A dynamically adjusted industrial dehumidifier and its control system

Abstract

The present invention relates to the field of dehumidification control technology, and specifically to a dynamically adjusted industrial dehumidifier and its control system. The industrial dehumidifier discloses an industrial dehumidifier body, wherein a dynamic adjustment controller is embedded in the industrial dehumidifier body, and the dynamic adjustment controller includes an environmental parameter prediction module and a dehumidification dynamic adjustment module. The present invention uses LSTM technology to predict various environmental characteristics around the industrial dehumidifier, and combines general finite element software to perform a virtual dehumidification effect simulation on the predicted environment with the current setting parameters of the industrial dehumidifier, accurately determines whether the current setting parameters of the industrial dehumidifier meet the dynamic requirements of the environment and the industrial dehumidifier itself, and after determining that the setting parameters need to be adjusted, combines the general finite element software to deeply analyze the setting parameters that meet the dynamic requirements, and selects appropriate optimization parameters to automatically control the industrial dehumidifier.

Description

Technical Field

[0001] The present invention relates to the technical field of dehumidification control, and more particularly to a dynamically adjusted industrial dehumidifier and a control system thereof. Background Art

[0002] Industrial dehumidifiers, as core equipment for industrial environmental control, are widely used in applications such as semiconductor manufacturing, lithium battery production, food processing, and chemical storage. Their control accuracy and operational efficiency directly impact product quality, equipment lifespan, and energy costs. With increasingly stringent humidity requirements in high-end manufacturing (for example, semiconductor wafer workshops must maintain a humidity of 20% ± 0.5% RH, and lithium battery filling processes require a dew point below -40°C), traditional dehumidification control technologies are struggling to meet the dynamic adjustment requirements under complex operating conditions.

[0003] Existing adjustment solutions generally rely on fixed parameters or manual experience, which are unable to adapt to dynamic changes in ambient humidity loads (such as pulsed humidity sources caused by production line startups and shutdowns, and high humidity air introduced by sudden changes in external weather). Typical issues include: manual adjustment lag: when humidity changes suddenly, manual response delays are long, which can easily lead to product defects caused by moisture; and severe energy waste: even under low-load conditions (such as when there is no production at night), equipment still operates at full power.

[0004] Traditional PID control algorithms are prone to oscillation under nonlinear operating conditions (such as evaporator frosting caused by low temperature and high humidity, and changes in air density at high altitudes), resulting in humidity fluctuations of more than ±2%RH. They are also unable to predict equipment failures (such as filter blockage and refrigerant leakage). Sudden parameter changes often trigger mechanical stress shocks (such as a sudden increase in compressor frequency causing bearing wear), shortening equipment life by more than 20%.

[0005] With the upgrading of industrial intelligence, multi-dimensional environmental data integration, dynamic predictive control, and virtual simulation debugging have become key development directions for dehumidification technology. However, existing solutions have the following technical gaps:

[0006] Lack of in-depth prediction of time-series characteristics such as ambient humidity, temperature, dew point, and air flow, making it impossible to plan control strategies in advance;

[0007] The precise mapping between device operating parameters and the virtual environment is not achieved, making it difficult to verify the validity of parameter combinations before physical debugging.

[0008] The risk of control oscillation and equipment damage caused by large-scale parameter adjustments has not been resolved, which can easily lead to cascading failures, especially in high-precision scenarios.

[0009] Based on the above content, the present invention proposes a dynamically adjusted industrial dehumidifier and its control system. Summary of the Invention

[0010] In view of the deficiencies in the prior art, the present invention aims to provide a dynamically adjusted industrial dehumidifier and a control system thereof.

[0011] To achieve the above object, the present invention provides the following technical solutions:

[0012] A dynamically adjusted industrial dehumidifier comprises an industrial dehumidifier body. A dynamic adjustment controller is embedded in the industrial dehumidifier body. The dynamic adjustment controller comprises an environmental parameter prediction module and a dehumidification dynamic adjustment module.

[0013] Furthermore, a control system for a dynamically adjusted industrial dehumidifier includes an environmental parameter prediction module and a dehumidification dynamic adjustment module;

[0014] An environmental parameter prediction module is used to set a dynamic adjustment period and regularly generate an environmental comprehensive feature prediction set based on the dynamic adjustment period;

[0015] The dehumidification dynamic adjustment module builds an environmental dehumidification model based on the environmental drawings of the industrial dehumidifier and the parameter configuration of the industrial dehumidifier to determine whether to dynamically adjust the industrial dehumidifier.

[0016] Furthermore, an environmental comprehensive feature prediction set is generated regularly based on the dynamic adjustment cycle: whenever a complete dynamic adjustment cycle ends, a real set of environmental comprehensive features is generated, and then a comprehensive time series environmental feature sequence is generated, and the comprehensive time series environmental feature sequence is imported into the comprehensive time series feature prediction model, and the comprehensive time series feature prediction model exports the environmental comprehensive feature prediction set.

[0017] Furthermore, the step of generating the comprehensive time series environmental feature sequence is as follows: collecting the real sets of environmental comprehensive features corresponding to the previous multiple dynamic adjustment periods, and integrating the real sets of environmental comprehensive features into a comprehensive time series environmental feature sequence in a time series manner.

[0018] Furthermore, the steps for generating a true set of comprehensive environmental features corresponding to a dynamic adjustment period are as follows: collecting various types of environmental data within a dynamic adjustment period, performing feature extraction processing on various types of environmental data, obtaining data features of various types of environmental data, and combining the data features of various types of environmental data in a set manner into a true set of comprehensive environmental features.

[0019] Furthermore, determine whether to dynamically adjust the industrial dehumidifier: determine the current setting parameters of the industrial dehumidifier, import the current setting parameters and the environmental comprehensive feature prediction set into the environmental dehumidification model, control the environmental dehumidification model to simulate a dynamic adjustment cycle, and after the simulation, obtain a dynamic adjustment value and set the dynamic adjustment threshold. When the dynamic adjustment value ≥ the dynamic adjustment threshold, generate an optimized control parameter, and adjust the setting parameters of the industrial dehumidifier according to the optimized control parameter. When the dynamic adjustment value < the dynamic adjustment threshold, maintain the current setting parameters of the industrial dehumidifier.

[0020] Further, the steps to obtain the dynamic adjustment value: During the simulation, every T fp Duration, obtain the real-time humidity of the virtual environment in the environmental dehumidification model and the power consumption of the industrial dehumidifier entity, set the humidity reliability range, when the real-time humidity is outside the humidity reliability range, increase the number of humidity changes by one, construct a rectangular coordinate system with time as the X-axis and power consumption as the Y-axis, mark all the obtained power consumptions in the form of coordinate points in the rectangular coordinate system, connect every two adjacent coordinate points in the rectangular coordinate system to obtain multiple power consumption dynamic line segments, obtain the slope of each power consumption dynamic line segment, combine all the power consumption dynamic line segments to form a power consumption dynamic curve, draw two perpendicular line segments from the two end points of the power consumption dynamic curve to the X-axis, and form a closed figure with the power consumption dynamic curve, the two perpendicular line segments and the X-axis, and comprehensively calculate the dynamic adjustment value based on the number of humidity changes, the area of ​​the closed figure and the slope of each power consumption dynamic line segment.

[0021] Furthermore, the steps for generating optimized control parameters are as follows: importing the environmental comprehensive feature prediction set into the environmental dehumidification model, controlling the environmental dehumidification model to perform multiple simulation runs, obtaining the dehumidification setting parameters corresponding to each simulation run, combining the dehumidification setting parameters with the current setting parameters of the industrial dehumidifier into a parameter comparison set, importing the parameter comparison set into the parameter comparison model, and the parameter comparison model deriving a parameter adjustment gap value, and marking the dehumidification setting parameter with the smallest parameter adjustment gap value as the optimized control parameter.

[0022] Compared with the prior art, the present invention has the following beneficial effects:

[0023] The control system of the present invention uses LSTM technology to predict various environmental characteristics around the industrial dehumidifier, and combines it with general finite element software to perform a virtual dehumidification effect simulation on the predicted environment with the current setting parameters of the industrial dehumidifier, and accurately determines whether the current setting parameters of the industrial dehumidifier meet the dynamic requirements of the environment and the industrial dehumidifier itself. After determining that the setting parameters need to be adjusted, the general finite element software is combined to deeply analyze the setting parameters that meet the dynamic requirements, and appropriate optimization parameters are selected to automatically control the industrial dehumidifier, thereby avoiding control oscillation and nonlinear distortion problems caused by large-scale adjustments to the industrial dehumidifier. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 It is a structural diagram of a dynamically adjusted industrial dehumidifier;

[0025] Figure 2 This is a functional block diagram of a control system for a dynamically adjusted industrial dehumidifier.

[0026] In the figure: 100, industrial dehumidifier body; 200, dynamic adjustment controller. DETAILED DESCRIPTION

[0027] Example 1, refer to Figure 1 A dynamically adjusted industrial dehumidifier includes an industrial dehumidifier body 100, on which a dynamic adjustment controller 200 is embedded. The dynamic adjustment controller 200 includes an environmental parameter prediction module and a dehumidification dynamic adjustment module.

[0028] Example 2, refer to Figure 2 , a control system for a dynamically adjusted industrial dehumidifier, including an environmental parameter prediction module and a dehumidification dynamic adjustment module.

[0029] The environmental parameter prediction module sets a dynamic adjustment cycle (the duration of the dynamic adjustment cycle is set according to the needs of the industrial scenario) and regularly generates an environmental comprehensive feature prediction set based on the dynamic adjustment cycle (the environmental comprehensive feature prediction set contains data features of multiple types of environmental data. These types of environmental data jointly determine whether the industrial dehumidifier needs to be adjusted. The types of environmental data include ambient humidity, ambient temperature, ambient dew point temperature, and air flow data. Each type of data contains corresponding data features. For example, ambient humidity includes data features such as humidity change rate, humidity mean, and humidity extreme value; ambient temperature includes data features such as temperature change rate, temperature mean, and temperature extreme value; ambient dew point temperature includes data features such as dew point change rate, dew point stability, and dew point extreme value; air flow data includes data features such as wind speed change rate and air volume peak / valley value).

[0030] An environmental comprehensive feature prediction set is generated regularly based on the dynamic adjustment cycle: whenever a complete dynamic adjustment cycle ends, a real set of environmental comprehensive features is generated, and then a comprehensive time series environmental feature sequence is generated. The comprehensive time series environmental feature sequence is imported into the comprehensive time series feature prediction model, and the comprehensive time series feature prediction model exports the environmental comprehensive feature prediction set.

[0031] The steps for generating the comprehensive time series environment feature sequence are as follows: collecting the real sets of comprehensive environment features corresponding to the previous multiple dynamic adjustment periods, and integrating the real sets of comprehensive environment features into a comprehensive time series environment feature sequence in the form of time series.

[0032] The steps for generating a true set of comprehensive environmental features corresponding to a dynamic adjustment period are as follows: collecting various types of environmental data within a dynamic adjustment period, performing feature extraction processing on various types of environmental data, obtaining data features of various types of environmental data, and combining the data features of various types of environmental data in a set manner into a true set of comprehensive environmental features.

[0033] The comprehensive time series feature prediction model is built based on the LSTM model. The steps to build the comprehensive time series feature prediction model are as follows: build the LSTM model, collect multiple comprehensive time series environment feature sequences, and the comprehensive time series environment feature sequence is a real set of comprehensive environmental features of multiple cycles (such as the first 12 cycles) spliced ​​in chronological order to form a time series sequence of length T. , where each St is the feature set of the t-th cycle. Using multiple comprehensive time series environmental feature sequences as basic data, the built LSTM model is trained. In this process, each comprehensive time series environmental feature sequence is given an environmental comprehensive feature prediction set (the comprehensive time series environmental feature sequence is converted into a supervised learning sample, that is, the input is the sequence of the past N cycles, and the output result is used to predict the next cycle. All features (such as humidity mean, temperature change rate, etc.) are normalized or standardized to eliminate the dimension effect, and the standardized parameters (such as mean, standard deviation) are saved so that the input data can be processed in the same way during prediction, and the output result is denormalized to restore the true value). The environmental comprehensive feature prediction set includes the corresponding data features of environmental humidity, ambient temperature, ambient dew point temperature, and air flow data. For example, ambient humidity corresponds to humidity change rate, humidity mean, and humidity extreme value. The ambient temperature includes temperature change rate, temperature mean, and temperature extreme value. The ambient dew point temperature includes dew point change rate, dew point stability, and dew point extreme value. The air flow data includes wind speed change rate and wind volume peak / valley value. The example of environmental comprehensive feature prediction set is python{

[0034] "humidity": {"rate": 0.5, "mean": 65, "max": 70, "min": 60},

[0035] "temperature": {"rate": -1.2, "mean": 22, "max": 25, "min": 18},

[0036] "dew_point": {"rate": -0.8, "stability": 0.9, "min": 12},

[0037] "air_flow": {"wind_rate": 0.3, "peak": 5.0, "valley": 2.0}}, the environmental comprehensive feature prediction set contains the data features of various types of environmental data in the next dynamic adjustment cycle. Then, multiple comprehensive time series environmental feature sequences are divided into training set, validation set and test set according to a specific ratio. The specific division ratio is determined to be 60%:20%:20%. First, the LSTM model is repeatedly trained using the training set. The training strategy is as follows: Batch training: Set the batch size (Batch Size) to 32 or 64 to reduce memory usage and improve training stability; Number of training rounds (Epochs): Initially set to 200 rounds, combined with the early stopping method (Early Stopping) to avoid overfitting; Hyperparameter tuning: Use grid search (GridSearch) or random search (Random Search) optimizes the following parameters: number of LSTM layers (1 or 2); number of neurons in each layer (32 / 64 / 128); dropout rate (to prevent overfitting, such as 0.1 / 0.2); learning rate (such as 1e-3 / 1e-4). Training process monitoring: plot loss curves for the training set and validation set to observe convergence and signs of overfitting; record evaluation indicators (such as MAE, RMSE) and compare prediction accuracy for different periods. Generalization ability verification: conduct offline testing in industrial scenarios: use newly collected periodic data to generate a sequence, input the model to predict the characteristics of the next period, and compare the evaluation error with the actual value; if the error exceeds the industrial allowable range (such as humidity prediction error >5%), it is necessary to re-check data quality, feature engineering or adjust the model structure, build a comprehensive time series feature prediction model, save the trained model as a .h5 or .pb file, load the model and initialize the normalizer during deployment, and preprocess and predict real-time input data.

[0038] The dehumidification dynamic adjustment module builds an environmental dehumidification model based on the environmental drawings of the industrial dehumidifier and the parameter configuration of the industrial dehumidifier, determines the current setting parameters of the industrial dehumidifier, and imports the current setting parameters and the environmental comprehensive feature prediction set into the environmental dehumidification model (the various types of environmental data in the virtual environment in the environmental dehumidification model will conform to the environmental comprehensive feature prediction set, and the industrial dehumidifier entity in the environmental dehumidification model can adjust the virtual environment according to the current setting parameters). The environmental dehumidification model is controlled to simulate a dynamic adjustment cycle. After the simulation is completed, a dynamic adjustment value is obtained (the dynamic adjustment value is a value generated after analyzing the simulation process of the entire environmental dehumidification model). A dynamic adjustment threshold is set. When the dynamic adjustment value ≥ the dynamic adjustment threshold, an optimized control parameter is generated. The setting parameters of the industrial dehumidifier are adjusted according to the optimized control parameter. When the dynamic adjustment value < the dynamic adjustment threshold, the current setting parameters of the industrial dehumidifier are maintained.

[0039] Steps to obtain dynamic adjustment value: During the simulation, every T fp duration, obtain the real-time humidity of the virtual environment in the environmental dehumidification model and the power consumption of the industrial dehumidifier entity, set the humidity reliability range (the humidity reliability range is the humidity range pre-set during the system design or deployment to ensure the normal operation of the equipment or process stability). When the real-time humidity is outside the humidity reliability range, increase the number of humidity changes by one, and mark the number of humidity changes as Ls (rk). Construct a rectangular coordinate system with time as the X-axis and power consumption as the Y-axis. Mark all the obtained power consumptions in the form of coordinate points in the rectangular coordinate system. Connect every two adjacent coordinate points in the rectangular coordinate system to obtain multiple power consumption dynamic line segments, obtain the slope of each power consumption dynamic line segment, sum and average the slopes of all power consumption dynamic line segments, calculate the average slope, and mark it as , a power consumption dynamic curve is formed by combining all the power consumption dynamic line segments. Two vertical segments are drawn from the two end points of the power consumption dynamic curve to the X-axis. The area of ​​the closed figure formed by the power consumption dynamic curve, the two vertical segments and the X-axis is marked as Vb (ar). Calculate the dynamic adjustment value The dynamic adjustment value is based on dimensionless calculation, where dg1 is the first adjustment coefficient, dg2 is the second adjustment coefficient, the value of the first adjustment coefficient is 1.67, and the value of the second adjustment coefficient is 0.98.

[0040] The steps for building the environmental dehumidification model are as follows: Based on the drawings of the environment in which the industrial dehumidifier is located, a virtual environment is created in general finite element software based on the drawings of the environment. The drawings of the environment in which the industrial dehumidifier is located are preprocessed to extract key geometric features (walls, doors and windows, equipment layout), and non-critical details (such as screws and small-sized pipelines) are simplified; material properties (wall thermal conductivity, ground moisture absorption rate) are annotated, and the material library of the general finite element software is imported to determine the location of the industrial dehumidifier. Based on the location, an industrial dehumidifier entity is created in the virtual environment. Various parameter configurations are assigned to the industrial dehumidifier entity according to the parameter configuration of the industrial dehumidifier. The industrial dehumidifier entity is refined, such as modeling the internal flow path of the dehumidifier (evaporator, condenser, fan impeller), using a porous media model to simulate the filter resistance, defining the phase change process (mass transfer coefficient of water vapor condensation on the evaporator surface), locally refining the mesh (size <2mm) in key areas such as the evaporator and air outlet, and using a coarse mesh (size 50mm) in non-critical areas to balance accuracy and computational efficiency; using mesh quality checking tools (such as ANSYS The skewness of meshing was less than 0.9. Three typical operating conditions (low, medium, and high humidity loads) were selected for actual measurement. The simulation results were compared and the evaporator heat transfer coefficient was corrected (for example, if the simulated value was 10% higher than the measured value, the coefficient was multiplied by 0.9). Error analysis tools (such as Matlab's corrcoef function) were used to calculate the correlation between the simulated and measured data (needed to be greater than 0.95). Finally, an environmental dehumidification model was constructed.

[0041] The steps for generating the optimized control parameters are as follows: import the environmental comprehensive feature prediction set into the environmental dehumidification model, control the environmental dehumidification model to perform multiple simulation runs, and obtain the dehumidification setting parameters corresponding to each simulation run (before each simulation run, the environmental dehumidification model will generate a dehumidification setting parameter, and the dehumidification setting parameter corresponding to each simulation run is different. A unique dehumidification setting parameter (such as target humidity, dehumidification intensity or operation time) is dynamically generated before each simulation run. The environmental comprehensive feature prediction set usually contains variable data (such as temperature and humidity at different time points), which causes the model to generate different dehumidification setting parameters before each run. Dehumidification setting parameters (for example, parameter A for high humidity scenarios, parameter B for low temperature scenarios). This helps simulate various operating conditions and verify the robustness of the model. After the run, dehumidification setting parameters with a "dynamic adjustment value ≥ dynamic adjustment threshold" are not counted, and only those with a "dynamic adjustment value < dynamic adjustment threshold" are counted. The dehumidification setting parameters are combined with the current setting parameters of the industrial dehumidifier into a parameter comparison set. The parameter comparison set is imported into the parameter comparison model, which derives a parameter adjustment gap value. The dehumidification setting parameter with the smallest parameter adjustment gap value is marked as the optimized control parameter.

[0042] The process of building a parameter comparison model: building a deep learning model, the model selects a fully connected neural network,

[0043] pythonfrom tensorflow.keras.models import Sequentialfrom

[0044] tensorflow.keras.layers import Dense, Dropout

[0045] model = Sequential([

[0046] Dense(64, activation='relu', input_shape=(F,)), # input layer

[0047] Dropout(0.2), # prevent overfitting

[0048] Dense(32, activation='relu'),

[0049] Dense(1, activation='linear') # output layer]), collect multiple historical parameter comparison sets, use the historical parameter comparison sets as basic data, and train the built deep learning model. During this training process, each historical parameter comparison set is decomposed into multiple normalized deviation values ​​of historical parameters, and each historical parameter comparison set is assigned a real parameter adjustment gap value as a label. The calculation method of the real parameter adjustment gap value is: , where i = 1, 2, ..., I, i represents the corresponding historical parameter, I represents the total number of historical parameters (historical parameters include but are not limited to humidity and speed), is the weight of the i-th historical parameter. For example, the humidity weight is higher than the speed. P1 to PI represent the normalized deviation value of each historical parameter in turn. When training the model, the model is trained based on the predicted parameter adjustment gap value output by the output layer and the actual parameter adjustment gap value to obtain a trained parameter comparison model. In this embodiment, the normalized deviation value of humidity and speed will be illustrated as an example. The normalized deviation value of humidity: For example, the minimum setting value of the humidity parameter of an industrial dehumidifier is 30%RH and the maximum setting value is 80%RH. The minimum difference of the humidity difference is , the maximum difference If the humidity parameter set in the dehumidification setting parameter is 60%RH and the humidity parameter set in the current setting parameter of the industrial dehumidifier is 50%RH, then the humidity difference between the dehumidification setting parameter and the current setting parameter of the industrial dehumidifier is , then the normalized deviation value of humidity Normalized deviation value of speed: If the minimum setting value of the speed parameter of the industrial dehumidifier is 500rpm and the maximum setting value is 2000rpm, the minimum value of the speed difference is Maximum difference , if the speed parameter set in the dehumidification setting parameter is 1500rpm, and the speed parameter set in the current setting parameter of the industrial dehumidifier is 1200rpm, then the speed difference is 1500-1200=300rpm, then the normalized deviation value of the speed is The parameter adjustment gap value range is set between -5 and 5. The size of the parameter adjustment gap value has a clear meaning. The larger the value, the greater the gap between the dehumidification setting parameters in the corresponding parameter comparison set and the current setting parameters of the industrial dehumidifier. That is, if the current setting parameters of the industrial dehumidifier are adjusted to the dehumidification setting parameters, the greater the adjustment range required.

[0050] The control system of the present invention uses LSTM technology to predict various environmental characteristics around the industrial dehumidifier, and combines it with general finite element software to perform a virtual dehumidification effect simulation on the predicted environment with the current setting parameters of the industrial dehumidifier, and accurately determines whether the current setting parameters of the industrial dehumidifier meet the dynamic requirements of the environment and the industrial dehumidifier itself. After determining that the setting parameters need to be adjusted, the general finite element software is combined to deeply analyze the setting parameters that meet the dynamic requirements, and appropriate optimization parameters are selected to automatically control the industrial dehumidifier, thereby avoiding control oscillation and nonlinear distortion problems caused by large-scale adjustments to the industrial dehumidifier.

[0051] The above formulas are all dimensionless and numerically calculated, and the preset parameters in the formulas are set by technicians in this field according to actual conditions.

[0052] The above embodiments can be implemented in whole or in part by software, hardware, firmware or any other combination. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer program are loaded or executed on a computer, the process or function described in the embodiment of the present application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from one website, computer, server or data center to another website, computer, server or data center via wired or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more available media sets. The available medium can be a magnetic medium (e.g., a floppy disk, a hard disk, a tape), an optical medium (e.g., a DVD), or a semiconductor medium. The semiconductor medium can be a solid-state drive.

[0053] It should be understood that in the various embodiments of the present application, the size of the serial numbers of the above-mentioned processes does not mean the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present application.

[0054] Those skilled in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0055] Those skilled in the art will clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0056] In the several embodiments provided in this application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are merely schematic. For example, the division of the units is merely a logical function division. In actual implementation, there may be other division methods, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0057] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium and includes several instructions for enabling a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the method described in each embodiment of the present application. The aforementioned storage medium includes various media that can store program codes, such as a USB flash drive, a mobile hard disk, a read-only memory (ROM), a random access memory (RAM), a magnetic disk or an optical disk.

[0058] The above description is merely a specific embodiment of the present application, but the scope of protection of the present application is not limited thereto. Any changes or substitutions that can be easily conceived by a person skilled in the art within the technical scope disclosed in this application should be included in the scope of protection of this application. Therefore, the scope of protection of this application should be based on the scope of protection of the claims.

Claims

1. A control system for a dynamically adjusted industrial dehumidifier, characterized in that: Including environmental parameter prediction module and dehumidification dynamic adjustment module; An environmental parameter prediction module is used to set a dynamic adjustment period and regularly generate an environmental comprehensive feature prediction set based on the dynamic adjustment period; A dehumidification dynamic adjustment module builds an environmental dehumidification model based on the environmental drawings of the industrial dehumidifier and the parameter configuration of the industrial dehumidifier, and determines whether to dynamically adjust the industrial dehumidifier based on the environmental comprehensive feature prediction set and the environmental dehumidification model; Determine whether to dynamically adjust the industrial dehumidifier: determine the current setting parameters of the industrial dehumidifier, import the current setting parameters and the environmental comprehensive feature prediction set into the environmental dehumidification model, and control the environmental dehumidification model to simulate a dynamic adjustment cycle. During the simulation, every T fp Duration, obtain the real-time humidity of the virtual environment in the environmental dehumidification model and the power consumption of the industrial dehumidifier entity. After the simulation is completed, obtain a dynamic adjustment value and set a dynamic adjustment threshold. When the dynamic adjustment value ≥ the dynamic adjustment threshold, generate an optimized control parameter and adjust the setting parameters of the industrial dehumidifier according to the optimized control parameter. When the dynamic adjustment value < the dynamic adjustment threshold, maintain the current setting parameters of the industrial dehumidifier. The steps for obtaining the dynamic adjustment value are as follows: setting a humidity reliability range; when the real-time humidity is outside the humidity reliability range, increasing the number of humidity changes by one; constructing a rectangular coordinate system with time as the X-axis and power consumption as the Y-axis; marking all the obtained power consumptions in the form of coordinate points in the rectangular coordinate system; connecting every two adjacent coordinate points in the rectangular coordinate system to obtain multiple power consumption dynamic line segments; obtaining the slope of each power consumption dynamic line segment; combining all the power consumption dynamic line segments to form a power consumption dynamic curve; drawing two perpendicular line segments from the two end points of the power consumption dynamic curve to the X-axis; forming a closed figure by the power consumption dynamic curve, the two perpendicular line segments, and the X-axis; and comprehensively calculating the dynamic adjustment value based on the number of humidity changes, the area of ​​the closed figure, and the slope of each power consumption dynamic line segment; The steps for generating optimized control parameters are as follows: importing the environmental comprehensive feature prediction set into the environmental dehumidification model, controlling the environmental dehumidification model to perform multiple simulation runs, obtaining the dehumidification setting parameters corresponding to each simulation run, combining the dehumidification setting parameters with the current setting parameters of the industrial dehumidifier into a parameter comparison set, importing the parameter comparison set into the parameter comparison model, and deriving a parameter adjustment gap value from the parameter comparison model. The dehumidification setting parameter with the smallest parameter adjustment gap value is marked as the optimized control parameter.

2. The control system of a dynamically adjusted industrial dehumidifier according to claim 1, characterized in that: An environmental comprehensive feature prediction set is generated regularly based on the dynamic adjustment cycle: whenever a complete dynamic adjustment cycle ends, a real set of environmental comprehensive features is generated, and then a comprehensive time series environmental feature sequence is generated. The comprehensive time series environmental feature sequence is imported into the comprehensive time series feature prediction model, and the comprehensive time series feature prediction model exports the environmental comprehensive feature prediction set.

3. The control system of a dynamically adjusted industrial dehumidifier according to claim 2, characterized in that: The steps for generating the comprehensive time series environment feature sequence are as follows: collecting the real sets of comprehensive environment features corresponding to the previous multiple dynamic adjustment periods, and integrating the real sets of comprehensive environment features into a comprehensive time series environment feature sequence in the form of time series.

4. The control system of a dynamically adjusted industrial dehumidifier according to claim 3, characterized in that: The steps for generating a true set of comprehensive environmental features corresponding to a dynamic adjustment period are as follows: collecting various types of environmental data within a dynamic adjustment period, performing feature extraction processing on various types of environmental data, obtaining data features of various types of environmental data, and combining the data features of various types of environmental data in a set manner into a true set of comprehensive environmental features.

5. A dynamically adjusted industrial dehumidifier, characterized in that: The invention comprises an industrial dehumidifier body (100), wherein a dynamic adjustment controller (200) is embedded in the industrial dehumidifier body (100), and the dynamic adjustment controller (200) comprises a control system of a dynamically adjusted industrial dehumidifier according to any one of claims 1 to 4.

Images