsed lasers typically release a large amount of energy in a very short time, resulting in a rapid rise in local temperature. The heat source term Qn may be defined in terms of specific parameters of the pulsed laser (e.g., pulse energy, pulse width, repetition rate, etc.). For example, a pulsed heat source term may be expressed as a heat source profile with a pulse function over time:
Q(t)=Q0·δ(t-t0);
Where Q0 is the pulse peak heat source intensity and delta (t-t0) is the time pulse function of the pulse at time t0. The transient temperature profile in each pulse period is obtained by this loading process. The transient temperature profile is subjected to a fast fourier transform to analyze the time-frequency characteristics of the temperature field. The fourier transform is able to transform the temperature distribution from the time domain to the frequency domain, revealing the main frequency components of the temperature variation. For example, the fourier transform may help identify which frequency components contribute most to temperature fluctuations, thereby guiding the design and optimization of the laser. And carrying out principal component analysis on the temperature field according to the time-frequency characteristic, and extracting a target temperature change mode under a pulse mode. Principal component analysis is a dimension reduction technique that extracts principal features by identifying the direction in the data where the variance is greatest. In the pulse mode, the target temperature variation pattern represents the temperature distribution pattern that has the greatest influence on the thermal behavior of the system. And (3) carrying out thermal-optical coupling analysis on the target temperature change mode, and researching how the temperature change influences the optical performance of the laser. The change in temperature affects the refractive index inside the laser, the excited state of the gain medium, the length of the resonant cavity, and the like, thereby changing the output characteristics of the laser. The influence function of the temperature on the pulse energy, peak power and pulse width is obtained through the thermo-optical coupling analysis. Assuming that the pulse peak power change due to temperature change Δt is Δp, the influence function can be expressed as:
ΔP=f(ΔT);
Where f (Δt) is a function of the influence of temperature variation on pulse power, describing how temperature fluctuations affect the output power of the laser. Based on the influence function, different pulse parameters are simulated, and the temperature accumulation effect of the pulse sequence is obtained. The temperature accumulation effect describes how the temperature in the material gradually increases under the influence of multiple laser pulses and has an effect on the thermal effects of subsequent pulses. The cumulative effect is particularly important in high repetition rate pulsed lasers, as the accumulation of temperature between pulses can lead to degradation or instability of the laser's performance. And performing nonlinear fitting on the temperature accumulation effect to obtain a mathematical model of heat accumulation between pulses. The mathematical model can be used to predict the magnitude and cumulative speed of the temperature rise of the laser under certain operating conditions. For example, assuming that the cumulative effect can be described by an exponential function, the model can be expressed as:
Tacc(t)＝T0+A·(1-e-βt);
Where Tacc (T) is the cumulative temperature, T0 is the initial temperature, A is the temperature rise, and β is the cumulative rate constant. The model predicts the temperature trend of the laser under long-time operation, thereby guiding the design of the cooling and thermal management system. Based on a mathematical model of the cumulative effect, pulse train simulation is performed on the laser output to obtain data of pulse energy stability and pulse shape variation. The simulation data can reflect the output characteristic change of the laser under different pulse conditions, and is a key basis for optimizing the design and the operation conditions of the laser. The pulse energy stability and pulse shape change data are wavelet transformed. Wavelet transformation can capture varying characteristics of the signal at different time scales, helping to identify multi-scale fluctuation characteristics of the pulse power fluctuations. By analyzing these characteristics, the influence of pulse power fluctuation on the laser performance is evaluated, and pulse power fluctuation influence data containing a pulse stability index and a heat accumulation threshold value is obtained.
In a specific embodiment, the process of executing step S400 may specifically include the following steps:
carrying out data fusion on continuous wave power fluctuation influence data and pulse power fluctuation influence data to obtain a power fluctuation feature matrix, and carrying out cluster analysis on a continuous wave mode and a pulse mode according to the power fluctuation feature matrix to obtain a mode feature boundary;
Carrying out fuzzy processing on the mode characteristic boundary to obtain a transition section of a continuous wave mode and a pulse mode, and constructing a dual-mode switching function according to the transition section to obtain a smooth transition strategy of mode switching;
Modeling a state space of the smooth transition strategy to obtain a state equation of the dual-mode system, and designing a linear quadratic regulator according to the state equation to obtain an optimal control law;
discretizing the optimal control law to obtain a differential equation of digital control, and constructing a feedforward compensator according to the differential equation to obtain a compensation quantity calculation function of temperature and power;
and carrying out self-adaptive adjustment on the compensation quantity calculation function to obtain a compensator, and optimizing the control parameters according to the compensator to obtain feedforward control parameters comprising a gain matrix and compensation coefficients.
Specifically, data fusion is carried out on continuous wave power fluctuation influence data and pulse power fluctuation influence data, and a power fluctuation feature matrix is obtained. The continuous wave mode and the pulse mode show different fluctuation characteristics in a time domain and a frequency domain, and data of the two modes are subjected to standardized processing, so that differences of different characteristics in numerical magnitude are eliminated, and different characteristics can have the same weight during fusion. Assuming that the power fluctuation of the continuous wave mode is characterized by Xcw and the power fluctuation of the pulse mode is characterized by Xpulse, the data fusion can be expressed as a form of a feature matrix X:
X=[Xcw|Xpulse];
Where X is a matrix containing all the power fluctuation features, each column representing a different feature, and each row representing a different observation point. In this way, the characteristics of the continuous wave mode and the pulse mode are combined to form a complete power fluctuation characteristic matrix. And carrying out cluster analysis on the continuous wave mode and the pulse mode, and classifying the continuous wave mode and the pulse mode into two types according to data in a feature matrix, wherein one type represents the features of the continuous wave mode, and the other type represents the features of the pulse mode. Feature boundaries in each mode are identified by a clustering algorithm such as K-means, which describe the feature differences between the two modes. For example, in the K-means cluster, the number of cluster centers is set to 2, and the data is divided to obtain feature boundaries of two modes. And (3) carrying out fuzzy processing on the mode characteristic boundary, and solving the transition problem of mode conversion in actual operation. And blurring the characteristic boundary to generate a transition interval between the continuous wave mode and the pulse mode. The interval may be represented as a fuzzy set describing the situation where the laser may exhibit both modes of characteristics at the same time under different operating conditions. Based on the fuzzy transition interval, a dual-mode switching function is constructed, so that smooth transition of the laser can be realized when the laser is switched from one mode to another mode. Assuming that the switching function is σ (t), its form can be represented by the Sigmoid function:
Where α is a parameter for adjusting the transition smoothness, and t0 is the switching time. In this way, σ (t) exhibits smooth transition characteristics before and after mode switching, avoiding abrupt changes in the laser during mode switching. And carrying out state space modeling on a smooth transition strategy of mode switching, and describing dynamic behaviors of the laser in different modes. Assuming that the system state vector is x (t), the control input is u (t), and the state equation can be expressed as:
Wherein A is a system matrix and B is a control matrix. The state equation describes the dynamic evolution of the system in continuous wave mode and pulse mode. Based on the state equation, a Linear Quadratic Regulator (LQR) is designed to find the optimal control law of the system. The goal of LQR is to optimize system performance by minimizing a quadratic performance index. The performance index may be expressed as:
wherein Q is a state weighting matrix, and R is a control weighting matrix. By solving the optimization problem, an optimal control law u (t) = -Kx (t), where K is the feedback gain matrix, is obtained. In order to enable the application of the optimal control law in a digital control system, discretization processing is carried out on the optimal control law, and the control law of continuous time is converted into a control strategy which can be executed on discrete time steps. Converting the continuous-time state equation into a differential equation by a differential method:
x[k+1]=Adx[k]+Bdu[k];
wherein, Ad and Bd are discretized system matrices. Based on the differential equation, a feed forward compensator is constructed for calculating the temperature and power compensation amounts of the system at each time step. The compensation amount calculation function can be expressed as:
u[k]=u0[k]+Fe[k];
Wherein u0 k is feedforward control quantity, e k is error vector, and F is compensation matrix. And carrying out self-adaptive adjustment on the compensation quantity calculation function to generate a self-adaptive compensator, so that the compensator can dynamically adjust the compensation coefficient according to the actual running condition, and the robustness and the adaptability of the system are improved. For example, when the working environment of the system changes, the compensator can automatically adjust the gain matrix and the compensation coefficient, so that the system can still maintain a stable and efficient working state. And (3) obtaining feedforward control parameters comprising a gain matrix and compensation coefficients through adjustment and optimization of the compensator. These control parameters ensure optimal performance of the laser in various modes of operation, avoid negative effects of power fluctuations on the output, and ensure smooth operation of the system upon mode switching.
In a specific embodiment, the process of performing step S500 may specifically include the following steps:
carrying out sliding window segmentation on real-time laser output data of a carbon dioxide laser therapeutic instrument to obtain a plurality of local time sequences, and carrying out fast Fourier transform on the plurality of local time sequences to obtain local frequency spectrum characteristics;
According to the local spectrum characteristics, performing self-adaptive filtering on each local to obtain denoised local output data, and performing nonlinear system identification on the denoised local output data to obtain a local dynamic model;
According to the local dynamic model, designing a plurality of local PID controllers to obtain a local PID parameter set, and carrying out fuzzy rule mapping on the local PID parameter set to obtain a global PID parameter adjustment strategy;
According to a global PID parameter adjustment strategy, a nonlinear PID control model is constructed, and stability analysis is carried out on the nonlinear PID control model to obtain Lyapunov stability conditions;
And generating an online parameter adjustment mechanism according to the Lyapunov stability condition, and performing numerical discretization on the online parameter adjustment mechanism to obtain a feedback control parameter comprising a gain matrix, an integration time and a differentiation time.
In particular, the continuous laser output signal is divided into a plurality of local time series having a certain time length. The sliding window segmentation generates a plurality of local time sequences by setting a fixed length time window and sliding the window across the entire signal sequence, progressively intercepting a local segment of the signal. Assuming that the original laser output signal is x (t), the sliding window length is L, and the sliding step is Δ, then the i-th local time series can be expressed as:
xi(t)=x(t+iΔ),t∈[0,L];
In this way, a plurality of local time series are generated over the entire time series, which local time series may reflect characteristic variations of the laser output signal at different time periods. And performing fast Fourier transform on the local time sequence to extract local spectrum characteristics. The fast fourier transform can transform a time domain signal into a frequency domain signal, revealing the distribution of the signal over different frequency components. For each local time series, a fast fourier transform is applied to obtain a spectral representation thereof:
Wherein Xi (f) represents the spectrum of the i-th local time series, f is the frequency, and L is the length of the time series. A set of local spectral features is obtained by Fourier transforming each local time series, reflecting the frequency distribution of the laser output signal over different time periods. And according to the local frequency spectrum characteristics, carrying out self-adaptive filtering on each local time sequence, removing noise components in the signal, and simultaneously preserving useful frequency information. The filter coefficient of the adaptive filter can be dynamically adjusted according to the characteristics of the input signal so as to achieve the optimal denoising effect. Assuming that the input signal of the filter is Xi (f), the output of the filter is Yi (f), and the adaptive coefficient of the filter is αi (f), the filtered signal can be expressed as:
Yi(f)＝αi(f)Xi(f);
Wherein, alphai (f) is adjusted according to the local characteristics of the signal, so as to ensure the optimization of the denoising effect. In this way, denoised local output data is obtained. And carrying out nonlinear system identification on the denoised local output data to construct a local dynamic model. By analyzing the input and output data, a mathematical model is built that describes the dynamic behavior of the system. Assuming the local output signal is yi (t) and the input signal is ui (t), the local dynamic model can be expressed as a nonlinear equation:
yi(t)＝f(ui(t),yi(t-1),...,yi(t-n));
Where f is a nonlinear function and n is the memory length of the system. And identifying the local data in different time periods to obtain a plurality of local dynamic models, and reflecting the dynamic characteristics of the laser output in different time periods. Based on the local dynamic model, a plurality of local PID controllers are designed. The PID controller is the most commonly used feedback controller in industrial control, and is characterized in that the accurate control of the system is realized by adjusting three parameters of proportion (P), integral (I) and derivative (D). Assuming that the control output of the ith local dynamic model is ui (t), the control error is ei(t)＝ri(t)-yi (t), where ri (t) is the reference signal, the output of the local PID controller can be expressed as:
Where Kp,i、Ki,i and Kd,i are proportional, integral and differential gains, respectively. By adjusting these gain parameters, the control effect is optimized. In order to realize unified control in the global scope, fuzzy rule mapping is carried out on the local PID parameter set. Fuzzy control is a control strategy based on fuzzy logic, and effective control is realized under the condition of larger uncertainty by defining fuzzy rules. The core of the fuzzy rule mapping is to convert the local PID parameters into a global PID parameter adjustment strategy through fuzzy logic. For example, the performance index of the local controller, such as steady state error and response time, is mapped into fuzzy variables, and then a global control strategy is obtained through fuzzy reasoning. And constructing a nonlinear PID control model according to the global PID parameter adjustment strategy, so that the control strategy can adapt to nonlinear behaviors of the system. Assuming that the nonlinear behavior of the system can be described by a state equation, the nonlinear PID control model can be expressed as:
Where x (t) is a system state vector and A and B are state-dependent nonlinear matrices. Stability of the nonlinear PID control model needs to be verified by Lyapunov stability conditions. The Lyapunov stability condition is an important tool for judging the stability of a nonlinear system. By constructing a Lyapunov function V (x), the system can be proven to be progressively stable if it can be proven to be positive in the global context of the system and its time derivative is semi-negative. Suppose the Lyapunov function is:
V(x)=xTpx;
Wherein, P is positive definite matrix, the stability condition is thatAfter the Lyapunov stability of the system is confirmed, an online parameter adjustment mechanism is generated. And adjusting parameters of the PID controller in real time according to actual running conditions so as to cope with the change of the dynamic characteristics and external interference of the system. In order to enable online adjustment to be realized in a digital control system, the online adjustment is subjected to numerical discretization processing to obtain feedback control parameters comprising a gain matrix, an integration time and a differentiation time.
In a specific embodiment, the process of executing step S600 may specifically include the following steps:
splicing the feedforward control parameters and the feedback control parameters to obtain a 30-dimensional input vector X containing a gain matrix, a compensation coefficient and PID parameters;
Inputting an input vector X into a first hidden layer, calculating by matrix multiplication W1X+b1, wherein W1 is a weight matrix of 64X30, b1 is a 64-dimensional offset vector, obtaining a 64-dimensional intermediate result, and then processing by a ReLU activation function max (0, X), obtaining a 64-dimensional first layer characteristic representation H1;
Inputting the first layer characteristic representation H1 into a second hidden layer, calculating by matrix multiplication W2H21+b2, wherein W2 is a weight matrix of 128x64, b2 is a 128-dimensional offset vector, obtaining a 128-dimensional intermediate result, and obtaining the 128-dimensional second layer characteristic representation H2 by ReLU activation function processing;
Inputting the second layer characteristic representation H2 into a third hidden layer, calculating by matrix multiplication W3H2+b3, wherein W3 is a weight matrix of 256x128, b3 is a 256-dimensional offset vector, obtaining a 256-dimensional intermediate result, and obtaining the 256-dimensional third layer characteristic representation H3 by ReLU activation function processing;
carrying out dropout processing on the third-layer characteristic representation H3, and randomly setting 50% of elements to 0 to obtain 256-dimensional noise reduction characteristic representation H3';
The noise reduction feature is expressed in an H3 'input and output layer, and is calculated through matrix multiplication W4H3' +b4, wherein W4 is a weight matrix of 10x256, and b4 is a 10-dimensional offset vector, so that an initial laser energy distribution result is obtained;
Applying a Softmax function to an initial laser energy distribution result to obtain multi-stage energy distribution probability distribution, and generating a plurality of groups of candidate energy distribution schemes by using an inverse transformation sampling method according to the multi-stage energy distribution probability distribution, wherein each group of schemes comprises energy values of corresponding stages;
and carrying out safety evaluation and energy utilization efficiency calculation on the multiple groups of candidate energy distribution schemes, and selecting the scheme with the highest energy utilization efficiency as a multi-stage laser energy distribution scheme.
Specifically, the feedforward control parameters typically include response time, target output, etc. of the system for predicting the behavior of the system, and the feedback control parameters include error information, actual output, and gain of the PID controller. These parameters need to be arranged in a certain order and converted into the same numerical range by a data preprocessing method, such as normalization or normalization. Assuming that the feedforward control parameter vector is f and the feedback control parameter vector is b, the 30-dimensional input vector X after splicing can be expressed as:
X=[f|b];
Wherein X includes information about all control parameters. The input vector X is input into a first hidden layer of the neural network. The first hidden layer obtains a 64-dimensional intermediate result through matrix multiplication and addition calculation. The input vector X is multiplied by the weight matrix W1 and added with the bias vector b1, and the calculation formula is:
Z1＝W1X+b1;
Where W1 is a 64x30 weight matrix representing the connection strength between each input dimension and each hidden layer neuron, and b1 is a 64-dimensional bias vector providing translational adjustment for each neuron. The calculated Z1 is a 64-dimensional intermediate result, each element of which represents a weighted sum of the input vector and the corresponding weight. To introduce nonlinearity, a ReLU activation function is applied to Z1, and the ReLU is calculated as:
H1＝ReLU(Z1)＝max(0,Z1);
The ReLU function truncates all negative values to 0, leaving positive values unchanged, generating a 64-dimensional first-layer feature representation H1. Nonlinear processing can help the network learn more complex patterns and features. The first layer feature representation H1 is input into the second hidden layer and the calculation process is similar to the first layer. The 128-dimensional intermediate result is obtained through matrix multiplication and addition calculation:
Z2＝W2H1+b2;
Where W2 is a 128x64 weight matrix and b2 is a 128-dimensional bias vector. Re-applying the ReLU activation function to Z2 results in a 128-dimensional second-layer feature representation H2. This process progressively extracts deep features in the input vector X, enabling the neural network to capture more complex patterns and relationships. The second layer feature representation H2 is input to the third hidden layer, and a 256-dimensional intermediate result Z3 is obtained through a similar calculation step:
Z3＝W3H2+b3;
Where W3 is a weight matrix of 256x128 and b3 is a 256-dimensional bias vector. After applying the ReLU activation function to Z3, a 256-dimensional third tier feature representation H3 is generated. To prevent overfitting of the neural network, a dropout process is performed on the third layer feature representation H3. In the dropout process, 50% of the elements are randomly set to 0, which means that only half of the features will be activated in each training, thus reducing the excessive dependence of the model on the particular features. The noise reduction process helps to improve the generalization ability of the model, resulting in a 256-dimensional noise reduction feature representation H'3. The denoised characteristic representation H'3 is input into the output layer. The calculation of the output layer is similar to the previous layers, and the initial laser energy distribution result is obtained through matrix multiplication and addition:
Z4＝W4H′3+b4;
wherein W4 is a weight matrix of 10x256, and b4 is a bias vector of 10 dimensions. At this time, Z4 represents initial values of 10 different energy allocation schemes. To convert the initial values into probability distributions, a Softmax function is next applied. The Softmax function can convert any real vector into a probability vector so that the sum of components is 1, and the specific calculation formula is:
Where Zi is the ith element in Z4, softmax (Zi) represents the probability of the ith energy allocation scheme. The Softmax function yields a probability distribution of the multi-stage energy distribution, which represents the probability of selection among the different energy distribution schemes. Based on the probability distribution, a plurality of sets of candidate energy allocation schemes are generated using an inverse transform sampling method. The inverse transform sampling method is an effective method for generating random samples from a given probability distribution, specifically, the method comprises the steps of generating a uniformly distributed random number, and then finding a corresponding distribution scheme according to a Cumulative Distribution Function (CDF). Each set of candidates contains energy values for the corresponding phase, representing the energy allocation that may be selected in actual operation. After multiple sets of candidate energy allocation schemes are generated, the schemes are subjected to security assessment and energy utilization efficiency calculation. Safety assessment ensures that each candidate does not produce unsafe laser outputs in clinical applications, while energy utilization efficiency calculations measure the efficiency of each regimen under given conditions. By comparing these schemes, the scheme with the highest energy utilization efficiency is selected as the final multi-stage laser energy distribution scheme.
Referring to fig. 2, fig. 2 is a schematic block diagram of an output power control system 200 of a carbon dioxide laser therapeutic apparatus according to an embodiment of the application, and as shown in fig. 2, the output power control system 200 of the carbon dioxide laser therapeutic apparatus includes:
The modeling module 210 is configured to perform finite element modeling on a light path, a laser tube and a cooling system of the carbon dioxide laser therapeutic apparatus to obtain a three-dimensional temperature field model;
the first mode analysis module 220 is configured to perform continuous wave mode analysis on the three-dimensional temperature field model to obtain continuous wave power fluctuation influence data;
The second mode analysis module 230 is configured to perform pulse mode analysis on the three-dimensional temperature field model to obtain pulse power fluctuation influence data;
The feedforward control module 240 is configured to perform comprehensive processing and dual-mode feedforward control analysis on the continuous wave power fluctuation influence data and the pulse power fluctuation influence data to obtain feedforward control parameters;
The calculation module 250 is used for performing multi-local control calculation on the real-time laser output data of the carbon dioxide laser therapeutic apparatus to obtain feedback control parameters;
The allocation module 260 is configured to perform multi-layer neural network analysis and laser energy allocation on the feedforward control parameter and the feedback control parameter to obtain a multi-stage laser energy allocation scheme.
Through the cooperative cooperation of the components, a high-precision three-dimensional temperature field model is obtained by carrying out finite element modeling on a light path, a laser tube and a cooling system of the carbon dioxide laser therapeutic instrument, continuous wave modes and pulse modes are respectively subjected to deep analysis, power fluctuation influence data under different working states are obtained, and the influence data of the continuous wave modes and the pulse modes are comprehensively processed, so that dual-mode feedforward control is realized, and the problem of power stability during switching among different working modes is solved. The real-time laser output data is processed by utilizing the multi-local control calculation method, so that the feedback control parameters which are dynamically adjusted are obtained, and the resistance of the system to external interference is improved. The multi-layer neural network is adopted to analyze feedforward and feedback control parameters, so that intelligent distribution of laser energy is realized, and the energy utilization efficiency is optimized. By an online parameter adjustment mechanism, the self-adaptive optimization of control parameters is realized, and the adaptability of the system under different working conditions is enhanced. And the feedforward compensation and the nonlinear PID control are combined, so that the control accuracy and stability of the output power are obviously improved. And a safety evaluation mechanism is introduced in the energy distribution process, so that the safety and reliability of laser treatment are ensured. The control method has stronger flexibility through fuzzy rule mapping and multi-stage energy distribution.
The application also provides an output power control device of the carbon dioxide laser therapeutic apparatus, which comprises a memory and a processor, wherein the memory stores computer readable instructions, and the computer readable instructions, when executed by the processor, cause the processor to execute the steps of the output power control method of the carbon dioxide laser therapeutic apparatus in the above embodiments.
The present application also provides a computer readable storage medium, which may be a non-volatile computer readable storage medium, and may also be a volatile computer readable storage medium, where instructions are stored in the computer readable storage medium, when the instructions are executed on a computer, cause the computer to perform the steps of the output power control method of the carbon dioxide laser therapeutic apparatus.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the above-described systems, systems and units may refer to the corresponding processes in the foregoing method embodiments, which are not repeated herein.
The integrated units, if implemented in the form of software functional units and sold or used as stand-alone products, may be stored in a computer readable storage medium. Based on such understanding, the technical solution of the present application may be embodied essentially or in part or all of the technical solution or in part in the form of a software product stored in a storage medium, including instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present application. The storage medium includes a U disk, a removable hard disk, a read-only memory (ROM), a random access memory (random access memory, RAM), a magnetic disk, an optical disk, or other various media capable of storing program codes.
While the application has been described in detail with reference to the foregoing embodiments, it will be understood by those skilled in the art that the foregoing embodiments may be modified or equivalents may be substituted for some of the features thereof, and that the modifications or substitutions do not depart from the spirit and scope of the embodiments of the application.