An environmental control method, system, equipment, and medium for a mobile high-voltage chamber.

Abstract

This invention discloses an environmental control method, system, equipment, and medium for a mobile high-voltage chamber, belonging to the field of high-voltage chamber environmental control technology. The method includes the following steps: dividing the mobile high-voltage chamber into A analysis zones, and establishing a temperature control intelligent agent in each analysis zone; collecting equipment status data of the analysis zones through the temperature control intelligent agents, reconstructing and analyzing the equipment status data using an autoencoder to obtain status tags; combining the temperature control intelligent agents with the status tags, and calculating a health status matrix by analyzing the equipment status data; collecting environmental data of each analysis zone in real time, combining the health status matrix to form a microclimate situation map, and generating optimal zone temperature control commands; adjusting the environment of each analysis zone through the optimal zone temperature control commands. The beneficial effect of this invention is that by dividing the mobile high-voltage chamber into multiple analysis zones and establishing a temperature control intelligent agent in each zone, a fundamental shift from traditional centralized control to distributed collaborative control is achieved.

Description

Technical Field

[0001] This invention relates to the field of high-pressure chamber environmental control technology, specifically to an environmental control method, system, equipment, and medium for a mobile high-pressure chamber. Background Technology

[0002] A mobile high-voltage room is a type of high-voltage electrical equipment that modularly integrates the primary and secondary equipment of a 10kV substation onto a movable base. It includes incoming line cabinets, busbar equipment, feeder cabinets, station service transformers, and typically also contains relay protection and automation devices, AC / DC power supply systems, and five-proof systems necessary for the complete functionality of medium-voltage switchgear. With the rapid expansion of modern urban power distribution systems, mobile high-voltage rooms play an increasingly important role in uninterrupted power supply construction, emergency power supply guarantees, and rapid power deployment tasks in cities.

[0003] Traditional high-pressure chambers rely heavily on centralized control and fixed-parameter strategies for environmental regulation. However, with the increasing complexity of high-pressure chamber structures, the denser deployment of equipment, and the dynamic overlap of multiple heat sources within the chamber, existing technologies face the following challenges: Single-sensor control struggles to address the non-uniformity and time-varying nature of the regional environment, particularly under conditions of frequent load fluctuations and varying heat source intensities in electrical equipment. Traditional temperature and humidity control systems often exhibit response lag, over-regulation, or ineffective coverage of regional hotspots. Equipment anomaly detection still primarily relies on threshold logic or manual inspection, lacking intelligent assessment methods for high-dimensional, continuous, and multimodal data, thus hindering early anomaly identification and warning. Summary of the Invention

[0004] In view of the above-mentioned problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by this invention is: how to realize zoned intelligent environmental control of mobile high-voltage rooms, and solve the problems of insufficient accuracy and limited equipment anomaly detection capability of traditional centralized control.

[0006] To solve the above-mentioned technical problems, the present invention provides the following technical solution: an environmental control method for a mobile high-pressure chamber, comprising the following steps: dividing the mobile high-pressure chamber into A analysis areas, and establishing a temperature control agent in each analysis area; collecting equipment status data of the analysis areas through the temperature control agent, reconstructing and analyzing the equipment status data using an autoencoder to obtain status labels; combining the temperature control agent with the status labels, and obtaining a health status matrix by calculating the equipment status data; collecting environmental data of each analysis area in real time, forming a microclimate situation map by combining the health status matrix, and generating optimal zone temperature control commands using a fuzzy PID controller and a multi-agent game mechanism; and adjusting the environment of each analysis area through the optimal zone temperature control commands.

[0007] As a preferred embodiment of the environmental control method for a mobile high-voltage chamber according to the present invention, the step of reconstructing and analyzing the equipment status data to obtain status labels includes: constructing a multimodal feature vector through the equipment status data; performing Gaussian filtering on the multimodal feature vector to form data subsets at multiple scales; constructing a corresponding encoder for each data subset at each scale, and training and optimizing the encoder using a multi-scale joint reconstruction loss function; inputting the current equipment status data into the trained encoder to obtain an equipment status encoding vector; calculating the distance between the equipment status encoding vector and the center of historical normal samples to generate an anomaly score; and determining the status label based on a preset threshold and the anomaly score; the status label includes a health label and an anomaly label.

[0008] As a preferred embodiment of the environmental control method for a mobile high-pressure chamber according to the present invention, the step of obtaining a health status matrix by calculating equipment status data includes: calculating the heat source output power of the equipment based on the equipment status data; obtaining the total regional heat source output power by analyzing the heat source output power of the equipment corresponding to all health tags; calculating the load risk index of the equipment corresponding to each health tag, and aggregating the obtained load risk indices to obtain the regional average load risk; combining the regional total heat source output power and the regional average load risk to form a regional state vector; and integrating the regional state vectors of all analyzed regions to generate a global health status matrix.

[0009] The beneficial effects of this preferred technical solution are: the use of differentiated temperature control strategies for healthy and abnormal equipment can avoid cascading failures caused by overheating of abnormal equipment.

[0010] As a preferred embodiment of the environmental control method for a mobile high-pressure chamber according to the present invention, the step of generating the optimal zone temperature control command includes: real-time acquisition of environmental data of each analysis area, fusion of the environmental data with the health status matrix to form a complete regional state vector; drawing a visualization heatmap based on the complete regional state vector to form a microclimate situation map; each temperature control agent constructing a multi-factor benefit function based on the microclimate situation map; initiating a collaborative game mechanism, with each temperature control agent sending the multi-factor benefit function to the main coordination node for global optimal strategy solution; and initiating a fuzzy PID controller to calculate a continuous control signal based on the game result, and superimposing the continuous control signal with the game result to form the optimal zone temperature control command.

[0011] The beneficial effects of this preferred technical solution are: it solves the control conflict problem of adjacent areas "competing" for cold source, reduces overall energy consumption compared with independent control, and minimizes temperature fluctuations in each area.

[0012] As a preferred embodiment of the environmental control method for a mobile high-pressure chamber according to the present invention, the step of adjusting the environment of each analysis zone through the optimal zone temperature control command includes: splitting the optimal zone temperature control command into an air supply control target, a cold / heat source control target, and a valve adjustment target; the air supply control subsystem adjusts the fan speed and controls the air volume according to the air supply control target; the cold / heat source control subsystem controls the compressor to cool or the heating element to heat according to the cold / heat source control target; and the valve adjustment subsystem performs opening changes according to the valve adjustment target to achieve directional air supply.

[0013] As a preferred embodiment of the environmental control method for a mobile high-pressure chamber according to the present invention, the step of dividing the mobile high-pressure chamber into A analysis regions includes: collecting the cabin structure data of the mobile high-pressure chamber to obtain point cloud data; performing a spatial clustering algorithm on the point cloud data to automatically divide the spatial regions according to the equipment distribution density; outputting M spatial regions through the clustering algorithm and assigning a unique number to each spatial region; allocating the equipment to the corresponding spatial region according to the spatial coordinates of the point cloud data and recording the binding relationship.

[0014] The advantages of this preferred technical solution are: it automatically adapts to different cabin layouts, eliminates the need for manual pre-setting of partitions, and provides finer temperature control granularity in densely populated equipment areas.

[0015] As a preferred embodiment of the environmental control method for a mobile high-pressure chamber according to the present invention, the steps for forming the microclimate situation map include: dividing the physical location coordinates of the mobile high-pressure chamber by analyzing the region; drawing a visual heatmap of the environmental data; overlaying and annotating the environmental data in the health status matrix as a situation map to form multiple layers; the multiple layers include a temperature and humidity distribution layer, a pressure difference distribution layer, a heat source load distribution layer, and a health status identification layer; merging the multiple layers to form the microclimate situation map.

[0016] This invention provides an environmental control system for a mobile high-pressure chamber.

[0017] To solve the above-mentioned technical problems, the present invention provides the following technical solution: an environmental control system for a mobile high-voltage chamber, comprising:

[0018] The data acquisition module is used to collect data on the cabin structure, equipment status, and environment of the mobile high-pressure chamber.

[0019] The intelligent analysis module is used to reconstruct and analyze the equipment status data to obtain status labels, calculate the equipment heat source output power and load risk index, and generate a health status matrix.

[0020] The situation map generation module is used to integrate environmental data with the health status matrix to form a microclimate situation map;

[0021] The collaborative decision-making module is used to generate optimal zone temperature control commands;

[0022] The environmental control module is used to execute optimal zone temperature control commands to adjust the environment of each analysis zone.

[0023] The present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, characterized in that the processor executes the computer program to implement the steps of the environmental control method for a mobile high-voltage chamber.

[0024] The present invention provides a computer-readable storage medium having a computer program stored thereon, characterized in that the computer program, when executed by a processor, implements the steps of the environmental control method for a mobile high-voltage chamber.

[0025] The beneficial effects of this invention are:

[0026] By dividing the mobile high-voltage chamber into multiple analysis zones and establishing a temperature-control agent in each zone, a fundamental shift from traditional centralized control to distributed collaborative control has been achieved. Each temperature-control agent can not only independently collect and process equipment status data within its zone, but more importantly, it can coordinate globally through a collaborative game mechanism, avoiding the decision-making limitations of a single controller facing complex and ever-changing environments. This multi-agent collaborative working mode produces an overall effect that transcends single-point control, making the environmental regulation of the entire mobile high-voltage chamber more efficient.

[0027] The technical approach of using autoencoders to reconstruct and analyze equipment status data and obtain status labels overcomes the limitations of traditional threshold logic detection. Encoders trained using multi-scale joint reconstruction loss functions can capture deep feature patterns in equipment operation, rather than simple numerical comparisons. When these intelligently identified status labels are combined with a temperature control agent to form a health status matrix, an unexpected synergistic effect is generated: environmental control strategies no longer passively respond to temperature changes, but actively predict and prevent heat source fluctuations under abnormal equipment conditions, achieving a deep integration of equipment health management and environmental regulation. Attached Figure Description

[0028] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0029] Figure 1 The above is a general flowchart of an environmental control method for a mobile high-pressure chamber provided in one embodiment of the present invention. Detailed Implementation

[0030] To make the above-mentioned objects, features, and advantages of the present invention more apparent and understandable, specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not all of them. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the protection scope of the present invention.

[0031] Example 1, referring to Figure 1 This is one embodiment of the present invention, which provides an environmental control method for a mobile high-voltage chamber, comprising the following steps S1 to S5:

[0032] S1. Divide the mobile high-pressure chamber into A analysis areas, and set up a temperature control intelligent agent in each analysis area.

[0033] The steps for dividing the moving high-pressure chamber into A analytical regions include S1.1 to S1.4:

[0034] S1.1 Collect structural data of the mobile high-pressure chamber to obtain point cloud data;

[0035] Using 3D laser scanning equipment or a BIM data interface, the structural information of the mobile high-voltage room is acquired in real time to obtain three-dimensional point cloud data. In this embodiment, the point cloud data includes the geometric coordinates of the internal space of the room, equipment location information, and spatial topology. The point cloud data obtained by laser scanning can achieve millimeter-level accuracy, accurately reflecting the spatial distribution of equipment such as incoming line cabinets, busbar equipment, feeder cabinets, and station transformers within the room.

[0036] S1.2 Execute a spatial clustering algorithm on the point cloud data to automatically divide the spatial regions according to the device distribution density;

[0037] The acquired point cloud data is processed using the density-based spatial clustering algorithm DBSCAN (Density-Based Spatial Clustering of Applications with Noise), with minimum cluster radius and minimum number of points set as clustering parameters. The algorithm automatically identifies densely populated areas of equipment as cluster centers based on equipment distribution density, grouping equipment points with similar spatial densities into the same cluster. By adjusting the clustering parameters, the system can adapt to different cabin layouts and equipment configurations, identifying spatial patterns in equipment distribution.

[0038] S1.3. Output M spatial regions using a clustering algorithm and assign a unique number to each spatial region;

[0039] After the DBSCAN clustering algorithm completes, it outputs M spatial regions (where M ≤ L, and L is the preset maximum number of regions). Each identified spatial region is assigned a unique region ID, with the ID format as follows: (Where j is the region index). Each region ID is associated with and stored along with its corresponding spatial coordinate range, the types of devices it contains, and their quantities.

[0040] S1.4. Assign the device to the corresponding spatial region according to the spatial coordinates of the point cloud data and record the binding relationship;

[0041] Based on the spatial coordinate information in the point cloud data, various devices within the cabin (including incoming line cabinets, busbar equipment, feeder cabinets, station transformers, relay protection devices, AC / DC power supply systems, etc.) are automatically assigned to their corresponding spatial areas. Simultaneously, sensor devices (wireless temperature sensors, humidity sensors, differential pressure sensors, current sensors, PM2.5 concentration sensors) and actuator devices (air valves, fans, air conditioners, humidifiers, purification modules) are physically bound according to their spatial coordinate relationships, and a detailed binding relationship table is recorded.

[0042] After completing the region division, the system automatically creates a virtual temperature control agent for each bound region ID, with a unique identifier. (Where j is the region index). Each All sensors and actuators within the region are bound together to form an independent perception-decision-execution closed loop. Initialization parameters, physical device binding relationships, and spatial distribution information for all temperature control agents are automatically uploaded to the distributed edge control network. Data backups are performed on the main control platform and on the local controller of each temperature control agent to ensure system reliability and fault tolerance.

[0043] S2. The device status data of the area is collected and analyzed by the temperature control intelligent agent, and the device status data is reconstructed and analyzed using an autoencoder to obtain status tags. The status tags include health tags and abnormal tags.

[0044] Specifically, each temperature control intelligent entity Real-time acquisition of equipment status data within the area, including multi-dimensional information such as equipment temperature, current, vibration amplitude, and equipment status codes, while simultaneously acquiring environmental data such as temperature, humidity, pressure difference, and PM2.5 concentration. The steps for reconstructing and analyzing the equipment status data to obtain status labels include S2.1~S2.4:

[0045] S2.1 Construct a multimodal feature vector using the device status data, and perform Gaussian filtering on the multimodal feature vector to form data subsets of multiple scales;

[0046] A multimodal feature vector is constructed based on the collected device status data, including current. ,temperature Vibration amplitude Pressure difference Equal physical quantities form characteristic vectors One-dimensional Gaussian filtering and normalization are performed on each dimension of the multimodal feature vector. The filter parameters... Adjustments are made based on signal characteristics. From the smoothed time series, a downsampling strategy is employed, sampling once at every other point to form a new time scale. By using different sampling intervals (e.g., 1×, 0.5×, 0.25×), feature subsets at multiple time scales are obtained. , where s represents the scale index.

[0047] S2.2 Construct a corresponding encoder for each scale subset of data, and train and optimize the encoder using a multi-scale joint reconstruction loss function;

[0048] Each temperature control agent loads multi-scale cached data within a specified time window (e.g., the past 24 hours), constructs an independent autoencoder (AE) subnetwork for each time scale, and the encoder... and decoder Composition. The reconstruction process is represented as:

[0049] ;

[0050] in, It refers to the multi-timescale features of the input. It is the reconstructed output.

[0051] The reconstruction error is calculated separately for each scale using the mean squared error (MSE) loss, and then a weighted average is applied across all scales to generate a multi-scale joint reconstruction loss function:

[0052] ;

[0053] in, The value represents the multi-scale joint reconstruction loss function, where S is the total number of scales and n is the total number of training samples. Let L2 norm reconstruction error be the L2 norm of the i-th sample at the s-th scale. This is a normalization factor used to average the loss across all samples and scales.

[0054] The Adam optimizer is used for minimization iteration, and an EarlyStopping strategy and error trend monitoring mechanism are introduced to save the encoder weights as the optimal model when the error reaches the historical optimum.

[0055] S2.3 Input the current device state data into the trained encoder to obtain the device state encoding vector;

[0056] By inputting the sampled data at the current time t into the optimal encoder corresponding to each scale, we can obtain the feature embedding representation of the current Agent at different scales:

[0057] ;

[0058] in, It is an embedded feature vector. Let be the trained encoder network corresponding to the s-th scale.

[0059] To further enhance anomaly detection capabilities, a multi-scale grouped Siamese structure is employed for contrastive learning. Embedded vectors from historical normal operating periods are used as reference healthy samples, forming positive sample pairs with the currently observed embeddings. Embedded vectors of the current healthy state of the anomalous device are paired with historical anomalous embeddings to form negative sample pairs. The Euclidean distance between the embedded vectors is calculated using a shared encoder, and a modified contrastive loss function is used for constrained training.

[0060] S2.4 Calculate the distance between the device status encoding vector and the center of historical normal samples, generate an anomaly score, and determine the status label based on a preset threshold and the anomaly score.

[0061] For each Agent, the multi-scale features collected at the current time are input into the encoder at the corresponding scale to obtain the current embedded features. The embedded features of all scales are combined to generate the current observation embedded representation set. The Euclidean distance S between the current observation embedded vector and the center of the historical normal sample at each scale is calculated. The spatial distances at all scales are weighted and fused to generate a unified anomaly score D.

[0062] Euclidean distance calculation:

[0063] ;

[0064] This is the embedding vector at the current time. The center embedding vector of the historical normal sample. It is an L2 norm.

[0065] An anomaly detection threshold A is set, and the mean score of healthy samples plus three standard deviations is calculated using the 3σ rule as the threshold. If the anomaly score D is greater than the threshold A, the corresponding device is marked as an anomaly and an anomaly label is obtained; otherwise, it is marked as healthy and a healthy label is obtained. The proportion of anomaly devices in the area is statistically analyzed. If the proportion of anomaly devices in the area exceeds a preset threshold I, the entire area is marked as an anomaly.

[0066] Through the multi-scale autoencoder reconstruction analysis and Siamese contrastive learning mechanism described above, deep feature patterns of equipment operation are captured at multiple time scales, enabling early anomaly identification and accurate status label generation, providing reliable data support for subsequent health status matrix construction and environmental control strategy formulation.

[0067] S3. Combine the temperature control intelligent agent with the status tag, and obtain the health status matrix by calculating the device status data.

[0068] Specifically, each temperature control intelligent entity Based on the equipment status tags (health tags and abnormal tags) obtained in step S2, heat source power calculation and load risk assessment are performed in conjunction with equipment status data to construct a regional status vector and ultimately form a global health status matrix. Specifically, this includes steps S3.1 to S3.5:

[0069] S3.1 Calculate the heat source output power of the equipment based on the equipment status data;

[0070] Each temperature control intelligent agent Heat source power calculations are performed for all equipment within the area, based on real-time acquired current data and equipment equivalent thermal resistance parameters:

[0071] ;

[0072] In this formula, Let be the heat source output power of the i-th device at time t. Let be the real-time current value of the i-th device at time t. Let be the equivalent thermal resistance coefficient of the i-th device. For time variables, For device index number.

[0073] S3.2. By analyzing the heat source output power of all devices corresponding to health tags, the total heat source output power of the area is obtained;

[0074] Devices marked with health labels were selected, and their heat source output power was aggregated and calculated to obtain the total regional heat source output power.

[0075] ;

[0076] In the formula, Let be the total output power of the heat source in the j-th region at time t. Let j be the set of indices of all health devices in the j-th region. Let be the heat source output power of the i-th health device, and j be the region index number.

[0077] S3.3 Calculate the load risk index of the device corresponding to each health tag, and aggregate the obtained load risk indices to obtain the regional average load risk;

[0078] Calculate the load risk index for each health device within the region:

[0079] ;

[0080] In the formula, Let be the load risk index of the i-th device at time t. The rated maximum thermal power output of the i-th device,

[0081] Aggregate the load risk index of all health devices within the region and calculate the regional average load risk:

[0082] ;

[0083] In the formula, Let be the average load risk index of the j-th region at time t. Let J represent the number of health devices in the j-th region. Let j be the set of health device indexes in the j-th region.

[0084] S3.4 Combine the total output power of the heat source in the region and the average load risk in the region to form a region state vector;

[0085] The total output power of regional heat sources and the average load risk of the region are combined to form a regional state vector:

[0086] ;

[0087] In the formula, Let j be the state vector of the j-th region at time t. Let J be the total output power of the heat source in the j-th region. Let be the average load risk index for the j-th region.

[0088] S3.5 Integrate the regional state vectors of all analysis regions to generate a global health state matrix.

[0089] Through the above steps, the system achieves the transformation from equipment-level health status identification to regional-level heat source-health status modeling. The global health status matrix not only reflects the heat load distribution in each region but also integrates equipment health status information, providing structured data support for subsequent microclimate situation map construction and multi-agent game control. This matrix can be updated in real time, reflecting changes in heat source distribution and the evolution trend of equipment health risks inside the mobile high-pressure room.

[0090] S4. Collect environmental data of each analysis area in real time, combine the health status matrix to form a microclimate situation map, and use a fuzzy PID controller and multi-agent game mechanism to generate the optimal zone temperature control command.

[0091] Specifically, each temperature control agent continuously collects regional environmental data, integrates it with the health status matrix constructed in step S3, and forms a microclimate situation map through visualization technology. Then, based on a multi-agent game mechanism and a fuzzy PID control algorithm, it generates the optimal control command. The steps for generating the optimal zone temperature control command include S4.1~S4.5:

[0092] S4.1. Collect environmental data of each analysis area in real time, and fuse the environmental data with the health status matrix to form a complete regional status vector;

[0093] Set the data acquisition period \Delta t, and each temperature control agent Agent_j continuously collects environmental data within the partition according to the set period to construct an environmental state vector:

[0094] ;

[0095] In this formula, Let be the environmental state vector of the j-th region at time t. Let j be the temperature value of the j-th region. Let j be the relative humidity value of the j-th region. Let j be the pressure difference value of the j-th region. Let be the PM2.5 concentration value of the j-th region.

[0096] Combine the environmental state vector with the health state matrix Data from corresponding regions is fused to form a complete state vector for the region:

[0097] ;

[0098] In the formula, Let j be the complete state vector of the j-th region at time t. Let J be the total output power of the heat source in the j-th region. Let be the average load risk index for the j-th region.

[0099] S4.2. Draw a visual heat map based on the complete state vector of the region to form a microclimate situation map;

[0100] The steps involved in forming a microclimate situation map include A1 to A3:

[0101] A1. The physical location coordinates of the mobile high-voltage chamber are divided by the analysis area, and the environmental data is visualized by drawing a heat map.

[0102] First, based on the region division results obtained in step S1, the physical location coordinate information of each analysis region is extracted, including the coordinates of the region's center point, boundary range, and spatial geometry. Based on this spatial location information, the system establishes a two-dimensional or three-dimensional cabin spatial coordinate system.

[0103] For the visualization of environmental data, the system employs a spatial interpolation algorithm to map discrete environmental data points collected from various regions onto a continuous spatial grid. The specific process includes: taking environmental parameters such as temperature, humidity, pressure difference, and PM2.5 as numerical inputs, and generating a smooth numerical distribution field within the cabin space using inverse distance weighted interpolation or Kriging interpolation methods. Subsequently, color mapping technology is used to map the numerical ranges to a predefined color gradient scheme, with high-temperature areas displayed in red and low-temperature areas in blue, forming an intuitive environmental data heatmap base layer.

[0104] A2. The environmental data in the health status matrix is ​​used as an overlay annotation for the situation map to form multiple layers;

[0105] Based on the global health status matrix constructed using the S3 steps, different types of status information are processed in layers to form four independent visualization layers:

[0106] The first layer is the temperature and humidity distribution layer, which spatially renders the real-time temperature and relative humidity data of each region. It adopts a dual-color mapping scheme, with temperature represented by red-blue spectrum and humidity represented by light and dark shades. The two are superimposed to form a composite color distribution.

[0107] The second layer consists of a pressure differential distribution layer and a PM distribution layer. The pressure differential data is displayed using contour lines to show the pressure gradient distribution inside the cabin. The PM2.5 concentration is represented by a scatter plot, with higher concentration areas having higher point density.

[0108] The third layer is the heat source load distribution layer. Based on the total output power data of regional heat sources in the health status matrix, it is rendered using a heat map method. Areas with high power density are highlighted to form a visual representation of the equipment's heat source distribution.

[0109] The fourth layer is the health status label layer. Based on the regional average load risk index and equipment abnormal status labels, status indicator symbols are superimposed at the corresponding spatial locations. Healthy areas display green labels, while abnormal risk areas display yellow or red warning labels.

[0110] The multiple layers include a temperature and humidity distribution layer, a pressure difference distribution layer, a heat source load distribution layer, and a health status indicator layer;

[0111] A3. Merge the multiple layers to form the microclimate situation map;

[0112] Layer blending technology is used to synthesize four independent layers according to preset transparency and weight parameters. During the merging process, the bottom layer temperature and humidity distribution map serves as the basic background, the pressure difference and PM distribution serve as the intermediate layers to provide local environmental details, the heat source load distribution map serves as the important information layer for weighted overlay, and the health status indicator map serves as the top layer to highlight key status information in symbol form.

[0113] The synthesized microclimate situation map possesses a unified ability to express multi-dimensional information, allowing maintenance personnel to simultaneously observe the distribution of environmental physical parameters, equipment heat source status, health risk areas, and air quality conditions on a single map. The system also provides layer visualization control functionality, allowing users to selectively display or hide specific layers as needed, achieving a personalized situation monitoring view.

[0114] The resulting microclimate situation map is presented in real time, with the data refresh cycle synchronized with the environmental data collection cycle, ensuring that the situation map reflects the real environmental status and equipment health of the mobile high-voltage room at the current moment.

[0115] S4.3 Each temperature control agent constructs a multi-factor benefit function based on the microclimate situation map;

[0116] Each agent constructs a multi-factor return function based on normalized metrics and global control input matrix data:

[0117] ;

[0118] In the formula, Let the value of the profit function for the Agent in the j-th region be . Let j be the current normalized temperature value of the j-th region. The normalized temperature value for the global target. Let j be the normalized relative humidity value for the j-th region. The humidity value is normalized to the global target. Let j be the normalized heat source load value for the j-th region. Let j be the normalized health abnormality score for the j-th region. , , , These are the weighting coefficients for temperature, humidity, heat source load, and health status.

[0119] S4.4. Initiate the collaborative game mechanism, and each temperature control agent sends the multi-factor payoff function to the main coordination node to solve the global optimal strategy.

[0120] Each agent sends its local reward function to the master coordinator node to perform Nash collaborative game optimization:

[0121] ;

[0122] In the formula, The optimal control instruction set for the j-th region. The set of control variables to be optimized for all regions. This represents the total number of regions. This represents the total energy consumption of the system at the current moment. This represents the upper limit of system energy consumption.

[0123] The game output is the optimal set of control instructions for each region's agent:

[0124] ;

[0125] In the formula, For the optimal air supply flow rate target of the j-th region, For the optimal cold and heat source power target of the j-th region, The target is the optimal valve angle for the j-th region.

[0126] S4.5. Based on the game result, start the fuzzy PID controller to calculate the continuous control signal, and superimpose the continuous control signal with the game result to form the optimal zone temperature control command.

[0127] Each agent activates a fuzzy PID controller based on the game output and the current temperature error:

[0128] ;

[0129] In the formula, Let J be the continuous control signal vector of the j-th region. Let J be the temperature deviation of the j-th region. ; , , Let the proportional, integral, and differential gain coefficients be those for the j-th region. This is the integral term for the temperature deviation. This is the differential term of the temperature deviation.

[0130] The final execution instruction is formed by superimposing continuous control signals with the game solution.

[0131] ;

[0132] In the formula, This is the final execution instruction vector for the j-th region. For the ultimate target air volume, For the ultimate power target, To achieve the final target valve opening, , , This refers to the dynamic correction increment of the game result by the fuzzy PID controller.

[0133] Through the above steps, the system realizes the fusion modeling of multi-dimensional environmental data and equipment health status, constructs an intuitive microclimate situation map, and generates zoned temperature control instructions that take into account both global optimization and local precision through the organic combination of Nash collaborative game mechanism and fuzzy PID controller.

[0134] S5. Adjust the environment of each analysis zone using the optimal partition temperature control command.

[0135] Specifically, the optimal zone temperature control command generated in step S4 is translated into specific equipment control actions, achieving precise adjustment of the environment in each zone through the coordinated work of three execution subsystems. This includes steps S5.1 to S5.4:

[0136] S5.1 The optimal zone temperature control command is divided into air supply control target, cold and heat source control target and air valve adjustment target;

[0137] Three sets of device control targets are extracted from the final execution instructions, and device interface matching is performed:

[0138] From the optimal zone temperature control command The breakdown yields: air supply control target, cold and heat source control target, and air valve regulation target.

[0139] Based on the equipment configuration list of each region, the control targets are mapped and matched with specific physical devices to establish the correspondence between control commands and device interfaces.

[0140] S5.2 The air supply control subsystem adjusts the fan speed and controls the air volume according to the air supply control target;

[0141] The air supply control subsystem extracts the target air volume from the final control command. The required frequency control signal for the fan is calculated by proportionally converting the signal based on the maximum allowable air volume of the equipment.

[0142] ;

[0143] In the formula, Let be the target frequency converter frequency for the j-th region's wind turbine. This is the operating frequency of the wind turbine foundation. Let be the maximum air volume of the fan in the j-th region.

[0144] The frequency signal is transmitted to the fan inverter via analog voltage output (0-10V) or Modbus communication protocol. After receiving the control signal, the inverter automatically adjusts the fan motor speed to achieve precise airflow control. Changes in fan speed directly affect the air velocity in the area, enabling the environment to respond quickly to changes in temperature requirements.

[0145] S5.3 The cold and heat source control subsystem controls the compressor to cool or the heating element to heat according to the cold and heat source control target;

[0146] The cold / heat source control subsystem selects the appropriate controlled object based on the current operating mode (cooling or heating). The target cooling / heating power extracted from the final command is directly used as the equipment power setpoint. The controller converts the power setpoint into a PWM pulse width modulation signal or a 0-10V analog voltage signal to drive the compressor or PTC heating element. The compressor transfers heat through refrigerant circulation, and the PTC heating element raises the temperature through resistance heating, ensuring that the heat source regulation can accurately respond to changes in environmental load requirements.

[0147] S5.4 The damper regulation subsystem performs opening changes according to the damper regulation target to achieve directional air supply;

[0148] The damper regulation subsystem reads the target damper opening degree and the maximum opening angle of the damper and performs a proportional conversion. The angle control signal is output to the damper electric actuator in the form of a pulse signal or a 0-10V analog signal. After receiving the control signal, the actuator drives the damper blades to rotate, adjusting the damper opening angle, thereby changing the cross-sectional area of ​​the airflow channel. Through precise control of the damper opening degree, the system can achieve directional guidance of airflow, accurately delivering hot or cold air to the target area, realizing a refined environmental control strategy of "air delivery to the point".

[0149] The execution status of all control actions, including actual airflow feedback, power output feedback, damper position feedback, and control command execution errors, is fed back to the temperature control agents in each area in real time and recorded in the local control log. The system continuously monitors the execution effect through a closed-loop feedback mechanism to ensure the accurate implementation of control commands and the effective achievement of environmental regulation goals.

[0150] Through the coordinated execution of the above four steps, the system achieves a complete transformation from abstract control strategies to specific physical actions, ensuring the accurate execution of the multi-agent game optimization results at the actual equipment level, and providing mobile high-voltage rooms with efficient, energy-saving, and precise zoned environmental control capabilities.

[0151] Example 2 is an embodiment of the present invention, which provides an environmental control system for a mobile high-pressure chamber, comprising:

[0152] The data acquisition module is used to collect data on the cabin structure, equipment status, and environment of the mobile high-pressure chamber.

[0153] The intelligent analysis module is used to reconstruct and analyze the equipment status data to obtain status labels, calculate the equipment heat source output power and load risk index, and generate a health status matrix.

[0154] The situation map generation module is used to integrate environmental data with the health status matrix to form a microclimate situation map;

[0155] The collaborative decision-making module is used to generate optimal zone temperature control commands;

[0156] The environmental control module is used to execute optimal zone temperature control commands to adjust the environment of each analysis zone.

[0157] Example 3: This example also provides an electronic device applicable to an environmental control method for a mobile high-voltage chamber, comprising: a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to implement the environmental control method for a mobile high-voltage chamber as proposed in the above examples.

[0158] This embodiment also provides a storage medium storing a computer program that, when executed by a processor, implements an environmental control method for a mobile high-pressure chamber as described in the above embodiments.

[0159] The storage medium proposed in this embodiment and the environmental control method for a mobile high-pressure chamber proposed in the above embodiments belong to the same inventive concept. Technical details not described in detail in this embodiment can be found in the above embodiments, and this embodiment has the same beneficial effects as the above embodiments.

[0160] Based on the above description of the implementation methods, those skilled in the art can clearly understand that the present invention can be implemented using software and necessary general-purpose hardware, and of course, it can also be implemented using hardware, but in many cases the former is a better implementation method. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as a computer floppy disk, read-only memory (ROM), random access memory (RAM), flash memory, hard disk, or optical disk, etc., including several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods of the various embodiments of the present invention.

[0161] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A method for environmental control of a mobile hyperbaric chamber, characterized by, The method comprises the following steps: dividing a mobile high-pressure chamber into A analysis areas, and setting up a temperature control agent in each analysis area; collecting device state data of the analysis area through the temperature control agent, using a self-encoder to perform reconstruction analysis on the device state data, and obtaining a state label; combining the temperature control agent with the state label, and obtaining a health state matrix by calculating the device state data; collecting environmental data of each analysis area in real time, combining the health state matrix to form a microclimate situation map, and using a fuzzy PID controller and a multi-agent game mechanism to generate optimal partition temperature control instructions; adjusting the environment of each analysis area through the optimal partition temperature control instructions; the state label comprises a health label and an abnormal label; the step of obtaining a health state matrix by calculating device state data comprises: calculating the heat source output power of the device according to the device state data; obtaining the total output power of the regional heat source by analyzing the heat source output power of the device corresponding to all health labels; calculating the load risk index of the device corresponding to each health label, and aggregating the obtained load risk index to obtain the average load risk of the region; combining the total output power of the regional heat source and the average load risk of the region to form a regional state vector; integrating the regional state vectors of all analysis areas to generate a global health state matrix.

2. A method of environmental control of a mobile high-pressure chamber as claimed in claim 1, characterized in that, The step of performing reconstruction analysis on the device state data to obtain a state label comprises: constructing a multi-modal feature vector from the device state data, performing Gaussian filtering processing on the multi-modal feature vector to form a plurality of scale data subsets; constructing a corresponding encoder for each scale data subset, and using a multi-scale joint reconstruction loss function to train and optimize the encoder; inputting the device state data at the current time into the trained encoder to obtain a device state encoding vector; calculating the distance between the device state encoding vector and the historical normal sample center to generate an abnormal score, determining according to a preset threshold and the abnormal score to obtain the state label.

3. A method of environmental control of a mobile high pressure chamber as claimed in claim 2, characterized in that, The step of generating the optimal partition temperature control instructions comprises: collecting environmental data of each analysis area in real time, combining the environmental data with the health state matrix to form a regional complete state vector; drawing a visual heat map according to the regional complete state vector to form a microclimate situation map; each temperature control agent constructs a multi-factor return function according to the microclimate situation map; starting a cooperative game mechanism, and each temperature control agent sends the multi-factor return function to a master coordination node for global optimal strategy solving; starting a fuzzy PID controller according to the game result to calculate a continuous control signal, and superimposing the continuous control signal and the game result to form the optimal partition temperature control instructions.

4. A method of environmental control of a mobile high-pressure chamber as claimed in claim 3, characterized in that, The step of adjusting the environment of each analysis area through the optimal partition temperature control instructions comprises: splitting the optimal partition temperature control instructions into air supply control targets, cold and heat source control targets, and air valve adjustment targets; an air supply control subsystem adjusts the air volume according to the air supply control target by adjusting the fan speed; a cold and heat source control subsystem controls the compressor refrigeration or heating element heating according to the cold and heat source control target. The damper adjustment subsystem adjusts the opening degree according to the damper adjustment target, and realizes directional air supply.

5. A method of environmental control of a mobile high pressure chamber as claimed in claim 4, characterized in that, The step of dividing the mobile high-pressure chamber into A analysis areas includes: Collecting cabin structure data of the mobile high-pressure chamber to obtain point cloud data; Performing a spatial clustering algorithm on the point cloud data to automatically divide the spatial areas according to the device distribution density; Outputting M spatial areas through the clustering algorithm and assigning a unique number to each spatial area; Assigning the devices to the corresponding spatial areas according to the spatial coordinates of the point cloud data and recording the binding relationship.

6. A method of environmental control of a mobile high pressure chamber as claimed in claim 5, characterized in that, The forming step of the microclimate situation map includes: Dividing the physical location coordinates of the mobile high-pressure chamber through the analysis area, and performing visual heat map drawing on the environmental data; Using the environmental data in the health status matrix as situation map annotation to form multiple layers; The multiple layers include temperature and humidity distribution layer, pressure difference distribution layer, heat source load distribution layer and health status identification layer; The multiple layers are merged into the microclimate situation map.

7. An environment control system for a mobile hyperbaric chamber, applying the environment control method for a mobile hyperbaric chamber according to any one of claims 1 to 6, characterized in that, It includes: A data acquisition module for collecting cabin structure data, device state data and environmental data of the mobile high-pressure chamber; An intelligent analysis module for reconstructing and analyzing the device state data to obtain a state label, calculating the heat source output power and load risk index of the device, and generating a health status matrix; A situation map generation module for fusing the environmental data and the health status matrix to form a microclimate situation map; A collaborative decision-making module for generating optimal partition temperature control instructions; An environmental regulation module for executing the optimal partition temperature control instructions to regulate the environment of each analysis area.

8. A computer device comprising a memory and a processor, the memory storing a computer program, characterized in that, The processor executes the computer program to realize the steps of the mobile high-pressure chamber environment control method in any one of claims 1 to 6.

9. A computer readable storage medium having stored thereon a computer program, characterized in that, The computer program is executed by the processor to realize the steps of the mobile high-pressure chamber environment control method in any one of claims 1 to 6.

Images