Medical waste treatment equipment temperature intelligent control method and system based on multi-physics field coupling

Abstract

The invention provides an intelligent temperature control method and system for medical waste treatment equipment based on multi-physics field coupling, and relates to the technical field of computers. The method comprises the steps of collecting multi-dimensional operation data in the operation process of medical waste treatment equipment, performing physical field feature mapping processing on the multi-dimensional operation data to obtain multi-dimensional processing data associated with multi-field features, and dynamically correcting model parameters through data and model deviation collaborative analysis; and analyzing the cross-field characteristics by using the temperature field reconstruction model, generating an equipment cavity global virtual temperature field and a temperature change trend thereof, and adjusting equipment operation parameters by combining multi-field coupling logic based on the temperature state of the core area so as to maintain the temperature in a preset interval. According to the invention, high-precision real-time reconstruction and intelligent control of the temperature field are realized, the problems of incomplete single-point measurement and response lag are solved, and the efficiency and safety of medical waste treatment are improved.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and more specifically, to a method and system for intelligent temperature control of medical waste treatment equipment based on multi-physics coupling. Background Technology

[0002] In the field of medical waste treatment, the efficient and stable operation of medical waste treatment equipment is crucial for ensuring public health safety. Among these aspects, temperature control during equipment operation is a key factor. Accurately monitoring the internal temperature of the equipment and rationally adjusting operating parameters directly affects the treatment effect of medical waste and energy utilization efficiency.

[0003] Currently, traditional medical waste treatment equipment generally uses conventional temperature sensors such as PT100 for single-point measurement. This single-point measurement method has many drawbacks. Due to the extremely complex temperature field distribution inside the equipment, single-point measurement cannot comprehensively and accurately reflect the temperature state throughout the entire device. Furthermore, traditional sensors have significant thermal inertia and slow response speeds. When the internal temperature changes, the sensors cannot promptly capture these dynamic temperature changes, causing the control system to be unable to adjust the equipment's operating parameters in a timely manner based on the actual temperature conditions. This not only affects the quality of medical waste treatment but also reduces energy efficiency and increases operating costs. Summary of the Invention

[0004] In view of the aforementioned problems, and in conjunction with the first aspect of the present invention, embodiments of the present invention provide a method for intelligent temperature control of medical waste treatment equipment based on multi-physics coupling, the method comprising:

[0005] Pre-construct a multi-physics coupling basic model for medical waste treatment equipment;

[0006] Collect multi-dimensional operational data during the operation of medical waste treatment equipment, and perform physical field feature mapping processing on the multi-dimensional operational data to obtain multi-dimensional processed data with associated multi-field features;

[0007] The multi-dimensional processed data is input into the multi-physics coupling basic model. Through the deviation analysis between the multi-field data and the model output, the heat conduction parameters and mechanical correlation coefficients in the multi-physics coupling basic model are adjusted to obtain a dynamically corrected multi-physics coupling model.

[0008] The pre-trained temperature field reconstruction model is invoked to perform cross-field feature analysis on the dynamically modified multiphysics coupling model, generating a target virtual temperature field for the entire cavity of the medical waste treatment equipment.

[0009] Based on the temperature distribution and temperature change trend of the core region of the cavity in the target virtual temperature field, the operating parameters of the medical waste treatment equipment are adjusted by combining multi-field coupling logic to maintain the temperature of the core region of the cavity within the preset operating range.

[0010] Furthermore, embodiments of the present invention also provide an intelligent temperature control system for medical waste treatment equipment based on multi-physics field coupling, characterized in that it includes:

[0011] A processor; a machine-readable storage medium for storing machine-executable instructions of the processor; wherein the processor is configured to execute the above-described intelligent temperature control method for medical waste treatment equipment based on multiphysics coupling by executing the machine-executable instructions.

[0012] In another aspect, embodiments of the present invention also provide a computer program product, the computer program product including machine-executable instructions stored in a computer-readable storage medium, the processor of the intelligent temperature control system for medical waste treatment equipment based on multi-physics coupling reading the machine-executable instructions from the computer-readable storage medium, the processor executing the machine-executable instructions, causing the intelligent temperature control system for medical waste treatment equipment based on multi-physics coupling to execute the above-described intelligent temperature control method for medical waste treatment equipment based on multi-physics coupling.

[0013] Based on the above, a multiphysics coupled basic model was constructed, incorporating the correlation between the thermal conduction characteristics and mechanical effects of the equipment cavity, friction mechanism, and motor components. This model simulated the thermal field and stress distribution under different operating conditions. Multi-dimensional operational data was then collected and processed using physical field feature mapping, enabling the data to better reflect the actual operating state of the equipment. The processed multi-dimensional data was input into the multiphysics coupled basic model, and the model parameters were dynamically corrected through deviation collaborative analysis. This resulted in a dynamically corrected multiphysics coupled model that better reflects the actual operating conditions of the equipment, effectively improving the accuracy and adaptability of the multiphysics coupled model. A pre-trained temperature field reconstruction model was used for cross-field feature analysis to generate a target virtual temperature field containing the temperature distribution and changing trends of various regions within the cavity. This enabled comprehensive, real-time, and high-precision monitoring of the internal temperature field of the equipment. Based on this target virtual temperature field, the equipment's operating parameters were adjusted to maintain the temperature in the core area of ​​the cavity within a preset range, achieving precise temperature control. This improved the medical waste treatment effect and energy utilization efficiency, significantly enhancing the sensitivity and accuracy of temperature monitoring in medical waste treatment equipment. Attached Figure Description

[0014] Figure 1 This is a schematic diagram of the execution flow of the intelligent temperature control method for medical waste treatment equipment based on multi-physics field coupling provided in an embodiment of the present invention.

[0015] Figure 2 This is a schematic diagram of exemplary hardware and software components of an intelligent temperature control system for medical waste treatment equipment based on multi-physics field coupling provided in an embodiment of the present invention. Detailed Implementation

[0016] The present invention will now be described in detail with reference to the accompanying drawings. Figure 1 This is a flowchart illustrating an intelligent temperature control method for medical waste treatment equipment based on multi-physics field coupling, according to an embodiment of the present invention. The following is a detailed description of this intelligent temperature control method for medical waste treatment equipment based on multi-physics field coupling.

[0017] Step S110: Pre-build a multi-physics coupled basic model of the medical waste treatment equipment;

[0018] In this embodiment, the application scenario is intelligent temperature monitoring of medical waste treatment equipment. The process of pre-constructing a multiphysics coupling basic model of the medical waste treatment equipment involves comprehensively considering the heat conduction and mechanical effects of various components of the equipment. The multiphysics coupling basic model includes the correlation between the heat conduction characteristics and mechanical effects of the equipment cavity, friction mechanism, and motor components. Furthermore, the multiphysics coupling basic model is obtained by simulating the heat field distribution and stress distribution under different operating conditions of the equipment.

[0019] Step S111: Based on the model features of the medical waste treatment equipment, extract the geometric dimensions of the equipment cavity, the assembly position of the friction mechanism, and the connection relationship between the motor and the shaft, construct a physical interaction network between equipment components, and identify the overlapping areas of heat conduction paths and mechanical action transmission paths;

[0020] In this embodiment, detailed model features of the medical waste treatment equipment are first obtained, including geometric dimensions such as the length, width, and height of the equipment cavity, for example, the length of the cavity is L1, the width is W1, and the height is H1. Simultaneously, the specific assembly position of the friction mechanism inside the cavity is determined, such as the distance from the left wall of the cavity being D1, and the height from the bottom being H2. The connection relationship between the motor and the shaft is also detailed in the drawings, such as the motor being connected to the shaft via a coupling, and the coupling model and connection method. Based on the above information, the various components of the equipment are abstracted as nodes, and the connection relationships between components are treated as edges, thereby constructing a physical interaction network between the equipment components. In this network, the heat conduction path is the channel for heat transfer, for example, heat generated from the friction mechanism is transferred outward through the cavity wall; the mechanical action transmission path is the path of force transmission, for example, the torque output by the motor is transferred to the friction mechanism through the shaft. By analyzing these two paths, their overlapping areas are identified. These overlapping areas are often where the thermal-mechanical coupling effect is more significant and need to be focused on in the subsequent modeling process.

[0021] Step S112: For the three key components of the equipment cavity wall, friction parts, and motor housing, collect data on the thermal conductivity, specific heat capacity, density, and elastic modulus of the materials under different temperature conditions, and construct a dynamic correlation model of material properties changing with temperature so that the material parameters match the temperature fluctuations in the actual operation of the equipment.

[0022] In this embodiment, the material used for the cavity wall of the equipment is ductile iron. By consulting material handbooks and relevant experimental data, data were collected for various temperatures, including K1 thermal conductivity, C1 specific heat capacity, ρ1 density, and E1 elastic modulus at temperature T1; and K2 thermal conductivity, C2 specific heat capacity, ρ2 density, and E2 elastic modulus at temperature T2. The friction components are made of wear-resistant alloy material, and corresponding material property data for different temperatures were also collected, such as K3 thermal conductivity, C3 specific heat capacity, ρ3 density, and E3 elastic modulus at temperature T3. The motor housing is generally made of cast iron, and data for K4 thermal conductivity, C4 specific heat capacity, ρ4 density, and E4 elastic modulus at temperature T4 were collected. Then, using the above data, a curve fitting method is employed to establish a functional relationship between material properties and temperature. For example, the relationship between thermal conductivity K and temperature T can be expressed as K = a × T + b (where a and b are fitting coefficients). Similarly, a dynamic correlation model between specific heat capacity, density, elastic modulus, and temperature is constructed. Therefore, when temperature fluctuations occur during equipment operation, the model can automatically adjust material parameters based on real-time temperature to match the actual situation.

[0023] Step S113: Based on the process requirements of medical waste treatment, determine the equipment operating conditions corresponding to different motor output power, different feed rates, and different ambient temperatures, establish operating condition classification rules, and mark the expected correlation direction of thermal field and stress field for each operating condition;

[0024] In this embodiment, the medical waste treatment process requires the equipment to operate under different conditions to achieve the best treatment effect. For example, when treating infectious medical waste, a higher treatment temperature and a longer treatment time are required, which corresponds to a higher motor output power; while when treating ordinary medical waste, the motor output power can be relatively lower. Based on the above process requirements, this embodiment determines the motor output power range to be P1 to P2, and divides this range into multiple power levels, such as P1, P1+ΔP, P1+2ΔP, ..., P2. The feed rate range is M1 to M2, also divided into multiple levels, such as M1, M1+ΔM, ..., M2. The ambient temperature considers the possible operating environment range of the equipment, such as Tmin to Tmax, and is also divided into multiple intervals. Different combinations of motor output power, feed rate, and ambient temperature constitute different equipment operating conditions. For each operating condition, this embodiment, based on experience and theoretical analysis, marks the expected correlation direction between the thermal field and the stress field. For example, when the output power of the motor increases, the heat generated by the friction mechanism increases, the thermal field intensity increases, and the force on the motor and shaft also increases, and the stress field intensity also increases accordingly. At this time, the expected correlation between the thermal field and the stress field is positively correlated. However, when the ambient temperature rises, it may affect the heat dissipation of the equipment, thereby increasing the thermal field intensity. But the effect on the stress field may be more complex and requires specific analysis and marking of its correlation direction.

[0025] Step S114: Based on the physical interaction network of equipment components, divide the three-dimensional geometric model of the equipment into regions, and mark the regions where heat conduction and mechanical action overlap as key coupling regions;

[0026] Based on the physical interaction network of device components constructed in step S111, this embodiment divides the three-dimensional geometric model of the device into regions. The division can be based on the component's function, location, and the degree of thermo-mechanical coupling. For example, the entire device can be divided into large regions such as the cavity region, motor region, friction mechanism region, and shaft region, with each large region further subdivided into several smaller sub-regions. After division, based on the overlap between heat conduction paths and mechanical action transmission paths, regions exhibiting both significant heat conduction and mechanical action are marked as key coupling regions. For example, the connection between the friction mechanism and the shaft involves both heat conduction due to friction and mechanical action due to torque transmitted by the shaft; therefore, this region is marked as a key coupling region. These key coupling regions require higher computational accuracy and attention in subsequent simulations and analyses.

[0027] Step S115: Embed the parameters in the dynamic correlation model of material properties into the modeling unit of the corresponding component, load different working condition parameters in sequence according to the working condition classification rules, simulate the diffusion process of the internal thermal field and the distribution process of the stress field under each working condition, and record the correspondence between the thermal field data and the stress field data.

[0028] In this embodiment, parameters from the dynamic correlation model of material properties constructed in step S112, such as thermal conductivity, specific heat capacity, density, and elastic modulus, are embedded into the modeling units of the corresponding components. For example, the dynamic property parameters of the cavity wall material are embedded in the modeling unit of the cavity wall, and the dynamic property parameters of the friction component material are embedded in the modeling unit of the friction component. Then, according to the working condition classification rules established in step S113, different working condition parameters are loaded sequentially. For each working condition, finite element analysis software is used to perform simulation calculations to simulate the diffusion process of the thermal field and the distribution process of the stress field inside the equipment. During the simulation, the software adjusts the material parameters of each component in real time according to the dynamic correlation model of material properties. After the simulation is completed, the thermal field data under the working condition, such as the temperature value and temperature gradient of each region, and the stress field data, such as the stress magnitude and stress direction of each region, are recorded, and the correspondence between the thermal field data and the stress field data is established, for example, the temperature value of each spatial location is correlated with the stress value at that location.

[0029] Step S116: Analyze the correspondence between the thermal field and the stress field under different working conditions, extract the coupling law between the heat conduction parameters and the mechanical correlation coefficient, and establish a parameter coupling rule base. The parameter coupling rule base contains the logic of the influence of the change of heat conduction parameters on the mechanical correlation coefficient under different working conditions.

[0030] Data on the correspondence between thermal and stress fields simulated under all operating conditions were collected and analyzed in depth. For example, the analysis examined how mechanical correlation coefficients, such as the elastic modulus, change when thermal conductivity increases under a specific operating condition; or the magnitude and trend of changes in mechanical correlation coefficients when thermal conductivity changes to a certain extent. Through extensive data statistics and analysis, the coupling rules between thermal conductivity parameters and mechanical correlation coefficients were extracted. For example, a functional relationship E=f(K) was found between thermal conductivity K and elastic modulus E, or that this relationship exhibits different forms within different temperature ranges. These coupling rules were organized into rules, establishing a parameter coupling rule base. Each rule in the rule base clearly defines the logic of the impact of changes in thermal conductivity parameters on mechanical correlation coefficients under specific operating conditions, such as "when the operating conditions are motor output power P3, feed rate M3, and ambient temperature T5, for every increase of ΔK in thermal conductivity, the elastic modulus increases by ΔE," etc.

[0031] Step S117: Based on the parameter coupling rule base, the simulated thermal field data and stress field data are correlated and integrated to construct a multi-physics coupling basic database, so that each data entry in the multi-physics coupling basic database contains operating condition parameters, thermal field characteristics, stress field characteristics and parameter coupling relationships.

[0032] Using the parameter coupling rule base established in step S116, the thermal field data and stress field data simulated in step S115 are correlated and integrated. For each set of operating parameters, such as motor output power P, feed rate M, and ambient temperature T, the corresponding thermal field characteristics (such as maximum temperature, average temperature, temperature distribution uniformity, etc.), stress field characteristics (such as maximum stress, average stress, stress concentration area, etc.), and parameter coupling relationships (such as the specific coupling law between heat conduction parameters and mechanical correlation coefficients under this operating condition) are integrated into a single data entry. Thus, each data entry in the multiphysics coupling basic database contains complete operating condition information, thermal field information, stress field information, and parameter coupling information.

[0033] Step S118: Establish a dynamic correlation model between thermal conduction characteristics and mechanical action through a multi-field feature mapping algorithm. With the support of a multi-physics coupling basic database, generate a multi-physics coupling basic model that includes the thermal-mechanical coupling relationship of the entire equipment. The multi-physics coupling basic model is used to output the corresponding thermal field distribution and stress distribution prediction results based on the input operating parameters.

[0034] Multi-field feature mapping (MMR) is an algorithm that maps and correlates thermal field features with stress field features. It uses a multi-physics coupling database as training data to establish a dynamic correlation model between thermal conductivity and mechanical effects. During model training, it continuously learns the relationship between thermal and stress fields under different operating conditions in the database, as well as the parameter coupling rules. The resulting multi-physics coupling database model can receive input operating parameters (such as motor output power, feed rate, ambient temperature, etc.) and, based on the dynamic correlation between thermal conductivity and mechanical effects within the model, output the corresponding predicted results for thermal and stress distributions. For example, when a specific operating parameter is input, the temperature value (thermal field distribution) and stress value (stress distribution) at various points inside the equipment can be predicted.

[0035] Step S120: Collect multi-dimensional operational data during the operation of the medical waste treatment equipment, and perform physical field feature mapping processing on the multi-dimensional operational data to obtain multi-dimensional processed data with associated multi-field features.

[0036] The multi-dimensional operational data includes temperature data at different locations within the equipment cavity, vibration data of the motor and shaft, and pressure data inside the cavity. For example, during the operation of medical waste treatment equipment, it is necessary to collect multi-dimensional operational data reflecting the equipment's operating status in real time. This multi-dimensional operational data includes temperature data at different locations within the equipment cavity, collected by temperature sensors (such as PT100 sensors) installed at different locations on the cavity wall; vibration data of the motor and shaft, collected by vibration sensors installed on the motor housing, shaft support, etc.; and pressure data inside the cavity, obtained by pressure sensors inside the cavity. After collecting this raw data, it is necessary to perform physical field feature mapping processing, associating the above data with physical field features such as thermal field and stress field, thereby obtaining multi-dimensional processed data with associated multi-field features, which can then be input into a multi-physics coupling model for analysis and calculation.

[0037] Step S121: Receive temperature data, vibration data, and pressure data transmitted in real time from each data acquisition component of the device. Based on the physical logic of the device operation, establish a spatiotemporal correlation between temperature data and vibration data. The spatiotemporal correlation reflects the time response logic of motor vibration changes and changes in the temperature around the motor.

[0038] The equipment's various data acquisition components, such as temperature sensors, vibration sensors, and pressure sensors, collect data in real time according to a set sampling frequency (e.g., 1kHz) and transmit it to the data processing center via communication methods such as industrial Ethernet. After receiving this data, the data processing center analyzes it based on the physical logic of the equipment's operation. For example, changes in motor vibration often cause changes in the temperature around the motor, because increased vibration may mean increased friction, thus generating more heat. By analyzing a large amount of historical data, a spatiotemporal correlation between temperature data and vibration data can be established. This correlation reflects the time response logic between changes in motor vibration and changes in the temperature around the motor. For instance, when the vibration amplitude of the motor increases by ΔV at time t1, the temperature around the motor will rise by ΔT at time t1+Δt, where Δt is the temperature response time to the vibration change.

[0039] Step S122: Based on the spatial structure and multi-physics coupling logic of the equipment cavity, the temperature data is mapped to the three-dimensional spatial coordinates of the cavity according to the acquisition position and associated with the thermal field characteristics of the corresponding area; the vibration data is mapped to the structural coordinates of the motor and shaft according to the acquisition position and associated with the stress field characteristics of the corresponding components; the pressure data is mapped to the spatial grid inside the cavity and associated with the thermal field-stress field interaction characteristics of the corresponding area.

[0040] The equipment cavity has a specific spatial structure, and a spatial coordinate system can be established based on its three-dimensional model. Temperature data collected by temperature sensors is mapped to the cavity's three-dimensional spatial coordinates according to the sensors' actual installation positions within the cavity. Each temperature data point corresponds to a specific spatial point coordinate (x, y, z). Simultaneously, based on multiphysics coupling logic, the thermal field characteristics of the region where this spatial point is located, such as temperature gradient and heat flux density, are associated with the temperature data. Similarly, vibration data is mapped to the structural coordinates of the motor and shaft based on the vibration sensor's installation position. For example, vibration data at a point on the motor housing corresponds to the (x1, y1, z1) point on the motor's structural coordinates, and is associated with the stress field characteristics of that component, such as the relationship between vibration amplitude and stress magnitude. Pressure data is mapped to a spatial grid inside the cavity, with each grid cell corresponding to a pressure value, and is associated with the thermal-stress field interaction characteristics of that region, as changes in pressure can affect heat conduction and stress distribution.

[0041] Step S123: Extract the dynamic change features from the mapped temperature data, vibration data, and pressure data. The dynamic change features include the rate of increase of temperature data, the frequency change of vibration data, and the fluctuation amplitude of pressure data. Establish the correlation constraints between the dynamic change features based on the multi-field coupling law.

[0042] Dynamic change features are extracted from the mapped temperature, vibration, and pressure data. For temperature data, the rate of temperature rise is calculated, i.e., the increase in temperature per unit time. For example, if the temperature rises from T0 to T1 in time Δt, the rate of rise is (T1-T0) / Δt. The dynamic change feature of vibration data can be its frequency variation. By performing spectral analysis on the vibration signal, the changes in different frequency components are obtained. The fluctuation amplitude of pressure data refers to the difference between the maximum and minimum pressure values ​​within a certain time. Based on the multi-field coupling law, there are correlation constraints among these dynamic change features. For example, an increase in the rate of temperature rise may lead to a change in the frequency of vibration data, and the fluctuation amplitude of pressure data may also change accordingly. By establishing the above correlation constraints, the changes in the operating state of the equipment can be described more comprehensively.

[0043] Step S124: Perform cross-field data anomaly correlation verification on dynamic change features according to correlation constraints, identify data where single data is normal but multi-field feature correlation is abnormal, and perform multi-field collaborative optimization on dynamic change features that pass the verification, and adjust data deviation based on the interaction relationship between thermal field features and stress field features.

[0044] Using the correlation constraints established in step S123 between dynamic change features, cross-field data anomaly correlation verification is performed on each dynamic change feature. For example, when the rate of temperature rise is within the normal range and the frequency of vibration data is also within the normal range, according to the correlation constraints, the relationship between the rate of temperature rise and the vibration frequency should satisfy a certain specific law. However, in the actual data, the above law is broken, indicating that there is a situation where a single data point is normal but the correlation of multiple field features is abnormal. For dynamic change features that pass the verification, i.e., the correlation between each feature meets expectations, multi-field collaborative optimization is then performed. Based on the interaction relationship between thermal field features and stress field features, such as thermal expansion causing stress changes, and stress changes potentially affecting heat conduction, the deviations in the data are adjusted to make the data more accurately reflect the actual operating status of the equipment.

[0045] Step S125: According to the dynamic cycle of equipment operation, integrate the optimized temperature data, vibration data, pressure data with the corresponding thermal field characteristics and stress field characteristics, and label the associated multi-field characteristic information for each time node;

[0046] The equipment operates within a dynamic cycle, such as a complete medical waste treatment cycle. According to this dynamic cycle, the temperature, vibration, and pressure data optimized in step S124 are integrated with their respective thermal field characteristics (such as temperature distribution and heat flow direction) and stress field characteristics (such as stress magnitude and strain). During the integration process, associated multi-field feature information is labeled for each time node (e.g., t0, t1, t2, ..., tn). For example, at time t0, the temperature data is T0, the corresponding thermal field characteristic is temperature gradient G0, the vibration data is V0, the corresponding stress field characteristic is stress value S0, the pressure data is P0, and the corresponding thermal-stress field interaction characteristic is I0, etc. This information is integrated to form a complete data record for that time node.

[0047] Step S126: Perform multi-field feature consistency verification on the integrated data to ensure that the thermal field features, stress field features and dynamic change features corresponding to each data entry conform to the multi-physics field coupling logic, and generate multi-dimensional processed data with associated multi-field features. The multi-dimensional processed data includes temperature data, vibration data, pressure data and corresponding multi-field feature labels.

[0048] The integrated data undergoes multi-field feature consistency verification, checking whether the thermal field features, stress field features, and dynamic change features of each data entry conform to the multi-physics coupling logic. For example, it checks whether the temperature distribution in the thermal field features matches the stress concentration area in the stress field features, and whether the temperature rise rate in the dynamic change features is consistent with the trend of vibration frequency change. If any data entry is found to be inconsistent with the multi-physics coupling logic, the data acquisition, mapping processing, or optimization process needs to be re-examined to identify and correct the problems. After consistency verification, corresponding multi-field feature labels are added to each data entry, such as thermal field feature labels (high temperature zone, normal temperature zone, etc.) and stress field feature labels (high stress zone, low stress zone, etc.). The final generated multi-dimensional processed data with associated multi-field features includes temperature data, vibration data, pressure data, and corresponding multi-field feature labels.

[0049] Step S130: Input the multi-dimensional processed data into the multi-physics coupling basic model, and through the deviation analysis between the multi-field data and the model output, adjust the heat conduction parameters and mechanical correlation coefficients in the multi-physics coupling basic model to obtain a dynamically corrected multi-physics coupling model.

[0050] The multi-dimensional processed data of the associated multi-field features obtained in step S120 is input into the multi-physics coupling basic model constructed in step S110. After receiving the input data, the multi-physics coupling basic model outputs the corresponding prediction results. By performing a deviation co-analysis between the actual multi-field data in the multi-dimensional processed data and the prediction results output by the model, the differences between the two are identified. Based on the above differences, the thermal conduction parameters (such as thermal conductivity, specific heat capacity, etc.) and mechanical correlation coefficients (such as elastic modulus, Poisson's ratio, etc.) in the multi-physics coupling basic model are adjusted so that the multi-physics coupling basic model can more accurately reflect the actual operating state of the equipment. After multiple adjustments and optimizations, a dynamically corrected multi-physics coupling model is obtained.

[0051] Step S131: Map the temperature data, vibration data, pressure data and corresponding multi-field feature labels in the multi-dimensional processed data to the thermal field input port, stress field input port and multi-field feature matching port of the multi-physics coupling basic model, respectively.

[0052] The multiphysics coupling fundamental model has different input ports, each used to receive different types of data. Temperature data from the multi-dimensional processed data is mapped to the model's thermal field input port according to the model's specified format and interface requirements; vibration data is mapped to the stress field input port; pressure data, depending on its correlation with the thermal and stress fields, may need to be mapped to corresponding input ports separately, or input through a dedicated integrated input port. Simultaneously, multi-field feature labels from the multi-dimensional processed data are mapped to the model's multi-field feature matching port for matching and comparing with the predicted feature labels output by the model.

[0053] Step S132: Run the multiphysics coupling basic model and output the predicted thermal field distribution, predicted stress distribution and predicted multi-field feature labels under the corresponding input data. Extract the temperature feature values ​​in the predicted thermal field distribution and the mechanical feature values ​​in the predicted stress distribution.

[0054] The multiphysics coupling basic model is launched, and it performs calculations and analyses based on the input temperature, vibration, and pressure data, as well as multi-field feature labels. After internal processing, the model outputs a predicted thermal field distribution corresponding to the input data, which describes the predicted temperature values ​​at various points inside the equipment. The predicted stress distribution provides the predicted stress levels for each component of the equipment. The predicted multi-field feature labels are feature labels related to the thermal and stress fields predicted by the model based on the input data. Temperature feature values, such as the predicted maximum, minimum, and average temperatures, are extracted from the predicted thermal field distribution; mechanical feature values, such as the maximum stress value, minimum stress value, and stress values ​​in stress concentration areas, are extracted from the predicted stress distribution.

[0055] Step S133: Extract the actual temperature feature value, actual mechanical correlation feature value and actual multi-field feature label from the multi-dimensional processed data, compare the deviation direction and deviation magnitude between the actual feature value and the model predicted feature value, and match the consistency between the actual multi-field feature label and the predicted multi-field feature label.

[0056] From the multi-dimensional data processing, actual temperature feature values ​​are extracted from the data records. These values ​​are obtained through actual sensor measurements and processing, such as the actual maximum temperature and the actual average temperature. Actual mechanical correlation feature values ​​are actual measured feature values ​​related to the stress field, such as the actual maximum stress value and the actual stress distribution uniformity. Actual multi-field feature labels are the real feature labels annotated during data processing. The extracted actual temperature feature values ​​are compared with the predicted temperature feature values ​​extracted in step S132 to determine their deviation direction (e.g., a positive deviation if the actual value is greater than the predicted value, and a negative deviation otherwise) and deviation magnitude (e.g., the size of the deviation value). Similarly, the deviation direction and magnitude of the actual mechanical correlation feature values ​​and the predicted mechanical feature values ​​are compared. Simultaneously, the actual multi-field feature labels are matched with the predicted multi-field feature labels to check their consistency, such as whether the type, quantity, and description of the labels are consistent.

[0057] Step S134: Based on the deviation direction, deviation magnitude, and label consistency results, and combined with the parameter coupling rule base in the multiphysics coupling basic model, determine the heat conduction parameters and mechanical correlation coefficients that need to be adjusted, and prioritize the parameter adjustments.

[0058] The model comprehensively considers the direction and magnitude of the deviation, as well as the consistency between the actual and predicted multi-field feature labels. If the deviation is significant or the label consistency is poor, it indicates that certain parameters in the model may require adjustment. The model incorporates a parameter coupling rule base within the multiphysics coupling foundation model, which stores the coupling patterns between thermal conductivity parameters and mechanical correlation coefficients under different operating conditions. Based on the current deviation and the rules in the rule base, it determines which thermal conductivity parameters (such as thermal conductivity and specific heat capacity) and mechanical correlation coefficients (such as elastic modulus and Poisson's ratio) need adjustment. For example, if there is a large positive deviation between the actual and predicted temperature feature values, and according to the rule base, the magnitude of thermal conductivity significantly affects the temperature distribution, then thermal conductivity may be one of the parameters requiring adjustment. Furthermore, based on factors such as the sensitivity of parameters to the model output and the magnitude of the deviation, the parameters with the greatest impact on the deviation are prioritized, such as adjusting the parameters most affected by the deviation first.

[0059] Step S135: Based on the deviation amplitude and parameter coupling law, calculate the coordinated adjustment ratio of heat conduction parameters and mechanical correlation coefficients, wherein the coordinated adjustment ratio reflects the interactive influence of changes in heat conduction parameters and changes in mechanical correlation coefficients;

[0060] Based on the deviation amplitude obtained in step S133 and the parameter coupling rules in the parameter coupling rule base, the coordinated adjustment ratio of the thermal conductivity parameter and the mechanical correlation coefficient is calculated. For example, assuming there is a coupling rule between thermal conductivity K and elastic modulus E, when K increases by ΔK, E will change by ΔE accordingly. Now, the deviation amplitude between the actual temperature characteristic value and the predicted value is ΔT. Based on the relationship between the deviation amplitude and thermal conductivity, the amount ΔK1 that the thermal conductivity needs to be adjusted can be determined. Then, according to the parameter coupling rules, the amount ΔE1 that the elastic modulus needs to be adjusted accordingly is calculated, thus obtaining the ratio ΔK1:ΔE1 of the adjustment amount of the thermal conductivity parameter (thermal conductivity) to the adjustment amount of the mechanical correlation coefficient (elastic modulus), i.e., the coordinated adjustment ratio. This coordinated adjustment ratio reflects the interactive influence between the change of thermal conductivity parameter and the change of mechanical correlation coefficient, ensuring that the two can coordinate with each other when adjusting parameters, jointly reducing the model prediction deviation.

[0061] Step S136: According to the coordinated adjustment ratio, the heat conduction parameters and mechanical correlation coefficients in the multiphysics coupling basic model are adjusted synchronously, and the numerical changes before and after the parameter adjustment and the corresponding multi-field feature correlation logic are recorded to obtain the adjusted multiphysics coupling basic model.

[0062] Based on the coordinated adjustment ratio calculated in step S135, the thermal conductivity parameters and mechanical correlation coefficients in the multiphysics coupling basic model are adjusted synchronously. For example, according to the ratio ΔK1:ΔE1, the thermal conductivity is adjusted from the original K_old to K_new=K_old+ΔK1, and the elastic modulus is adjusted from E_old to E_new=E_old+ΔE1. During the adjustment process, the numerical changes of each parameter before and after the adjustment are recorded in detail, such as K_old, K_new, E_old, E_new, etc., as well as the changes in the multiphysics characteristic correlation logic corresponding to these parameter adjustments, such as how the correlation between the thermal field and the stress field changes due to the parameter adjustment. After the above adjustments, the adjusted multiphysics coupling basic model is obtained.

[0063] Step S137: Run the adjusted multiphysics coupling basic model, output new predicted thermal field distribution, predicted stress distribution and predicted multi-field feature labels, extract the predicted feature values ​​again and compare them with the actual feature values ​​in the multi-dimensional processed data, and analyze the changing trend of multi-field deviation.

[0064] The adjusted multiphysics coupling basic model is launched, with the same multidimensional processed data as before. After the model runs, it outputs new predicted thermal field distribution, predicted stress distribution, and predicted multi-field feature labels. Predicted temperature and mechanical feature values ​​are extracted again from the new prediction results and compared with the actual temperature and mechanical correlation feature values ​​in the multidimensional processed data. The comparison results are analyzed to observe the changing trend of the multi-field deviation, i.e., whether the deviation direction has changed and whether the deviation amplitude has increased or decreased. For example, if the deviation amplitude of the temperature feature value before adjustment is ΔT1, and the deviation amplitude after adjustment becomes ΔT2, if ΔT2 < ΔT1, it indicates that the deviation amplitude is decreasing, and the adjustment has had a positive effect.

[0065] Step S138: If the biases of multiple fields all tend to decrease and the consistency of feature labels improves, then continue to optimize the parameters according to the current adjustment logic; if the bias of any physical field increases or the label consistency decreases, then recalculate the collaborative adjustment ratio and adjust the parameter coupling rules.

[0066] The judgment is made based on the multi-field bias trend and feature label consistency analysis in step S137. If the biases of all physical fields (such as thermal field and stress field) are decreasing, and the consistency between the actual multi-field feature labels and the predicted multi-field feature labels is improving, it indicates that the current parameter adjustment logic is effective, and the model parameters can continue to be optimized according to this logic, such as further fine-tuning the parameters to achieve a smaller bias. Conversely, if the bias of any physical field increases after adjustment, or the consistency of feature labels decreases, it indicates that there may be a problem with the current collaborative adjustment ratio or parameter coupling rule. The collaborative adjustment ratio needs to be recalculated, and the relevant rules in the parameter coupling rule base may need to be adjusted and corrected to adapt to the new situation.

[0067] Step S139: Repeat the parameter adjustment, model running, and deviation analysis steps until the deviations of multiple fields are all within a reasonable range and the consistency of feature labels reaches the preset standard, thus obtaining the dynamically corrected multiphysics coupling model.

[0068] Following steps S136 to S138, the parameter adjustment, model running, and deviation analysis are repeatedly performed. In each cycle, the parameters are adjusted based on the previous deviation, then the model is run and the new deviation is analyzed. This process is repeated until the deviations of all physical fields are within a preset reasonable range, such as temperature deviation not exceeding ±ΔT, stress deviation not exceeding ±ΔS, etc., and the consistency between the actual multi-field feature labels and the predicted multi-field feature labels reaches a preset standard, such as a consistency matching rate of over 90%. When these conditions are met, the parameter adjustment process is stopped, and the resulting model is the dynamically corrected multiphysics coupling model.

[0069] Step S1310: Record the parameter adjustment process and final parameter values ​​of the dynamically corrected multiphysics coupling model, and establish a correlation file between parameter adjustment and multi-field deviation changes;

[0070] After obtaining the dynamically corrected multiphysics coupling model, the adjustment process of the model parameters is recorded in detail, including the parameter name for each adjustment, the value before adjustment, the value after adjustment, and the basis for adjustment (such as deviation, coordination adjustment ratio, etc.). Simultaneously, the final determined values ​​of each heat conduction parameter and mechanical correlation coefficient are recorded. This information is then compiled to establish a correlation file between parameter adjustments and changes in multiphysics deviation. This correlation file can demonstrate how parameter adjustments affect changes in multiphysics deviation and also facilitates rapid adjustment of model parameters when significant changes occur in equipment operating conditions.

[0071] Step S140: Call the pre-trained temperature field reconstruction model to perform cross-field feature analysis on the dynamically modified multiphysics coupling model to generate the target virtual temperature field of the entire cavity of the medical waste treatment equipment;

[0072] In this embodiment, the target virtual temperature field includes the temperature distribution and temperature change trends of each region of the cavity. While the dynamically modified multiphysics coupling model can provide information such as the thermal and stress fields of the device, a full-domain target virtual temperature field needs to be generated to gain a more comprehensive understanding of the temperature conditions inside the cavity. By calling a pre-trained temperature field reconstruction model, cross-field feature analysis is performed on the dynamically modified multiphysics coupling model. This involves comprehensively analyzing the thermal field features, stress field features, and other relevant physical field features in the model. Utilizing the learning and reasoning capabilities of the temperature field reconstruction model, a target virtual temperature field describing the entire cavity of the medical waste treatment device is generated. This target virtual temperature field not only includes the current temperature distribution of each region of the cavity but also predicts temperature change trends.

[0073] Step S141: Extract the thermal field distribution characteristics, stress field distribution characteristics, material thermal property characteristics, and multi-field coupling rules from the dynamically modified multi-physics coupling model, and establish the correlation mapping relationship between thermal field characteristics and stress field characteristics based on multi-field collaborative logic;

[0074] From the dynamically modified multiphysics coupling model, the following characteristics are extracted: thermal field distribution features, such as temperature values, temperature gradients, and heat flux densities in each region; stress field distribution features, such as stress magnitude, stress direction, and strain distribution; material thermal properties, i.e., the characteristics of thermal conductivity, specific heat capacity, etc., of each component material in the model as a function of temperature; and multi-field coupling rules, i.e., the coupling laws between the thermal field and stress field in the model. Based on the multi-field synergistic logic, i.e., the laws governing the mutual influence and interaction between the thermal field and stress field, a correlation mapping relationship between thermal field features and stress field features is established. For example, analysis reveals that when the stress in a certain region increases by ΔS, the temperature in that region will increase by ΔT, thus establishing a correlation mapping between stress change and temperature change.

[0075] Step S142: Integrate the thermal field distribution characteristics, stress field distribution characteristics, and material thermal property characteristics according to the correlation mapping relationship to generate a multi-field fusion feature set, wherein the multi-field fusion feature set contains coupling relationship information between each feature;

[0076] Based on the correlation mapping relationship established in step S141, the thermal field distribution characteristics, stress field distribution characteristics, and material thermal property characteristics are integrated. During the integration process, not only is the numerical information of each characteristic included, but also the coupling relationship information between them is preserved. For example, the temperature value T, stress value S, material thermal conductivity K at a certain spatial location, and their correlation relationships (such as the functional relationship between T and S, the influence of K on T, etc.) are integrated to form a fused feature unit. Multiple such feature units are combined to form a multi-field fused feature set, which comprehensively reflects the physical field characteristics inside the equipment and their mutual coupling relationships.

[0077] Step S143: Input the multi-field fusion feature set into the feature encoding module of the pre-trained temperature field reconstruction model to generate a temperature field association encoding vector;

[0078] The pre-trained temperature field reconstruction model comprises multiple functional modules, one of which is the feature encoding module. The multi-field fusion feature set generated in step S142 is input into the feature encoding module. The feature encoding module processes the input multi-field fusion feature set, performing extraction, transformation, and compression operations on each feature to map the high-dimensional multi-field fusion feature set into a low-dimensional temperature field correlation encoding vector. This temperature field correlation encoding vector contains key information from the multi-field fusion feature set and can represent features such as the thermal field, stress field, and material thermal properties, as well as their coupling relationships.

[0079] Step S144: Based on the spatial structure and multi-physics coupling logic of the equipment cavity, the temperature field correlation encoding vector is mapped to the three-dimensional spatial grid of the cavity, the correspondence between encoding features and spatial positions is established, and the multi-field feature weights of different spatial positions are divided.

[0080] Based on the spatial structure of the equipment cavity, such as its shape, size, and the distribution of internal components, a three-dimensional spatial mesh is constructed, dividing the cavity into multiple small mesh units, each corresponding to a specific spatial location (x, y, z). Based on multiphysics coupling logic, each element in the temperature field association encoding vector is mapped to the corresponding mesh unit in the cavity's three-dimensional spatial mesh according to certain rules, thus establishing a correspondence between encoded features and spatial locations. Simultaneously, based on the importance of different spatial locations in the thermal and stress fields, and the degree of influence of multiphysics features on the temperature field, multiphysics feature weights are assigned to different spatial locations. For example, for spatial locations near the friction mechanism, due to their strong thermo-mechanical coupling, the multiphysics feature weights can be set higher; while for locations far from the heat source and stressed components, the weights can be relatively lower.

[0081] Step S145: Based on the correspondence between coding features and spatial location, combined with the thermal characteristics of materials and multi-field coupling rules, calculate the initial temperature value of each spatial grid node, and initially construct the temperature field framework of the entire cavity.

[0082] Based on the correspondence between the encoded features and spatial locations established in step S144, and the multi-field feature weights for each spatial location, combined with material thermal properties (such as thermal conductivity and specific heat capacity of the material corresponding to each grid cell) and multi-field coupling rules (such as heat conduction equations and thermo-mechanical coupling equations), the initial temperature value of each spatial grid node is calculated. During the calculation, the influence of the multi-field information represented by the encoded features on temperature, as well as the material thermal properties and multi-field coupling effects, are considered. Through the calculation of the initial temperature values ​​of all spatial grid nodes, a preliminary temperature field framework for the entire cavity is constructed, which provides a preliminary temperature estimate for each grid node within the cavity.

[0083] Step S146: Using the actual temperature data and corresponding multi-field feature labels in the multi-dimensional processing data, the temperature of the corresponding spatial node in the temperature field framework is corrected, and the temperature value of the node not covered by the actual data is adjusted based on the multi-field feature correlation of the surrounding nodes.

[0084] The multi-dimensional data processing includes temperature data actually measured by sensors, corresponding to specific spatial node locations within the cavity. This actual temperature data and its corresponding multi-field feature labels are matched with the corresponding spatial nodes in the temperature field framework. For nodes with actual temperature data, the initial temperature value of that node in the framework is corrected using the actual temperature value, making the corrected temperature value closer to the actual situation. For nodes not covered by actual data, their temperature values ​​are adjusted based on the temperature values ​​of their surrounding corrected nodes and the multi-field feature relationships between these nodes (such as the proximity effect of heat conduction, the continuity of stress distribution, etc.). For example, if the surrounding nodes of an uncovered node have higher temperatures, and the thermal conductivity of this node is similar to that of its surrounding nodes, it can be inferred that the temperature of this node is also relatively high, and its temperature value is adjusted accordingly.

[0085] Step S147: Analyze the temperature change trend of each region in the corrected temperature field framework, combine the stress field distribution and multi-field coupling rules output by the dynamically corrected multi-physics coupling model, adjust the temperature of the region in the temperature field that does not match the stress field characteristics, and perform global optimization of the temperature field to generate a target virtual temperature field for the entire cavity of the medical waste treatment equipment; the target virtual temperature field includes the temperature distribution state, temperature change trend and corresponding multi-field feature annotations of each region of the cavity;

[0086] The corrected temperature field framework is analyzed to observe the temperature change trends in each region, such as which regions are experiencing temperature increases, which are experiencing temperature decreases, and the rates of increase or decrease. Simultaneously, the matching between the temperature field and stress field is analyzed by combining the stress field distribution output from the dynamically corrected multiphysics coupling model. If a mismatch is found between the temperature characteristics and stress field characteristics in certain regions of the temperature field—for example, according to the stress field distribution, this region should have significant stress concentration and a correspondingly high temperature, but the temperature in this region is low—it indicates that the temperature field may be unreasonable. In this case, the temperature of the aforementioned mismatched regions is adjusted according to the multiphysics coupling rules. After local adjustments, the entire temperature field is then globally optimized to make the overall temperature field distribution more reasonable, smooth, and in line with physical laws. The final generated target virtual temperature field includes detailed information on the temperature distribution status of each region of the cavity (such as the temperature values ​​of each grid node), temperature change trends (such as temperature change curves over time), and corresponding multiphysics feature annotations (such as thermal field feature labels and stress field feature labels for this region).

[0087] Step S148: Collect historical multi-dimensional data and corresponding historical measured temperature field data of the medical waste treatment equipment under different operating conditions. The historical multi-dimensional data includes historical temperature data, historical vibration data, historical pressure data and corresponding historical multi-field feature labels. The historical measured temperature field data is a set of measured temperatures at multiple locations in the cavity.

[0088] To train the temperature field reconstruction model, a large amount of historical data needs to be collected. This data comes from the operating records of the medical waste treatment equipment under different operating conditions, including different motor output power, different feed rates, and different ambient temperatures. The historical multi-dimensional data includes historical temperature data (temperature values ​​measured by temperature sensors in the past), historical vibration data (historical records of vibration sensors), historical pressure data (historical data from pressure sensors), and corresponding historical multi-field feature labels (thermal field and stress field feature labels marked during past operation). The historical measured temperature field data is a collection of temperatures obtained through actual measurements at multiple locations within the cavity. These locations cover as many different areas of the cavity as possible to reflect the actual temperature distribution within the cavity.

[0089] Step S149: Based on the multi-field coupling law, feature extraction is performed on historical multi-dimensional data to obtain historical thermal field correlation features, historical stress field correlation features, and historical material thermal property features. The historical thermal field correlation features, historical stress field correlation features, and historical material thermal property features are then integrated into a historical multi-field fusion feature set.

[0090] Based on the multi-field coupling principle, feature extraction is performed on the collected historical multi-dimensional data. For historical temperature data, historical thermal field correlation features reflecting thermal field characteristics are extracted, such as historical maximum temperature, historical temperature change rate, and historical temperature distribution uniformity. Historical vibration data is analyzed to extract historical stress field correlation features related to the stress field, such as historical vibration frequency, historical vibration amplitude, and the relationship between stress. Historical material thermal characteristic features are extracted based on material property data at historical operating temperatures. The extracted historical thermal field correlation features, historical stress field correlation features, and historical material thermal characteristic features are integrated according to a certain structure and rules to form a historical multi-field fusion feature set, with each set corresponding to multi-field feature information under a specific historical operating condition.

[0091] Step S1410: Extract spatial features from historical measured temperature field data to obtain the temperature feature value of each spatial node, the temperature correlation between nodes and the corresponding spatial multi-field feature weights, and construct a historical temperature field feature model.

[0092] Historical measured temperature field data is a collection of measured temperatures from multiple locations. Spatial feature extraction is performed on this data. First, the spatial node corresponding to each measured temperature is determined, and the temperature feature values ​​of each spatial node are obtained, such as the measured temperature value and temperature fluctuation range. Then, the temperature correlation between different spatial nodes is analyzed, such as the temperature difference between adjacent nodes and the consistency of temperature change trends. Simultaneously, based on historical operating conditions and multi-field coupling patterns, the spatial multi-field feature weights corresponding to each spatial node are determined. These weights reflect the importance of the node in the historical temperature field and the degree of influence of multi-field features on its temperature. The above spatial feature information is integrated to construct a historical temperature field feature model, which can describe the spatial distribution characteristics and correlations of the historical measured temperature field.

[0093] Step S1411: Combine the historical multi-field fusion feature set with the corresponding historical temperature field feature model to form training sample pairs, and classify the training sample pairs according to the working condition type;

[0094] The historical multi-field fusion feature set obtained in step S149 is matched one-to-one with the corresponding historical temperature field feature model constructed in step S1410 to form training sample pairs. Each training sample pair contains input data (historical multi-field fusion feature set) and expected output data (historical temperature field feature model). Then, the above training sample pairs are classified according to the equipment's operating conditions, such as high-power high-feed-rate conditions, low-power low-feed-rate conditions, and different ambient temperature conditions. The purpose of this classification is to enable the model to learn for different operating conditions during training, thereby improving the model's adaptability and accuracy under various operating conditions.

[0095] Step S1412: Construct the network structure of the temperature field reconstruction model, wherein the network structure of the temperature field reconstruction model includes a feature encoding module, a spatial mapping module, a temperature calculation module, and an optimization and adjustment module;

[0096] To meet the requirements of temperature field reconstruction, a network structure for the temperature field reconstruction model was designed and constructed. This network structure comprises several functional modules: a feature encoding module encodes the input multi-field fusion feature set to generate low-dimensional feature vectors; a spatial mapping module maps the encoded feature vectors to a three-dimensional spatial grid of the cavity, establishing a correspondence between features and spatial locations, and assigning multi-field feature weights; a temperature calculation module calculates the initial temperature value of each spatial node based on the spatial mapping results, material thermal properties, and multi-field coupling rules, constructing a temperature field framework; and an optimization and adjustment module uses actual temperature data to correct and optimize the temperature field framework, improving the accuracy of the temperature field. These modules collaborate to complete the temperature field reconstruction process.

[0097] Step S1413: The classified training samples are used to reconstruct the model for the input temperature field. The feature encoding module encodes the historical multi-field fusion feature set. The spatial mapping module maps the encoded features to the spatial grid and assigns corresponding multi-field feature weights. The temperature calculation module calculates the initial temperature field based on material properties and multi-field coupling rules. The optimization and adjustment module corrects the temperature field based on the historical temperature field feature model.

[0098] The training samples, categorized by operating condition, are input one by one into the temperature field reconstruction model for training. First, the feature encoding module receives a set of historical multi-field fusion features, encodes them using algorithms such as neural networks, extracts key features, and generates encoded feature vectors. Next, the spatial mapping module maps the encoded feature vectors onto a pre-constructed three-dimensional spatial grid of the cavity, assigning corresponding multi-field feature weights to each spatial grid node based on the spatial multi-field feature weight information in the historical temperature field feature model. Then, the temperature calculation module uses the encoded features, spatial weights, material properties (obtained from the historical multi-field fusion feature set), and multi-field coupling rules (such as the heat conduction equation) to calculate the initial temperature value for each spatial grid node, forming the initial temperature field. Finally, the optimization and adjustment module compares the initial temperature field with the historical temperature field feature model, calculates the differences between the two, and adjusts the model parameters through backpropagation and other methods to correct the initial temperature field, making the reconstructed temperature field closer to the historical measured temperature field.

[0099] Step S1414: During the training process, based on the difference between the temperature field output by the temperature field reconstruction model and the historical temperature field feature model, adjust the parameters of each module of the temperature field reconstruction model, optimize the multi-field feature interaction algorithm and the spatial multi-field feature weight allocation rule, so that the temperature field reconstruction model can learn to capture the influence of multi-field coupling on the temperature field.

[0100] During the training and iteration of the temperature field reconstruction model, the differences between the temperature field output by the model and the historical temperature field feature models are continuously calculated, such as mean squared error and mean absolute error. Based on these differences, optimization algorithms (such as gradient descent) are used to adjust the parameters of each module of the model, such as the neural network weights and biases in the feature encoding module, the weight allocation parameters in the spatial mapping module, and the calculation coefficients in the temperature calculation module. Simultaneously, the multi-field feature interaction algorithm is optimized, that is, the calculation method of the interaction between different physical field features is improved, enabling the model to more accurately simulate the multi-field coupling effect. Furthermore, the spatial multi-field feature weight allocation rules are optimized. Based on the actual impact of multi-field features on the temperature field under different operating conditions, the weight values ​​at each spatial location are adjusted, enabling the model to better learn and capture the influence of multi-field coupling on the temperature field.

[0101] Step S1415: Through multiple rounds of training, the temperature field output by the temperature field reconstruction model is made to match the historical measured temperature field data to the preset requirements. At the same time, the temperature field reconstruction model can stably output a temperature field containing multi-field feature annotations under different working conditions.

[0102] The training process requires multiple iterations. Each iteration uses all training samples to train the model and adjust its parameters. After multiple training iterations, the model's predictive ability gradually improves, and the consistency between the output temperature field and historical measured temperature field data continuously increases. Training stops when the consistency reaches a preset requirement (e.g., the mean squared error is less than a preset threshold). Simultaneously, during training, it is crucial to ensure that the temperature field reconstruction model can stably output a temperature field containing multi-field feature annotations under different operating conditions; that is, the model possesses good generalization ability and can adapt to various possible operating conditions.

[0103] Step S1416: Collect new operating data of medical waste treatment equipment, generate a new validation sample set, and use the new validation sample set to test the generalization ability of the temperature field reconstruction model. If the generalization ability reaches the preset standard, a pre-trained temperature field reconstruction model is obtained; if it does not reach the preset standard, new operating condition training samples are added to continue training.

[0104] To verify the generalization ability of the temperature field reconstruction model, i.e., its adaptability to new operating conditions not encountered during training, operational data of the medical waste treatment equipment under these new conditions were collected. These new operating conditions could be combinations of factors not previously covered, such as motor output power, feed rate, and ambient temperature. A new validation sample set was generated using this new operating data. Each sample pair in the set consisted of a multi-field fusion feature set under the new operating conditions and the corresponding measured temperature field feature model. The new validation sample set was then input into the trained temperature field reconstruction model, and the degree of agreement between the model's output temperature field and the measured temperature field was observed. If the agreement reached a preset standard (e.g., the mean square error under the new operating conditions was still less than a preset threshold), the model's generalization ability met the requirements, and a pre-trained temperature field reconstruction model was obtained. If the preset standard was not met, training samples under the new operating conditions needed to be added to the training set for further training until the model's generalization ability under the new operating conditions met the requirements.

[0105] Step S150: Based on the temperature distribution and temperature change trend of the core region of the cavity in the target virtual temperature field, the operating parameters of the medical waste treatment equipment are adjusted in combination with multi-field coupling logic to maintain the temperature of the core region of the cavity within the preset operating range.

[0106] The target virtual temperature field provides temperature information for the entire cavity, with the temperature in the core region being crucial for the effectiveness of medical waste treatment and needing to be maintained within a preset operating range (e.g., 135℃ ± 5℃). Based on the temperature distribution (current temperature values ​​at various points) and temperature change trends (whether the temperature is rising, falling, or stabilizing) in the core region of the cavity within the target virtual temperature field, and combined with multi-field coupling logic (such as the impact of motor speed changes on frictional heat generation, and the impact of feed rate on the amount of material and heat transfer within the cavity), a strategy for adjusting motor speed and feed rate is constructed. By adjusting these two parameters, the temperature in the core region is controlled and kept within the preset range. The equipment operating parameters may include, but are not limited to, motor speed, motor forward and reverse rotation frequency, liquid water spray flow rate and spray response cycle, and characteristics of subsequent continuous feeding, etc.

[0107] Step S151: Analyze the target virtual temperature field, determine the spatial range of the cavity core region, extract the temperature distribution characteristics, temperature change trends and corresponding multi-field feature annotations within the cavity core region, and identify the temperature control target of the cavity core and the associated multi-field influencing factors.

[0108] A detailed analysis of the target virtual temperature field is conducted. Based on the process requirements of medical waste treatment and the structural characteristics of the equipment, the spatial range of the core area of ​​the cavity is determined. For example, the core area of ​​the cavity can be defined as the area where the friction mechanism is located and the space within a certain range around it. The temperature in this area directly affects the treatment effect of medical waste. After determining the spatial range, the temperature distribution characteristics within the core area are extracted, such as the highest temperature, lowest temperature, average temperature, and temperature distribution uniformity; the temperature change trend, i.e., how the temperature in the core area changes over time, whether it gradually increases, decreases, or remains stable; and the corresponding multi-field feature annotations, such as thermal field feature labels (high temperature zone, heat flow direction, etc.) and stress field feature labels (high stress zone, stress concentration, etc.). At the same time, the temperature control target of the cavity core is identified, i.e., the preset core area temperature operating range (e.g., T_target±ΔT). The related multi-field influencing factors refer to various physical field factors that affect the temperature of the core area, such as motor output power (affecting frictional heat generation), feed rate (affecting the amount of material and residence time, thus affecting heat transfer and reaction exothermics), ambient temperature (affecting equipment heat dissipation), and cavity pressure (affecting the physicochemical state of the material and heat transfer efficiency), etc.

[0109] Step S152: Based on the dynamic correction multiphysics coupling model and multi-field coupling rules, combined with the temperature control target of the cavity core, determine the allowable fluctuation range of the core area temperature. Based on the associated multi-field influencing factors, identify the stress field and pressure field related parameters that may be affected by adjusting the motor speed and feeding speed. Derive the parameter correlation logic between the core area temperature change and the motor speed and feeding speed.

[0110] Using the dynamic correction multi - physical - field coupling model and multi - field coupling rules, simulate the temperature changes in the core area under different conditions. Combining the temperature control target of the cavity core (such as T_target ± ΔT), determine the allowable fluctuation range of the temperature in the core area, that is, the temperature can fluctuate between T_min and T_max (T_min = T_target - ΔT, T_max = T_target + ΔT). Based on the associated multi - field influencing factors identified in step S151, analyze the possible impacts of the adjustment of the motor speed and the feeding speed on other physical fields. For example, an increase in the motor speed will cause an increase in the frictional heat generation of the friction mechanism, thereby enhancing the intensity of the thermal field. At the same time, the stress on the motor and the rotating shaft will also increase, that is, affecting the stress - field parameters; the change in the feeding speed will affect the filling rate of the materials inside the cavity, and further affect the cavity pressure, affecting the pressure - field parameters. Through the analysis of the above - mentioned impacts, deduce the parameter - association logic between the temperature change in the core area and the motor speed and the feeding speed. For example, deduce the relationship that when the motor speed increases by ΔN, the temperature in the core area rises by ΔT1; when the feeding speed increases by ΔF, the temperature change in the core area is ΔT2, etc., and establish a quantitative or qualitative association logic between the temperature and the motor speed and the feeding speed.

[0111] Step S153: According to the temperature change trend and the parameter - association logic, combined with the temperature control target, judge the current temperature deviation direction, and based on the multi - field influencing factors, screen the adjustable target parameters to determine the target parameters that need to be adjusted;

[0112] Observe the temperature change trend in the core area of the target virtual temperature field. If the current temperature is rising and has approached or exceeded T_max, or the current temperature is falling and has approached or fallen below T_min, combined with the temperature control target, judge the current temperature deviation direction. For example, if the current temperature is T_current, if T_current > T_max, the deviation direction is a positive deviation; if T_current < T_min, it is a negative deviation. Based on the multi - field influencing factors, analyze which parameters can be adjusted to change the temperature in the core area, and screen out the adjustable target parameters. In this embodiment, the main adjustable target parameters are the motor speed and the feeding speed. According to the current temperature deviation direction and the parameter - association logic, determine the target parameters that need to be adjusted. For example, if it is a positive deviation (the temperature is too high), according to the parameter - association logic, reducing the motor speed can reduce the frictional heat generation and thus lower the temperature. Therefore, determine that the motor speed is one of the target parameters that need to be adjusted; or increase the feeding speed. By introducing more low - temperature materials to absorb heat, it may also lower the temperature. At this time, the feeding speed may also become an adjustment target.

[0113] Step S154: Query the preset temperature-speed correlation table and the preset temperature-feed speed correlation table to obtain the initial adjustment range corresponding to the current temperature sub-interval. The initial adjustment range includes the initial speed adjustment range and the initial feed speed adjustment range.

[0114] The preset temperature-speed correlation table and temperature-feed rate correlation table are pre-constructed based on historical data and experimental results. The temperature-speed correlation table records the relationship between different temperature sub-ranges and the corresponding motor speed adjustment range; the temperature-feed rate correlation table records the relationship between different temperature sub-ranges and the feed rate adjustment range. Based on the current temperature value T_current of the core region of the cavity, its corresponding temperature sub-range is determined. For example, the temperature sub-range can be divided into T... <T_min-ΔT1、T_min-ΔT1≤T<T_min、T_min≤T≤T_max、T_max<T≤T_max+ΔT1、T> T_max+ΔT1, etc. Query the preset association table to obtain the initial speed adjustment range (e.g., the motor speed can be reduced from the current speed N_current by ΔN1 to N_current-ΔN1, or increased by ΔN2 to N_current+ΔN2) and the initial feed rate adjustment range (e.g., the feed rate can be increased from the current speed F_current by ΔF1 to F_current+ΔF1, or decreased by ΔF2 to F_current-ΔF2).

[0115] Step S155: Based on the dynamic correction multiphysics coupling model, set the simulation termination condition with the temperature control target as the benchmark, combine the multi-field influencing factors to screen the simulation scenario, simulate the temperature change results and multi-field characteristic changes in the core area under the adjustment range of different target parameters, and establish the correlation model of parameter adjustment range, temperature change amount, and multi-field characteristic change amount.

[0116] Simulation analysis was performed using a dynamically modified multiphysics coupling model. Based on the temperature control target, simulation termination conditions were set, such as stopping the simulation when the core region temperature entered a preset operating range (T_min to T_max) and stabilized for a period of time. Combining the associated multi-field influencing factors identified in step S151, possible simulation scenarios were selected. For example, combinations of different motor speed adjustment ranges and feed rate adjustment ranges were considered as simulation scenarios. For each simulation scenario, i.e., different target parameter adjustment ranges (such as motor speed adjustments ΔN3, ΔN4, etc., and feed rate adjustments ΔF3, ΔF4, etc.), the dynamically modified multiphysics coupling model was run to simulate the temperature change results in the core region (such as a decrease in temperature ΔT3, an increase in temperature ΔT4, etc.) and the changes in multi-field characteristics (such as a change in stress value ΔS3 in the stress field, a change in pressure value ΔP3, etc.). By analyzing the results of multiple simulation scenarios, a correlation model was established between parameter adjustment ranges (motor speed adjustment range ΔN, feed rate adjustment range ΔF), temperature changes (ΔT), and changes in multi-field characteristics (ΔS, ΔP, etc.). This correlation model can be represented by functional relationships, such as ΔT=f(ΔN, ΔF), ΔS=g(ΔN), ΔP=h(ΔF), etc.

[0117] Step S156: Based on the correlation model and the initial adjustment range, calculate the adjustment range required to bring the core area temperature back to the preset range. The adjustment range includes the motor speed adjustment range and the feed speed adjustment range. At the same time, verify whether the multi-field characteristic changes corresponding to the adjustment range are within the range allowed for safe operation of the equipment based on multiple influencing factors, and generate motor speed adjustment instructions and feed speed adjustment instructions. The motor speed adjustment instructions and feed speed adjustment instructions include the adjustment direction, adjustment range, and the stable maintenance time after adjustment.

[0118] Based on the correlation model established in step S155 and the initial adjustment range obtained in step S154, the required motor speed adjustment and feed rate adjustment range are calculated to bring the core area temperature back from the current temperature T_current to the preset operating range (T_min to T_max). For example, if the current temperature T_current > T_max, it is necessary to calculate the reduction of motor speed ΔN or the increase of feed rate ΔF, so that the temperature decreases by ΔT = T_current - T_target. According to the correlation model ΔT = f(ΔN, ΔF), the appropriate ΔN and ΔF are solved. After calculating the adjustment range, based on multiple influencing factors (such as the maximum allowable stress and maximum allowable pressure of the equipment), it is verified whether the multi-field characteristic changes (such as stress change ΔS and pressure change ΔP) corresponding to the adjustment range are within the allowable range for safe operation of the equipment. If they are within the safe range, motor speed adjustment commands and feed rate adjustment commands are generated. The adjustment instructions specify the direction of adjustment (e.g., reducing motor speed, increasing feed speed), the adjustment range (e.g., reducing ΔN, increasing ΔF), and the stable maintenance time after adjustment (i.e., the time to maintain the speed and feed speed after adjustment to observe whether the temperature is stable within the target range).

[0119] Step S157: Send the motor speed adjustment command and the feed speed adjustment command to the equipment control module, receive the operation feedback data after the equipment parameter adjustment in real time, and analyze whether the temperature change in the core area meets the temperature control target and whether the multi-field characteristic changes meet the prediction of multi-field influencing factors in combination with the update results of the target virtual temperature field.

[0120] The generated motor speed adjustment commands and feed rate adjustment commands are sent to the control module of the medical waste treatment equipment via the communication module. The equipment control module adjusts the motor speed and feed rate according to the commands. Simultaneously, the data acquisition system collects real-time operational feedback data after the equipment parameters are adjusted, such as the adjusted motor speed, feed rate, new temperature data, vibration data, and pressure data. This feedback data is input into the dynamically corrected multiphysics coupling model to update the target virtual temperature field. Based on the updated target virtual temperature field, the system analyzes whether the temperature change in the core area of ​​the cavity meets the temperature control target, i.e., whether the temperature gradually returns to and stabilizes within the preset operating range. Simultaneously, the system analyzes whether the changes in multi-field characteristics conform to the predictions of multi-field influencing factors. For example, after the motor speed decreases, does the vibration data decrease? Do the stress field characteristics change as expected? After the feed rate is adjusted, do the pressure data changes conform to the predictions?

[0121] Step S158: If the temperature change is as expected and the multi-field characteristics change normally, then maintain the currently adjusted parameters; if the temperature change is as expected but the multi-field characteristics are abnormal, or the temperature change does not meet expectations, then based on the latest target virtual temperature field data and multi-field coupling logic, combined with the temperature control target, re-correct the parameter association logic, re-select the target parameters based on the multi-field influencing factors, and adjust the motor speed adjustment command and the feed speed adjustment command until the temperature of the core area of ​​the cavity stabilizes within the preset operating range and the multi-field characteristics remain normal.

[0122] The judgment is made based on the analysis results of step S157. If the temperature change in the core area meets expectations, i.e., the temperature successfully returns to and stabilizes within the preset operating range, and the changes in multiple field characteristics are normal (such as stress, pressure, etc., all within the safe range and in line with the prediction), then the currently adjusted motor speed and feeding speed are maintained. If the temperature change meets expectations, but multiple field characteristics are abnormal (such as stress exceeding the equipment's safe allowable value), or the temperature change does not meet expectations (such as the temperature not returning to the preset range), then readjustment is required. At this time, based on the latest target virtual temperature field data and multi-field coupling logic, combined with the temperature control target, the parameter correlation logic between the core area temperature change and the motor speed and feeding speed is revised, which may require consideration of multiple field influencing factors that were not fully considered before. Based on the revised parameter correlation logic and multiple field influencing factors, the target parameters are re-selected (which may still be the motor speed and feeding speed, or other parameters may need to be considered), and the motor speed adjustment command and feeding speed adjustment command are adjusted (such as changing the adjustment direction, adjustment range, or stabilization time). The process of repeatedly sending instructions, receiving feedback, and analyzing results continues until the temperature in the core area of ​​the cavity stabilizes within the preset operating range, and all multi-field characteristics remain normal.

[0123] Step S159: Combining the process requirements of the medical waste treatment equipment with the multi-field coupling rules, based on the temperature control target, the temperature control interval is divided into multiple temperature sub-intervals, each temperature sub-interval corresponding to different operating states of the equipment and corresponding multi-field characteristic ranges.

[0124] Based on the technological requirements of medical waste treatment equipment, such as the processing temperature and time requirements for different types of medical waste, and the multi-field coupling rules (the interaction laws of thermal field, stress field, pressure field, etc.), the temperature control range is divided into multiple temperature sub-ranges based on the temperature control target (T_target±ΔT). For example, it can be divided into a low-temperature sub-range (T_min to T_target-ΔT / 2), a medium-temperature sub-range (T_target-ΔT / 2 to T_target+ΔT / 2), and a high-temperature sub-range (T_target+ΔT / 2 to T_max). Each temperature sub-range corresponds to different operating states of the equipment. For example, the low-temperature sub-range may correspond to the initial feeding stage or low-load operation, the medium-temperature sub-range corresponds to the normal processing state, and the high-temperature sub-range corresponds to the high-load or special waste processing state. At the same time, a corresponding multi-field characteristic range is determined for each temperature sub-interval, such as the thermal field characteristic range (smaller temperature gradient, lower heat flux density), stress field characteristic range (smaller stress value), and pressure field characteristic range (lower pressure) corresponding to the low temperature sub-interval; while the high temperature sub-interval corresponds to a larger temperature gradient, higher heat flux density, larger stress value, and higher pressure.

[0125] Step S1510: For each temperature sub-interval, combined with the multi-field influencing factors, simulate the core area temperature change results and multi-field characteristic changes under different motor speeds by dynamically correcting the multi-physics coupling model, and record the correspondence between motor speed, temperature change, and multi-field characteristic changes.

[0126] For each temperature sub-interval divided in step S159, a dynamic modified multiphysics coupling model is used for simulation, taking into account the associated multi-field influencing factors (such as motor output power, feed rate, ambient temperature, etc.). During the simulation, other influencing factors are kept constant, and only the motor speed is changed to simulate the temperature change results (such as the amount of temperature increase or decrease ΔT) and multi-field characteristic changes (such as stress change ΔS, pressure change ΔP, etc.) of the core region under different motor speeds. For example, in the medium-temperature sub-interval, the motor speed is simulated to gradually increase from N1 to N2, and the core region temperature change ΔT and multi-field characteristic changes ΔS, ΔP, etc. corresponding to each speed value N are recorded, thereby establishing the correspondence between motor speed, temperature change, and multi-field characteristic changes.

[0127] Step S1511: Analyze the correspondence obtained from the simulation, and extract the influence coefficient of motor speed on temperature and the influence coefficient of motor speed on multi-field characteristics. The influence coefficient reflects the difference in influence within different temperature sub-intervals.

[0128] A thorough analysis is conducted on the correspondence between motor speed, temperature change, and multi-field characteristic changes recorded in step S1510. Through data fitting and statistical analysis, the influence coefficient of motor speed on temperature is extracted. This influence coefficient represents the temperature change in the core region caused by a unit change in motor speed, for example, α = ΔT / ΔN, where α is the influence coefficient of motor speed on temperature. Similarly, the influence coefficients of motor speed on multi-field characteristics are extracted, such as the influence coefficient of motor speed on stress β = ΔS / ΔN, and the influence coefficient on pressure γ = ΔP / ΔN. These influence coefficients may have different values ​​in different temperature sub-intervals, reflecting the differences in the influence of motor speed on temperature and multi-field characteristics within different temperature sub-intervals. For example, in the high-temperature sub-interval, due to changes in the physicochemical properties of the material, the influence coefficient of motor speed on temperature may differ from that in the medium-temperature sub-interval.

[0129] Step S1512: Based on the influence coefficient of motor speed on temperature and the influence coefficient of motor speed on multi-field characteristics, and combined with the allowable fluctuation range in the temperature control target, determine the motor speed adjustment range corresponding to each temperature sub-interval, so that adjusting the speed within the motor speed adjustment range can stabilize the temperature within the corresponding temperature sub-interval and keep the multi-field characteristics within the allowable range.

[0130] Based on the extracted influence coefficients α of motor speed on temperature and β and γ of motor speed on multi-field characteristics, combined with the allowable fluctuation range ΔT in the temperature control target (i.e., the core area temperature is allowed to fluctuate between T_min and T_max), and the safe allowable range of multi-field characteristics (such as maximum allowable stress S_max, maximum allowable pressure P_max, etc.), the motor speed adjustment range corresponding to each temperature sub-interval is determined. For example, in the mid-temperature sub-interval, the current temperature is T_mid. To maintain the temperature within this sub-interval, the allowable temperature change is ΔT_sub. Based on the influence coefficient α, the allowable motor speed adjustment ΔN_sub = ΔT_sub / α can be calculated. Thus, the motor speed adjustment range is determined to be from N_current - ΔN_sub to N_current + ΔN_sub. Simultaneously, based on the influence coefficients β and γ, the changes in multi-field characteristics within this speed adjustment range are calculated as ΔS = β × ΔN_sub and ΔP = γ × ΔN_sub, ensuring that ΔS ≤ S_max - S_current and ΔP ≤ P_max - P_current, i.e., the multi-field characteristics remain within the allowable range. Based on the above calculations, a suitable motor speed adjustment range is determined for each temperature sub-range.

[0131] Step S1513: Simulate the temperature change results and multi-field characteristic changes of the core region under different feed rates by dynamically correcting the multi-physics coupling model, extract the influence coefficient of feed rate on temperature and the influence coefficient of multi-field characteristics, and determine the feed rate adjustment range corresponding to each temperature sub-interval.

[0132] Similar to steps S1510 and S1511, for each temperature sub-interval, while keeping other influencing factors fixed, the core region temperature change and multi-field characteristic changes under different feed rates are simulated using a dynamically modified multiphysics coupling model. Data such as feed rate F, temperature change ΔT, and multi-field characteristic changes ΔS and ΔP are recorded. These data are analyzed to extract the influence coefficients of feed rate on temperature (δ=ΔT / ΔF) and on multi-field characteristics (ε=ΔS / ΔF, ζ=ΔP / ΔF). Then, combining the allowable fluctuation range in the temperature control target and the safe allowable range of multi-field characteristics, the feed rate adjustment range corresponding to each temperature sub-interval is determined to ensure that adjusting the feed rate can stabilize the temperature within the corresponding sub-interval and maintain normal multi-field characteristics.

[0133] Step S1514: Organize the temperature sub-intervals, the corresponding motor speed adjustment range, the influence coefficient of motor speed on temperature, and the influence coefficient of motor speed on multi-field characteristics in tabular form, so that each table entry includes the temperature range, speed adjustment direction, upper and lower limits of adjustment range, the influence coefficient of motor speed on temperature, and the influence coefficient of motor speed on multi-field characteristics, and generate a preliminary temperature-speed correlation table.

[0134] The information for each temperature sub-interval, such as the temperature range (T_low to T_high), the corresponding motor speed adjustment range (ΔN_low to ΔN_high), the speed adjustment direction (determined based on whether the temperature needs to be increased or decreased; for example, when the temperature is below the lower limit of the sub-interval and needs to be increased, the speed adjustment direction is to increase), the upper and lower limits of the adjustment range (N_max_adjust and N_min_adjust), the influence coefficient α of motor speed on temperature, and the influence coefficients β and γ on multi-field characteristics, are organized in tabular form. Each table entry corresponds to a temperature sub-interval, clearly listing the above information, thus generating a preliminary temperature-speed correlation table.

[0135] Step S1515: Organize the temperature sub-intervals, the corresponding feed rate adjustment range, the influence coefficient of feed rate on temperature, and the influence coefficient of feed rate on multi-field characteristics to generate a preliminary temperature-feed rate correlation table;

[0136] Similarly, the temperature sub-intervals, the corresponding feed rate adjustment range (ΔF_low to ΔF_high), the feed rate adjustment direction (e.g., when the temperature is too high, the feed rate needs to be increased to reduce the temperature), the upper and lower limits of the adjustment range (F_max_adjust and F_min_adjust), the influence coefficient δ of the feed rate on the temperature, and the influence coefficients ε and ζ of the multi-field characteristics are compiled into a table to generate a preliminary temperature-feed rate correlation table.

[0137] Step S1516: Based on the historical data of actual equipment operation, optimize the adjustment range and influence coefficient of the preliminary temperature-speed correlation table and the preliminary temperature-feed speed correlation table. Based on the actual effect of parameter adjustment in different temperature sub-intervals during historical operation and the changes in multiple field characteristics, correct the values ​​in the preliminary correlation table that do not meet the set requirements due to deviations from the actual operation.

[0138] Historical data on actual equipment operation is collected. This data includes actual records of motor speed and feed rate adjustments within different temperature sub-ranges, as well as corresponding temperature changes and multi-field characteristic changes. The historical data is compared with preliminary temperature-speed and temperature-feed rate correlation tables to analyze whether there are deviations between the adjustment ranges and influence coefficients in the preliminary correlation tables and the actual operating conditions. For example, historical data may show that in a certain temperature sub-range, after adjusting the motor speed according to the range in the preliminary correlation table, the temperature change deviates significantly from the result calculated based on the influence coefficients, or multi-field characteristics exceed the allowable range. In this case, it is necessary to optimize and correct the adjustment ranges and influence coefficients in the preliminary correlation tables based on the actual effects of historical operation and the changes in multi-field characteristics, so that the values ​​in the correlation tables better reflect the actual operating characteristics of the equipment.

[0139] Step S1517: Through multiple rounds of actual operation verification and correction, the parameter adjustment range corresponding to each temperature sub-interval in the preliminary temperature-speed correlation table and temperature-feed speed correlation table can achieve effective temperature control and stable multi-field characteristics, and finally obtain the preset temperature-speed correlation table and the preset temperature-feed speed correlation table.

[0140] The optimized preliminary correlation table is applied to the actual operation of the equipment for multiple rounds of verification. In each round of verification, the motor speed and feed rate are adjusted according to the correlation table, and the temperature control effect and the stability of multi-field characteristics are observed. If it is found that the parameter adjustment range corresponding to certain temperature sub-intervals still cannot effectively control the temperature, or the multi-field characteristics become unstable, the correlation table is corrected again. After multiple rounds of actual operation verification and correction, the parameter adjustment range corresponding to each temperature sub-interval in the temperature-speed correlation table and the temperature-feed rate correlation table can accurately and effectively control the core area temperature, and ensure that the multi-field characteristics are stable within the safe allowable range. The final correlation table is the preset temperature-speed correlation table and the preset temperature-feed rate correlation table, which is used for querying in step S154.

[0141] For example, the method further includes: step S210: extracting the temperature distribution characteristics, temperature change trends and corresponding multi-field feature annotations of each region in the target virtual temperature field, and combining the multi-field coupling rules to determine the temperature feature range, temperature change trend range and corresponding multi-field feature range of each region under normal operating conditions.

[0142] The temperature distribution characteristics of each region of the cavity are extracted from the target virtual temperature field, such as the average temperature, temperature gradient, and temperature extreme values ​​of each region; the temperature change trend, i.e., the rate of temperature change and direction of change of each region over time; and the corresponding multi-field feature labels, such as thermal field feature labels, stress field feature labels, and pressure field feature labels. Combining multi-field coupling rules, the range of temperature characteristics that each region should fall within under normal operating conditions is analyzed (e.g., the normal average temperature range of region A is T_A1 to T_A2), the range of temperature change trends (e.g., the normal range of temperature change rate of region B is -ΔT_rate to ΔT_rate), and the corresponding multi-field feature ranges (e.g., the normal stress range of region C is S_C1 to S_C2, and the normal pressure range is P_C1 to P_C2, etc.). These ranges are determined based on the equipment's design parameters, material properties, process requirements, and historical normal operating data.

[0143] Step S220: Extract vibration features, pressure features and corresponding multi-field feature labels from the multi-dimensional processed data. Based on the physical logic of equipment operation and multi-field coupling rules, establish a correlation model between vibration features, pressure features and temperature features. The vibration features include vibration frequency and vibration amplitude, and the pressure features include pressure fluctuation period and pressure peak value.

[0144] Vibration features are extracted from multi-dimensional data processing, including vibration frequency (the number of vibrations per unit time) and vibration amplitude (the maximum displacement or acceleration of the vibration); pressure features, including pressure fluctuation period (the time interval between pressure peaks) and pressure peak (the maximum value in pressure fluctuations); and corresponding multi-field feature labels. Based on the physical logic of equipment operation, such as vibration generated by the movement of components like motors and shafts, and vibration changes potentially related to component temperature and stress changes; pressure changes are related to factors such as feed rate, internal chemical reactions, and temperature. Combining multi-field coupling rules, a correlation model between vibration features, pressure features, and temperature features is established through analysis and modeling of historical data. For example, a functional relationship between vibration frequency f and temperature T can be established: T = f(f, A) (where A is the vibration amplitude), or a relationship between pressure peak P_peak and temperature change rate dT / dt.

[0145] Step S230: Integrate the temperature characteristics, vibration characteristics, pressure characteristics and corresponding multi-field characteristic information of each region according to the correlation model to generate a multi-field feature vector of the equipment operating status. The multi-field feature vector contains the coupling relationship information between each feature and the degree to which the feature deviates from the normal range.

[0146] Following the correlation model established in step S220, the temperature characteristics (such as temperature value and temperature change trend), vibration characteristics (vibration frequency and vibration amplitude), pressure characteristics (pressure fluctuation period and pressure peak value) of each region are integrated with the corresponding multi-field feature information (multi-field feature labels for each feature). During the integration process, not only are the specific values ​​of each feature included, but also the coupling relationship information between each feature (such as the correlation degree between temperature and vibration, the correlation coefficient between pressure and temperature, etc.) and the degree to which each feature deviates from the normal range (such as the percentage of temperature deviating from the normal range, the amount of vibration amplitude exceeding the normal range, etc.). The integrated information is represented as a high-dimensional vector, namely the multi-field feature vector of the equipment operating status, which comprehensively reflects the current operating status of the equipment.

[0147] Step S240: Compare the multi-field feature vector with the preset normal operation multi-field feature template, identify abnormal feature items in the multi-field feature vector that do not match the preset template, determine the type of abnormal feature, the type of abnormal feature is temperature abnormality, vibration abnormality or pressure abnormality, and analyze the coupling correlation between abnormal features based on the correlation model.

[0148] The preset multi-field feature template for normal operation is a standard template established based on the multi-field feature vectors of the equipment under normal operating conditions. It includes the numerical range, coupling relationship information, and deviation degree of each feature during normal operation. The multi-field feature vectors of the current equipment operating state are compared with this template. By calculating the similarity between vectors and the deviation of each feature item from the corresponding feature item in the template, abnormal feature items in the multi-field feature vectors that do not conform to the preset template are identified. Based on the type of the abnormal feature item, the type of abnormal feature is determined. For example, if a feature item is a temperature feature and exceeds the normal range, it is a temperature abnormality; if it is a vibration feature abnormality, it is a vibration abnormality; if it is a pressure feature abnormality, it is a pressure abnormality. Simultaneously, the coupling relationship between these abnormal features is analyzed based on a correlation model. For example, does a temperature abnormality cause a vibration abnormality, or do pressure abnormalities share a common cause with temperature abnormalities?

[0149] Step S250: Based on the type and coupling correlation of the abnormal features, combined with the dynamic correction multiphysics coupling model and multi-field coupling rules, deduce the possible equipment component failures corresponding to the abnormal features, and classify the causal relationship between the failures and the abnormal features.

[0150] Based on the types of anomalous features (temperature, vibration, pressure anomalies) and their coupling relationships, analysis is performed using a dynamically modified multiphysics coupling model and multiphysics coupling rules. For example, if a coupling relationship exists between abnormally high temperature and abnormally high vibration, combining the model and rules, it might be deduced that excessive wear of friction components leads to increased friction, which in turn causes increased temperature and vibration. The corresponding equipment component failure might be wear of the friction plates. Through the above analysis, the possible equipment component failures corresponding to the anomalous features are deduced, and the causal relationship between the failure and the anomalous feature is classified, such as direct causal relationship (failure directly causes anomalous features) and indirect causal relationship (failure indirectly causes anomalous features by affecting other physical fields).

[0151] Step S260: Based on the causal relationship between abnormal features and faults, query the equipment fault association database, which contains corresponding records of multiple characteristic abnormal patterns, fault causes, and faulty components in historical fault cases, and determine the possible fault causes corresponding to the current abnormal features.

[0152] The equipment fault association database stores a large number of historical fault cases. Each case records the corresponding information for multiple abnormal patterns (such as abnormal temperature, abnormal vibration, abnormal pressure, and combinations thereof) at the time of the fault, the cause of the fault (such as component wear, insufficient lubrication, motor failure, etc.), and the faulty components (such as friction plates, motor bearings, seals, etc.). Based on the causal relationship between the abnormal features and the fault obtained in step S250, a query is performed in the equipment fault association database to find historical cases similar to the current abnormal feature pattern. Through comparison and matching, the possible fault causes corresponding to the current abnormal feature are determined. For example, if the current abnormal feature pattern is similar to the abnormal pattern of "friction plate wear" in historical cases, then the possible fault cause is friction plate wear.

[0153] Step S270: Combining the spatial distribution of the target virtual temperature field and the multi-field feature annotation, locate the spatial position of the equipment corresponding to the abnormal features, identify the specific component or area where the fault occurred, and divide the relationship between the faulty component and the surrounding components.

[0154] The spatial distribution of the target virtual temperature field can display the temperature conditions of each region. Combined with multi-field feature annotations (such as the spatial distribution of stress and pressure fields), the specific spatial location of abnormal features (such as temperature anomalies and stress anomalies) can be pinpointed. For example, if the target virtual temperature field shows an abnormally high temperature in a local area of ​​the cavity, and the stress field annotations reveal an abnormally high stress in that area, then that spatial location can be located. Based on the equipment's structural drawings and component layout, the specific component or region corresponding to that spatial location can be identified, thereby determining the specific component or region where the fault occurred. Simultaneously, analyzing the connection and interaction relationships between the faulty component and surrounding components, and classifying the correlations between the faulty component and surrounding components (such as mechanical connections, heat conduction, and mechanical transmission), helps assess the potential impact of the fault on surrounding components.

[0155] Step S280: Integrate abnormal feature types, abnormal feature coupling correlations, possible causes of faults, and fault location information to generate fault diagnosis results. The fault diagnosis results include the specific manifestations of multiple field feature anomalies, the physical location of the faulty component, the possible mechanism leading to the fault, and the potential impact of the fault on equipment operation.

[0156] The abnormal feature types (temperature anomaly, vibration anomaly, pressure anomaly) determined in step S240, the abnormal feature coupling correlations analyzed in step S240 (the relationships between various abnormal features), the possible causes of the fault determined in step S260, and the fault location information located in step S270 are integrated to generate a complete fault diagnosis result. The fault diagnosis result describes in detail the specific manifestations of multiple abnormal features, such as temperature anomalies manifested as the core area temperature exceeding the preset range by 5°C, and vibration anomalies manifested as the motor vibration frequency increasing by 10Hz and the amplitude increasing by 20%, etc.; the physical location of the faulty component, such as "the friction plate of the friction mechanism is located below the left wall inside the cavity"; the possible mechanisms leading to the fault, such as "friction plate wear leads to an increase in the friction coefficient, an increase in frictional heat generation, causing a temperature rise, and at the same time, intensified vibration"; and the potential impact of the fault on equipment operation, such as "if not handled in time, it may lead to further wear of the friction plate, motor overload, or even overheating and damage to the cavity wall, affecting the medical waste treatment effect and equipment safety."

[0157] For example, the method further includes: step S310: collecting historical operating data of the medical waste treatment equipment under normal operating conditions and various fault conditions, the historical operating data including historical target virtual temperature field data, historical multi-dimensional processing data, corresponding fault types, fault location information and fault processing results;

[0158] To build a fault diagnosis model, it is necessary to collect historical operating data of the equipment under different operating conditions. This data includes data under normal operating conditions and data under various fault conditions (such as motor faults, friction component faults, sensor faults, etc.). Historical target virtual temperature field data is a virtual temperature field generated during historical operation; historical multi-dimensional processed data includes historical temperature data, historical vibration data, historical pressure data, and corresponding historical multi-field feature labels; the corresponding fault type is a classification of historical faults (such as mechanical faults, electrical faults, thermal faults, etc.); fault location information is the specific component or area where the historical fault occurred; and fault handling results record the handling measures taken for the historical faults and their effects.

[0159] Step S320: Extract features from the historical target virtual temperature field data to obtain the historical temperature distribution features, historical temperature change trend features, and corresponding historical multi-field feature annotations for each region; extract features from the historical multi-dimensional processed data to obtain historical vibration features, historical pressure features, and corresponding historical multi-field feature labels.

[0160] Feature extraction is performed on historical virtual temperature field data to obtain historical temperature distribution characteristics for each region, such as historical average temperature, historical temperature gradient, and historical temperature extremes; historical temperature change trend characteristics, such as historical temperature change rate and historical temperature fluctuation period; and corresponding historical multi-field feature annotations, such as historical thermal field feature labels and historical stress field feature labels. Feature extraction is also performed on historical multi-dimensional processed data to obtain historical vibration characteristics (historical vibration frequency and historical vibration amplitude), historical pressure characteristics (historical pressure fluctuation period and historical pressure peak value), and corresponding historical multi-field feature labels. The extracted features will be used as input features for the fault diagnosis model.

[0161] Step S330: Based on the multi-field coupling rule, the historical temperature features, historical vibration features, historical pressure features and corresponding multi-field feature information are associated and integrated to generate a multi-field feature vector corresponding to each historical operating state, and the corresponding fault type label and fault location label are labeled at the same time.

[0162] Based on multi-field coupling rules, the extracted historical temperature features, historical vibration features, historical pressure features, and their corresponding multi-field feature information are correlated and integrated. For example, the temperature distribution features, vibration frequency, pressure peak value, and corresponding thermal field and stress field labels of the same historical operating state are integrated into a single vector to generate a multi-field feature vector corresponding to that historical operating state. Simultaneously, based on the fault type and fault location information in the historical operating data, each multi-field feature vector is labeled with a corresponding fault type label (such as "motor bearing wear," "sensor fault," etc.) and a fault location label (such as "motor housing," "right side wall of the cavity," etc.).

[0163] Step S340: Classify the training samples composed of multi-field feature vectors and labels according to the fault type;

[0164] Training samples, consisting of multi-field feature vectors and corresponding fault type and fault location labels, are classified according to fault type. For example, all samples belonging to "mechanical faults" are grouped into one category, "electrical faults" into another, and "thermal faults" into a third, and so on. This classification helps the fault diagnosis model to learn and identify different types of faults specifically, improving the model's diagnostic accuracy for various fault types.

[0165] Step S350: Construct the network structure of the fault diagnosis model, wherein the network structure of the fault diagnosis model includes a multi-field feature input module, a feature association module, an anomaly identification module, a fault location module, and a causal analysis module;

[0166] The network structure of the fault diagnosis model is designed, which includes multiple functional modules. The multi-field feature input module receives multi-field feature vectors as input to the model; the feature association module analyzes the correlation and coupling strength between features in the input feature vectors based on multi-physics coupling rules, such as analyzing the correlation between temperature and vibration features; the anomaly identification module identifies anomalous items and their degrees of anomalousness in the correlation, i.e., finding feature combinations that do not conform to normal correlations and their deviations; the fault location module locates the specific location of the fault based on the spatial distribution of anomalous features in the target virtual temperature field and equipment structure; and the causal analysis module derives the causal relationship between the anomaly and the fault based on the anomalous features, fault location, and multi-field coupling rules, thus determining the cause of the fault.

[0167] Step S360: Input the classified training samples into the fault diagnosis model. The multi-field feature input module receives multi-field feature vectors. The feature association module analyzes the association relationship and coupling strength between features based on the multi-physics field coupling rules. The anomaly identification module identifies the abnormal items and the degree of abnormality in the association relationship.

[0168] The classified training samples are input one by one into the fault diagnosis model. The multi-field feature input module receives the multi-field feature vectors from the training samples and passes them to the feature association module. The feature association module loads the multi-physics coupling rule base and uses the coupling rules in the rule base to analyze each feature in the multi-field feature vectors, calculating the correlation between features (such as positive correlation, negative correlation) and the coupling strength (the degree of correlation). The anomaly identification module compares the feature correlation and coupling strength obtained by the feature association module with the preset normal correlation template, identifying anomalies in the correlation (such as two features that should be positively correlated showing a negative correlation) and the degree of anomaly (the degree of deviation from the normal correlation strength).

[0169] Step S361: The fault location module locates the fault location based on the spatial distribution of abnormal features, and the causal analysis module deduces the causal relationship between the abnormality and the fault.

[0170] The fault location module receives abnormal feature items identified by the anomaly recognition module. Combining this with the spatial distribution information of the abnormal features in the target virtual temperature field (obtained from multi-field feature vectors or historical target virtual temperature field data), as well as the equipment's structural information and component layout, it locates the spatial position corresponding to the abnormal features, thereby determining the specific location of the fault. The causal analysis module, based on the abnormal feature items, fault location information, the correlation and coupling strength analyzed by the feature association module, and combining multi-field coupling rules and the physical logic of equipment operation, derives the causal relationship between the abnormal features and the fault. That is, it analyzes how the anomaly is caused by the fault, or how the fault leads to the appearance of the abnormal features, determining the possible mechanisms leading to the fault.

[0171] Step S370: During the training process, based on the differences between the fault type, fault location, causal relationship and label output by the fault diagnosis model, adjust the parameters of each module of the fault diagnosis model, focusing on optimizing the feature association rules, anomaly identification threshold and causal relationship derivation logic, so that the fault diagnosis model can learn to identify different types and severity of faults.

[0172] During the training of the fault diagnosis model, the fault type, fault location, and causal relationship output by the model are compared with the labels (true fault type, fault location) in the training samples, and the differences between them (such as classification error rate, location bias, etc.) are calculated. Based on these differences, optimization algorithms such as backpropagation are used to adjust the parameters of each module of the model, such as the association weights in the feature association module, the anomaly identification threshold (the critical value used to determine whether a feature is abnormal) in the anomaly identification module, and the derivation logic parameters in the causal analysis module. The focus is on optimizing the feature association rules to more accurately reflect the true association between multiple features; optimizing the anomaly identification threshold to improve the ability to identify anomalies of different severity; and optimizing the causal relationship derivation logic to enable the model to more accurately deduce the causal relationship between anomalies and faults. Through this training, the fault diagnosis model can learn to identify different types (mechanical, electrical, thermal, etc.) and different severity levels (minor, moderate, severe) of faults.

[0173] Step S380: Through multiple rounds of training, ensure that the fault diagnosis model achieves the preset requirements in terms of fault type identification accuracy, fault location accuracy, and causal relationship deduction accuracy on the validation sample set.

[0174] The training sample set is divided into a training set and a validation set. The training set is used for model parameter learning, and the validation set is used to evaluate the model's performance on untrained data. The model parameters are continuously adjusted through multiple training iterations. After each training iteration, the model is tested using the validation set, and the following metrics are calculated: fault type identification accuracy (the proportion of correctly identified fault types out of the total number of validation samples), fault location accuracy (the proportion of correctly located fault locations out of the total number of validation samples), and causal relationship derivation accuracy (the proportion of correctly derived causal relationships out of the total number of validation samples). Model training stops when all these accuracy metrics meet preset requirements (e.g., all greater than 90%).

[0175] Step S390: Collect operational data of new fault types of medical waste treatment equipment, generate a new validation sample set, and use the new validation sample set to test the fault diagnosis model's ability to identify new fault types. If the identification ability reaches the preset standard, a pre-trained fault diagnosis model is obtained; if it does not reach the preset standard, new fault type training samples are added to continue training.

[0176] To test the fault diagnosis model's ability to identify new fault types not present in the training samples, operational data for these new fault types was collected. These new fault types could be faults that had never occurred before or those not included in the training samples. A new validation sample set was generated using this data, containing multi-field feature vectors and corresponding fault labels for each new fault type. This new validation sample set was then input into the trained fault diagnosis model to test its ability to identify new fault types, such as fault type identification accuracy and fault location accuracy. If the identification ability met a preset standard (e.g., an accuracy rate of at least 85% for identifying new fault types), the pre-trained fault diagnosis model was obtained. If not, training samples for the new fault types were added to the training set, and the model was further trained until its ability to identify new fault types met the requirements.

[0177] In one exemplary embodiment, a temperature intelligent control system for medical waste treatment equipment based on multi-physics field coupling is provided. This system can be a terminal, server, etc., and its internal structure diagram can be as follows: Figure 2As shown, the intelligent temperature control system for medical waste treatment equipment based on multi-physics coupling includes a processor, memory, input / output interface, communication interface, display unit, and input device. The processor, memory, and input / output interface are connected via a system bus, and the communication interface, display unit, and input device are also connected to the system bus via the input / output interface. The processor provides computational and control capabilities. The memory includes a non-volatile storage medium and internal memory. The non-volatile storage medium stores the operating system and computer programs. The internal memory provides the environment for the operation of the operating system and computer programs in the non-volatile storage medium. The input / output interface is used for information exchange between the processor and external devices. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, mobile cellular networks, Near Field Communication (NFC), or other technologies. When executed by a processor, this computer program implements a method for intelligent temperature control of medical waste treatment equipment based on multi-physics coupling. The display unit of this intelligent temperature control system for medical waste treatment equipment, used to form a visually visible image, can be a display screen, a projection device, or a virtual reality imaging device. The display screen can be an LCD screen or an e-ink screen. The input device of this intelligent temperature control system for medical waste treatment equipment can be a touch layer covering the display screen, or buttons, a trackball, or a touchpad set on the casing of the intelligent temperature control system for medical waste treatment equipment, or an external keyboard, touchpad, or mouse, etc.

[0178] It should be noted that, in order to simplify the description of the present invention and thus help to understand one or more embodiments of the invention, multiple features may sometimes be grouped into one embodiment, drawing or description thereof in the foregoing description of the embodiments of the present invention.

Claims

1. A method for intelligent temperature control of a medical waste treatment device based on multi-physical field coupling, characterized in that, The method includes: Pre-construct a multi-physics coupling basic model for medical waste treatment equipment; Collect multi-dimensional operational data during the operation of medical waste treatment equipment, and perform physical field feature mapping processing on the multi-dimensional operational data to obtain multi-dimensional processed data with associated multi-field features; The multi-dimensional processed data is input into the multi-physics coupling basic model. Through the deviation analysis between the multi-field data and the model output, the heat conduction parameters and mechanical correlation coefficients in the multi-physics coupling basic model are adjusted to obtain a dynamically corrected multi-physics coupling model. The pre-trained temperature field reconstruction model is invoked to perform cross-field feature analysis on the dynamically modified multiphysics coupling model, generating a target virtual temperature field for the entire cavity of the medical waste treatment equipment. Based on the temperature distribution and temperature change trend of the core region of the cavity in the target virtual temperature field, the operating parameters of the medical waste treatment equipment are adjusted by combining multi-field coupling logic to maintain the temperature of the core region of the cavity within the preset operating range.

2. The intelligent temperature control method for medical waste treatment equipment based on multi-physics coupling according to claim 1, characterized in that, The pre-built multiphysics coupling basic model of the medical waste treatment equipment includes: Based on the model features of medical waste treatment equipment, the geometric dimensions of the equipment cavity, the assembly position of the friction mechanism, and the connection relationship between the motor and the shaft are extracted. A physical interaction network between equipment components is constructed, and the overlapping area of ​​heat conduction path and mechanical action transmission path is identified. For three key components—the cavity wall, friction parts, and motor housing—we collect data on the thermal conductivity, specific heat capacity, density, and elastic modulus of materials under different temperature conditions, and construct a dynamic correlation model of material properties changing with temperature to match the material parameters with the temperature fluctuations in actual equipment operation. Based on the process requirements of medical waste treatment, the operating conditions of the equipment corresponding to different motor output power, different feed rates, and different ambient temperatures are determined, operating condition classification rules are established, and the expected correlation direction of thermal field and stress field is marked for each operating condition. Based on the physical interaction network of equipment components, the three-dimensional geometric model of the equipment is divided into regions, and the regions where heat conduction and mechanical action overlap are marked as key coupling regions. The parameters in the dynamic correlation model of material properties are embedded into the modeling unit of the corresponding component. Different working condition parameters are loaded in sequence according to the working condition classification rules to simulate the diffusion process of the thermal field and the distribution process of the stress field inside the equipment under each working condition, and the correspondence between the thermal field data and the stress field data is recorded. The correspondence between thermal field and stress field under different working conditions is analyzed, the coupling law between heat conduction parameters and mechanical correlation coefficient is extracted, and a parameter coupling rule base is established. The parameter coupling rule base contains the logic of the influence of changes in heat conduction parameters on mechanical correlation coefficient under different working conditions. Based on the parameter coupling rule base, the simulated thermal field data and stress field data are correlated and integrated to construct a multi-physics coupling basic database, so that each data entry in the multi-physics coupling basic database contains operating condition parameters, thermal field characteristics, stress field characteristics and parameter coupling relationships. A dynamic correlation model between thermal conductivity and mechanical action is established through a multi-field feature mapping algorithm. Supported by a multi-physics coupling basic database, a multi-physics coupling basic model containing the thermal-mechanical coupling relationship of the entire equipment is generated. The multi-physics coupling basic model is used to output the corresponding thermal field distribution and stress distribution prediction results based on the input operating parameters.

3. The intelligent temperature control method for medical waste treatment equipment based on multi-physics coupling according to claim 1, characterized in that, The process involves collecting multi-dimensional operational data during the operation of the medical waste treatment equipment, and performing physical field feature mapping on this multi-dimensional operational data to obtain multi-dimensional processed data with associated multi-field features, including: The device receives real-time temperature, vibration, and pressure data from various data acquisition components. Based on the physical logic of the device's operation, it establishes a spatiotemporal correlation between temperature and vibration data. This spatiotemporal correlation reflects the time response logic between changes in motor vibration and changes in the temperature around the motor. Based on the spatial structure and multi-physics coupling logic of the equipment cavity, temperature data is mapped to the three-dimensional spatial coordinates of the cavity according to the acquisition position and associated with the thermal field characteristics of the corresponding area; vibration data is mapped to the structural coordinates of the motor and shaft according to the acquisition position and associated with the stress field characteristics of the corresponding components; pressure data is mapped to the spatial grid inside the cavity and associated with the thermal field-stress field interaction characteristics of the corresponding area. Dynamic change features are extracted from the mapped temperature data, vibration data, and pressure data. These dynamic change features include the rate of temperature rise, the frequency change of vibration data, and the fluctuation amplitude of pressure data. Correlation constraints between dynamic change features are established based on multi-field coupling laws. Based on the correlation constraints, cross-field data anomaly correlation verification is performed on dynamic change features to identify data where a single data point is normal but multiple field features are abnormally correlated. Multi-field collaborative optimization is then performed on the dynamic change features that pass the verification, and data deviations are adjusted based on the interaction relationship between thermal field features and stress field features. Based on the dynamic cycle of equipment operation, the optimized temperature data, vibration data, and pressure data are integrated with the corresponding thermal field characteristics and stress field characteristics, and the associated multi-field characteristic information is labeled for each time node; The integrated data undergoes multi-field feature consistency verification to ensure that the thermal field features, stress field features, and dynamic change features corresponding to each data entry conform to the multi-physics field coupling logic, thereby generating multi-dimensional processed data with associated multi-field features. The multi-dimensional processed data includes temperature data, vibration data, pressure data, and corresponding multi-field feature labels.

4. The intelligent temperature control method for medical waste treatment equipment based on multi-physics coupling according to claim 1, characterized in that, The process of inputting the multi-dimensional processed data into the multi-physics coupling basic model, and adjusting the heat conduction parameters and mechanical correlation coefficients in the multi-physics coupling basic model through collaborative analysis of the deviation between the multi-field data and the model output, to obtain a dynamically corrected multi-physics coupling model includes: Temperature data, vibration data, pressure data, and corresponding multi-field feature labels from the multi-dimensional processed data are mapped to the thermal field input port, stress field input port, and multi-field feature matching port of the multi-physics coupling basic model, respectively. Run the multiphysics coupling basic model and output the predicted thermal field distribution, predicted stress distribution and predicted multi-field feature labels under the corresponding input data. Extract the temperature feature value and mechanical feature value from the predicted thermal field distribution and the predicted stress distribution. Extract actual temperature feature values, actual mechanical correlation feature values, and actual multi-field feature labels from multi-dimensional processed data. Compare the deviation direction and deviation magnitude between the actual feature values ​​and the model predicted feature values. At the same time, match the consistency between the actual multi-field feature labels and the predicted multi-field feature labels to obtain the label consistency results. Based on the deviation direction, deviation magnitude, and label consistency results, and combined with the parameter coupling rule base in the multiphysics coupling basic model, the heat conduction parameters and mechanical correlation coefficients that need to be adjusted are determined, and the priority of parameter adjustment is divided. Based on the coupling law between deviation amplitude and parameters, the coordinated adjustment ratio of heat conduction parameters and mechanical correlation coefficient is calculated. The coordinated adjustment ratio reflects the interactive influence of changes in heat conduction parameters and changes in mechanical correlation coefficient. According to the coordinated adjustment ratio, the heat conduction parameters and mechanical correlation coefficients in the multiphysics coupling basic model are adjusted synchronously. The numerical changes before and after the parameter adjustment and the corresponding multi-field feature correlation logic are recorded to obtain the adjusted multiphysics coupling basic model. Run the adjusted multiphysics coupling basic model to output new predicted thermal field distribution, predicted stress distribution and predicted multi-field feature labels. Extract the predicted feature values ​​again and compare them with the actual feature values ​​in the multi-dimensional processed data to analyze the changing trend of multi-field deviation. If the biases of multiple fields all tend to decrease and the consistency of feature labels improves, then continue to optimize the parameters according to the current adjustment logic; if the bias of any physical field increases or the label consistency decreases, then recalculate the collaborative adjustment ratio and adjust the parameter coupling rules. Repeat the steps of parameter adjustment, model running, and deviation analysis until the deviations of multiple fields are within a reasonable range and the consistency of feature labels reaches the preset standard, thus obtaining a dynamically corrected multiphysics coupling model.

5. The intelligent temperature control method for medical waste treatment equipment based on multi-physics coupling according to claim 1, characterized in that, The process involves calling a pre-trained temperature field reconstruction model to perform cross-field feature analysis on the dynamically modified multiphysics coupling model, generating a target virtual temperature field covering the entire cavity of the medical waste treatment equipment, including: Extract the thermal field distribution characteristics, stress field distribution characteristics, material thermal property characteristics, and multi-field coupling rules from the dynamically modified multiphysics coupling model, and establish the correlation mapping relationship between thermal field characteristics and stress field characteristics based on multi-field synergistic logic; The thermal field distribution characteristics, stress field distribution characteristics, and material thermal property characteristics are integrated according to the correlation mapping relationship to generate a multi-field fusion feature set, which contains the coupling relationship information between each feature; The multi-field fusion feature set is input into the feature encoding module of the pre-trained temperature field reconstruction model to generate a temperature field association encoding vector. Based on the spatial structure and multi-physics coupling logic of the equipment cavity, the temperature field correlation encoding vector is mapped to the three-dimensional spatial grid of the cavity, the correspondence between encoding features and spatial positions is established, and the multi-field feature weights are divided for different spatial positions. Based on the correspondence between coding features and spatial location, combined with the thermal characteristics of materials and multi-field coupling rules, the initial temperature value of each spatial grid node is calculated, and a preliminary temperature field framework for the entire cavity is constructed. By utilizing the actual temperature data and corresponding multi-field feature labels in the multi-dimensional processing data, the temperature of the corresponding spatial node in the temperature field framework is corrected. Based on the multi-field feature correlation of the surrounding nodes, the temperature value of the node not covered by the actual data is adjusted. The temperature change trends of each region in the corrected temperature field framework are analyzed. Combined with the stress field distribution and multi-field coupling rules output by the dynamically corrected multi-physics coupling model, the temperature of the region in the temperature field that does not match the stress field characteristics is adjusted, and the temperature field is globally optimized to generate a target virtual temperature field for the entire cavity of the medical waste treatment equipment. The target virtual temperature field includes the temperature distribution state, temperature change trend and corresponding multi-field feature annotations of each region of the cavity.

6. The intelligent temperature control method for medical waste treatment equipment based on multi-physics coupling according to claim 5, characterized in that, The training process of the pre-trained temperature field reconstruction model includes: Collect historical multi-dimensional data and corresponding historical measured temperature field data of medical waste treatment equipment under different operating conditions. The historical multi-dimensional data includes historical temperature data, historical vibration data, historical pressure data and corresponding historical multi-field feature labels. The historical measured temperature field data is a set of measured temperatures at multiple locations in the cavity. Based on the multi-field coupling law, feature extraction is performed on historical multi-dimensional data to obtain historical thermal field correlation features, historical stress field correlation features, and historical material thermal property features. These historical thermal field correlation features, historical stress field correlation features, and historical material thermal property features are then integrated into a set of historical multi-field fusion features. Spatial features are extracted from historical measured temperature field data to obtain the temperature feature value of each spatial node, the temperature correlation between nodes and the corresponding spatial multi-field feature weights, and to construct a historical temperature field feature model. The historical multi-field fusion feature set and the corresponding historical temperature field feature model are combined to form training sample pairs, and the training sample pairs are classified according to the working condition type. A network structure for constructing a temperature field reconstruction model is provided, comprising a feature encoding module, a spatial mapping module, a temperature calculation module, and an optimization and adjustment module. The classified training samples are used to reconstruct the model for the input temperature field. The feature encoding module encodes the historical multi-field fusion feature set. The spatial mapping module maps the encoded features to a spatial grid and assigns corresponding multi-field feature weights. The temperature calculation module calculates the initial temperature field based on material properties and multi-field coupling rules. The optimization and adjustment module corrects the temperature field based on the historical temperature field feature model. During the training process, based on the difference between the temperature field output by the temperature field reconstruction model and the historical temperature field feature model, the parameters of each module of the temperature field reconstruction model are adjusted, and the multi-field feature interaction algorithm and spatial multi-field feature weight allocation rules are optimized, so that the temperature field reconstruction model can learn to capture the influence of multi-field coupling on the temperature field. Through multiple rounds of training, the temperature field output by the temperature field reconstruction model is made to match the historical measured temperature field data to the preset requirements. At the same time, the temperature field reconstruction model can stably output a temperature field with multi-field feature annotations under different working conditions. Collect new operating data of medical waste treatment equipment to generate a new validation sample set. Use the new validation sample set to test the generalization ability of the temperature field reconstruction model. If the generalization ability reaches the preset standard, a pre-trained temperature field reconstruction model is obtained; otherwise, new operating condition training samples are added to continue training.

7. The intelligent temperature control method for medical waste treatment equipment based on multi-physics coupling according to claim 1, characterized in that, The adjustment of the medical waste treatment equipment's operating parameters based on the temperature distribution and temperature change trend of the core region of the cavity in the target virtual temperature field, combined with multi-field coupling logic, includes: The target virtual temperature field is analyzed to determine the spatial range of the cavity core region. The temperature distribution characteristics, temperature change trends and corresponding multi-field feature annotations within the cavity core region are extracted to identify the temperature control target of the cavity core and the associated multi-field influencing factors. Based on the dynamic correction multiphysics coupling model and multi-field coupling rules, the allowable fluctuation range of core area temperature is determined by combining the temperature control target of the cavity core. Based on the associated multi-field influencing factors, the stress field and pressure field related parameters that may be affected by the adjustment of motor speed and feeding speed are identified. The parameter correlation logic between core area temperature change and motor speed and feeding speed is derived. Based on the temperature change trend and parameter correlation logic, combined with the temperature control target, the current temperature deviation direction is determined, and the adjustable target parameters are screened based on multiple influencing factors to determine the target parameters that need to be adjusted. Query the preset temperature-speed correlation table and the preset temperature-feed speed correlation table to obtain the initial adjustment range corresponding to the current temperature sub-interval. The initial adjustment range includes the initial speed adjustment range and the initial feed speed adjustment range. Based on the dynamic correction multiphysics coupling model, the simulation termination condition is set with the temperature control target as the benchmark. The simulation scenario is selected by combining multiple influencing factors. The temperature change results and multi-field characteristic changes in the core area are simulated under the adjustment range of different target parameters. A correlation model of parameter adjustment range, temperature change amount, and multi-field characteristic change amount is established. Based on the aforementioned correlation model and initial adjustment range, the adjustment range required to bring the core area temperature back to the preset range is calculated. The adjustment range includes the motor speed adjustment range and the feed rate adjustment range. Simultaneously, based on multiple influencing factors, it is verified whether the multi-field characteristic changes corresponding to the adjustment range are within the range allowed for safe operation of the equipment, and motor speed adjustment commands and feed rate adjustment commands are generated. The motor speed adjustment commands and feed rate adjustment commands include the adjustment direction, adjustment range, and the stable maintenance time after adjustment. The motor speed adjustment command and the feed speed adjustment command are sent to the equipment control module. The operation feedback data after the equipment parameter adjustment is received in real time. Combined with the update results of the target virtual temperature field, the temperature change in the core area is analyzed to see if it meets the temperature control target and whether the multi-field characteristic changes meet the prediction of multi-field influencing factors. If the temperature change is as expected and the multi-field characteristics change normally, then maintain the current adjusted parameters. If the temperature change is as expected but the multi-field characteristics are abnormal, or the temperature change does not meet expectations, then based on the latest target virtual temperature field data and multi-field coupling logic, the parameter association logic is revised in conjunction with the temperature control target. The target parameters are re-selected based on the multi-field influencing factors, and the motor speed adjustment command and the feed speed adjustment command are adjusted until the temperature of the core area of ​​the cavity stabilizes within the preset operating range and the multi-field characteristics remain normal.

8. The intelligent temperature control method for medical waste treatment equipment based on multiphysics coupling according to claim 7, characterized in that, The process of constructing the preset temperature-speed rotation table and the preset temperature-feed rate correlation table includes: Combining the process requirements of medical waste treatment equipment with multi-field coupling rules, and based on the temperature control target, the temperature control interval is divided into multiple temperature sub-intervals, each temperature sub-interval corresponding to different operating states of the equipment and corresponding multi-field characteristic ranges. For each temperature sub-range, combined with the aforementioned multi-field influencing factors, the core region temperature change results and multi-field characteristic changes under different motor speeds are simulated by dynamically correcting the multi-physics coupling model, and the correspondence between motor speed, temperature change, and multi-field characteristic changes is recorded. The correspondence obtained from the simulation was analyzed, and the influence coefficients of motor speed on temperature and motor speed on multi-field characteristics were extracted. The influence coefficients reflect the differences in influence within different temperature sub-intervals. Based on the influence coefficients of motor speed on temperature and motor speed on multi-field characteristics, and combined with the allowable fluctuation range in the temperature control target, the motor speed adjustment range corresponding to each temperature sub-interval is determined so that adjusting the motor speed within this adjustment range can stabilize the temperature within the corresponding temperature sub-interval and keep the multi-field characteristics within the allowable range. By dynamically modifying the multiphysics coupling model, the core region temperature change and multi-field characteristic change under different feed rates are simulated. The influence coefficient of feed rate on temperature and the influence coefficient of multi-field characteristics are extracted, and the feed rate adjustment range corresponding to each temperature sub-interval is determined. The temperature sub-intervals, the corresponding motor speed adjustment ranges, the influence coefficient of motor speed on temperature, and the influence coefficient of motor speed on multi-field characteristics are organized in tabular form, so that each table entry includes the temperature range, speed adjustment direction, upper and lower limits of adjustment range, the influence coefficient of motor speed on temperature, and the influence coefficient of motor speed on multi-field characteristics, thus generating a preliminary temperature-speed correlation table. Organize the temperature sub-intervals, the corresponding feed rate adjustment ranges, the influence coefficient of feed rate on temperature, and the influence coefficient of feed rate on multi-field characteristics to generate a preliminary temperature-feed rate correlation table. Based on historical data from actual equipment operation, the adjustment range and influence coefficients in the preliminary temperature-speed correlation table and the preliminary temperature-feed speed correlation table are optimized. Based on the actual effect of parameter adjustment in different temperature sub-intervals during historical operation and the changes in multiple field characteristics, the values ​​in the preliminary correlation table that do not meet the set requirements are corrected. Through multiple rounds of actual operation verification and correction, the parameter adjustment range corresponding to each temperature sub-range in the preliminary temperature-speed correlation table and temperature-feed speed correlation table can achieve effective temperature control and stable multi-field characteristics, and finally obtain the preset temperature-speed correlation table and the preset temperature-feed speed correlation table.

9. A temperature intelligent control system for medical waste treatment equipment based on multi-physics coupling, characterized in that, include: processor; A machine-readable storage medium for storing machine-executable instructions of the processor; The processor is configured to execute the intelligent temperature control method for medical waste treatment equipment based on multi-physics coupling as described in any one of claims 1 to 8 by executing the machine-executable instructions.

10. A computer program product, characterized in that, The computer program product includes machine-executable instructions stored in a computer-readable storage medium. The processor of the intelligent temperature control system for medical waste treatment equipment based on multi-physics coupling reads the machine-executable instructions from the computer-readable storage medium and executes the machine-executable instructions, causing the intelligent temperature control system for medical waste treatment equipment based on multi-physics coupling to perform the intelligent temperature control method for medical waste treatment equipment based on multi-physics coupling as described in any one of claims 1 to 8.

Images