Method, device, system, controller and agricultural machine for controlling the pressure in an enclosed space

Abstract

The present disclosure relates to the technical field of fluid control, and provides a closed space air pressure control method, device, system, controller and agricultural machine. The method comprises: in response to detecting that a target object for opening and closing a closed space generates a preset action, acquiring a current environment parameter of the closed space, the current environment parameter at least comprising an air conditioner air supply speed parameter, an air conditioner return air pressure parameter and a target object movement speed parameter; inputting the current environment parameter into a pre-trained prediction model to acquire a current optimal pressure relief valve opening area output by the prediction model; and generating an air pressure control signal of the closed space according to the current optimal pressure relief valve opening area. By using the method, the global optimal pressure relief valve opening degree can be dynamically solved according to the real-time collected air conditioner working conditions and target object movement parameters, the ear pressure fluctuation peak value in the opening and closing process can be reliably controlled within the comfort interval, and the vehicle sealing performance and low wind noise requirements can be simultaneously considered.

Description

Technical Field

[0001] This disclosure relates to the field of fluid control technology, and in particular to a method for controlling air pressure in a confined space, a device for controlling air pressure in a confined space, a control system for air pressure in a confined space, a controller, and agricultural machinery. Background Technology

[0002] In enclosed spaces such as the cockpits of agricultural machinery, automobiles, and ships, in order to ensure the comfort of operators' ears during the opening and closing of doors and avoid eardrum pressure caused by sudden changes in air pressure, a method is currently used to install pressure relief valves on the shell of the enclosed space to quickly balance the internal and external air pressure at the moment of opening and closing the door.

[0003] Currently, commonly used pressure relief valves are mainly divided into two categories: passive pressure relief valves and active electronic pressure relief valves. Passive valves rely on the pressure difference between the inside and outside of the vehicle to overcome the valve plate's own weight or spring force to open. Passive valves have a slow response speed and significant lag, making it difficult to keep up with the instantaneous and drastic changes in air pressure during door opening and closing. Active electronic pressure relief valves collect pressure signals from inside and outside the vehicle through sensors, and the controller calculates the target opening degree and drives the actuator to adjust the valve port area. However, the control logic of existing active pressure relief valves is usually extremely simplified. Most of them use simple design variables and control the valve through pre-calibrated empirical lookup tables or simple linear rules. Such simplified models cannot truly reflect the complex nonlinear effects of the flow field and pressure dynamic distribution inside the confined space, and therefore cannot dynamically solve for the globally optimal pressure relief valve opening degree based on actual operating data. Summary of the Invention

[0004] Therefore, it is necessary to provide a method for controlling air pressure in a confined space, a device for controlling air pressure in a confined space, a control system for air pressure in a confined space, a controller, and agricultural machinery to address the aforementioned technical problems.

[0005] In a first aspect, this disclosure provides a method for controlling air pressure in a confined space, the method comprising: In response to the detection of a target object for opening or closing the enclosed space generating a preset action, the current environmental parameters of the enclosed space are acquired; the current environmental parameters include at least the air conditioning supply air speed parameter, the air conditioning return air pressure parameter, and the movement speed parameter of the target object; The current environmental parameters are input into a pre-trained prediction model to obtain the current optimal pressure relief valve opening area output by the prediction model. The prediction model is obtained by fitting and training a sample dataset. Each sample data pair in the sample dataset consists of the sample environmental parameters of the sealed space and the corresponding optimal pressure relief valve opening area. The optimization constraints of the optimal pressure relief valve opening area include preset ear pressure comfort constraints. Based on the current optimal pressure relief valve opening area, a pressure control signal for the sealed space is generated.

[0006] In one embodiment, the step of constructing the sample data pairs includes: The fluid domain of the enclosed space is modeled to obtain a three-dimensional fluid domain model that includes the air conditioning inlet, the air conditioning return outlet, the pressure relief valve opening, and the target object. The current sample environment parameters are input into the three-dimensional fluid domain model for fluid simulation to determine the pressure change range of the ear region during the movement of the target object and all pressure relief valve opening areas that satisfy the ear pressure comfort constraint, thus obtaining the set of pressure relief valve opening areas corresponding to the current sample environment parameters. Based on the set of pressure relief valve opening areas, the optimal pressure relief valve opening area is determined, and the current sample environment parameters and the optimal pressure relief valve opening area are used to form a sample data pair.

[0007] In one embodiment, determining the optimal pressure relief valve opening area based on the set of pressure relief valve opening areas includes: From the set of pressure relief valve opening areas, select the smallest pressure relief valve opening area as the optimal pressure relief valve opening area.

[0008] In one embodiment, the steps for constructing the prediction model include: The sample dataset is divided into a training set and a test set; The preset initial model is fitted and trained based on the training set to construct a mapping function relationship between the sample environmental parameters and the optimal pressure relief valve opening area; The prediction error of the mapping function relationship is calculated based on the test set, and if the prediction error is less than a preset threshold, the current mapping function relationship is determined as the prediction model.

[0009] In one embodiment, the preset initial model is a surrogate model, which is any one of a multinomial response surface model, a Kriging model, a radial basis function neural network model, or a support vector regression model. The step of fitting and training a preset initial model based on the training set to construct a mapping function relationship between the sample environmental parameters and the optimal pressure relief valve opening area includes: The sample environment parameters are converted into multi-dimensional input feature vectors, and the corresponding optimal pressure relief valve opening area is converted into output labels. The multidimensional input feature vectors in the training set are input into the preset initial model to obtain the predicted opening area; Based on the loss function value between the predicted opening area and the output label, the internal parameters of the preset initial model are iteratively updated; When the loss function value meets the preset convergence condition, the iteration stops, and the preset initial model after training is determined as the mapping function relationship between the multidimensional input feature vector and the output label.

[0010] In one embodiment, the current environmental parameters further include at least one of the current temperature, current humidity, and current air pressure within the enclosed space, and the sample environmental parameters are of the same parameter type as the current environmental parameters.

[0011] In one embodiment, the ear pressure comfort constraint condition is that the pressure fluctuation amplitude in the ear region during the movement of the target object does not exceed a preset value. The ear region includes a left spherical region with a preset radius centered on the three-dimensional coordinates of the left ear position and a right spherical region with a preset radius centered on the three-dimensional coordinates of the right ear position. The pressure fluctuation amplitude is the average of the average static pressure fluctuation peak value of the left spherical region during the movement of the target object and the average static pressure fluctuation peak value of the right spherical region during the movement.

[0012] Secondly, this disclosure also provides a closed space air pressure control device, the device comprising: The acquisition module is used to acquire the current environmental parameters of the enclosed space in response to the detection of a target object for opening or closing the enclosed space and a preset action. The current environmental parameters include at least the air conditioning supply air speed parameter, the air conditioning return air pressure parameter, and the movement speed parameter of the target object. The model processing module is used to input the current environmental parameters into the pre-trained prediction model to obtain the current optimal pressure relief valve opening area output by the prediction model. The prediction model is obtained by fitting and training a sample dataset. Each sample data pair in the sample dataset consists of the sample environmental parameters of the sealed space and the corresponding optimal pressure relief valve opening area. The optimization constraints of the optimal pressure relief valve opening area include preset ear pressure comfort constraints. The control generation module is used to generate a pressure control signal for the sealed space based on the current optimal pressure relief valve opening area.

[0013] Thirdly, embodiments of the present invention provide a closed-space air pressure control system, including a data acquisition component, a controller, and a pressure relief valve actuator, wherein the controller is signal-connected to both the data acquisition component and the pressure relief valve actuator. The data acquisition component is configured to acquire the current environmental parameters of the enclosed space; the current environmental parameters include at least the air conditioning supply air speed parameter, the air conditioning return air pressure parameter, and the movement speed parameter of the target object; the target object is used to open and close the enclosed space. The controller is configured to, in response to detecting the target object and generating a preset action, acquire the current environmental parameters and input the current environmental parameters into a pre-trained prediction model to obtain the current optimal pressure relief valve opening area output by the prediction model; the prediction model is obtained by fitting and training a sample dataset, wherein each sample data pair in the sample dataset consists of the sample environmental parameters of the sealed space and the corresponding optimal pressure relief valve opening area, and the optimization constraints of the optimal pressure relief valve opening area include preset ear pressure comfort constraints; based on the current optimal pressure relief valve opening area, a pressure control signal for the sealed space is generated and sent to the pressure relief valve actuator; The pressure relief valve actuator is configured to adjust the opening area of ​​the pressure relief valve based on the air pressure control signal; the opening of the pressure relief valve is in communication with the sealed space.

[0014] Fourthly, embodiments of this disclosure provide a controller, including a memory and a processor, wherein the memory stores a computer program, and the processor executes the computer program to implement the steps of a closed space air pressure control method.

[0015] Fifthly, embodiments of this disclosure provide an agricultural machine, including a closed-space air pressure control system or controller.

[0016] One of the above technical solutions has the following advantages or beneficial effects: by acquiring current environmental parameters in response to the preset actions of the target object, and these current environmental parameters include at least the air conditioning supply air velocity parameter, the air conditioning return air pressure parameter, and the target object's motion velocity parameter, a pre-trained prediction model is used to output the current optimal pressure relief valve opening area that satisfies the ear pressure comfort constraint in real time, and a pressure control signal is generated accordingly. Since the types of current environmental parameters at least cover the air conditioning supply air, return air, and the target object's motion state, the sample data of the prediction model is also constructed based on considering the influence of the air conditioning inlet and return air outlet on the internal flow field of the enclosed space. Therefore, the mapping function from environmental parameters to the optimal pressure relief valve opening area learned by the prediction model inherently reflects the complex coupling effect of the supply and return airflows on the transient pressure field during door opening and closing during air conditioning system operation. In actual control, the predictive model can output the globally optimal pressure relief valve opening degree under the current operating conditions within milliseconds based on the real-time collected air conditioning conditions and door motion parameters. This breaks through the technical bottlenecks of passive pressure relief valve response lag and active electronic pressure relief valve control logic simplification and inability to solve for optimal flow, and reliably controls the peak of ear pressure fluctuation during door opening and closing within the comfort range. Attached Figure Description

[0017] Figure 1 This is a schematic flowchart of a method for controlling air pressure in a confined space in one embodiment; Figure 2 This is a flowchart illustrating the construction and application of a prediction model in one embodiment; Figure 3 This is a structural block diagram of a closed-space air pressure control device in one embodiment; Figure 4 This is a structural block diagram of a closed-space air pressure control system in one embodiment; Figure 5 This is a diagram of the internal structure of the controller in one embodiment. Detailed Implementation

[0018] To make the objectives, technical solutions, and advantages of this disclosure clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this disclosure.

[0019] In this document, the term "embodiment" means that a particular feature, structure, or characteristic described in connection with an embodiment may be included in at least one embodiment of this disclosure. The appearance of this phrase in various places throughout the specification does not necessarily refer to the same embodiment, nor is it a separate or alternative embodiment mutually exclusive with other embodiments. It will be explicitly and implicitly understood by those skilled in the art that the embodiments described herein can be combined with other embodiments.

[0020] In enclosed spaces such as the cockpits of agricultural machinery, automobiles, and ships, air conditioning systems are typically installed to regulate temperature, humidity, and airflow in order to maintain a comfortable internal environment. The air conditioning inlet supplies a large amount of air into the cabin, while the return air inlet draws in air. The speed, direction, and flow rate of these airflows significantly alter the pressure field distribution within the enclosed space. Especially during the brief process of opening and closing the door, the compression / expansion effects of the air conditioning airflow and the door movement are superimposed, causing the instantaneous pressure change in the ear area to far exceed the amplitude caused by the door movement alone. However, currently, no pressure relief valve control system incorporates parameters such as the air conditioning inlet speed and return air pressure into its control model. This results in low control accuracy, difficulty in further optimizing ear pressure comfort, and an inability to balance overall vehicle sealing and wind noise performance while ensuring ear pressure comfort. Therefore, this disclosure provides a method for controlling air pressure in enclosed spaces.

[0021] In one embodiment, such as Figure 1 As shown, a method for controlling air pressure in a confined space is provided. Taking the application of this method to an intelligent vehicle-mounted controller in the operating room of agricultural machinery as an example, the method includes the following steps: S102, in response to detecting a target object for opening or closing the enclosed space and generating a preset action, the current environmental parameters of the enclosed space are obtained; the current environmental parameters include at least the air conditioning supply air speed parameters, the air conditioning return air pressure parameters, and the movement speed parameters of the target object.

[0022] The enclosed space can be a limited, enclosed space such as the operating room or driver's cab of agricultural machinery, a ship's cabin, or a train carriage, where rapid changes in air pressure can cause ear discomfort; the specific type is not limited. The target object is a component used to open and close the enclosed space, such as a vehicle door / cabin door, window, sunroof, or rear door. The number of target objects in this scheme can be set according to actual needs; for example, it can be just a vehicle door, or the operation of both the vehicle door and window can be monitored simultaneously, etc., and the specific type is not limited here. The preset actions of the target object may include opening and / or closing actions, and of course, other actions, as long as the action may cause a sudden change or a sudden trend in the sealing state or internal volume of the enclosed space, causing a momentary imbalance in the internal and external air pressure or a momentary air pressure fluctuation sufficient to cause ear pressure discomfort, it can fall within the scope of the above-mentioned preset actions; for example, the preset actions of the target object may specifically include at least one of the following: the moment of moving from a fully closed state to an initial open position, the moment of moving from a fully open state to an initial closed position, the moment of being about to fully close at the end of the closing stroke, the moment of continuing to move after a short pause during the opening and closing action, and the moment of unlocking or locking the locking mechanism of the target object.

[0023] In this step, taking a door as an example, the current environmental parameters of the enclosed space can be obtained by detecting the preset actions of the target object through a door status sensor, door handle sensor, or door pivot angular velocity sensor. Specifically, generating a preset action means detecting that the motion state of the target object (such as rotational angular velocity or displacement) exceeds a preset micro-motion threshold (for example, a door rotational angular velocity greater than 0.5 rad / s, or a displacement greater than 5 mm; specific values ​​can be calibrated according to the vehicle model). The threshold is set to effectively filter out invalid micro-shaking caused by vehicle driving bumps or ambient wind, ensuring that the prediction model is only activated and the current motion speed parameters are obtained when the occupant actually performs the opening or closing operation. This ensures timely control while reducing system energy consumption and extending the life of the actuator. Of course, the above method of obtaining the current environmental parameters of the enclosed space can also be applied to situations where the target object is a window, sunroof, tailgate, etc., which will not be elaborated here.

[0024] Data acquisition components for obtaining current environmental parameters can be configured according to the type of current environmental parameter, and may include, for example, a driver's cab interior temperature and humidity sensor, a driver's cab air conditioning vent speed sensor, a driver's cab return air vent pressure sensor, and a door hinge angular velocity sensor. The air conditioning supply speed parameter corresponds to the air conditioning vent speed, the air conditioning return air pressure parameter corresponds to the return air vent pressure, and the target object's movement speed parameter corresponds to the door hinge angular velocity. In some embodiments, current environmental parameters may also include the current temperature, current humidity, and current air pressure within the enclosed space. Correspondingly, the data acquisition components for obtaining current environmental parameters may include temperature sensors, humidity sensors, and air pressure sensors arranged in the enclosed space. These sensors can be integrated into a multi-functional environmental sensor module or arranged independently. Specifically, temperature and humidity sensors can be arranged near the operator's ears (such as on both sides of the seat headrest or in front of the control panel) to more accurately obtain the temperature and humidity of the occupant's breathing area; they can also be arranged on the inner walls, control panel, or top lining of the enclosed space to obtain indoor temperature and humidity reflecting the overall space status. The barometric pressure sensor can be positioned at the geometric center of a confined space or at the height of the occupant's head to directly measure the ambient static pressure reference value acting near the occupant's eardrum. Alternatively, multiple barometric pressure sensors can be placed in the confined space, and the average value can be taken as the current barometric pressure parameter to improve the representativeness of the measurement. The parameters collected by the above sensor components can be transmitted to the controller in real time via the vehicle bus or hardwired connection.

[0025] S104, Input the current environmental parameters into the pre-trained prediction model to obtain the current optimal pressure relief valve opening area output by the prediction model; The prediction model is obtained by fitting and training the sample dataset. Each sample data pair in the sample dataset consists of the sample environmental parameters of the closed space and the corresponding optimal pressure relief valve opening area. The optimization constraints of the optimal pressure relief valve opening area include the preset ear pressure comfort constraints.

[0026] Understandably, the prediction model is a pre-trained offline proxy model that can be deployed in an intelligent vehicle controller. Internally, the prediction model establishes an implicit mapping relationship from multi-dimensional environmental parameters to the optimal pressure relief valve opening area. Upon inputting the current environmental parameters, the prediction model can immediately output the pressure relief valve opening area that satisfies ear pressure comfort constraints and optimizes sealing. Furthermore, ear pressure comfort constraints include, for example, controlling the peak pressure change in both ears within ±100 Pa, or even ±50 Pa, during door opening and closing. Since the current environmental parameters at least cover air conditioning supply air velocity, air conditioning return air pressure, and the target object's movement speed, and the prediction model's sample data is constructed based on sample environmental parameters consistent with the current environmental parameter types, this means that the model has already learned the complex flow field characteristics within a confined space under the superposition of air conditioning intake and return airflow and the opening and closing actions of the target object through a large amount of sample data during the offline training phase. Therefore, this prediction model can accurately and completely reflect the nonlinear influence of air conditioning operation on the transient pressure field of door opening and closing, as well as the pressure fluctuation law in the ear area caused by it. It avoids the problem that the control strategy is out of touch with the actual physical process due to the neglect of air conditioning factors or the oversimplification of complex flow fields in existing technologies. Thus, it outputs a more accurate and robust optimal pressure relief valve opening area in practical applications.

[0027] S106, generate a pressure control signal for the sealed space based on the current optimal pressure relief valve opening area.

[0028] In this step, the intelligent vehicle controller outputs a corresponding electrical signal to the pressure relief valve actuator based on the obtained optimal opening area. Optionally, the pressure relief valve actuator uses an electrically controlled proportional solenoid valve to adjust the opening area of ​​the pressure relief valve in real time, thereby balancing the internal and external air pressure at the moment of closing or opening the door to ensure ear pressure comfort. In addition, the pressure relief valve actuator can also use a digital stepper motor drive mechanism, a combination of multiple parallel on / off solenoid valves, a piezoelectric or shape memory alloy drive mechanism, etc., which can be configured according to specific needs and are not limited here.

[0029] The air pressure control method for a confined space described above utilizes a pre-trained prediction model to instantaneously calculate the optimal pressure relief valve opening area based on real-time collected air conditioning and door movement parameters. This enables precise control of ear pressure fluctuations throughout the entire door opening and closing process, breaking through the upper limit of traditional control methods where pressure peaks are difficult to control within ±50Pa, and achieving the optimal balance between comfort and sealing.

[0030] In some embodiments, the steps for constructing sample data pairs specifically include: modeling the fluid domain of a confined space to obtain a three-dimensional fluid domain model containing an air conditioning inlet, an air conditioning return outlet, a pressure relief valve opening, and the target object; inputting the current sample environment parameters into the three-dimensional fluid domain model for fluid simulation to determine the pressure change range of the ear region during the movement of the target object and satisfy the ear pressure comfort constraint for all pressure relief valve opening areas, thereby obtaining a set of pressure relief valve opening areas corresponding to the current sample environment parameters; determining the optimal pressure relief valve opening area based on the set of pressure relief valve opening areas, and forming a sample data pair with the current sample environment parameters and the optimal pressure relief valve opening area.

[0031] The modeling of the 3D fluid domain can be achieved using common 3D modeling software in this field, such as UG, SolidWorks, Catia, and Pro-e, without specific limitations. Taking UG modeling as an example, the fluid domain of the control room is modeled using full parametric methods. Full parametric modeling means that all geometric features involved in the model (such as the size of the air conditioning inlet, the position of the air outlet, the opening area of ​​the pressure relief valve, the internal volume of the control room, the shape of the door, and the clearance between openings) are defined as parametric variables, rather than being fixed as a set of geometric values. In this way, during subsequent iterative optimization, the parameter optimization platform can directly modify the values ​​of these parametric variables, and the model can be automatically updated and new geometry generated, eliminating the need for manual remodeling. Furthermore, the modeling objects can include simplified models of the air fluid domain inside the control room, the air fluid domain outside the door, the air conditioning inlet, the air conditioning return vent, the pressure relief valve opening, and moving parts such as the door. After modeling is completed, the model file can be converted into a common exportable format (e.g., .x_t format) so that it can be recognized and imported by subsequent fluid simulation software.

[0032] Import the aforementioned model file into the fluid simulation software. Commonly used computational fluid dynamics (CFD) software in this field can be used, such as Fluent, StarCCM+, StarCD, FIDAP, STREAM, Shipflow, and XFLow. Fluent is used as an example here, but it is not a limitation. After importing, mesh generation is performed in the simulation software, that is, the continuous fluid space is discretized into a finite number of tiny cells for numerical solution. The type, size, and density of the mesh can be selected based on the balance between computational accuracy and computational resources. Then, a physical model suitable for describing the transient flow of opening and closing doors is selected, such as a compressible flow model or a turbulent flow model. Next, the boundary conditions for the simulation are set. Boundary conditions are the input prerequisites for fluid simulation calculations, and their values ​​are directly set based on the current sample environment parameters. The sample environment parameters are a set of specific parameter values ​​obtained, such as the current temperature, current humidity, current air pressure, air conditioning inlet velocity, air conditioning return air pressure, and door rotation speed in the operating room. These parameters are applied to the simulation model as temperature boundary, velocity inlet boundary, pressure outlet boundary, and moving mesh motion boundary, respectively. In addition, the initialization conditions for the simulation also need to be set, such as the initial velocity distribution and pressure distribution of the flow field.

[0033] To efficiently evaluate the performance of sample environmental parameters under different pressure relief valve opening areas, the above operations such as mesh generation, physical model selection, boundary condition setting, and initialization condition setting can be written into a pre-processing script file, and the output operations of simulation results (especially the extraction of pressure data in the ear region) can be written into a post-processing script file, thereby automating the simulation calculation; through the script, simulations of different pressure relief valve opening areas can be performed in batches.

[0034] Under each set of sample environmental parameters, the pressure relief valve opening area is used as an adjustable design variable. One or more fluid simulations are run. The post-processing script automatically extracts the pressure change data of the occupant's ear area during the process and determines whether the peak pressure change falls within the preset ear pressure comfort constraint (e.g., not exceeding ±100Pa). By continuously adjusting the pressure relief valve opening area for trial calculations, all pressure relief valve opening areas that can meet the ear pressure comfort constraint under the specific sample environmental parameters can be found. They together constitute a set of pressure relief valve opening areas.

[0035] Furthermore, in some embodiments, to improve the computational accuracy of the simulation model, it can be first corrected by benchmarking with one or more sets of known experimental data. For example, a door opening and closing test is conducted on an actual vehicle to obtain the pressure change curve in the ear region. The simulation results under the same conditions are compared with these data. By adjusting the parameters in the simulation model (such as turbulence model coefficients, mesh density, damping coefficients, etc.), the simulation results are made to match the experimental data well, forming a validated high-precision simulation model. Subsequent batch simulations are all based on this corrected model.

[0036] Furthermore, based on the set of pressure relief valve opening areas, the optimal pressure relief valve opening area can be determined, and the current sample environment parameters and the optimal pressure relief valve opening area can be used to form a sample data pair.

[0037] One criterion for determining the optimal pressure relief valve opening area is to select the smallest pressure relief valve opening area from the set of samples as the optimal pressure relief valve opening area corresponding to the environmental parameters of that set of samples. In this way, the opening area of ​​the pressure relief valve can be minimized while ensuring ear pressure comfort, thereby maximizing the sealing performance of the confined space and reducing wind noise during driving.

[0038] Finally, this set of sample environmental parameters (such as temperature, humidity, air pressure, air conditioning inlet velocity, air conditioning return vent pressure, and door rotation speed) are combined with their corresponding optimal pressure relief valve opening area to form a sample data pair; a large number of sample data pairs covering different operating conditions are collected to form a sample dataset for training the prediction model.

[0039] Through the closed-loop process of fully parametric modeling, simulation automation, experimental calibration, and parameter optimization that integrates air conditioning factors, it is possible to systematically generate accurate sample data of opening area that meets ear pressure comfort and has optimal sealing performance. This allows the subsequent training of the prediction model to learn the globally optimal control law, thus effectively overcoming the shortcomings of traditional control schemes that ignore the influence of air conditioning on the transient flow field when opening and closing doors, resulting in a mismatch between the control strategy and the actual physical process and poor ear pressure control effect.

[0040] In some embodiments, determining the optimal pressure relief valve opening area based on the set of pressure relief valve opening areas includes: selecting the smallest pressure relief valve opening area from the set of pressure relief valve opening areas as the optimal pressure relief valve opening area.

[0041] The optimal solution is selected from the set of pressure relief valve opening areas based on the following technical considerations: the pressure relief valve opening area must simultaneously consider ear pressure comfort and the sealing performance of the enclosed space, creating a mutually restrictive technical contradiction. If the opening area is insufficient, the instantaneous flow rate of the pressure relief channel cannot balance the internal and external pressure difference caused by the opening and closing of the target object, resulting in pressure fluctuations in the ear area exceeding the comfort threshold. If the opening area is too large, the inherent sealing characteristics of the enclosed space are excessively weakened, introducing increased wind noise, increased air conditioning load, and the risk of external pollutant intrusion. Existing passive fixed-opening or active pressure relief valves based on simplified rules cannot accurately determine the minimum effective opening area that just meets the ear pressure comfort constraint under different operating conditions. This solution pre-sets ear pressure comfort as a quantitative constraint (e.g., peak pressure fluctuation in both ear areas ≤ ±100Pa). Under this constraint, minimizing the opening area of ​​the pressure relief valve is the optimization direction, thus transforming the multi-objective trade-off into a single-objective constraint optimization problem. The resulting sample data pairs naturally contain the optimization characteristic of minimizing the opening while satisfying comfort. After training, the opening value output by the prediction model can control the pressure relief area within the lower bound of the feasible region under the current operating conditions, while ensuring that ear pressure fluctuations do not exceed the limit. In principle, this suppresses the deterioration of sealing caused by excessive opening, effectively reduces wind noise and additional energy loss of the air conditioning system during high-speed driving, and achieves a dynamic optimal balance between pressure balance and sealing performance.

[0042] In some embodiments, the steps of constructing the prediction model include: dividing the sample dataset into a training set and a test set; fitting and training a preset initial model based on the training set to construct a mapping function relationship between sample environmental parameters and the optimal pressure relief valve opening area; solving the prediction error of the mapping function relationship based on the test set, and determining the current mapping function relationship as the prediction model if the prediction error is less than a preset threshold.

[0043] The training and test sets can be divided in a preset ratio (e.g., 8:2). This division aims to use independent subsets of data to perform model fitting and generalization ability evaluation separately, preventing overfitting. The preset initial model is a surrogate model, which is a mathematical model that constructs an approximate functional relationship between input and output using limited sample data. Its computational cost is far lower than the original physical simulation. Optional surrogate models include multinomial response surface models, Kriging models, radial basis function neural network models, and support vector regression models, selected based on sample size, degree of nonlinearity, and real-time requirements. The training process essentially drives the adjustment of the surrogate model's internal parameters through the sample dataset, enabling it to learn the mapping law from multidimensional input feature vectors (i.e., sample environment parameters) to single-dimensional output labels (i.e., the optimal pressure relief valve opening area). After training, the error of the obtained mapping function relationship is evaluated using the test set. If the coefficient of determination R...2 If the accuracy is greater than a preset value (e.g., 0.95) or the root mean square error is less than a preset value, it indicates that the prediction accuracy of the mapping function relationship meets the requirements of engineering applications. This can be solidified into a prediction model and deployed in the vehicle controller. Therefore, in the actual control process, the controller does not need to call a computational fluid dynamics (CFD) simulation program; it only needs to input the real-time collected environmental parameters into this proxy model to obtain the current optimal pressure relief valve opening area within milliseconds, thus ensuring the real-time performance of the control from an architectural perspective.

[0044] In some embodiments, the preset initial model is a Kriging model. Sample environmental parameters are transformed into multi-dimensional input feature vectors, and the corresponding optimal pressure relief valve opening area is transformed into an output label. The multi-dimensional input feature vectors from the training set are input into the preset initial model to obtain the predicted opening area. Based on the loss function value (e.g., mean squared error) between the predicted opening area and the output label, the internal parameters of the preset initial model (e.g., regression coefficients and correlation parameters in the Kriging model) are iteratively updated using an optimization algorithm (e.g., gradient descent or maximum likelihood estimation). Iteration stops when the loss function value meets a preset convergence condition (e.g., the change over N consecutive iterations is less than a set threshold), and the trained preset initial model is determined as the mapping function relationship. This embodiment utilizes the Kriging model to effectively fit strongly nonlinear relationships and accurately reflect the dynamic pressure characteristics within a confined space.

[0045] In some embodiments, the current environmental parameters also include at least one of the current temperature, current humidity, and current air pressure within the enclosed space, and the sample environmental parameters are of the same parameter type as the current environmental parameters. Adding these parameters can more comprehensively reflect the impact of air properties on pressure fluctuations and improve the accuracy of the prediction model.

[0046] In some embodiments, the ear pressure comfort constraint is: the pressure fluctuation amplitude in the ear region during the movement of the target object does not exceed a preset value (e.g., ±50Pa). The ear region includes a left spherical region centered on the three-dimensional coordinates of the left ear with a preset radius and a right spherical region centered on the three-dimensional coordinates of the right ear with a preset radius. The pressure fluctuation amplitude is the arithmetic mean of the maximum change (or peak fluctuation) of the average static pressure in the left spherical region during the movement of the target object and the maximum change (or peak fluctuation) of the average static pressure in the right spherical region during the movement. This definition quantifies the ear pressure impact felt by the occupant's ears, accurately mapping comfort requirements. Through the predictive model provided in this application and the linkage control of the proportional solenoid valve, the peak value of the average static pressure fluctuation in the spherical region of the ears can be effectively controlled at a low level (e.g., between 30Pa and 40Pa) during the instant the target object (e.g., a car door) moves. This significantly overcomes the comfort limit of traditional passive pressure relief valves or simple lookup table control schemes, which struggle to control the pressure peak within ±50Pa.

[0047] like Figure 2 As shown, in a more specific embodiment, the construction of the above sample data pairs and the training of the prediction model can be accomplished through a real-time virtual numerical system integrating modeling, simulation, optimization, and model training. The system's workflow mainly includes the following steps: First, a fully parameterized fluid domain model of the confined space is constructed. Using 3D software, a fully parameterized model of the fluid domain in the confined space is created, generating a 3D fluid domain model file that includes the air conditioning inlet, air conditioning return inlet, pressure relief valve opening, and target objects (such as vehicle doors). This model can respond to control commands from the parameter optimization module, allowing for rapid adjustment of key design variables such as the pressure relief valve opening area.

[0048] Second, fluid simulation calculations are performed and corrected against experimental data. The fully parameterized fluid domain model is imported into the fluid simulation software. Pre-processing scripts are written to set simulation boundary conditions (including sample environmental parameters from the parameter optimization module, such as temperature, humidity, air pressure, air conditioner inlet velocity, air conditioner return vent pressure, and target object movement velocity). Mesh generation is then performed, and the solver is run. Post-processing scripts automatically extract the average static pressure fluctuation peaks of two spherical regions with preset radii centered on the left and right ear coordinates during the target object's movement. Simulation results are fed back to the parameter optimization module. Simultaneously, actual experimental data is used to correct the simulation model, forming a closed-loop verification process of simulation-experimental benchmarking-model correction-re-simulation, thus improving simulation accuracy.

[0049] Third, iterative optimization is achieved through a multidisciplinary parameter optimization platform. Using parameter optimization software as the central element, with the pressure relief valve opening area as the design variable and the objective function being that the peak pressure fluctuation in both ears does not exceed the preset ear pressure comfort constraint, the fluid simulation is driven to perform multiple rounds of automated iterative calculations. In each iteration, the parameter optimization platform outputs updated sample environmental parameter mean values ​​and control commands to the fluid simulation module, and receives the static pressure data of the ear region from the simulation output, filtering out the set of pressure relief valve opening areas that meet the ear pressure comfort constraint.

[0050] Fourth, numerical model training is performed to accelerate optimization. A large number of sample environmental parameters generated during the iteration process of the parameter optimization platform, along with their corresponding optimal pressure relief valve opening areas (usually the minimum opening area satisfying the constraints), are used as sample data pairs. These are input into a pre-set initial model for fitting training, constructing a mapping function relationship from the multi-dimensional input feature vector (sample environmental parameters) to the output label (optimal pressure relief valve opening area). The trained mapping function relationship is fed back to the parameter optimization platform, which can replace some time-consuming fluid simulation calculations, significantly accelerating the subsequent optimization process.

[0051] Fifth, the final prediction model is output. After sufficient iterative training and testing, the mapping function relationship becomes a prediction model that can be used for real-time control. It is then deployed in the intelligent vehicle controller to realize the function of instantaneously outputting the current optimal pressure relief valve opening area based on real-time environmental parameters.

[0052] By constructing a predictive model using the aforementioned real-time virtual numerical system that integrates fully parametric modeling, fluid simulation and experimental benchmarking, multidisciplinary parameter optimization, and numerical model training, at least the following technical effects can be achieved: Firstly, by constructing a fully parameterized fluid domain model that includes the air inlet and outlet of the air conditioner, and incorporating the air supply velocity and return air pressure into the simulation boundary conditions, the sample data can truly reflect the coupling effect of the supply and return airflow on the transient pressure field of the door opening and closing during the operation of the air conditioning system. This makes up for the deficiency of traditional control schemes that generally ignore air conditioning factors when designing pressure relief valves, and ensures from the source that the input and output mapping relationship of the prediction model is based on the complete physical scenario.

[0053] Secondly, through a closed-loop verification mechanism of simulation-experimental benchmarking-model correction-re-simulation, the simulation model is continuously corrected using measured data, which effectively improves the calculation accuracy of fluid simulation. This makes the sample data used for training more closely approximate the real working conditions, thereby ensuring that the prediction model has good control accuracy after actual deployment.

[0054] Third, by using the large amount of sample environmental parameters—optimal pressure relief valve opening area data accumulated by the parameter optimization platform during the iterative optimization process—for numerical model training, a mapping function relationship from multidimensional environmental parameters to the optimal opening area is constructed. This mapping function is then fed back to the optimization platform to replace some of the time-consuming fluid simulation, significantly accelerating the search process for the optimal solution and making the entire training process practical for engineering applications.

[0055] Fourth, the predictive model finally deployed in the intelligent vehicle controller has already completed a large number of simulation iterations and model training in the offline stage. In actual operation, no online fluid simulation is required. It only needs to receive the real-time collected air conditioning conditions and door motion parameters, and can output the globally optimal pressure relief valve opening area under the current conditions in milliseconds. This reliably controls the peak ear pressure fluctuation during the door opening and closing process within ±50Pa or even better comfort range, while maintaining the minimum pressure relief valve opening area to take into account the vehicle's sealing and low wind noise. This achieves a comprehensive optimal control effect in terms of real-time performance, comfort, and sealing that cannot be achieved by existing passive or active pressure relief valve solutions.

[0056] The deficiencies of the above solutions and the proposed solutions are the result of the inventor's practice and careful research. Therefore, the discovery process of the above problems and the solutions proposed in this disclosure below should be considered as the inventor's contribution to this disclosure.

[0057] It should be understood that, for the foregoing method embodiments, although the steps in the flowcharts are shown sequentially as indicated by the arrows, these steps are not necessarily executed in the order indicated by the arrows. Unless explicitly stated herein, there is no strict order restriction on the execution of these steps, and they can be executed in other orders. Moreover, at least some steps in the flowcharts of the method embodiments may include multiple sub-steps or multiple stages. These sub-steps or stages are not necessarily completed at the same time, but can be executed at different times. The execution order of these sub-steps or stages is not necessarily sequential, but can be performed alternately or in turn with other steps or at least a portion of the sub-steps or stages of other steps.

[0058] Based on the same inventive concept, this disclosure also provides a closed-space air pressure control device for implementing the closed-space air pressure control method described above. The solution provided by this device is similar to the solution described in the above method; therefore, the specific limitations in one or more closed-space air pressure control device embodiments provided below can be found in the limitations of the closed-space air pressure control method described above, and will not be repeated here.

[0059] In one embodiment, such as Figure 3 As shown, a closed space air pressure control device 300 is provided, including: a data acquisition module 301, a model processing module 302, and a control generation module 303.

[0060] The acquisition module 301 is used to generate a preset action in response to the detection of a target object for opening and closing a sealed space, and to acquire the current environmental parameters of the sealed space; the current environmental parameters include at least the air conditioning supply air speed parameters, the air conditioning return air pressure parameters, and the movement speed parameters of the target object.

[0061] The model processing module 302 is used to input the current environmental parameters into the pre-trained prediction model to obtain the current optimal pressure relief valve opening area output by the prediction model. The prediction model is obtained by fitting and training the sample dataset. Each sample data pair in the sample dataset consists of the sample environmental parameters of the closed space and the corresponding optimal pressure relief valve opening area. The optimization constraints of the optimal pressure relief valve opening area include preset ear pressure comfort constraints.

[0062] The control generation module 303 is used to generate a pressure control signal for the sealed space based on the current optimal pressure relief valve opening area.

[0063] In some embodiments, the enclosed space air pressure control device 300 further includes a sample data generation module, specifically used to model the fluid domain of the enclosed space to obtain a three-dimensional fluid domain model including the air conditioning inlet, air conditioning return outlet, pressure relief valve opening, and target object; input the current sample environment parameters into the three-dimensional fluid domain model for fluid simulation to determine the pressure change range of the ear region during the movement of the target object and satisfy the ear pressure comfort constraint for all pressure relief valve opening areas, thereby obtaining a set of pressure relief valve opening areas corresponding to the current sample environment parameters; based on the set of pressure relief valve opening areas, determine the optimal pressure relief valve opening area, and form a sample data pair with the current sample environment parameters and the optimal pressure relief valve opening area.

[0064] In some embodiments, the sample data generation module is specifically used to select the smallest pressure relief valve opening area from the set of pressure relief valve opening areas as the optimal pressure relief valve opening area.

[0065] In some embodiments, the enclosed space air pressure control device 300 further includes a model training module, specifically used to divide the sample dataset into a training set and a test set; to fit and train a preset initial model based on the training set to construct a mapping function relationship between sample environmental parameters and the optimal pressure relief valve opening area; to solve the prediction error of the mapping function relationship based on the test set, and to determine the current mapping function relationship as the prediction model if the prediction error is less than a preset threshold.

[0066] In some embodiments, the preset initial model is a surrogate model, which can be any one of a multinomial response surface model, a Kriging model, a radial basis function neural network model, or a support vector regression model. The model training module is specifically used to convert the sample environment parameters into multidimensional input feature vectors and convert the corresponding optimal pressure relief valve opening area into an output label. The multidimensional input feature vectors in the training set are input into the preset initial model to obtain the predicted opening area. Based on the loss function value between the predicted opening area and the output label, the internal parameters of the preset initial model are iteratively updated. When the loss function value meets the preset convergence condition, the iteration stops, and the trained preset initial model is determined as the mapping function relationship between the multidimensional input feature vectors and the output label.

[0067] In some embodiments, the current environmental parameters also include at least one of the current temperature, current humidity, and current air pressure in the enclosed space, and the parameter type of the sample environmental parameters is consistent with that of the current environmental parameters; and / or, the ear pressure comfort constraint is that the pressure fluctuation amplitude in the ear region during the movement of the target object does not exceed a preset value, wherein the ear region includes a left spherical region with a preset radius centered on the three-dimensional coordinates of the left ear position and a right spherical region with a preset radius centered on the three-dimensional coordinates of the right ear position, and the pressure fluctuation amplitude is the average of the average static pressure fluctuation peak value of the left spherical region during the movement of the target object and the average static pressure fluctuation peak value of the right spherical region during the movement.

[0068] Each module in the aforementioned enclosed space air pressure control device can be implemented entirely or partially through software, hardware, or a combination thereof. These modules can be embedded in the processor of the controller in hardware form or independent of it, or stored in the memory of the controller in software form, so that the processor can call and execute the corresponding operations of each module.

[0069] Furthermore, in the above-described embodiment of the enclosed space air pressure control device, the logical division of each program module is merely illustrative. In actual applications, the functions described above can be assigned to different program modules as needed, for example, for the sake of corresponding hardware configuration requirements or the convenience of software implementation. That is, the internal structure of the enclosed space air pressure control device can be divided into different program modules to complete all or part of the functions described above.

[0070] In one embodiment, such as Figure 4 As shown, a closed-space air pressure control system is provided, including a data acquisition component, a controller, and a pressure relief valve actuator. The data acquisition component includes sensors for interior temperature and humidity in the driver's cab, air conditioning vent speed, return air pressure, and door hinge angular velocity, used to collect parameters such as current temperature, humidity, air conditioning supply speed, return air pressure, and door hinge angular velocity. The controller can be an intelligent vehicle controller, in which a pre-trained predictive model is deployed. The pressure relief valve actuator is a proportional solenoid valve that receives electrical signals, and its opening is connected to the control compartment. When a preset door action is detected, the controller acquires sensor data, calls the predictive model to determine the optimal pressure relief valve opening area, generates an air pressure control signal to drive the pressure relief valve to adjust the opening area, thereby balancing the internal and external air pressure in real time during door opening and closing, and controlling the peak ear pressure within ±50Pa.

[0071] In one embodiment, a controller is provided, such as an intelligent vehicle controller for agricultural machinery, the internal structure of which can be shown in the following diagram. Figure 5As shown, the controller includes a processor, memory, input / output interfaces, and a communication interface. The processor, memory, and input / output interfaces are connected via a system bus, and the communication interface is also connected to the system bus via the input / output interfaces. The processor provides computational and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides the environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The input / output interfaces are used for exchanging information 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, NFC (Near Field Communication), or other technologies. When the computer program is executed by the processor, it implements a method for controlling the air pressure in a confined space.

[0072] Those skilled in the art will understand that Figure 5 The structure shown is merely a block diagram of a portion of the structure related to the present disclosure and does not constitute a limitation on the controller to which the present disclosure is applied. A specific controller may include more or fewer components than those shown in the figure, or combine certain components, or have different component arrangements.

[0073] In one embodiment, an agricultural machine is provided, including the aforementioned enclosed space air pressure control system or the aforementioned controller.

[0074] In one embodiment, a computer-readable storage medium is provided having a computer program stored thereon that, when executed by a processor, implements the steps in the above method embodiments.

[0075] In one embodiment, a computer program product is provided, including a computer program that, when executed by a processor, implements the steps in the above method embodiments.

[0076] Those skilled in the art will understand that all or part of the processes in the above embodiments can be implemented by a computer program instructing related hardware. The computer program can be stored in a non-volatile computer-readable storage medium, and when executed, it can include the processes of the embodiments of the above methods. Any references to memory, databases, or other media used in the embodiments provided in this disclosure can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetic random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. By way of illustration and not limitation, RAM can take many forms, such as Static Random Access Memory (SRAM) or Dynamic Random Access Memory (DRAM). The databases involved in the embodiments provided in this disclosure may include at least one type of relational database and non-relational database. Non-relational databases may include, but are not limited to, blockchain-based distributed databases. The processors involved in the embodiments provided in this disclosure may be general-purpose processors, central processing units, graphics processing units, digital signal processors, programmable logic devices, quantum computing-based data processing logic devices, etc., and are not limited to these.

[0077] It should be noted that the acquisition, transmission, storage, use, and processing of data in the technical solution of this application all comply with relevant laws and regulations. In the embodiments of this application, certain existing industry solutions such as software, components, and models may be mentioned. These should be considered exemplary, intended only to illustrate the feasibility of implementing the technical solution of this application, and do not imply that the applicant has already used or necessarily used such solutions.

[0078] The technical features of the above embodiments can be combined arbitrarily. For the sake of brevity, not all possible combinations of the technical features in the above embodiments are described. However, as long as the combination of these technical features does not contradict each other, it should be considered within the scope of this specification. In the above embodiments, the descriptions of each embodiment have their own emphasis. For parts not described in detail in a certain embodiment, please refer to the relevant descriptions of other embodiments.

[0079] The terms “comprising” and “having”, and any variations thereof, in the embodiments herein are intended to cover non-exclusive inclusion. For example, a process, method, system, product, or apparatus that includes a series of steps or (module) units is not limited to the steps or units listed, but may optionally include steps or units not listed, or may optionally include other steps or units inherent to such process, method, product, or apparatus.

[0080] In this article, "multiple" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent: A alone, A and B simultaneously, or B alone. The character " / " generally indicates that the preceding and following related objects have an "or" relationship.

[0081] The terms "first" and "second" used herein are merely to distinguish similar objects and do not represent a specific ordering of objects. It is understood that "first" and "second" can be interchanged in a specific order or sequence where permissible. It should be understood that the objects distinguished by "first" and "second" can be interchanged where appropriate so that the embodiments described herein can be implemented in an order other than those illustrated or described herein.

[0082] The embodiments described above are merely illustrative of several implementations of this disclosure, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of protection of this disclosure. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of this disclosure, and these all fall within the scope of protection of this disclosure. Therefore, the scope of protection of this disclosure should be determined by the appended claims.

Claims

1. A method for controlling air pressure in a confined space, characterized in that, The method includes: In response to the detection of a target object for opening or closing the enclosed space generating a preset action, the current environmental parameters of the enclosed space are acquired; the current environmental parameters include at least the air conditioning supply air speed parameter, the air conditioning return air pressure parameter, and the movement speed parameter of the target object; The current environmental parameters are input into a pre-trained prediction model to obtain the current optimal pressure relief valve opening area output by the prediction model. The prediction model is obtained by fitting and training a sample dataset. Each sample data pair in the sample dataset consists of the sample environmental parameters of the sealed space and the corresponding optimal pressure relief valve opening area. The optimization constraints of the optimal pressure relief valve opening area include preset ear pressure comfort constraints. Based on the current optimal pressure relief valve opening area, a pressure control signal for the sealed space is generated.

2. The method according to claim 1, characterized in that, The steps for constructing the sample data pairs include: The fluid domain of the enclosed space is modeled to obtain a three-dimensional fluid domain model that includes the air conditioning inlet, the air conditioning return outlet, the pressure relief valve opening, and the target object. The current sample environment parameters are input into the three-dimensional fluid domain model for fluid simulation to determine the pressure change range of the ear region during the movement of the target object and all pressure relief valve opening areas that satisfy the ear pressure comfort constraint, thus obtaining the set of pressure relief valve opening areas corresponding to the current sample environment parameters. Based on the set of pressure relief valve opening areas, the optimal pressure relief valve opening area is determined, and the current sample environment parameters and the optimal pressure relief valve opening area are used to form a sample data pair.

3. The method according to claim 2, characterized in that, The step of determining the optimal pressure relief valve opening area based on the set of pressure relief valve opening areas includes: From the set of pressure relief valve opening areas, select the smallest pressure relief valve opening area as the optimal pressure relief valve opening area.

4. The method according to claim 1, characterized in that, The steps for constructing the prediction model include: The sample dataset is divided into a training set and a test set; The preset initial model is fitted and trained based on the training set to construct a mapping function relationship between the sample environmental parameters and the optimal pressure relief valve opening area; The prediction error of the mapping function relationship is calculated based on the test set, and if the prediction error is less than a preset threshold, the current mapping function relationship is determined as the prediction model.

5. The method according to claim 4, characterized in that, The preset initial model is a surrogate model, which can be any one of the following: multinomial response surface model, Kriging model, radial basis function neural network model, or support vector regression model. The step of fitting and training a preset initial model based on the training set to construct a mapping function relationship between the sample environmental parameters and the optimal pressure relief valve opening area includes: The sample environment parameters are converted into multi-dimensional input feature vectors, and the corresponding optimal pressure relief valve opening area is converted into output labels. The multidimensional input feature vectors in the training set are input into the preset initial model to obtain the predicted opening area; Based on the loss function value between the predicted opening area and the output label, the internal parameters of the preset initial model are iteratively updated; When the loss function value meets the preset convergence condition, the iteration stops, and the preset initial model after training is determined as the mapping function relationship between the multidimensional input feature vector and the output label.

6. The method according to any one of claims 1 to 5, characterized in that, The current environmental parameters also include at least one of the current temperature, current humidity and current air pressure in the enclosed space, and the sample environmental parameters are of the same parameter type as the current environmental parameters; And / or, the ear pressure comfort constraint condition is that the pressure fluctuation amplitude in the ear region during the movement of the target object does not exceed a preset value, wherein the ear region includes a left spherical region with a preset radius centered on the three-dimensional coordinates of the left ear position and a right spherical region with a preset radius centered on the three-dimensional coordinates of the right ear position, and the pressure fluctuation amplitude is the average of the average static pressure fluctuation peak value of the left spherical region during the movement of the target object and the average static pressure fluctuation peak value of the right spherical region during the movement.

7. A closed-space air pressure control device, characterized in that, include: The acquisition module is used to acquire the current environmental parameters of the enclosed space in response to the detection of a target object for opening or closing the enclosed space and the generation of a preset action. The current environmental parameters include at least the air conditioning supply air speed parameters, the air conditioning return air pressure parameters, and the target object's movement speed parameters; The model processing module is used to input the current environmental parameters into the pre-trained prediction model to obtain the current optimal pressure relief valve opening area output by the prediction model. The prediction model is obtained by fitting and training a sample dataset. Each sample data pair in the sample dataset consists of the sample environmental parameters of the sealed space and the corresponding optimal pressure relief valve opening area. The optimization constraints of the optimal pressure relief valve opening area include preset ear pressure comfort constraints. The control generation module is used to generate a pressure control signal for the sealed space based on the current optimal pressure relief valve opening area.

8. A closed-space air pressure control system, characterized in that, The system includes a data acquisition component, a controller, and a pressure relief valve actuator. The controller is connected to both the data acquisition component and the pressure relief valve actuator via signal connections. The data acquisition component is configured to acquire the current environmental parameters of the enclosed space; the current environmental parameters include at least the air conditioning supply air speed parameter, the air conditioning return air pressure parameter, and the movement speed parameter of the target object; the target object is used to open and close the enclosed space. The controller is configured to, in response to detecting the target object and generating a preset action, acquire the current environmental parameters and input the current environmental parameters into a pre-trained prediction model to obtain the current optimal pressure relief valve opening area output by the prediction model; the prediction model is obtained by fitting and training a sample dataset, wherein each sample data pair in the sample dataset consists of the sample environmental parameters of the sealed space and the corresponding optimal pressure relief valve opening area, and the optimization constraints of the optimal pressure relief valve opening area include preset ear pressure comfort constraints; based on the current optimal pressure relief valve opening area, a pressure control signal for the sealed space is generated and sent to the pressure relief valve actuator; The pressure relief valve actuator is configured to adjust the opening area of ​​the pressure relief valve based on the air pressure control signal; the opening of the pressure relief valve is in communication with the sealed space.

9. A controller comprising a memory and a processor, the memory storing a computer program, characterized in that, When the processor executes the computer program, it implements the steps of the method according to any one of claims 1 to 6.

10. An agricultural machine, characterized in that, Includes the system as described in claim 8 or the controller as described in claim 9.

Images