Fumigation phototherapy instrument control method and device and fumigation phototherapy instrument

Abstract

The invention relates to the technical field of medical instruments, and discloses a control method and device of a fumigation phototherapy instrument and the fumigation phototherapy instrument, and the method comprises the steps: obtaining a state parameter value collected by at least one sensor; predicting state parameter predicted values corresponding to at least part of the state parameters based on the state parameter values acquired by the sensor by using a prediction model, and generating a first predicted value of a control parameter based on the state parameter predicted values; generating a corresponding adjustment control instruction based on the first predicted value of the control parameter; and adjusting a corresponding control parameter value based on the adjustment control instruction. According to the invention, dynamic and accurate control of the fumigation phototherapy instrument can be realized.

Description

Technical Field

[0001] This invention relates to the field of medical device technology, specifically to a control method, device, and fumigation phototherapy device. Background Technology

[0002] A fumigation phototherapy device is a device that combines traditional Chinese medicine fumigation with specific wavelength light irradiation. It is a two-in-one device, comprising a fumigation (vaporization) section and a phototherapy (irradiation) section. The fumigation (vaporization) section heats and vaporizes water or traditional Chinese medicine liquid, producing medicinal vapor containing medicinal components. The phototherapy (irradiation) section simultaneously emits light of specific wavelengths (commonly such as infrared, red light, yellow light, etc.) to synergistically irradiate the fumigated areas.

[0003] The fumigation phototherapy devices provided in the related technologies have asynchronous parameter adjustments and status responses. For example, the temperature adjustment of the fumigation phototherapy devices in the related technologies mostly adopts on / off control, that is, heating stops after reaching the preset temperature and restarts when it falls below another preset threshold. This temperature adjustment method cannot accurately meet the temperature requirements. Summary of the Invention

[0004] This invention provides a control method, device, and fumigation phototherapy device to solve the problem that the parameter adjustment of the fumigation phototherapy device cannot accurately meet the relevant needs.

[0005] In a first aspect, the present invention provides a control method for a fumigation phototherapy device, the method comprising: Acquire the status parameter values ​​collected by at least one sensor; Using a prediction model, at least some state parameters are predicted based on the state parameter values ​​collected by the sensor, and a first predicted value of the control parameter is generated based on the predicted state parameter value. Generate corresponding adjustment control commands based on the first predicted value of the control parameters; Based on the adjustment control command, the corresponding control parameter values ​​are adjusted.

[0006] The control method for the fumigation phototherapy device provided in this embodiment generates adjustment control commands for control parameters by predicting changes in state parameters using a predictive model, thereby improving the accuracy of control.

[0007] Secondly, the present invention provides a control device for a fumigation phototherapy device, the device comprising: A measurement data acquisition module is used to acquire state parameter values ​​collected by at least one sensor. The optimized control module is used to use a prediction model to predict state parameter prediction values ​​corresponding to at least some of the state parameters based on the state parameter values ​​collected by the sensor, and to generate a first prediction value of the control parameters based on the state parameter prediction values. A control command generation module is used to generate corresponding adjustment control commands based on the first predicted value of the control parameters. The parameter adjustment module is used to adjust the corresponding control parameter values ​​based on the adjustment control command.

[0008] Thirdly, the present invention provides a fumigation phototherapy device, comprising: a memory and a processor, wherein the memory and the processor are communicatively connected to each other, the memory stores computer instructions, and the processor executes the computer instructions to perform the control method of the fumigation phototherapy device of the first aspect or any corresponding embodiment described above.

[0009] Fourthly, the present invention provides a computer-readable storage medium storing computer instructions for causing a computer to execute the control method of the fumigation phototherapy device according to the first aspect or any corresponding embodiment thereof.

[0010] Fifthly, the present invention provides a computer program product, including computer instructions for causing a computer to execute the control method of the fumigation phototherapy device described in the first aspect or any corresponding embodiment thereof. Attached Figure Description

[0011] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the drawings used in the description of the specific embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0012] Figure 1 This is a schematic flowchart of a control method for a fumigation phototherapy device according to an embodiment of the present invention; Figure 2 This is a schematic diagram of closed-loop control according to an embodiment of the present invention; Figure 3 This is a structural block diagram of the control device of the fumigation phototherapy instrument according to an embodiment of the present invention; Figure 4 This is a schematic diagram of the hardware structure of the fumigation phototherapy device according to an embodiment of the present invention. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0014] It is understood that before using the technical solutions disclosed in the various embodiments of the present invention, users should be informed of the types, scope of use, and usage scenarios of the personal information involved in the present invention and their authorization should be obtained in accordance with relevant laws and regulations through appropriate means.

[0015] The terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance or implicitly specifying the number of technical features indicated. Thus, a feature defined as "first" or "second" may explicitly or implicitly include one or more of that feature. In the description of this invention, "a plurality of" means two or more, unless otherwise explicitly specified.

[0016] This invention provides a control method for a fumigation phototherapy device, which uses a predictive model to predict changes in state parameters and adjusts control parameters to improve control accuracy.

[0017] According to an embodiment of the present invention, a control method for a fumigation phototherapy device is provided. It should be noted that the steps shown in the flowchart in the accompanying drawings can be executed in a computer system such as a set of computer-executable instructions. Furthermore, although a logical order is shown in the flowchart, in some cases, the steps shown or described may be executed in a different order than that shown here.

[0018] This embodiment provides a control method for a fumigation phototherapy device. Figure 1 This is a flowchart of a control method for a fumigation phototherapy device according to an embodiment of the present invention, such as... Figure 1 As shown, the process includes the following steps: Step S101: Obtain state parameter values ​​collected by at least one sensor.

[0019] Specifically, the state parameter value here can be the state parameter value at the current moment. In this embodiment, each sensor collects the corresponding state parameter value at a certain frequency during the operation of the fumigation phototherapy device. The sampling frequencies of different sensors can be the same or different.

[0020] Step S102: Using a prediction model, predict the state parameter prediction values ​​corresponding to at least some of the state parameters based on the state parameter values ​​collected by the sensor, and generate the first prediction value of the control parameters based on the state parameter prediction values.

[0021] In this embodiment, the first predicted value of the control parameter is obtained in real time based on the latest collected state parameter value using the prediction model.

[0022] Step S103: Generate a corresponding adjustment control command based on the first predicted value of the control parameter.

[0023] Step S104: Based on the adjustment control command, adjust the corresponding control parameter values. Specifically, based on the adjustment control command, adjust the control parameter values ​​of the execution structure, which includes a heating module, a phototherapy module, a fan module, etc.

[0024] The control method of the fumigation phototherapy device provided in this embodiment is a real-time rolling control mechanism, for example, the control parameters can be adjusted once every 1 second (1 second can be the minimum sampling interval of all sensors).

[0025] The control method for the fumigation phototherapy device provided in this embodiment generates adjustment control commands for control parameters by predicting changes in state parameters using a predictive model, thereby improving the accuracy of control.

[0026] In some optional embodiments, the state parameter values ​​include at least one of the following: steam temperature Ts, steam flow rate Fs, ambient temperature Ta, skin temperature Tk, light power density Ip, and working distance D. Different state parameter values ​​are acquired using corresponding sensors.

[0027] The control parameters include at least one of the following: heating power Ph, light power Po, and fan speed Vf.

[0028] The working distance refers to the distance between the fumigation phototherapy device and the user. The fan here refers to the fan used in the fumigation phototherapy device to diffuse the steam. Heating power Ph refers to the power of the fumigation heat source, light power Po refers to the power of the phototherapy light source, and fan speed Vf is an auxiliary parameter for steam diffusion.

[0029] Steam temperature Ts (in °C) reflects the real-time energy level of the steam heat source. Steam flow rate Fs (in mL / min) affects the amount of steam acting on the skin per unit time. Ambient temperature Ta (in °C) affects the steam heat dissipation rate and the skin's baseline temperature. Skin temperature Tk (in °C) is directly related to safety (avoiding burns) and effectiveness. Light power density Ip (in mW / cm²) determines the energy penetration depth of phototherapy. Working distance D (in cm) determines the degree of light power density attenuation and steam heat loss.

[0030] In this embodiment, the ambient temperature Ta and working distance D are used as disturbance inputs and may not be used as prediction objects of the prediction model. That is, the prediction model may not predict the ambient temperature Ta and working distance D.

[0031] In related technologies, fumigation phototherapy devices only monitor heating power and light power during operation, without considering the impact of the usage environment, working distance, and total working time on the effect. Therefore, the effect cannot be guaranteed under different usage conditions. This embodiment introduces state parameters such as temperature, flow rate, and distance, specifically including multimodal state parameters such as steam temperature, steam flow rate, ambient temperature, skin temperature, light power density, total working time, and working distance. Based on these multimodal state parameters, control parameters such as heating power, light power, and diffusion fan speed are adjusted to achieve closed-loop regulation and control.

[0032] In addition, embodiments of the present invention introduce state parameters such as ambient temperature and skin temperature, thereby addressing temperature fluctuations caused by factors such as human metabolism and ambient temperature.

[0033] In this embodiment of the invention, multiple sensors are required to collect different state parameter values, and the multiple state parameter values ​​collected by the sensors are required to predict the future values ​​of multiple state parameters. When there are differences in the sampling frequency of different sensors (e.g., skin temperature Tk sampling frequency 1Hz, light power density Ip sampling frequency 5Hz), all parameters need to be uniformly interpolated to a 1Hz time scale to ensure that the parameters are fused at the same time point (e.g., t=1s, t=2s) to avoid prediction errors caused by time misalignment.

[0034] In some optional embodiments, the prediction model includes at least one of a heat transfer sub-model, a fluid sub-model, and an optical sub-model; The heat transfer sub-model is used to describe the dynamic relationship between skin temperature, steam temperature, and ambient temperature. The fluid sub-model is used to describe the relationship between steam flow rate and heating power, and pipe resistance. The optical sub-model is used to describe the relationship between optical power density and optical power, working distance, and light incident angle. The light incident angle can be a fixed value or a variable value. If the light incident angle is variable, it can be measured using sensors such as a digital angle meter.

[0035] In this embodiment, the prediction model used for state parameter prediction is based on the three core principles of heat transfer, fluid mechanics, and optics, and is constructed into three sub-models: the heat transfer sub-model is used to predict skin temperature and steam temperature, the fluid sub-model is used to predict steam flow rate, and the optics sub-model is used to predict light power density.

[0036] Specifically, the heat transfer sub-model is constructed based on the following heat transfer assumptions: the area of ​​action (human skin and superficial subcutaneous tissue) is a homogeneous semi-infinite medium, and the heat of vapor is transferred through convection (vapor-skin surface) and conduction (skin surface-deep layer), while radiative heat transfer is neglected (accounting for <5%, which can be ignored).

[0037] The fluid sub-model is constructed based on the following fluid dynamics assumptions: the steam is in steady laminar flow at the transmission pipeline and outlet, and the flow rate change is only affected by the heating power and pipeline resistance, turbulent disturbances are ignored (the equipment pipeline diameter is <50mm, the flow velocity is <0.3m / s, which meets the laminar flow conditions).

[0038] The optical sub-model is constructed based on the following optical assumptions: the phototherapy light source is a Lambertian light source, which only undergoes diffuse reflection and absorption in skin tissue, and the absorption coefficient is only positively correlated with skin temperature (the absorption coefficient increases by 0.02 cm⁻¹ for every 1°C increase in temperature). -1 ).

[0039] Accordingly, the heat transfer sub-model, based on Fourier's law and Newton's cooling formula, describes the dynamic relationship between skin temperature, vapor temperature, and ambient temperature:

[0040] in, The value is the steam-skin convection heat transfer coefficient, ranging from 5 to 8 W / (m²). ℃), The skin-environment convective heat transfer coefficient ranges from 2 to 4 W / (m²). ℃), This is the distance-dependent heat loss coefficient, with a value ranging from 12 to 15 W·cm / (m²·℃). , , The calibration coefficient is used in the experiment.

[0041] The fluid sub-model is based on Bernoulli's equation, relating steam flow rate to heating power and pipe resistance:

[0042] in, This represents the fixed loss corresponding to pipeline resistance, with a value ranging from 5 to 8W. These are inherent parameters of the equipment. This is the flow coefficient, with a value ranging from 0.03 to 0.05 mL / (min). √W), The calibration coefficient is used in the experiment.

[0043] The optical sub-model, based on the inverse square law and Lambert's law, describes the relationship between optical power density and optical power, working distance, and incident angle:

[0044] Where θ is the incident angle of light, which is 0° by default in the device, therefore cosθ = 1. The light absorption efficiency of the skin is 0.6-0.8.

[0045] In some optional implementations, the prediction model includes the heat transfer sub-model, the fluid sub-model, and the optical sub-model, and also includes a coupled prediction module.

[0046] Step S102, namely, using a prediction model to predict state parameter prediction values ​​corresponding to at least some state parameters based on the state parameter values ​​collected by the sensor, and generating a first predicted value for the control parameters based on the predicted state parameter values, includes: In step S1021, the heat transfer sub-model, the fluid sub-model, and the optical sub-model in the prediction model predict the changes in the corresponding state parameter values ​​based on some of the state parameter values ​​collected by the sensor, and obtain the corresponding state parameter prediction values.

[0047] In step S1022, the coupled prediction module in the prediction model integrates the predicted values ​​of the state parameters and obtains the first predicted value of the control parameters through the coupling term prediction.

[0048] The coupling terms include: a coupling term between heating power and steam temperature, and a coupling term between optical power and skin temperature. Specifically, the heating power Ph simultaneously affects the steam temperature Ts of the heat transfer sub-model and the steam flow rate Fs of the fluid sub-model; the optical power Po indirectly affects the skin temperature Tk through the optical power density Ip of the optical sub-model; and the fan speed Vf adjusts the dynamic response speed of the steam flow rate Fs and the steam temperature Ts by changing the steam diffusion rate.

[0049] Specifically, each sub-model predicts corresponding state parameter values ​​based on the values ​​acquired by sensors. Different sub-models require different input data (i.e., state parameter values) and output data (i.e., predicted state parameter values). Specifically, the input data for the heat transfer sub-model includes steam temperature Ts, ambient temperature Ta, working distance D, and skin temperature Tk; the output data is the predicted skin temperature (Tk). pred The input data for the fluid sub-model includes the current heating power (Ph). current ), steam flow rate Fs, output steam flow rate prediction value (Fs) pred The input data for the optical sub-model includes the current optical power (Po). current The output data includes the working distance D, optical power density Ip, and the predicted optical power density value (Ip). pred ).

[0050] The coupling prediction module integrates the predicted state parameter values ​​and obtains the first predicted value of the control parameter through the coupling term. Specifically, the coupling term Ph between the heating power Ph and the steam temperature Ts can be calculated based on the physical relationship. couple The calculation formula can be, for example, Ph couple =0.6·Ph+0.4·Ts, and can also be used to calculate the coupling term Po between optical power Po and skin temperature Tk. couple =0.7·Po+0.3·Tk. Then, based on the skin temperature prediction value (Tk) output by the sub-model... pred ), steam flow forecast (Fs) pred ) and predicted optical power density (Ip) pred This, combined with the coupling term, yields the first predicted value of the control parameter. Specifically, if Tk > Tk opt (Tk opt To achieve the optimal treatment temperature for the skin (e.g., 40°C), lower the pH. pred (Predicted heating power), and simultaneously calculate the required reduction amount (e.g., Tk). pred Every higher than Tk opt 1℃, Ph pred Reduce by 5W); if Ip pred <Ip ref (Ip ref The target optical power density is, for example, 50 mW / cm². 2 If this increases Po, then... pred (Predicted optical power value), the amplitude is adjusted according to D (e.g., for every 1cm increase in D, Po...). pred Increase by 3W); if Fs pred If the flow rate deviates from the target flow rate (e.g., target flow rate is 50 mL / min), adjust Vf. pred (Fan speed prediction), for example, Fs pred For every 5 mL / min decrease, Vf pred Increase the rpm by 100.

[0051] Furthermore, after the coupled prediction module in the prediction model predicts the predicted value (i.e., the first predicted value) of the control parameter, Ph can also be verified. pred Po pred Vf pred Is it within a safe range (e.g., Ph)? pred <1500W, Po pred <500W, Vf pred If the speed exceeds 3000 rpm, the prediction will be truncated and corrected. Finally, the compliant prediction value will be output to the intelligent control algorithm layer to generate the corresponding adjustment control command.

[0052] In this embodiment, three sub-models are first constructed based on the three core principles of heat transfer, fluid mechanics, and optics. Then, based on the multi-parameter coupling characteristics of the fumigation phototherapy device, the three sub-models are integrated into a multi-parameter coupled prediction model through coupling terms. That is, a prediction model is established through parameter correlation, forming a closed-loop parameter correlation network.

[0053] In some alternative embodiments, step S102, which involves using a prediction model to predict state parameter prediction values ​​corresponding to at least some of the state parameters based on the state parameter values ​​collected by the sensor, and generating a first predicted value for the control parameters based on the predicted state parameter values, includes: Step S1021: Using a prediction model, at least some state parameter prediction values ​​are predicted based on the state parameter values ​​collected by the sensor and the candidate control parameter values.

[0054] Specifically, given the current state parameter value, we can predict the state parameter values ​​for the next N (N is a positive integer) time steps, resulting in N predicted state parameter values. For example, the prediction time step can be 1 second, meaning the interval between the N predicted time steps is 1 second, and N can be 10, meaning we predict the trend of state parameter values ​​over the next 10 seconds. The value of N needs to balance prediction accuracy and real-time performance; predictions that are too long can lead to error accumulation, while predictions that are too short cannot anticipate parameter fluctuations.

[0055] Step S1022: Based on the deviation between the predicted value of the state parameter and the corresponding expected value of the state parameter, adjust the value of the candidate control parameter, and use the prediction model again to predict a new predicted value of the state parameter based on the adjusted value of the candidate control parameter, until the deviation meets the preset condition, and use the latest value of the candidate control parameter as the first predicted value of the control parameter.

[0056] Specifically, candidate control parameter values ​​can include candidate control parameter values ​​for the next M (M is a positive integer) time steps, where M ≤ N. After adjustment, once the deviation meets preset conditions, the candidate control parameter value for the next time step from the latest candidate control parameter values ​​can be used as the first predicted value of the control parameter. For example, the value of M can be 5, meaning that control parameter values ​​for the next 5 time steps are predicted each time, ensuring control flexibility.

[0057] In this embodiment, the prediction model may also include a heat transfer sub-model, a fluid sub-model, and an optical sub-model, as detailed in the above embodiments, and will not be repeated here.

[0058] As described in the above embodiments, the predicted values ​​of the state parameters include steam temperature Ts, steam flow rate Fs, skin temperature Tk, and light power density Ip. The predicted values ​​of the control parameters include heating power Ph, light power Po, and fan speed Vf.

[0059] In the process of using a prediction model to predict the state parameters and adjusting the candidate control parameter values ​​based on the deviation between the predicted state parameter values ​​and the corresponding expected state parameter values, the following constraints must be met: 1. Constraints on control parameters: Ph min ≤Ph(t)≤Ph max Po min ≤Po (t)≤Po max Vf min ≤Vf(t)≤Vf max For example, Ph min =300W, Ph max =1500W; Po min =50W, Po max =500W, Vf min =500rpm, Vf max =3000rpm.

[0060] 2. Constraints on state parameters: Tk min ≤Tk(t)≤Tk max Ts min ≤Ts(t)≤Ts max IP min ≤Ip(t)≤Ip max For example, Tk min =32℃, Tk max =45℃, to avoid burns; Ts min =40℃, Ts max =95℃, to avoid steam overheating or failure; Ip min =30mW / cm 2 IP max =70mW / cm 2 This ensures therapeutic efficacy while avoiding photodamage.

[0061] The above constraints are determined based on equipment safety standards and human tolerance levels. They are hard constraints on control parameters and state parameters, that is, safety boundaries that cannot be broken.

[0062] Furthermore, embodiments of the present invention may also include soft constraints, which are performance objectives, i.e., expected values ​​of state parameters, that need to be approximated as closely as possible. For example, Tk target =40℃, Ts target =60℃, Ip target =50 mW / cm 2 , Fs target =50mL / min.

[0063] In this embodiment of the invention, the optimal adjustment control command for the fumigation phototherapy device is generated through a predictive future-rolling optimization loop mechanism, while satisfying the device safety constraints.

[0064] Specifically, regarding the deviation between the predicted value and the corresponding expected value of the state parameter, if the predicted value is higher than the expected value, the candidate control parameter value in subsequent optimization needs to be adjusted to reduce the subsequent predicted value. That is, when adjusting the candidate control parameter value, adjustments can be made based on the aforementioned deviation. Specifically, if there are multiple state parameters, there are corresponding deviations for multiple state parameters. When adjusting the candidate control parameter value, the deviation corresponding to each state parameter can be considered separately, or a weighted sum of the deviations can be performed, and then the candidate control parameter value can be adjusted based on the weighted sum. The weight of each deviation can be assigned differently based on the degree of influence of different state parameters on safety and efficacy, using the Analytic Hierarchy Process (AHP). For example, skin temperature should have the highest weight because it is directly related to safety, and the skin temperature error weight is ωTk=0.4; light power density affects the efficacy of phototherapy, so the light power density error weight is ωIp=0.3; steam temperature affects the efficacy of fumigation, so the steam temperature error weight is ωTs=0.2; steam flow rate affects the fumigation effect, so the steam flow rate error weight is set to ωFs=0.1.

[0065] Furthermore, for each sub-model in the prediction model mentioned in the above embodiments, the calibration coefficients of the sub-model can be adjusted based on the error between the predicted state parameter values ​​and the actual state parameter values ​​at the corresponding time (specifically, those acquired by sensors). , , , Make minor adjustments (adjustment magnitude <5%, i.e., fine-tuning). These adjustments can be made periodically (e.g., every 5 minutes) to avoid predictive model drift caused by equipment aging.

[0066] In some optional implementations, the preset condition is the minimum weighted sum of the deviation and the change in the control parameter value.

[0067] Specifically, in this embodiment of the invention, the objective of adjusting the candidate control parameter values ​​is to minimize the weighted sum of the deviation and the changes in the control parameter values, so as to balance efficacy and control stability. Therefore, the following objective function J can be constructed:

[0068] in, The state deviation term represents the predicted values ​​of the state parameters at N future time points. With the corresponding expected value of the state parameters ( The sum of squared deviations of the state parameters is given by Q, which is a weight matrix. Q is a diagonal matrix where the diagonal elements are the weights of each state parameter. These weights can be consistent with the weights used when adjusting the candidate control parameter values ​​to ensure that the optimization direction of the control parameters is consistent with the safety and efficacy objectives. For example, Q... 11 =0.4, Q 22 =0.2, Q 33 =0.3, Q 44 =0.1.

[0069] The control variation term represents the amount of change in the control parameter over the next M time points. The sum of squares, where R is the weight matrix, and the diagonal elements are set to smaller values, for example, R0. 11 =0.05, R 22 =0.03, R 33 =0.02, to avoid frequent and large changes in control parameters that could lead to unstable equipment operation, such as sudden increases or decreases in heating power.

[0070] The process of solving the above objective function may specifically include: First, the prediction model is linearized: the nonlinear coupling equations of the prediction model (such as the differential equations of the heat transfer sub-model) are expanded using Taylor at the current operating point ((X(t), U(t-1)), approximating a linear state-space model: X(t+1|t)=A·X(t|t)+B·U(t|t)+C·E(t), where A (state matrix), B (control matrix), and C (perturbation matrix) are the linearized coefficient matrices, obtained through numerical differentiation. Specifically, the state matrix A is composed of state parameters (including steam temperature, steam flow rate, skin temperature, and light power density); the control matrix is ​​composed of control parameters (including heating power, light power, and fan speed); and the perturbation matrix is ​​composed of perturbation parameters from the state parameters (including ambient temperature Ta and operating distance D).

[0071] Then, constraint transformation: transform the above hard constraints (Ph) min ≤Ph(t)≤Ph max Convert to linear inequality constraints: U min ≤U(t+k|t)≤U max (k=0,1,..,M-1), X min ≤X(t+k|t)≤X max (k=1,2,…,N), to ensure the optimization results are compliant.

[0072] Finally, a quadratic programming solver (such as OSQP or IPOPT) can be called, with the linearized model, objective function, and constraints input, to solve for the optimal control parameter vector sequence for the next M time steps: .

[0073] In other words, embodiments of the present invention can use a quadratic programming (QP) algorithm to solve the objective function, and find the control parameter vector that minimizes the objective function while satisfying hard constraints. .

[0074] In this embodiment, since the dynamic process of the fumigation phototherapy device is nonlinear (e.g., the steam temperature and heating power are not simply proportional, but are also affected by the ambient temperature), the complex nonlinear optimization problem can be transformed into an easily solvable quadratic programming problem by using Taylor expansion to approximate it as a linear model, while ensuring the accuracy near the current working point (e.g., the approximate error within a certain skin temperature range is extremely small).

[0075] In this embodiment of the invention, a rolling implementation strategy is adopted, that is, based only on the first optimal control parameter value of the optimal control parameter vector sequence ( The first predicted value of the control parameter (i.e., the initial value of the control parameter) is used to generate the corresponding adjustment control command. Subsequent adjustment control commands are re-optimized and generated in the next control cycle.

[0076] In some optional implementations, step S103, namely generating the corresponding adjustment control command based on the first predicted value of the control parameter, includes: Step S1031: Obtain the error value fed back after the first L (positive integer) control parameter adjustments, wherein the error value is the error between the predicted value of the state parameter and the corresponding measured value of the state parameter; Step S1032: Based on the error value and / or the corresponding error change rate, dynamically adjust the proportional coefficient, integral coefficient, and derivative coefficient of the PID using a fuzzy rule base; Step S1033: Correct the first predicted value of the control parameter based on the adjusted PID to obtain the second predicted value of the control parameter; Step S1034: Based on the first predicted value and the second predicted value of the control parameters, generate corresponding adjustment control commands.

[0077] In this embodiment of the invention, in order to further improve control accuracy and response speed, fuzzy PID is used to locally correct the first predicted value of the control parameters obtained by the prediction model.

[0078] Specifically, fuzzy PID controllers target a single control parameter (such as heating power Ph) and base their decisions on the magnitude of the error and the rate of change of the error (such as...). The proportional coefficient (Kp), integral coefficient (Ki), and derivative coefficient (Kd) of the PID are dynamically adjusted using a fuzzy rule base. For example, when eTk(t) > 2℃ and When the temperature rises rapidly by more than 0.5℃ / s (skin temperature increases rapidly), the rule library outputs adjustment instructions to "increase Kp, decrease Ki, increase Kd" to quickly reduce the heating power Ph and prevent the skin temperature Tk from exceeding the safe threshold.

[0079] Corrected control parameter Ph PID (That is, the second predicted value of the control parameter) and the optimal control parameter Ph obtained based on the prediction model. MPC Weighted fusion is performed, and the fusion formula is Ph final =0.7·Ph MPC +0.3·Ph PID It retains the advantages of global optimization based on predictive models, and utilizes the fast response capability of fuzzy PID to achieve a synergistic control effect of "global stability + local fast correction".

[0080] In addition, in this embodiment of the invention, the error can be limited to the feedback error of multiple state parameters only when the feedback error exceeds the threshold (e.g., eTk(t) > 2℃, eIp(t) > 5mW / cm). 2 When this occurs, the local correction function of the fuzzy PID controller is triggered. That is, during the steady-state operation phase (e.g., Tk stabilizes at 39-41℃, Ip stabilizes at 48-52mW / cm²), the local correction function of the fuzzy PID controller is activated. 2 It only requires global optimization using a predictive model to generate smooth adjustment control commands, avoid parameter fluctuations, and ensure stable system operation.

[0081] Multivariable control of heating power: Physical meaning: ; in, Indicates total power. Indicates the base power. Indicates PID correction. This indicates a fuzzy correction.

[0082] Control principle: .

[0083] This indicates the ambient temperature compensation term; a larger temperature difference indicates that more power is required. This represents the distance compensation term; the greater the distance, the greater the heat loss. This indicates the steam flow rate; a higher flow rate requires more heat.

[0084] PID control: ; in, The proportional term responds to the current error, while the integral term eliminates the steady-state error. Differential terms improve response speed.

[0085] Multi-objective error definition: At the same time, skin temperature error and steam temperature error are taken into account.

[0086] Adaptive control of optical power: ,in, Indicates the nominal optical power. This represents the distance efficiency factor. This represents the heat effect factor.

[0087] Distance compensation algorithm: ,in, The law of inverse square law states that light intensity is inversely proportional to the square of the distance. cos(θ) is an angle correction factor, taking into account the influence of the incident angle of light. =30cm, which is the standard reference distance.

[0088] Thermal feedback adjustment: ; Gaussian function form, at the optimum temperature =The efficiency is highest around 40℃, λ=0.1, which is a parameter to control temperature sensitivity. The light power is automatically reduced when the skin is too hot.

[0089] Light intensity closed-loop control: , among which, Ip ref The target optical power density is 50 mW / cm², Ip is the measured optical power density, and Po is... min This is the minimum sustaining power.

[0090] Fan speed fuzzy control: Vf=Vf min +△Vf fuzzy , among which, Vf min =500rpm, which is the minimum operating speed, △Vf fuzzy ΔVf is the adjustment amount of the fuzzy inference output. fuzzy =∑(μi·Wi·Vfi) / ∑(μi·Wi), where μi is the activation strength (membership) of the i-th rule, Wi is the weight of the i-th rule, and Vfi is the rotational speed output corresponding to the i-th rule. The weighted average method is used for defuzzification.

[0091] In some optional implementations, the prediction model may further include a data preprocessing module, which is set before the three sub-models to clean and normalize the state parameter values ​​collected by the sensors, so as to provide high-quality input data for the subsequent sub-models.

[0092] Specifically, the data preprocessing module includes a data cleaning unit and a normalization unit. The data cleaning unit removes outliers, employing the 3σ criterion to remove data exceeding reasonable ranges (e.g., steam temperature Ts > 100℃, skin temperature Tk > 45℃) in real time, preventing outliers from interfering with the model's predictions. The normalization unit maps all input parameters (i.e., state parameter values) to the [0,1] interval. The normalization formula is: ,in, For the minimum and maximum values ​​of each state parameter (such as the ambient temperature Ta), =15℃ =35℃).

[0093] Additionally, the data preprocessing module may include a feature extraction unit. This unit extracts derived features based on the input state parameter values. These derived features, along with the input state parameter values, serve as input to the subsequent three sub-models. Examples of derived features include temperature difference features and distance-power features. Temperature difference features, for example, include the temperature difference between steam temperature and skin temperature (i.e., Ts). Tk), the temperature difference between the steam temperature and the ambient temperature (i.e., Ts) Ta), etc., distance-power characteristics can be, for example, the product of the working distance and the optical power density (i.e., D). Ip).

[0094] In other embodiments, the data preprocessing module may not be set in the prediction model, but may be independent of the prediction model, so as to perform the above-mentioned preprocessing on the state parameter values ​​collected by the sensor, and then input them into the prediction model.

[0095] In some optional implementations, before using a prediction model to predict state parameter prediction values ​​corresponding to at least some of the state parameters based on the state parameter values ​​collected by the sensor, and generating a first predicted value of the control parameters based on the state parameter prediction values, the method further includes: Kalman filtering is used to fuse multiple identical state parameter values ​​collected by different sensors to obtain a single state parameter value.

[0096] In other words, such as Figure 2 As shown, after acquiring the state parameter values ​​collected by multiple sensors, if there are multiple state parameter values ​​of the same type collected by different sensors, such as multiple steam temperatures collected by temperature sensors at different locations, or light power densities collected by sensors at different locations, then Kalman filtering (KF) is used to fuse the same state parameter values ​​collected by multiple different sensors in order to reduce measurement noise and improve data accuracy.

[0097] The Kalman filter process consists of two steps: prediction and update. In this embodiment, fusion can be completed every second.

[0098] The prediction process includes: A linear state-space model based on a prediction model is used to predict the state vector and covariance matrix at the current time step (based on the state parameter values ​​collected by the sensor at the previous time step). The state prediction equation is: ,in This is the prior state estimate at time t (without considering current measurement data (i.e., data collected by sensors)). Let A and B be the posterior state estimate at time t-1 (after fusing measurement data from time t-1), where A and B are the linearized state and control matrices, respectively. The covariance prediction equation is: Where P(t) is the prior covariance matrix at time t (reflecting the uncertainty of the prior estimate), P(t-1) is the posterior covariance matrix at time t-1, and Q... k The process noise covariance matrix (diagonal elements are the process noise variances of each state parameter, such as Q) k11 =0.05℃, Q k33 =0.2(mW / cm 2 ) 2 (), obtained through experimental calibration.

[0099] The update process includes: Collect multimodal measurement data at the current moment (such as the measurement values ​​Tk from two skin temperature sensors). meas1 (t), Tk meas2 (t), the measured values ​​Ip from two optical power density sensors meas1 (t), Ip meas2 (t)), construct the measurement vector:

[0100] Calculate the Kalman gain (K(t)). Where H is the measurement matrix (mapping the state vector to the measurement vector, e.g., the first row of H is [1,0,0,0,0,0], corresponding to Tk). meas1 (t)=1·Tk(t)+0·Ts(t)+...), R k To measure the noise covariance matrix (where the diagonal elements are the noise variance of each measurement, e.g., R0) 11 =0.1℃, R 44 =0.3(mW / cm 2 ) 2 The value is determined by the sensor manual and experimental testing.

[0101] Posterior state estimation (data fusion result): ,in To measure the residual (the difference between the measured value and the prior prediction), the Kalman gain K(t) dynamically adjusts the weights according to the magnitude of the residual. The larger the residual, the larger K(t), indicating that the measured data contributes more to the fusion result; the smaller the residual, the smaller K(t), indicating that the prior prediction is more reliable, thus achieving "optimal weighted fusion of the predicted value and the measured value".

[0102] Posterior covariance update: P(t)=(IK(t)·H)·P(t), where I is the identity matrix. The updated P(t) is used for covariance prediction in the next period, reflecting the uncertainty of the posterior estimate. The smaller the uncertainty, the higher the accuracy of subsequent fusion.

[0103] In addition, anomaly detection can be performed on the state parameter values ​​acquired by the sensors. Specifically, anomaly detection is used to identify outliers in multimodal data (such as sudden increases in Tk caused by sensor failure or Ip jumps caused by electromagnetic interference), to prevent abnormal data from affecting the fusion results and control commands. The specific process consists of four steps: "anomaly detection algorithm selection - feature extraction - anomaly determination - anomaly handling". 1. Anomaly detection algorithm selection: Considering the multimodal data characteristics of the fumigation phototherapy device (high real-time requirements and low data dimensionality), a combination of two algorithms, "statistical threshold-based anomaly detection" and "sliding window-based trend anomaly detection", is selected to balance detection accuracy and real-time performance.

[0104] 2. Feature Extraction: Two types of anomaly features are extracted from the raw measurement data before Kalman filtering: Static characteristics: the absolute deviation of a single measurement, i.e., |X| meas (t)-X avg (t-10)|, where X avg (t-10) is the average value of this parameter over the past 10 seconds (e.g., Tk). avg (t-10) reflects the deviation between the current measured value and the historical steady-state value; Dynamic characteristics: the rate of change of the measured value, i.e. It reflects the instantaneous rate of change of the measured value. ), to capture rapidly changing anomalies.

[0105] 3. Anomaly detection: Static anomaly detection: If |X meas (t)-X avg If (t-10)|>3·σx (σx is the standard deviation of this parameter over the past 10 seconds), then it is considered a static anomaly. For example, Tk avg (t-10)=40℃, σTk=0.5℃, if Tk measIf (t) = 42℃, then |42-40| = 2℃ > 3 × 0.5 = 1.5℃, which is determined to be a static anomaly.

[0106] Dynamic anomaly determination: If >ΔX max (ΔX max This is the maximum allowable rate of change for this parameter. If the temperature is 1.5℃ / s, it is determined to be a dynamic anomaly.

[0107] Comprehensive judgment: If a measurement value simultaneously meets both static and dynamic anomalies, or meets a single anomaly rule for three consecutive cycles (3s), it is ultimately judged as "abnormal data"; if only a single cycle meets a single anomaly rule, it is judged as "suspected anomaly" and will not be removed for the time being, and the data of the next cycle will continue to be observed.

[0108] 4. Exception handling mechanism: Outlier removal: For measurements identified as "outliers," in the measurement update step of the Kalman filter, the corresponding measurement noise variance Rk is increased by a factor of 10 (e.g., Rk11 from 0.1℃). 2 Increase to 1℃ 2 This reduces the weight of the abnormal data on the fusion result, which is equivalent to indirectly removing the influence of the abnormal data.

[0109] Sensor fault alarm: If the measured value of the same sensor is judged as abnormal data for 5 consecutive cycles (5s), a sensor fault alarm will be triggered, and a message will be displayed on the device screen: "Sensor X is faulty, please check." Simultaneously, the device will automatically switch to a backup sensor (e.g., Tk). meas1 (t) Switch to Tk meas2 (t)) ensures the continuity of data fusion.

[0110] Fusion result correction: If all sensors (e.g., 2 Tk sensors) for a certain parameter are identified as abnormal data, then the prior state estimation of the dynamic prediction model is used. (t) is the result of the fusion of this parameter, and the control command is adjusted to a safe mode (e.g., Ph is reduced to Ph). min =300W, Po drops to Po min =50W), to prevent the device from operating in an unsafe state when data is abnormal.

[0111] In summary, the timing relationship of a control method according to an embodiment of the present invention is shown in Table 1.

[0112] Table 1. Timing Relationship of a Control Method

[0113] The control method of the fumigation phototherapy device in this invention forms a complete "data acquisition-fusion-prediction-control" closed loop. Specifically, through the timing coordination shown in Table 1, a complete closed loop is achieved from the acquisition of multimodal data to fusion and then to the generation of control commands, completing one cycle every 1 second. This ensures both the accuracy of data fusion and the effective handling of abnormal data, as well as the real-time performance and optimization of control commands, providing data support for the safe and efficient operation of the fumigation phototherapy device.

[0114] In this embodiment of the invention, by employing a multimodal monitoring feedback control model, the photothermal power applied to the human body can be accurately predicted. Fumigation can complement phototherapy, achieving effective closed-loop control and significantly improving treatment efficacy and safety. Furthermore, the combination of fumigation and phototherapy modes can be adjusted according to different symptoms to create personalized treatment plans.

[0115] This embodiment also provides a control device for a fumigation phototherapy device, which is used to implement the above embodiments and preferred embodiments; details already described will not be repeated. As used below, the term "module" can refer to a combination of software and / or hardware that performs a predetermined function. Although the device described in the following embodiments is preferably implemented in software, hardware implementation, or a combination of software and hardware, is also possible and contemplated.

[0116] This embodiment provides a control device for a fumigation phototherapy instrument, such as... Figure 3 As shown, it includes: The measurement data acquisition module 301 is used to acquire state parameter values ​​collected by at least one sensor; The optimized control module 302 is used to use a prediction model to predict state parameter prediction values ​​corresponding to at least some of the state parameters based on the state parameter values ​​collected by the sensor, and to generate a first prediction value of the control parameters based on the state parameter prediction values. The control command generation module 303 is used to generate corresponding adjustment control commands based on the first predicted value of the control parameters; The parameter adjustment module 304 is used to adjust the corresponding control parameter value based on the adjustment control command.

[0117] In one optional embodiment, the state parameter values ​​include at least one of the following: steam temperature, steam flow rate, ambient temperature, skin temperature, optical power density, and working distance; and / or, The control parameters include at least one of the following: heating power, light power, and fan speed.

[0118] In one optional implementation, the prediction model includes at least one of a heat transfer sub-model, a fluid sub-model, and an optical sub-model; The heat transfer sub-model is used to describe the dynamic relationship between skin temperature, steam temperature, and ambient temperature. The fluid sub-model is used to describe the relationship between steam flow rate and heating power, and pipe resistance. The optical sub-model is used to describe the relationship between optical power density and optical power, working distance, and light incident angle.

[0119] In one optional implementation, the prediction model further includes a coupled prediction module; The step of using a prediction model to predict state parameter prediction values ​​corresponding to at least some state parameters based on the state parameter values ​​collected by the sensors, and generating a first predicted value for control parameters based on the predicted state parameter values, includes: The heat transfer sub-model, the fluid sub-model, and the optical sub-model in the prediction model predict the changes in the corresponding state parameter values ​​based on some of the state parameter values ​​collected by the sensor, and obtain the corresponding state parameter prediction values. The coupled prediction module in the prediction model integrates the predicted values ​​of the state parameters and obtains the first predicted value of the control parameters through the coupling term prediction. The coupling terms include: a coupling term between heating power and steam temperature, and a coupling term between light power and skin temperature.

[0120] In one optional implementation, the optimization control module includes: The prediction unit is used to use the prediction model to predict state parameter prediction values ​​corresponding to at least some of the state parameters based on the state parameter values ​​and candidate control parameter values ​​collected by the sensor. An optimized modulation unit is used to adjust the candidate control parameter value based on the deviation between the predicted value of the state parameter and the corresponding expected value of the state parameter, and then use the prediction model again to predict a new predicted value of the state parameter based on the adjusted candidate control parameter value, until the deviation meets a preset condition, and then use the latest candidate control parameter value as the first predicted value of the control parameter.

[0121] In one optional implementation, the preset condition is the minimum weighted sum of the deviation and the change in the control parameter value.

[0122] In one optional embodiment, the parameter adjustment module includes: An error value acquisition unit is used to acquire the error value fed back after the previous L adjustments of the control parameters. The error value is the error between the predicted value of the state parameter and the corresponding measured value of the state parameter. A coefficient modulation unit is used to dynamically adjust the proportional coefficient, integral coefficient, and derivative coefficient of the PID based on the error value and / or the corresponding error change rate using a fuzzy rule base. A correction unit is used to correct the first predicted value of the control parameter based on the adjusted PID, so as to obtain a second predicted value of the control parameter; The instruction generation unit is used to generate corresponding adjustment control instructions based on the first predicted value and the second predicted value of the control parameters.

[0123] In one optional embodiment, the control device of the fumigation phototherapy instrument further includes: The fusion module is used to fuse multiple identical state parameter values ​​collected by different sensors using Kalman filtering to obtain a single state parameter value.

[0124] The control device for the fumigation phototherapy instrument provided in this embodiment of the invention can execute the control method for the fumigation phototherapy instrument provided in any embodiment of the invention, and has the corresponding functional modules and beneficial effects for executing the method. Further functional descriptions of the above modules and units are the same as those in the corresponding embodiments described above, and will not be repeated here.

[0125] Figure 4 This is a schematic diagram of the structure of a fumigation phototherapy device provided in an embodiment of the present invention.

[0126] The following is a detailed reference. Figure 4 The diagram illustrates a structural schematic suitable for implementing the fumigation phototherapy device in embodiments of the present invention. The fumigation phototherapy device may include a processor (e.g., a central processing unit, graphics processor, etc.) 401, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 402 or a program loaded from a memory 408 into a random access memory (RAM) 403. The RAM 403 also stores various programs and data required for the operation of the fumigation phototherapy device. The processor 401, ROM 402, and RAM 403 are interconnected via a bus 404. An input / output (I / O) interface 405 is also connected to the bus 404.

[0127] Typically, the following devices can be connected to I / O interface 405: input devices 406 including, for example, touchscreens, touchpads, etc.; output devices 407 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; memory devices 408 including, for example, magnetic tapes, hard disks, etc.; and communication devices 409. Communication device 409 allows the fumigation phototherapy device to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 4 A fumigation phototherapy device with various devices is shown, but it should be understood that it is not required to implement or have all of the devices shown, and it is possible to implement or have more or fewer devices instead.

[0128] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via a communication device 409, or installed from a memory 408, or installed from a ROM 402. When the computer program is executed by the processor 401, it performs the functions defined in the control method of the fumigation phototherapy device according to embodiments of the present invention.

[0129] Figure 4 The fumigation phototherapy device shown is merely an example and should not be construed as limiting the functionality and scope of use of the embodiments of the present invention.

[0130] This invention also provides a computer-readable storage medium. The methods described above according to embodiments of the invention can be implemented in hardware or firmware, or implemented as computer code that can be recorded on a storage medium, or implemented as computer code downloaded via a network and originally stored on a remote storage medium or a non-transitory machine-readable storage medium and then stored on a local storage medium. Thus, the methods described herein can be processed by software stored on a storage medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware. The storage medium can be a magnetic disk, optical disk, read-only memory, random access memory, flash memory, hard disk, or solid-state drive, etc.; further, the storage medium can also include combinations of the above types of memory. It is understood that the computer, processor, microprocessor controller, or programmable hardware includes storage components capable of storing or receiving software or computer code. When the software or computer code is accessed and executed by the computer, processor, or hardware, the control method of the fumigation phototherapy device shown in the above embodiments is implemented.

[0131] A portion of this invention can be applied as a computer program product, such as computer program instructions, which, when executed by a computer, can invoke or provide the methods and / or technical solutions according to the invention through the operation of the computer. Those skilled in the art will understand that the forms in which computer program instructions exist in a computer-readable medium include, but are not limited to, source files, executable files, installation package files, etc. Correspondingly, the ways in which computer program instructions are executed by a computer include, but are not limited to: the computer directly executing the instructions, or the computer compiling the instructions and then executing the corresponding compiled program, or the computer reading and executing the instructions, or the computer reading and installing the instructions and then executing the corresponding installed program. Here, the computer-readable medium can be any available computer-readable storage medium or communication medium accessible to a computer.

[0132] Although embodiments of the invention have been described in conjunction with the accompanying drawings, those skilled in the art can make various modifications and variations without departing from the spirit and scope of the invention, and such modifications and variations all fall within the scope defined by the appended claims.

Claims

1. A control method for a fumigation phototherapy device, characterized in that, The method includes: Acquire the status parameter values ​​collected by at least one sensor; Using a prediction model, at least some state parameters are predicted based on the state parameter values ​​collected by the sensor, and a first predicted value of the control parameter is generated based on the predicted state parameter value. Generate corresponding adjustment control commands based on the first predicted value of the control parameters; Based on the adjustment control command, the corresponding control parameter values ​​are adjusted.

2. The method according to claim 1, characterized in that, The state parameter values ​​include at least one of the following: steam temperature, steam flow rate, ambient temperature, skin temperature, optical power density, and working distance; and / or, The control parameters include at least one of the following: heating power, light power, and fan speed.

3. The method according to claim 1 or 2, characterized in that, The prediction model includes at least one of a heat transfer sub-model, a fluid sub-model, and an optical sub-model; The heat transfer sub-model is used to describe the dynamic relationship between skin temperature, steam temperature, and ambient temperature. The fluid sub-model is used to describe the relationship between steam flow rate and heating power, and pipe resistance. The optical sub-model is used to describe the relationship between optical power density and optical power, working distance, and light incident angle.

4. The method according to claim 3, characterized in that, The prediction model also includes a coupled prediction module; The step of using a prediction model to predict state parameter prediction values ​​corresponding to at least some state parameters based on the state parameter values ​​collected by the sensors, and generating a first predicted value for control parameters based on the predicted state parameter values, includes: The heat transfer sub-model, the fluid sub-model, and the optical sub-model in the prediction model predict the changes in the corresponding state parameter values ​​based on some of the state parameter values ​​collected by the sensor, and obtain the corresponding state parameter prediction values. The coupled prediction module in the prediction model integrates the predicted values ​​of the state parameters and obtains the first predicted value of the control parameters through the coupling term prediction. The coupling terms include: a coupling term between heating power and steam temperature, and a coupling term between light power and skin temperature.

5. The method according to claim 1, characterized in that, The step of using a prediction model to predict state parameter prediction values ​​corresponding to at least some state parameters based on the state parameter values ​​collected by the sensors, and generating a first predicted value for control parameters based on the predicted state parameter values, includes: Using the prediction model, at least some of the state parameters are predicted based on the state parameter values ​​collected by the sensor and the candidate control parameter values. Based on the deviation between the predicted state parameter value and the corresponding expected state parameter value, the candidate control parameter value is adjusted, and the prediction model is used again to predict a new state parameter value based on the adjusted candidate control parameter value, until the deviation meets the preset condition, and the latest candidate control parameter value is used as the first predicted value of the control parameter.

6. The method according to claim 5, characterized in that, The preset condition is the minimum weighted sum of the deviation and the change in the control parameter value.

7. The method according to claim 1, characterized in that, The generation of corresponding adjustment control commands based on the first predicted value of the control parameters includes: Obtain the error value fed back after the previous L control parameter adjustments, where the error value is the error between the predicted value of the state parameter and the corresponding measured value of the state parameter; Based on the error value and / or the corresponding error change rate, the proportional coefficient, integral coefficient and derivative coefficient of the PID are dynamically adjusted through a fuzzy rule base. The first predicted value of the control parameter is corrected based on the adjusted PID to obtain the second predicted value of the control parameter; Based on the first and second predicted values ​​of the control parameters, corresponding adjustment control commands are generated.

8. The method according to claim 1 or 2, characterized in that, Before using a prediction model to predict state parameter prediction values ​​corresponding to at least some state parameters based on the state parameter values ​​collected by the sensor, and generating a first predicted value for the control parameters based on the predicted state parameter values, the method further includes: Kalman filtering is used to fuse multiple identical state parameter values ​​collected by different sensors to obtain a single state parameter value.

9. A control device for a fumigation phototherapy instrument, characterized in that, The device includes: A measurement data acquisition module is used to acquire state parameter values ​​collected by at least one sensor. The optimized control module is used to use a prediction model to predict state parameter prediction values ​​corresponding to at least some of the state parameters based on the state parameter values ​​collected by the sensor, and to generate a first prediction value of the control parameters based on the state parameter prediction values. A control command generation module is used to generate corresponding adjustment control commands based on the first predicted value of the control parameters. The parameter adjustment module is used to adjust the corresponding control parameter values ​​based on the adjustment control command.

10. A fumigation phototherapy device, characterized in that, include: The device includes a memory and a processor, which are communicatively connected to each other. The memory stores computer instructions, and the processor executes the computer instructions to perform the control method of the fumigation phototherapy device according to any one of claims 1 to 8.

Images